Copyright is owned by the Author of the thesis. Permission is given for a copy to be downloaded by an individual for the purpose of research and private study only. The thesis may not be reproduced elsewhere without the permission of the Author.

SEROLOGICAL DIAGNOSIS OF BRUCELLA INFECTION

A the s i s presented in partial fu l fi lment of the requirement s

for the degree o f Master o f Veterinary Science

at Massey University, New Zealand

Derek Vernon T imbs

B . V . Sc. ( Queens land )

November , 1978

( i i i )

ABSTRACT

The automated complement fixation te s t (CFT ) and the

bruce l losis card test (BCT ) have been wide ly used as o fficial t e s t s

in the New Zealand Bovine Bruce l losis Eradication Scheme . During

the course of the eradication programme i t was observed that a

s igni ficant proportion o f catt le reacted to the BCT yet remained

negative to the CFT and thi s o ften occurred on more than one

occas ion for any particu lar anima l .

Twenty cows , from reactor herds , that had been BCT+/CFT- on

at least three success ive occasions were s laughtered . De spi te

extensive samp l ing , attempts at i solating Bruce l l a abortus organi sms

from t i s sues o f these animals were un�ucce s s ful . Serum from one

cow was found to be pos i tive to a wide range of sero logical tests

and i t a l so caused a strong prozone react ion in the CFT , which

could easily have been overlooked. The possibility that the automated CFT , which i s e ssential ly a one d i lution te s t , was unab le

to detect such prozoning sera was inves t igated. I t was shown that

providing a s uitab l e choice of ant igen concentrat ion was made ,

such sera would be de tected by the automated te st.

B ruce l la - spec i fic I gG 1 , I gG2 , and IgM leve l s in prozoning

and non prozoning sera were measured u sing the single radial immuno­

d i f fus ion test. I t was shown that serum containing a high

proportion of speci f i c IgG2 was l ike ly to exhibi t prozoning and

that various degree s of prozoning could be induced by varying the

ratio o f speci fic I gG 1 to specific IgG 2•

Cat t le , previous ly sens i t i zed by cal fhood B r . abortus s train 19

vaccination , were experimentally inoculated with kil led �· abortus .

I t was shown that a l though serum agglutination test ( SAT ) and CFT

t i tres appeared for a short period , t i tres to the BCT in some catt l e

tended t o remain longer thus a l lowing an animal t o be BCT+/CFT- .

An analysis o f herd test ing data indicated that BCT+/CFT­

animal s were more l ike ly to exis t in infected herds than in non­

infected herds . In heavily infected herds up to 1 6% o f CFT- animal s

were BCT+ whereas in non- infected or very l i ghtly infected herd s

l e s s than 4% were CFT-/BCT+. I t was concluded that i n sensiti zed

cat t l e at least exposure to the organism without true infection i s

capabl e o f stimulating antibody which i s detected by the BCT, but

not necessar i ly abl e to provoke positive CFT titre s .

!

( iv)

The performance o f the Auto-Analyzer adaptat ion o f the CFT as

used in the New Zealand erad icat ion scheme was assessed . Various

pro zoning sera from known infected animals were te s ted and the e ffect

of varying ant igen concentrations on these and o ther sera was noted .

S i gnificant d i f ferences in ant igen concentrat ion required for

more antigen than non-prozoning sera and even sera that d id not

exhibit prozoning had varying optimal antigen requirements .

By using L125 labe l led bovine gama-globul in.

the d i lution gradient

of serum within the Auto-Analyzer sys tem was e st imated . Knowledge o f

the serum d i lu t ion gradient be ing obtained was e s sent ial for proper

understanding of unusual traces given by prozoning sera .

ACKNOWLEDGEMENTS

The work for thi s thes i s was carried out whi l e in the

emp loyment o f , and with the a s s i s tance o f , the New Zealand

Mini s try o f Agr icu l ture and Fi sheries .

Particular thanks are due to Mr J . W . Moxham , Superintendent ,

Central Bruce l lo s i s Laboratory , Wal lacev i l le Animal Re search

Centre , Upper Hut t , and to Drs R . B . Mar shal l and K .M . Moriarty

o f Mas sey Univers i ty for their interes t , advice and encouragement .

Special thanks go to Mr A .Wo Barkus for as s i s tance in

preparation of photographs and figure s and to Mrs N .M . We thera l l

for typing the manuscrip t .

Acknowledgements

List o f f igures

L i s t of t ables

Introduc tion

Chapter 1 .

Chapter 2 .

Chapter 3.

Chapter 4 .

Chapter 5 .

Chapter 6 .

Chapter 7 .

Appendix I .

References

TABLE OF CONTENTS

Li terature review

Cul tura l and serological s tudies

Exper imental inoculat ion of cat t le

with k i l led Bruce l la abortus

S tudi e s on Bruce l la spec i f ic serum

ant ibody fract ions

Appra i sa l of the Auto -Analyzer

adaptat ion of the comp lement

fixat ion test

Ana ly s i s of f ie ld data

General di scuss ion

Page

(v )

(vii )

( ix)

1

4

48

69

7 2

84

105

1 10

1 13

1 1 6

\V

LIST OF FIGURES

Figure

1 Bruc e l losis card test

2 Auto-Analyzer adaptation of the

complement fixation te s t for brucel losis

3 His togram of titres given by cattle

inocu lated wi th ki l led Bruce l la abortus

s train 19 .

4 Elution gradient of phosphate buffer

used in DEAE ce l lulose column

5 Typ ical trace of absorbance at 254 � of

eluate of DEAE ce l lulose co lumn

6 Typica l trace of absorbance o f 254 � of

e luate from Sephadex G200 co l umn

7

8

Typical electrophoresis p late

Typical single radial immunodiffus ion

p late

9 Schematic diagram of automated complement

fixation test

10

1 1

Principle o f continuous flow analys is as

adapted to the complement fixation test

Continuous d i lution method of titration

1 2 Auto-Analyzer trace given b y continuous ly

d i lu t ing complement in the absence of

s amp les

(vii )

Between pp .

38 - 39

42 - 43

70 - 7 1

7 3 - 74

7 3 - 74

7 3 - 74

7 3 - 74

75 - 7 6

8 4 - 85

84 - 8 .5'

88 - 8 9

9 3 - 9 4

(viii )

Figure Between PP•

1 3 Graph drawn b y p lotting values derived

from chart shown in Figure 1 2

14 �il�tien �radiant •iven by 112319�elled serum

15 His togram of relative serum d i lutions

in each a l iquot taken from Auto-Analy'zer

1 6 Series o f typica l antigen t i trat ion

charts from Auto-Analyzer

1 7 Ant igen ti tration with varying serum

dilutions of a prozoning serum

18 Antigen t i tration wi th varying serum

d i lutions of a non-prozoning serum

19 Antigen t i tration wi.th varying serum

d i lutions of a non-prozoning serum

20 Effect o f variat ion in antigen

concentrat ion on prozoning sera in

automated complement fixat ion test

I•

9 3 - 94

96 - 9 7

96 - 9 7

9 8 - 9 9

9 8 - 99

98 - 99

98 - 99

98 - 99

Table

I

LIST OF TABLES

S train 19 antigen t i tration for

agglutination test , against labgratgry standard serum

II Whole ce l l s train 19 ant igen , t itration

for warm fixation CFT

II Sonicated strain 19 antigen , ti trat ion

for warm fixation CFT

IV Ether-water treated strain 19 ant igen ,

titration for warm fixation CFT

V Titres to agglutination type te s t s o f

repeat BCT+/CFT- animal s s laughtered

VI Complement fixation test resu lts for

repeat BCT+/CFT- animals· '

VII Complement fixation test resul t s for

repeat BCT+/CFT- animals

VIII Complement fixation te st resu l t s

I

IX Rivanol comp lement fixation test

commercia l s train 99 antigen

X Immunoglobul in concentration and Brucella

specific antibody percentage rela ted to

degree of prozoning of warm CFT t i tre

XI Speci fic and total immunoglobulin

concentrations in al iquots of IgG1 and

IgG2

XII Results of warm fixation CF te sts for

mixtures of specific I gG1 and IgG2

XIII Reaction o f I gG1 and I gG2 to serum

agglutination and card te sts

XIV Summary o f leve l s of .. radioactivity

recorded before and after d i lution o f

samples by AutoAnalyzer

(lx)

Page

59

59

60

60

62

63

64

65

65

78

80

80

80

97

(x)

Tab le Page

XV Manual CFT results for sera giving

peaks shown in figure 6 . 3 99

XVI BCT and CFT reac tor ratea at initial

herd te sts 107

I•

INTRODUCTION

Bovine brucellosis is a contagious disease of cattle resulting

from infection with the bacterium Brucella abortus.

The disease is primarily manifest as a placentitis in the

pregnant female with a resultant abortion. Humans may also be

infsGtQd by ths GaYsative g��aniem 1n which casQ thQ disQQ&Q s

known as undulant fever.

Because of the contagious nature of the disease, the

prevalence of infection within an unprotected herd may be very high

and persons in close contact with such an infected herd run a very

real risk of contracting undulant fever�

Apart from the human involvement the importance of the disease

lies in the loss of calves and in the resultant loss of milk

production from aborted cows.

The organism responsible for contagious abortion of cattle

was first described by Bang (1897), and termed Bacillus abortus.

Some years earlier, Bruce who had been investigating the

cause of Malta fever in troops succeeded in culturing a micrococcus

from the spleens of four fatal cases (Bruce,l887). He termed this

organism Micrococcus melitensis.

The close relationship between Bacillus abortus and

Micrococcus melitensis was not recognised until 1918 when Evans

(1918) correctly identified M. melitensis as a bacillus and

suggested the designation Bacillus melitensis. She also noted

that the agglutination test performed on human serum could not

distinguish infection due to B. melitensis from that caused by

B. abortus.

Meyer and Shaw (1920) confirmed the work of Evans and

proposed the term Brucella be adopted as the generic name in

honour of the pioneering work of Bruce.

In the 1920's the role of Brucella abortus in human infection

was established and Brucella suis was recognised as a distinct

species.

Three further species of Brucella were later isolated.

Buddle and Boyes (1953) described Brucella ovis for the first

time and in 1957 Stoenner and Lachman (1957) described Brucella

neotomae which they had isolated from the Desert Wood Rat. The

latest member of the genus, Brucella canis was first described by

Carmichael and Bruner (1968). A recent review of the disease

in dogs has been prepared by Carmichael (1976).

By the late 1920's it had been established that Brucella

abortus could infect man and that raw milk was an important source

of infection. By this time most of the basic epidemiology of the

disease had been established and the serum agglutination test had

been shown to be a useful diagnostic tool.

2 .

Formal attempts at eradication or control of the disease in

cattle began in the United States in 1934 as a drought relief

programme. In Denmark serological testing began in 1936. These

.schemes along with initial attempts in other Scandanavian countries

did not ,however, make great progress. Difficulties in test

standardisation and interpretation and the advent of World War II

hindered eradication attempts.

Renewed attempts at eradication gathered momentum in the late

1940's and these were aided by the development and use of the live

attenuated vaccine Br. abortus strain 19. (Cotton and Buck ,l934).

Another important aid to eradication was the introduction of the

Milk Ring test (Fleischhauer, 1937), which eliminated the need for

bleeding of cattle.

Success in eradicating bovine brucellosis has largely been a

function of the intensity of effort applied along with the type of

cattle husbandry practised. Scandanavian countries have applied

considerable legislative and testing pressure since the 1920's

(Thomsen,l957) and have generally been free of the disease�nce

1970.

In the United States the Co-operative State-Federal Brucellosis

Eradication Programme commenced in 1934, but was given emphasis

only in the early 1950's. It aimed for completion by 1975 but due

to various reinfection problems a more realistic date of 1984 has

now been set (Schilf, 1972; Becton,l976).

The European system of farming with small intensely managed

herds has favoured eradication attempts. Control of the disease in

countries which practise grassland cattle grazing with large herds

and outdoor calving patterns is more difficult. Such differences

in cattle management and in public acceptance of the resources

required , mean that the technology used successfully in Northern

Europe may not be as effective in the more extensive cattle

raising nations.

The New Zealand brucellosis eradication scheme began in

September 1971 with the stated objective of bringing every eligible

animal under compulsory test by August 1977. By this time 69% of

the beef herds and 81% of the dairy herds in the country were

accredited free of brucellosis.

Because of the rapidity with which the New Zealand scheme has

moved and the limited resources available, there has been little

continuing assessment of the effectiveness of technical procedures

used.

The object of this thesis is to describe laboratory aspects

of the New Zealand scheme, to assess the performance of automated

serological techniques and to investigate some aspects of the

serology of certain sera which have given anomalous test results.

CHAPTER 1 . LITERATURE REVIEW

Pathogenesis o f Bruce l l a abortus infect ion in Cat t le

Bruce l la abortus may be descr ibed as a facu l tat ive intra­

cellular microworganiam. It is � Gr&m·n�gaeiv� baeillua about

4 .

400 nm in d iameter and a l though aerobic i t requires 5 t o 1 0 percent

co2 to sustain growth during initial isolation. The characteristic c l inical feature o f bruce l lo s i s in the

bovine i s the loca l isat ion and mu l t ipl icat ion of Br . abortus in

the epi the l ial ce l ls o f the chorion ( Smi th, 1919 ) . Later , by

spread o f infect ion into the p lacental fluid a foetal pneumonia

may be induced ( Smith, l 9 25 ) .

Accord ing to Hudd leson ( 1 943 ) , the cause o f foetal death and

abort ion may be due to interrup t ion of the vital functions o f the

placent a , or to the toxic effects of an endo-antigen .

Payne ( 1 95 9 ) at tempted to fo l low the course o f Br . abortus

infect ion in eleven pregnant cows experimentally infected via the

conj unct iva l s ac . After one week the paro tid nodes were found to

contain organisms suggest ing that Br . abortus penetrates the

conj unct iva l epithel ium and proceeds directly via the lymphat ics

to the loca l lymph gland . After two weeks organisms could be

iso lated from the sp leen and supramammary glands and after four

weeks B r . abortus was cons is tent ly isolated from the u ter ine lumen .

Payne ' s conc lus ion was that fo l lowing co lonisat ion of the local

lymph node the organism is d i st r ibu t e d by the blood and loca lises

in the interst i tial connec tive t i s sue . Inflammation resu l t s and

the bacteria find their way into the u terine glands and then into

the uterine lumen . Mul t ip l ication of Br . abortus occurs in macro­

phages and the exudate contains .these ce l l s packed wi th bacteria .

In the pregnant uterus the interstitial inflammation leads to

a severe u lcerat ive endometrit i s and by the time of abortion

nearly a l l o f the endometrial mucosa i s eroded and rep laced by

sub-acute inflammatory granulat ion tis sue .

Infection o f the p lacental cotyledons fo l lows the invo lvement

o f the uterine lumen a l though this part o f the process i s s low in

onset . Around the periphery of the cotyledons �· abortus induces

u lceration o f maternal tis sue and multipl ies extens ive ly within

chorionic trophoblast ce l l s . Bacteria then migrate through the

connect ive t is sues o f the al lantochor ion and enter the foetal

b lood vesse l s eventual ly invading the p lacental f lu ids and the

foetus .

After ca lving or abort ion Br . abor tus leave s the u terus to

loca l i se in the udder and supramammary lymph nodes (Manthei and

Carter, 1950; Lambert ll .!!.·, 19.6.1) . Phi lippon ll &· ( 1970 )

recovered Br . abortus from the supramammary lymph nodes o f 44 out

of 46 pregnant animals infected a t f ive to s ix and a ha l f months

s.

o f gestat ion and s laughtered s ix weeks after partur ition . The

recovery rate from the udder was only 60% and from the u terus only

5 1% . Other t i ssues where the organism frequently pro l i ferates are

the i l iac lymph nodes , pharangeal lymph nodes and spleen . Bruce l lae

are also found in a high proport ion of knee hygromas (Doyle , 1935 ) .

In an at tempt to determine why the reproduct ive organs should

be so recept ive to Br . abortus , Payne ( 1960 ) experimental ly

infected non-pregnant cows . He d ivided these into two groups and

infected a l l animals wi th one group rece iving dai ly treatments o f

proges terone . The invasiveness o f B r . abortus was much reduced in

the non-pregnant animals and al though proges terone treatment

induced profound changes in the geni tal ia i t had no effec t on the

pathogenesis o f bruce l los i s .

In an invest igat ion o f the chemical factors associated wi th

the growth of Br . abortus Pearce � .!1,. (1962 ) found that erythr itol

was present in high leve ls in bovine foetal fluids and that i ts

presence was required to enable these f luids to s t imulate growth

of B r . abortus in bovine phagocytes. A search of various maternal

and foetal t i s sues of pregnant cows showed that erythrito l was

concentrated in those t issues (chorion , coty ledons and foetal

fluid s ) which in bruce l los i s are prone to heaviest invasion .

Erythritol s timulated the growth of �· abortus l£ vi tro and

1£ � in one to five day o ld calves , and hence was incriminated

as the cause for the pred i lect ion of !!_. abortus for foetal "t i s sue

in bovine contagious abort ion (Wi l l iams � �. , 1962 ; Smi th ll �.,

1962 ) . S imi larly in the bul l , bruce l lae tend to localise in the

genitalia and on invest igation erythri to l was found in the seminal

ves ic les and tes tes (Keppie �!!., 1965 ) .

In later s tudies Keppie � �. ( 19 6 7 ) and Meyer ( 1967) found

that erythritol inhibited the growth o f �· abortus s train 19 and

thi s difference in compari son with viru lent strains was sugge sted

as a chemical bas i s for the re lative "avirulence" o f s train 19 .

6 .

Braude ( 195 la ) detai led the expected e ffects o f chal lenge with

d i f ferent species of Bruce lla and highl ighted the fact that the

granuloma is the basic his tological lesion in the t i s sues o f

animals and man a f ter infect ion . I n a further s tudy Braude ( 195 lb )

detai led the evo lut ion o f the hepatic granuloma in experimental

infect ion o f guinea pigs and mice .

After chal lenge Bruce l lae are rap idly c leared from the b lood

as they are inges ted by polymorphonuclear phagocytes and carr ied to

the l iver , sp leen and ce l l s of the r�t icu lo-endo the lial sys tem .

The pr imary hos t re sponse appears to be a mass ive pro l i feration

and/or migrat ion o f po lymorphonuclear and later mononuclear ce l l s

at or towards the focus o f infect ion where they form a granuloma

wal l ing off the infected ce l l s (Braude , 1 95 lb ) . In the case o f

tuberculosis infection, the granuloma is termed a tubercle - the

analagous name in bruce l lo s i s , a brucercle , does not seem to have

been serious ly proposed .

Another characteristic feature o f Bruce l la infect ion is the

abi l i ty of the organi sm to survive wi thin the phagocytes o f the

hos t (Ho l land and P ickett , 1 958 ; S t inebring and Ke sse l , 1959 ;

Elberg , 1960) . Smith ( 19 1 9 ) de·s'cribed the intrace l lu lar

loca li sation o f Br . abortus within the e p i the lial ce l l s o f bovine

foetal membranes . · The bacteria were not only contained in the s�

ce l l s , but were protected so that mu l t ip l icat ion of the organism

could proceed wi thout interrupt ion . Dickey and Forbus ( 1945 )

noted the immediate phagocytos i s o f Br . suis by polymorphonuc lear

ce l l s and the re sul t ing mul tipl icat ion o f the organism. They

postu lated that such phagocytos i s might actually be detrimental

to the hos t since the bacteria may not be ki l led but protected

by this local isat ion .

The intrace l lu l ar loca l isat ion and growth of Bruce l l a has been

confirmed by many investigators . Even in the b lood , unless there

is overwhelming infect ion , the bacteria are present within

leukocytes (Braude , 195 lb ) . S tudy of the abi l i ty of Bruce l l a to

mu l t ip ly within mononuclear phagocytes maintained in t i s sue cu lture

has shed much l ight on the nature of virul ence . Virulent s trains

have thi s abi l ity , whi le aviru lent s trains do not (McCul lough ,

1970) .

7 .

In the infected bu l l two c l inical syndromes have been reported

and these may or may not occur in the same animal . The first

involving the tes t ic l e and epid idymis i s o f ten characteri sed

c l inica l ly by orchi t i s (Buck � � . , 1 9 1 9 ; King , 1932 ; Bend ixen and

B lom , l947 ; Rankin , l9 65 ) . The second , and apparent ly more common

•yndrome tnvolve• ehe aemina l vesie l es and ampullae (nuek �al.,

19 19 ; Bendixen and B lom, l947; Rankin.l965 ) . Few reports re lating

to exper imental · infect ion and pathogenes i s of the disease in bu l l s

are avai lable ( Seddon, l919 ; Lubbehusen and Fi tch, l926 ; Chris tensen .

1948) . The bu l l , however , is not recognised as p laying a

s igni ficant role in the transmiss ion of the disease (Thomsen, l943 ;

Rankin , l965 ) .

Thus , in summary , a Bruce l la chal l enge can lead to a chronic

long term infect ion , characteri sed by a low leve l , intermittent or

f luctuating ant ibody t i tre . The infected cow tends to abort once ,

and then settles down into a chronic carrier cond i t ion . The

Brucellae become localised, particularly in the suprarnarnrnary

lymph nodes and udde r , and may be excreted in the milk , inter­

mittent ly , for many years (Morgan , l969 ) .

The Epidemiology of Bovine Bruce l lo s i s

A sound knowledge o f the sources o f infection , hos t

suscept ib i l i t ies and organism survival are required before any

di sease contro l programme can be attemp ted . A great deal has

been wri tten about the epidemio logy of bovine bruce l lo s is and a

certain amount o f confus ion has arisen because o f ear l ier

conflicting s tudies .

Sources o f Infect ion

( a ) The infected cow

8 .

The importance o f the aborted cow as a source o f infection

was recognised by B ang ( 1897 ) . Large numbers of bruce l lae are

shed in genital discharges fo l lowing abor tion or after an

apparent ly normal ca lving from an infected cow . Numbers rapidly

decrease one to two weeks a fter calving . Chronica l ly infected

cows may excrete organisms fol lowing subsequent ca lvings . Thus

the primary hazard wi thin a herd is the aborting cow during the

first 10 days after the abortion (Fi tch � �. , 1938 ; Manthei and

Carte� 1950) .

In the chronic d i sease organisms loca lise in the udder and

re lated lymph nodes. Up to 2 x 105 organisms per ml may be found

in the colos trum or mi lk wi thin the first few days fo l lowing

ca lving or abortion . Numbers decrease rapidly after the first

week but intermit tent excretion wi l l o ften continue for the

remainder of this and subsequent lactations . S train 19 vaccination

wi l l reduce but not necessari ly e liminate thi s excretion (Morgan

and McDiarmid , l960 ) .

( b ) The infected bul l

Infected bul l s may shed bruce l lae via the semen , seminal

fluid , faeces , urine and hygroma fluid . S emen may contain large

numbers o f organisms during the ini tial s tages o f infection but

as the d i sease progresses excretion is reduced and may cease

entire ly (Lambert�� . , 196 3 ; McCaughey and Purce l l , 197 3 ) .

( c ) Secondary or transient host s

The FAO/WHO Expert Commit tee on Bruce l lo s i s (Anon . , 197 1 ) notes

that the wide hos t range i s one o f the important characteri s tics

o f the genus Bruce l l a , even though particular biotypes may have

a l imited hos t range .

As mentioned previous ly �· abortus infection o f man i s we l l

known a lthough there are no reports o f the agent caus ing abortion

in women. Tran smi s sion from man to catt l e is not known to occur .

9 .

O f the domestic anima l s , infect ion o f the horse i s recognised

and is usually associated wi th 'fistulous wi thers ' and arthri t i s

( S tab le for th , 1959) . Hutchins and Lepherd ( 1968 ) noted that horses

in contact wi th infected dairy herds had a higher incidence o f

pos i t ive Bruce l la t i tre s than · horses wi thout such contact.

McCaughey and Kerr ( 1967 ) , Shortridge ( 1967 ) , Cros sman and Bonson

( 1968 ) and Robertson � �· ( 197 3 ) have a l l reported abort ions in

mare s apparent ly due to Br . abortus infect ion.

B r . abortus appears to have low pathogenici ty for sheep but

sporadic natural infect ions have been reported (Luchs inger and

Anderson , 1967 ; Al lsup , 1969) . Shaw ( 19 7 6a ) recovered Br . abortus ,

biotype I from the foetuses o f f ive out o f eight abort ing ewe s .

These ewe s were from a group o f 2 0 known t o have had contact wi th

catt le having the same biotype . In a subsequent s tudy Shaw ( 1976b )

experimentally infected four ewes by adminis tering a suspens ion o f

infected bovine cotyledon oral ly. Thus a l though sheep can sust•in

infection and may abort , the low prevalence and incidence o f the

di sease on a f lock basis po ints to sheep as being of minor

importance as reservoirs of infect ion (Al l sup , 1974 ) .

A case in New Zealand reported by. Wal lace (1959) suggests

that p i gs shou ld be cons idered a s a pos s ible re servoir or carrier

o f Br . abortus. A sow which had access to infect ive material from

infected cows aborted and al though organisms cou ld not be cul tured

from foeta l membranes , which showed a necrot ic placent i t i s , they

were recovered from foetal s tomach contents.

In the USA , McCul lough � �· ( 195 1 ) obtained 35 B ruce l la

isolates , 10 o f which were Br. abortus , from cu ltures o f sub­

maxi l lary lymph nodes of 5 000 p igs at s laughter . Surve i l l ance

of human bruce l lo s i s indicates that Br. abortus infect ions o f

swine may b e more common than p reviously thought . Four o u t o f

eight i so lates o f Br . abortus i n humans i n the Uni ted S tates in

10 .

1 9 7 1 were from patients infected as a resu l t o f contact with p igs

only ( Anon . , 1 9 7 1 ) . Abort ion in the bi tch due to !s· abortus has been reported by

Morse � � . ( 1 953 ) in the USA , Phi l ippon � �. ( 1969 ) in France ,

Ehr lein � � . ( 1963 ) in Germany and by Taylor � � . ( 1975 ) and

Bi�knell £t £l, ( l976 ) in Great Britain, It has been su�,ested that the prevalence o f bruce l lo s i s in dogs is higher than can be

inferred from the infrequent reports in the li terature (C legg and

Rorrison , l969 ) .

Of the wi ld animals known to be infected wi th Br. abortus ,

mos t reports have come from Europe and North America . Hares ,

re indeer and rodents have been shown to be infected . Came l s ,

yaks , buffalo , mink , foxes and various other ungulates have a l so

been shown to be reservo irs o f infection (Anon . , 197 1 ) . Meyer

( 1966 ) in a comprehens ive review of the ro le of o ther animal s

concludes that Bruce l l a organisms are not read i ly transmis s ib l e

from their pre ferent ial hos t t o d i s s imi lar hosts and that no

ser ious or threatening reservoir o f infect ion e xists in wi ld

animals in the United S tates.

Danie l ( 1966, 1967 ) and McAl lum ( 1976 ) have tes ted

New Zealand deer sera for Br . abortus but only one pos i t ive and

three suspicious t i tres were obtained from some 700 sera .

In the absence o f any evidence to the contrary i t appears

that no animal other than the bovine has any s i gnificant ro le to

p l ay in the s pread of bruce l lo s i s in New Zealand.

( d ) Mechanical hosts

Hudd leson ( 1943 ) has given the fo l lowing data for the

survival t ime o f Bruce l la abortus

Direct s unlight

Dry so i l

Wet so i l

S ter i le tap water

Milk

Urine

Faeces

Genital discharge

Foetus

. .

(Room temperature )

11 11

11 11

11 "

11 "

11 "

( Ice box)

( Shade )

4� hours

1 - 2 months

2- 3 months

2 - 3 months

2 -4 days

2 -4 days

3-4 months

7 months

6-8 months

Ice cream

Butter

Cheese

( e ) Milking machines

1 month

1 - 2 months

(Room temperature ) 2 months

There is no doub t that Brucellae are excreted in the milk . However , op inions vary on the ro le p layed by the milking machine

in the transmiss ion of bruce l losis . Lapraik � � ( 1 964) found

1 1 • .

no evidence of spread by excre t ing cows at milking . Leech � a l . ( 1964 ) also conc luded that brucellosis was not read ily spread by

the milking machine . In Northern Ireland spread via infected mi lk

at milking was thought a rea l pos sibility in infected herds (Kerr

and Rankin, 1959 ) and mastit i s workers read i ly accept that

Staphylococc i and S treptococci can be trans ferred from udder to

udder by the mi lking machine (Davidson and Slavin, l958 ;

Whittles tone � � . , ( 1968 ) .

Methods o f Infect ion

Brucellae can be transmi tted to a susceptible cow by :

( a ) Inges t ion

B ang ( 19 06) or igina l ly showed that this was a route o f

infection . I t i s thought that this i s the most usual natural

route a l though the dose of bacteria needs to be large (Cotton

and Buck, l931 ; W i l son and Mi les , l967 ) .

( b ) Inhalation

Infection by inhalation occurs in guinea p igs ( Elberg and

Henderson ,l948 ) and man (Anon . , 197 1 ) . Challenge via the respiratory

tract must be cons idered a real hazard in cattle .

( c ) Direct contact

Transmiss ion may occur through unbroken skin and more

readi ly through broken skin . Challenge via the conjunct iva has

been commonly used in experiments and low doses induce infect ion

(Cotton , and Buck , l 9 31 ) . Infection is a l so said to be acquired

through the udder at mi lking (Kerr and Rankin, l959 ) .

1 2 .

( d ) Coi tus

Cows may be infected via the genital tract when served by an

infected bul l (Thomsen, l 9 36 ) or by art i ficial insemination wi th

infected semen ( Se i t , l944) . I t appears that natural service by

the bul l i s o f negl igib l e importance in the transmiss ion o f

bru eel l osis (King,l94o; Thomsen , 194 3 ; Rankin, l965; Barc lay, 1977).

However , art i ficial inseminat ion has been held responsib le for

caus ing widespread infection in cows (Bendixen and B lom , l947 ;

Manthei � � . ,1950 ) . In natural inseminat ion semen i s depos i ted

in the vagina whereas when art i ficial insemination is used semen

i s introduced into the u terus . I t appears that a much higher

dose of organisms is required to infect via the vagina than via

the uterus (Manthei � �. , 1950) .

( e ) Congenital pas s ive transfer

Fi tch � �. ( 194 1 ) reared 56 hei fer ca lves in isolation from

natura l ly infected mothers . No evidence o f bruce l los i s was found

in these animals dur ing the seven years o f the experiment . This

work led to the genera l as sumpt ion that congenital transfer of

infect ion did not occur . �lommet � �· ( 1 9 7 3 ) proved other­

wise . They showed that ca lves born of infected dams cou ld harbour

the organism without any c l inical or sero logical manifestation

unti l dur ing pregnancy. At this s tage the disease became apparent and usua l ly resul ted in abortion .

The Susceptible Animal

�· abortus has a special affinity for genital t i ssues in

the pregnant cow. The non-pregnant cow and the mature unserved

heifer are said to show some innate res i s tance to chal lenge

( Edington and Donham , l 9 39 ) but the s i ze o f the challenge dose i s

important . I n a series o f ini tial herd tests i n Great Britain 28%

o f cows , 3 . 4% of he i fers and 13 . 2% o f bu l l s reacted ( S tab le fo�th ,

195 9 ) . Calves are re lative ly res i s tant to infect ion and many

experiments have been conducted to determine thei r suscep t ib i l i ty

by feeding them infected mi lk and letting them s uckle infected

cows . In mos t cases they fai l to produce any d emonstrable serum

antibodies (Huddleson , 1942 ) . Carpenter ( 1924) c laimed that

1 3 .

Bruce l lae can be recovered from the organs o f ca lves fed infected

m i lk but disappear short ly after feeding is discontinued .

However , the susceptib i l i ty of young animal s to Br . abortus

remains a controversial subj ect as Plommet � !!.1•( 1973 ) and Lapraik

� &. ( 1975 ) have shown that ca lves born o f infected mothers and

reared in isolation can retain a latent infection . There is also

much circumstant ial evidence l inking breakdowns o f Bruce l la- free

herds to re infect ion from carrier ca lves (Tacken, l964 ; Rankin ,

1 965 ; Cunningham and O ' Connor , l97 1 ; Tarala , l97 5 ) . Breakdowns

o f Bruce l l a- free herds , apparent ly due to carry-over o f latent

infect ion in calves , have also been seen in New.Zealand ( J . Wa l lace

Pers . Corn. ) . Thus it appears that al though congeni tal and ca l f­

hood infect ion wi th Br . abortus can in fact occur , the ca l f remains

re lative ly insuscep t ib l e and such infect ion is uncommon . It i s ,

neverthe less , an extreme ly important cons iderat ion in the context

o f eradicat ion programmes because of the long interval between

infect ion and the deve lopment o f serological t i tre and/or c l inical

symptoms .

Incubation Period

Many of the problems associated with the control o f

bruce l losis i n cat t le are due t o the long and variable incubat ion

per iod of the disease .

The length o f the incubat ion period varies according to the

state of pregnancy o f the cow. Thomsen ( 19 37 , 1949 ) found that

for hei fers infected a t the t ime o f mat ing the average period

between infect ion and abort ion was 2 25 days , whereas in cows

infected at s ix and seven months o f pregnancy , the incubat ion

per iod was 67 and 5 3 days respective ly . Animals infected as

calves may incubate the disease unt i l late pregnancy which can be

up to two and a half years later (Nagy and Hignet t , 1 967 ) .

There seems to be a relat ionship between the s i ze o f the

challenge dose and the development o f s ero logica l t i tre and

abortion . McEwan � &. ( 1939 ) infected hei fers in their fi f th

month of gestation and found that whereas a dose o f 1 . 46 x 109

5 o rganisms gave an incubation period o f 3 2 - 5 9 day s , 1 . 46 x 10

o rganisms gave 83- 15 1 days .

Animal s vaccinated with s train 19 Bruce l l a abortus are said

to take much longer to deve lop serological t i tre s to infect ion

than do unvaccinated animals ( S chi l f , 1 968 ) . Pre sumab ly the

same app l i e s for the deve lopment o f infect ion .

Course o f the Disease Wi thin the Herd

1 4 • .

In an acute outbreak in a fu l ly susceptib le herd the abort ion

rate may exceed 50% . More usual ly the rate of spread i s s low

extending over several seasons with a peak abortion rate o f about

30% . He i fers do not usual ly acquire infection unt i l brought

into the herd at first calving . Once infect ion is es tab l i�hed

it tends to pers i s t indefinitely ( S tab leforth 195 9 ) .

Ep idemiological S tudies Associated Wi th Eradicat ion Programmes

The obj ect o f any epidemiological s tudy in bruce l lo s i s i s to

further knowledge of the way in which the d isease spreads so that

appropriate targets may be ident i f ied for particular at tent ion in

the event o f an attempt at erad ication or contro l of the d isease .

The real test o f the value of s tudies previously made comes when

eradicat ion is tried . In the f inal stages of an eradicat ion

scheme prob lems arise and the inves tigation of these can lead to

a greater understanding of the ep idemiological factors as sociated

with test e ffect iveness and d i sease spread .

The deve lopment o f tes ting methods i s de tai led in a .later s-ection .

Bacteriological s tud ies o f bruce l losis "problem" herds have

been made by Ne lson � �· ( 19 66 ) and Luchs inger � � . ( 19 7 3 ) .

These workers conclude that the b iotype o f �· abortus respons ible

for a par t i cular infection has no bearing on the type o f

eradication problem that may exi s t. Luchs inger � �. ( 19 7 3 ) also

point out the value of routine biotyp ing in determining sources of

infect ion in the final s tages o f an eradication programme .

Ke l l ar � al . ( 19 7 6 ) in a comprehensive s tudy o f a series o f

reinfected herds in Canada us ing "case control techniques" , were

ab le to wei gh the importance of the various factors involved in

caus ing herd breakdowns . A.mos t interesting finding in this s tudy

was that s train 19 vaccinat ion�� did not appear to adverse ly

inf luence the interpretation o f the sero logical tests used nor did i t appear to protect the individual animal .

The Immune Response o f the Bovine to Bruce l la abortus

Infect ion

( a ) Ce l lu lar response

S tudies on the ro le of ce l l med iated immuni ty in the bovine

response to bruGe l losis have only recent ly been undertaken with any vigour . There has been some interes t in de layed type

hypersens it ivity ( DTH ) skin tests for d iagnosing Bruce l la

me l i tens is infection in sheep and goats in Eas tern Europe and

1 5 .

China (Didovet s , l 9 65 ) and widespread tes t ing has been carried out .

Current stud ies are mainly concerned with the deve lopment o f DTH

skin tests (Jones , l9 76 ) and lyrnphocyte transformation and inhibi t ion

tes t s ( Swiderska ,!;! &. , 19 7 1 ; Renoux ,!;! &·, 19 7 6 ; Kaneene e t al . ,

1978 ) .

Experiments by Jones and Berman ( 19 75 ) have shown that

circulating ant ibody is made in re sponse to the po lysaccharide

determinants of the ce l l wal l lipopo lysaccharide ( LPS ) in animals

infected wi th smooth Bruce l la , and not to the protein components .

Apparently , the fact that Bruce l l a i s an intrace l lular

paras i te makes i t unneces sary to pos sess a comp lete or typ ical

( compared with the enteric baci l l i ) LPS to protect agains t

phagocytos i s and comp lement des truct ion . Bruce l lae are res i s t ant

to intrace l lu lar des truct ion by inhib i t ing or b locking the

degranulation process normal ly observed fo l lowing ingestion . I t

has been shown that " l ipid A" i s the mitogenic principle o f LPS ;

thus i t appears that Bruce l la has a unique " l ip id A" component

(Hase and Th . Rie tsche l , l976 ) .

According to the National Research Council Subcommittee on

bruce l losis research (Anon . , 1977b ) : " S tudies are non-exis tent on the

ro le o f the comp lement components in sero logical and immuno-

logical phenomena in bruce l losis; membrane receptor s tud ies and

surface fixation of the ant ibody substructures onto ce l l s ;

studies on the components in , and mode o f action of , co los tra l

substances ; re investigat ion o f the bacteriocida l act ion o f b lood

and i t s components in the newer era of immunoglobul in and

complement chemis try ; genetics o f the his tocompatability sys tem

and susceptibility to Bruce lla infection . These top ics have the

capaci ty to change radically viewpoints and technology in applied

immunisation to brucellosis".

( b ) Humoral re sponse

There are five bovine immunoglobul ins present ly recognised :

IgG1 , I gG2 , IgM , IgA and IgE (Hammer � � . , 19 7 1 ; Duncan � �. ,

197 2 ) . I gG1 and IgG2 are quantat ive ly and functionally important

in serum and certain s ecret ions .and are not re s tricted to the

b lood vascu lar space (Duncan � �. , 1 9 7 2 ) . I gGl is select ive ly

concentrated in bovine colos trum (Mach and Pahud , l97 1 ) .

IgM is a large mo lecule and because o f its s i ze tends to be

more res tricted to intravascular spaces than IgG (Husband � �. ,

197 2 ) .

In most animals , the ant ibody response to an antigenic

s t imulus cons ists of two phases - an ini t ial IgM fo l lowed by IgG .

The mechanism o f the convers ion o f IgM to I gG i s not known but i t

can b e b locked, e . g . by 6 -mercaptopurine ( Sahiar and Schwatz , l965 ) .

There is also evidence that IgG inhibits 19S ( IgM) synthes i s

by neutra l i sing the antigen , thus leading to an IgG s teady s tate

(F ink and LoSpal luto , l967 ; H�l l iday, l 968 ) .

The pat tern o f immunglobulin product ion in Bruce l la

vaccinated or infected cat t le has rece ived much attent ion s ince

the ear ly 1960 ' s when the nature and s tructure of immunoglobul ins

was e luc idated .

The first ant ibod ies detected in sera of cat t le exposed to

s train 19 or a virul ent s train of Br . abortus are heat lab i le

macroglobul ins ( IgM) fol lowed shortly after by the heat s tab le

IgG . (Amerault � � . , 196 2 ; Rose and Ameraul t , 1 964 ; Rose and

Roepke , l9 64 ; Rice � � . , 1966 , 1967 ; Wi lkinson , 1966 ) . In the

case o f vaccinated calves both types o f ant ibody usual ly

disappear wi thin a few months a l though agglut inins thought to be

due to persi stent IgM may o ften be demons trated in a proport ion

of such animals . In f ield cases o f bovine bruce l los i s the ant i -'

bodies have been reported to be predominant ly or ent ire ly of the

1 6 .

7 S ( I gG ) c lass (Rice� �. , 1966 ; Rice and Boyes , l97 1 ; Beh, l9 74 ) .

Fo l lowing s train 19 vaccinat ion of ca lves , IgM ant ibody

appears at about five days whi l s t IgG appears s imul taneous ly or

a few days later . Whereas peak level s of IgM are reached at

about 1 3 days , IgG leve ls peak at 28-42 days (Rice � �. , 1 9 66 ;

Beh , l974 ) . Because o f the o ften long and variab l e incubat ion

period o f the disease , IgM and IgG ant ibodies . may not appear for

some t ime after infec t ion. IgM antibody usually appears first

and this is fo llowed a few days later by I gG. IgM leve ls then

decline whi l s t IgG remains to be the predominant and o ften the

only immunoglobu l in pres ent (Morgan , l967 ) .

1 7 .

18 .

Vaccination Against Bruce l l a abortus in Catt le

Nearly every nation that has a ttempted to contro l bruce l losis

in catt le has emp loyed the use o f vaccination at some s tage o f the

programme . These e f forts to provide immunity have u t i l ised a wide

rangG of o�gont•m�l ttving and dead& virulent and attenuated; re lated and unre lated .

S train 19

( a ) Deve lopment o f s train 19

As ear ly as 1919 the U . S . Bureau o f Animal Indus try permi tted

commercial manufacture and distribut ion of l ive Br . abortus

vaccines . These virulent vaccines were used in the be l i e f that

if non-pregnant cows were inoculated they would not abort in

subsequent pregnancies .

Fol lowing the work of Cotton ( 19 3 2 ) , who demons trated that

organisms could be estab l i shed in the udder of cows vaccinated

whi l s t lactat ing , the production of vaccines from virulent

organisms was prohibited (Mohler � �., 1941 ) .

Buck (1930) administered the live attenuated strain 19 Br . abortus to three ca lves five to e i ght months o f age and

challenged them with virulent organi sms at the fourth month of

pregnancy . Al l three he i fers calved normal ly whereas contro ls

became infected and aborted . The conclusion drawn from thi s

experiment was that immunity could b e induced by early vaccination

and a resultant reduct ion in the leve l of agglut inins produced

added further benefi t . In fo l low-up experiments Cotton � �·

( 1933) and B uck � al . ( l934) concluded that a s train o f low

virulence ( s train 19 ) conferred as much or more immunity as d id

more virulent s trains . In subsequent experiments Buck � !!·

( 1938 ) Haring and Traum ( 1937, 1943 ) and Haring ( 1938 ) veri fied

ear l ier resul t s and concluded that the immunity persis ted unt i l

the animals matured .

In 1936 the u.s. Bureau of Animal Industry carried out large

scale f i e ld trials and the resu l t s were so encouraging that ca l f­

hood vaccinat ion was adopted as o ff icial pol icy (Mohler � !!.,

1941). S ince then s tudies by Gi lman and Wagner (1959), King and

Frank (1961) and Lambert � !l.(l96l) have added further evidence

' '

that calfhood vaccination is indeed as effective as adult

vaccination . Redman � �. ( 1967 ) concluded that the protect ion

afforded by vaccinating at two to three months was comparable to

that found in cattle vaccinated at four to eight months of age .

There has recently been renewed interest in strain 19

vaccinati<:m of adults . In some "problem" herds, where the eradicat ion o f the d isease is proving d i fficult , previously

unvaccinated adults have been vaccinated with the obj ect of

containing the spread o f infection . I t i s claimed that by

applying a ser ies of serological tests vaccinat ion t itres can be

d i fferentiated from those due to infect ion (Worthington � � . ,

1 9 7 3 ; Nicoletti , l97 7 ) .

(b ) Pathogenicity

Strain 19 d i ffers from more virulent strains o f �· abortus

in that even in large doses it produces in guinea p igs no more

than a slight enlargement of the s pleen and a blood serum t itre

seldom over 1 : 50 . Organi sms are rarely found in the spleen s ix

weeks after infect ion (Alton � �. , 1975a ) .

19.

Meyer and Nelson ( 19 6 9 ) reported that re s idual localization

o f strain 19 occurred following immunisat ion of female cattle but

was infrequent . The re s idual infect ion involved the mammary glands

but did not spread to or involve the reproduct ive tract . N icoletti

and Muraschi ( 1966 ) and Nicoletti ( 1969 ) have reported on the

i s olation o f strain 19 organisms from animals in problem herds .

Because no more than one s train 19 infected animal was ever found

in any one herd it was concluded that there was no evidence to

s uggest that this biotype is contagious .

Lambert � �. ( 1964) reported the development o f orchitis

and the i solation of strain 19 from the tes ticles o f two bulls

vaccinated at five to s ix months . More recently Lambert � � ·

( 1965 ) administered strain 19 to 15 bulls four to ten months o f

age . Mild transient post-vaccinal orchit i s developed i n e ight o f

these animals . Post-vaccinal serum agglutination ti tres pers i sted

at diagnostic levels longer than in hei fers . When slaughtered at

18 months of age s train 19 was unable to be isolated .

( c) E f fectivenes s

The resis tance conferred by: s train 19 vaccination has been

I

20.

the subj ect o f a number o f s tudies , the most extens ive being a

review o f U.S . Nat ional Animal D isease Laboratory data by Manthei

( 1959 ). This data indicated that 65- 75% of vaccinated animals

were completely protected whilst of the remaining 25-35%, although

infected , most did not abort.

Safford (l9S9) reviewed the effeGtiveness of strain 19

vaccinat ion in Montana and S tuart � �. ( 1959) have reviewed the

Cali fornian experience . Jones and Berman ( 1976) have reviewed the

overall U . S . experience wi th s train 19 particularly emphas i s ing the

use o f the vaccine in Wiscons in .

McDiarmid ( 1957 ) made an extensive s tudy of the durat ion o f

res i s tance conferred by calfhood vaccinat ion with s train 19 . From

a seven year experiment involving 500 cattle he concluded that one

dose o f s train 19 at six months o f age conferred adequate immunity

for five pregnancies and probably for the complete normal milking

life o f the animal . Other s tud ies on the durat ion o f immunity

conferred have been reported by Goode � �. ( 1956 ) and Manthei

� �. ( 195 1) .

Although there i s li ttle doubt about the effectiveness o f

s train 1 9 vaccinat ion in protect ing individual animals and

reducing the overall prevalence o f d isease , strain 1 9 vaccination

will not keep brucellris is out o f a clean herd or prevent i t from

spreading after i t has entered a herd (Nadler, 1978) . This view

has also been s tated by King ( 19 7 1) who found that herd

vaccinat ion s tatus did not s i gni ficantly reduce herd infection

rate . S tud ies in Ontario have also failed to detect s i gnificant

differences in vaccination levels between infected and non­

infected herds (Kellar � � . , 1976) .

( d ) Route o f administration

The use o f intracutaneous or intradermal admini s tration o f

�· abortus vaccines was f i r s t reported b y Cotton ( 19 3 2 ) and by

Cotton � �. ( 1933) . They found that s train 19 organisms became

localised in the udder when vaccine was given intradermally but

not when administered subcutaneously .

S everal inves t i gators have found that good agglu t inat ion

t i tres have been produced by intracutaneous adminis tration o f

smaller doses than are required wit� subcutaneous inoculation

(Campbell and Rodwell,l 945; McDiarmid,l 948, 1950; Cotton,l 953 ) .

2 1 .

Cotton ( 19 5 3 ) found that agglutinat ion t i tres s t imulated by intia­

cutaneous admini stration of s tr�ln 19 reached peaks as high as

those obtained by subcutaneous inoculation but fe l l to low leve ls

s ignificantly sooner .

In a recent report P lommet and Fens terbank ( 19 7 6 ) , indi cated

t h a t intraconj unc t iva l admin i s t r a t ion o f s t rain 1 9 vacc ine wa s

at least as e ffective as subcutaneous vaccination and that the

vaccine cou ld be . adminis tered at any age without risk of producing

a serological re sponse that would interfere with d iagnos i s .

( e ) Effect on diagnosis

The d iagnost ic prob lems created by s train 19 vaccinat ion are

we l l known . In both the ca l f and adu l t agglut inins appear within

a few days and reach a peak at two to three weeks , when t i tres may

be s imi lar to those caused by natural infect ion , i . e . about 2000

i . u . /ml . In animals vaccinated subcutaneously with the usual

dose, titres soon decline and most are negAtive six mont h s a f ter

vaccination . In calves vaccinated a t s ix t o nine months about

80% can be expected to be negat ive 12 months later and 90% two

years later ( S tableforth , l959 ) .

The various modificat ions to the s erum agglutinat ion test

( SAT ) as detailed in a later sect ion were large ly deve loped wi th the

obj ect of f inding a tes t that would not be sens i t ive to

agglut inins produced by vaccinat ion , but would detect those

promoted by infection . Al though the comp lement f ixat ion test

(CFT ) had been used in the early 1900 ' s and was known to be a

part icularly e f f icient tes t , i t was not unt i l its re lat ive

insens it ivity to vaccinat ion and o ther non specific t i tres ·was

rea l i sed that i t found ready acceptance (Wisniowski , l95 7 ;

Thomsen, l95 7 ) . Even though the CFT and various other supp le-

mental tests may help in d i s t inguishing vaccinat ion t i tres from

infection t i tre s , it is c laimed that as yet there is no val id

data indicating that any s ingle test i s capable o f making this

d i f ferent iation (Anon . , 1 9 7 7b ) .

Schi l f ( 19 6 8 ) reported on the exis tence of a "masking" effect

o f s train 1 9 vaccination. Vaccinated cattle were claimed to take

longer to deve lop reactor t i tres after exposure to infect ion than d id

non-vaccinated animals . Data from US e pidemiologists presented

by S chi l f ( 1968) showed that �· abortus was cul tured from 462 cows

I

2 2.

vaccinated as calves ; o f these 194 ( 42%) were not classed as

reactors in the standard test . In a group of vaccinated and non-

vaccinated cattle exposed to equal numbers of virulent organisms ,

75% of non-vaccinated animals reacted to the agglut ination tes t

30 days post-exposure while in the vaccinated group i t took 90 days

for 75% of the animals to react .

Renoux � �. ( 19 7 1) exper imentally infected a group of strain

19 vaccinated and unvaccinated he i fers at be tween five and s ix

and a half months of ges tation . Ant ibody responses as measured

by the SAT and �FT appeared later in the vaccinated group than in

the unvaccinated controls . This delayed development o f t i tre in

vaccinated animals was evident in groups of animals that eventually

abo rted as well as in those whose pregnancy did not terminate in

abortion . Fensterbank ( 1 9 7 3 ) found that in an experimentally

infected group o f vaccinated and unvaccinated he i fers , the

vaccinated animals developed card and complement fixat ion test

t i tres well before the unvaccinated group . Unfortunately the

relat ive appearance of agglut ination test t i tres was not detaile d .

Anderson � �. ( 196 2) noted that vaccinat ion with s train 1 9

vaccine appears to increase the d ifficulty o f de tecting infected

animals with the agglut inat ion tes t . They sugges t that detect ion

o f exposure to Brucella would be s impli fied if vaccination were

not practi sed in areas of very low prevalence where the ri sk o f

exposure i s negligible .

S train 45/ 20 Killed Adj uvant Vaccine

McEwan and Samuel ( 1955 ) used an avirulent s train (45 ) , and

a s train der ived from i t by guinea p ig passage ( 20 ) , heat killed

in an oily base . The "rough" phase induced a level o f immunity

comparable to that of the " smooth" phase and raised the poss ibility

of developing from the "rough" phase a vaccine which would not

produce agglutinins to s tandard antigen s trains and so interfere

with serological tests . This s train ( 45/20) could not be used

as a live vaccine as earlier wor k ( Edwards ��. , 1945 ; Taylor and

McDiarmi d,l 949 ) had shown t hat it could re vert to a fully virulent

smo ot h form .

The ma jor theoretical a dvanta ge o f strain 45 / 20 vaccine is

2 3 .

that agglutinat ion t i tres fo llowing vaccination are transient .

However , Cunningham ( 1968 ) and Cunningham and O ' Re i l ly ( 1968)

found low leve l persis tent agglut inin and complement fixation

test responses . Corbel and Bracewe l l ( 19 76 ) and Hal l et �. ( 19 7 6 )

have also noted such responses part icularly fo l lowing a second

d o s e of 45 / 20 . Beh ( 19 7 5 ) and Corbe l ( 19 7 6 ) have ehara e t e r i s e d

the ant ibodies produced by such a response .

According to Corbe l ( 1976 ) animals which have not been

previous ly exposed to smooth Bruce l la ant igens initial ly produce

ant ibodies o f the IgM c las s against rough B r . abortus when

inoculated with s train 45 / 20 vaccine . As the immune response

proceeds , I gGt and IgG2 ant ibodies appear and become predominant .

In animah previously exposed to smooth !£• abortus through

virulent infect ion or s tra in 19 vaccination , ant ibodies to both

smooth and rough ant igens are rap idly produced and are o f both

I gM and IgG classes .

A s igni ficant di sadvantage of s train 45 / 20 vaccine i s the

deve lopment o f abcesses at the s i te of inj ect ion . Such abcesses

caused by the o i ly adj uvant used in the vaccine may las t for more

than 12 months and cause s erious carcase damage (Cunningham , l966 ;

Ha l l � �. , 1 9 76 ) .

Whereas with s train 1 9 vaccinat ion one dose has proven to

g ive adequate protection , i t has been shown that two doses 10- 1 2

weeks apart are required for optimal protect ion with 45/ 20 . Even

with two doses the protect ion afforded is no better than and may

be inferior to that given by s train 19 (McDiarmid , l97 2 ;

Worthington and Horwe l l , l 9 74 ; Ray and Hendricks , l974 ) .

Ray ( 19 7 6 ) has reviewed inves tigat ions conducted in the USA

into the e ffectivenes s o f 45/ 20 and concludes that 45 / 20 does not

confer adequate res i s tance on catt le in infected herds subj ected

to mult iple exposures of virulent organisms .

S t rain 5 3H38

This vaccine is made from a s train of B r . me l itens i s , heat

ki l led and used with an adjuvant . Renoux and Valette ( 1967a ,

1967b , 1967c ) studied the e f fectiveness of 5 3H38 vaccine and

found it to be at least as e fficient as l ive strain 19 vaccine . \

Agglutination titres produced in response to vacc ination were

24.

found to d isappear more rapid ly than s train 19 induced t i tres .

Pos s ib ly because o f prob lems with local reactions a t the point o f

inj ect ion 5 3H38 vaccine has not found widespread acceptance

(Dhennin , l97 3 ) .

vaeeine "P . n . "

Pi le t-Bonneau vaccine cons i s ts o f formalin inact ivated

s train 19 organisms which have b een saturated wi th specific immune

sera to reduce agglutinin format ion . I t was deve loped by P i le t

and Bonneau (Bonneau � �. , 1 9 7 0 ) but has not been wide ly used

(Dhennin, 1 9 73 )

Rev . I .

Rev . I vaccine i s a l ive a t tenuated s train of Br . me l i tens i s

and is wide ly used for the control of �· me li tensis infection in

she.ep and goats . I t wi l l pro tect agains t bovine bruce l los i s but

is not recommended for use in countries where B t . me l i tens i s does

not exi s t (Anon . , 19 7 1 ) .

Solub l e Antigens as Potent ial Vaccines

Although very l i ttle work has been done in cat t le , several

laboratories have examined solub le preparat ions as potent ial

immunogens agains t challenge with viru lent s trains o f Bruce l la in

laboratory animals . Fos ter and R i b i ( 19 6 2 ) showed that a subs tance extracted from ce l l wal ls by aqueous ether had immunis ing potency

superior to kil led whole ce l l s or ce l l wal ls . Markenson ll � .

( 19 6 2 ) examined �he immunis ing potential in mice o f sediment from

a sonic extract . Rasooly ll a l . ( l967 , 1968) have shown that

Bruce l l a ce ll wal l s can protect laboratory animals against

infection with Br . me li tensis and Br . abortus . - -

Guinea p igs

inj ected with ribosomes derived from s train 19 organisms , and

mixed with an adj uvant , were shown by Corbe l ( 1975b ) to induce

immunity to chal l enge by virulent organisms .

Future deve lopments

There is strong evidence that protection against B ruce l la

infection i s dependent on ce l l -mediated immunity and not on the

presence of humoral ant ibody (Al ton, l977a) . Par i sh ( 19 7 2 ) has

suggested that the presence of antibody cou ld interfere with the

deve lopment of ce l l -medi ated immunity and that chemical mod i fic­

a tion of the specific ant i gen could leave intact the part that

s t imulat d c e l l �med iated immunity whil e ina c t iva t ing the por t i on

respons ible for s t imulat ing ant ibody producing lymphocytes .

25 .

According to the National Research Counci l S ubcommittee on

Bruce l losis (Anon . , 1 9 7 7b ) ; ' ' I t i s unl ike ly that anything wi l l

rep lace s train 19 al though data on laboratory animals indicates

ce l l fractions (e . g . aqueous ether or sodium dodecyl su lphate

extracts of ce l l wal l s ) p lus proper adj uvants may rep lace heat

ki l led who le ce l l vaccines which cause extens ive t i s sue necrosis

and abscess formation a t the s i te o f inj ection" . The subcommi ttee

s uggest that improvements in d iagnostic tests are more urgently

required than the deve lopment o f o ther immunising agents .

Deve lopment o f Diagnostic Tests for Bovine Bruce l losis

Tests used to diagnose bruce l los i s may be convenient ly

d ivided into three groups .

( i ) Tes ts to demons trate the presence of specific immuno­

globulins in the b lood , mi lk , vaginal mucous or semen .

( i i ) Tes ts to demonstrate the presence of Brucel la abortus

( i i i ) Tests to demonstrate a speci f i c al lergic react ion to

bruce l losis .

2 6 .

A s mentioned i n a previous section the maj or specific immuno­

globul ins act ive in these te sts are be l ieved to be IgG1, IgG2 , and IgM .

There i s a comp l i cation in that the ant igenic s t imulus may have

come from:

( i ) Field infect ion with B r . abortus .

( i i ) Vaccinat ion wi th l ive s train 19 vaccine .

( i i i ) Vaccinat ion wi th ki l led 45 / 20 vaccine .

( iv) Other antigens with common de terminates , e . g . Yersinia

enteroco l i tica .

Incontrovert ible evidence of Bruce l l a infect ion i s obtained

by isolation and ident i f icat ion of the organism . S ince i t i s not

always pos s ible to i solate the causal organism from infected

pat ient s , sero logical tests play a maj or role in the rout ine

diagnosis of bruce l losis (Alton � �. , . 1975a) .

The first recorded d iagnostic test for bruce l losis in animal s

was an agglutination tes t . A Mal tese phys ician tested the sera

o f s ix goats prior to his at tempt at experimental ly infect ing

them with the recent ly i so lated Micrococcus me l i tens is . The

b lood of five of the s ix goats showed a s trong reaction and thus

the first l ink between Malta fever and goat mi lk was e s tab l ished f

( S p ink , l956 ) . An agglutination test had previous ly been app l i ed

to the d ifferentiation o f typoid and Mal ta fever in man by

Wri ght ( 1897 ) .

Full deta i l s o f the more commonly accepted test methods

currently used are given by Alton � a l . ( l975a ) .

Agglut inat ion Tests

Tradi t ional ly the serum agglutina t ion tes t has been the main

27 .

test used for the d iagnosis o f bruce l losis in man and �nimals . I t

i s relat ively easy to perform and has been wel l standardi sed . I t

was recognised by many o f the ear ly workers that the t i tre obtained

with a given serum is influenced by the tes t methods used . A

dried reference serum was prepared in Britain in 1933 ( S tab leforth ,

1 9 36 ) and this was adopted by the Office Internat ional des

Epizooties in 1937 fo l lowing recognit ion of the divergence of

methods used in d ifferent countries ( S tab leforth, l959 ) . In 195 2

a new batch o f dried serum equivalent in t i tre to the original

s tandard was estab l i shed by the WHO Expert Commi ttee ' on B io logical

S tandardi sat ion as the International Standard for Ant i -Bruce l la

abortus Serum ( ISAbS ) . I t was decided that the Internat ional uni t

should be one thousandth o f the dried material from one ml o f the

original serum (Anon . , 1 954) .

Because s tocks o f the first ISAbS became dep leted a rep lace­

ment s tandard was prepared and the second ISAbS was estab l i shed

in 1968 . The Internat ional Unit was then redefined as the

activi ty contained in 0 . 0955 2 mg of the second ISAbS i . e . one

thousandth o f the mater ial from one ml . The ant ibodies contained

in thi s serum are almo s t exclus ive ly IgG (Anon . , 19 7 1 ) .

Tradi tional ly the s tandard agglutination test i s carried out

in tes t tubes wi th the antigen suspended in phenol - sa l ine ( 0 . 5%

phenol 0 . 85% saline ) . Because o f the occurrence of the prozone

phenomena some workers prefer to minimise thi a e ffe�t by us ing 5%

sal ine as the di luent ( Diaz and Levieux , l97 2 ; Fenske , l97 7 ) .

Because i t is a comparat ive ly easy test to perform the

agglut inat ion test has been readi ly accepted . Desp i te i t s wide­

spread use in Europe and America i t has in recent years been

subj ected to more cr i ti cal s tudy and has been found to be less

efficient than o ther tes ts . Nico letti and Muraschi ( 1966)

reported that i t fai led to clas s i fy 39% of 1 35 cul ture - po s i t ive

catt le as reactors . In a later report �ico letti ( 1969) c la imed

that i t fai led to correctly identify 48% o f a group o f 1 65

cul ture pos i t ive catt l e . Alton � �. ( l975b) found that 1 1% o f

culture - pos i t ive cat t le had l e s s than lOO i . u . o f ant ibody per

ml of serum and 4% had less than 30 i . u . Fenske ( 19 7 7 ) in a

large s tudy claimed that in acute ly infected cattle the

agglutinat ion tes t detected only 50% o f the infected animals .

I t i s also acknowledged that the agglutination test becomes

28 .

pos itive later than does the complement fixation test and that i t s tays fa lse ly positive longer in s train 1 9 vacc inated animal s than

o ther diagnos t i c tes t s (Morgan and Richards , l974) .

Corbe l ( 1 9 7 2 ) , Beh ( 19 7 3 ) , Al lan � �. ( 19 7 6 ) and Patterson

� �. ( 19 7 6 ) have investigated the re lative importance of the

a pecifie immunog lobu l in e L 4 s s � s in the agglutination , complement

f ixation and card (Rose Bengal ) tes t s . I t is general ly agreed

that a l l the maj or immunoglobul in c lasses p lay a ro le in the

agg lutination tes t . IgG1 , however , is said to agglutinate more

e f ficiently with buffered antigen or antigen in 5% sodium

chloride (Diaz and Levieux , l97 2 ) a l though Pat terson � � . ( 1 9 7 6 )

found this t o be s o only wi th sera from strain 1 9 vacc inated catt le .

Variat ions o f the Agglut ination Test

Because the standard agglutinat ion test requires serum

d i lutions to be made in tubes , and incubated overnight , a number

of yariat ions in methods used have emerged .

The Rap id or P late agglutination test has been used in the

USA for a number of years as one of the o fficial tests ( Hudd leson ,

1 943 ) . I t can be s tandardi sed with the international s tandard

serum and i t s interpretat ion can be d irect ly re lated to the tube

test (Anon . , 1 9 65b ) .

In an attempt to differentiate be tween the specific Bruce l la

agglutinins and non-specific agg lutin ins Rose and Roepke ( 1 95 7 )

introduced a modi fication to the p late test whereby the ant igen

was buffered_ a t pH 4 . 0 immediately before use . This test i s

termed the acidi fied p late test . Mi l l er ( 1 9 7 1 ) has described an

automated vers ion of this test which provides a semi- quantat ive

result .

Numerous o ther modifications inc l uding varying incubat ion

t imes and temperatures , prior serum inactivat ion , centri fugat ion

to hasten depo s ition of agglutinins , and -the addit ion of various

sa l ts have been described . O l i tzki ( 19 70 ) has extensive ly

reviewed the l i terature regarding these procedures .

The Bruce l lo s i s Card Tes t (BCT ) or Rose Bengal Plate Test ( RBPT )

Fo l lowing the observat ions_ o f Rose and Roepke ( 1957 ) that an

29 .

acidif ied antigen would destroy the act ivity of non- specif ic

agglut inins a card test us ing a stained antigen buffered at pH 3 . 6

was developed (Nicolett i , l967 ) . The card has been used as an

o ff ic ia l test in the US bruce l losis eradication programme s ince

1966 .

I n the UK a s imi lar deve lopment took p lace at Weybr idge whieh

re sul ted in the Rose Bengal P l ate tes t , this was introduced as an

off icial test in 1970 (Morgan and Richards , l974 ) .

The two tes ts are essentia l ly the same , both uti l i s ing the

mixing of equal vo lumes (usua l ly 0 . 03 ml ) of serum and antigen on

a whi te card or t i le and then rocking gent ly for four minutes .

Any degree o f agglutination i s regarded as pos itive , there i s no

doub t fu l clas s i f ication .

Whereas the British RBPT antigen i s made from Br . abortus

s train 99 , the American BCT antigen is derived from s train 1 1 1 9 - 3

ce l l s . Otherwise preparat ion o f the antigens are s imi lar . The

RBPT i s conducted on a white t i le or in haemagglut ination tray

we l l s whi l s t the BCT is carried out on a white card ( Brewer Card

Test Kit , Hynson , Westcott and Dunning , Inc . , Ba l t imore ,

Maryland) . Antigen for the ki t is prepared by the Nat ional Animal

Diseases Laboratory , Ames .

The card test is usually used as a screening tes t with

animal s showing a pos it ive reaction be ing subj ect to further

tes ting .

The Coombs ( anti -bovine globu l in) Test

Another test based on an extens ion o f the agglut ination test

i s the Coombs te s t . Thi s is one o f the most sens i t ive sero logical

tests for the detection o f antibody , part icularly for incomp lete

or b locking antibodies (Morgan , l967 ) .

An anti -bovine globul in preparation i s added to a washed non­

agglutinat ing ant igen-antibody comp lex . so that the anti-globul in

wi l l combine with and agglut inate the previously formed comp lex .

Detai l s o f the test procedure are given by Morgan ( 19 67 ) . The

maj or disadvantage o f the test is the need for carefu l repeated

washing of the ce l l s a fter the first react ion .

There has been some confus ion over the use o f the terms

30 .

"b locking" and "incomplete" antibodies . Jones � &. ( 1957 ) advised

l imiting the term "blocking" ant ibodies to those revealed by the

"blocking" test and the term " incomplete" to those revealed . by the

Coombs tes t . A serum shown to have blocking antibodies by the

b locking tes t wi l l usual ly have higher t i tres of incomp lete anti · �ad t e s a s �evaal�a a y ehe C aom�s te s t ( Janes and Wi i san, l�5 l a M � � l and Manion , l 953 ) , the Coombs test i s therefore preferred .

Cunningham ( 1967 ) demons trated the pre sence of ant i -bovine

globulin t i tres in several sera that were negat ive to the

agglutination and complement fixation tests and conc luded that

the Coombs test detect s a specif ic incomp lete ant ibody for

B ruce l l a . Hadj u ( 19 6 3 ) reported o n the usefulness o f the test in

detecting chronic infect ions in Czechos lovakia and Fenske ( 1977 )

has reported on its value but warns that i t is often over ­

sens i tive . The particu lar prob lem o f i t s over- sensi t ivity in

s train 19 vaccinated animals has been invest igated by Beh and

Lasce l les ( 19 73 ) who found that i t was the IgG1 incomp lete anti ­

body that persisted . They suggested that the u se of an ant i- IgG2 immunoglobu l in in the Coombs test would be he lpful in

differentiat ing vaccination from infection titre s .

Disulphide Bond Reduction Tests

IgM , which is a pentametre , i s broken down by reduct ion o f

the disulphide bonds by certain compounds such a s mercaptoethano l ,

cys te " ne and di thiothreito l . After reduct ion the mo lecule loses

its antibody act ivi ty . These tes t s are used as presumpt ive

evidence for the presence of IgG by comparing agglut ination

t i tres before and after treatment . The mercaptoethano l vers ion

origina l ly described by Anderson � &· ( 1964) is used as one o f

the s o cal l ed "supp lemental" tes t s i n the US . I ts e ffectivenes s '

has been reported on by Nico let t i and Muraschi ( 19 66 ) and

�ico lett i ( 1969 ) . The dithiothrei to l test is used in the UK

(Morgan � a l. l 9 7 1 ) .

Rivanol Tes t

Rivano l (2 -ethoxy-6 , 9-diaminoacridine lactate ) precipi tates

all serum proteins except gamma g lobul ins ( Frommhagen and Mart ins ,

196 3 ; Hudd leson , l965 ) and has been used a� the bas i s o f a

3 1 .

d iagnos t ic test for bruce l lo s i s (Nico letti and Muras chi , l9 66 ) . The

principle of interpretation i s the same as that app l ied for the

d isulphide bond reduction tests .

The Comp lement Fixation Tes t (CFT )

According to O l itzki ( 19 70 ) the complement fixat ion test (CFT )

was u sed along with the agglutination tes t in the ear liest

inve s t igat ions on the sero logica l diagnosis of bovine bruce l losis

in Europe in the ear ly 1 900 ' s . There have been many reports in

the l i terature on the superiority o f the CFT over the agglut ination

tes t , but despite these the test found l i ttle use , except as a

confirmatory test in some European countries , unt i l relative ly

recent ly (O l i t zki 1970 ) .

A number o f s tudies in recent years have indicated that the

CFT i s more specific than the agglutination tes t and is espec ially

use ful in cattle that have been vaccinated wi th s t ra i n 1 9 vaccine

( Burki , l963 ; MacKinnon , l963 ; Nico letti and Muraschi , l966 ;

N ico l etti , l969 ; Morgan and Richards , l974 ; Alton � �. , 1975b ) .

A maj or prob lem in the deve lopment o f the CFT has been the

lack of a s tandardi sed method be tween l aboratories . Thi s has

l ed to a wide variety of t i tres being given for tes t s on standard

sera (Morgan et al . , 1 9 7 3 ) . - -

Vaginal Mucus Agglutination Test

This test was introduced by Kerr ( 1955 ) and has been further

s tudi ed by Chris tie � �. ( 1968 ) . I t is said to be part icu lar ly

· use fu l in testing cows fo l lowing suspect calvings when sero logica l

t i tres may not have deve loped . The presence o f ant ibodies is not

s i gni ficantly influenced by s train 19 vaccination .

The Mi lk Ring Test

The mi lk r ing test (MRT ) was deve loped by F le i schhaeur ( 1937 )

for detecting the presence o f Bruce l la antibodies in mi lk us ing

a s tained antigen . I t s mos t usefu l app lication is in the testing

3 2 .

of bu lked mi lk samples from farm vats to indicate the presence or

absence o f infect ion in the herd (Roepke and S t i les, l 9 70 ) .

Pat terson and Deyoe ( 19 7 7 ) provide an up to date discuss ion on the

mechanisms o f the test whi le O l i t zki ( 19 70 ) has extensive ly

reviewed i t s deve lopment and the manufacture of the s tained

antigen .

Microscopic Examination

S taining methods usually used for the direct examination o f

foetal material , p lacenta , discharges and f luids from j o ints e tc . ,

are the Kos ters and the modi fied Ziehl Nee lson methods ( S tamp

.!:.!_ al , , 1950 ) . Leech .!:.!_ 2.1 ·( 1964) recorded the re sults o f the

examinat ion of 569 membranes . Pos i t ive smears were found in only

1 . 2/. of p lacentae which were declared non- infected using cul ture

or animal inocu lat ion test results as the criterion of infection .

Lapraik .!:.!_ i!l-(1967) drew at tention to the difficu lty in di fferentiat ing Br . abortus from Coxie l la burnett i by this method .

In New Zealand where Coxiel la burnet t i is not known to exi s t this

di fficu l ty does not ar ise .

Corbe l ( 1973a) has asses sed the direct fluorescent antibody

test and obtained cons is tent resu l t s . This method was ab le to

d i fferentiate C . burnetti from Br . abortus .

Cu ltural Tests

Various workers have compared d i fferent med ia for the cul ture

/ of B r . abortus , (Morga� , l960 ; Painter � & . , 1 9 66 ; Farre l l and

Robertson, 1 9 7 2 ) . Al ton � & .( 1975� detail the suggested t i ssue

and types of cul ture media recognised for optimal isolat ion of

Br . abortus . Serum dextrose agar p l ates are commonly used with

a variety of selective inhibitors - e . g . bacitracin , po lymixin B

and cyc lohexamide . P lates are incubated in 5 to 10% C02 in air

as most b iotypes wi l l grow only in this a tmosphere on initial

i solat ion . Robertson .!:.!_ & . ( 197 7 ) have discussed the deve lop­

ment of methods of isolation of Bruce l la from contaminated

sources and have reviewed the deve lopment o f modern se lect ive

media .

Corbel and Morgan ( 1975 ) and Morgan and Corbe l ( 19 75 ) have

proposed minimal s tandards for the descriptions o f species and

b iotypes o f Bruce l l a .

The need for b iotyping of cul tures a s an aid t o the

ep idemiologic knowledge o f the di sease has been expounded by

Ne lson � �. ( 1 9 6 6 ) and Luchs inger � �. ( 197 3 ) .

Nine biotypes o f Br . abortus are recognised and these are

identif ied on the basis of dye sens i t ivity tes ts , C0 2 requirement

and H 2S product ion . Al ton � �. ( 1975a ) g ive fu l l details of

b iotyping procedures . In New Zealand only three b iotypes have

been recognised from 1 140 cultures typed a t the Ruakura Animal

Hea l th Laboratory . These are :

. biotype 1 93 . 2"1 •

biotype 2 6 . 4"1.

biotype 4 0 . 1 8"1.

s train 19 0 . 1 8"1. (Anon . , 1 9 7 7 a )

Gu inea p ig inoculat ion is often used in conj unction wi th

p late culture when maximum sensitivi ty is required . Homogenised

tissue is normal ly inj ected into two guinea pigs which are

ki l led at three and six weeks post inocu lation . Spleens and

any o ther inflammed organs or lymph nodes are cul tured and a

serum agglutination te s t is carried out a t the time of necropsy .

Ful l de tai ls are given by Al ton � �. ( 1975a ) .

De layed Type Hypersensit ivity ( DTH ) Tests

A l though DTH t e s t s for bruce l l o s i h ave been widely used in

sheep , goats and man in Eas tern Europe (Didovets, 1965 ) , they have

not been recommended for use in cat t le .

Olitzki ( 1 9 7 0 ) has reviewed a l l ear l i er at tempts at DTH

tes t ing . Prob lems which have ari sen inc lude d i f ferentiat ion o f

infected from s train 1 9 vaccinated anima l s , false pos itive

reactions a t repeat tests and false pos i t ive react ions in cat t le

infected with tubercu los is .

I t is unl ikely that a DTH test wi l l rep lace serological

testing of cat t l e in the near future . However , renewed interes t

in this test may b e s timulated now that the allergen has b�en

characterised ( Jones and Berman, l975 , 1 9 7 6 ) .

3 3 .

'

34 .

O t her Tes t s

Numerous ot her s erologica l a nd milk t es t s ha ve been us ed f or

t he dia gnos is of brucellos is in ca t t le. Among t hes e a re t he

conglut ina t ion complement - a bs orpt ion ( complement f ixa t ion) t es t

( R ice, 195 2) , pa s s ive ha ema gglut ina t ion t es t ( F reema n � �. , 195 5 ),

indirect ha ema gglut ina t ion t es t ( Ca rrere a nd Roux , l95 2) ,

precipit in t es t ( R eit er, 193 6 ) , f loccula t ion t es t ( Hunt er a nd

Colbert ,l95 6 ) indirect f luores cent a nt ibody t es t ( Biegeleis en

� �. , 196 2), s urf a ce f ixat ion t es t ( Ca s t a neda , 195 0) , gel

dif f us ion t es t ( Bruce a nd J ones , 195 8) , 45 / 20 a na mnes t ic t es t

( Cunningha m a nd O ' Connor, 1971) , indirect ha emoly s is t es t ( Pla cket t

� �. , 1 97 6 ) , ra dioimmunoa s s a y (Chappel � �. , 197 6 ) , indirect

enzy me- labelled ant ibody t es t ( Sa unders and Clina rd , 197 6 ) , milk

pla t e t es t ( Blake � �. , 195 2), whey a gglut ina t ion t es t ( Ca meron

and Kendrick, 195 5 ) , whey complement f ixat ion t es t ( R obert s on a nd

F a rrell, 196 8) , and ly mphocy t e t ra ns f orma t ion t es t s ( Swiders ka

� �. , 197 1). O f t hes e t he indirect ha emoly s is , indirect enzyin e­

la belled a nt ibody t es t a nd ly mphocy t e t rans f orma t ion t es t s a ppea r

t o s how mos t promis e for fut ure us e.

The Evaluation of Diagnostic Tests

When a new test for a d i sease is be ing eva luated it is

cus tomary to perform the test in two se lected groups o f animals ;

one group known not to have the d i sease and the o ther group a l l

known t o be infected . Te s t r e s u l t s are u s u a l l y �iven a s po e i t 1ve

or negative or by some numerica l value which can later be

evaluated in terms of a particular degree of pos i t ivi ty or

negativi ty .

An a l ternative method , part i cu larly useful in pre l iminary

observat ions of the effect ivenes s o f a tes t , is to compare the

new test resu l ts with those o f another accepted re l iab le te s t .

A third method which has rece ived l i t t le at tent ion but which

can be app l i ed in certain areas o f veter inary medicine , i s the

trial use o f a particular te s t in an eradication programme . This

method i s at present be ing used in a large scale field eva luat ion

of the ind i r e c t haemo lys i s t e s t in V ictor ia , Australia, and wi ll

be the eventual means by which the automated comp lement f ixat ion

test is j udged in the New Zea land eradication scheme .

Pub l ications describing the mathematical concepts o f test

evaluation inc lude those of Thorner and Reme in ( 1961 ) and S chwabe

et a l . ( 19 7 7 ) . S ens i tivity and specificity are two important attr ibutes o f

a tes t . S ens i t ivity is the ab i l i ty of a test t o correc t ly detect

a diseased anima l . Spec ificity is the ab i l i ty of a test to

correctly de tect animals which are not di seased .

Sens i tivity =

Speci f ic i ty =

di seased animals wi th pos i tive test diseased animal s tested

non-d i seased animals negative to test non-d i seased animals tes ted

35 .

Martin ( 19 7 7 ) has d iscussed the eva luat ion of tests with

respect to s pecificity and sens i t ivity and graphically i l lustrated

errors inherent in us ing inadequate tes ts . Vecchio ( 19 66 ) showed

how the "predictive value" o f tes t s varies according to the

infection s tatus of the population under tes t .

Predictive value (PV+) ( of pos it ive tes t )

= di seased animals with pos itive tes t tota l animal s with pos itive te st

36 .

Predictive value (PV- ) ( o f negat ive tes t )

= non-diseased animals wi th negative test total animals with negative test

Predict ive va lu� o f posi tive and negative tests at varying

d i s ease prevalences when sensitivity and specificity each equa l

95% are given in the fol lowing table taken from Vecchio ( 1966 ) .

Actual disease prevalence PV+ PV-% % %

1 16 . 1 99 . 9

2 27 . 9 9 9 . 9

5 50 . 0 9 9 . 7

10 67 . 9 9 9 . 4

20 82 . 6 98 . 7

5 0 95 . 0 95 . 0

75 . .

98 . 3 8 3 . 7

lOO 100 . 0

The important concept i l lus trated by this examp le is that in

terms of an eradicat ion scheme where the disease prevalence is

d iminishing , a test that was adequate early in the programme may

not be suitab le later when di seased animals are uncommon .

Another aspect o f the evaluat ion o f tests i s that of the

practicab i l i ty and precis ion o f a given tes t . Even though a

tes t may be we l l sui ted according to its sens itivi ty and

specificity i t may be d i fficult to perform , subj ect to errors in

i t s performance and interpretat ion , or require expens ive reagents

or equipment . The United S tates experience with the comp lement

f ixation te s t for bruce l losis has been that it i s too di fficu l t

to standardise and that a combination o f s impler tests (Rivanol ,

mercaptoethanol , rap id p late) wi l l give an equivalent result for

l e s s e f fort (Nicoletti , 1 96 9 ; Pietz , l97 7 ) . Other workers have

obvious ly fe l t that the extra t ime and care required for comp lement

f ixation tes t ing is j u s t i fied by its superior s ens itivity and

specifici ty . Tests such a s the radioimmunoassay procedure

(Chappel � � . , 1976 ) require expens ive and compl ex equipment and

for this reason are pr�bably limited to use as a research too l .

;

37 .

An important cons iderat ion when evaluating te s t s against

resu l t s of cul ture is that whereas the sens itivity of a given test

is readi ly derived, the spec i f icity is not . S ince a negative

cu l ture resu l t cannot be taken as proof o f the absence o f infect ion

i t i s usual for e st imations o f specificity to be made in known non­

infeeted populat ions ( e . g . McKinnon , 1963 ) . However , such

popu lations may not be representative and deve lopment of t i tres by

animal s exposed to , but not infected by , the disease may cause the

test s pecificity to be qui te d i fferent in a partial ly infected

popu lat ion from that in a d i s ease free popu lation .

38 .

Eva luat ion of the Bruce l losis Card Test (BCT ) or Rose Bengal

P late Test (RBPT )

The bruce l losis card t e s t was deve loped fo l lowing the

observations o f Rose and Roepke ( 1 95 7 ) that the rap id p late

agg l u t ina t i on t e s t wou l d , wi th an ac i d i f i ed ant i gen , d e s t roy the

act ivity of non-specific agglut inins .

Both the card and Rose Bengal p late tests ut i l i se a Rose

B engal stained ant igen buffered at pH 3 , 65 and adj us ted to contain

8% o f ce l l s by vo lume . The two tests are somet imes known as

Buffered Bruce l la Antigen tests (BBA) . De tai l s o f ant igen

preparat ion and te s t methods are given by Al ton � � . ( 1975a ) .

Nico le t t i ( 1967 ) ini tial ly evaluated the use o f the card test

and found that i t was pos i t ive in all o f 184 cu l tura l ly pos i t ive

cat t le whereas the tube agglut inat ion test detected only 48% o f

these a s be ing reactors . Morgan � �. ( 1969) us ing the same

ant igen as that prepared for the card test but us ing serum ins tead

o f p lasma for the tes t , found good agreement wi th the comp lement

f ixat ion and serum agglut inat ion tests .

Fo l lowing this ear ly work the card tes t gained acceptance as

a rapid screening te s t whereby any sera positive to it were

retes ted us ing trad i tional techniques . The ear ly acceptance o f

the card test in the US was a lso prompted by increas ing farmer

agi tation for a more rap id test ing method , particular ly from beef

cat t le ranchers in the southern s tates (Becton, l97 6 ) .

S tud ies re lating bacteriologica l cul ture resu l ts to the test

have been few, Nico letti ( 1967 ) showed that all 1 84 cu l ture

pos i t ive cat t le that he examined gave pos i t ive card test react ions .

Fensterbank ( 19 7 3 ) in a s tudy o f exper imental ly infected hei fers

found that 1 . 6% of the serum samples from known infected animal s

were negat ive to the Rose Bengal tes t . My lrea ( 19 7 2 ) in an

investigation o f serological tests in natural ly infected cows

found that two out of 4 2 cul ture pos i t ive animal s gave a negat ive

RBPT ( i . e . 4 . 8% false negatives ) . In a further s tudy (Mylrea and

Fraser, l97 6 ) a l l o f 40 cul ture pos i t ive cows gave a pos i t ive RBPT .

Alton � �· ( 1975b) in a s tudy o f natural ly infected animals ,

found that only one o f 7 9 animal s gave a negative RBPT . The

worst correlations appear to be those obtained in trials

Figure 1 Bruce l los i s card te s t

Nos . 1 3 5 9 1 0 negat ive

No . 4 ++ posi t ive

Nos . 2 6 7 8 +++ pos i t ive

BRUCELLOSIS U. S. PAT. N O . 3 , 0 7 4 , 8 5 3

preceding the New Zea land eradicat ion scheme when 27 out o f 2 1 3

cu l ture posit ive animals ( 1 2 . 7% ) were negat ive t o the card test

(Adlam , l 9 7 8 ) .

39 .

S tudies in Great Britain and Ire land have attempted to re late

the RBPT to the comp lement fixation (CFT ) and serum agg lu t ination

(SAT ) t e s t s . Mor gan � !1 . ( 1969 ) t e s ted 6424 uns e l e c t e d c a t t l e

sera and found the maj ori ty o f sera wi th positive CF tests were

also pos i t ive to the RBPT .

There seems to have been some difficu l ty in the ear ly s tages

in deciding whether to compare the RBPT with the CFT or the SAT or

both . When compared to the CFT the RBPT has shown good sens itivity

(Morgan � � . , 1969 ; Davies , l97 1 ; Morgan and Richards, 1 9 74 ) ,

whereas when compared to the SAT , especially in the suspicious

(marginal ly pos itive ) range , it has been less enthusias t ica l ly

rece ived (O ' Re i l ly and Cunningham, l9 7 1 ; Prior � �. , 1 9 75 ) .

Canadian investigators examining sera from a large ly Bruce l l a -

free popu l ation found that 99 . 8% o f the negative card t e s t resu l t s

were confirmed by the serum agglut inat ion tes t , which was the

official tes t . On the other hand 5 6 out o f 15 2 ( 37%) agglutination

test reactors were mis sed by the card test whereas only 32 out o f

482 ( 6 . 6% ) sera suspicious o r posi tive t o the complement fixation

tes t were card tes t negat ive (Pr,ior � �· , 1975 ) . The que s t ion of

whether to accept the CFT or SAT test as the defini tive test has

been co loured somewhat by the traditiona l acceptance of the SAT as

the official te s t and the practical dif ficu l t ies of us ing the CFT

on a wide range o f unse lected s era or sera from known non- infected

herds .

In terms o f test sensi t ivity the card te s t is genera l ly

recognised as being too sens it ive to use as a definit ive test but

because of its ease o f app l icat ion and good specificity i t has

been used extensive ly as a screening tes t . The bacterio logical

s tudies reported by Nico lett i ( 19 67 ) , Mylrea ( 19 7 2 ) , Al ton � al .

(1975b ) , Mylrea and Fraser ( 19 7 6 ) and Ad lam ( 1978 ) a l l inc lude

observations on cul ture negative animal s which ind icate that the

card t e s t is oversensi t ive , especially in s train 1 9 vaccinated

animals .

In New Zea land the re lationship between the BCT and the CFT

, has been investigated by T imbs � �. ( 1978a) who emphas ised that

40 .

the dynamics o f the population under test mus t be considered when assessing the s ensitivity of a tes t . I t was found that in

heavily in_fected herds 78 . 67. of BCT pos itive sera were a l so CFT

positive whereas in l ight ly infected herds only 1 2 . 6% o f the BCT

positives were CFT pos it ive . Thus the sens itivity of the BCT

wi th r e s p e c t to the CFT had changed f�om . 7 8 6 eo . 1 2 6 s imp ly by

considering a d i fferent samp le . No doubt the sens itivity o f the

CFT changes in a like manner with respect to true infection .

Nicoletti ( 19 6 7 ) found a s imi lar change in sens i t ivi ty when

evaluat ing the BCT wi th respect to cul t ure in two groups o f herds

wi th different infect ion rates .

Reasons for the apparent oversens itivity of the card test are

not clear and have apparently not been speci fica l ly invest igated .

Al lan � �. ( 19 7 6 ) have drawn at tention to at least two types o f

non- speci ficity in agglut ination re�ct ions .

( i ) That due to immunologica l ly non- specific agglutination .

This i s due t o non- specific agglutinins which can

agglutinate a variety o f unre lated bacterial antigens .

( i i ) Tha t due to immunologica l ly s peci fic but non-

diagnostica l ly specific agglutins , e . g . prior s train

19 vaccination .

Innumo logical ly non- speci fic agglutination has been

investigated by Hess ( 195 3a , 1 953b ) and Rose and Roepke ( 195 7 ) .

Organisms that have been reported as cross-react ing wi th Bruce l l a

inc lude Pas ture l la (Mal lman, l930 ; B erman, l956 ; King, l961 ;

Morse e t �. , 1 95 3 ) . Vibrio (Morse � �· , 1953 ; Kiggens � �· ,

1955 ) , Salmonel la (Morse � �. , 195 3 ; Corbe l, l975a) and Yers inia

(Ahvonen � �. , 1 969 ; Corbe l and Cu l len, l970) .

Another aspect o f the evaluation of the card test which

requires cons ideration is its effectiveness in detecting ear ly

infect ions . Because of the long incubation period o f the disease

it would be advantageous to emp loy a test that is efficient at

detecting early or latent infect ions . Ear ly invest igators

observed that with infected animals the card test became pos i t ive

before the CFT or the SAT (Nico le t t i, 1967 ; Morgan � �. , 1969 ;

Davies, 1 9 7 1 ) . As thorough invest igat ion of thi s aspect requires

experimental infection o f catt le , �here have been few

comprehensive s tudies o f i t . Fensterbank ( 197 3 ) , in an experiment

41 .

involving the inoculation of 50 two year o l d he ifers , found that

the RBPT became pos i tive earlier than d id the CPT or SAT . On average

the RBPT became pos i tive 40 . 5 days after infection for unvaccinated

cat t le and 29 . 5 days after infection for animals vaccinated as

ca lves with s train 1 9 . These figures compare wi th 49 . 5 and 34 . 3

days res�e�t ive ly fgr the CFT . The serYm e��lYt ina t iQn t e s t tggk an average o f 6 3 . 2 days to become pos i t ive in the group as a whol e .

Pietz ( 19 7 7 ) found that in experimental ly infected animals i t

took an average o f 6 2 days after exposure for the agglut inat ion

test to detect infected animal s whi l s t the card test took only

43 days .

Immunoglobu l ins act ive in the Card Test

Results o f s tudies to identify the c lass or classes o f immuno ­

globu l in act ive in the card or RBP tests have been conflicting .

Corbe l ( 197 2 , l 9 7 3b , 1973c ) Diaz and Levieux ( 19 7 2 ) and Wood and

Corbe l ( 19 73 ) cons idered that the only immunoglobu l in act ive in

the test was IgG1 • Levieux ( 1974) found that IgG1 and IgM were

act ive but that IgG2 activity was inhibi ted by the low pH . Other

s tudies have indicated that IgG1 , I gG2 , and IgM are a l l act ive

(Beh , l97 3 ; Al lan � � . , 1 97 6 ; Pat terson � �. , 19 7 6 ) . A further

attempt at identi fying immunoglobu l in c las ses act ive in the

card test is made in this thes is .

I t has been sugges ted that the discrepancies in re sults may

be due to di f ferent methods chosen for analysis and because sera

vary so much in their immunoglobu l in content (Jones , 1977 ) . In the

knowledge that it is IgM agglutinins that �nd to pers i s t after

s train 19 vaccinat ion , and that the card test is more "over

sens i t ive' ' in s train 19 vaccinated popu lat ions i especial ly in . younger animals (T imbs � �. , l978a ) , then the theory that I gM is

active in the card tes t reaction tends to carry mos t weight .

Further circums tantial evidence for the ro le o f I gM in the card

test is provided by the fact that IgM i s the first immunoglobul in

produced fo l lowing irtfect ion and the card tes t detects infection

ear l ier than o ther tests (Rice � � . , 19 6 6 , 1967 ; Rice and B oyes ,

197 1 ; Fensterbank , l 97 3 ) .

Evaluat ion of the Complement Fixation Test for Bruce l lo s i s

The complement f ixation te s t was used a long with the

agglutination tes t in the ear l ie s t invest igat ions into bovine

abortion in Denmark and England . Al though most of the early

workers agreed tha t the tes t was more efficient than the

agglut inat ion tes t , i t was apparently cons idered too comp lex for

rout ine use . Jones � � . ( 1 9 6 3 ) and O l i tzki ( 1 970 ) have

extens ive ly reviewed the early l i terature and Jones � �. ( 19 6 3 )

have highl ighted the observation that comprehens ive reviews

o f bruce l losis by Sp ink ( 1 95 6 ) , S tab leforth (l95 9 ) and Dalrymple­

Champneys ( 1 960) have d i smi sseq the test with a brief statement .

42 .

Of the many variab les in the test method , two which have

great inf luence on its e ffect iveness are the t ime and temperature

of the pr imary incubation and the type o f antigen used .

Some workers have expressed a preference for the co ld fixa tion

(4 °C for 1 8 - 24 hours ) method (Zeissig and Mans field , l930 ;

Tri lenko , 1 95 7 ; I sayama , 196 1 ) and i t has been the experience o f

workers a t Wal lacevi l l e that the cold te s t i s ab le t o detect a

smal l number o f infected animals which are otherwise mis sed by

the warm method (Te Punga , unpublished ) . Many laboratories prefer

to use the warm f ixat ion ( 37 °C for 30-60 minutes ) technique and

o f 25 laboratories in 1 8 countries participating in a recent survey

10 used warm fixation and 1 3 used cold fixation with two

laboratories us ing both (Morgan � �. , 197 3) . Preference for

the warm method is general ly due to its convenience and the

rapidity with which results are obtained .

Various different antigen preparations have been used

including a trichloracetic acid extract (Renoux and Al ton, l95 7 )

. and soluble antigens ( Renoux , l95 7 ) . The heat ki l led who le ce l l

antigen a s used for the agglut ination test has , however , proved

to be quite satis factory and is a lmost universal ly used as the

standard antigen .

There are re lative ly few reports o f s tudies relating

comp lement fixation test resul ts to bacteriological findings .

McKinnon ( 1963 ) , in a comprehens ive work , laid the framework

for the International S tandard Ant i-Brucel la abortus Serum . He

examined sera from 1 1 34 vaccinated , unvaccinated , Bruce l la -free

and infected cattle and proposed determinant t i tres for

Figu re 2 Au to -An alyzer adaptation o f the complemen t - �

fixation tes t for brucellos is

I

43 .

classi fication o f negative , suspect and reactor animals .

Lambert and Ameraul t ( 19 6 2 ) demonstrated pos itive CFT t i tres

in 16 experimentally infected catt le , whi le among 20 which res i s ted

challenge only one showed a trans ient ti tre . In more recent

s tudies Nico le t t i and Muraschi (196 6 ) found that 1 14 out o f 1 17

e• t t l • ( 9 7 . 5% ) whieh gave po s i t ive e u l t u r e a a l s o gave po s i t ive CF tests ; two o f the three catt le improperly identi fied gave

suspicious react ions . Nicoletti ( 19 6 9 ) in a further s tudy

obtained 1 1 6 CFT pos itive t i tres from 1 1 9 cu lture pos i tive cat t le

( 9 7 . 5%) . Renoux � � . ( 197 1 ) found that 10 out o f 47 experimentally

infected hei fers gave negat ive or doubtful agglut inat ion tests at

abort�on or ca lving whereas only three gave negat ive comp lement

fixation tes t s . Fens terbank ( 19 7 3 ) experimental ly infected 43

he ifers in mid-gestat ion and they a l l gave a pos i t ive CFT response ,

the average t ime between infect ion and deve lopment o f titre was

34 . 3 days for s train 19 vaccinated he ifers and 49 . 5 days for those

unvaccinated . Mylrea ( 19 7 2 ) isolated Br . abortus from 42 cows and

sera from 4 1 o f these gave a pos i t ive CF ti tre and in a later s tudy

(Mylrea and Fraser , l 9 7 6 ) 47 pos i tive CF ti tres were obtained from

the sera o f 47 cul ture pos i t ive cows . Al ton � � . ( 1975b ) showed

that the CFT detected a l l of 7 9 cat t l e giving pos it ive lymph node

cu ltures . New Zealand trials undertaken to as sess the potential

o f various t e s ts also showed the co ld CFT to be the mos t sens itive

test al though only 1 10 out o f 1 3 1 infected cat tle ( 84%) were

detected by i t . This compared with 84% de tected by the card test

and 80 . 1% de tected by the SAT (Ad l am , l978 ; Te Punga ,unpub l i shed ) .

An interesting aspect o f the New Zealand trials was that 1 9 o f the

1 3 1 cu lture pos itive animals were negative to a l l sero logica l tests

app l ied .

Al l o f the above investigators have concluded that the

complement f ixation test is the best s ingle serological t e s t 1

available for bruce l losis in catt l e .

Irnmunoglobul ins Act ive in the Compl ement F ixation Test

Inve s tigations into the various c las ses of irnmunoglobul ins

active in the CFT have general ly been conducted in paral l e l with

investigations of the card and serum agglutination tes t s . Rice

44 .

� a l . ( l966) have c laimed that lgM does not fix complement , however ,

Beh ( 1973 ) , Levieux ( 1974 ) , Allan et al . ( l976 ) and Patterson e t al . -- --- - --

( 19 76 ) all agree that bo th lgG1 and IgM immunoglobul ins fix·

comp lement whi l s t lgG2 does not . Allan � !1 · ( 1 9 7 6 ) maintain that

IgM probab ly fixes comp lement twice as efficient ly as lgG1 on a

W@ t lh t b iM t l , n�wevo f , beeau•• t e t• heat lab i l e in serum, i t may be des troyed during the inactivation proce ss commonly performed

prior to conduct ing the CFT .

The Prozone Phenomenon

The use o f the comp lement fixat ion te st for bruce llosis is

comp l icated by the occasional occurrence of "prozones" . Sera

exhib i t ing this e f fect appear to give negat ive reactions in low

tes t d i lutions whi le giving pos i tive react ions in higher di lut ions .

Al though the occurrence o f prozones in the agglutinat ion test i s

we l l known and has been inves t igated (Gl enchur � �. , 19 61 ; Cho

and Ingram , l9 7 2 ) there has been little comment unt i l recently on

the s igni ficance o f prozones in the CFT a l though they were

recogni sed by Burki ( 195 7 ) . Al ton � !1. ( 1975b) noted the

exis tence of pro zones in both the warm and co ld fixation tests but

they went to much higher dilutions in the warm tes t . Placket t and

Al ton ( 1975 ) and McNaught � !1 . ( 1977 ) investigated the immunogenic

mechani sm of pro zones and found that specific IgG2 ant ibodies to

B r . abortus cou ld b lock comp lement fixation by IgG1 and IgM

ant ibod ies . McNaught � !1 • ( 1 9 7 7 ) also . no ted that the relat ive

ant i gen concentrat ion has a marked influence on the extent o f

prozoning , a high ant igen concentration inhibits prozoning whi l e

a low concentrat ion enhances i t .

The importance o f this IgG1 /IgG2 interaction is highl ighted

by the fact that the ratio · .o f the two c las ses can vary widely in

cat t l e and in particular , serum IgG1 is dep leted shortly before

parturi tion by trans fer to the mammary gland whi le serum I gG2 leve ls remain high (Brandon � !! . , 1972 ; Wi l l i ams and Green , 1 9 7 6 ) .

Thus prozone react ions may be more l ike ly to occur at about the

t ime o f parturi tion.

Deve lopment o f Automated Serological Tes ting Methods for

Bovine Bruce l lo s i s

4 5 0

Automated serological test ing methods have been a recent

deve lopment compared wi th the much ear l ier introduct ion of automated

biochemi s try . An exce l lent review of the methods and machinery in

current use for automated sero logy is given by Kwantes ( 1 9 76 ) .

A bib l iography on rap id methods and automation in microbiology and

immunology has recent ly been pub l ished by Pa lmer and LeQuesne

( 19 76 ) .

Automated tes ting methods are general ly divided into continuous

flow and d iscrete samp l ing sys tems ( S tevens 1 9 7 3 ) . In cont inuous

flow systems samp les s eparated by air bubb les are fed through a

tube and reagents are added to the s tream so that react ions take

p lace and are measured within the tubes . In the discre te samp l ing

systems individual samp les are p rocessed separate ly . -/(

The Technicon Auto-Analyzer was deve loped dur ing the 1950 ' s

to provide a rapid method o f performing c l inical biochemica l

analyses� The principle used was that o f cont inuous flow analys is

and in 1957 the Auto -Analyzer became the - first commercia l ly

avai lab l e fu l ly automated testing system.

Joubert � � . ( 19 6 7 ) provided the first descript ion of a

comp lement fixation te s t for bruce l losis adapted to the Auto­

Analyzer , and later , Mi l ler � � . ( 1 9 7 3 ) , described a s imi lar

sys tem deve loped in B r i tain . Auto-Analyzer adap tat ions o f the

agglutination tes t have been described by Joubert � �. ( 1967 ) ,

Vargues � al . ( l9 68 ) and Mi l ler ( 19 7 l a ; 19 7 lb ) . An automated

Rose Bengal P late t e s t has been deve loped at the Central Veterinary

Laboratory , Weybridge , and has been documented by Gower � �. ( 1974 ) . The Autotape machine used in this app l ication has the

capacity to perform 1 200 tests per hour .

The automated comp lement f ixat ion test used in the New Zealand

bruce l lo s i s eradicat ion scheme was deve loped a t the Wal l acevi l l e

Animal Research Centre and was origina l ly described b y Te Punga

( 19 7 1 ) . It has received further mention by E l l io t and Pul l an

( 19 7 3 ) and T imbs . � �. ( 1978b ) . The init ial evaluat ion o f the

* Technicon Equipment Corp . N .Y .

I

test with respect to its corre lation wi th animals known to be

infected and free of d isease , and with other manual tests has not

been pub l ished . This thes is further examines the efficiency o f

the tes t wi th particu lar respect t o its capacity to detect

prozoning sera .

4 6 .

Anothe� s emi - au t oma t e d t e s t in; me thod whi�h i 8 be comin; wi de ly

used in sero logy is the micro adapt ion of s tandard tube tests . A

machine may be used to perform d i spens ing and ser ial d i lution

functions thus automat ing an otherwise manua l micro- tes t sys tem .

Micro comp lement fixat ion tes ts are used in the Aus tralian and

New Zealand bruce l losis erad ication schemes , Al ton ( 19 7 7 b ) and

T imbs � �. ( 1978b) have described the methods use d .

B ruce l lo s i s Eradicat ion S chemes

B ovine bruce l losis contro l or eradication schemes have been

at tempted in many countries . Such programmes have been j us t i fied

on the grounds of e conomic l o s s by infected catt le and because o f

pub l i c h ea l th cons idera t ions . In New Z e a l and the maj or s t imu l u s

for eradicat ion was the threat o f d iscontinued market acces s for

meat products , particular ly to the European Economic Community .

47 .

The init ia l approach to contro l has genera l ly been to dec lare

bruce l losis a not i fiable diseas e . An Act o f 1903 in Norway

required catt l e owners to not i fy cases o f contagious abortion and

in 1 9 20 bruce l lo s i s was made a noti fiab l e di sease in Denmark

(Thomsen , l95 7 ) . Fo l lowing the deve lopment of the serum

agglut inat ion test contro l p ro grammes based on a test and s laughter

pol icy were begun in 1934 in the USA , Denmark 1936 , Norway 1935 ,

F inland 1938 and Sweden 1938 , (Thomsen , l 95 7 ; S chi l f , l 9 7 2 ) .

Countries having comp l e ted or vir tua l ly comp l e ted bovine

bruce l losis control or eradicat ion schemes include Denmark

(dec lared free 1970 ) , Norway ( 195 2 ) , Sweden ( 195 7 ) , West Germany ,

United S tate s , Canada , F inland , Northern Ireland , I ta ly ,

Luxembourg , Swi t zerland , Czechos lova l ia and Tasmania .

By 1976 extensive programmes were operating in Japan , Aus tria ,

Austra l i a , New Zea land , Hungary , East Germany , Yugos l avia , USSR ,

and Great Britain .

48 .

CHAPTER 2 . CULTURAL AND SEROLOGICAL STUDIES

Introduction

Isolation o f Bruce lla abor tus from the tissues o f an infected

animal is the only indisputab le evidence availab le that the animal

is infe c t e d . Because of the abs ence o f any c l inical s i gns or gros s pathology , and the need to cu lture a var iety of tis sues , s laughter

of the suspect animal is genera l ly requ ired be fore an assessment

o f its true infection s tatus can be made .

In the New Zealand Bruce l lo s i s Eradication Scheme the

bruce l l o s i s card tes t (BCT ) is o ften used as a screening test and

pos i tive ly react ing sera are then submi t ted to the comp lement

fixation test ( CFT ) for definit ive diagnosis . In many ins tances

all sera from herd s are card tes ted and submi tted for CF testing

desp i te the card test resu lt .

Herds are retes ted regu lar ly at 2-4 month interva ls unt i l

declared free o f brucellosis (Ad l am � � . , 1978a ) .

I t has been not iced that many animals may be repeated ly BCT

pos itive and CFT negat ive for some months (T imbs � � . , 1978a ) . Such

BCT+/CFT- reactions could be the result o f some related or non­

specific antibody which is detected by the card test but not by

the CF tes t .

This section describes cu l tural and serologica l s tud ies

carr ied out on a group of BCT+/CFT- animals .

Mater ials and Method s

1 Selection of Cat tle for S laughter

Twenty cows which had been BCT+/CFT- for at least three

previous tests were selected for s l aughter from 12 herds in South

Taranaki . The age of these an��als ranged from three to 10 years .

As far as could be de termined , a l l had been vaccinated a s calve s

with l ive Brucel l a abortus strain 1 9 vaccine .

S e lection was solely on the bas is o f each having had at least

three previous tests BCT+/CFT- , no account was taken o f the previous

herd testing his tory . · . · The ·herds from whi ch these animal s came had

histories of continuing reactors despite repeated te sting . These

reactors d id , however , tend to occur sporadica l ly so that the total

number removed was not unduly l arge and the herds could not be

49 .

said to be heavily infected .

Co l lection o f S pecimens

Because o f financial constraints i t was neces sary for the

purchased cows to go d irectly for s l aughter . I t was not pos s ib l e

to co l lect serum o r mi lk specimens or to observe them for a per iod

pr ior to s laughter . The animal s were routine ly s l aughtered a t

an export freez ing works and approximately 3 i b lood and various

t i s sues were co l lected from each animal during carcase procees s ing .

Sections o f u terus , sp l een , udder , and mesenter i c , supra­

mammary , i l iac , retropharyngeal , sub-maxi l lary , and parotid lymph

nodes were co l l ected in individual p lastic bags and refrigerated

overnight before cul turing . Sera were stored at - 20 °C .

Culture Technique

Each t i s sue was defatted , dipped in 95% ethanol , flamed and

p laced in a s teri le p lastic bag with an approximate ly equa l vo lume

o f 0 . 857. sal ine . �·c

The bag was p laced in a Co lworth S tomacher and

the tissue pounded for approximate ly 30 seconds . Two drops o f

the resul tant homogenate were spread onto cul ture media and 0 . 5 m l

was retained for guinea p i g inocu lat ion .

* A . J . Seward .Ltd . , Lond .

5 0 .

Bruce lla se lect ive medium was prepared according to the

method o f Al ton and Jones ( 1975a) us ing a serum dextrose agar base

and cycloheximide , baci tracin and po lymixin B as ant ibacterials .

B lood agar was prepared by adding 5% bovine blood to a nutrient

agar base . Homogenate from each tis sue was inoculated onto the

mo d f ym , All PlA t O • w• �• tneub a e e d at 3 7 °C in an a emo aphere o £ 5% co

2-in -air and examined at three and f ive days for evidence o f

growth . Each batch o f med i um was tes ted for its ab i l ity to sus tain

growth by inocu lating wi th �· abortus b iotype 2 .

Suspect Bruce l l a co lonies were examined microscopica l ly and

i f neces sary rep lated and later tes ted us ing a s l ide agglutination

tes t agains t contro l ant i-serum . Any other uniform growth on

blood agar p lates was also examined care f u l ly and subj ected to

rout ine biochemi ca l tests as required to establ ish ident i ty . Use *

was made of APl 20 mul t i - te s t s trips to identify entero-

bacteriaceae .

Guinea P ig Inocu lat ion

Fol lowing homogenisation 0 . 5 ml o f each samp le was placed in

a s ingle sterile bo t t le so that a compos i te samp le of homogenates

from each anima l was avai lab le for gu inea pig inoculation .

One ml o f e ach of the 20 compos i te homogenates was inj ected

intramuscularly into the left hind leg of each of 20 guinea pigs .

S ix weeks later the guinea p igs were b led and euthanased . At

autopsy a l l organs were examined care fu l ly for any sign o f

1 abnorma l i ty and a l l sp leens were taken asep t ical ly for d irect

culture onto b lood agar and Bruce l la select ive med ium . Serum

agglut ination and comp lement fixat ion tests were performed on a l l

sera .

Preparat ion o f Who l e Ce l l Antigen

For most o f the standard serology a s tandard ised concentrate

o f B r . abortus strain 99 preserved wi th · 0 . 5% pheno l was used .

This was obtained from a commercial source+ .

When required , fresh unpreserved antigen was prepared by

* Analytab Produc t s Inc . N . Y . +

Bruce l l a abortus standardi sed concentrate . I . C . I . Tasman Vaccine Ltd . , Upper Hut t , New Zealand .

I I

cul ture o f strain 99 or s train 19 �· abortus on potato infusion

agar in Roux flasks as described by Alton � � . ( 197Sa) . After

harvesting , the organisms were heat ki l led at 60 °C for 1 hr ,

centri fuged and resuspended in phosphate buf fered saline ( PBS )

pH 6 . 4 . S uspens ions were kept at 4 °C unt i l required , but being

unpre1 �ved were ua•d wi ehtn !4 daya .

Preparation o f Ultrasound Treated Antigen

51 .

We l lcome opacity tubes+

were used to prepare a suspens ion o f

1 x lO ll ce l l s /ml from fre shly harves ted , heat ki l led , B r . abortus

s train 99 or strain 1 9 . The suspens ion was made in d i s t i l led

water . Ce l l s were sonicated for 20 minutes in an M . S . E . ul trasound

d is integrator+

. An ice j acket w�s used to prevent excess ive

heating . Fol lowing u l trasound treatment the suspens ion was centri ­

fuged at 4000 g for 30 minutes t o remove ce l l debris . The

supernatant was then used as a solub le ant igen .

Preparation of Ether Treated Antigen

The method used was essentially that o f Ribi � �. ( 1959 ) .

Both commercially prepared and preserved Br . abortus s train 99

and freshly prepared s train 1 9 organisms were used .

Washed ce lls were suspended in 0 . 85% sodium chloride to a

concentration of 1 x 10" ce l ls/ml , and shaken with exces s ether in

a separatory funne l for 1 min . The suspens ion was left to s tand

overnight . The aqueous phase which had settled to the bo ttom was

then removed and air was bubb led through i t to remove any exce s s

ether . Insoluble res idue was removed by centri fuging at 4000 g

for 30 minutes .

Bruce llos is Card Tes t ( BCT )

In i t s commerci a l ly avai lab le form this test i s known as the *

Brewer Card Test • Fu l l ins tructions accompany the kit .

+ Wel lcome Research Laboratories , Beckenham , Kent .

+ Measuring and Scient ific Equipment Ltd . Lond .

* Brewer Diagno s t ic Kits , Hynson Wes tco t t and Dunning Inc . , Baltimore , Maryland , U . S .A.

.

0 . 03 ml volumes of serum and anti gen were measured on to

whi te card , each card he ld 10 samples . After mixing with a

s t i rrer the card was placed on a rocking machine and rocked for

four minutes , results were then read immediately .

Reactions were scored a s fo l lows : -

no visible agglutination

+ very fine particles pre sent

++ fine partic les observed wi th some clumping

+++ comp lete clumping of antigen

Serum Agglutination Test ( SAT )

This test was performed accord ing to the method o f Al ton

5 2 .

� �. ( 1975 a ) . A series o f doub ling d i lut ions o f each test serum

was made in phenol saline and an equal vo lume of d i luted

' 's tandardi sed concentrate'' antigen added . Incubation took place

overnight at 37 °C . Titres were read according to the degree o f

agglutination present after incubation . The degree o f

agglutination in each tube was graded from O , no agglut inat ion ,

to 4 , 100% agglutination .

Coombs Test

Fo l lowing reading o f the agglutination tes t , tubes wi th 2+

or more agglut ination were d iscarded . All remaining tubes were

centr i fuged at 2000 g for 20 minutes and the supernatant discarded .

I One ml of pheno l - sa l ine was added to each tube and the deposit

resuspended with the aid of a pas teur p ipette or vortex mixer .

Centri fugation and resuspens ion was repeated and fol lowing the

third centrifugat ion 0 . 45 ml of pheno l - s a l ine was added and the

depo s it resuspended . At this s tage 0 . 05 ml of d i luted rabbi t

ant i -bovine globu lin serum was added and the tubes incubated

overnight . Results were read as for the agglutination tes t .

T i tration o f Rabb it Anti-bovine Globul in (Coombs reagent )

A known high t itre Coombs pos i t ive serum was taken and '

d i luted in doubling d i lutions to 15 tube s . This was repeated for

five sets of tubes . A serum agglutination test was performed by

5 3 .

adding ant igen and incubating . The fo l lowing day tubes showing

less than 2+ agglut inations were centr i fuged and the ant igen washed

as descr ibed above . After the f inal centr i fugat ion 0 . 45 ml phenol ­

sal ine was added t o each tube . Sui tab le d i lutions o f rabbi t ant i ­

bovine g lobu l in (Coombs reagent ) were made , 1 : 5 , 1 : 10 , 1 : 20 , 1 : 40

and l t 80 . 0 . 05 ml o f e ach d i l u t ion o f Coombs reagent was added

to each tube in i t s corresponding series , thus each test series

had a d i f ferent di lut ion of Coombs reagent . The tubes were

incubated overnight and resu l t s read the fo l lowing day as for an

agglutinat ion test .

The Coombs reagent d i lut ion in the series o f tubes having the

highest t i tre was taken as the optimal di lution to use in the t e s t

proper .

Rivanol Agglut inat ion Tes t

A 1 % solution o f Rivanol ( 2 ethoxy-9 diamino acrid ine lactate )

in d i s t i l l ed water was prepared . Equa l vo lumes o f 1% Rivanol and

serum ( 0 . 5 ml each) were prepared , shaken and al lowed to s tand a t

room t emperature for 15 - 30 minu te s . The suspens ion was centr i ­

fuged a t 1000 g for 1 0 minutes and the supernatant used in an

agglut ination tes t , replacing normal serum . Account was taken o f

the d i f fer ing equivalent serum d i lution obtained afte r Rivanol

treatment i . e . the supernatant contained the equival ent o f 50%

serum .

1 2-Mercapto-ethanol Test ( 2-ME )

Two methods were used for this test .

( a ) Serum was treated b y preparing a 0 . 1 M solut ion of

2-ME containing serum in a 1 : 5 concentrat ion ( i . e .

0 . 7 m l sa l ine , 0 . 2 m l serum , 0 . 1 ml o f lM 2-ME ) . Thi s

solution was then used as the f irst tube in an

agglut inat ion test series which used normal sa l ine as

the d i l uent in the remaining tubes .

(b ) The agglutination tes t was performed in the s t andard

fashion except that a Q . 05M solution o f 2-ME in norma l s a l ine was used in p lace of pheno l - s a l ine as the d i l uent .

I I

54 .

The antigen concentrate used in the 2 -ME tes t was centri fuged

to remove the pheno l - s a l ine di luent and resuspended in normal sal ine

to a d i lut ion equiva l ent to that obtained by diluting the

standardised concentrate 10 times . This phenol- free ant igen was

then used in e ach 2-ME tes t and in the agglutination tests

per formed i n para l l e l t o them .

Comp lement F i xation Test (CFT )

Both automa ted and manual comp lement fixation tests were used .

The automated CFT is described in Chap ter 6 .

The method used for manual C F tests resemb led that descr ibed

in the U . S . Pub l ic Health Service Monograph No . 74 (Anon . , 1 965 a ) .

Both macro and micro test var iat ions were used . Barbital buffered

sal ine was used as the d i luent for a l l reagents ( see Append i x I ) . *

Reagents were titrated in macro vo lumes . Haemo lysin was

t i trated by the p lateau method des i gned to give op tima l ly

sens itised erythrocytes . F ive 50% haemolytic units of guinea pig + comp lement were used . A commerci a l ly avai lable standard i sed

+ concentrate o f Br . abor tus as used in the agglut ination tests was

used as the antigen . Sonicated and ether-water extracts o f

Br . abortus prepared a s described previou sly were also used as

antigens . Ant igens were t i trated in a block ti tration against a

positive serum . Sheep blood was co l lected in Alsever ' s solution

( see Append ix I ) and left to stand at 4 °C for at least one week

before being centri fuged and the ce l ls washed three times wi th

barbital buffe r . The erythrocytes were standard ised to 2% by

centri fuging an approximately 50% suspens ion at 1000 g for

10 minutes , in a graduated tube , record ing the concentra tion , and

d i luting accordingly .

Both warm fixat ion ( 37 °C for 30 mins ) and co ld fixation ;

( 4 °C for 18- 24 hrs ) methods were used . For the warm fixation test

doub l ing serum d i lutions commencing at 1 : 4 were used whi le for co ld

fixations the tes t series started a t 1 : 10 . Reactions were described

accord ing to the degree o f haemo lys i s in each tube .

* Commonweal th Serum Laboratories , Parkvi lle , Vie .

+ Becton , Dickinson and Co . , Cockeysvi lle , Maryland .

+ I . C . I . Tasman Vaccine Ltd . Upper Hutt, N . Z .

I

5 5 .

0% haemo lys is 4

25% " 3

5 0% " 2

75% 11 1

100% 11 0

Serum and comp lement controls and a test serum o f known t i tre

were inc luded with each batch of tests .

Sera were inact ivated at 58 °C for 5 0 minutes . For the

macro test 0 . 25 ml vo lumes o f d i luted serum , �omp lement , ant igen

and haemo lytic sys tem were used . For the micro system 0 . 0 25 ml

uni t vo lumes were used .

Rivano l Comp lement Fixation Test

After inact ivation at 58 °C for 5 0 minutes serum was treated

wi th an equal vo lume of 1% Rivanol . In some

tes t s any surp lus Rivano l was removed from the supernatant by

treat ing i t wi th 5% w/v NaC l and leaving to stand overnight ,

A precipi tate containing the Rivano l formed and was centri fuged

off . The treated serum was then dialysed agains t barbital buffer

( pH 7 . 4 ) to reduce the salt content . In o ther tests the super­

natant recovered d irect ly after Rivano l treatment was used des p i te

the presence o f some co lour due to surp lus Rivano l .

The Rivano l treated serum was used in the comp lement fixat ion

test as a direct subs titute for normal serum except that i t was

only hal f normal s erum s trength due to d i lution with the Rivano l

solution .

Gel Diffus ion Test

This tes t was performed according to the method o f Ouchterlony

( 195 3 ) . The d i f fusion medium was 1 . 2% Agarose in 0 . 85% saline

containing 0 . 1% sodium azide . Sonicated and ether-water treated.

anti gens were used , p lates were incubated at room temperature

and read daily for three days .

Ind irect Haemo lys i s Test

The method o f Placke t t � !1. ( 1976) was used . After making

I

56 .

doub ling d i lutions o f serum in barb ital buffer ( pH 7 . 4 ) in micro­

titre p lates , an equa l vo lume of washed ant igen- sens i t ised bovine

red blood ce lls was added and after 15 minutes incubat ion at room

temperature one drop o f a 1 : 10 d i lution o f guinea p ig serum

( comp lement ) was added . Plates were incubated for 60 min at 37 °C

wi th Qgn s t ant e hakini and e tgked ove �n i ah t a t 4 °C tg a l lgw un lysed

ce l l s to settle . The endpo int was taken as the highe s t d i lution

showing comp lete haemo lysis .

Record ing and Interpretat ion o f T i tres

To faci l i tate result notat ion whi l e s t i l l provid ing an

indication of the degree o f react ion at each dilut ion o f the tes t

series the fol lowing system o f result recording was adopted .

The degree o f reaction wi thin each tube of a series was

graded between 0 and 4 . In the case o f agglutinat ion tests

0 = no agglutination

1 = 25% "

2 = 50% "

3 = 75% "

4 = 100% "

For comp lement fixation tests

0 = 100% haemo lys is

1 = 75% "

2 = 50% "

3 = 25% 11

4 = no "

This degree o f reaction in each tube o f the tes t series was

recorded . Thus for a series in which the first two tubes exhibi ted

a reaction of 4 and the third tu�e showed a reaction o f 2 the

notation used was 442 .

To s imp l i fy the expres s ion o f higher t i tres an index figure

was used to indicate that a par t i cu lar reaction occurred in a series

o f consecutive tubes . e . g . The series , 444443 may be expressed

as 45 3 where the first five tubes had a reaction o f 4 and the

sixth t ube had a 3 reaction . Us ing thi s system i t i s under­

s tood that the reader is fami l i ar wi th which d i lutions were made

5 7 .

in a part icu lar tes t . For a l l agglutinat ion tests serum-di lutions

were 1 : 10 , 1 : 20 , 1 : 40 , 1 : 80 etc . For the warm f ixation CFT

d i lutions were 1 : 4 , 1 : 8 , 1 : 16 e t c . For the co ld fixa tion CFT

d i lutions were 1 : 10 , 1 : 20 , 1 : 40 e t c .

U s ing the resu l t notation descr ibed above some examp les o f

equiva lent t i tre va lues are : -

SAT 4 23 ( 443 ) = 3 a t 1 : 40

Warm CFT 442 2 a t 1 : 64

Cold CFT 46 4 at 1 : 320

It shou ld be noted that when referring to agglutination

tests the d i lution f igure refers to the proport ion of serum in the

f inal test reaction volume . In the case of the comp lement

f ixation test the d i lu t ion figure refers to the serum d i lution

before o ther reagents such as ant igen , comp lement and haemolytic

sys tem are added .

O ther Sera

A select ion of o ther sera was avai lable from known infected

animals and this was used on occas ions in compari sons o f various

t e s t s . A l aboratory standard serum , of known t i tre with respect

to the 2nd International S tandard Anti -Bruce l la abortus Serum

( 2nd ISAbS ) , was included in each series o f test s .

Re sults

Microbio logical Examinat ion

No growth appeared on any of the Bruce l la se lective media

p la te s . Various l ight mixed growths were observed on b lood-agar

p la tes and a range of organisms were identified including !· co l i ,

Klebsie l l a , S treptococcus , Pseudomonas and Corynebacterium

pyrogene s . Again no Bruce l la were identi fied .

Guinea Pig Inoculation

At necropsy no abnormal i ties were seen in any of the guinea

p igs . S p l een culture reve aled no growth on either b lood-agar or

B ru ce l la selective media . Serum agglutination and comp lement

fixation te s t s on the serum - taken immediate ly prior to death were

negative .

Sero logical Tes t s

( i ) Titration of ant igens

Because commercial ly prepared antigen is pre served in

pheno l - saline , ce l l wal l s o f the organisms tend to be

relative ly re s i s tant to di srupt ion , thus freshly

58 .

prepared organi sms were used for u l tra- sound and e ther­

water treatments . Freshly prepared ant igen (Br . abortus

s train 19 ) was f irst - t i trated agains t a l aboratory

standard serum and a " s tandardi sed concentrate"

prepared . A port ion - o f thi s concentrate was then

treated and t i trations o f this treated antigen were

compared with untreated ant igen for use in comp lement

fixat ion te sts .

Tab l e I indicates the t i tration of ant igen against a

standard serum to f ind the d i lution at whi ch 50%

agglutinat ion was g iven against a 1 : 500 d i lution of

the s tandard serum .

1 : 16 was the d i lut ion of ant igen required for use in

the serum agglutination test and for preparat ion of

s tandardised concentrate the antigen was d i luted 1 : 1 . 6

with PBS .

( i i ) Comp lement fixation test t i trat ions

Tab les II , Ill and IV show comp lement f ixation test

ant igen ti trations for untreated , ul tra- sound treated

and e ther-water treated antigens .

I I

Table I S train 19 ant igen t i tra tion for agglutination tes t ,

against l aboratory s t andard serum ( equivalent to 2nd ISAbs ) .

Serum d i lution 1 : 300 1 : 400 1 : 500 1 : 600 1 : 700

Antigen d i lution

1 : 10 2

1 : 1 2 3 4

1 : 14 3 4 1

1 : 1 6 4 4 2 1

1 : 18 4 4 4 3

Tab le II . Who le ce l l s train 19 ant igen , t i trat ion for warm

fixation CFT .

Serum d i lution 1 : 4 1 : 8 1 : 16 1 : 32 1 : 64 1 : 1 28

Antigen d i lution

1 : 100 4 4 4

1 : 200 2 4 4 4 . 3

1 : 400 1 4 3

1 : 800

1 : 1 600

59 .

Table I I I . Sonicated s train 19 ant igen , t i trat ion for warm fixation CFT .

Serum d i lu t ion 1 : 4 1 : 8 1 : 16 1 : 3 2 1 : 64 1 : 1 2 8

Antigen d i lution

1 : 50 4 4 4

1 : 100 4 4 4 2

1 : 200 1 3 4 3

1 : 400 1 3 4

1 : 800

1 : 1600

Table IV . E ther-water treated strain 19 antigen , titrat ion for

warm fixation CFT ,

Serum d i lut ion 1 : 4 1 : 8 1 : 16 1 : 32 1 : 64 1 : 1 2 8

Antigen d i lut ion

1 : 5 0 4 4 2

1 : 100 4 4 4

1 : 200 4 4

1 : 400 2 3 2

1 : 800

1 : 1600

. 60 .

6 1 .

( i i i ) S ero logical resu l t s

Tab le V d isplays results obtained for the agglutinat ion

type test s for sera from each o f the repeat BCT+/CFT­

animal s s laughtered .

Complement f ixat ion tes ts resul ts are given in Tables VI ,

VII and VIII for tests us ing s tandard commercial s train 99 ant i gen ,

sonicated s train 19 and e ther-water treated strain 19 ant igens .

A l aboratory s tandard serum with equivalent comp lement f ixing

activity to that o f a 1 : 30 di lution o f the 2nd International

S tandard Anti-Bruce l la abor tus Serum ( 2nd I SAbs ) was inc luded in

each series of tes t s .

The Rivanol comp lement f ixat ion te$t was negative for a l l

samples in the series 1 - 20 except for numbers 1 and 20 , see

Tab le IX .

Ge l -di ffus ion tests were conducted u s ing both sonicated and

ether- treated ant igens . Of the twenty te s t sera (Numbered 1- 20 )

only serum 1 showed a precipitat ion l ine . Both sonicated and

e ther treated ant igens cou ld produce this s ingle l ine which

appeared to be common to them both . The l ine was a lso produced

with high t i tred s era from known infected animals . Upon d i l ut ing

serum 1 to 1 : 10 l ines d isappeared .

The indirect haemo lys i s test was negat ive for sera 2- 19 .

Serum 1 gave a t i tre o f 46

Serum 20 gave a t i tre o f 4 2· and the labor�tory s tandard gave 4 •

Table V . Ti tres to agglut ination

animals s laughtered .

Animal No . BCT SAT

1 +H- 47 2 2 ++ 41

3 ++ 4

4 4 2

5 + 42 2

6 4 2 7 + 4 2

8 ++ 432

9 43

10 + 4 2

11 4 2

12 43

13 43 1

14 + 2

15 4 2

16 +++ 42 3

17 ++ 43

18 + 42 2

19 ++ 42 31

20 +++ 43 2

( 1) method ( a) see text

( 2 ) method (b ) see text

Riv

47

2

4

3

4

3

2

3

4

6 2 .

type tests o f repeat BCT+/CFT-

ME ( l ) ME ( Z ) Coombs

47 2 ,_/1 4 1 3 2 4 2 41 41

3 3 4 2

21 1 4 2

421 42 42 2

4 2 4 4 2

43 4 42 2

43 43 432

43 3 4 2

4 2 43 42 3

31 3 4 2

4 2 2 42

43 42 4\

2 1 4 2

4 2 4 4 2

422 43 42 3

43 4 43

43 43 42 2

423 43 4 231

44 43 . 45

6 3 .

Tab le VI . Comp lement f i xa t ion ' te s t results for repeat BCT+/CFT-

animals .

( a ) Co ld fixation: commercial s train 99 antigen

Animal No . Antigen concentrat ion

1 : 100 1 : 200 1 : 400 1 : 600

1 47 2 47 3 224 7 _ 742 3

2

3 4

5 6 2 1 7 2 41 2 1

8 2 2

9

1 0

1 1

1 2

1 3

1 4

1 5 1 6 1 3 3 l

1 7

1 8

1 9 2 3 4 1

20 43 432 42 1 1 3

S td ( l / 30 ISAb S ) 4 42 43 4 2

( b ) Warm fixation : comme r i cal s train 99 antigen

Animal No , 1 : 100 1 : 200 1 : 400 1 : 600

*

1 -4

1 32 22 -41 2 332 _5 2

332 -5 1 33 1

20 32 4 2 4 2 43

S td ( l/30 ISAb S ) 43 42 42 42

* S era 2- 1 9 al l negat ive .

Tab l e VII . Comp lement f ixa t ion test results for repeat BCT+/CFT­

anima l s .

( a ) Co ld fixation: sonicated s tra in 1 9 antigen

Animd No . Ant i gen concentra t i on

�'r 1 : 20 1 : 5 0 1 : 100 1 : 200

i 46 2 46 3 2 245 3 - 2 3 24 33

7 3 2 43 43 3 2 2

8 2 1 2

1 9 3 1 4 4

20 3 43 1 4 2 1 42 2

Std ( 1/ 30 ISAb S ) 4 1 42 4 2 4

( b ) Warm f ixa t ion : sonicated strain 19 ant igen

Animal No . Ant i gen concentration �'r 1 : 20 1 : 50 1 : 100 1 : 200

1 - 5 34 -5 2 1 -\ 2 2

20 4 2 3 2 1 43 23 2 2

S td ( 1 / 30 ISAbS ) 4 4 1 4 2 42

* Resu l t s for sera no t l i s ted - a l l negat ive .

64 .

Tab le VIII . C omp l ement fixation t e s t resu l t s :

Use o f e ther treated , sonica ted and who le ce l l ant i gens in

tes ting of pro zoning sera from known infected animals .

Samp l e No .

S4 982

13846

A5 639 684

5 98

1 65

199

1399 1

E ther Aij 1 : 100

WCFT''r CCFT '0'r

20 20

20

- 5 243 -5 34 2 3

1 346 48

s o n 't- e. � u a A! 1 : 100

WCFT CCFT

20

20

- 2 146 2

_ 3346

45 3

- 2 3443

_ 347

48

- 243

wn� h �u l. A a 1 : 400

WCFT CCFT

20

20

- 3 234 2 2 345 3

-4

2343 2 347 3

49 4 1 0

43 3

S td ( l/ 30 ISAbS ) 4 2 1 4 2 4 1 4 3 4 2 4 2

,•: WCFT �df CCFT

= Warm f ixation , comp lement fixation tes t Cold f ixa t ion , comp l ement fixat ion test

Table IX . Rivano l comp l ement f ixa t ion te s t commerc i a l s train 99

antigen 1 : 400

weFT'' . ; CCfT'h'r

S erum No .

l _5 2 23 2 2 246 2

20 4 3

''r . WCFF Warm f ixation , Gomp lement fixation test ' ·

** CCFT = C o ld f ixation , comp l ement f ixation te s t

65 .

Discus sion

The antigenic re lat ionships o f Bruce l la to other organisms

have been reported on by Harris ( 1 950 ) , S p ink ( 19 56 ) , Berman

( 1 95 6 ) , Nico letti and Ho 1mes ( 1968) , Ahvonen � � ( 1969 ) and

o thers . Al though Pas ture l l a , Sa lmone l l a , Shige l la and Vibrio sp .

have a l l been imp l icated, only Yers inia enteroco l i tica type IX has

cons is tent ly shown s trong cross-reactivity (Ahvonen � � . , 1969 ;

Corbe l and Cul len , l970 ) . I t was cons idered that in the absence

of B r . abortus infect ion it may have been some other agent

s t imulating a cross -reacting antibody that caused the card test

to react but did not fix comp l ement .

In the event it transpired that nei ther Bruce l la abortus nor

any o ther l ike ly non- specific agent cou ld be iso l ated from any o f

the twenty cat t l e examined . A series o f sero logical tests was

conducted in an at tempt to c lar ify the s i tuat ion .

6 6 .

Animals purchased for this experiment were selected o n the

bas i s of having reacted to previous card tests but not to previous

comp lement fixat ion tests . Herds from which they came genera l ly

had a long testing his tory al though in some cases the actual

number of reactors taken over the series o f tests was very low .

The prevalence o f such repeat BCT+/CFT- animals in infected herds

and in non- infected herds is not known . This i s because herds

are not necessar i ly subj ected to card testing at each test and

herds which become accredited are not s ubj ected to any further

test ing , thus there i s no opportuni ty to app ly the card te s t on

a repetitive bas i s in non- infected herd s . Timbs � � . ( 1978a)

investigated the subsequent testing his tory of a group o f these

animal s and concl uded that the probab i l i ty o f any animal eventual ly

becoming CFT pos i t ive was not re lated to its previous card testing

his tory . The prob lem o f repeat BCT+/CFT - animals is probab ly not

unique to New Zealand a l though it has not been recorded e lsewhere

as be ing of any s i gnificance . A further considerat ion of this

matter is made in chapter 7 .

Fol lowing ini tial preparation , the s train 19 ant igen was

t itrated in a checkerboard fashion (Table I ) to enable i t to be

s tandardised prior to al iquots being taken for e ther treatment

and sonication . In this way th� : antigenic activity was equated

to the commercial s train 99 antigen used for the s tandard

agglut inat ion and comp lement fixat ion tests . Tab le II compares

complement fixation test antigen t i trat ions o f commercial s train

99 ant igen and the freshly prepared s train 19 antigen . (Both

ant igens having the s ame agglut ination activi ty) . Thus , al though

strain 19 antigen is a l i t t le less sens itive than s train 9 9 i t

remains quite suitab l e . Tab les I l l and I V compare the effective­

ness o f sonicated and ether treated s train 1 9 ant igens ; there

appears to be l i t t l e d i fference in the sens i t ivit ies of these two

antigens and in each case a di lution o f 1 : 100 was taken as the

optimal d i lut ion to use in the CF te s t .

Sonicated and e ther-water treated antigens were used in the

CFT spec if ica l ly to de termine if prozoning could be avoided by

altering the type o f ant igen emp loyed . As indicated by resu lts

i l lus trated in Tables VI , VII and VIII ne i ther type of soluble

antigen appears to o f fer any advantage over the s tandard whole

67 .

ce l l ant igen . Ether-water treatment o f pheno l-preserved commercial

who le ce l l strain 99 B r . abortus antigen proved to be quite

succe s s ful . On ti tration the comp lement fixing activity o f this

extract proved to be equivalent to that obtained from treatment o f

freshly prepared ce l l s .

The outs tanding features o f the resu l t s obtained in this

sect ion are the high ti tres given by serum from animal No . 1 .

When the ini tial comp lement fixati0n tests were performed on this

animal ' s serum , prior to it be ing selected for this experiment ,

i t was thought to be negative . This may have been because o f

true absence o f t i tre or because there was such a high degree o f

prozoning that no complement was f ixed . Al though Br . abortus

was not isolated , the sero logical resu l t s indicate that this

animal may wel l have been infected . The nature of the CFT

react ion was such that it required further invest igation and this

formed the bas is of s tudies out l ined in chapter 5 .

Al though serum 20 had a low CF t i tre , the marked reduct ion �

fo l lowing Rivanol treatment , in both the CF and agglut ination

tests , indicated that the t i tre may have been due to res idual

vaccinat ion ( IgM) ant ibodies (Morgan , l967 ) .

A l l other sera in thi s series were CFT negat ive and were

regarded as having insignificant agglutination ti tres .

Two mercapto- ethanol methods were compared wi th the Rivanol

t e s t . I f it is accepted that animal s 2-20 were not infected and

that the agglut ination t itre present was due to residua l s train

. 68 .

1 9 stimulated IgM or a cros s react ing antibody , then i t appears

that the Rivanol test is best at clarifying the s i tuat ion as i t

markedly reduced the SAT t i tre s . The mercapto-e thanol test

performed us ing buffer containing O . OSM 2-ME appeared to be more

e ffective than the method of us ing O . lM 2-ME in the first tube

prior to making dilut ions . Despite the extra time required to

perform the Rivanol tes t i t proved to be superior and a lso avoided

the use of mercapto-e thanol which is dangerous to use and has a

mos t unp leasant pungent odour .

The high Coombs t i tre of serum l was notewor thy ( 1 : 8 1920 ) ,

t i tres of this order are often seen in other s trongly prozoning

sera .

Ge l diffus ion tes t s were conducted in an effort to detect any

abnormal pat tern o f ant ibody di ffus ion that serum l may have had .

The single l ine that was detected was identical to a l ine produced

by another serum from a known infected cow . No abnormal pat tern

was seen . No other sera in the series �roduced any detectab le

l ines . Only serum 1 reacted to the indirect haemolys is te s t and

no evidence o f prozoning was seen . This test was specifica l ly

d eve loped to overcome the problem of prozoning in the complement

fixation test (Placket � � &. , 1976) .

CHAPTER 3 . EXPERIMENTAL INOC��TlON OF CATTLE WITH KILLED

BRUCELLA ABORTUS .

Introduct ion

69 .

Data co l lected and evaluated in chapter 7 ind icated that

animal s BCT pos i tive yet negative to the CFT were more prevalent

in infected herds than in non- infected herds . A possib le

exp lanation for the phenomenon i s that animal s exposed to the

ant igen but which re s i s t infect ion , deve lop a transient BCT t i tre

but do not react to the CFT .

Cul l en and Corbe l ( 19 70 ) inoculated adu lt mi lking cows , that -

had been vaccinated as calve s , with various doses o f l iving and

dead s train 19 Br . abortus . A rapid r i se in t i tre was noted .

Titres achieved maximum leve l s after two weeks and began to fa l l

again after one month . A dose/response e f fect was noted

sugge s t ing that the vaccine was act ing par t ly as a non- infect ing

ant igen . Beck � �· ( 1964) conducted s imi lar exper iments with

s imi lar re sults and s tressed the importance of us ing properly

cleaned vaccinat ion equipment .

The obj ect ive o f the experiment reported in thi s section was

to examine the serological re sponse to ki l led Br . abortus by cows

whi ch had been vaccinated as calves but which did not have any

detectable t i tres to the SAT , CFT or BCT at the commencement o f

thi s invest igat ion .

Materials and Methods

Animals : N ine cul l dairy cows were available for use , these

animal s were a l l more than e i ght years of age and had a l l been

vaccinated wi th s train 19 �· abortus as calve s .

negative t o the SAT , CFT and BCT .

They were a l l

Inoculation : - The contents · o f a vial o f commercia l s train 1 9 *

Br . abortus vaccine - were reconstituted with d i s t i l led water and

d i luted with the aid o f Wel lcome opac i ty tubes+ to final ce l l 9 6 concentrations o f 1 x 10 . and 1 x 10 organisms/ml . The d i luted

* I . C . I . Tasman Vaccine Ltd . , Upper Hut t , New Zealand .

+ Wel lcome Research Laboratories , Beckenham, Kent .

7 0 .

organi sms were then heat ki l led in a water bath at 80 °C for

60 mins and checked for s ter i l ity by inoculating onto b lood-agar .

Two cows were each given 1 x 109 ki l led organisms by sub-6 cutaneou s inj ection and one further cow rece ived 1 x 10 organi sms

s/c . Three other animals were each dosed by spraying 1 ml o f a

1 x 109 organism suspension into one eye . This was achieved by

forcing i t through a 26G syringe needle which had been f lat tened

to cause a fine spray .

The final three cows each rece ived organisms intra-nasally by 9 spraying 1 ml of a 1 x 10 organisms/ml suspension onto and into

the external nares .

Animals were b led at approximately weekly interva l s fo l lowing

inoculation and a second dose , by the same route as the fir s t , was

given after eight weeks . 0 Sera were col lected and store d at - 20 C until the las t b lood

co l lect ion was made . Al l sera were then tested together as one

batch . Card , serum agglutinat ion and complement fixation tests

were per formed as described in chapter 3 .

fixat ion CF tests were carried out .

Resu l t s

Both warm and co ld

BCT , SAT and CFT t i tres for the nine cows before and after

inocu lat ion with ki l led s train 19 vaccine are shown in F igures 3a ,

3b and 3c .

As indicated by the hi stograms , agglut ination t i tre s tended

to be read i ly produced whereas complement f ixation and card test

titres were rather low and transient . T i tres to a l l t e s t s reached

peak leve l s within 14 days of the ini tial inoculat ion . Fol lowing

the repeat dose t i tres again rose but general ly did not exceed 1

those obtained after the init ia l inoculation . Highes t t i tres

were s t imul ated by subcutaneous inoculation . Conj unct ival and

nasal routes of administration resul ted in only low t i tre s .

I I

Figure 3 Histogram of t i tres given by cat t le

inoculated with ki l led Bruce lla abortus

s train 19 .

(a ) Cows inoculated subcutaneous ly

Nos . 5 7 & 70 1 x 109 organisms

No . 9 2 1 x 106 organisms

BCT +++

++

+

Figure 3 Hi stogram o f t i tres given by cattle -

inoculated with ki l led Brucel la abortus

s train 1 9 .

(b) Cows inoculated via conjunctiva .

I

+++ BCT ++

+

36 , 62 , 95 I I I

A 1 : 1.0 SAT 1 : 20

<1> 1.... 1 : 10

........

..... 0 I I

1 :1.0 CFT (cold fixation) 1 :20

1 : 10

0 36, 62 I I I I N N N (.[) w en

(J1 ln ln en en

Date

. I

I I I I _.. N w ':--1 en w 0 en Cl Cl '-J

A of sampling

36

� I I I

I I I _. .t'- _.. .t'- . CX> � CX> CX>

I _..

c.D

f I

Figure 3 Histogram of ti tre s given by cattle

inoculated with ki lled Bruce l la abortus

s train 19 .

( c) Cows inoculated via nasa l route . '

Q) '-

+++

++

+

1 :40

1 :20

1 : 10

0

1 :40

1 : 20

1 : 1 0

0

BCT

1 . 25 , 63 I I I I 4

SAT 25

/:::: I I I I I 4

I I

CFT (cold fixation )

25

63

� I I '

63 25 � � � I I I I I I I I I I I I ..... _. NI:X N U) ..... N w ':'I _. � N tp � . C1l w 0 l'

tn C1l C1l C1l en -...] CO Vl VlVl C1l "" A Date of sampling

1 , 25,63 I I ..... _.

CO CO U)

I ;

7 1 .

Discuss ion

Al though the adminis trat ion of organisms subcutaneous ly was

satisfactor i ly achieved the same cou ld not be said o f the

conj unctival and nasal routes . When spraying ef the suspension onto the conj unctiva was at tempted , the cows tried to avoid the

spray by turning the ir heads and clos ing their eye l ids . Attempts

at intra-nasa l spraying resul ted in s imi lar resentment wi th the

added comp l icat ion of fierce exp iration . Consequent ly somewhat

l e s s than the ful l dose of 1 x 109 organisms was rece ived by

animals inoculated by these routes .

S ince a l l animal s had been vaccinated as calves wi th l ive

s train 19 Br . abortus vaccine the r i s e in t i tre fo l lowing

inoculat ion wi th dead organisms was not altogether unexpected .

Animal s inoculated subcutaneously gave the mos t s igni ficant

t i tre enhancement , as might have been expected . There was a

suggest ion

received 1

animal No .

o f a dose-response e ffect in that animal s

X 109 organisms gave higher t itres in a l l

3 which received only 1 X 6 . 10 organisms .

1 and 2 which

tests than

Animals treated intra-nasa l ly and intra- conj unctiva l ly d id

not respond we l l sero logically but the observation that some

animal s did in fact respond , particu lar ly to the SAT , supports the

hypothesis that sensi t i sed animal s natura l ly exposed to dead

organisms , may deve lop ti tres . T i tres rose rap idly to achieve

maximum l eve l s after two weeks ; fo l lowing the repeat dose o f

organisms mos t fe l l to pre- inoculation leve l s wi thin two months .

The response fo l lowing repeat inoculat ion was no greater than that

s een after the first dose which init iated the anamnes tic response .

Beck � � .( 1964) and Cul l en and Corbe l ( 1970) inocu lated cows

to tes t the e ffect o f us ing syringes contaminated wi th s train 1 9

Br . abortus on subsequent titre deve lopment . They were concerned

with the SAT and CFT response and did not investigate the pos t­

inoculat ion deve lopment o f BCT t i tre s . This experiment has

d emonstrated that BCT t itres may be s t imul ated by ki l led B r . abortus

organisms and maintained even after significant SAT and

CFT titres have d isappeared . Further inves tigation o f thi s aspect 1

us ing a f i e ld s train o f Bruce l l a and dosing via the oral route i s

required .

CHAPTER 4 . STUDIES ON BRUCELLA-SPECIFIC SERUM ANTIBODY

FRACTIONS

Introduct ion

The "Prozone" phenomena which occurs in Bruce l la complement

f i xat ion t Q & t & haa b&en de s GribQd ifi ChAP t9� z .

7 2 .

Because the automated comp lement f ixation tes t , on which the

New Zealand Bruce llosis Eradicat ion S cheme is based is es sentia l ly

a one dilution te s t , i t has been critici sed �n the grounds that i t

may fai l to detect prozoning sera which do not react at the

particu lar di lut ion of the automated te s t sys tem. One of the

obj ectives o f thi s thesis is to j us t i fy the use o f the automated

CFT and to show that it can and does detect such prozoning sera .

Ant ibody produced in response to B ruce l la abortus infect ion

or Br . abortus s train 19 vaccination is found in the IgG1 , I gG2 and lgM immunoglobul in c lasses (Rice and Boyes, l97 1 ; Beh, l974) .

The re lat ive proport ions o f each of these immunoglobul in classes

in the sera of infected and vaccinated animals may vary (Beh and

Lasce l les , l97 3 ) . I t has a l so been shown that the various

sero logical tes t s used in the diagnosi s of bruce l lo s i s vary

considerably in their abi l ity to detect antibody of a particular

immunoglobulin c lass (Beh 1 9 7 3 ; Al lan � al . , 197 6 ) .

The experiments out l ined in this section were undertaken to

provide an unders tanding of the immunological bas is of prozoning

in the CFT . This was achieved by us ing the procedure o f Nash

and Heremans ( 19 6 9 ) for the quantitative determination of antibody

be longing to the various immunological c lasses and re lating

findings to comp lement fixat ion t i tres of sera under tes t ,

Further understanding o f the pro zone phenomenon was gained by

conducting compl ement fixat ion tests on serum fract ions and

mixtures o f serum fract ions derived from prozoning sera .

Materials and Methods

Ion Exchange Chromatography

A 30 cm x 1 . 5 cm diam . column o f diethylaminoethyl ce l lulose

* (DEAE) was prepared and equil ibrated with 0 . 2M phosphate buffer

pH 8 . 0 . Two mls serum or bovine colos trum which had previous ly

been dialysed against thi s buffer for two days was introduced .

7 3 .

The co lumn was eluted with starting buffer and then wi th a gradient

to 0 . 6M phosphate buffer pH 8 . 0 us ing a Varigrad Chamber+. The

type of e lution gradient obtained is presented in Figure 4 .

Protein e lut ion was monitored by continuous ly recording the

absorbance , at 254 nm, of the e luate . A typical trace is shown in

Figure 5 . Fract ions of e luate were col lected , poo led where

required , and concentrated by d ialysis against po lyethylene glyco l

(M . W. 20 , 000 ) .

Ge l Fil tration

Sephadex G- 200 beads were swol len , degassed and poured into

a 60 cm x 2 . 5 cm d iam . co lumn . Fo l lowing equi libration with

0 . 015 M phosphate buffered saline pH 7 . 2 , 1 ml of previous ly

dialysed serum in which 0 . 2 g sucrose had been di s so lved was

layered onto the top o f the ge l . Protein e lut ion was monitored

by recording the absorbance of the buffer at 254 nm and fract ions

were concentrated by d ialysis against polyethylene glycol .

A typical absorbance pattern i s shown in Figure 6 .

Immunoe lectrophores i s ( IEP )

IEP agar was prepared as described in appendix I . Immuno-

electrophoresis was performed against rabbi t anti-bovine who le

serum and against rabbit anti-bovine immunoglobulin .

IEP p late is shown in Figure 7 .

* DE-52 Whatman , England .

+ · Varigrad . Bachler Instruments , Fort Lee , N.Y .

A typical

Figure 4 E lution gradient of phosphate buffer

used in DEAE cellulose column

M 0 ...-)(

(/) 0 ...c :L ::J.

>. ......

:� ..... u :J

' · ' 'U c 0 u

3 0

2 0

10 8 6 4 2

4 8 12 1 6 2 0 24 28 32 36 40 44 48 52 56 60 64 68 72 76 1 80 84 88

Fraction number

0 · 6 M

0·02 M

Figure 5

Figure 6

Typ ical trace o f absorbance at 254 nm o f e luate o f DEAE cellulo se co lumn

Typical trace o f absorbance at 254 nm

o f e luate from sephadex G200 co lumn .

Fract ion numbers indicate location o f

eluate fract ion taken for e lectrophores i s

a s d i splayed in figure 7 .

I I I

I IT m IV V VI Fraction number

Figure 7

I

Typical e lectrophores is plate for- - serum

fract ionated on DEAE - cellulose

( ) I \

I II III IV V VI I VII

Number s refer to fractions taken as indicated in figure 5 .

Troughs contain rabbi t anti-bovine whole serum, wells contain sample

o f appropriate fraction .

/•

74 .

Immunodi ffus ion

Immunodi ffus ion agar was prepared as described in Appendix I ,

this agar was ident ical to that used for IEP . Immunodi ffus ion was

used where appropriate to detect contamination with unwanted

e l as s e s o! immunoglobu l in . I e was part icular ly . usefu l in moni to r ing

cross adsorption of speci fic antisera .

Preparation of Antisera

Solut ions of approximate ly 1 - 2 mg ./ml of specific immuno­

g lobul ins were prepared . 1 ml of each of IgG1 , IgG2 and IgM was *

emulsi fied with 1 ml Freund ' s complete adj uvant and inj ected in

two equal doses into each hind leg of each of three rabbits .

Further inoculat ions of immunoglobul in were given two and

s i x weeks later . Freund ' s incomp lete adj uvant was used to emu l s i fy

the immunog lobu l in preparat ion a t tho se later trea tment s . Rabb i t s erum was harvested a t two , four and six weeks fo l lowing the final

inoculation and preserved with 0 . 1% sodium azide .

Anti I gG1 serum was absorbed by s lowly adding IgG2 unt i l no

Anti IgG1- IgG2 precipi t in l ine could be detected on immunodiffus ion ,

l ikewise Anti IgG2 serum was absorbed with IgG1 and Anti IgM was

absorbed with IgG1 • In each case 0 . 005 mg. of specific immuno­

globulin per ml of anti serum was added approximate ly every hour

unt i l absorption was comp lete . Approximate ly 0 . 05 mg of specific

immunoglobul in was required per ml of antiserum in each case .

Quantitation of Immunoglobul ins

Approximate estimat ions of immunoglobul in concentrations

were obtained by measurement of absorbance of solutions at 280 nm

in an ul tra vio let spectrophotometer and reading agains t a

s tandard curve .

Standard so lutions o f immunoglobul ins were prepared by

taking a pure solut ion o f immunoglobul in of unknown concentration

and d ialys ing for five days against frequent changes of 0 . 05 M

ammonium carbonate . The immunoglobul in preparation was then

* Difco , Detroit .

75 .

distributed into pre-weighed vials and freeze-dried . Following

lyophi l isation the vials were again weighed and the quantity of

dried immunoglobulin material calculated . This mater ial was then

recons t i tuted wi th phosphate buffered sal ine to the required

concentration .

S ingle Radial Immunodiffusion (SRID) tests

Quantitative single radial immunodiffus ion tests were

performed accord ing to the method of Mancini � �. ( 1965 ) . Plates

were incubated for two days and then washed for three days in

sal ine and for a further day in distilled water . They were then

dried and stained with Amido b lack and decolourised before ring

diameters were read ( see Appendix I ) . Each plate contained four

s tandards from which a standard curve was prepared . I f any samp le

gave a ring diameter which fe l l outside the range o f the s tandards

it was concentrated or d i luted and rete s ted .

Adsorption of Bruce lla Spec i f ic Antibody

A standardised concentrate of Brucel la abortus antigen

( s train 99) was centrifuged at 4000 g for 20 minutes to deposit

ce l l s . These were resuspended to one- tenth of the original

volume of concentrate in phosphate buffered sal ine ( PBS ) . This

super -concentrated antigen suspens ion was used to a dsorb test

sera .

Fo� each serum a series of four tubes was prepared as

fo llows :

Serum Antigen PBS (ml ) (ml ) (ml )

T . l 0 . 2 0 . 2 0 . 6

T . 2 0 . 1 0 . 2 0 . 7

C . l 0 . 2 o . s C . 2 0 . 1 0 . 9

0 After incubation at 4 C for two days tubes were centrifuged

at 4000 g for 20 mins and the supernatants removed and tested by

SRID.

The amount of specific immunoglobul in measured by each test

was calculated by plotting the concentration of each s tandard

Figure 8

I

. - :::. ' ,... . '

Typical s ingle radial immunodiffus�n

p late

..

·'

7 6 .

immunoglobul in agains t the square of i t s corresponding precipit in

ring d iameter , concentrations of unknown samples were read d irect ly

from this chart . Five sets o f four samp les (T 1 , T2 , c 1 , c 2 ) and

four s tandards were included on each p late .

After determining the immunoglobul in concentration for each

s amp le , Bruee l l a spee i f i e p ereentage s were ealeul a ted a s fo l lows :

Ig cone . o f absorbed serum [r] 2[T2] + [T l ] 2

Ig cone . o f control serum [CJ = 2[c2] + [c l ] 2

Bruce lla specific % of serum = [c ] - [T] X 100

[c]

Inhibi t ion of Specific IgG1 Comp lement Fixing �ctivi ty by

Specific IgG2

Aliquots of IgG1 and IgG2 fractions of a s trongly prozoning

serum , were taken from a DEAE co lumn as described in previously

and were each adsorbed with Br . abortus antigen us ing the method

given in above .

Protein concentrations of adsorbed and unadsorbed a l iquots

were measured by the Fol in-Ciocal teu method ( Lowry � �. , 195 1 ) .

The concentration of speci fic IgG1 and IgG2 was then determined

by cal culating the difference between test and contro l samples .

A series of test samp les was prepared such that each

contained a known quantity of anti-Brucel la specific IgG1 and

Complement fixation tests were then performed on each

sample to determine the extent to which prozoning occurred .

Results

The qual i ty of the IgG1 , IgG2 and IgM preparations recovered

from the DEAE ce l lulose and . sephadex co lumns was monitored by

immunoelectrophoresis and immunodiffusion .

preparations were general ly of a high quality and after cross

7 7 .

adsorption to remove anti- light chain antibody were rendered

sub-class specific . IgM preparat ions were a l so adsorbed with

IgG1 to remove anti - l ight chain antibody . Most IgM preparations

were found on electrophores i s to be contaminated with a2 macro•

gleb\iUn .

Only pure preparations were u sed to s t imulate antiserum

production .

Immunoglobul in Standards

Dialysis against ammonium carbonate fo l lowed by lyophi l i sat ion

proved to be a very satis factory method o f estimating the protein

concentration of various immunoglobul in fract ions . Al l vials into

which a control samp le of dialysed PBS was p laced showed no weight

gain thus indicating that dialysi s had been complete .

SRID Test s on Adsorbed and Unadsorbed Sera

. Tab le X indicates IgG1 , I gG2 and I gM concentrat ions for each

of fourteen sera tested .

and IgG2 are also given .

Bruce l l a specific percentages for IgG 1 The "prozone" index number indicates

the number of d i lutions in which - prozoning occurred in the warm

fixation CFT . The figure in the ti tre co lumn indicates the

number o f doub l ing d i lutions in w4ich react ions were obtained , I thus a figure of four means the titre i s 1 : 32 (where 1 : 4 i s

the serum d i lution in the first tube of the series and in the

fourth doubling d i lution the t i tre is 1 : 32 .

An assessment o f the degree o f error occurring within the I

SRID test system was made by calculating the correlat ion

coefficient ( r ) of the sets of results obtained for [c1] [c2] and (T l] (T2] where

For

For

(T 1] = 2(T2] .. .

(t 1] (T2] " [cl] [c2J

r

r

and [c l ] = 2(c2] = 0 � 88

= 0 . 93

Serum No .

. 1

2

3

4

5

6

7

8

9

10

1 1

1 2 , .

1 3

14

Table X . Immunoglobul in concentration and Bruce l la specific antibody percentage

related to degree of prozoning of warm CFT t i tre

Immunoglobulin Cone . (mg/ml ) IgGl IgG2 I gM

10 . 6 10 . 2 2 . 4

10 . 9 10 . 5 1 . 5

1 1 . 7 10 . 9 2 . 7

1 1 . 4 1 1 . 8 1 . 8

1 1 . 4 1 2 . 6 4 . 6

8 . 2 8 . 7 2 . 5

10 . 6 1 2 . 5 2 . 0

14 . 1 18 . 6 2 . 6

6 . 3 6 . 8 1 . 6

1 3 . 2 16 . 6 2 . 4

10 . 8 16 . 0 3 . 0

9 . 3 15 . 5 3 . 3

7 . 5 7 . 8 3 . 1

8 . 2 4 . 75 2 . 2

Bruce l la specific IgG "/. IgG 1 2

10 . 8 1 2 . 2

1 . 0 8 . 5

. 9 3 . 7

13 . 6 1 3 . 5

8 . 1 0

0 0

9 . 0 0

3 . 6 10 . 8

24 . 8 33 . 6

10 . 4 9 . 3

8 . 3 5 . 5 I

3 . 5 2 . 2

0 0

0 0

.

* Sera gave no t i tre at a l l to warm fixation CFT - very s trong prozone

Prozone Index

4 -

-

* -

-

-

-

* *

3 ·

2

-

-

Titre

10

2

4

9+

9

3 -

6

10+

9+

9

7

5

2

+ ' Titre given is for co ld fixation CFT , and indicates the number o f doubl ing dilutions tto which comp lete fixation was given

" (X)

I

Serology on IgG1 and IgG2 Fractions and Combinations of IgG1 and I gG2

79 .

Table XI shows concentrations o f total and specific IgG1 and

IgG2 , prepared as described previous ly and estimated by the

methed e f LeWTy � � ( l �� l ) ,

Table XII gives typical results for warm fixat ion CFT ' s with

varying proportions of specific IgG1 and IgG2 •

Increas ing the proportion of specific IgG2 induced prozoning

and eventua l ly inhibited a l l complement fixation . Co ld fixation

tests were not as sens it ive to IgG2 in terms of prozone formation

but were neverthe less affected .

Reactions of immunoglobul in al iquots to the SAT and card test

are shown in Table XIII .

I

Table XI . Specific and total immunoglobul in concentrat ions in

al iquots of IgG1 and I gG2

Total Ig cone . Specific Ig cone . % specific mg(ml mg(ml

tgtll .9 . 8 . 6 ? 1 1 . 6 IgG2 6 . 4 . 8 2 1 2'. 8

Table XII . Results o f warm fixat ion CF test s for mixtures of

specific IgG1 and I gG2

IgG1 IgG2 1 : 4 1 : 8 1 : 16 1 : 32 1 : 64 1 : 1 28

IJ.g/ml IJ.g/ml

250 0 4 4 4 4 4

200 50 2 4 4 4

150 100 2 3

100 150

50 200

0 250

Table XIII . React ion of IgG1 and IgG2 to serum agglutination

and card tests

Immunoglobulin wt . specific SAT* Card Test

Ig mg/ml i . u . t itre

I gG1 . 67 37 2 443 +++

IgG2 . 8 2 536 45 1 +++

SAT = serum agglutination test

i . u . = international units . Second International S tandard

Anti-Bruce lla abortus serum by definit ion contains

1 000 i . u . per ml .

80 .

Ig

. I

8 1 .

Discuss ion

The exis tence of the p rozone phenomenon in the comp lement

fixation test has rece ived l i t t le at tention unt i l recently .

Although the e f fect was undoubted ly recognised by early workers i t

WA I no e d o J er i bed o r inves t i gated i n any depth . This may have

been because of its transient appearance or s imp ly that i t was not

thought s igni ficant or unders tood . Al ton and Jones ( 1 967 ) in the ir s tandard monograph on Laboratory Techniques in Bruce l losis do not

mention the e f fect al though its importance is recognised in the

second edit ion (Al ton � � . , 1975a ) .

Al ton � �. ( 1975b) in a s tudy of cul tural and sero logical

re lat ionships , obtained a pos i t ive cu l ture from an animal which

a l so gave a very s trong prozone to the CFT . They drew at tention

to the danger of us ing insufficient d i lutions when performing the

test in case such "prozoners" were not de tected .

P 1 acket t and A1 ton ( 1975 ) and McNaught � � . ( 1 9 7 7 ) have

examined various aspects o f prozoning . Both groups of workers

have noted that speci fic IgG2 i s capable o f inducing prozoning and

can eventual ly block comp lement fixat ion o f spec i fic IgG1 and I gM

entirely if present in sufficient quantity .

In the course o f the present s tudy a number of s trong

"prozoning" sera were col lected from known infected animals . One

very s trong prozoning sample was taken from an animal whi ch despite

extensive cul ture attempt s could not be proved to be infected .

These sera formed the bas is of this series o f exper iments de signed

to invest igate the respective roles o f Bruce l la specific I gG1 ,

I gG2 and IgM . B y us ing the technique of rad ial immunodi f fus ion

i t was hoped to quantify specific IgG1 , IgG2 and IgM in various

s era and relate these findings to serological t i tres .

Quant i tative measurements o f Bruce l la -speci fic ant ibody in

various immunoglobulin c lasses have previous ly been made on immuno­

g lobul ins fo l lowing fractionat ion . The only recorded s tudy of

this nature is that o f Al lan � �. ( 19 7 6 ) who a fter adsorbing the

various s erum fract ions wi th !£. abortus measured init ia l and post­

adsorpt ion protein leve l s by the method o f Lowry � !1. ( 195 1 )

using Fol in-Ciocal teu reagent . They also used a rad io - iodination

method to measure adsorbed immunoglobul in d irectly and concluded

that this was a super ior procedure to that of protein measurement .

8 2 .

Beh ( 19 74 ) used a method s imi lar to the one described in thi s the s i s to measure Bruce l la s pecific antibody concentrat ion changes

in cattle sera fo l lowing s train 19 vaccination or infect ion .

The s ingle r adial immunodiffus ion test (SRID ) i s one o f the

few methods avai lable for d i rect �y measuring specif ic classes or

sub - c lasses o f immunoglobul ins . I t s sens itivity i s claimed to be

capab le o f detect ing microgram quanti t ies of immunoglobul in

(Mancini, 1965 ) , but i t is a l so di fficult to prepare and s tandardise .

In this series o f experiments it was found that preparat ion for the

te s t inc luding serum fract ionat ion , preparat ion o f ant i sera and

s tandardisation of p lates and standardised antigen was indeed t ime

consuming . I t was also noted that the test could be very sens i t ive

and cons iderable time was spent adj usting reagents so that it

eventual ly gave precipitation rings o f 0 . 5 - 1 . 5 cm for the range of

samp les used . A disappoint ing aspect o f the use o f thi s test was

its poor precis ion . Al though sui t ab le for e s t imat ing total immuno­

g lobu l in content of a samp le , when estimat ing the Bruce l la speci fic

percentage by a process o f subtract ion o f adsorbed from unadsorbed

values , the errors were compounded to a degree that made re sults

suspect . For IgG1 and IgG2 the large differences be tween adsorbed

and unadsorbed values for p rozoning and strong CFT pos i t ive sera

meant that for these sera resu lts were more meaningful . However ,

for e s t imations of IgM no s i gnificant d ifference s were found

between adsorbed and unadsorbed sera . This was more l ikely due

to the lack of precis ion of the test than the real absence of

specific immunoglobu l in .

Pfeiffer � �. ( 19 7 7 ) e s t imated the error o f the SRID test

to be in the order of 5 - 10% for measuring bovine immunoglobul in

leve ls . In this series o f tests i t was estimated to be more like

10- 20% . When the d i f ference between two te sts , each wi th errors

of this magnitude is taken the error of the d ifference can eas i ly

become qui te unacceptab le . Wi th further experience in the

technique and refinement o f some of the methods used i t i s

exp ected that the magnitude of errors could be reduced somewhat .

However , the problem remains that unless the tests are very

precise and are repeated a number of t imes , an accurate measure o f

the smal l amounts o f Bruce l l a speci fic antibody i n serum made

d irectly , wi thout prior fractionat ion , wi l l be extremely d ifficult .

8 3 .

In the l i ght of experience gained i n this s tudy i t is cons idered that the SRID test is ab le to offer sufficient accuracy for

measurement of sma l l amounts of speci fic ant ibody only aftei the

experimenter has become thoroughly fami l iar with the technique and experienced in i ts use . I t does , however , offer cons iderable sens i t ivity coup led with the use o f very sma l l vo lumes of serum

as l i t t le as 20 � 1 per determinat ion .

Mancin i � �. ( 1965 ) in thei r ini tial eva luat ion o f the SRID

test c laimed that the s tandard deviat ion from the mean o f antigen

measurements was only 2% . Nash and Heremans ( 1969 ) in a s tudy o f

specific mouse immunoglobul in leve ls c laimed an error leve l of 3%

for each measurement . It appears that with more care fu l at tention

to detai l , error leve l s for detect ion o f bovine immunoglobul ins

may be ab le to be reduced to leve l s approaching these .

Pfei f fer � �. ( 19 7 7 ) a l so evaluated commercial SRID ki ts and found that errors were somewhat greater than in tests performed

on the ir l aboratory prepared p late s .

Des p i te the test l imitations , re sults obtained for p ercentages

of specific IgG1 and IgG2 in the sera tes ted concur with the find­

ings o f McNaught � �. ( 197 7 ) and indicate that in prozoning sera

there is a higher proport ion of s pecific I gG2 than in non­

prozoning sera . These re sults are supported by the find ings that

add i tion of spec ific IgG2 to spec i fic I gG1 wi ll induce prozoning .

Brandon � �. ( 197 1 ) and W i l l iams and Green ( 19 7 6 ) s tudied the

manner in which bovine serum I gG 1 and I gG2 levels change at about

the t ime o f parturi t ion . S ince I gG1 i s act ively drawn from serum

into the udder (Brandon � �. 1 9 7 1 , Mach and Pahud, l 97 l ) , serum

leve l s fa l l and the I gG1 : IgG2 ratio can change quite dramatica l l y .

Serum I gG 1 leve l s can fa l l t o one quarter of prepartur i t ion leve l s

whi l s t I gG2 remains relat ive ly constant (Wi l l iams and Green, l97 6 ) . I

Such a change in ratio may induce prozoning in serum from a

Bruce l l a infected animal which has otherwise only low serum leve l s

o f speci fic IgG2 • Further detai led s tudy o f the way in which

Bruce l l a specific immunoglobul in l eve l s vary at about . the t ime o f

parturit ion are required t o assess the s i gnificance o f prozoning

during this period . In the infected animal the rap id pro-

l i ferat ion of the organism and consequent major s t imulation to

the hos ts immune system combined with the changing I gG1 : IgG2 ratio

at this t ime should generate an interes t ing and chal lenging s tudy .

CHAPTER 5 . APPRAISAL OF THE AUTO-ANALYZER ADAPTATION OF

THE COMPLEMENT FIXATION TEST

Introduction

84 .

Au t o -Ana l y zer ada p t a t ions o f the CFT for bruce l l o s i s have

previous ly been descr ibed by Joubert � �· ( 1967 ) and Mi l ler

� �· ( 19 7 3 ) . The sys tem at · pre sent used in New Zealand for

routine testing o f sera submitted as part o f the New Zea land

Brucellosis Eradicat ion Scheme , has been descr ibed by Te Punga

( 197 1 ) , E l l io t and Pul lan ( 19 7 3 ) and T imbs � al . ( 1978b ) . None

o f these . authors have ful ly described the dynamics - o f the sys tem

nor have they d irect ly re lated resu l t s to those obtained by use

o f class ica l manual CF tests . In h i s ini tial eva luat ion o f the

automated CFT Te Punga ( unpub l i shed ) found that the sera of 80

o f 82 cul ture pos i tive catt le reacted to the automated CFT , the

two sera that were negat ive were a l so negat ive to the manual CFT .

The experiments out l ined in this section were de signed to

cr it ica l ly examine the e f fectivenes s of the automated CFT . The

dynamics o f the te s t system were f ir s t investigated to enab le an

understand ing of the reaction proce s s , then various prozoning

sera were tes ted to f i t their observed pattern of react ion wi th

the expected resu l t .

Materia l s and Methods

System Descr ipt ion

Schematic d iagrams of the equipment configurat ion for the

automated CFT are shown in Figure s 9 and 1 0 . The automated CFT

i s essentia l ly a warm fixation CFT carried out . a t a s ingle

d i lut ion whi le flowing a long a tube . Serum whi ch has previou s ly

been inact ivated a t 5 8 °C for 50 minutes i s aspirated from the

samp le cup and di luted with buffer , comp lement and ant igen : an

initial incubation last ing 14 minutes then takes p lace . . Sheep

erythrocytes and haemolysin ( rabbi t anti - sheep erythrocyte

Figure 9

I

Schematic diagram of automated complement fixation tes t

HEATING BATH

Air

0-32 Haemal sin

O.J2 2% Shee Red Blood Cells

.---... •L-oo 3 -9 Saline f2j w.

.__ __________ __. ...... To sampler II 0

r- � :ASTE 2-9 FlowceU wash receJ:tacle

I STIRRER

I I

------�----L _

_

_

_ _u.� JloRo

li

'•

.-, lit ·

. '

COLORIMETER 1 5mm Tubular 'tic 660rrVA Filters

Figure 10

I

Principle of continuous f low analysis as adapted to the complement fixation test

c'+ Ag

Serum

Air

C = Comp lement

Ag = Ant i gen

RBC's Haemolysin

RBC ' s = Sheep red b lood cel l s

Bubbles removed

Chart rn recorde r �

Colorimeter

v Waste

serum) are then added and after a further 10 minutes incubat ion

the degree of haemo lysis o f the sheep erythrocytes i s measured

in a co lorimeter and recorded on a chart recorder .

85 .

According to Beer ' s law the quanti ty o f monchromatic l ight

absorbed bi a so l u t ion i s d i r e c t ly propor t i ona l to the

concentrat ion of the solute , wi thin certain l imits , and the

absorbance or optical dens i ty (OD) is re lated to the transmiss ion

(T ) of l i ght through the solut ion by the expression

O . D . = 1 log (T

) = - log T

In the system de scribed , the pre sence o f particulate

erythrocytes in the f low s tream prevents the direct measurement

o f soluab le haemoglobin for the de terminat ion of the degree

of haemolysis of the ce l l s . Ins tead , an indirect method mus t

b e u sed whereby the transmi t t ance o f e ry throcytes i n s u s pens ion

is used .

Thu s , when there is no haemolys i s , the presence o f a high

concentration o f erythrocytes prevents transmi ssion o f l ight and

when there is 100% haemolys i s , the absence o f erythrocytes

al lows l i ght to pass unre stricted . The e ffect o f a varying

haemoglobulin concentration - due to partial or comp lete lys i s

o f ce l l s - i s minimised b y us ing a 660 nm fi lter .

S ince the incident l ight i s large ly di spersed by re flect ion

from part iculate erythrocytes rather than be ing truly absorbed

by a solute , Beer ' s law cannot be s trictly app lied . However ,

in contrast to mos t app l ications of colorimetry where a uni form

quantitative re sult is required , a simple chart showing only

the re lative number of erythrocytes in suspension for a given

sample , i s a l l that is required from thi s system .

Reagents

Reagents used for the automated CFT were the same as those

used for the manua l test excep t that t i trations were carried

# : :

8 6 .

out on the Auto-Analyzer and d i lutions were therefore d i f ferent from

those used in the manual tes t .

T i trat ion of Reagents

Sheep erythrocytes were co l lected and pre served in Al sever ' s

so lut ion as de scr ibed by Alton � �· ( l975a ) . After washing

three t imes in barbital buffer , pH 7 . 4 ; ce l ls were standardi sed

at 2% by centri fuging at 2000 g for 10 mins .

In the p resence o f excess comp lement a minimal haemo lyt i c �'r

dose o f rabbi t haemolysin was determined by introducing decreasing

concentrat ions of haemo lysin unt i f the s teady state repre senting

lOO% lys is o f the ce l l s was d i srupted . The concentrat ion o f

haemo lysin which marked the beginning o f incomplete lys i s was

termed the "minimal haemo lyt ic dose" . Routine working s trength

haemolysin was prepared to three t ime s the minimal haemo lyt i c

concentrat ion . The haemo lysin t i tre obtained by thi s method was

general ly s imilar · to that given by the manual method o f Al ton

� �· ( l9 75 a ) .

Ant igen was t i trated by samp l ing a weak pos i tive serum and

varying the antigen concentration . The ant igen concentrat ion

g iving the highest peak for �he samp le was taken as the opt imal

concentrat ion for use in the test .

Comp lement+ was supp l ied as lyophi l i sed guinea-pig serum

preserved by a modi fication (Cruickshank � �. , 19 7 2 ) o f the

method of Richardson ( 1941 ) . After sheep red b lood ce l ls ,

haemolysin and ant igen had been t i t rated the opt imal comp lement con­

centrat ion was determined with re ference to a "standard" serum . The

* Commonwea l th Serum Laboratories , Melbourne , Aust .

+ Becton Dickinson and Co . , Cockeysvi l le , Maryland .

8 7 .

complement concentrat ion was adj us ted so that the "s tandard" serum ,

when samp led , gave a peak he ight equivalent t o 50% lys is o f the

sheep erythrocytes . In pract ice i t was d i fficult to keep the

standard he ight at exactly this leve l so a tolerance of be tween 25%

and 75% lys i s was usual ly al lowed . Once the standard height

moved outs ide this range the complement concentrat ion was adj us ted

to re turn i t to within these l imits .

Preparation of S tandard Serum . .

A feature of the Auto-Analyzer adaptat ion of the CF tes t s is

that all tes t s era are compared wi th a predetermined standard and

are clas sed as be ing pos i tive or negat ive with respect to thi s

s tandard . The FAO/WHO Expert Commi ttee on Bruce l losis (Anon . , 1 97 1 )

suggests that a serum containing more than the equivalent o f 1o i . u .

o f the comp lement fixing act ivi ty o f the 2nd International S tandard

Anti-Bruce l l a abortus Serum ( 2nd ISAbS ) should be c lassed as

posi tive .

A "standard serum" was prepared by taking a large vo lume o f

serum from a cow previous ly shown t o have a high CFT t i tre . This

was then d i luted with Bruce l la negat ive serum unt i l a serum with

equivalent CF t i tre to the second ISAbS was obtained . For

convenience 1 ml al iquots of thi s serum were lyophyl i sed for use

as a laboratory "s tandard serum" .

Continuous D i lution Method of T i trat ion

Titrat ion of reagents for use in the Auto-Analyzer adaptat ion

o f the CFT i s usual ly accomp li shed by prepar ing a series o f d i screte

concentrat ions of the reagent under test and applying each reagent

to the sys tem in order o f increas ing or decreas ing concentrat ion .

To inve s t igate de tai ls o f the e f fects of changes in

concentrat ion of reagents , the Auto-Analyzer lends itse l f to the

use of a continual ly changing concentrat ion gradient . Thi s

gradient is achieved by asp irat ing a concentrated solut ion o f the

reagent under test from a beaker and at the same t ime d i lut�ng

the remaining reagent with buffer or o ther sui tab le d i luent . As

long as the f low into and out o f the reagent receptac le is the

same , and the contents are kept wel l mixed , the reagent

concentration at any g iven t ime can be calcu lated by use of the

expres s ion : -

C = C e-kt t 0

wh•�o et w e oneen t r a t t on a t t ime t

C = original concentration 0

k = flow rate into or out Vo lume of reagent in

t = time

88 .

The derivat ion of this equat ion is given in the caption to Figure

1 1 .

This cont inuous di lution gradient method was used in

inves t igations o f antigen and complement interact ion in the

automated CFT .

Determinat ion of Von Krogh ' s Equat ion Characteristics for the

Automated Complement Fixation Test

(a ) Introduction

In the pas t , haemolytic activity of comp lement has been

e s t imated usual ly in terms o f the smal lest amount o f fresh serum

which wi l l produce complete lys is o f a specified number o f

sens i t i sed red ce l l s . However , during recent years there has

been recognit ion o f the advantage of us ing a 50% haemolys i s end-

point . The re lat ionship be tween the amount of complement used

and the proport ion of ce l l s lysed is not l inear , but fo l lows a 1

s i gmoidal curve . Relat ive ly large amounts of comp lement are

required to ini t ia te and to complete lys i s but in the central

region the degree o f lys i s i s sens i t ive .to smal l changes in

complement concentrat ion . Von Krogh ( 19 1 6 ) deve loped a

mathematical descr iption of the observed data on haemo lys i s and

the Von Krogh equation has been commonly employed to quantify

various component s ·of the complement f ixation tests system.

Figure l l Continuous d i lution method o f ti trat ion

A beaker or conical f lask is p laced on a magnetic

s tirrer . A measured volume (V ) of reagent with con-

centration (C ) i s p laced in the flask and continually 0

s t irred whi l e a d i luent i s added at constant f low rate f .

At the same time , reagent i s withdrawn from the flask at

constant f low rate f .

Thus

Concentrat ion of substance in initial reagent

c = J:L where M = mass V . .

d M = M = - f X c M = change in

d t mass

f c c =

V V

which has the solut ion

- kt c Ae

where A = c = ·c at t ime t 0 0

and K f =

V

= et c - ( f/V ) t e 0

buffer in = f1 mls/min

Vo lume = Vt

buffer out = fa mls/min

...._ _________

! magnet i c st i rrer

1

X = k (_y_)n 1-y

where X = amount of comp lement (ml s )

89 .

y • degree o f haemo l y s i s ( i . e . 100 y • % haemo lys i s )

k = 50% uni t o f comp lement (CH50 uni t )

when y

X = k.

= 0 . 5 ( i . e . 5 0% haemolys i s ) , then -1-1 1 and therefore -y

Thus the CH50 uni t i s expres sed in mls of comp lement

required to give 5 0% haemo lys i s . The magnitude o f the exponent ,

l, depends on the experimental condit ions but is usua l ly 0 . 2 + 1 0% ; n its va lue determines the shape o f the s i gmoidal curve (Kabat and

1 Mayer , l 9 6 1 ) . A knowledge of the value o f ; for the automated

comp lement fixation test sys tem wi l l there fore lead to a greater

understanding of the dynamics of the react ion .

( b ) Method

A cont inuous d i lu tion method of comp lement t i trat ion was used .

Values o f % transmiss ion given by the chart recorder were

trans formed into % haemolys i s and the comp lement concentrat ion

required for a given degree of haemo lysis of red ce l l s calculated .

The colorimeter and chart recorder were balanced so that d i s t i l led

water gave 100% transmiss ion and a comp lete obs truct ion to the -l ight path gave 0% transmis s ion . The % haemolys i s ( H ) f igure for

any· g iven % transmiss ion (T ) was calculated by the expres s ion : ­

OD - OD

where

i.H = 1 -X 100 X lOO

ODo- ODlOO

OD = - log T X X

OD lOO - - log T lOO

OD = ·- log 0 0

' I

90 .

and T 7. Transmiss ion a t Eoint X X = lOO

T lOO = 7. Transmiss ion for 1007. haemo l,:t:s is

100

T - '· Transmiss ion for Oi. haemo lys is 0 100

The re lationship between this degree o f haemolys is and %

transmis s ion was also obtained by substituting appropriate va lues

in the simul taneous equations .

and

y 0 = M . OD 0 + c M . ODlOO + C

then so lving for M and C

where y 0 y

l =

OD = 0

OD lOO=

degree

degree

op tical

opt ical

o f haemolysis of 0 ( i . e . 07. haemolys i s )

of haemolysis of 1 ( i . e . 1007. haemolys i s )

dens i ty at 07. haemolys is

dens i ty at 1007. haemo lys is

M = s lope of l ine

Values of y cou ld then be ob tained for any values o f ODX '

represented by ODx = - log T X , taken direct ly from the particular

Auto-Analyzer chart by subs t i tution in the equat ion y = M. ODX + C .

l .

Two methods were used for the ca lculat ion o f CH50 uni t .

A graph of comp lement concentrat ion [c] plotted against � was prepared and the comp lement concentration required for

50% haemolysis (-1-1 = 1 ) was read directly from the graph . -y By thi s method the neatness of fit o f the s traight l ine (when

drawn on Log/Log paper ) could be readi ly observed and gave an

indicat ion of the accuracy o f the method .

2 . The %T figure corresponding t o 50% haemolysis was read d irect­

ly from the chart and the complement concentrat ion required

to give this value was then calculated direct ly .

OD for SO% lys i s =

and %T for SO% lys is =

0Dl00 +

ODo 2

10 _oDso X 100

9 1 .

= ODSO

1 ThO V4 1 U ' O f thO oxpOnOnt , n • t n determined direc t ly from the graph of

the Von Krogh equat ion W6 1 [c ] v -1- where l i s

1 -y n represented by the s lope of the s traight line , or by trans formation

o f the equat ion to

1 log X - log K =

n log -1-1-y

When y = 0 . 5 ( i . e . 507. haemo lysi s ) , then

_1_ 1 -y 1 and log -1-1

= 0 , also log X = log K so that -y

log X - log K = 0 .

Thus to use this equat ion a va lue o f y o ther than 0 . 5. mus t be

used with its corresponding value' o f x .

Investigat ion o f Resul tant Serum Dilut ions in the Automated

Complement F ixation Test

( a) Introduct ion

In the automated CFT samp le is asp irated at 0 . 23 ml /min

for 12 seconds and inj ected into the antigen/comp lement f low s tream .

This is fo l lowed by aspiration o f buffer unti l the next samp le i s

taken . At the point o f introduct ion into the main flow s tream

the samp le i s d i luted 0 . 23 : 3 . 13 or 1 : 1 3 . 6 ( sample l ine 0 . 2 3 ml/

min , antigen/complement l ine 2 . 9 ml /min) . However , because o f ..

carry-over in the f low s tream the samp le wi l l not s tay a t this

concentration but wi l l become pro_gres s ive ly more di lute . To tes t

the extent o f this d i lution - carry-over effect radio- labe l l ed

serum was used and traced through the f low-s tream .

( b ) Method

1 25 An al iquot o f serum containing I labe l led bovine gamma

g lobul in was prepared . Three 0 . 25 ml samples were taken · and the gamma

I

9 2 .

emis s ion o f each was counted in ' an auto-gamma scinti l lat ion �'r

spectrometer • S amp le ( 1 ) was aspirated by the samp ler for the

standard time o f 1 2 secs and the reagent mixture reaching the end

o f the system after the addit ion of red b lood ce l l s and haemolysin

and f inal incubation was co l l ected in sma l l al iquots . These

a liquots were then p laced in the scint i l lat ion spectrometer and

the number o f counts/min for each was measured . Samples ( 2 ) and

.( 3 ) were asp irated in a s imi lar way except that the co l lect ion o f

aliquots was made d irect ly after the ini tial ' incubat ion and prior

to the addition o f haemo lytic system . In this way the di lut ion

gradient of the samp le was estimated at �bout the time of the

primary antigen-antibody react ion rather than at the end of the

haemo lytic reaction as with sample ( 1 ) .

Invest igation of the Effect of Antigen Concentration on Comp J ement

Fixing Ability

( a ) Introduction

The cont inuous f low nature of the Auto-Analyzer adaptation o f

the complement f ixat ion tes t lends itse l f to use as a too l to

investigate various aspects of the kinetics of the tes t . Antigen

titrations are carried out as standard practice in the performance

of complement fixat ion te sts with the obj ective of finding the

optimal antigen concentrat ion for rout ine use in th� tes t . The

s tandard manual antigen t i tration such as that given by Alton

� � . ( 1975a) cal ls for the use of a known pos it ive serum agains t

whi ch the antigen is t i trated . I t has been noted that d i f ferent

"positive" serums may have different opt imal antigen requirements

for the CFT . The Auto-Analyzer proved to be a convenient too l

for investigating the extent o f these d i f ferences .

( b ) Method

By choosing suitab le serum and comp lement concentrations and

us ing a continuous di lution method of a l tering antigen concentrat ion ,

the trace obtained on the Auto-Analyzer chart recorder d irec t ly

* •,

Packard Gamma- scint i l lation spectrometer .

9 3 .

reflected the complement fixing abi l ity o f var ious ant igen/antibody ratio s . A number o f d i fferent sera and serum dilutions were

tested .

( a ) Introduct ion

The automated CFT is e s sentia l ly a one d i lut ion warm fixat ion

type test . As such , i t s princip le has been critici sed on the

grounds that i t wi l l fail to detect prozoning sera . A previous

sect ion in thi s chapter exp lained how an experiment was de signed

to show how the sample within the Auto-Analyzer did not retain

a di screte di lution level but rather formed a continuous d i lut ion

gradient . Thus the automated CFT measures the degree o f complement

fixation on a wide range of serum d i lut ions for each samp le .

( b ) Method

A series of prozoning sera were tested at var ious antigen

d i lutions by the automated complement fixat ion tes t .

Re sults

Detenmination o f Von Krogh ' s Equat ion

A typical trace obtained by continuously d i lut ing complement

in the absence of samp les i s shown in Figure 12 . The sp ike

labe l led "A" represents an air bubbl e introduced to indicate the

s tart o f the continuous d i lut ion o f comp lement . The curve between

B and C represents the change in haemo lys is of ce l l s from 100% to

0% . By calcu lating the % transmiss ion at which 50% of the ce l ls

are lysed and measuring the d i s tance o f this point from the spike A

the d ilut ion o f comp lement required to lyse 50% o f the red ce l l s

may b e determined .

A graph

Figure 1 3 .

as fol lows .

p lo tted from the chart in Figure 1 2 i s shown in

Values o f [ c] and ..:L.1 were --derived for this graph -:Y

Figure 1 2 Auto-Analyzer trace given by continuous ly

d i luting compl ement in the absence o f

samp les

0

1 0

20

30

40

z 0 (./)

50 (./) :L (./) z <( 0:: 1- 60

� 0

70

80

90

100

A 0 1 HAEMOLYSI S

50 1 H : Q . 4 Q 7 T D

13 . 9 cm [ C' ) :: 1 :493

----------------�8 1 0 0 /.

HA E M O LYSIS

c

F i gure l 3 Graph drawn by p lo t t ing values der ived

from chart shown in f igure 1 2 .

By measuring the s lope o f this l ine a 1 value for - in the von· Krogh equation n

i s obtained .

0.010

0.008

0.006

[ C l 0.004

0.003

0.002

�

0.001 0 .1

slope = 1 .4 = {

0.2 0.3 0.4 0.6 0.8 1

9

degree of haemolysis,

9

2 3 y

1 -y

6 8 10

94 .

The distance ( dX) from the origin ( sp ike A in Figure 1 2 ) to

the point on the chart representing a particular '7. transmiss ion

is converted into a measure of time ( t X) by

d X t x

= s

where s = chart speed in cm/m in (

et e -kt now = e 0

where et = concentrat ion at time t

e = original concentrat ion . 0

k = flow rate into or out Vo lume o f reagent in

For this series o f experiments f = 2 . 5 1 mls/min

V = 20 mls t

To f ind degree o f haemo lys i s (y )

y

where ODX =

and TX =

= 1 -OD X

- log T

OD 100

% Transmi ss ion at point X 100

= � 7 62 cm/min)

Thus the series of points for the �raph shown in Figure 1 3 are : -

95 •

Tx [ c] ....L 1-y

0 . 2 1 0 . 00146 0 . 1 19

0 . 25 0 . 00 1 66 0 . 2 2 6

0 . 3 0 . 00 1 7 9 0 . 468 0 . 5 0 . 00 2 17 1 . 638

0 . 6 0 . 00 235 2 . 67 6

0 . 8 0 . 00 2 7 1 8 . 804

The s lope o f the curve can be measured and found to be 0 . 14 which

� s the value o f -1 · h V K h · • �n t e on rog equat�on . n

Al ternative ly for any part icu lar point on the chart in Figure 10

1 n log X - log k

log .J_ 1-y

where x and k

=

= amount o f comp lement for degree o f lys i s y

amount o f comp lement for 50% lysis ( i . e . CH50 uni t )

From the chart in Figure 1 2 the CH50 uni t was measured d irect ly

and found to be 1 : 49 3 or 0 . 00 203 . · From the graph Figure 1 3

the point of intersect obtained when � = 1 represents the CH50 unit and is seen to be 0 . 00 2 . Thus for a value of T = 0 . 5 for

examp le :

1 n = log 0 . 002 1 7 - log 0 . 00203

log 1 . 63 8

= . 135 1

For a series of experiments us ing the same batch o f reagents , 1 values o f the CH50 uni t and o f n were calculated as : -

CH 50

( d i lution o f C )

1 : 49 3

1 : 498

1 : 49 3

1 : 495

l n

. 14

. 1 38

. 137

. 14

96 .

l When antigen was added to t�e system values for - remained at n

0 . 138-0 . 14 but the CH50 d i lution o f complement reduced to

approximately l : 485 . for a 1 : 1000 d i lution of antigen and to 1 : 470

when a 1 : 500 antigen d i lution was used . Thus the degree o f anti -

complementary activity in the ant igen could be quanti fied in terms

o f CH50 uni t s . Values o f the CH50 unit also var ied s l ight ly

according to the batch o f red cel l s u sed but again the value o f

l remained constant at about 0 . 1 38 . Tt'

Determinat ion of the Dilution Gradient of Serum in the Automated CFT

Figure 14 i l lus trates the d i lution gradient for each radio­

l abe l led samp le after trave l l ing through the Auto-Analyzer system .

S amp le l traversed the comp lete system whi l s t samp les 2 and 3 were

retrieved fo l lowing the initial incubation and prior to the addit ion

o f haemolytic system . For sample l aliquots were each o f 0 . 7 mls

whereas for samp les 2 and 3 aliquot vo lumes were 0 . 25 ml each .

Figure 15 indicates the re lative serum concentrat ion in each

a l iquot and compares i t to the serum concentration immediate ly

fo l lowing aspirat ion and with the serum concentrations in a set

o f tubes in which a manua l fixation CFT is be ing performed .

The ini t ia l serum concentrat ion in the · Auto-Analyzer during the

p eriod which serum is asp irated i s 1 : 1 3 . 6 ( 0 . 07 34 ) .

S ample cone . in aliquot retrieved = counts/min/a l iguot x Total counts/min re­

covered initial vo l . of sample introduced

vo l . o f al iquot retrieved Table XIV summarises counts per minute recorded at the various

s tages of the experiment . The number of counts was fairly constant

b etween samp les and would have more near ly approximated the number

o f counts asp irated i f more al iquots had been co l lected and

counted for each samp l e_ .

Figure 14 Di lut ion gradient given by serum

containing r 1 25 label led bovine

gamma globul in .

150 000

2 100 000 :J

.£ E

--1! �) 50 000 0

u

Total counts

i .. ' • . . . . . .

Run 1 -- 315 136 Run 7 - - - - - 314 437

. .

A f : Run J · . . . . .... . 315 567 I \ \ \

\ : I I i I \ : I \ i I \' \ I :\

I i \ I i \ I : \ I . \ I \ I ', \, I ' • • ... , ·· .. I • ' •.

:--":�--:---:-oooo:--:--"'1"'"-�-t. . . . . . . . .... --,-,-..:;=:;=::"�-�·y·. �· llloiofio.;;-.�.,-���,_, I I I I I I I I I I I I I I I 2' I. 6 8 10 12 14 16 18 20 22 24 26 28 30

Aliquot number

•

Figure 1 5 H i s togram of relative serum di lutions in

each a liquot taken from Auto-Analyzer

-...... 0 � . Q" 0

0-06

0 -04

0 -02

0

E 0-04 -­Q)

g-� 0 -02

E

c 0

........ 0 '­

........ c Q) u c 0

u

0

0 -04

0-02

0

Sample 1

Sample 2

Sample 3

1

Al iquot number

Initial 1 :4 :8 :16 :32 :64

cone. Eq..�ivalent cone. in

of sample manual tube series

97 .

Tab le XIV . Summary o f leve ls o f radioact ivity recorded before

and after d i lution o f samp les by Auto-Analyzer

Sample No . 1 2 3

Ini tial counts per m in ( cpm) 1 267606 1 2 6 7606 1267606

Counts remaining afte r sample aspirated (cpm) 898702 89 7 164 897045 .

• counts asp irated (cpm) 368904 370442 370561 .

Total counts retrieved ( cpm) 3 15 1 36 3 1 4437 315569

I \·

Effect of Antigen Concentration on Bruce l l a abortus Complement

F ixation Test

98 .

Figure 16 i l lustrates a series o f typical antigen ti trat ion

charts obtained by using a continuous d i lution method o f antigen

t i trat ion wi th the complement concentrat ion adj us ted to give mid­

range peak heights . The corresponding antigen d i lut ion for each

peak height has been calculated and noted on the chart . Whereas

samp le C requires an antigen d i lution of about 1 : 3000 for optimal

comp lement fixation , s amp le A requ ires in the region of 1 : 1000

antigen . At antigen d i lut ions of 1 : 3000 or more samp le A fai l s

t o show evidence o f s i gni ficant comp lement fixation .

Figure 17 shows how d i f ferent di lutions o f the same serum

a l so have d i fferent opt imal antigen concentrations . The serum

used for this set o f curves was a s trongly pro zoning serum and

i t can be seen that in general the concentrat ion o f ant igen

required is relative ly high , i . e . less than 1 : 750 .

F igures 18 and 19 chart optimal antigen d i lutions for

d i f ferent d i lutions o f pos itive non-prozoning sera . In these

cases , al though less antigen is required as the samp le is d i luted ,

an antigen more concentrated than 1 : 1000 is not required .

The Characteristics of Auto-Analyzer Chart Traces Given by

Prozoning Sera

A number o f sera known to prozone in manual comp lement

fixation tests were te s ted on the Auto-Analyzer . Figure 20

i l lustrates a series o f typical peaks for prozoning and non-

prozoning sera at various antigen d i lutions . I t can be seen

that while the highes t s tandard peak height is obtained when the

ant igen d i lution is at 1 : 1 000 , prozoning serum peak heights are

greates t at [Ag] = 1 : 500 . Tab le XV gives manual CFT resul t s for

each o f these sera .

The mos t notab le p eaks are those given by sample numbers 6 , 7

and 10 . These very strongly prozoning sera each have a

characteristic diphasic peak .

Figure 1 6

I

Series of typical antigen �i tration chart s from Auto-Analyze r ; figures repre sent the

reciprocal of the e f fect ive d i lut ion of

antigen at tip of peak.

Each chart was constructed by repeatedly

sampl ing a pos i tive serum while the antigen

was being continuous ly di luted .

The series of peaks represent the s trength

of reaction given by the serum at each

antigen di lution .

-0 0

10 0

CJ) 0

.....:! 0

3 3 25 2720

545

0'1 0

477

6 3 4

71 805973 8 4 87

l11 0

I' 0

w 0

564 667 '

2301

762

5052

1 3 9 2

S,0 2 948

1 34f 1140

19141 592

901

4 2 74

1 6 4 6

1066 1 260 ��&r

2 081

3515 33892

1 9792 3 3 4

281 2 3325 3930 4 647 �493

780g 02

9384 11 2 7 9

131 1 1

N 0

_. 0 0

)>

(lJ

(")

Figure 17 Antigen t i tration with varying serum

d i lutions o f a prozoning serum.

Curves were constructed from a series

o f charts s imilar to those disp layed

in figure 1 6 .

I•

Optical Density

. 7

,,

0 % haemolysis

1 :20 ----· 30 ----'\

���� ': � .... � I \ ··.\.

1 -'"""''''" 1 :40 .. _

1 :50 ---i � �, :

� ... : .-· ..... ,

"'\. 0 -·-·-1 :6 --· 1 :70 -/

·,, .. I '·' · .. �

.

·� "'·�

I ·-···- '" . . ,

·-·

. , ---\. ......... "·�

-------==.:-.:--

'/ �- ·-.

------:._ _ _ _ _

. ,�, �\:....,t:h.. . ----�- --·-·

l ·s

I • - . ---� .• "'!!i;,... ,.,.= ··· · ··· · · ..... .. .

" haemo ysl

. , - - -----=--=- -

---�10�0,.::to:.r---

/ , · I

===.---��-� � 6000 �1,1------==-

5000 1000 2000 30 00 4000

Dilution of antigen .

' '.

Figure 18 Antigen ti tration with varying serum

d i lut ions o f a non-prozoning serum

Optical Density

0 .6

O.L.

0.2

1 l% haemolysis

G e rum Dilu t ion,:; �- - -/ .......... I .......... I ..... .....

I - --

1 : 1 � I - ..... ..... I ..... .....

I -..... ..... I · · · · · · · · · · · · · · · · · · · · · · · ............... ..... .. . ·

· . . . . . .. . . . . . . . . . . . . I .. · · · · · · ·

' -..... .

,,

,.•' • · · · · · · ' -... I ; . I .

· · · · · · · · · · · · · · · · · · · - - -·

... · . . I · · · · · · · · · · '

•' ' ' • • • • •·l ' I j l

_,/ ,/ 100% haemolysis

0�----��----�------�----�------�----�----� 1000 2000 3000 6000 7000

Dilution of antigen · .

Figure 1 9 Antigen t i tration wi th varying serum

d i lut ions of a non-prozoning serum-

Optical Density

---.,....,... ...............

, "' ........ ........ / ......._......._ 'I : eO I ....... �-

. � -

0% haemolysis

/ � ....... '•.) � ....... � ....... _ 1 \ �-

/ ...

I .......... . .. ..

. I ; .) .. ..

100% haemolysis

D ilution of antigen

Figure 20 E f fect of variation in antigen

concentration on prozoning sera in

automated comp lement fixation tes t

� ·----2------�-----�----�----� 0 ..

0 .. :il

"'

-"11

.. § N

99 .

Tab le XV . Manual CFT resul t s for sera giving peaks shown

in Figure 20

Sample No . CFT t i tre

Warm f i xa t ion Co l d fixa t ion

l - 323422 3453

2 49 410

3 - 42345 2 3473

4 346 1 462

5 46 46

6 20 _ 5 343

7 20 _ 6243

8 42 42

9 422 432

10 -41 2 332 47 2

1 1 246 47

1 2 43 44

1 3 - 2 2422 2344

S tandard 42 42

Ti tres are given in the notat ion exp lained in section 3 . 2 where

422 indicates that a react ion of 4 (++++) was seen in the first

two tubes of the ser ies and a reaction o f 2 (++) was s een in the 20 third tube . = No reaction seen in any o f a 20 tube

series .

Discuss ion

Use of the Auto-Analyzer adaptat ion of the complement

fixation te st for Bruce l la abortus antibodies has enab led the

New Zealand bruce l losis eradicat ion scheme to proceed at a rate

beyond that which would have been attainable if manual testing

systems had been emp loyed .

Sobota and Gil len ( 1965 ) , Vargues � �. ( 1965 ) and Gai l lon

� �. ( 1 966 ) have described Auto-Analyzer adaptat ions of the

comp lement fixation te st particu larly with respect to the sero­

d iagnosis of syphi lis . Joubert � �. ( 1 967 ) l ater described an

automated comp lement fixation tes t for bruce l losis but did not

e l aborate on its practica l i ty or properly evaluate i t against

manual tests or known infected animals .

100 .

Te Punga ( 19 7 1 ) fir st described the automated CFT as used in

the New Zeal and eradication scheme . He described the mechanics

of the test but did not detail the epidemio logica l s tudies that

led to its acceptance as a definit ive tes t . In the unpub l i shed

s tudies associated with the deve lopment of the automated te st ,

Te Punga concluded that the co ld fixation CFT related better to

the automated test than did the warm fixation method . I t was a lso

observed that the co ld f ixation test was more eff icient at detect-

ing infected animals than was the warm fixation te st , this

advantage of the co ld test has a l so been noted by Ze issig and

Mans field ( 19 30 ) , Tri lenko ( 195 7 ) and Isayama ( 19 6 1 ) .

Minor modifications to the original (Te Punga 197 1 ) mani fo ld

l ayout and testing method s have been made and are detai led by Timbs

� �· ( l978a) .

O f utmost importance in the performance of the automated CFT

i s the use o f a "s tandard" serum wi th each group of 39 samp les

( each Auto-Analyzer carousel ho lds 39 unknown , p lus one "s tandard" ,

a total o f 40 samp le s ) . The t i tre o f this "s tandard" is set

according to accepted international op inion and for this series o f

experiments was used a t a comp lement fixation t i tre equivalent to

l / 30 th that of the Second International S tandard anti -Bruce l la abortus serum .

The continuous d i lution method o f t itration has been used in

s tudies of the kinetics of comp lement fixation sys tems by Vargues

101 .

and Ayera ( 19 63 ) , Vargues ( 19 65 ) and Vargues � �· ( 1966 ) . In

the pre sent study it proved to be a most useful means by which

antigen and comp lement affinity relationships could be accurately

measured .

I t has been shown that various prozoning sera can produce

peak heights be low that o f the "s tandard" despite being pos i t ive

to manual tests . The choice o f antigen - concentration i s most

important in thi s regard . Experiments out l ined in this chapter

have shown how sera have d i f ferent optimal antigen concentrations

and how a particular concentration chosen after a t i trat ion wi th

one serum may not be entire ly sui tab le for another serum . The

e ffect of an increased concentration of antigen in suppressing

prozoning is o ften seen when performing manual antigen t i trations

but has only recent ly been formally described (McNaught � �. ,

1977 ) .

In manua l t e s t s a wide l a t i tude o f ant i gen concentrat ion i s

tolerab le provided sufficient serum d i lutions are made ( e . g .

1 : 100- 1 : 400 Ag) . However , in the automated tes t the use o f a

sub-optimal antigen concentration cannot be compensated for by

other factors . Unfortunately , one particular concentration wi l l

n,o t b e optimal for a l l sera s o a compromise must b e reached . I t

i s thus important to choose for antigen titrations , a serum which

i s known to require a mid-range optimal antigen concentrat.ion

and to err on the side o f a more concentrated reagent when in

doub t . I n practise , the serum deve loped for use a s the laboratory

standard should be chosen with thi s particu lar qua l i ty in mind .

From the re sults shown in this section and those o f McNaught � �·

( 19 7 7 ) it is apparent that prozoning sera require high ant igen

concentrations to effect optimal comp lement fixation , and the

extension of thi s argument is that sera with high concentrations

of speci fic I gG2 require more antigen than do those with lower

amounts o f I gG2 antibody .

Von Krogh ( 19 1 6 ) app l i ed mathematical formulae to various

chemical reaction rate s , Among the systems he investigated was

that o f sensi t i sed erythrocyte lys is by comp lement . According

to Mayer ( 19 6 1 ) the value o f l in the Von Krogh equation 1 n

[x= K(G' >1rJ determines the shape o f the sigmoidal curve and i s

usually o : 2 + 10% .

102 .

Manua l complement t i trations p erformed in connection with this

proj ect invariab ly gave this value for l. On the other hand the n automated comp lement fixat ion test sys tem was found to give a

consis tent value o f l = 0 . 14 . The value o f l can be rai sed by n n ei ther increas ing the concentration o f erythrocytes , or

increas ing the calcium ion concentration [ea++

] .

S ince the same buffer was used in both manual and automated

test systems it cou ld not be a change in [ea++

] that affected the 1 value o f -. In the manual eFT a 2% erythrocyte suspension i s used n

and upon d i lution in the te s t system equates to a final

concentrat ion of 0 . 25% . In the automated te st the final

erythrocyte concentration i s calcu l ated by d ividing the erythrocyte

pump tube f low rate by the total of the reagent pump tube f low rates

and mul t ip lying by 2/ 100 ( Ini tial red ce l l cone . = 2%) .

Thus the final concentration o f erythrocytes in the automated

test i s

0 . 3 2 =

2 . 9 + . 23 + . 3 2 + . 3 2 X 2 100 0 . 17%

which is somewhat less than the 0 . 25% concentrat ion obtained in the

manual test 1 �en ­n

wi l l re su l t

1 and exp lains the re lat ive ly low value o f -. n is sma l l , a sma l l change in comp lement concentration

in a large change in the degree of lysi s . Thus the

automated test as de scribed i s relative ly sensit ive to small

di fferences in the t i tre of samp les which have approximately the

same antibody concentrat ion as does the s tandard serum , i . e . when

they achieve sufficient fixat ion o f comp lement to give 20-80%

haemo lys i s o f erythrocytes .

The I 1 25 radio- labe l led pro te in method proved to be a useful

way o f measuring the d i lution gradient of the samp le wi thin the

Auto-Analyzer system . Earl ier a t tempts to measure thi s samp le

di lution gradient with dyes genera l ly proved to be unsat i s factory

due to d i fficu lties in balancing the co lorimeter/recorder to

accurate ly measure a sufficient ly large gradient of dye co lour , and

due to dyes tending to adhere to the internal wal l s of the tubing .

I t has been s tated that the main problem with the automated eF te s t

i s the d i fficu l ty in choosing a suitably representative s erum

di lution to use (Mi l ler , 197 la) . With a knowledge o f the serum

dilution gradient eventual ly achieved - as shown in f igures 14 and

15 - the way in which a given reaction i s represented on the chart

103 .

recorder is more readi ly comprehended . I t i s the overall balance

o f the system and the way in which reagents are ti trated as we l l as

the serum di lution chosen that eventua l ly determines the validity

o f results for a wide range o f samp les0

Prozoning samp les present a pecu l iar prob lem . I t has been

shown by McNaught � �· ( 1 9 7 7 ) and in the experiments out l ined in

chapter 5 that specific IgG2 is capab le of inhibi t ing complement

fixation by speci fic IgG1 . The nature o f thi s blocking e ffect i s

unknown , but i ts intens ity can b e reduced by increasing antigen

concentration . Re sults i l lus trated by F igure s 17 and 18 show how

increas ing anti gen concentrat ions wi l l a l so inhibit the e ffect o f

pro zoning in the automated test .

Figures 1 7 , 18 and 19 i l lus trate how d i f ferent sera have

d i f ferent optimal antigen concentrations for complement fixation .

Although i t has not been conc lus ive ly proven , i t appears that the

mechanisms responsible for prozoning are the same as those mediating

the unusual ly high antigen b ind ing capacity , for optimal comp lement

fixation , in some sera , as the degree o f prozoning exhibited by a

serum is antigen concentration dependent . Sera that do not

normal ly show prozoning in manual tes t s may in fact s t i l l have

Bruce l la- speci fi c IgG2 present , the amount of which may we l l

determine the optimal antigen concentration for best comp lement

fixation .

Thus the concentration of speci fic IgG2 within a sample

appears to determine its characteri s tics for :_-

( i ) Prozoning

( i i ) Opt imal antigen concentration for comp lement fixation .

I t a l so appears that "prozoning" is simp ly a visual

manifestation of a serious antigen/ant ibody imbalance , within a

test system , which is mediated by speci fic IgG2 • A less serious

imbalance may not necessari ly be vi s ib le but i t may we l l cause

d i f ferences in t i tre between two simi lar sera when different

antigen concentrations are used in the CF te s t . This e ffect was

o ften seen when two sera o f apparently identi cal t itre were found

to have d i f fering t itres when tested u s ing a different ant igen

concentration . Genera l ly titre d i fferences within acceptab le

antigen dilution ranges were smal l enough to be o f l i t t le

consequence but occasional sera , par t icularly those that prozoned

104 .

in manual tests , did show larger titre variations . F igure 20

i l lustrates how the sensitivity of the automated test varie s wi th

antigen concentration and in particular how the degree of pro zoning

o f a serum influence s di fferences in peak heights obtained . By

using sufficiently s trong ant igen the peak heights o f pro zoning sera can be elevated . However , thi s may be at the expense of

suppress ing peak hei ghts o f normal sera if unduly high antigen

concentrations are u sed .

I t i s important therefore , to choose an antigen concentration

sufficient ly high to enable detection of prozoning sera yet not

so high that normal - particular ly low t i tre - peaks are suppres sed .

Fo l lowing from these observations it can be appreciated that

a simp le direct interpretat ion of chart traces is not a l l -together

possib le . Al though a l l peaks above that of the standard may be

taken as posit ive i t cannot be said that a l l peaks be low the

standard are negative . Some of these " smal l " peaks may be due to

prozoning sera . U sual ly such peaks , especially those given by

strongly prozoning sera , are recognised by the traine d observer .

Prozone peaks are generally proportiona l ly wide and o ften have a

ragged top c . f . F i gure 2 0. Peaks due to low ti tred normal sera

are f ine and have a sharp spike . To properly interpret charts it

is therefore nece ssary to repeat manual ly any samp le giving a

de finite peak be low the standard peak he i ght : fortunate ly the

prevalence of such peaks is normal ly very low .

105 .

CHAPTER 6 . ANALYSIS OF FIELD DATA

Introduction

The experiments and concepts developed in thi s thesis have

l arge l y evo lved in re sponse to the fund ame n t a l q ue s t i on "wha t i s

the signi ficance o f animals which are repeatedly card te st posi tive

and comp l ement f i xa t ion , t e s t n e g a t ive , in the N ew Ze a l and brucel lo s i s eradication scheme ? " T o enable a proper appreciation

o f the re lationship between the - card and comp lement fixation tests

nat ional and regional data was analysed to assess the way in which

the two te sts interacted as the scheme progressed . T imbs � �·

( 1978a ) investigated the exi stence and eventual fate o f animals

that were repeatedly BCT+ and CFT- . This study examines the

prevalence of BCT+/CFT- animals in herds with varying initial CFT

reactor rates . ·

Materials and Methods

Certain bas ic information concerning details of all tests

carried out during the N . Z . bruce l losis eradicat ion scheme is

stored within a central computer f i le .

Unti l August 1 9 7 7 a l l sera from initial dairy herd te sts were

card te sted and card test positive sera were then comp lement

fixation tes ted . Init ial bee f herd tests were treated in thi s way

unti l August 1975 . At tests other than these the card test was

not nece s sarily performed and the computer fi les did not

di fferentiate herds which were card tested and had no BCT positives

from those which were not card tes ted . Thus data from a l l repeat

herd te sts and some initial beef tests could not be used .

106 .

Resu lts

Table XVI l i s ts detai l s of BCT and CFT reactor rates for dairy

and beef herds at ini tial herd tests .

( a ) For da iry herds

Linear regre ss ion y a + bx

y No BCT+ No CFT+ X lOO No animal s - No CFT+

X No CFT+ lOO No animal s X

a = 2 . 5 b 0 . 225

F l , 5 = 60 . 7 1 ·k·-k ( li. ) R2 9 2 . 4 i.

(b ) For beef herds

a 3 . 7 b 0 . 346

F l , 5 18 . 10 ";'('•/( ( l i. ) R2 = 7 8 . 47.

Thus a s ignificant corre lation exi s ts between the overal l herd

infect ion rate and the proport ion of animals wi thin herds that

are BCT+ but CFT- .

107 .

Tab le XVI . BCT and CFT reactor rates at ini tial herd tests

( a ) Dairy herds

Herd CFT No . herds No . animals No . BCT+ No . CFT+ i. CFT-reactor rate that are

BCT+

0 4895 5 20 1 3 7 1 7903 3 . 44

0- 1 1435 244045 5 9 24 1586 l . 79

1 -5 4147 5 33 9 9 2 2 7 9 2 7 1417 3 2 . 64

5 - 10 2439 343042 3834 1 249 7 7 4 . 20

10- 15 1 305 183006 3 1496 2 240 1 5 . 6 6

15- 20 608 8 3 1 2 2 18898 1429 1 6 . 69

> 20 418 49090 1 5 2 3 1 1 2 35 7 7 . 8 2

( b ) Beef herds

Herd CFT No . herds No . animals No . BCT+ No . CFT+ i. CFT-reactor rate that are

BCT+

0 6 207 8 3 2 1 84 33904 4 . 07

0 - l 8 2 9 2 7 7 155 9 434 1 1 98 2 . 9 7

1 - 5 1 274 2 20418 1 6 1 88 5 3 19 s . os

5 - 10 406 807 29 1 1427 5 7 7 9 7 . 5 3

10- 15 155 30945 58 78 367 3 8 . 09

1 5 - 20 5 8 8 3 7 1 1 842 14 26 6 . 00

> 20 6 3 5 0 10 1 85 4 1 3 1 8 14 . 5 2

108 .

Discuss ion

Fo l lowing the inab i l ity to cul t ure s i gnificant organi sms from

20 repeat BCT+/CFT- cows the theme o f this proj ect swi tched to an

investigation of the means by which animal s might react to the

card test but not to the complement fixat ion tes t .

are : -

Pos s ib l e exp lanations for the occurrence of BCT+/CFT- animal s

( i ) I f the animal i s infected an unusual ly high specific

I gG1 : IgG2 rat io may cause the comp lement f ixat ion test

to prozone to such an extent that i t i s erroneous ly

cons idered to be negat ive .

( i i ) An animal , previous ly s ens i t i sed by ca l fhood s train 19

vaccinat ion , may come into contact wi th ki l l ed

( i i i )

� · abortus organisms o r wi th low doses o f l ive

organisms , such that an immune response is generated

wi thout infection becoming estab l i shed .

A chronica l ly infected animal may have only low leve l s

o f spec i f ic IgG1 and IgG 2 . Such low leve l s may cause

the card test to react but may no t cause a comp lement

f ixat ion react ion e i ther because o f the very low

antibody lev·e ls or because o f an unsuitab le I gG1 : IgG2 ratio .

( iv ) In the early s tages o f infect ion an IgM response

genera l ly precede s that of IgG . The greater s ens i t ivity

o f the card te st to specific I gM (Al lan � � 1 9 7 6 ) may

resul t in a card test reaction in the absence o f any

compl ement f ixat ion t i tre .

In Case ( iv ) i t i s to be expected that at a subsequent samp l ing

the animal would have become BCT+/CFT+ whi l s t in case s ( i ) , ( i i )

and ( i i i ) i t i s conceivab le that an animal could cont inue to be

BCT+/CFT- for some t ime .

In thi s the s i s an attempt has been made to inves t i gate some

of the circums tance s l eading to the exis tence of such BCT+/CFT­

animal s part icularly those which may be explained by po ints ( i )

and ( i i ) above .

109 .

Analys i s o f the data pre sented in Table XVI i l lus trates

that animals in heavi ly infec ted herds are more l ikely to be BCT+/

CFT- than animal s in non- infected or l ight ly infected herds . Thus

it is evident that the card test reactivi ty may be s t imu lated by

exposure to - b u t no t n e c e s s ar i ly i n fe c t i on b y - B ru c e l l a

organisms .

l t h a s b e e n obs erved by some worke r s that a card t e s t r e a c t ion

deve lops ear l ier than a CFT t i tre in infected animal s (Nico l e tt i ,

1 9 6 7 ;

19 7 7 ) .

Morgan � � . , 19 69 ; Davis , l 97 l ; Fens terbank , l9 7 3 ; P i e t z ,

I t cou ld be that some o f the BCT+/CFT- animals inve s t igated

above be long to this group . Resu lts of exper iments reported by

Fens terbank ( 1 9 7 3 ) and Pietz ( 19 7 7 ) indi cate that the average

per iod be tween deve lopment of a pos i t ive BCT reaction and a

po s it ive CFT reaction i s o f the order o f 10- 20 days . I t is

unl ikely therefore that this phenomena contributes s igni ficant ly

to the overa l l numbers o f BCT+/CFT- animals detected . T imbs

� � . ( l978a ) found that approximately . 1 6% of BCT+/CFT- anima l s ,

in infected herd s , detected at any part icular herd test might

be expected to become CFT+ at a later test . Inc luded in this

number would be animal s in the ear ly s t ages o f t i tre deve lopment

which had deve loped a BCT t i tre prior to producing a CFT t i tre .

1 10 .

CHAPTER 7 . GENERAL DISCUSSION

In choos ing a t e s t on whi ch to base a bovine bruce l lo s i s

contro l scheme i t has been cus tomary for each country t o carry out

p i lo t s tudies to determine not only the preva l ence and d i s tribut ion

of the d i se a s e but a l so to inve s t i gate the r e l a t ive mer i t s o f the various d iagnostic te s t s . Unt i l recently i t has been the SAT

that has invariably been chosen for use in routine te sting whi l s t

various modifications o f this t e s t such as mercapto-ethano l ,

rivano l , heat- lab i l e and Coombs tests have been used as supp le ­

mental tests . Among the deve loped countries New Zealand was

relative ly late in formu lating an eradi ca t ion programme .

Technol ogical improvements did , however , a l low the deve lopment o f

an automated comp l ement fixation test which was chosen a s the

definit ive test for use in the s cheme .

A part icu lar prob lem met with in j us t i fy ing the use o f the

Auto-Analyzer adaptat ion of the CFT was the theoret ical argument

that as the test is e s sentia l ly carried out at one d i lut ion , i t

may no t detect sera which cause a strong prozone react ion .

Inves t igat ion of thi s aspect has shown that in practice such

sera are in fact detected , and examination o f the serum di lut ion

gradient within the test sys tem has exp lained the ab i l i ty of the

automated te s t to detect such sera .

Examinat ion o f prozoning sera recovered from known infected

anima l s revealed that i t is the Bruce l l a specific I gG1 : IgG2 ratio that determine s the degree o f prozoning o f a serum in the

CFT . I t was a l so shown that a high antigen concentrat ion can

suppre s s the extent to which such prozoning occurs . The re su l t s

obtained confirm the findings o f P lackett and Al ton ( 1975 ) and

o f McNaught � �· ( 19 77 ) that sufficient quanti ties o f Bruce l l a

spec i f i c IgG2 can cause prozoning and abnormal reactions in the

CFT or prevent comp l ement fixat ion by I gG 1 altogether .

A l though a s er ious imbalance or the Bruce l l a speci f ic I gG1 :

I gG2 ratio wi l l l ead to obvious prozoning·, a l e s s serious imbalance

wi l l not be manife s t in this way but may be re f lected in sub­

optimal comp l ement fixat ion l eading to a reduced t i tre . The

extent to which a t i tre may be reduced i s d ependent upon the

degree to which the antigen concentration u sed varies from the

optimal ant igen concentration required for the part icu lar serum.

To ensure that the comp l ement fixat ion te st system is made as

s en s it ive as po s s ib le to prozoning sera , i t i s es s ential that

1 1 1 .

adequate care i s taken i n the t itrat ion o f ant igen . A ser ie s o f

known po s i t ive sera mus t be used i n antigen t i trat ions t o enable

an appreciation of the r e l at ive range o f op t imal values l ike ly to

b e obtained for a wide range o f sera . Thi s princip le s app l i e s

equal ly t o t i trat ions for the automated o r manual te s t s .

The exi stence o f prozoning s era can exp lain why some sera are

BCT+ yet apparent ly CFT- . S uch sera can be tho se wi th an

unusua l ly high proportion o f spec i fi c IgG2 which comp le te ly blocks

any comp l ement fixat ion . The frequency _wi th which such sera

occur and their l ikely s i gnificance in an erad icat ion programme i s

unknown . Animal s y ie lding prozoning sera have o ccas iona l ly been

re-b led after 2 -6 weeks and the prozone e ffect has usua l ly been

very much reduced - and o ften non- exi s tent - in the repeat sample .

B randon et a l . ( 197 1 ) and Wi l l iams and Green ( 19 7 6 ) found that at

about the t ime of parturi t ion serum IgG1 leve ls fa l l as thi s immuno­

globul in i s selective ly concentrated in the udder . The serum IgG 1 IgG2 ratio therefore undergoes cons iderable change . A revers ion

to a more normal rat io i s achieved some 4-6 weeks after partur it ion .

B e cause o f the repet i t ive nature o f testing in an eradicat ion

programme the exi s tence o f a sma l l number of cat t le with low IgG1 I gG2 rat ios at any par t icular t ime i s unl ikely to unduly affect

t e s t ing progress . However , care mus t be taken in the interpretat ion

o f test results from recent ly calved or aborted cows , and the

exi stence of potentia l ly fa l se negative CFT resu l t s should be

recognised .

Inve st igations o f the way in which the IgG1 changes in bovines are practica l ly non-exis tent .

I gG2 ratio

Because of the

e ffect a change in the B ruce l la specific IgG 1 : I gG2 ratio may have

on re sul t s o f the CFT , examination o f the physio logical changes in

this ratio particular ly during the incubat ion phase o f infect ion

and a l so at about the t ime o f partur i tion , is required .

Upon examinat ion o f scheme record s , i t was apparent that more

animal s were l ike ly to b e BCT+/CFT-- in infected herds than in non­

infected herd s . Thi s lead to . the - hypothes i s that animal s

previous ly vaccinated with l iving s train 1 9 Br . abor tus as calve s ,

then exposed to , but not infected by , the viru l ent organism as

1 1 2 .

adu l t s , may deve lop a BCT t i tre in the absence of any t i tre to the

CFT . A lthough studies u s ing l i ght cha l lenges of viru lent organi sms

were not carried out , inoculation with ki l l ed organisms did

stimul ate sero logical response s and proved that such exposed

animal s could in fact be BCT+/CFT- . On the other hand i t was a l so

shown that other animals s imi lar ly exposed could be BCT- /CFT+ .

Another mechani sm by whi ch the BCT may produce fa l se p o s i t i ve react ions but which has not been invest igated in this thes i s i s the

pre sence of res idual strain 19 vacc inat ion titres . As IgM is the

princip l e immunoglobul in s t imu lated in re sponse to Br . abortus

strain 19 vaccination , it has been pos tu l ated that BCT+/CFT­

reactions in non- infected cat t l e may be due to the e f fect o f a

residual vaccinat ion t i tre (Al ton � �. , l975b ) . A l l an � �·

( 19 75 ) demonstrated that spec i f ic IgM ant ibody is up to ten t imes

more e fficient on a we ight bas i s at producing a BCT reaction than

i s e i ther IgG 1 or IgG2 .

S train 19 vaccination o f a l l female ca lves was compul sory in

New Zea land between 1966 and 1 97 6 . The pre sent vo luntary

vaccination programme wi l l eventua l ly be phased out when the

prevalence of the d i sease i s reduced to an acceptab le leve l . As

cows which were not vaccinated as calves begin to form a s ignif i cant

propor t ion o f a l l cattle be ing te s ted , i t i s expected that the

prevalence of unusual sero logical reactions wi l l dec l ine .

1 1 3 .

APPENDIX I REAGENTS

( i ) B arbi tal buffered sa l ine (Kabat and M ayer 1 9 6 1 ) pH 7 . 3 )

( i i )

( i i i )

Sod ium chlor ide 85 g

S odium 5 , 5 -diethyl-barbi turate 3 . 7 5 g

5 , 5 -diethy lbarbitur ic acid 5 . 75 g

Magnes ium chloride ( anhydrous ) 0 . 47 5 g

Calc ium chlor ide ( anhydrous ) 0 . 185 g

Make up to 2 l i tres wi th disti l led water to prepare a

concentrated stock so lut ion . Di lute stock so lution

1 : 5 ( l ml stock p lus 4 m l water ) to prepare working

s trength buffer .

Alsever ' s solut ion

Glucose 18 . 6 6 g

Sod ium ch loride 4 . 18 g

Sod ium c i trate 8 . 00 g

Citric acid 0 . 5 5 g

Dis t i l led water to 1000 ml

The solution p laced in screw- capped bott le s and

au toc laved .

Phosphate buffered s a l ine l ml

Di- sod ium hydrogen phosphate Na2HP04 1 2 g

Sodium d i-hydrogen phos phate NaHlO 42H20 2 . 5 g

Sodium chloride NaC l 8 5 g

Dis t i l led water to 1000 ml

( iv)

(v)

Phos phate buffer

S o l u t ion A :

Sodium d ihydrogen phosphate NaH 2PO 4 2H 20

So lu t ion B :

D i - sodium hydrogen phosphate Na2HP04

Adj u s t solut ion A to pH 8 with

Pheno l sal ine

Sodium chloride

Phenol

D i s t i l led water to

1 M

31 . 202 g/ 1

28 . 39 g/ 1

solution B .

8 . 5 g

5 . 0 g

1000 m l

( vi.) Immunoelectrophores i s ( IEP ) agar

Rad ial immunodiffus ion (RID) agar

Ge l diffus ion (GD) agar

( a ) Prepare barbit urate buffer pH 8 . 6

S od ium die thy lbarbi turate

0 . 1 M Hydrochloric acid

Sodium azide

D i s t i l led water to

( b ) Add 3 g Ion agar or Nob l e agar

to lOO ml barbiturate , heat to

d i ssolve .

9 g

65 ml

0 . 5 g

1000 ml

Dis tribute in sui tab l e vo lumes into screw capped bot t les .

When required for use add an equal vo lume o f barbi turate

buffer and heat in a bo i l ing water bath until d i s so lved ,

then pour onto glas s p l ates . For RID agar add anti-

serum at suitable d i lut ion to buffer , heat buffer and

agar to 58 °C , mix qui ckly and pour p late .

( vi i )

L l5 .

I . E . P . , R . I . D . , G . D . S tain

( a ) Rinse s o lution

Methanol 5 parts

D i s t i l led water 5 parts

Glacial acetic acid 1 part

( b ) S tain

D i s so lve 1 gm Amido b lack in 1 l itre rinse solut ion .

( c ) Procedure

After deve lopment of reaction in agar p late rinse p late

for 3 days in frequent change s of 1 . 5% NaCl .

a further day in d i s t i l led water .

Wash for

Cover p l ate wi th f i l ter paper and dry at 37 °C overnight .

S tain by immersing in stain solution for approx 30 mins

then transfer to r inse so lution for 15-20 minute s to

decolouri se . Dry .

1 1 6 .

REFERENCES

Anon . ( 1954) : B io logical standard ization . World Heal th

Organisat ion Technical Report Series No . 86 . WHO Geneva .

Anon . ( 19 65a ) : S tandard diagnostic complement fixat ion method and

adapt i on to mic�o t e s t . u . s . Pub l i c He a l th S e rvice , Pub l i c He a l th Monograph , No . 7 4 ; U . S . Government Printing Office , Washington ,

D . C .

Anon . ( 19 75b ) : S tandard agglutination test procedures for the

diagnosis of bruce l los i s . USDA. Animal Heal th Divis ion ,

National Animal Diseases Laboratory Diagnostic Reagents Manual

6 5D . Ames , Iowa .

Anon . ( 19 7 1 ) : Expert Committee on Bruce l los i s . 5 th Report ( 19 7 1 ),

WHO Technical Report Series No . 464, FAO Agricu l tura l S tud ies

No . 85 . Geneva .

Anon . ( 1 97 7a ) : Surve i l lance . 197 7 . No . 2 . Animal Hea l th Divi s ion ,

N . Z . Ministry o f Agricu l ture and Fisherie s .

Anon ( 19 7 7b ) : Bruce l los is Re search : An Eva luat ion . Subcommi ttee

on Bruce l losis Research , Committee on Animal Hea l th , Nat ional

Research Counci l , Nat ional Academy o f Science s ; Washington D . C .

Ad lam , G . H . ( 19 78 ) :

New Zealand :

42-43 .

The eradi cat ion of bovine bruce l losis in

H i s tory and obj ectives . N . Z . ve t . J . 26 :

Ahvonen , P . ; Jansson , E . ; Aho , K . ( 1969 ) : Marked cross­

agglutinat ion between Bruce l la and a subtype o f Yersinia

enteroco l i t ica . Acta path . microbial . scand . ll= 291- 295 .

Al lan , G . S . ; Chapp e l , R . J . ; Wi l l iamson , P . ; McNaught , D . J . ( 1 97 6 ) :

A quantitat ive compari son o f the sens i t ivity o f serological

tests for bovine bruce l losis to d i f ferent ant ibody classes .

J . Hyg. Camb . �: 287 - 29 8 .

Al l sup , T . N . ( 1969 ) : Abortion i n sheep associated with Bruce l l a

abortus infection , Vet . Rec . �: 104- 108 .

Allsup , T . N . ( 19 74 ) : Failure to demonstrate Bruce l la infect ion in

ewes exposed to natura l bovine infection . Ve t . Rec . 94 :

1 83- 186 .

Al ton , G . G . ( 19 7 7 a) : Experiences with· Bruce l la vaccine s . In

Bovine Bruce l losis . An Internat ional Symposium . Editors ,

R . P . Cra�ford and R . J . Hidalgo . Texas A & M University

Press . ----- ---- - ·----

117 .

'

Alton , G . G . ( 19 7 7b ) : S tandardised complement- fixation test for

bovine bruce l los is . Aus t . ve t . J , 12= 394-400 .

Al ton , G . G . ; Jones , L . M . ( 1965 ) : Laboratory Technigues in

Bruce l los i s . WHO Monograph series No . 55 . World Heal th

Organisation , Geneva .

Al ton , a . G . ; Jones , L . M . ; �ietz , D . E . ( l9 7§a ) : Laboratory

Technigues in Bruce l los i s . WHO Monograph S eries No , 5 5 ,

2nd ed . World Hea lth Organisation , Geneva .

Al ton , G . G . ; Maw , J . ; Rogerson , B .A . ; McPherson , G . G . ( 1975b ) :

The serological d iagnosis o f bovine bruce l los i s : An

evaluation o f the complement fixat ion , serum agglut inat ion

and Rose Bengal tes t s . Aus t . vet . J , 5 1 : 5 7- 6 3 .

Ameraul t , T . E . ; Lambert , G . ; Manthei , C .A . ( 19 62 ) : The heat

s tab i l i ty of Bruce l la agglut initts in bovine serum .

Am . J . vet . Res . 11= 1023- 1026 .

Amerau l t , T . E . ; Manthei , C .A. ; Goode , E . R. ; Lambert , G . ( 196 1 ) :

A heat- inact ivat ion test for d i f ferent iating specif ic and

non- specific agglut inat ing reactions for bovine bruce l losi s .

Am . J . vet . Res . 2 2 : 564- 5 69 .

Anderson , R . K. ; Jenness , R . ; Brumfield , H . P . ; Gough , P . ( 19 64 ) :

Bruce l l a-agglut inating ant ibodies . Re lat ion of mercaptoethanol

s tab i l i ty to comp lement fixat ion . Science 143 : 1 334- 1 335 .

Anderson , R . K . ; P ie t z , D . E . ; Nel son , C . J . ; Kimberling , C . V . ;

Werring , D . F . ( 19 6 2 ) ; Epidemiologic s tudi e s of bovine

bruce l losis in prob lem herds in Minnesota . Proc . U . S .

l ive s tk . sani t . As s . 66th Ann . Mtg. pp 109- 1 1 8 .

B ang , B . ( 1897 ) : The e t iology o f ep i zootic abortion . J . camp.

Path . Ther . 10 : 1 25 - 149 .

Bang , B . ( 19 06) : Infect ious abortion in cat t l e .

Ther . 1 9 : 19 1 - 202 .

J . comp. Path .

Barclay , D . F . ( 19 7 7 ) : Bruce l los i s : Retention o f bu l l s . - adult herd

s laughters . \ I '

S tate vet . J . 3 2 : 107- 109 .

Beck , C . C . ; E l l i s , D . J . ; Fichter , G . J . ; Laiho , E .R . ; Whi tehead , G . L . I·

( 1964) : Bruce l l a tube agglutination t itres resul t ing from the

use of syringes containing viab le s train 19 vaccine .

J . Am . vet . med . Ass . �: 6 20-625 .

1 1' 8 .

Becton , P . ( 1976 ) : Symposium : B iological agents caus ing infect ions

·in dairy cat t le . Bruce l los i s s tatus report . J . Dairy Sci . 11:

l 16 3- 1 165 .

Beh , K . J , ( 1 9 73 ) : Distribut ion of Bruce l la antibody among immuno­

globul in classes and a low molecu lar weight ant ibody fract ion

in serum and whey o f Gatt l e . Rgs. vet. Ssi. li= JS l - 324 .

Beh , K . J . ( 1974) : Quantitat ive Dis tribut ion of Bruce l la ant ibody

amongst immunoglobul in c l asses in vaccinated and infected

catt le . Re s . vet . Sci . 1 7 : 1 -4 .

Beh , K . J . ( 1 975 ) : Immunoglobul in class s pecificity o f non­

agglutinating ant ibody produced in cattle fo l l owing Bruce l l a

abortus 45 / 20 vaccinat ion . Aus t . vet . J. 5 1 : 48 1 - 483 .

Beh , K . J . ; Lasce l les , A .K . ( 19 7 3 ) : The use of the ant iglobu l in

test in the diagnosis o f bovine bruce l los is . Re s . ve t . Sc i .

14 : 239- 244 .

Bendixon , H . C . ; B lom , E . ( 1947 ) : Invest igat ions on bruce l los is in

the bovine male , with speci�l regard to spread o f the di sease

by art i ficial insemination . Br . ve t . J. 10 3 : 3 3 7 - 345 .

Berman , D . T . ( 195 6 ) : The e f fect of bacterins containing Pas teure l la

mult icida on agglutinins for Bruce l l a in cat t le . Proc . U . S .

l ive stk . s anit . Ass . 60th Ann . mtg. pp 97- 103 .

B ickne l l , S . R . ; Be l l , R .A . ; Richards , P .A . ( 19 76 ) : Bruce l la

abortus in the bitch . Vet . Rec . 99 : 85 -86 .

B iege le i sen , J . Z . ; Moody , M . D . ; Marcus , B . B . ; Flynt , J .W. ( 19 6 2 ) :

The use of fluorescin- label led ant i -Bruce l la suis globu l in for

demons trat ing Bruce l la antigen in animal t issues . Am . J .

ve t . Res . 2 3 : 592-595 .

B lake , G . E . ; Manthei , C .A . ; Goode , E . R . ( 1952) : A mi lk p late

test for the detect ion of bruce l los i s . J. Am. vet . med . Ass .

1 20 : 1 - 6 .

Bonneau , M . ; Valette , L . ; F i le t , C . ( 1970 ) : Etude pre l iminaire

e t comparee du vaccin P . B . chez les bovine . Int . Symp . on

B ruce l los i s , Tunis 1968 . Symp. Ser . Immunobiol . s tandard . 1 2 :

6 3-68 .

Brandon , M .R . ; Watson , D . L . ; Lasce l les , A .K . ( 19 7 1 ) : The mechani sm

o f transfer o f immunoglobul in into mammary secretion o f cows .

Aust . J . exp. B iol . med . Sci . �: 6 1 3-623 .

1 1 9 .

Braude , A . I . ( 195 la ) : S tudies in the pathology and pathogenes i� of

exper imental bruce l los i s . I . A comparison of the patho­

genici ty of Bruce l l a abortus , Bruce l l a me litens i s and

Bruce l l a � for guinea pigs . J . infect . Dis . �: 7 6-86 .

Braude , A . I . ( 195 lb ) : · S tudies in the pathology and pathogeneis o f

oxp e � im•n t A l bruee l lo• i • · I I . The fo�ma tion o f the hep a t i c granuloma and i ts evolution . J . infect . Dis . 89 : 8 7 -94 .

Bruce , D . ( 1887 ) :

Ma l ta fever .

Note on the discovery o f a micro-organism in

Practitioner 3� : 1 6 1 - 1 70 .

Bruce , W . ; Jones , L . M . ( 1958 ) : The use of ge l diffus ion precipitin

plates in the s tudy of Bruce l l a . Bu l l . Wld . H l th . Org. li=

187 - 1 9 6 .

Buck , J .M . ( 1930 ) : S tudies o f vaccination during ca l fhood to

prevent bovine infectious abortion . J . agric . Re s . 4 1 :

667- 689 .

Buck , J .M . ; Cotton , W . E . ; Smi th , H . E . ( 1938 ) : Vaccinat ion o f

calves and year l ings agains t Bang ' s d isease , USDA Technical

Bu l l e t in No . 658: 1 - 6 .

Buck , J .M . ; Creech , G . T . ; Ladson , H . H . ( 19 1 9 ) : Bacterium abortus

infect ion in bu l l s . J . agric . Res . ll: 239- 246 .

Buddle , M . B . ; Boye s , B .W . ( 195 3 ) : A Bruce l la mutant caus ing

genital d i sease o f sheep in New Zealand .

145 - 1 5 3 .

Aus t . ve t . J . 29 :

"

Burki , F . ( 195 7 ) : Uber Immunant ikorper gegen Bruce l l a abortus

im B lutserum von Rindern , mog l ichkei ten 24 ihrem Nachwei s und

ihre klmische Bewertung . Zb l . ve t . med . 4 : 833-87 0 .

Burki , F . ( 1963 ) : The complement fixat ion test for d i s t inguishing

infected cat t le from those vaccinated against bruce l los i s .

Bu l l . Off . int . Epiz , �: 4 19 -43 2 .

Cameron , H . S . ; Kendrick, J .W . ( 1955 ) : The diagnos i s o f bruce l losis

o f the mammary g land by a s ero logic mi lk tes t . Pre l iminary

report . Proc . U . S . l ive s tk . sanit . Ass . 59th Ann . mtg.

pp . 1 38- 1 4 1 .

Campbe l l , A . D . ; Rodwe l l , A .W . ( 1945 ) : The relationship o f

dosage and s i te of inoculation t o the agglutinin response to

Bruce l la abortus s train 19 vaccine ; a compari son of the

subcutaneous , intracutaneou s and intracaudal rou te s .

J . comp. Path , Ther . · .21= 277·- 289 .

------- --- --- . ------ --,......-

1 20 .

Carmicahel . L . E . ( 1976 ) : Canine bruce l losis : An annotated review

with se lected cautionary comment s . Theriogeno logy 6 : 105 - 1 1 6 .

Carmichael , L . E . ; Bruner , D .W. ( 1968 ) : Characteris tics of a newly­

recognised species o f Bruce l la responsible for infectious

canine abortions . Cornel l Vet . �: 579-59 2 .

Carpenter , C . M . ( 19 24 ) : Bacterium abortus invas ion of the t i ssues

o f ca lves from the ingestion of infected mi lk . Corne l l Vet . 1 4 : 1 6 - 3 1 .

Carrere , L . ; Roux, J . ( 195 2 ) : Hemaggl�tination pas sive

d ' hematies sensibi l i sees per antigenes bruce l l iques ou des

subs tances solubles speci fique s . Annls . Ins t . Pasteur, Paris .

8 3 : 810- 8 1 3 .

Castaneda , M . R . ( 1950 ) : Surface fixation . A new method o f

detecting certain immunologic react ions . Proc . Soc . exp.

B io l . Med . 73 : 46-49 .

Chappel , R . J . ; Wi l l iamson , P . ; McNaught , D . J . ; Da l l ing , M . J . ;

Al lan , G . S . ( 197 6 ) : Radioimmunoas say for antibodies against

Bruce l la abortus : A new serological test for bovine

bruce l los i s . J . Hyg. , Camb .ZL: 369- 376 .

Cho , H . J . ; Ingram , D . G . ( 19 7 2 ) : Mechanisms of prozone formation

in agglut ination reaction . Can . J . Microbia l . 18 : 449- 45 6 .

Christensen , N . O . ( 1 948 ) : S tudies on the agglutinin formation

in bruce l l ar infection of the genitals of the bul l .

Acta path . microbia l . scand . 25 : 20 2 - 209 .

Chris tie , T . E . ; Kerr , W . R . ; McCaughey , W . J . ( 1968 ) : Bruce l losis

eradicat ion in Northern Ire land . Vet . Rec . 8 2 : 1 7 6 - 183 .

C legg , F . G . ; Rorri son , J .M . ( 19 68 ) : Bruce l la abortus infect ion

in the dog . A case of po lyarthri tis . Res . vet . Sci . 9 :

183- 185 .

Corbe l , M . J . ( 197 2 ) : Identification o f the immunoglobul in c lass

active in the Rose Bengal P late test for bovine bruce l los i s .

J . Hyg.;> Camb . 1Q: 7 79 - 795 .

Corbe l , M . J . ( 1973a ) : The d irect f luorescent antibody tes� for

detection of Brucel la abortus in bovine abortion materia l .

J . Hyg. , Camb . ll= 1 23- 1 29 .

Corbel , M . J . ( 1973b ) : Character�sation o f antibodies active in

the Rose Bengal p late test . Vet . Rec . 90 : 484-485 .

1 2 1 .

Corbe l , M . J . ( 19 7 3c ) : S tudies on the mechanism o f the Rose Bengal

p late test for bruce l los i s . Br . vet . J . ill' 157- 166 ,

Corbe l , M . J . ( 19 7 5a ) : The sero logica l re lationship between

Brucella spp . , Yers inia enteroco l i tica serotype IX and

S almone l la sero types o f Kauffmann-Whi te group N . J . Hyg. ,

£!.m!l· ll• 1 5 1 • 1 7 1 .

Corbel , M . J . ( 1975b ) : The immuno logical properties o f Bruce l l a

r ibosomal preparations . International sympos ium on

B rucellosis I I . Deve lop . biol . s t andard . 1.!.: 1 15- 1 2 2 .

Corbe l , M . J . ( 197 6 ) : The immune· re sponse to Bruce l la abortus 45 / 20

adj uvant vaccine in terms o f immunog lobulin c las s . Inter-

nat ional Sympos ium on Bruce l losis II . Deve lop. bio l .

s t andard . 3 1 : 141 - 144 .

Corbe l , M . J . ; Bracewe l l , C . D . ( 197 6 ) : The sero logical re sponse

to rough and smooth Bruce l l a antigens in ca ttle vaccinated

wi th Bruce l l a abor tus S train 45 / 20 adj uvant vaccine . Inter-

nat ional Sympos ium on Bruce l losis II .

standard . 3 1 : 35 1 - 35 7 .

Deve lop. b io 1 .

Corbe l , M . J . ; Cul len , G . A . ( 1970 ) : Differentiat ion o f the

serological re sponse to Yers inia enterocoli tica and Bruce l la

abortus in cat t le . J . Hyg.', Camb . .2.§_: 5 19-530 .

Corbe l , M . J . ; Day , C . A. ( 19 7 3 ) : As ses sment of ind irect

haemagglutinat ion procedures for the serologica l diagno s i s

o f bovine br-uce llos i s . Br . ve t . J . 1 29 : 480-49 1 .

Corbe l , M . J . ; Morgan , W . J . ( 1975 ) : Proposal for . minimal standards

for descript ions o f new species and b iotypes o f the genus

Bruce l l a . Int . J . sys t . Bact . 11: 83-89 .

Cotton , W . E . ( 19 32 ) : Efficacy of different s tra ins o f Bruce l la

abortus as immunis ing agents agains t infect ious abortion • ..

J . agric . Res . �: 705 - 7 24 .

Cotton , C .M . ( 1953 ) : An intens ive s tudy o f post-vaccination

responses in groups o f ca lves vaccinated intracutaneously and

sub-cutaneous ly wi th s train 19 Bruce l la abortus vaccine .

Am . J . vet . Res . li= ,337- 342 .

Cotton , W . E . ; B uck , J .M . ( 19 3 1 ) : Bureau o f Animal Industry

researches on infectious abortion . J . Am . ve t . med . As sn .

li= 306- 325 .

. 1 2 2 .

Cotton , W . E . ; Buck, J .M . ( 1934) : The s tatus o f vaccination against

B ang ' s d isease . J . Am. vet . med . Ass . 84: 329- 340 .

Cotton , W . E . ; Buck , J . M . ; Smi th , H . E . ( 1933 ) : Efficacy and safe ty

o f abort ion .vaccines prepared from Bruce l la abortus s trains o f

di fferent degrees Q f viru lenQe , .;.J.a., ...:i;a.lilg.:ior.ilii.olls., ...... Rio:ie�s • �: 4- 8 .

Cot ton , W . E . ; Buck , J .M . ; Smi th , H . E . ( 1934 ) : S tudies of five

Bruce l l a abortus s trains as immunis ing agents against Bang ' s

d isease ( Infectious abortion ) , J . Am . ve t . Med . Assn , 38 :

232- 247 .

Cros sman , P . J . ; Bonson , M . D . ( 1968 ) : Abort ion in a donkey associated

wi th Bruce l la abortu s . Ve t . Rec , �: 607- 608 .

Cruickshank , R . ; Du guid , J . P . ; Marmion , B . P . ; Swain , R . H . A . ( 1975 ) ;

Med ical Microb io logy Vol . l l , 1 2 th ed . Churchi l l Livings tone ,

Ed inburgh .

Cul len , G . A . ; Corbe l , M . J . ( 1970 ) : Observations on some pos sible

causes o f variat ion in the t i tre of Bruce l l a antibod ies in

catt l e . Vet . Re c . 8 7 : 101- 106 .

Cunningham , B . ( 1966 ) : Observat ions on the loca l react ions

fo l lowing the inocu lat ion of cattle wi th ki l led Bruce l l a 45 / 20

adj uvant vaccines . Ir . ve t . J . 20 : 205 - 209 .

Cunningham , B . ( 1967 ) : The Coombs test for bruce l losis in cattle .

Vet . Rec , 80 : 5 27 - 5 28 ,

Cunningham , B . ( 1968 ) : The contro l and erad icat ion of Bruce l los i s .

Vet . Rec . 8 2 : 7 - 1 1 .

Cunningham , B . ; O ' Connor , M . ( 197 1 ) : The use o f ki l led 45 / 20

adj uvant vaccine a s a d iagnostic agent in the fina l stages

of the eradication of bruce l losis . Ve t . Rec . 89 : 680- 686 .

Cunningham , B . ; O ' Re i l ly , D . J . ( 1968 ) : Bruce l l a abortus vaccines :

Agglut inin responses in b lood serum and mi lk fol lowing

vaccination of cattle o f various ages wi th live S . 19 and

ki l led 45 / 20 adj uvant vaccines . Vet . Rec . �: 678- 680 and 689 .

Dalrymp le-Champneys , W . ( 1960 ) : Bruce l la Infection and Undu lant

Fever in Man . Oxford U�ivers i ty Press . London .

Daniel , M . J . ( 1966) : A pre l iminary survey o f the incidence o f

bruce l losis , leptosp�ros i s and salmone l losis i n red deer in

New Zealand . N . Z . J l . Sci . 9 : 399-408 .

J. 23 .

Danie l , M . J . ( 1967 ) : A survey o f d i seases in fal low , Virginia and

Japanese deer , chamois , thar and feral goats and p igs in

New Zealand . N . Z . J l . Sci . 1 0 : 949-963 .

Davidson , I . , S lavin , G . ( 1958 ) : Dis infection o f mi lking uni t s in

the control of mas t i t i s due to S treptococcus agalactiae .

Vet . Rec . 7 0 : 893-898 .

Davies , G . ( 19 7 1 ) : The Rose Benga l test . Ve t . Rec . 88 : 447 -449 .

Dhennin , L . ( 19 7 3 ) : Resultat de L ' e tude comparee de sept vaccines

ant i-bruce l l iques . Bu l l . Acad . Ve t . Fr . 46 : 17 1 - 190 .

Diaz , R . ; Levieux , D . - ( 197 2 ) : Role respectif en sero logie de la

bruce l lose bovine des antigenes e t de immunoglobu l ines G l

et G2 , dans les tests d ' agg lutinat ion , de Coombs e t au Rose

Bengal a insi que dans le phenomene de zone . C . r . hebd . Seanc .

Acad . S c i . , Pari s . 2 7 4 : 15 92- 15 9 6 .

Dickey , J .W . ; Forbus , W . D . ( 1945 ) : Chemotactic properties o f

Bruce l la suis . Am . J . Path . 2 1 : 195 - 202 .

D idovets , S . R . ( 1965 ) : Programme d ' eradication de la bruce l lose

ovine en Ukraine . B u l l . Off . int . Epizoot . �: 1 2 25 - 1 23 3 .

Doy l e , T .M . ( 19 35 ) : The d i s tribut ion o f Br . abortus in the sys tem

of "carr ier" cows . J , Comp. Path . Ther . 48 : 1 9 2- 2 17 .

Duncan , J . R . ; Wi lkie , B . N . ; Hiestand , F . ; Winter , A . J . ( 1 9 7 2 ) : The

serum and secre tory immunoglobul ins of catt le : character isation

and quantitat ion . J . Immun . 108 : 965-976 .

Edington , B . H . ; Donham , C . R . ( 1939 ) : Infection and reinfect ion

exper iments wi th Bang ' s di sease . J . agric . Res . 5 9 : 609 - 6 1 8 .

Edwards , S . J . ; de Ropp , R . S . ; McLeod , D . H . ( 1945 ) : A s tudy o f the

immuno logical properties and infectivity o f Bruce l la abortus

strain 45 / 20 , McEwen , in catt le . Vet . Rec . 5 7 : 259- 265 ,

Ehr lein , H . J . ; Schimmelp fennig , I . R . ; B i sping , W . ( 196 3 ) :

Bruce l losis in dogs . D t . t ierarzt l . Wschr . 7 0 : 35 3-35 4 .

E lberg , s . s . ( 1960 ) : Ce l lu lar Immunity . Bact . Rev . �: 67-95 .

Elberg , S . S . ; Henderson , D .W . ( 1948 ) : Respiratory pathogenicity

· o f Bruce l la . J . infect , Dis . �: 302- 306 ,

Ellio t , R . E .W . ; Pullan , N . B . ( 1�( 3 ) : Bovine bruce l los i s in

New Zealand : A control program based on the automated

124.

comp l ement fixation tes t .

8 7 - 90 .

Advance s in Automated Analys is 7 :

Evans , A . C . ( 19 18 ) : Further s tudies on Basterium ibortus and

re l ated bacteria . J . infect . Dis . 11: 580- 5 9 3 .

Farre l l , I . D . ; Robertson , L . ( 19 7 2 ) :.

A comparison o f various

se lective med i a , inc lud ing a new se lect ive medium for the

isolation of Bruce l lae from mi lk . J . appl . Bact . ]l : 625- 630 .

Fenske , G . ( 1977 ) : Sens itivity and specificity o f serological

tes ts for bovine bruce l los i s . Mh . VetMed . 3 2 : 18- 2 3 .

Fens terbank , R . ( 1 9 7 3 ) : Assessment o f the va lue o f the react ion

with Rose Benga l on the sera of he i fers experimenta l ly

infected wi th Bruce l la abor tus . Report No . 109 from the

XLi s t Gene r a l S e s s ion o f the O . I . E . Commi t tee , Paris 2 1 - 26

May 19 7 3 .

F ink , C . W . ; Lo Spal luto , J . J . ( 1967 ) : The effect o f 7 S and 19S

ant ibod ies on the pr imary response to Sa lmone l la typhi

antigens . Immunology 11= 259- 266 .

�i tch , C . P . ; Boyd , W . L . ; B i shop , L . M . ( i9 38 ) : A study o f the

vaginal content of pregnant Bang- infected cows for the

presence o f Bruce l la abortus . J . Am . vet . med . Ass . 9 2 :

1 7 1 - 177 .

F i tch , C . P . ; Boyd , W . L . ; Ke l ly , M . D . ; B ishop , L . M . ( 1 941 ) : An

extended study of fema le o ffspr ing o f pos itive Bang ' s

d iseased catt l e . J . Am . ve t . med . Ass . 22= 413-414 .

Fleischhauer , G . ( 1937 ) : D ie Abortus -Bang-Ringprobe (ABR) .

Zur Feststel lung von Bangverdachtigen Vol lmi lchiproben .

Berl . Much . tierarztle Wchnschr . 21= 5 27-5 28 . C i ted in

O litzki , A. ( 1970) in Immunological me thods in Bruce l lo s i s

I . p . 136 S . Karger , B a se l .

Fos ter , J .W . ; Ribi , E . ( 1962 ) : Immunological ro l e o f -Bruce l l a

abortus ce l l wal l s . J . Bact . 84 : 258- 268 .

Freeman , N . C . ; Felsenfe ld , 0 . ; Eve land , W . C . ( 1955 ) : S l ide

haemagglutinat ion te s ts with 0 ant igens of enteric

organi sms and B ruce l la . Am . J . clin. Path . 25 : 332- 335 .

Freeman , B .A . ; McGhee , J . R . ; Baughn , R . E . ( 1970 ) : Some

phys i ca l , chemica l , and taxonomic features o f the so lub le

antigens of the Bruce l lae . J . infect . Dis . 1 2 1 : 5 2 2-5 2 7 .

1 2 " · .

Frommhagen , L . H . ; Martins , M . J . ( 1 9 6 3 ) : A comparison of f luoresccin

labe 1 led y-globu l ins pur i fied by rivanol and DEAE chromato­

graphy . J . Immun . 90 : 1 1 6- 1 20 .

Gai l lon , R . ; Ripau l t , J . ; S tudievic , C . ; Dausse t , J . ( 1966 ) ;

U t i l i sa t ion pratique de L ' Auto -Analyseur pour de sero­

d iagnos tic de la syphi l i s (e tude de 1009 serums ) u t i l i sant

un ant igene card io l ipid ique . Path . B iol . , Paris 14 : 95 2- 95 4 .

Gi lman , H . L . ; Wagner , W . C . ( 1959 ) : The eva luat ion of bruce l lo s i s

vaccinat ion a t 4 and 8 months o f age . Corne l l Vet . XLIX :

399-408 .

Glenchur , H . ; Zinneman , H . H . ; Hal l , W . H . ( 1961 ) : S i gni f icance

of the b locking ant ibody in experimental bruce l los i s .

J . Immun . 86 : 421 -426 .

Goode , E . R . ; Manthei , C .A . ; Amerau l t , T . E . ( 1956 ) : Fur ther

stud i e s on durat ion of immuni ty to bruce l lo s i s induced in

cal f-vaccinated cat t le with s train 19 vaccine . Proc . U . S .

l ive s tk . sanit . Ass . 1 95 6 : 89-96 .

Gower , S . G .M . ; Wright , E . C . ; Davie s , G . ; Morgan , W . J . B . ;

Hopkinson , W . I . ; Gibbs , D . F . ; Bennett , E . J . ( 1 9 76 ) : An

automated Rose Bengal agglu t inat ion test us ing the ADAM

sys tem . Vet . Rec . 21= 544-547 .

Hadj u , S . ( 1963 ) : Antiglobul in tests for diagnosis of bovine

bruce l los i s . Bu l l . Off . int . epizoot . 60 : 433·445 .

Hal l , W .T . K . ; Ludford , C . G . ; Ward , W .H . ( 1976 ) : Infect ion and

sero logical responses in catt le given 45 / 20 vaccine and

later challenged with Bruce l la abortus . Aus t . vet . J . 11:

409-41 3 .

1 2 6 .

Hal l , W . H . ; Manion , R . E . ( 195 3 ) : Compari son· o f the Coombs test wi th

o ther methods for Bruce l la agglut inins in human serum .

J . c l in . Inve s t . 3 2 : 96- 106 . �

Hal l iday , R . ( 1968 ) : The e f fect o f pas s ive immunisation agains t

Bruce l la abortus on act ive product ion o f Bruce l l a abortus

agglut inins in young rats . J . Path . B ac t . . .2.2, : 137- 148 .

Hammer , D . K . ; Kickho fen , B . ; Schmid , T . ( 19 7 1 ) : Detect ion o f

homocytotrop ic antibody as sociated with a unique immunoglobu l in

c lass in the bovine species . Eur . J . Immun . 1 : 249 - 25 7 .

Har ing ,C .M . ( 19 38 ) : Vaccinat ion agains t Bang ' s disease in an

infected dairy herd with United S tates Bureau o f Animal

Indus try Bruce l la abortus s train 19 . J . Am. vet . med . As s . 9 2 :

5 2- 6 0 .

Har ing , C .M . ; Traum , J . ( 19 37 ) : Observat ions of pathogenic and

ant igenic e f fe cts o f Bruce l l a abortus , United S tates Bureau

o f Animal Indus try s train 19 . J . agr ic . Re s . IV : 1 1 7 - 1 28 .

Har ing , C .M . ; Traum , J . ( 1943 ) : The e f fect o f Bruce l la abortus

S train 19 on catt le o f var ious ages and its bearing on adu l t

cattle vaccinat ion . Proc . U . S . l ive s tk . sani t . As s . 47 :

42-46 .

Harr i s , H . J . ( 1950 ) : Bruce l losis (Undu lant Fever) . C l inical and

subclinica l . 2nd Ed . p . 285 Pau l B . Hoeber Inc . N . Y .

Hase , S . ; Th . Rietsche l , E . ( 19 7 6 ) : I so lation and analys i s o f the

lipid A backbone . Eur . J . B iochem . 6 3 : 101- 107 .

Hes s , W . R . ( l953a ) : S tudies on a non- specific Bruce l l a­

agglutinat ion substance in bovine serum . I . The

different iat i on of the spec ific and non-spec ific agg lutinins

by heat treatment . Am . J . ve t . Re s . Ji: 1 92- 194 .

Hes s , W . R . ( l953b ) : S tudies on a non- specific Bruce l l a-agglut inat ing

substance in bovine serum . II . I s o lation and pur i fic�t ion o f

the Bruce l la-agglutinating substances . Am . J. vet . Res . 1 4 :

. 1 95- 197 .

Hinsdi l l , R .D . ; Berman , D . T . ( 1968 ) : Ant igens o f B ruce l l a abortus .

I I . Toxicity for macrophages in cul ture . J . infect . Dis . �:

307- 3 1 6 .

12 7 .

Ho l land , J . J . ; Picke t t , M . J . ( 1958 ) : A ce l lu lar bas i s o f immuni ty

in experimental Bruce l l a infect ion . J . exp. Med . 108 : 343- 359 .

Hudd leson , I . F . ( 1942 ) : Immunity in bruce l losis . Bact . Rev . 6 :

1 1 1- 142 .

Hudd leson , I . F . ( 1943 ) : Bruce l los i s in man and animal s . The

Commonwealth Fund , N . Y .

Hudd leson , I . F . ( 1965 ) : Electrophoretic and serologic s tudies o f

ant iserum gamma- globu l in after rivanol fractionat ion and

tryptic diges tion . Am . J . ve t . Res . 26 : 108 - 1 1 1 .

Hunter , C . A . ; Colbert , B . ( 1956 ) : F loccu lat ion test for

Bruce l losis . J . Immun . 7 7 : 232-241 .

Husband , A . J . ; Brandon , M . R . ; Lasce l les , A . K . ( 19 7 2 ) : Absorpt ion

and endogenous product ion of immunoglobu l ins in calves .

Aus t . J . exp. B io l . med . Sci . 5 0 : 49 1 -498 .

Hutchins , D . R . ; Lepherd , E . E . ( 1968 ) : The occurrence o f agglut inins

to Bruce l la abor tus in horses . Aus t . vet . J . 44 : 323- 325 .

I s ayama , Y . ( 196 1 ) : Sero logica l s tud ies on bruce l los i s . Jap. J ,

anim . infec . D i s . 34 : 1 1 86 - 1 203

Jones , L .M . ( 197 6 ) : A po tency as say for Bruce l l a a l lergens in

gu inea p igs . International Sympos ium on Bruce l lo s i s II .

Deve lop. biol . S tandard . 3 1 : 165 - 1 70 .

Jones , L . M . ( 1977 ) : Bruce l la antigens and sero logical test re su l ts .

In Bovine Bruce l los is , An Internat ional Symposium . Ed i tors ,

R . P . Crawford and R . J . Hidalgo . Texas A & M University Pre s s .

pp . 40-48 .

Jones , L .M . ; Berman , D . T . ( 1975 ) : Ant ibody-mediated and de layed

type hypersens i t ivity react ions to Bruce l la skin test antigens

in guinea p igs . Infec . and Immunity 11: 360- 364 .

Jones , L .M . ; Berman , D . T . ( 1 97 6 ) : The ro le o f l iving vaccines in

prophylaxis . Internat ional Symposium on Bruce l lo s i s I I .

Deve lop. biol . standard , ]1: 328- 334 .

Jones , _ L .M . ; Bij fenga , G . ; Kaplan , M .M . ( 1957 ) : Aberrant

agglutinat ion phenomena in Bruce l la sera . FAO/WHO Expert

Committee on Bruce l los i s . WHO/Bruc/ 149 (September 4 , 1957 ) .

1

Jones , L .M . ; Hendricks , J . B . ; Berman , D . T . ( 1963 ) : The

s tandardisat ion and use of the comp lement-fixat ion test for

the diagnosis of bovine bruce l los is , wi th a review of the

l i terature . Am . J . ve t . Res . 24 : 1 143- 1 15 1 .

Jones , L .M . ; Wilson , M .M . ( 1 95 1 ) : Serum agglutinins in Bruce l losis .

Nature , 1 67 : 558-55 9 .

Joubert , L . ; Roumiantzeff , M . ; Vale tte , L . ( 1967 ) : L ' au tomat ion

en serologie spec i fique appl icat ion au dep is tage de la

bruce l lose . Revue med . vet . 1 1 8 : 307- 320 .

Kabat , E . A . ; Mayer , M .M . ( 19 6 1 ) : Experimental Immunochemistry.

2nd Ed . Char les C . Thomas , Springfield , I l l inois , USA .

Kaneene , J .M . ; Johnson , D . W . ; Anderson , R .K . ; Angus , R . D . Pietz ,

D . E . ; Muscoplat , C . C . ( 197 8 ) : Kine tics of in vi tro bovine

lymphocyte immunos t imulat ion with a Brucel la abortus antigen .

Am . J , vet . Re s . 39 : 2 35- 239 .

Ke l lar , J . ; Marra , R . ; Mar t in , W . ( 19 7 6 ) : . Bruce l losis in Ontar io .

A case control s tudy . Can . J . comp. Med . 40 : 1 19 - 1 2 8 .

Keppie , J . ; Wi l l iams , A . E . ; Wi t t , K . ; Smi th , H . ( 19 65 ) : The ro le

of erythri tol in the t i ssue loca lisation of the bruce l lae .

Br . J . exp . Path . 46 : 104- 108 .

Keppie , J . ; W i t t , K . ; Smi th , H . ( 1967 ) : The effect of erythri tol

on the growth of S 19 and other at tenuated s trains o f Bruce l la

abortus . Res . vet . Sc i . 8 : 294-29 6 .

Kerr , W . R . ( 1955 ) : Vagina l and uterine ant ibodies in cat tle wi th

particu lar reference to Br . abortus . Br . ve t . J . 3 : 1 6 9 - 1 7 8 .

Kerr , W . R . ; Rankin , J . E . F . ( 195 9 ) : The spread o f bruce l losis within

herds - The milk prob lem. Vet . Rec . 7 1 : 1 78- 179 .

Kiggens , E . M . ; Plastridge , W . N . ; Wi l l iams , L . F . ; Eas terbrook , H . L .

( 1955 ) : Cross agglutination be tween Vibrio fetus and Bruce l l a

abortus . Am. J . ve t . Res . Ji: 29 1 - 294 ,

King , R . O . C . ( 1932 ) : Con tagious abort ion in bu l l s . A review o f

recent agglutinat ion tests . Aus t . ve t , J , 8 : 226- 230 .

King , R .o . c . ( 1940 ) : Bruce l la infection in the bul l . A progress

report o f mating experiment s with natura l ly infected bu l l s .

Aus t . vet , J . Ji: 1 1 7 - 1 19 .

1 29 .

King , N . B . ( 196 1 ) : Effect o f three shipp ing fever bacterins on

Bruce l la t i tre s in ca t t l e . J . Am . ve t . med . As s . 1 39 :

6 78 -680 .

King , H . C . ( 197 1 ) : Progres s o f the S tate-Federal bruce l losis

eradicat ion program. Proc . U . S . Anim . Hlth . Assn . 75th

Ann . Mtg. pp 1 1 3- 122 .

King , N . B . ; Frank , N . A . ( 196 1 ) : Effect o f age on re s i s tance and

re tention of t i ter in ca t t le vaccinated with S train 19

Bruce l la abortus vaccine . J . Am . ve t . med . Ass . 1 39 : 1 00- 1 0 3 .

Kwantes , W . K . ( 197 6 ) : Au tomat ion in sero logy . 2nd Internat ional

Symposium on Rapid Methods and Automation in Microbiology.

Ed i tors H . H . Johns ton and S .W . B . Newsom . Learned Information

( Europe) Ltd . Oxford . pp . 1 9 7 - 20 1 .

Lamber t , G . ; Amerau l t , T . E . ( 19 6 2 ) : Comparat ive s tudy o f three

sero logic tes ts for detecting the response in ca t t le to

viru lent Bruce l la abor tus . Am . J . ve t . Re s . 2 3 : 5 29-533 .

Lamber t , G . ; Amerau l t , T . E . ; Manthei , C . A . ; Goode , E . R . ( 1960 ) :

Further s tudies on the per s i s tence o f Bruce l la abor tus

infection in cat t l e . Proc . U . S . l ive s tk. sani t . Assn .

64th Ann . Mtg. pp 109 - 1 1 7 .

Lamber t , G . ; Ameraul t , T . S . ; Manthei , C . A . ; Geode , E . R . ( 1 96 1 ) :

Immunogenic response o f ca lves vaccinated at d i f ferent ages

with Bruce l la abortus s train 1 9 . Pro c . U . S . l ive stk .

sani t . Ass . 65 th Ann . Mtg. 93-99 ,

Lamber t , G . ; Deyoe , B . L . ; Painter , G .M . ( 1964) : Pos t-vaccinal

persistence o f Bruce l la abortus s train 19 in bul l s .

J . Am . ve t . med . Ass . 145 : 909 -9 1 1 .

Lamber t , G . ; Deyoe , B . L . ; Painter , G . M . ( 19 65 ) : Further studies

on the persis tence o f Bruce l la abortus strain 1 9 in bu l l s .

Pro c . U . S . l ive stk . sani t , As s . 69th Ann . Mtg. pp 124- 1 29 .

Lamber t , G . ; Manthei , C ,A . ; Beyoe , B . L . ( 196 3 ) : S tudies on

B ruce l la abortus infect ion in bul l s , Am . J . ve t . Re s . 24:

1 15 2- 1 157 .

Lapraik , R . D . ; Brown , D . D . ; Mann , H . ; Brand , T. ( 19 75 ) :

Bruce llos i s : A study o f f ive ca lves from reactor dams .

Vet . Rec . 22: 5 2-54 .

•

Lapraik , R . D . ; MacKinnon , D . J . ; S l avin , G . ( 1967 ) : Bovine

abortion : Po ss ible confus ion o f Br . abortus and "Q"

fever infect ions . Ve t . Rec . �: 414-415 .

1 3• • ·

Lapraik , R . D . ; Morgan , W . J . B . ; Lawson , J . R . ( 1964) : The possible

transmis s ion of bruce l los i s by mi lking machine .

vet . J . 19 : 87 - 88 .

S tate

Leech , F . B . ; Vessey , M . P . ; Macrae , W . D . ; Lawson , J . R . ; MacKinnon ,

D . J Q ; Morgan , W . J . B . ( 1 964) : Bruce l los is in the Br i ti sh

dairy herd .

HMSO London .

MAFF Animal Di sease Surveys . Report No . 4 .

Levieux , D •. ( 1 974) : Bovine immunoglobul ins and bruce l lo s i s . II .

Activity of IgG1 , I gG2 , and IgM of serum in agglutinat ion ,

Coombs , comp lement fixat ion react ions and in the Rose Bengal

te st . Ann . Rech . Ve t . 5 : 343 - 35 4 .

Lowry , O . H . ; Rosebrough , N . J . ; Farr , A . L . ; Randal l , R . J . ( 1 95 1 ) :

Protein measurement wi th the Fol in-Phenol reagent . J . B iol .

chem . 1 9 3 : 265 - 275 .

Lubbenhusen , R . E . ; F i tch , C . P . ( 19 2 6 ) : A report of exper imental

work on the bull as a factor in the spread o f infectious

abortion . J. Am . ve t . med . As s . 68: 467-48 1 .

Luchs inger , D . W . ; Anderson , R . K . ( 1 967 ) : Ep izootio logy o f

bruce l losis in a f lock of sheep . J . Am . ve t . med . Ass . 150 :

101 7 - 102 1 .

Luchsinger , D . W . ; Angus , R . D . ; Crue , C . S . ; Anderson , R . K . ( 19 7 3 ) :

The u t i l ization of Bruce l la abor tus cu lturing and biotyping

resu lts in the ep izootiologica l invest igat ion of bovine

bruce l losis . Proc . U . S . Anim . H l th . Ass . 7 7 th Ann . Mtg.

pp 85 -99 .

Mach , J . P . ; Pahud , J . J . ( 19 7 1 ) : S ecretory IgA , a maj or immuno­

globu l in in mos t bovine external secre tions . J . Immun . 106 :

5 5 2- 5 6 3 .

Mackinnon , D . J . ( 1 9 63 ) : The comp lement fixa t ion te s t in bruce l los i s .

B u ll . O f f . i n t . F.p i 7. , C10 : 1 H 3 - 400 .

Ma l lrnnn , W . L . ( l ' t iO ) a ' l ' l 1 1 • ' ' ' ' " " '" '· ' '· ' ' i l l i i l l l • l l l t y " ' "'' '"" " ' ' ' { " ' ' ' '" l l r. I I C< d . l. n 1 1 1 H I l ' n : t L I I r u l l. n �.jt ! l l < ! l.' / 1 , , f , Aur , v r t l , t l l r • d , A H H . 'j_L1

6 3 6 - 6 38 .

1 3 1 .

Mancini , G . ; Carbonara , A . O . ; Heremans , J . F . ( 1965 ) : Immuno­

chemical quanti tat ion o f antigens by s ingle rad ial immuno­

diffusion . Immunochemis try, 2 : 235 - 254 .

Manthei , C .A . ( 1 95 9 ) : Summary o f contro l led research wi th s train

19 . Proc . U . S . l ive s tk . sanit . As s . 63rd Ann . Mtg. 9 1 - 9 7 .

Manthei , C . A. ; Carter , R . W . ( 1950 ) : Persis tence o f Bruce l l a abortus

infect ion in cat t l e . Am . J . ve t . Re s . 1 1 : 1 7 3 - 180 .

Manthei , C .A . ; DeTray , D . E . ; Goode , E . R . ( 1950) : Bruce l la

infect ion in bu l l s and the spread o f bruce l losis in cat t l e

by ar tificial inseminat ion . I . Intrauterine inj ection .

Proc . Am . ve t . med . As s . 8 7 th Ann . Mtg. pp 17 7 - 184 .

Manthe i , C .A . ; Mingle , C . K . ; Carter , R . W . ( 1 95 1 ) : Durat ion o f

immunity t o bruce l losis induced in ca tt le with s train 19

vacc ine . Proc . Am . ve t . med . Ass . 88th Ann . Mtg. pp 1 28 - 141 .

Markenson , J . ; S u l i t zeanu , D . ; O l i t zki , A . L . ( 196 2 ) : Immunogenic

activity o f Bruce l l a ce l l wa l l . B r . J . exp. Path . 43 : 67 - 7 6 .

Mar t in , S .W . ( 19 7 7 ) : The evaluat ion o f tests . Can . J . comp. Med .

41 : 19 - 25 .

Mayer , M .M . ( 19 6 1 ) : Comp lement and comp lement fixation . In

Experimenta l Immunochemi s try. Ed s . Kabat , E .A . and Mayer ,

M .M . 2nd Ed . 196 1 . C . Thomas , I l l inois .

McAl lum , M . J . F . ( 19 76 ) : Animal hea l th and deer farming . In

Deer Farming in New Zealand p 15 - 1 9 . Editors Drew , K .R .

and McDonald , M . F . New Zea land Society of Animal Product ion

Occas ional Publ icat ion No . 5 .

We l l ington .

Ed i tor ial Services Ltd . ,

McCaughey , W . J . ; Kerr , W . R . ( 1967 ) : Abor tion due to bruce l losis

in a thoroughbred mare . Vet . Rec . 80 : 186- 187 .

McCaughey , W . J . ; Purce l l , D . A . ( 19 7 3 ) : Bruce l losis in bulls .

Vet . Rec . 9 3 : 336-337 .

McCul lough , N . B . ( 1970 ) : Microb ial and Host factors in the

pathogenes is of Brucel losis . In Infectious Agents and Hos t

Reactions . Ed . S . Mudd . W . B . Saunders . Phi lade lphia & London .

132 .

McCu l lough , N . B . ; E i se le , C . W . ; Pave lchek , E . ( 1 95 1 ) : Survey o f

bruce l losis in s laughtered hogs .

Washington �: 205 - 208 .

Pub l ic Hea l th Report ,

McDiarmid , A . ( 1948 ) : A compar i son o f the immuni ty produced in

guinea p igs by the inocu lat ion of S . 1 9 Br . abortus , vaccine

intradermal ly and subcu taneous ly . Ve t . Rec . 60 : 227- 228 .

McDiarmid , A. ( 1950 ) : A compari son of the immuni ty produced in

catt le by the inocu lat ion o f Br . abortus strain 19 intra­

dermal ly , intracuada l ly and subcutaneously . Ve t . Rec . 6 2 :

3 6 1 - 364 .

McDiarmid , A. ( 1 957 ) : The degree and durat ion of immunity in

cat t le resu l t ing from vacc inat ion wi th S . 19 Br . abortus and

i t s imp l ication in the fu ture contro l and eventual

eradicat ion of bruce l lo s i s . Vet . Rec . 69 : 8 7 7 - 8 79 .

McDiarmid , A . ( 1 9 7 2 ) : Immunisat ion contre la bruce l lose au moyen

d ' une souche ne determinant pas la formation d ' agglutinins .

Annls . Ins t . Pas teur, Par is 102 : 7 92 - 795 .

McEwen , A . D . ; Prie s t ly , F .W . ( 1 9 38 ) : Experiments on contagious

abortion : Immunisat ion s tudies with vaccines o f graded

viru lence . Ve t . Rec . 5 0 : 1097 - 1 106 .

McEwan , A . D . ; Prie s t ly , F .W . ; Paterson , J . D . ( 19 39 ) : An e s t imate

of a su itab le infect ive dose of Br . abortus for immuni sat ion

tests on cat t l e . J . comp. Path . Ther . 5 2 : 1 1 6 - 1 28 .

McEwen , A. D . ; Samuel , J .McA . ( 195 5 ) : Bruce l la abortus heat s tab le

protect ive ant igen revealed by adj uvant and pre sent in a

"rough" var iant , s train 45 / 20 : immuni sation exper iments on

guinea pigs . Ve t . Rec . 67 : 546- 548 .

McNaught , D . J . ; Chappe l , R . J . ; Al lan , G . S . ; Bourke , J . A. ;

Rogerson , B . A . ( 197 7 ) : The e ffects o f IgG2 and o f antigen

concentration on prozoning in fue comp lement fixation te st for

bovine bruce l los is . Re s . ve t . Sci . 2 2 : 194- 197 .

Meyer , M . E . ( 1966 ) : Host-paras i te re lationships in bruce l losis .

Proc . U . S . l ive stk . sani t . Ass . 70th Ann . Mtg. pp 1 29- 134 .

Meyer , M . E . ( 1967 ) : Metabo l ic characterisation o f the genus

Bruce l la . VI . Growth s t imulat ion by erythritol compared

with s train viru lence for guinea p igs . J . Bact . 21: 996- 1000 .

1 3 \ "

Meyer , M . E . ; Nelson , C . J . ( 19 69 ) : Pers i s tence o f Bruce l la

abortus , strain 1 9 infect ion in immunised catt le . Proc . U , S .

l ive s tk . sani t . As s . 7 3rd Ann . Mtg. pp 15 9 - 165 .

Meyer , K . F . ; Shaw, E . B . ( 1 9 20 ) : A comparison of the morphologic ,

cu l tura l and biochemica l character is tics o f B . abortus and

B . me l i tens i s . J . infect . Dis . 27 : 1 7 3 - 1 84 .

Mi l ler , J . K . ( 19 7 la) : Automated sys tem for Bruce l la ant ibody

screening . Lab . Pract . 20 : 7 1 3 - 7 15 and 730 .

Mi l ler , J . K . ( 1 97 lb ) : Prel iminary eva luat ion of an automated

Bruce l la antibody screening sys tem. Re s . ve t . Sci . 1 2 :

199- 202 .

M i l ler , J . K . ; Ne tt leton , P . F . ; Robertson , A .M . ( 19 7 3 ) : Evaluat ion

o f a Two- channe l au toma ted sys tem for the sero-d iagnosis o f

bruce l losi s . Ve t . Rec , 9 2 : 492-496 ,

Mohler , J . R . ; Wight , A . E . ; O ' Rear , H .M. ( 194 1 ) : Cal fhood

vaccinat ion as an aid in co-operative Bang ' s di sease .

J . Am . ve t . med . As s . 98 : 1 - 9 .

Morgan , W . J . B . ( 1960) : Compari son of var ious media for the

growth of Bruce l la . Re s . ve t . Sci . 1 : 47-5 2 .

Morgan , W . J . B . ( 196 7 ) : The sero logical diagnosis o f bovine

bruce l l osis . Ve t . Rec . 80 : 6 1 2 - 6 2 1 .

Morgan , W . J . B . ( 1969 ) : Bruce l los i s in anima l s . Diagnosis and

Contro l . Proc . R . Soc . Med . 6 2 : 1050- 105 2 .

Morgan , W . J . B . ; Corbe l , M . J . ( 197 5 ) : Recommendat ions for the

description of species and biotypes o f the genus Bruce l la .

Internat ional Symposium on Bruce l lo sis . II . Dev . biol ,

S tandard , ll= 27 - 37 .

Morgan , W . J . B . ; Davidson , I . ; Herbert , C . N . ( 19 7 3 ) : The Use of

the second Internat ional S tandard for anti�Bruce l l a abortus

serum in the comp lement f ixat ion te s t . J, bio l . S tandard ,

1 : 43- 6 1 .

Morgan , W . J . B . ; McDiarmid , A . ( 19 60 ) : The excret ion o f Bruce l la

abortus in the milk o f exper imenta l ly infected catt le .

Res . ve t . Sci . 1 : 5 3-56 .

1 34 .

Morgan , W . J . B . ; MacKinnon , D . J . ; Gi l l , K . P .W . ; Gower , S . G . M . ;

Norris , P . I . W . ( 197 1 ) : .S tandard Laboratory technique s for

the diagnos i s o f bruce l lo s i s . Minis try of Agriculture ,

F i sheries and Food , Central Ve ter inary Laboratory , Weybr idge .

Morgan , W . J . B . ; MacKinnon , D . J . ; : Lawson , J . R . ; Cul len , G .A . ( 1969 ) :

The Rose B engal p late agglutinat ion test in the d iagno s i s o f

bruce l los i s . Vet . Rec . � : 636-641 .

Morgan , W. J . B . ; Richard s , R . A . ( 1974 ) : The diagnos i s , control

and erad i cat ion o f bovine bruce l lo s i s in Great Britain .

Vet . Rec . 9 4 : 5 10-5 1 7 .

Mor se , E . V . ; Re s t ic , M . ; Roberts tad , G . W . ; Schnieder , D . W . ( 195 3 ) :

Cross agglut ination react ions among Bruce l l a , V ibrio , and

o ther micro organisms . Am . J . ve t . Res . 14 : 324- 327 .

Morse , E . V . ; R i s t ic , M . ; Wi t t , L . E . ; Wip! , L . ( 195 3 ) : Canine

abort ion apparently due to Bruce l la abortus . J . Am . ve t .

med . As s . 1 2 2 : 1 8 - 20 .

Myl re a , P . J . ( 19 7 2 ) : The d i agno s i s o f bruce l losis in dairy herds .

Aus t . ve t . J . 48 : 369- 375 .

Myl re a , P . J . ; Fraser , G . C . ( 1 976) : The use of supp lementary te sts

in the sero logical d iagnosis o f bovine bruce l lo s is . Aus t .

ve t . J . 5 2 : 26 1- 266 .

Nad ler , H . E . ( 19 78 ) : Bovine di sease control prob lems in the

Northeas t . lA Bruce l lo s i s : An overview. Corne l l Ve t . 68 :

164- 1 7 2 .

Nagy , L . K . ; H i gnett , P . G . ( 1967 ) : The long term e f fects o f

Bruce l l a infection o f newly born ca lves . Res . vet . Sci . �:

247 - 255 .

Nash , D . R . ; Heremans , J . F . ( 1 969 ) : A quantitat ive ant ibody

binding me thod for the determination of specific antibody

within d i fferent immunoglobu l in classe s . Immunology 12:

685 - 6 94 .

Ne l son , C . J . ; Anderson , R .K . ; Kimber ling , C . V . ; Pie t z , D . E . ( 1 9 66 ) :

Ep i zootiologic fac tors o f bovine bruce l losis : Comparat ive

bacterio logic s tudies o f infected herd s . Am . J . ve t . Re s . 27 :

1 5 1 5 - 1 520 .

Nico lett i , P . ( 196 7 ) : U t i l isat ion of the card te st in bruce l lo s i s

erad ication . J . Am . ve t . med . As s . 15 1 : 1 778- 1 7 8 3 .

Nico le t t i , P . ( 1 969 ) : Fur ther eva luat ions o f sero logic test

procedures used to diagnose bruce l los i s . Am . J . ve t . Re s . 30 :

1 8 1 1 - 18 1 6 .

Nico lett i , P . ( 197 7 ) : Adu l t vaccination . In Bovine Bruce l lo s i s

An Internat ional Sympos ium . Ed itor s R . P . Crawford and R . J .

Hida lgo . Texas A&M Univers i ty Pre s s . pp 20 1 - 208 .

Nico le t t i , P . ; Holmes , J . ( 1 968 ) : At tempts to produce cross ­

reactions for Bruce l la in ca t t le treated wi th haemorrhagic

sept icaemia bac terins . Corne l l ve t . 58 : 421-424 .

Nicole t t i , P . ; Muraschi , T . F . ( 19 66 ) : Bac terio logic eva luat ion o f

s ero logic test procedure s for the diagnosis o f bruce l lo s i s in

prob lem cat t l e herds . Am . J . ve t . Re s . 27 : 689- 694 .

O l i tski , A . ( 1970 ) : Immuno logica l Me thods in Bruce l lo s i s Re search .

Par t I . In vi tro Procedure s . Part II . In � Procedures .

S . Karger , Base l .

O ' Re i l ly , D . J . ; Cunningham , B . ( 197 1 ) : An assessment o f the

bruce l losis card test . Ve t . Rec . 88 : 590-594 .

Ouchte r lony , 0 . ( 1953 ) : Antigen-antibody react ions in ge l s .

Act a path . microbia l . scand . 3 2 : 2 3 1 - 240 .

Painter , G .M . ; Deyoe , B . L . ; Lambert , G . ( 1966 ) : Compari son of

var ious media for the isolation o f B ruce l la . Can . J . comp.

Med . Vet . Sci . 30 : 2 18- 223 .

Palmer , W . J . ; LeQuesne , S . E . ( 1 9 7 6 ) : Rapid Methods and

Automation in Microbiology and Immunology. Information

Retrieva l Ltd . London .

1 36 .

Parish , C .R . ( 19 7 2 ) : The relationship between humoral and ce l l -

mediated immunity . Transplantat ion Reviews , 11= 35 - 66 .

Patterson , J .M . ; Deyoe , B . L . ( 197 7 ) : E ffect of physical propert ie s

o f mi lk fat globu les on Bruce l la r ing test sens i t ivity .

J . Dairy Sc i . 2Q: 85 1 - 85 6 .

Patterson , J .M . ; Deyoe , B . L . ; S tone , S . S . ( 19 76 ) : Identi fi cation

of Immunoglobu lins associated wi th complement fixat ion ,

agglu t inat ion , and low pH buffered ant igen tests for

bruce l los i s . Am . J . vet . Res . 37 : 3 19- 324 .

Payne , J .M . ( 1959 ) : The pathogenesis o f experimental bruce l lo s i s

in pregnant cow . J . Path . Bact . 7 8 : 447 -46 3 .

Payne , J .M . ( 19 60 ) : The pa thogenes i s o f experimental bruce l lo s i s

in virgin hei fers with and wi thout cont inuous proge s terone

treatment . J . Endocr . 20 : 345 - 354 .

P earce , J . H . ; W i l l iams , A . E . ; Harris-Smith , P .W . ; F i t z-George , R . B . ;

Smi th , H . ( 1 962 ) : The Chemica l bas is o f the virulence o f

Bruce l la abortus . II . Erythr i to l a const ituent o f bovine

foetal fluids which s t imulates the growth of Br . abortus in

bovine phagocytes . Br . J . exp. Path . 43 : 3 1 - 37 .

P fei ffer , N . E . ; McGuire , T . C . ; Bende l , R . B . ; Weike l , J .M . ( 19 7 7 ) :

Quanti ta tion of bovine immunoglobul ins : comparison o f s ingle

radial immunodi ffusion , zinc su lphate turbidi ty , serum

electrophores is , and re fractome ter methods . Am . J . ve t . Re s . 38 :

693- 698 .

Phi l ippon , A . ; Renoux , G . ; Plommet , M . ( 1 970 ) : Experimental

bruce l lo s i s in cat t l e . II . Distr ibut ion of Bruce l la abortus

six weeks a fter partur i tion and 3-5� months after experimental

infection . Ann . Rech . Ve t . 1 : 203- 2 1 3 .

Phi l ippon , A . ; Roumy , B . ; Renoux , G .

bi tch due t o Bruce l la abortus .

9 23-9 28 .

( 1969 ) : Bruce l losis in a Bu l l . Acad . ve t . Fr . 4 2 :

P iet z , D . ( 19 7 7 ) : Bruce l la ant igens and sero logic t e s t re su l t s .

In Bovine B ruce l losis, An Internat ional Symposium. Editors

R . P . Crawford and R . J . Hidalgo . Texas A & M Univers i ty

Press . pp 49- 60 .

13 7 .

P lacke t t , P . ; Al ton , G . G . ( 19 75 ) : A mechani sm for prozone

format ion in the comp lement fixation te�t for bruce l los i s .

Aust . ve t . J . 21= 374- 37 7 .

P lackett , P . ; Co t tew , G . S . ; Be s t , S . J . ( 1 9 7 6 ) : An ind irect

haemolysis test ( IHLT ) for bovine bruce l los i s . Aus t . ve t . J .

5 2 : 136- 140 .

Plommet , M . ; Fensterbank , R . ( 19 76 ) : Vacc inat ion agains t bovine

bruce l los is wi th a low dose of strain � admini stered by the

conj unct iva l route . Ill . Sero logical response and immuni ty

in the pregnant cow . Ann . Rech . Ve t . 7 : 9- 23 .

P lomme t , M . ; Fensterbank , R . ; Renoux , G . ; Gestin , J . ; Phi l ippon , A .

( 1 973 ) : Experimental bovine bruce l los i s . XII . Pers i s tence

to adu lt age o f congeni tal infect ion in the he ifer .

Ann . Rech . Vet . 4 : 4 1 9 -435 .

Prior , M . G . ; N i i lo , L . ; Reeker , W. H . ( 19 7 5 ) : Use o f the

bruce l los is card test for screening cattle in Saskatchewan .

Can . J . camp. Med . 39 : 107- 109 .

Rank in , J . E . F . ( 1 9 65 ) : Bruce l la abortus in bu l l s . A study o f

twe lve natura l ly infected cases . Vet . Rec . 7 7 : 1 3 2 - 1 35 .

Rasooly , G . ; O l i t zki , A . L . ; S u l i tzeanu , D . ( 1967 ) : Immunisat ion

against Bruce l la with ki l led vaccines . II . Immunizing

activity o f ki l led Bruce l la in gu inea pigs . I srae l J , med ,

S ci . 3 : 8 14- 8 2 1 .

Rasoo ly , G . ; Boros , D . L . ; Ger ichter , C . H . ( 1968 ) : Immunizat ion

against Bruce l l a wi th ki l led vacc ine s . Ill . Exper iments

in mice . I srael J . med . Sci . 4 : 246- 25 1 .

Ray , W . C . ( 19 7 6 ) : An assessment o f inves t igat ions conducted in

the USA on Bruce l la abortus strain 45 / 20 bacter ins .

Internat ional Symposium on Bruce l los i s . II , Develop . b io l .

standard . 3 1 : 335 - 342 .

Ray , W.C . ; Hendr icks , J . B . ( 1 974) : 45 / 20 vaccination in a Bruce l l a

infected herd . Proc . U . S . Anim . H l th . As s . 78th Ann . Mtg .

p p .. ·88- 103 .

Redman , D . R . ; Deyoe , B . C . ; King , N . B . ( 19 67 ) : Re s i s tance o f

cattle t o Bruce l la abortus fo l lowing vaccination a t two and

three months o f age . J . Am . vet . med . Ass . 11QL 403·407 .

138

Re i ter , D . O . ( 1936 ) : S tudies on the extract ion of a precip i tab le

subs tance from the genus B ruce l la . J. infect . D i s . 5 8 :

45 -58 .

Renoux , G . ; Val e tte , L . R . ( 1 967a ) : Evo lut ion des ant icorps chez

les bovins apres immunisat ion contre la bruce l lose par le

vaccin tue 5 3H38 . Bu l l . Acad . Vet . Fr . 40 : 2 7 - 36 .

Renoux , G . ; Valette , L . R . ( 1967b ) : Evo lut ion des anticorps che z

les bovins apre s immunisation contre la bruce l lose par le

vaccin tue 5 3H38 . Rev . Immuno l . 3 1 : 99 - 107 .

Renoux, G . ; Va lette , L . R . ( 1967c ) : Vaccinat ion mixte de genisses

contre la bruce l lo se et la fievre aphteuse . Resultats de

l ' immunisation ant ibruce l l ique . Rev . Immuno l . 3 1 : 341 - 35 0 .

Renoux , G . ( 195 7 ) : E tudes sur l a Bruce l lose ovine e t capr ine .

XV . Du diagnost ic sero logique de la bruce l lose ind ividue l le

de chevres arti f ic ie l lement infectees par B r . me l i tens i s .

Archs . Inst . Pasteur, Tuni s . 34: 207 - 2 25 .

Renoux , G . ; Al ton , G . G . ( 1 95 7 ) : Etudes sur la bruce l lose ovine

et caprine . XII I . Reactions d ' agg lut inat ion et de fixation

du comp lement che z de chevres suedoises apres la vacc ination

et apres l ' inoculation infe ctante d ' epreuve . Archs . Inst .

Pas teur, Tuni s . 34: 29- 35 .

Renoux, G . ; Plommet , M . ; Phi l ippon , A. ( 1 9 7 1 ) : Exper imental

bovine bruce l los i s . VIII . Development of agglutinat ing and

comp lement- fixing ant ibod ies after inoculat ion of Bruce l l a

abortus a t 5 t o 6 . 5 months o f pregnancy . Ann . Rech . Vet . 2 :

15 9 - 17 1 .

Renoux , - M . ; Palat , A . ; Gui l laumin , J . M . ; Renoux , G . ( 1 9 7 6 ) :

Hemagglutination pass ive , transformation lymphob last ique et

migration des l eucocytes app l iques au d iagnostic des bruce l lo se s .

International Symposium on B ruce l lo s i s . II . Develop. b i o l .

Standard ]l: 145 - 15 6 .

Rib i , E . ; Mi lner , K . C . ; Perr ine , T . D . ( 1959 ) : Endotoxic and

ant igenic fractions from the ce l l wal l o f sa lmone l la

enter i t idis - Methods for separat ion and some bio logic

act ivi t ies . J . immun . 8 2 : 75 - 84 .

139 .

Rice , C . E . ; Alexander , D . C . ; Barrett , B . B . ( 1967 ) : Chromatographic

stud ies of sera from ca lves vacc inated wi th Bruce l la abortus ,

s train 1 9 . Can . J . comp. Med . ]l: 1 14- 1 2 1 .

Rice , C . E . ; Boulanger , P . ; Mackie , C . ; Moore , T . ( 1 9 5 2 ) : The

conglutinat ing- comp lement fixation te st as a supp lementary

method of detect ing act ivi ty with Bruce l la abor tus antigen .

Can . J . comp. Med . Ve t . Sci . 1 6 : 348- 35 6 .

Rice , C . E . ; Boye s , B .W . ( 19 7 1 ) : Serum Immunoglobul ins in bovine

bruce l los i s . N . Z . ve t . J . 19 : 1 46- 154 .

Rice , C . E . ; Tai lyour , J . ; Cochrane , D . ( 1 966 ) : U l tracentri fuged

studies of sera from cattle vacc inated or natural ly infected

wi th Bruce l la abortus . Can. J . comp. Med . 30 : 270- 278 .

Richardson , G .M . ( 194 1 ) : The preservation of l iquid comp lement

serum . Lancet 2 : 696- 697 .

Robertson , F . J . ; Mi lne , J . ; S i lver , C . L . ; C lark , H . ( 19 7 3 ) :

Abort ion associated wi th Bruce l l a abortus (bio type I ) in a

T . B . mare . Ve t . Rec . 9 2 : 480-48 1 .

Robertson , L . ; Farre l l , I . D . ( 1 968 ) : The use o f the Bruce l l a

whey comp lement fixation te st in a herd outbreak . Vet . Rec .

8 3 : 643- 646 .

Rober tson , L . ; Farre l l , I . D . ; Hinchli ffe , P .M . ( 19 7 7 ) : The

isolation of bruce l lae from contaminated source s . A review.

Br . ve t . J . 1 3 3 : 1 9 3 - 200 .

Roepke , M . H . ; S t i les , F . C . ( 1970) : Potent ial efficiency o f mi lk

ring test for detect ion of bruce l losis . Am . J . ve t . Re s . 1 2 :

145- 249 .

Rose , J . E . ; Amerau l t , T . E . ( 1964) : E lec trophore t i c and ul tra­

centri fugation stud ies on serum from pregnant heifers after

exposure to viru lent B ruce l l a abortus . · · Am . J . ve t . Re s . 12,:

998- 100 1 .

Rose , J . E . ; Lambert , G . ; Roepke , M . H . ( 1 964) : U l tracentr ifugat ion

and heat- inact ivat ion s tud ies on sero-agglut inins o f pregnant

heifers artificia l ly infected wi th viru lent B ruce l l a abortus .

Am . �J . vet . Res . 25 : 3 29-332 .

�

140 .

Rose , J . E . ; Roepke , M . H . ( 195 7 ) : An acid i fied antigen for detect ion

o f non- spec i fic reactions in the p late- agglutinat ion te st for

bovine bruce l los i s . Am . J . ve t . Re s . 18 : 550-55 5 .

Rose , J . E . ; Roepke , M . H . ( 1964 ) : Physiochemica l studies on pos t­

vaccinal Bruce l la agglutin ins in bovine serum . Am . J . ve t .

Res . 25 : 325- 328 .

Safford , J . W . ( 1 95 9 ) : Montana beef-cattle bruce l los i s vaccina tion

survey 195 2- 5 8 . Proc . U . S . l ive s tk . sani t . As s . 6 3rd Ann .

� pp 98- 103 .

Sahiar , K . ; Schwar tz , R . S . ( 19 65 ) : The immunoglobulin sequence .

I . Arre st by 6-mercap topur ine and re s t i tution by antibody ,

ant igen or sp leenectomy . J . Immun . 95 : 345 - 354 .

Saunder s , G . C . ; C l inard , E . H . ( 1 97 6 ) : Rap id microme thod o f

s creening for antibod ies t o di sease agents us ing the indirect

enzyme- labe l l ed antibody te s t . J . c l in . microb io l . 3 :

604- 608 .

Schi l f , E .A . ( 1968 ) : Bruce l lo s i s eradi cat ion program , i ts pre sent

and future s tatus . J . Dairy Sci . il= 1121- 1 1 25 .

Schi l f , E .A . ( 1 97 2 ) : Comp leting the erad i cation of bruce l losis by

197 5 . J . Am . vet . med . Ass . 1 6 1 : 1 5 25- 1527 .

Schwabe , C . W . ; Ri emann , H . P . ; Frant i , C . E . ( 19'7 7 ) : Epidemio logy

in Veter inary Prac tice . Lea & Febiger , Phi lade lphia .

Seddon , H . R . ( 1 9 19 ) : S tud ies in abort ion disease . J . Comp. Path .

The r . 3 2 : 1 - 34 .

Seit , B . ( 1944 ) : Transmiss ion o f contagious abort ion through

art i ficial inseminat ion . Maaneds skr . Dyrlaeg. 5 6 : 1 2- 37 .

Abs tract No . 1890 . Vet . B ul l . 14 : 296 ( 1944 ) .

Shaw, W . B . ( 1976a ) : Bruce l la abortus infect ion in sheep .

I . Field case . Br . ve t . J . 1 3 2 : 18- 27 .

Shaw , W. B . ( 1976b ) : Bruce l la abortus infection in sheep .

II . Experimental infection of ewes . B r . ve t . J . 1 3 2 : 143- 15 1 . I

Shortridge , E . H . ( 1967 ) :

abortion in mares .

Two cases o f suspected B ruce l la abortus

N . Z . ve t . J . 15 : 33-34 .

._ I

Smith , T . ( 19 19 ) : A character i s t i c loca l i sation o f Bac i l lus aborLus

in the bovine fetal membrane s . J . exp . med . 29 : 45 1 .

Smi th , T . ( 19 25 ) : Pneumonia as soc iated with Baci l l us abor tus (Bang )

in fetuses and new born ca lve s . J . exp . Med . �: 639- 647 .

Smith , H . ; Wi l l iams , A . E . ; Pearce , J . H . ; Kepp ie , J . ; Harr i s - Smi th ,

P . W . ; F i t z-George , R . B . ; Wi t t , K. ( 1 96 2 ) : Foe ta l erythr i to l :

a cause of the loca l isat ion o f Bruce l la abor tus in bovine

contagious abort ion . Na ture , London , 1 9 3 : 47 -49 .

Sobota , J . T . ; Gil l em , H . G . ( 1965 ) : Automated performance o f the

syphi l it ic comp lement- fixation test . Am . J . C l in . Path . 44 :

5 76 .

Spink , W . W . ( 1 956 ) : Th� Nature o f Bruce l losis . The Univers i ty o f

Minnesota Pres s , Minneapo l i s .

S tab leforth , A . W . ( 1 9 36 ) : A Br . abortus suspens ion o f uni form

agglut inabi l i ty standard ised by means o f a dry stab l e

standard ant i -abortus serum . J . comp. Path . Ther . 49 : 25 1 - 2 6 2 .

S tab leforth, A .W . ( 1 95 9 ) : Bruce l lo s i s . In Infectious Di seases o f Animal s . Diseases due to Bacteria . A . W . Stableforth and I . A .

Ga l loway (Ed i tor s ) . Volume 1 : 53- 159 . Butterworths , London .

Stamp , J . T . ; McEwen , A . D . : Wat t , J . A .A . ; N i sbe t , D . I . ( 1950 ) :

Enzoo t i c abortion in ewes . I . Transmiss ion o f the d i sease .

Ve t . Rec . 6 2 : 25 1 - 25 4 .

S tevens , J . F . ( 197 3 ) : Current trend s i n automation : a review.

Med . Lab . Techno l . 30 : 139- 144 .

S t inebring , W . R . ; Kes se l , R . ( 1959 ) : Cont inuous growth o f

Bruce l la abortus in mononuc lear phagocytes o f rats and

guinea p i gs . Proc . Soc . exp. B iol . Med . 101 : 4 1 2 -4 1 5 .

S toenne r , H . G . ; Lackman , D . B . ( 1 957 ) : A new species o f B ruce l l a

iso lated from the desert wood ra t , Neotoma lepida , Thomas .

Am . J . ve t . Res . 1 8 : 947 - 95 1 .

S tuart , J . E . ; B i l l s , C . B . ; DeMat t i c , J . D . ; Mace , D . L . ( 195 9 ) :

The resu l t s o f e leven years vacc inat ing wi th s train 19 .

Proc . U . S . l ive stk . sani t . Ass . 63rd Ann . Mtg . pp 83-90 .

Swiderska , H . ; Osuch , T . ; Br zoska , W . J . ( 197 1 ) : Periphera l b lood

lymphocy'te blast trans format ion test as app l ied for the

diagno s i s o f bruce l losis , Exp. Med . Microbiol . 12: 133- 138 .

t4l .

Tachen , P . H . W . ( 19 64) : Tes t vacc ination o f young cat t le to detect

infect ion wi th Br . abortu s . Tijd schr . Diergeneesk. 89 :

1 703- 1 708 .

Tarala , J . ( 1975 ) : Bruce l los i s infect ion in the he i fer . S tate

vet . J . 30 : 7 1 - 7 6 .

Tay lor , A .W . ; McDiarmid , A . ( 1 949 ) : The stab i l i ty o f the avirulent

characters of Bruce l l a abortus , s train 19 and s train 45 / 20 in

lactating and pregnant cows . Ve t . Rec . 6 1 : 3 1 7 - 3 18 .

Taylor , D . J . ; Renton , J . P . ; McGregor , A . B . ( 1 975 ) : Bruce l la

abortus B io type I as a cause o f abort ion in a b i tch . Vet . Rec .

9 6 : 428-429 .

Te Punga , W .A . ( 19 7 1 ) : An automated comp lement fixat ion test for

the ident i f i ca t ion of cat t le infected wi th Bruce l la abortus .

Advance s in Automated Analys i s . Technicon Sympos ium , Sydney

197 1 . pp 1 - 3 .

Thomsen , A . ( 1 9 36 ) : Transmi ss ion o f bovine contagious abort ion

by co i tus . Maaned sskr . Dyr laeg. 48 : 225 - 2 33 .

Abstract Veter inary Bul let in L: 263 ( 1937 ) .

Thomsen , A . ( 19 37 ) : On the occurrence o f copulat ion infect ion

among cat t le wi th infect ious abor t ion . J , comp. Path . Ther .

5 0 : 1 - 9 .

Thomsen , A . ( 1 943 ) : Does the bull spread infectious abort ion in

cattle ? Experimental s tud ies from 1 936 to 1942 .

J . comp. Path . Ther . 5 3 : 199- 2 1 1 .

Thomsen , A . ( 1949 ) : The incubation per iod o f bovine contagious

abortion . Nord . ve t . Med . l= 7 9 7 - 804 .

Thomsen , A . ( 1957 ) : The erad icat ion of bovine bruce l losis in

S candanavia . Adv . ve t . S c i . 3 : 1 9 7 - 240 .

Thorner , R .M . ; Remein , Q . R . ( 1 9 6 1 ) : Pr inc iples and procedures in

the evaluation of screening for d i sease . U . S . Pub l ic Hea l th

Service , Divi sion of Chronic Diseases . Monograph No . 67 .

Government Printing Office , Washington , D . C .

Timb s , D . V . ; Digby , J . G . ; Doe , I . ( 1 978a) : The relat ionship

between the bruce l losis card test and the comp lement f ixation

test u sed in the bruce l losis eradication scheme .

N . Z . vet . J . 1&: 67-70 .

143 .

Timb s , D . V . ; Moxham , J . W. ; Liberona , H . E . ( 1978b ) : The use of

automated comp l ement fixation techniques . in the bruce l losis

eradication scheme . N . Z . vet . J . � : 52 - 5 6 .

Tri lenko , P .A . ( 1957 ) : New modi ficat ion o f the comp lement fixation

test for the d iagnosis of bruce l los i s . Zh . Mikrobio l . ,

Suppl . �: 43- 44 .

Vargues , R . ( 1965 ) : The use o f the Auto-Analyzer for the automatic

t i tration o f antigenic preparations by means o f comp lement

fixation . Ann . N . Y . Acad . Sci . 1 30 ( 2 ) : 8 19 - 8 26 .

Vargues , R . ; Ayera , R . ( 1963 ) : U t i l i sat ion de L ' Auto-Analyseur

Technicon pour L ' e tude de la cinet ique de la react ion de

fixation due comp lement . C . r . S eanc . Soc . B iol . 15 7 :

8 7 2 - 87 6 .

Vargues , R . ; S tudievic , C . ; Audurier , A . ( 19 66 ) : Le dosage des

anticorps fixant le comp lement . Comparai son entre les

techniques manue l les et les technique s automatiques .

Annls . Ins t . Pas teur, Paris . 1 10 ( 3 ) : 2 38- 243 .

Vargues , A . ; S tud ievic , C . ; Ripou l t , C . ( 19 65 ) : L ' Automatisat ion

de la react ion de Bordet-Was sermann . Annls . B io l . C l in . 2 3 :

6 2 3- 640 .

Vargues , R . ; Va lette , L . ; Rive , M . ; S tudievic , C . ( 1968 ) :

De termination o f bovine bruce llos i� us ing haemagglut ination

or comp lement f ixation wi th the Auto-Analyzer . In Inter­

national Symposium on Bruce l losis, Tunis, 1968 . Ed . R . H .

Regamey . Symposia ser ies in Immunobio logica,l standardisat ion

Vol . 1 2 . pp 383- 388 . S . Karger , Base l , 1970 .

Vecchio , T . J . ( 1966 ) : Pred ic tive value o f a single diagnostic te s t

in unse lected populations . New Engl � J . med . 274 : 1 1 7 1� 1 17 3 .

Von Krogh , M . ( 19 1 6 ) : Col loidal chemis try and Immunology .

J . infect . Dis . 1 9 : 45 2-47 7 .

144.

Wa1 1 ace , N .M . ( 1959 ) : Bruce l la abortus associated with abort ion

in a cow . N . Z . vet . J . I: 106 .

Whi t t les tone , W . G . ; de Langan , H . ; Twomey , A. ( 1968 ) : Machine

mi lking and mast i t i s . Inter-unit cross infection , Annual

Report o f Research Divis ion; New Zea la.nd Department o f Agricul ture l � 68 - 6 9 s 5 4 .

Wi lkinson , P . C . ( 1 966) : Immunoglobulin patterns o f ant ibodies

against Bruce l la in man and anima l s . J . Immun . 96 : 45 7-46 3 .

W i l l iams , A . E . ; Kepp ie , J . ; Smi th , H . ( 196 2 ) : The chemical bas i s

o f the virulence o f B ruce l la abortus . Ill . Foe tal

erythrito l , a cause of the local isat ion of Bruce l la abortus

in pregnant cows . Br . J . exp. Path. 43 : 5 30 - 5 37 .

Wi l l iams , M . R . ; Green , J . R . ( 197 6 ) : Changes in bovine immuno­

globu l in leve ls during a re sponse to homologous erythrocyte

membrane antigen . Re s . vet . Sci . 2 1 : 1 68- 17 2 .

W i l son , G . S . ; Miles , A .A . ( 1967 ) : Undulant Fever . In Top ley and

Wilson ' s Principles o f bacterio logy and immunity. Edi tors

G . S . Wil son and A .A . Miles . Vo l . 2 pp 2039- 2078 . Arno ld ,

London .

Wi sniowski , J . ( 1957 ) : Interpretat ion of re su lts o f sero logical

examinations o f cat t l e for bruce l losis . Med . ve t . 1 3 : 6- 9 .

Wood , W ,A . ; Corbe l , M . J . ( 1 9 7 3 ) : Concentrations o f bovine serum

protein c lasses in re lation to reactivity in sero logica l

tests for bruce l lo s i s . J . comp . Path . 8 3 : 143- 150 .

Worthington , R .W . ; Horwe l l , F . D . ( 1974) : An inve s tigat ion o f the

eff icacy of three B ruce l la vaccines in catt le . Jl . S . Afr .

ve t . med . Ass . 45 : 8 7 - 9 . 1 "

Wor thington , R . W . ; Mu lders , M . S , G . ; McFar lane , I . S . ; Becker , D .

( 197 3 ) : A serological invest igation on adu l t cattle

vaccinated with Bruce l l a abortus S train 1 9 , Onders tepoort J .

vet . Res , 40 : 7 - 1 2 .

I·

1 4 •

Wright , A . E . ; Smith , F . ( 1897 ) : On the app licat ion o f the serum

test to the differential d iagnosis o f typhoid and Ma l ta

fever . Lance t 1 : 65 6 .

Zei s s i g , A . ; Mans field , H . L . ( 1930 ) : A compari son o f the

agglutination and comp lement fixation te sts for the detection

o f B rucel la abortus infection . J . Am . vet . med . As s . 29 :

2 1 1 - 230 .