Microbiological Process Report

Analytical Microbiology

I. The Test Organism

JOHN J. GAVIN

Food Research Laboratories, Inc., Long Island City, N. Y.

Received for publication June 28, 1956

Analytical microbiology may be defined as thatbranch of microbiology in which microorganisms areused as reagents for the quantitative determination ofcertain chemical compounds. The procedures used foranalysis are based on the reaction of a particular micro-organism to its environment. If a suitable microorgan-ism will react with a measurable response to a certainchemical entity, an analytical method for the quantita-tive estimation of the substance may be devised.Theoretically, any chemical might be assayed. Inpractice, a wide range of substances, from simplemineral elements to complex organic compounds, areassayed. The methods available for the assay of thesemany different materials range from qualitativethrough semiquantitative to quantitative procedures.Many articles and reviews have been written on

specific methodology for given chemical entities. Thisseries of papers deals with the theoretical aspects ofanalytical microbiology which are common to all typesof methods, regardless of the material being assayed.The first paper will deal with the test organism, as

this is the essential part of any microbiological pro-cedure. The following papers of the series will considerthe procedures. For the purposes of this series, allmethods used for microbiological assay have beenclassified in one of the following categories: (1) diffusionmethods; (2) turbidimetric methods; (3) metabolicresponse methods; (4) gravimetric methods.

THE TEST ORGANISM1. Requirements for an Ideal Test Organism

In order to perform a microbiological assay, it isnecessary to select a microorganism that can be used asa reagent. As with the reagents used in a chemicalassay, there are specifications that this microbial re-agent must meet. The ideal test organism should havefour basic characteristics: (a) it must be sensitive to thesubstance being assayed; (b) it must be easily culti-vated; (c) it must have some metabolic function or

response that is measurable; (d) it must not be suscepti-ble to variation in either sensitivity or phase.

In addition, the following two characteristics are

desirable: (a) it should have specificity, if possible; (b) itshould be, preferably, nonpathogenic. Barton-Wright(1952) rules out pathogens for the microbiologicalassay of vitamins because pathogens are unstablemutants. This would not be sufficient ground to elimi-nate them for the assay of antimicrobial substances.However, it is of major importance in a laboratorywhere routine assays are performed by technicians thatpathogens are not used. The use of a nonpathogenicorganism eliminates precautionary measures thatmight be necessary to protect personnel.

2. Types of Microbial Response

The type of response measured is determined by thesubstance being assayed. It is dependent upon the bio-chemical effect of the substance on the metabolism ofthe organism. All types of microbial response may beplaced in either of the following categories.

(a) Growth response. Most microbiological proceduresthat are used for analysis depend upon the growth re-sponse of the microorganism to the environment. Thisgrowth response may be positive or negative; positive,if the organism responds to the substance being assayedwith increased growth; negative, if the organism failsto grow in the presence of a particular material. Thegrowth or lack of it may be measured by numericalcounts, by optical density, by weight, or by area. Theresponse may be graded in proportion to the concentra-tion of the test material or it may be a definite endpoint, an all-or-none response.

(b) Metabolic response. In addition to the growth re-sponse of an organism to a particular substance, theremay also be a metabolic reaction to this material.Certain organisms produce metabolic products whichare measurable, and others may have a change in somefunction that can be measured. As in growth response,the microorganism's reaction is either negative or posi-tive. Among the measurable metabolic responses areacid production, carbon dioxide production, oxygen up-take (Schopfer, 1935), reduction of nitrates (Hoaglandand Ward, 1942), hemolysis of red blood cells (Dimick,1943), antiluminescent activity (Rake et al., 1943), in-

323

on April 3, 2020 by guest

http://aem.asm

.org/D

ownloaded from

J. J. GAVIN

hibition of spore germination (Brian and Hemming,1945), and the sterilizing or curing of the growinghyphae of fungus mycelium (Brian and Curtis, 1946).

3. Types of Microorganisms Used for AssayThe spectrum of microbial reagents used for assay is

broad. Procedures have been devised using such diverseorganisms as Micrococcus pyogenes var. aureus andEuglena gracilis. Artificially induced mutants of organ-isms, such as Neurospora and Escherichia coli, have beenused for assay purposes. The organism which is to beused in any given assay must be chosen by the operatorin light of his needs: the type and amount of materialto be assayed, what response is desired, time availablefor assay, and accuracy of the results required.

(a) Bacteria. Bacteria are used for the assay of aminoacids, antibiotics, and vitamins, and in the evaluationof antiseptics, disinfectants, and chemotherapeuticagents. The use of bacterial reagents, in general, posesrelatively fewer problems than the use of other organ-isms. There are cases where their use is excluded by thenature of the material being assayed, that is, evaluationof antifungal compounds.Although many organisms are available for the assay

of a variety of substances, certain bacteria have beenfound to be better suited as reagents. In actual labora-tory practice the variety used is limited.

Lactic acid-fermenting bacteria are used for the assayof vitamins and amino acids. Commercially availableantibiotics may be assayed using only five or six organ-isms. Most official procedures for testing antiseptics anddisinfectants specify Micrococcus pyogenes var. aureusor Salmonella typhosa as the test organism.The interests of a particular laboratory will deter-

mine the number and variety of organisms used. Forexample, a limitation of the number of bacteria used forassay purposes is of practical value in a laboratorywhere routine analyses of a variety of substances arebeing done. Valuable time is saved in the preparationof media and in the transfer of stock cultures. In alaboratory where investigational work is of major im-portance, the number of organisms used will be greater.Screening compounds for antibacterial activity requiresthe use of many organisms. Special problems, such asthe determination of substances in blood or naturalproducts, might require an organism sensitive to smallconcentrations of the particular substance. Again, itmust be emphasized that the needs and interests of theindividual laboratory or investigator determine whatorganisms should be used.

(b) Yeasts. Assay procedures, using yeasts as the re-agent, have been developed for thiamine (Williams etal.; Wooley, 1941), nicotinic acid (Williams, 1946),pyridoxine (Schultz et al., 1939), pantothenic acid(Williams et al. 1938), biotin (Snell et al., 1941), inositol(Williams et al. 1941), and cocarboxylase (Ochoa and

Peters, 1938). These organisms are generally consideredto be unsatisfactory. Snell et al. (1941) preferred theiruse over lactic acid bacteria in the assay of vitaminsfrom the standpoint of simplicity and minimum ex-penditure of time. However, Barton-Wright (1952)cautions against the use of yeasts for assay purposes onthe basis that the nitrogen requirements of these organ-isms need further definition. He criticizes the sources ofnitrogen used in basal media, indicating that testsamples might contain stimulatory polypeptides whichwould invalidate an assay where these substances in-creased the growth of the organism. In any case, wherean alternative method using another test organism isavailable, it should be used in preference to a methodusing a yeast.

In order that these organisms grow properly, aerationis necessary. This is usually accomplished by shakingthe tubes or flasks. In many laboratories this wouldinterfere with the processing of large numbers ofsamples.

(c) Fungi. Fungi are used in a number of methodsdeveloped for the assay of antibiotics, trace metals, andvitamins, and in the evaluation of fungicidal and fungi-static materials.

Fungi, in general, have the disadvantage of being slowgrowers, thus increasing the time required for assay.The Neurospora mutants, in particular, have twomajor disadvantages (Mitchell, 1950). (1) Growth issatisfactory on synthetic media, but materials such asyeast or liver extracts can double this growth. It isprobably due to nonspecific sparing actions by sub-stances in the extracts. An accurate assay using syn-thetic media cannot be expected unless the substancebeing assayed is present in sufficient amount that thesupplementary effects can be diluted out. (2) Dis-turbances in metabolism occur as a result of single genemutation, which at times would not be desirable inassay work.Fungi do not have as wide an application as assay

organisms as the bacteria do. They are used in a numberof procedures where no other organism is available orwhere fungicidal or fungistatic evaluation is desired.It is claimed that mutation techniques make it theo-retically possible to produce an assay organism for anyconstituent known to be a normal component of proto-plasm (Foster, 1949).

(d) Protozoa. Protozoa (Hamilton et al., 1952) havebeen found to have a variety of nutritional require-ments. In addition to vitamins and amino acids, certainof these microorganisms require nucleic acid derivativesand fatty factors. Vitamin B12 may be assayed usingEuglena gracilis (Hutner et al., 1949). Another procedureuses a member of the hemoflagellate group of parasiticprotozoa as an assay organism. The use of this groupof microorganisms for assay purposes is comparativelynew. The work, which has been reported by Hamilton

324 [VOL. 4

on April 3, 2020 by guest

http://aem.asm

.org/D

ownloaded from

ANALYTICAL MICROBIOLOGY. I

et al. (1952), seems promising. An organism which re-sponds to cholesterol has been found. There is anindication that hormones might be assayed micro-biologically using protozoa as test organisms. The useof these organisms might considerably broaden thefield of analytical microbiology.

4. Factors InvolvedIn microbiological assay procedures, controlled

growth of the microorganism is essential. The particularapproach to controlled growth depends upon whether apositive or negative growth response is desired. Apositive response is not necessarily the normal optimalgrowth response for the organism, but optimal onlywithin a range that is conditioned by the concentrationof essential nutrient supplied. On the other hand, anegative growth response is controlled within a rangethat is conditioned by the quantity of inhibitory sub-stance present, with all nutritional requirements foroptimum growth supplied to the organism. In eithercase, the factors which influence the growth of an organ-ism must be considered.

Gunsalus (1951) lists the factors which controlnormal microbial growth as follows: (a) Factors de-pendent upon the nature of the organism, that is, con-ditions essential for the initiation and continuation ofgrowth. (b) Factors dependent upon the environment,that is, composition and concentration of medium.(c) Influence of temperature, that is, minimum, maxi-mum, and optimum.For analytical procedures the factors are similar. It is

necessary to know how an organism grows under op-timum conditions before it can be used as a testorganism.

In any given assay, further emphasis is placed on:(a) age and/or condition of culture; (b) size of the inocu-lum; (c) exact temperature relationship, optimum onlyto the extent of the effect on the particular assay;(d) culture media; (e) oxygen relationships.

(a) Age and/or condition of culture. The culture mustbe maintained in such a manner that it will give asimilar response each time it is used in an assay pro-cedure. This involves both the age and the conditionof the culture. In assays that involve short incubationperiods, the culture should be used in the assay whenit is in the log phase of its growth cycle to minimize theeffect of lag phase. On the other hand, if an overnightincubation period is employed, the age may vary from18 to 24 hr, for the lag phase is then of less importance.The use of cultures older than 24 hr is to be avoided.Older cultures may not respond adequately for at leastfour reasons: (1) preponderance of dead organisms,resulting in insufficient growth response; (2) variationof essential growth factor requirements or changes insusceptibility to antimicrobial substances; (3) inabilityto standardize inoculum because of the variation in the

number of live cells; (4) possibilitv of autolysis with theresulting release of substances which might interferewith the assay.

Frequently, organisms on repeated transfer in thesame culture media mutate to the extent that they nolonger require certain growth factors (Pennington,1946; Snell, 1948). This is especially true in organismsused in the assay of growth factors. Fastidious organ-isms become less fastidious until their need for a givengrowth factor declines to such an extent that reliableassays are not obtained. Barton-Wright (1952) recom-mends the technique of Nymon and Gortner (1946) as aroutine procedure for all lactic bacteria used in assaywork. He maintains that only when the bacterial cellsare in a state of optimum vigor will they respond suc-cessfully as test organisms.A second important requirement of the stock culture

is its purity. A variety of results may be obtained witha contaminated culture. In a growth factor assay, fullgrowth mav be obtained at all concentrations. In anantimicrobial activity assay the contaminantmaynot besusceptible to the material being assayed. Both caseswould result in invalid assays and wasted time. Amicroscopic examination, together with a streak cultureof the test organism, should be a routine procedure inanalytical laboratories.

Several methods have been reported which may beused to insure a constant physiologic age of the inocu-lum. Nymon et al. (1945) used lyophilized cultures incertain vitamin assays. Foster and Woodruff (1943)pasturized spore suspensions which may be stored in arefrigerator for long periods of time. A further extensionof the refrigeration of cultures is their storage in dry ice(Heilman, 1944). Berridge and Barrett (1952) main-tained their inocula in the logarithmic phase by con-stant dilution before the optical density of the culturereached its maximum. The simple method of using anactively growing young culture of some predeterminedoptical density is the easiest and probably the mostpractical method of controlling the age of the culture.Harrison et al. (1951) carried this procedure further andprepared an inoculum reference graph. This graphrelated the density of the inocula to the volume re-quired to yield zones of the best character and a dosageresponse slope giving the greatest difference for stand-ard dose increments. They assumed that the variationin the ratio of viable cells to total bacterial count fromday to day was insignificant.

(b) Size of inoculum. Lack of control over the size ofthe inoculum used for an assay may contribute to thevariation found in day-to-day assay of the same ma-terial. This is particularly applicable to assays of shortduration, that is, 3 hr or less up to 18 to 24 hr, or inassays where the rate of growth rather than total growthis measured. In longer assays, differences in the size of

19561 325

on April 3, 2020 by guest

http://aem.asm

.org/D

ownloaded from

J. J. GAVIN

the inoculum used from one assay to another becomeless significant.

In turbidimetric and titrimetric assay of vitamins,the inoculum is commonly prepared by incubation ofthe culture for a specified period with subsequent dilu-tion. It is then added dropwise to each assay tube. Snell(1948) states that the final amount of growth is de-pendent almost solely on the concentration of the limit-ing vitamin and not on fluctuation in drop size. Inrecent years, the trend has been to standardize thedilution using a photoelectric colorimeter. An arbi-trary turbidity reading is selected and the inoculumdiluted to this value for each assay. The actual turbiditywill vary with the type of instrument used since theinstruments vary in sensitivity. After this standardiza-tion, the inoculum is added dropwise to the assay tubes.Certain authors have carried the standardization to agreater degree. Bulk inoculation of the assay mediumwith yeasts, use of a specific volume (Flynn et al., 1951)and use of a syringe rather than a pipette to controldrop size have been used (Black and Arnold, 1940;Thompson et al., 1950). The latest official procedurescall for the preparation of calibration curves for theinstrument used in the assay which relate transmit-tance for the photoelectric colorimeter, under condi-tions of the assay, to cell density (Official Methods ofAnalysis, 1955; The Pharmacopeia of the U.S.A.). Im-provements have been stimulated by day-to-day ob-servations of variation in inoculated blank tubes anddifferences in turbidity values for tubes of the sameconcentration of the same assay. Control of the inocu-lum size in these procedures can only lead to betterturbidity readings and greater reproducibility of results.

Turbidimetric antibiotic assays may vary with theinoculum size. Florey et al. (1949) mention that theapparent antibacterial potency of some antibioticsmay be many times greater with small inocula than withlarge, while the potency of others may not be affectedby this factor. Table 1 shows how the amount of inocu-lum effects the results obtained in the dilution assay ofpolymyxin.

TABLJ, 1. Effect of amount of inoculum on polymyxinassay values*

Inhibitory Conc.Inoculum Polymyxin (ug/MI)in Drops Broth

Media per100 ml v.BueiBroth vs. E. coli bronchel

septica

Difco broth ...................... 2 0.7 0.013Difco broth ...................... 10 6.0 0.37Difco broth ...................... 50 17.0 1.1Trypticase soy broth ............. 2 2.0 0.1Trypticase soy broth ............. 40 17.0 3.3

* Reese and Eisenherg, 1949.

TABLE 2. Effect of inoculum density*

Diametert of Zone of Exhibition for VitaminVolume of Inoculum per Plate B12 Concentration of(12.5 Ml of Agar Medium)t

0.05 pg/ml 0.2 pg/ml

ml mm mm

0.01 21.2 24.40.02 20.8 23.60.05 19.9 23.50.1 19.6 23.60.2 17.9 20.4

* Cuthbertson et al., 1951.t 15-16 Hr culture diluted

Wellcome turbidity scale.t Average of four cups.

to turbidity 4 on Burroughs-

Several authors have reported the same experiencesin the assay of this substance (White et al., 1949;Brownlee et al., 1949a and 1949b). Large inocula alsointerfere with the activity of Aureomycin (Paine et al.,1948a and 1948b) and Terramycin (Baron, 1950).In diffusion assays, for both antibiotics and vitamins,

the density of the inoculum or the quantity of organismsused in the inoculum affects the zone size obtained.Changes in inoculum sizes are correlated with changesin zone size (table 2).

In the assay of vitamin B12 using Lactobacillus lactis(ATCC 8000) Cuthbertson and Lloyd (1951) found thatlarge inocula are required for their plate procedure.

In all assay procedures the size of the inoculum canbe controlled. In method development, the microbiolo-gist should determine experimentally what size inocu-lum is optimum for a new procedure and standardize itat that particular point, thereby eliminating anothervariable.

(c) Temperature relationships. Although temperatureaffects the growth rate of the test organism and theoptimum temperature is defined (Gunsalus, 1951) as"that at which the organism has the minimum gener-ation time, that is, maximum growth rate," it is notnecessary that the incubation temperature used in anassay procedure be the optimum temperature for thetest organism. It should be, however, a temperature atwhich good growth will occur. The selected temperatureshould give the desired results in relation to both thetest organism and the test solution within the timelimitation set for the particular assay.For example, in the assay of antibiotics, Bacillus

subtilis will give results after 3jfi4 hr incubation at 37 Cor overnight at 25 to 28 C (Florey et al., 1949). In theassay of streptomycin by E. coli, better results are ob-tained when the incubation temperature is 28 C thanwhen it is 37 C (Brown and Young, 1947). Most tests forantiseptics, disinfectants, and fungicides are performedat an incubation temperature close to optimum for thetest organism.The range of optimum temperatures for organisms of

the lactic-acid group is wide. The official compendia

326 [VOL. 4

on April 3, 2020 by guest

http://aem.asm

.org/D

ownloaded from

ANALYTICAL MICROBIOLOGY. I

(Official Methods of Analysis, 1955; The Pharmacopeiaof the U.S.A.) permit selection of a temperature be-tween 30 and 37 C for the assay of riboflavin, folic acid,niacin, and vitamin B12, but require that the tempera-ture be kept constant to 40.5 C. Schweigert and Snell(1947) select the same range for the assay of aminoacids. However, in the assay of vitamin B12, J. F. Gain,Jr. (personal communication) has found that decidedlybetter growth is obtained when the assay is carried outat 37.5 than at 34 C.The selected temperature may be dictated by the

nature of the substance being assayed. An example ofthis is chlortetracycline, which is unstable at 37 C, thusnecessitating a lower temperature for the microbiologi-cal assay of this material. Experiments by Borek andWaelsch (1951) have demonstrated that temperaturechanges can affect the nutritional requirements ofLactobacillus arabinosus for amino acids. At tempera-tures up to 35 C, the organism grows readily in theabsence of phenylalanine or tyrosine. At 37 C, thismicroorganism does not grow unless one or the otheramino acid is supplied. At 39 C, aspartic acid is re-quired for growth. Yet this same organism may be usedto assay nicotinic acid at any temperature between 30to 37 C (Snell, 1950).The most important aspect of temperature in assay

procedures is the control of variations in incubators.Variations of from 3 to 4 C in an incubator set at 37 Cwere found by Price and Graves (1944). The changes intemperature caused variation in results in their assayof riboflavin. Circulating-air incubators and controlledwater-bath incubators are most effective in control ofthese temperature fluctuations.The selection of the proper incubation temperature is

important, but the actual choice often may be arbitrary.In the development of an assay procedure, the tempera-ture will depend upon the test organism, the materialto be assayed, and the amount of time in which theassay is to be completed. Further investigation into thetemperature relationships of assay organisms appearswarranted. Snell (1950) suggests that perhaps betterassays could be devised at temperatures outside theranges now used.

(d) Culture medium. It was previously mentionedthat microbiological assay procedures are dependentupon the reaction of the test organism with its environ-ment. The culture medium is, therefore, an importantfactor in the success of a given assay.An assay that is satisfactory must give reproducible

results. The culture medium used in any assay mayaffect this reproducibility within or among laboratories.First, within a laboratory, consistent results can be ob-tained only by rigid control of the composition of theculture medium. Variations in the concentration ofcertain ingredients is to be avoided in routine assays.When the culture medium is not quantitatively pre-

pared, environmental conditions are changed and thetest organism may exhibit variation in growth to agreater or lesser degree. In analytical procedures whichmay vary to the extent of ±15 per cent, control of theseapparently minor factors may aid in reducing the over-all error.

Secondly, a further extension in the control of thecomposition of the culture medium is necessary forlaboratories to obtain results that other laboratoriescan reproduce. It is essential to standardize the culturemedium used for a particular assay. Neter et al. (1952)determined that the composition of the culture mediuminfluenced the activity of several antibiotics. In theassay of antiseptics, disinfectants, and fungicides, theresistance of the test organism will vary with changesin the specified culture media (Brewer, 1942). Stand-ardization of media for testing such compounds is re-ported by several investigators (Klarmann and Wright,1945; Ostrolink and Brewer, 1949; Wolf, 1945).A variety of media are available for the assay of

different growth factors. This presents a problem in theevaluation of results obtained using different media.A single medium has been devised for the assay of folicacid and riboflavin, and nicotinic acid and vitamin B,2(Flynn et al., 1951). Mayernick and Ewald (1951) haveprepared a dry composite medium for the determinationof ten amino acids.

It has been suggested by Price et al. (1947) thatlaboratories maintain a reference-standard mediumwhich would be used with a sensitive test organism totest new batches of medium. They have suggested use ofa factor to interpret results from different laboratories.An example of the effect of culture media upon the

activity of a compound is in the evaluation of strepto-mycin. Table 3 (May et al., 1947) shows the differencein streptomycin activity in two different culture mediaagainst a number of organisms.

It is obvious that the effect of the culture media onthe antimicrobial activity of this compound is great.Other studies have shown similar differences (Bondiet al., 1946). Variations in tryptone (Donovick andRake, 1946), the nitrogenous constituents of themedium (Green and Waksman, 1948), and the substi-tution of neopeptone for asparagine in McLeod's mediumhave all changed the minimum inhibitory concentrationof streptomycin in in vitro tests (Lenert and Hobby,1947). Amino acid requirements of certain organismsmay vary with different basal media (Snell, 1945). Thelysine, threonine, and alanine requirements of L. casei,L. arabinosus, and L. delbrueckii are eliminated whenpyridoxine in the basal medium is replaced by an equalamount of pyridoxamine (Stokes and Gunness, 1945).For all assay culture media, there must be some mini-

mum standards. The requirements for an ideal culturemedium for use in analytical procedures are as follows:(1) It must contain all factors necessary to support

1956] 327

on April 3, 2020 by guest

http://aem.asm

.org/D

ownloaded from

TABLE 3. Comparison of activity of streptomycin hydrochloride against bacteria in two different media*

Range of Concentrations Completely Inhibiting Growth, 1 Part

Organism Number ofStrains Studied

In broth In glucose phenol redserum water

Bacterium friedlanderi ................. .......................... 5 30,000-1,500,000 2,000,000-12,000,000Bacterium coli .................. ..................... 18 40,000-160,000 500,000-2,000,000Pseudomonas pyocyanea ....................................... 5 20,000-40,000 1,000,000-4,000,000Proteus ......................................................... 6 10,000-20,000 500,00 -2,000,000Salmonella typhi . ................................................ 8 20,000-80,000 500,000-4,000,000Salmonella paratyphi A ................ .......................... 4 20,000-40,000 1,000,000-2,000,000Salmonella paratyphi B ................ .......................... 5 20,000-40,000 500,000-1,000,000Salmonella paratyphi C .......................................... 1 20,000 1,000,000Salmonella sp .................................................... 6 20,000-160,000 500,000-2,000,000Shigella sp ...................................................... 4 160,000 1,000,000Streptococcus pyogenes ................. .......................... 1 160,000 1,000,000Streptococcus viridans ..................... ....................... 7 50,000-400,000 1,000,000-6,000,000Staphylococcus .................................................. 2 320,000 4,000,C00Hemophilus influenzaet .......................................... 4 100,000 600,000

* May et al., 1947.t Fildes extract added to medium for H. influenzae.

growth. The test organism must be supplied with all thenutrients that it needs for normal growth and metabo-lism, including (a) source of energy; (b) source ofnitrogen; (c) mineral salts (specific ions); (d) growthfactors when applicable (employment of syntheticmedia).

In obtaining a positive growth response, as in theassay of growth factors, this requirement is met bysupplying graded amounts of the material being as-sayed. In this case, the culture medium must be pre-pared so that all of the necessary nutrients are availableto the test organism, with the exception of the one thatwill control growth. Any absence of growth must bedue only to the lack of test material, or the assay is notvalid.

This also applies to the negative growth response.Lack of growth must be due only to the presence of someinhibitory substance, rather than to a missing nutrientwhich is essential to the normal growth and develop-ment of the test organism. The variation in the nutri-tional requirements of the organisms used for analyticalpurposes is as great as the number of organisms.Culture media range from the relatively simple mediaused in the assay of trace metals by fungi and the assayof vitamins by Neurospora mutants to the highly com-plex media used in the assay of vitamins and anti-biotics.

(2) It must not contain any constituent which affectseither the substance being assayed or the test organism.This requirement may be affected by substances whichfall into three categories-stimulants, substitutes, orantagonists. Many substances have been found to havea stimulatory effect on microorganisms. For example,fat in certain concentration has a stimulatory effect onthe test organism used in the assay of riboflavin(Bauernfeind et al., 1942; Strong and Carpenter, 1942),

pantothenic acid (Bauernfeind et al., 1942; Neal andStrong, 1943), and biotin (Williams and Fieger, 1945).

Starch, in the presence of riboflavin, has thissame effect on Lactobacillus casei (Scott et al., 1941;Andrews et al., 1942). The zones of inhibition in theassay of antibiotics are also affected by stimulatorysubstances. Sugars (Bond, 1952; Knowlden, 1955),peptones, salt concentration (Loo et al., 1945), andyeast extracts (Grenfell et al., 1947) modify the zonesobtained. A growth increase in certain antibiotic assaysis said to be the result of subbacteriostatic concentra-tions of antibiotics (Dufrenroy and Pratt, 1947). In theassay of viomycin, a subinhibitory concentration ofsulfadiazine increases sensitivity of the assay (Dye etal., 1952). The use of impure amino acids in culturemedia prepared for use in the assay of individual aminoacids will lead to high blanks and invalid assays (Barton-Wright, 1950). The minerals present in water usedin the assay of trace metals must be removed if theassay is to be valid.

Stimulation of the activity of the material beingtested is also considered in this category. For example,serum albumin will enhance the activity of gramicidinagainst a hemolytic streptococcus several times (Floreyet al., 1949). The inclusion of sorbitan monooleate(Tween-80) in the culture medium will enhance the ac-tivity of subtilin against M. tuberculosis, and increasethe sensitivity of polymyxin assay (table 4).The second category of substances includes those

whose activity will affect the test organism in the samemanner as the material under test. These may be calledsubstitutes. Thymidine in excessive quantities will sub-stitute for folic acid when Streptococcus faecalis is usedas a test organism in the assay of the latter compound(Snell and Mitchell, 1941; Stokes, 1944). Thiamineaffects Neurospora sitophilia so that the organism does

328 J. J. GAVIN [VOL. 4-

on April 3, 2020 by guest

http://aem.asm

.org/D

ownloaded from

ANALYTICAL MICROBIOLOGY. I

TABLE 4. Titration of polymyxin with and withoutTween-80*

Zone DiameterConc. Polymyxin

Control Tween-80

units/ml mm mm

128 26.7 28.864 24.3 25.732 19.7 23.2516 16.7 20.08 0 16.04 0 0

* Stansly and Schlosser, 1947.

not respond quantitatively to pyridoxine (Morris et al.,1949). The desoxyribosides and thymidine may replacevitamin B12 in the nutrition of L. lactis and L. leich-mannii (Cuthbertson et al., 1951).

Finally, there is an antagonistic effect. An evaluationof sulfa drugs would be invalid if there were an excess ofpara-aminobenzoic acid in the culture medium. Strepto-mycin is inactivated by incubation with cysteine(Florey et al., 1949), glucose, sulphydryl, and ketonecompounds (Baron, 1950). Chlortetracycline activity isreduced 55 to 96 per cent by decomposition in variousculture media (Price et al., 1948) and oxytetracycline,similar in structure to chlortetracycline, also deterioratesin culture media. Penicillin is affected by amino acids(Baron, 1950). Growth factors also have their antago-nists. It was previously mentioned that fats may stimu-late the response of L. casei to riboflavin. Dependingupon the concentration, they may also inhibit the growthresponse (Snell, 1950). Amino acid antagonisms are re-

ported in the literature (Brickson et al., 1948; Glad-stone, 1939; Gunsalus, 1951). This can be overcome byeither the number or the concentration of the aminoacids in the basal medium (Barton-Wright, 1952;Gunsalus, 1951; Schweigert and Snell, 1947).

It is important that the culture medium prepared foran assay of any given material be free from componentsthat would cause invalid results.

(3) It must have a pH which will not interfere witheither the activity of the substance being assayed or thegrowth of the test organism. The pH of the culturemedium must be in a range that will allow the organismto grow readily. In the case of organisms producing by-products which are acid in nature, suitable buffers mustbe incorporated in the medium. As indicated previously,absence of growth should be due only to the lack oftest material when determining positive growth re-

sponse, and due only to the presence of an inhibitorysubstance in negative response.A second problem is involved in setting limits for pH.

The pH must be favorable for the optimum activity ofthe substance being tested. Generally, this can be con-

trolled by the use of buffer solutions in preparing thesample for assay. In order to gain optimum conditions,

it might be necessary to vary both the pH of the culturemedium and the pH of the assay solution.

(e) Oxygen relationships. This factor is of relativelyminor importance. Several reports in the literature indi-cate that the oxidation-reduction potentials of culturemedia might contribute to vitamin assay variation(Bohonos et al., 1942; Geiger et al., 1946; Koditscheket al., 1949; Niven and Smiley, 1943). The addition ofcysteine, thioglycollate, or ascorbic acid will helpminimize this, but in most assay procedures this pre-caution is unnecessary.

Certain organisms, such as Acetobacter suboxydansused for the assay of panthenol, and yeasts in general,require a supply of oxygen for heavy growth. This isaccomplished either by continuous shaking during in-cubation or by the use of flasks which will give a favor-able ratio of surface to volume.

Parrish et al. (1955) reduced variation between repli-cate tubes in vitamin B6 assay with Saccharomycescarlsbergensis by placing a glass bead in each assaytube. They attribute this reduction to an increase in theuniformity of agitation during the incubation period.

Streptomycin (Geiger et al., 1946) and chlortetra-cycline (Paine et al., 1948b) are both affected byanaerobiosis. The examples cited above show that thereare special cases where the oxygen requirement of thetest organism is important to the success of the assayand should not be overlooked.

REFERENCES

ANDREWS, J. S., BOYD, H. M., AND TERRY, D. E. 1942 Ribo-flavin analysis of cereals. Application of the microbio-logical method. Ind. Eng. Chem., Anal. Ed., 14, 271-274.

BARON, A. L. 1950 Handbook of Antibiotics. ReinholdPublishing Corporation, New York, N. Y.

BARTON-WRIGHT, E. C. 1950 Some aspects of the micro-biological assay of amino acids. Chemistry & Industry,69, 845-848.

BARTON-WRIGHT, E. C. 1952 The Microbiological Assay ofthe Vitamin B-Complex and Amino Acids. Pitman Pub-lishing Company, New York, N. Y.

BAUERNFEIND, J. C., SOTIER, A. L., AND BORUFF, C. S. 1942Growth stimulants in the microbiological assay for ribo-flavin and pantothenic acid. Ind. Eng. Chem., Anal.Ed., 14, 666-671.

BERRIDGE, N. J. AND BARRETT, J. 1952 A rapid method forthe turbidimetric assay of antibiotics. J. Gen. Micro-biol., 6, 14-20.

BLACK, T. L. AND ARNOLD, A. 1940 Inoculation procedure.Ind. Eng. Chem., Anal. Ed., 12, 344.

BOHONOS, N., HUTCHINGS, B. L., AND PETERSON, W. H. 1942Pyridoxine nutrition of lactic acid bacteria. J. Bacteriol.,44, 479-485.

BOND, C. R. 1952 The Staphylococcus aureus plate assay ofpenicillin: Effects of sugars on zone edges and accuracy ofresults. Analyst, 77, 118-123.

BONDI, A., JR., DIETZ, C. C., AND SPAULDING, E. H. 1946Interference with the antibacterial action of streptomy-cin by reducing agents. Science, 103, 399-401.

BOREK1, E. AND WAELSCH, H. 1951 The effect of temperature

1956] 329

on April 3, 2020 by guest

http://aem.asm

.org/D

ownloaded from

J. J. GAVIN

on the nutritional requirements of microorganisms. J.

Biol. Chem., 190, 191-196.BREWER, C. M. 1942 Use and abuse of Staphylococcus aureus

as a test organism. Am. J. Public Health, 32, 401-405.BRIAN, P. W., CURTIS, P. J., AND HEMMING, H. G. 1946 A

substance causing abnormal development of fungal hyphae

produced by Penicillium janczewskii Zal. I. Biologicalassay, production and isolation of "curling factor."

Brit. Mycol. Soc. Trans., 29, 173-187.BRIAN, P. W. AND HEMMING, H. G. 1945 Gliotoxin, a fungi-

static metabolic product of Trichoderma viride. Ann.Appl. Biol., 32, 214-220.

BRICKSON, W. L., HENDERSON, L. M., SOLHJELL, I., AND EL-

VEHJEM, C. A. 1948 Antagonism of amino acids in the

growth of lactic acid bacteria. J. Biol. Chem., 176, 517-528.

BROWN, A. M. AND YOUNG, P. A. 1947 A dilution method

for the assay of streptomycin. J. Gen. Microbiology, 1,

353-360.BROWNLEE, G., BUSHBY, S. R. M., AND SHORT, E. I. 1949a

Comparative biological studies of polymyxin A and D.

Ann. N. Y. Acad. Sci., 51, 891-896.BROWNLEE, G., BUSHBY, S. R. M. AND SHORT, E. I. 1949b

The pharmacology of polymyxin A, B, and D. Ann.

N. Y. Acad. Sci., 51, 952-967.

CUTHBERTSON, W. F. J. AND LLOYD, J. T. 1951 The assay

of vitamin B12. I. Factors affecting the response of L.lactis Dorner (ATCC 8000) to vitamin B12. J. Gen. Micro-

biol., 5, 416-420.

CUTHBERTSON, W. F. J., PEGLER, H. F., AND LLOYD, J. T.

1951 The assay of vitamin B12. Part III. Microbiological

estimation with Lactobacillus lactis Dorner by the platemethod. Analyst, 76, 133-141.

DIMICK, K. P. 1943 Quantitative method for the determina-tion of tyrothricin. J. Biol. Chem., 149, 387-393.

DONOVICK, R. AND RAKE, G. 1946 Influence of certain sub-stances on activity of streptomycin. I. Modifications

in test medium. Proc. Soc. Exp. Biol. Med., 61, 224-227.DUFRENROY, J. AND PRATT, R. 1947 Cytochemical mecha-

nisms of penicillin action. I. Oxidation-reduction levels

J. Bacteriol., 53, 657-666.DYE, W. E., LYNCH, H. P., CULLEN, T. M., AND SPRAGUE, E.

M. 1952 Use of synergism to increase the sensitivity

of a microbiological method for the assay of viomycin.

Antibiotics & Chemotherapy, 2, 610-614.FLOREY, H. W., CHAIN, E., HEATLY, N. G., JENNINGS, M. A.,

SANDERS, A. G., ABRAHAM, E. P., AND FLOREY, M. E. 1949Antibiotics, Vol. I, Chapter 3. Oxford University Press,London, New York.

FLYNN, L. M., WILLIAMS, V. B., O'DELL, B. L., AND HOGAN,A. G. 1951 Medium for the assay of vitamins withlactic acid bacteria. Anal. Chem., 23, 180-185.

FOSTER, J. W. 1949 Chemical Activities of Fungi, p. 248.Academic Press Inc., New York, N. Y.

FOSTER, J. W. AND WOODRUFF, H. B. 1943 Improvements inthe cup assay for penicillin. J. Biol. Chem., 148, 723.

GEIGER, W. B., GREEN, S. R., AND WAKSMAN, S. A. 1946 Theinactivation of streptomycin and its practical applica-

tions. Proc. Soc. Exp. Biol. Med., 61, 187-192.GLADSTONE, G. P. 1939 Interrelationships between amino

acids in the nutrition of B. anthracis. Brit. J. Exptl.Pathol., 20, 189-200.

GREEN, S. R. AND WAKSMAN, S. A. 1948 Effect of gluicose,peptone and salts on streptomycin activity. Proc. Soc.

Exp. Biol. Med., 67, 281-285.GRENFELL, E., LEGGE, B. J., AND WHITE, T. 1947 Micro-

biology of submerged culture of P. chrysogenum. J.Gen. Microbiol., 1, 171-185.

GUNSALUS, I. C. 1951 Growth of bacteria. Bacterial Phys-iology. Ed. by C. H. Werkman and P. W. Wilson. Aca-demic Press Inc., New York, N. Y.

HAMILTON, L. D., HUTNER, S. H., AND PROVASOLI, L. 1952The use of Protozoa in analysis. Analyst, 77, 618-628.

HARRISON, E., LEES, K. A., AND WOOD, F. 1951 The Assay

of Vitamin B12. Part IV. Microbiological estimationwith a mutant of Escherichia coli by the plate method.Analyst, 76, 696-705.

HEILMAN, D. H. 1944 Method for standardizing penicillin.Am. J. Med. Sci., 207, 477-483.

HOAGLAND, C. I. AND WARD, S. M. 1942 The quantitative

determination of factor V by measurement of nitrate pro-

duced by Hemophilus influenzae. J. Biol. Chem., 146,115-122.

HUTNER, S. H., PROVASOLI, L., STOKSTAD, E. L. R., HOFFMANN,C. E., BELT, M., FRANKLIN, A. L., AND JUKES, T. H. 1949Assay of antipernicious anemia factor with Euglena.Proc. Soc. Exptl. Biol. Med., 70, 118-120.

KLARMANN, E. G. AND WRIGHT, E. S. 1945 Synthetic and

semisynthetic media for disinfectant testing. SoapSanit. Chemicals, 21, 113-119.

KNOWLDEN, N., BROOMFIELD, E., AND HERMAN, J. 1955 In-terference of sucrose in the plate assay of polymyxin B.

J. Am. Pharm. Assoc., Sci. Ed., 44, 231-234.KODITSCHEK, L. K., HENDLIN, D., AND WOODRUFF, H. B. 1949

Investigations on the nutrition of Lactobacillus lactisDorner. J. Biol. Chem., 179, 1093-1102.

LENERT, T. F. AND HOBBY, G. L. 1947 Observations on theaction of streptomycin in vitro. Proc. Soc. Exp. Biol.Med., 65, 235-249.

Loo, Y. H., SKELL, P. S., THORNBERRY, H. H., EHRLICH, J.,MCGUIRE, J. M., SAVAGE, G. M., AND SYLVESTER, J. C.1945 The assay of streptomycin by paper disc platemethod. J. Bacteriol., 50, 701-709.

MAY, J. R., VOUREKA, A. E., AND FLEMING, A. 1947 Someproblems in titration of streptomycin. Brit. Med. J.,I, 627-630.

MAYERNIK, J. J. AND EWALD, S. M. 1951 Dry basal mediafor rapid microbiological assay of amino acids. J. Am.Pharm. Assoc., Sci. Ed., 40, 462-464.

MITCHELL, H. K. 1950 Vitamins and metabolism in Neuro-spora. Vitamins and Hormones, vol. VIII, p. 127. Aca-demic Press Inc., New York, N. Y.

MORRIS, S., HERWIG, G., AND JONES, A. 1949 A study of theStokes method for the assay of Vitamin B6 with Neurosporasitophilia M. 299. Analyst, 74, 37-43.

NEAL, A. L. AND STRONG, F. M. 1943 Microbiological de-termination of pantothenic acid. Ind. Eng. Chem., Anal.Ed., 15, 654-657.

NETER, E., MURDOCK, M. R., AND KUNZ, E. H. 1952 Theinfluence of the composition of culture media on growthinhibition of various microorganisms by antibiotic discs.Antibiotics & Chemotherapy, 2, 35-41.

NIVEN, E. F., JR. AND SMILEY, K. L. 1943 A microbiologicalassay for thiamine. J. Biol. Chem., 150, 1-9.

NYMON, M. C. AND GORTNER, W. A. 1946 Some culturestudies on Lactobacillus arabinosus and Lactobacilluscasei. J. Biol. Chem., 163, 277-282.

NYMON, M. C., GUNSALUS, I. C., AND GORTNER, W. A. 1945An application of the lyophile process to the maintenanceof cultures for microbiological assay. Science, 102, 125-126.

Official Methods of Analysis of the Association of Official Agri-cultural Chemists, 8th ed. 1955 Washington, D. C.

330 [VOL. 4

on April 3, 2020 by guest

http://aem.asm

.org/D

ownloaded from

ANALYTICAL MICROBIOLOGY. I

OCHOA, S. AND PETERS, R. A. 1938 Vitamin B1 and cocar-boxylase in animal tissues. Biochem. J., 32, 1501-1515.

OSTROLENK, M. AND BREWER, C. M. 1949 A bactericidalspectrum of some common organisms. J. Am. Pharm.Assoc., Sci. Ed., 38, 95-106.

PAINE, T. F., JR., COLLINS, H. S., AND FINLAND, M. 1948aBacteriologic studies on Aureomycin. J. Bacteriol.,56, 489-498.

PAINE, T. F., JR., COLLINS, H. S., AND FINLAND, M. 1948bLaboratory studies with Aureomycin. Ann. N. Y. Acad.Sci., 51(2) 228-230.

PARRISH, W. P., Loy, H. W., AND KLINE, 0. L. 1955 A studyof 'the yeast method for vitamin B6. J. Assoc. Offic.Agr. Chemists, 38, 506-513.

PFLNNINGTON, D. 1946 Assay of p-aminobenzoic acid.Science, 103, 397.

The Pharmacopeia of the United States of America, FifteenthRevision U.S.P. XV. Mack Publishing Company, Eas-ton, Pa.

PRICE, S. A. AND GRAVES, H. C. H. 1944 Microbiologicalassay of riboflavin. Nature, 153, 461.

PRICE, C. W., RANDALL, W. A., CHANDLER, V. L., AND REEDY,R. J. 1947 Observations on the in vivo and in vitro de-velopment of bacterial resistance to streptomycin. J.Bacteriol., 53, 481-488.

PRICE, C. W., RANDALL, W. A., AND WELCH, H. 1948 Bac-teriological studies of Aureomycin. Ann. N. Y. Acad.Sci., 51, 211-217.

RAKE, G., McKEE, C. M., AND JONES, H. 1943 A rapid testfor the activity of certain antibiotic substances. Proc.Soc. Exptl. Biol. Med., 51, 273-274.

REESE, E. T. AND EISENBERG, G. M. 1949 Polymyxin assayprocedures. Ann. N. Y. Acad. Sci., 51, 968-979.

SCHOPFER, WV. H. 1935 Crystalline B1 as growth hormone formicroorganisms (Phycomyces). Arch. Microbiol., 61,139-140.

SCHULTZ, A. S., ATKIN, L., AND FREY, C. N. 1939 VitaminB6, a growth promoting factor for yeast. J. Am. Chem.Soc., 61, 1931.

SCHWEIGERT, B. S. AND SNELL, E. E. 1947 Microbiologicalmethods for the estimation of amino acids. NutritionAbstr. & Rev., 16, 497-510.

SCOTT, M. L., RANDALL, F. E., AND HESSEL, F. H. 1941 Amodification of the Snell and Strong microbiologicalmethod for determining riboflavin. J. Biol. Chem., 141,325-326.

SNELL, E. E. 1945 The microbiological assay of amino acids.Advances in Protein Chem., II, 85-118.

SNELL, E. E. 1948 Use of microorganisms for assay of vita-mins. Physiol. Revs., 28, 255-282.

SNELL, E. E. 1950 Microbiological methods in vitamin re-search. Vitamin Methods, Vol. I, ed. by Paul Gybrgy.Academic Press, Inc., New York, N. Y.

SNELL, E. E., EAKIN, R. E., AND WILLIAMS, R. J. 1941 Assaymethod for biotin. Univ. Texas Publ., 4137, 18-21.

SNELL, E. E. AND MITCHELL, H. K. 1941 Purine and pyrimi-dine bases as growth substances for lactic acid bacteria.Proc. Natl. Acad. Sci., 27, 1-7.

STANSLY, P. G. AND SCHLOSSER, M. E. 1947 Studies on poly-myxin: method of assay. J. Bacteriol., 54, 585-597.

STOKES, J. L. 1944 Substitution of thymine for "folic acid"in the nutrition of lactic acid bacteria. J. Bacteriol., 48201-209.

STOKES, J. L. AND GUNNESS, M. 1945 Pyridoxamine and thesynthesis of amino acids by Lactobacilli. Science, 101,43-44.

STRONG, F. M. AND CARPENTER, L. E. 1942 Preparation ofsamples for microbiological determination of riboflavin.Ind. Eng. Chem., Anal. Ed., 14, 909-913.

THOMPSON, H. T., DIETRICH, C. S., AND ELVEHJEM, C. A.1950 The use of Lactobacillus leichmannii in the estima-tion of vitamin B12 activity. J. Biol. Chem., 184, 175-180.

WHITE, H. J., ALVERSON, C. M., BAKER, M. J., AND JACKSON,E. R. 1949 Comparative biological studies of poly-myxin and "aerosporin." Ann. N. Y. Acad. Sci., 51,879-890.

WILLIAMS, W. L. 1946 Yeast microbiological method for thedetermination of nicotinic acid. J. Biol. Chem., 166.397-406.

WILLIAMS, V. R. AND FIEGER, E. A. 1945 Growth stirnu-lants for microbiological biotin assay. Ind. Eng. Chem..Anal. Ed., 17, 127-130.

WILLIAMS, R. J. and associates 1941 Studies on the vita-min content of tissues, I., Univ. Texas Publ. 4137, pp. 1-87.1942 Studies of the vitamin content of tissues, II., Univ.Texas PubI. 4237, pp. 1-145.

WILLIAMS, R. J., TRUESDAIL, J. H., WEINSTOCK, H. H. JR.,ROHRMANN, E., LYMAN, C. M., AND MCBURNEY, C. H.

1938 Pantothenic acid. II. Its concentration and puri-fication from liver. J. Am. Chem. Soc., 60, 2719-2723.

WOLF, P. A. 1945 A medium containing an acid casein hy-drolysate for use in testing disinfectants. J. Bacteriol.,49, 463-472.

WOOLEY, D. W. 1941 Destruction of thiamine by a substancein certain fish. J. Biol. Chem., 141, 997-998.

1956] 331

on April 3, 2020 by guest

http://aem.asm

.org/D

ownloaded from