CLINICAL MICROBIOLOGY REVIEWS, Jan. 1991, p. 20-34 Vol. 4, No. 10893-8512/91/010020-15$02.00/0Copyright © 1991, American Society for Microbiology

Rhodococcus equi: an Animal and Human PathogenJOHN F. PRESCOTT

Department of Veterinary Microbiology and Immunology, University of Guelph, Guelph, Ontario NIG 2WI, Canada

INTRODUCTION................................................................. 20TAXONOMIC STATUS................................................................. 20EPIDEMIOLOGICAL ASPECTS OF INFECTION ................................................................. 21INFECTIOUS DISEASE: CLINICAL AND PATHOLOGICAL MANIFESTATIONS ...........................21

Horses.................................................................. 21Clinical manifestations in foals ................................................................. 22Clinical manifestations in adults................................................................. 22Pathological manifestations in foals................................................................. 22

Humans................................................................. 23Natural Infections in Other Species ................................................................. 23Experimental Infections in Animals ................................................................. 24

VIRULENCE AND PATHOGENIC MECHANISMS................................................................. 24IMMUNITY TO INFECTION .................................................................. 25Humoral Immunity in Horses.................................................................. 25Cellular Immunity in Horses ................................................................. 25Immunity to Infection in Humans .................................................................. 26Immunotherapy in Horses ................................................................. 26

LABORATORY DIAGNOSIS ................................................................. 27Isolation and Identification of the Organism ................................................................. 27

Isolation and colonial morphology ................................................................. 27Microscopic morphology and staining properties ................................................................. 27Identifying biochemical characteristics ................................................................. 27Other identifying characteristics ................................................................. 28Capsular serotypes ................................................................. 28

Diagnostic Procedures.................................................................. 28TREATMENT AND CONTROL.................................................................. 29

Antimicrobial Susceptibility ................................................................. 29CONCLUSIONS ................................................................. 30ACKNOWLEDGMENTS .................................................................. 30REFERENCES ....................................................................... 30

INTRODUCTION

Rhodococcus equi is a well-recognized bacterial pathogenin veterinary medicine. First isolated from Swedish foals byMagnusson in 1923 (96), it causes an important chronicgranulomatous pneumonia and lung abscesses in foals agedunder 4 months and is a common isolate from cervical lymphnodes in swine. Although rare, infection also occurs in awide variety of other mammals, often following immunosup-pression of various causes. Infections in these unusual hostscommonly include granulomatous pneumonia which devel-ops into lung abscesses, lymphadenitis (often of the mesen-teric, bronchial, or cervical lymph nodes), wound infections,and abscesses in various parts of the body. Before 1983, only12 cases had been reported in humans (176). At least 20additional cases have been described since then, the major-ity of which have been in patients with AIDS. The increasednumber of human cases reported recently is partly the resultof the spread of AIDS but may also reflect the increasingawareness by medical laboratories of this opportunisticpathogen and their improved ability to identify it rather thanto dismiss it as a contaminating "micrococcus" or "diph-theroid." Our current understanding of R. equi comes pri-marily from equine research, though it can be applied in partto the organism as a human opportunist pathogen. Otherreviews are available (6, 42, 43). The present review empha-

sizes the questions that a medical microbiologist might askwhen faced with a bacterial isolate that needs confirming asR. equi, as well as the questions that physicians might askwhen treating a patient with confirmed R. equi infection.

TAXONOMIC STATUS

Magnusson in 1923 proposed the name Corynebacteriumequi for isolates from foals with pyogranulomatous pneumo-nia (96). The subsequent confusing taxonomic history of theorganism, which is currently described as both Corynebac-terium and Rhodococcus in the Approved List of BacterialNames (153), has been generally resolved with improve-ments in the methods used to classify the nocardioformactinomycetes (62, 63). The genus Rhodococcus ("red-pigmented coccus") belongs to the phylogenetic group de-scribed as nocardioform actinomycetes, which contains thegenera Caseobacter, Corynebacterium, Mycobacterium,Nocardia, Rhodococcus, and the "aurantiaca" taxon (61,62), the last of which has recently been proposed as thegenus Tsukamurella (33). These genera are gram-positive,aerobic, catalase-positive bacteria which, in part because oftheir diverse microscopic and macroscopic morphology, arebest characterized by biochemical criteria (22, 61-64).

Within the nocardioform actinomycetes, the genus Rho-dococcus is characterized by rod-to-coccus morphologic

20

on June 29, 2020 by guesthttp://cm

r.asm.org/

Dow

nloaded from

RHODOCOCCUS EQUI 21

variation during the growth cycle, by the presence of diphos-phatidylglycerol, phosphatidylethanolamine, and phospha-tidylinositol mannosides; mycolic acids of carbon length 34to 64 with up to four double bonds; dihydrogenated mena-quinones with either eight or nine isoprene units as the majorisoprenalog; large amounts of straight-chain, unsaturatedfatty acids; and tuberculostearic acid. The genus Rhodococ-cus (61) is heterogeneous and can be divided into twogroups. Those species originally classified in the genusGordona (R. bronchialis, R. rubropectinus, and R. terrae)have 48- to 66-carbon-atom mycolic acids and nine isoprenedihydrogenated menaquinones and produce mycobactins(61). All other species produce 34- to 52-carbon mycolicacids with up to two double bonds and eight isoprenedihydrogenated menaquinones, but not mycobactins (61, 67,173). Recently, Stackebrandt and others (157) have rein-stated the genus Gordona for the three species originally inthis genus while keeping the genus Rhodococcus for thesecond group. This redefined genus Rhodococcus contains12 recognized species (3, 63, 157, 195) that are widelydistributed in nature, particularly in soil and herbivoremanure. Besides R. equi, which is the species with the mostpathogenic potential for animals including humans, otherspecies with pathogenic potential are four Rhodococcusspecies listed as species incertae sedis (61), isolated from thesputa of patients with pulmonary disease (174, 175), and R.fascians, a plant pathogen. Unidentified Rhodococcus spe-cies have been isolated from a variety of noncaseatinggranulomatous lesions in the lungs, lymph nodes, pleura,meninges, pericardium, and skin of human patients, oftenassociated with immunosuppressive disease processes ordrug treatments (175). Little is known about the pathogenic-ity for animals or humans of rhodococci other than R. equi.

EPIDEMIOLOGICAL ASPECTS OF INFECTION

R. equi is largely a soil organism with simple growthrequirements, which appear to be met perfectly by herbivoremanure and summer temperatures in temperate climates(126). The organism is widespread in grazing animals andtheir environment. Organisms have been isolated from thefeces or intestines of a high proportion of herbivores oromnivores, including cattle, deer, goats, horses, pigs, andsheep (23, 184, 188); and from the manure of a high propor-tion of wild birds but uncommonly in the manure of chickens(23). By contrast, the organism appears rare in the feces ofdogs and was not isolated from cats (23), and only twoisolates have been made from 521 human fecal samples(115).As an obligate aerobe (61), R. equi is unlikely to multiply

in the anaerobic environment of the large bowel of herbi-vores (7, 119, 188) and therefore is unlikely to be normalflora in the intestine. It might, however, multiply in the moreaerobic environment of the small bowel. Although there isno direct evidence that it multiplies in the intestine of adultherbivores, the organism multiplies in the first 8 weeks of lifewithin the foal intestine (167). This multiplication, however,ceases by 12 weeks of age, probably when an adult type ofintestinal flora has developed (78, 167). The widespreadpresence of the organism in adult herbivore feces seemslargely to reflect acquisition from feed contaminated with R.equi (7, 169). The widespread presence of R. equi antibodyin the horse population (discussed below) suggests thatantigenic stimulation by organisms either transiently passingthrough or perhaps growing in the intestine is a commonevent. The highest numbers of R. equi are found in the

surface soil on infected-horse farms (166), which would beexpected of aerobic organisms dependent for growth onsimple nutrients derived from herbivore manure. The multi-plication of R. equi in herbivore manure can be striking. Forexample, Barton and Hughes reported that R. equi multi-plied 10,000-fold in horse manure kept under Australianspring temperatures (daytime temperatures fluctuating from15 to 37°C; nighttime, from 5 to 9°C) for 14 days (7).The multiplication of R. equi in soil depends on environ-

mental temperatures, the presence of volatile fatty acidsfrom herbivore manure, and on soil pH (78, 160). Theorganism does not multiply at 10°C or below (78, 160, 166).Differences in environmental conditions between years mayexplain the annual variation in number of cases of R. equidiagnosed in foals (144). In one study, the number of R. equiisolated from the air on horse farms rose with environmentaltemperatures and was highest on dry and windy days (160).Aerosol infection via dust is thought to be the major route offoal infection (155).

R. equi infection in horses occurs endemically on somehorse farms and sporadically on others and is not recognizedon most (144). Despite this difference in epidemiology, mosthorses develop immunological evidence of infection (71, 131,162), although horses on farms where disease is endemic aremost likely to have high levels of antibody (151). Extrapo-lating from these epidemiological characteristics as well asfrom studies of macrophage-R. equi interactions in vitro(69), it seems that young foals can overcome infection withlow numbers of R. equi; heavy or continuous exposurepredisposes them to disease, particularly in the absence ofantibody or fully competent cell-mediated immune mecha-nisms. On horse farms, there is progressive increase ofenvironmental contamination with R. equi (133), related tothe length of use of the farm for horses (presumably areflection of concentration of horses and manure disposalpractices, summer temperatures, soil type, and whether ornot the farm is used to breed foals). Clinical isolates fromfoals are usually more virulent for mice than environmentalisolates (16).The suggestion that pneumonia in foals results from mi-

gration of R. equi-infected helminth larvae through thelungs, and therefore that helminth control may prevent thedisease (5), has not been confirmed.

INFECTIOUS DISEASE: CLINICAL ANDPATHOLOGICAL MANIFESTATIONS

R. equi is a facultative intracellular pathogen, survivinginside macrophages to cause granulomatous inflammation.With the eventual destruction of macrophages, the granulo-mas may become purulent and progress to caseous necrosis.In all species, the lung is the most frequent organ affected byR. equi, but intestinal ulceration and lymphadenitis mayfollow heavy intestinal infection (84). The organism mayinfect wounds as well as disseminate from a large focus ofinfection, for example, in the lung, to cause abscessesthroughout the body.

Horses

The foal is unique in its susceptibility to naturally occur-ring R. equi suppurative bronchopneumonia, which is char-acteristically a chronic disease associated with the develop-ment of multiple, large lung abscesses as well as abscesses inthe bronchial and mesenteric lymph nodes. Foals withchronic R. equi pneumonia may also develop both ulcerative

VOL. 4, 1991

on June 29, 2020 by guesthttp://cm

r.asm.org/

Dow

nloaded from

CLIN. MICROBIOL. REV.

colitis and granulomatous mesenteric lymphadenitis afterswallowing infected sputum. In rare instances, such intesti-nal lesions may be found in the absence of lung pathology.

Clinical manifestations in foals. The disease occurs wher-ever foals are raised (144), but in most cases the diseaseoccurs sporadically. Infection may be subclinical, beingrecognized incidentally by the presence of individual lungabscesses in animals autopsied for other reasons (144, 199),or by growth of the organism in transtracheal cultures ofapparently healthy foals (4). Foals are affected up to 6months of age; most clinical cases occur by 2 months of age(4, 96, 97, 137, 199). In one large study of 89 foals that diedof pneumonia and/or enterocolitis, duration of illness was <1week in 26%, between 1 and 3 weeks in 25%, and >3 weeksin 49%. The disease occurs in the summer months, whichusually coincides with the presence of foals of peak agesusceptibility. While not well documented, infection may bepredisposed by viral respiratory disease in foals (144).

Early signs of the usual, chronic progressive form of thedisease associated with multiple, massive lung abscesses arefever (up to 41°C), increased respiratory rate with broncho-vesicular sounds over large airways and wheezing over smallairways, cough and, sometimes, bilateral nasal dischargeand depression (4, 5, 49). As the lung abscesses develop,foals show progressive increases in respiratory rate anddepth, and movement becomes increasingly distressful. Thechronic disease may progress inexorably in untreated ani-mals until eventually they die of asphyxiation (43, 137, 199).In foals with the chronic form of the disease, severe diarrheamay develop as a result of colonic mucosal invasion by R.equi (31, 144, 199). Rarely, colitis without lung involvementmay occur (144). There are no pathognomonic signs of R.equi pneumonia in foals, although a chronic, active, nonsep-tic synovitis characterized by joint distension has beendescribed in about one-third of affected animals (159). Foalswith pneumonic disease show elevated total leukocytecounts, predominantly neutrophils. Plasma fibrinogen valuesare elevated, to levels correlating well with the degree oflung damage (49). Occasionally, the disease develops veryacutely (144), characterized by sudden onset of respiratorydisease and death within 24 to 48 h. In some cases this isassociated with a sudden and overwhelming exposure of thelung to many organisms (144).

Radiographically, in the acute stages of infection, foalswith R. equi pneumonia tend to have a prominent interstitialpattern of infiltrates which progresses to a consolidated,alveolar pattern with nodular and cavitary lesions. Particu-larly in severely ill foals, these lesions are accompanied bylymphadenopathy (49). Serial thoracic radiographs are usedto monitor the progression of the disease (4).

Ulcerative lymphangitis caused by R. equi has occasion-ally been reported to develop on foal legs (5, 48, 122, 154).This condition appears to be a wound superinfection, theorigin of which may be migrating Strongyloides westerilarvae in some cases (38, 39). Occasionally, R. equi maydisseminate from lung abscesses to intervertebral or otherjoints (such as those of the long bones) or to other body sites,including the eye, to cause localized infections (97, 137, 199).Pleurisy and peritonitis are uncommon presentations (5).

Clinical manifestations in adults. Disease due to R. equi israre in adult horses. It manifests as a sporadically occurringillness similar to that observed in foals, involving primarilylungs or colon and related lymph nodes, or rarely as woundinfection (43, 59, 142, 150, 199). Acquired combined immu-nodeficiency of unknown origin was identified as the predis-posing cause in one case of lung abscessation in an adult

FIG. 1. Lungs of foal with extensive antero-ventral pyogranulo-matous abscesses typical of the lesions of R. equi pneumonia.

horse (54). The organism has been isolated from the uterus ofinfertile mares and from aborted fetuses (5, 18), but therehave been few recent reports of such isolations (59, 199).

Pathological manifestations in foals. The most commonlesions are subacute to chronic suppurative bronchopneu-monia with extensive abscessation and an associated suppu-rative lymphadenitis (190). Lung abscesses vary from pea-sized to hen's egg or considerably larger size and commonlycontain inspissated pus (Fig. 1). Up to half of infected foalsmay also exhibit multifocal ulcerative colitis and typhlitis. R.equi behaves as a facultative intracellular pathogen, capableof persisting within phagocytes, which the organisms even-tually destroy. Experimental infection of the lung or intes-tine of foals has shown a characteristic progression of lesionsassociated with this behavior (83, 100). Early lung lesions arecharacterized by massive influx of phagocytic cells into thealveolar spaces. These cells are predominantly large macro-phages often in the form of multinucleated giant cells. R.equi are found in large numbers within the macrophages andgiant cells, as well as in the less common neutrophils, butrarely elsewhere. Interalveolar septa are intact. Grossly, thelungs of experimentally infected foals may show massiveconsolidation in the early disease stages (83). The eventualdegeneration of the macrophages coincides with the devel-opment of focal lytic lesions in, and destruction of, the lungparenchyma. Caseous necrosis is the dominant lesion inadvanced lung disease in some experimentally and naturallyinfected foals (83), but in most naturally infected foalsadvanced lesions are suppurative rather than caseous (199).

In the intestine, a pyogranulomatous process similar tothat described in the lungs begins in the Peyer's patches,which are destroyed with formation of ulcerated areas. As inthe lung, infection spreads to involve local lymph nodeswhich may become considerably enlarged (84). Fixed mac-

rophages in the body appear to destroy R. equi better thanalveolar macrophages since, despite the bacteremia that mayoccur during infection, lesions in the liver and spleen are rare

(190). Alveolar macrophages are relatively inefficient com-

pared with macrophages from other sites in providing the

22 PRESCOTT

on June 29, 2020 by guesthttp://cm

r.asm.org/

Dow

nloaded from

RHODOCOCCUS EQUI 23

TABLE 1. Predisposing factors identified in humanR. equi infections

Patients with

Predisposing No. of possiblefactor patients animal Reference(s)

source ofR. equi

AIDS 11 1 13, 51, 66, 90, 95,a146, 147,a 156,178

Hemolymphatic tumors 8 5 12, 24, 57, 60, 85,(prednisone treatment) 98, 176

Renal transplant (pred- 6 3 56,a 85,a 121, 145,nisone, azothioprine 148, 176, 181treatment)

Corticosteroid therapy 2 2 73, 95Penetrating eye injury 2 41, 76Alcohol overuse 1 91Raw carrots, ingestion 1 1 172Hepatic fistula 1 56,a 85a

a Cases reported in references 56 and 85 and those reported in references 95and 147 describe the same patients.

accessory function required for initiation of the immuneresponse (104).

HumansIn immunosuppressed patients, R. equi infections occur

mainly in the lungs. Of 32 cases reported (Table 1), 88% wereassociated with immunosuppression following the develop-ment of AIDS, treatment for hemolymphatic tumors, or theprevention of rejection following renal transplantation. Twocases of panophthalmitis followed penetrating eye injuries.In two cases (cervical lymphadenitis in a child, hepaticfistula in an adult) infection appeared to follow disseminationfrom the mouth or intestine (85, 172).Pneumonia was the main presenting problem in two-thirds

(21 of 32) of patients. Exceptions were two cases of pene-trating eye wounds (41, 73), one AIDS patient with aninflammatory mass in the pelvis (51), one AIDS patient withbloody diarrhea and cachexia (51), one patient with pleuraleffusion (91), one patient with osteomyelitis (which followeda pneumonic episode) (121), one renal transplant patient witha paraspinal abscess (85), one AIDS patient with a psoasabscess (51), one asymptomatic renal transplant patient witha lung abscess recognized on radiography (176), one AIDSpatient with an "inflammatory pseudotumor" in the lung(13), and a child with cervical lymphadenitis (172).

In the usual pneumonic presentation of infection, patientsgradually developed fever of several days to several weeksin duration, with malaise, dyspnea, and nonproductivecough; in some cases hemoptysis is described. Patients oftencomplain of chest pain. Radiographic abnormalities in theearly stages are typically infiltrative, with opaque lesionscommonly localized to one upper lung lobe. Lesions mayvary in size from 2-cm nodules to those involving a large partof the affected lobe. If untreated or treated inappropriately,lesions persist, often enlarge, and, typically in 2 to 4 weeks,develop into cavities characterized by the presence of anair-fluid line within the lesion. In a few cases, pleuraleffusions have been the prominent lesions (91, 176).The changes commonly described are also characteristic

of those associated with tubercular or fungal infectionswhich should be differentiated by using appropriate skin ormicrobiologic tests, although the impaired immune response

in AIDS patients reduces the value of immunologic tests(41). The fluorochromes used to detect mycobacterial organ-isms in clinical specimens may give positive results with R.equi, so that infection may be misdiagnosed in the earlystages of the disease (51, 90). R. equi is one of the causes ofcavitatory pulmonary disease in AIDS patients and may beregarded as a possible indicator of AIDS (66). Disseminationfrom focal lung sites to brain, skin, paraspinal tissue, andbone has occasionally been described (12, 60, 85, 121, 176).No age predisposition is apparent, but 78% of patients

were male, largely because of the greater association ofAIDS with males. A possible animal source of infection wasrecorded in 12 of 32 patients but only one of these was anAIDS patient, a farm worker with exposure to horses andother animals (178). Since the major route of R. equiinfection is by soil contaminated with manure of herbivorousanimal origin, it is not surprising that some patients werelivestock farmers (74, 95), worked with animal manure in agarden (12), or cleaned out dusty cattle, sheep, and pig pensin stockyards (60). Nevertheless, most patients have had nosuch history of exposure that might account for their infec-tion, although bird feces might be one unrecognized sourceof infection (23).

Natural Infections in Other Species

R. equi has been isolated from many species other thanhumans and horses (Table 2), but with the marked exceptionof pigs which develop submaxillary lymphadenitis, reportsof the isolation from other species are uncommon. Lesionsare generally typical of those described in horses and hu-mans. Table 2 illustrates the predominance of disease inherbivores and the tendency for the organism to be isolatedfrom lungs, granulomatous lymph nodes, abscesses, orwound infections. This spectrum of disease is similar to thatobserved in horses and in humans. Miscellaneous olderreports record the isolation of R. equi from aborted fetusesand from animals with mastitis and metritis (18, 19, 35, 138).The organism has been isolated from ulcerative lymphangitisin cattle (120), apparently following skin wounds made bythorns. In goats the organism may have a tendency to causeliver abscesses, possibly following penetration from theintestinal tract.

In cattle, as in swine, the organism may be isolated fromgranulomas in lymph nodes, usually respiratory tract nodes(105). In both species, the lesions from which R. equi havebeen isolated strikingly resemble those of tuberculosis andhave therefore generated considerable interest among veter-inarians. Although there are many reports of the isolation ofR. equi from the submaxillary lymph nodes of pigs with"tuberculous" lesions (7, 34, 43, 50, 86, 183), the organismmay also be recovered from a similar proportion of normalsubmaxillary lymph nodes in healthy pigs (34, 50, 86, 168).For example, Karlson and co-workers (86) isolated R. equifrom 22.4% of 89 lymph nodes from tuberculous swine butalso from 24% of 25 lymph nodes from healthy pigs. Theyand others were unable to reproduce submaxillary lymphad-enitis in swine by feeding pigs cultures of R. equi (34, 86).The causative role of R. equi in granulomatous lymphaden-itis in swine thus remains unproven, though possible, basedon the type of inflammatory changes and the similarity to R.equi lymphadenitis in other species. In some cases thetuberculous lesions observed can be explained by the con-current presence of various Mycobacterium species (50,168).

VOL. 4, 1991

on June 29, 2020 by guesthttp://cm

r.asm.org/

Dow

nloaded from

CLIN. MICROBIOL. REV.

TABLE 2. Species and site of R. equi isolation from animals other than horses or humans

Animal species Isolation site Frequency Selected reference(s)

Pigs Submaxillary lymphadenitis Common 34, 50, 86, 139, 171, 183Pneumonia Rare 183

Cattle Lymphadenitis (mesenteric, bronchial) Rare 105, 189Pneumonia, chronic Rare 77, 107Pyometra Rare 35Ulcerative lymphangitis Rare 120

Buffalo Pyometra Rare 138Mastitis Rare 136

Goat Pneumonia, liver and spleen abscesses Rare 25, 179Liver abscess Rare 25

Sheep Pneumonia Rare 1, 141Abortion Rare 36

Deer Lungs, abscesses Rare 23Cat Lymphadenitis, pyogranulomas, abscess Rare 47, 72, 80Dog Unspecified skin lesions Rare 124Koala Purulent rhinitis and pneumonia Rare 135Seal Lung abscess and lymphadenitis Rare 8Marmoset Pyogranulomatous bronchopneumonia Rare 158Alligator, crocodile Fulminating bacteremia Rare 81

Experimental Infections in Animals

Experimental pneumonic infection mimicking that seen innatural animal or human infection has not been consistentlyinduced in any animal other than the foal. Intranasal oraerosol administration of cultures to the pig and mouseresulted in a subacute, macrophage-rich, interstitial pneumo-nia which resolved, rather than progressing to the pulmonaryabscess stage typical of natural infection (16, 196).The normal mouse lung can progressively (but remarkably

slowly) clear a heavy inoculum of R. equi (112). Studies ofthe early events that occur in mice infected intrabronchiallywith defined bacterial doses demonstrated clearance of aproportion of bacteria within the first 24 h by nonspecific,largely neutrophil, phagocytes. In addition, a number oforganisms survive, apparently within macrophages. Suc-cessful defense against this surviving population appeared torequire specific cell-mediated immunity (17). Mice chal-lenged intranasally with R. equi following cyclophosphamideadministration often developed fatal infection in contrast tountreated mice, suggesting the importance of functionalcellular immunity in protection against infection (112).

Extensive studies of the susceptibility of mice to experi-mental Nocardia asteroides infection are relevant to under-standing immunity to R. equi, a closely related organism.The susceptibility and tissue response of mice to experimen-tal N. asteroides infection depend on the strain of mouseused, the most susceptible mice having genetic defectsaffecting T-lymphocyte function (9). For example, athymicnude (Nu/Nu) mice were protected from the lethal effects ofintranasally administered N. asteroides organisms by adop-tive transfer of T lymphocytes from preimmunized hetero-zygous mice (9). Protection appears to be related in part tomacrophage activation but also to a direct cytotoxic effect ofT lymphocytes against the organisms (9). Alveolar macro-phages from immunized normal (BALB/c) mice were leasteffective in killing N. asteroides when compared with mac-rophages isolated from liver, the peritoneal cavity, and thespleen (9). In studies of mice infected by R. equi by theintravenous, intranasal, or intratracheal route normal micewithout genetically defined defects in immune function wereused (16, 116, 165).

Intratracheal inoculation of pigs or guinea pigs resulted in

suppurative pulmonary lesions similar to those of the naturaldisease in foals, but in guinea pigs abscesses did not developand lesions resolved within a relatively short time (79, 171,196). The experimental infection in swine has been less welldescribed (171).

VIRULENCE AND PATHOGENIC MECHANISMSSusceptibility of the young foal to R. equi pneumonia

remains largely unexplained but must relate in part to acombination of factors, including heavy challenge by therespiratory route coinciding with declining maternally de-rived antibody and absence of fully competent cellularimmune mechanisms. Variation in virulence of R. equiisolates has been identified with experimentally infectedmice and foals. The type strain isolated by Magnusson(ATCC 6939, NCTC 1621) does not cause pneumonia in foals(100, 163) and is avirulent for mice (116, 165). Lethality ofstrains for mice infected intravenously was related to theability of organisms to resist both clearance from the liverand spleen and phagocytosis and intracellular killing bymouse macrophages (116, 165). Based on data from experi-mental infections, clinical isolates are more pathogenic thanenvironmental isolates (16, 52). The basis of the variation invirulence has not been determined. Differences in mousevirulence in Nocardia spp. have been attributed to certainmycolic acids (10). Two distinct mycolic acid fractionationpatterns have been described for a small number of R. equiisolates, including the avirulent type strain (22).Among possible candidates as virulence factors are cap-

sular polysaccharide, which might inhibit phagocytosis ofthe organism, and cholesterol oxidase and phospholipase Cexoenzymes ("equi factors"). Cholesterol oxidase is aprominent product of R. equi (53, 93). The combined actionof the equi factors may confer membranolytic activity on R.equi (11, 93), but their role in virulence needs to be defined.Cholesterol oxidase is produced in different amounts bydifferent strains, but whether this correlates with virulence isunknown (2). Mycolic acid-containing glycolipids of R. equiare likely to promote granuloma formation as do those ofother rhodococci, but their role has not been investigated(149).The ability of R. equi to persist in and eventually to

24 PRESCOTT

on June 29, 2020 by guesthttp://cm

r.asm.org/

Dow

nloaded from

RHODOCOCCUS EQUI 25

destroy macrophages is the basis of its pathogenicity. Thepathological change elicited by R. equi is similar to that ofsome other parasites within the group Corynebacterium-Mycobacterium-Nocardia, with which it shares lipid-richcell wall components (61, 106). Foal alveolar macrophagescultured in vitro rapidly ingested opsonized R. equi, but 75%of ingested organisms remained viable after 4 h of incubation(198). Electron microscopic examination of the behavior ofR. equi in cultured foal and adult horse macrophages showedthat organisms evaded killing by preventing phagosome-lysosome fusion, thus multiplying in and eventually killingthe phagocytes (69, 197). In addition, nonspecific degranu-lation of lysosomes in R. equi-infected macrophages hasbeen suggested to be the major survival and pathogenicmechanism ofR. equi (69). Such degranulation in vivo wouldcontribute to the tissue destruction and neutrophil influx thatis characteristic of R. equi lung lesions. Opsonization withspecific antibody enhanced ingestion, phagosome-lysosomefusion, and killing of R. equi by cultured macrophages (69).

In vitro, studies of R. equi killing by isolated horseneutrophils have shown excellent bactericidal activity ofopsonized organisms and no age-related changes in bacteri-cidal activity (70, 99, 191). Foal neutrophils were mostefficient in ingesting and killing cells when opsonized withspecific antibody (69). One study, however, identified amoderate proportion of neonatal foals whose neutrophilbactericidal activity increased with age (103). A heat-stablesurface component of R. equi, possibly a capsular polysac-charide, has been identified and shown to inhibit the oxygen-dependent cytotoxic mechanisms of adult horse neutrophilsagainst Staphylococcus aureus (44). Although this inhibitiondoes not appear to be sufficient to impair overall bactericidalactivity of equine neutrophils against opsonized R. equi invitro (70, 99, 191), this impairment may be significant invivo, perhaps in the absence of specific antibody. In contrastto some other Rhodococcus species (67), R. equi does notproduce the siderophore mycobactin under iron-limitinggrowth conditions.

IMMUNITY TO INFECTION

With the exception of the young foal, disease in otherspecies is rare unless, as in most human patients, the host isimmunosuppressed. The basis of immunity in horses, par-ticularly that relating to cell-mediated immune mechanisms,is largely unknown. Current understanding of immunity ofhorses to R. equi has recently been reviewed (187).

Humoral Immunity in Horses

Early studies of naturally or experimentally infected foalsshowed a generally inconsistent or low level of antibodydetectable by agglutination, passive hemagglutination, com-plement fixation, or precipitation (20, 26, 97, 117, 129, 137,180). The conclusion from these early studies was that thereis a negligible humoral response to infection and that immu-nity is primarily based on cell-mediated mechanisms (20,129, 131). The later development of more sensitive detectionmethods based on enzyme-linked immunosorbent assay(ELISA) with various antigens, indirect immunofluores-cence, or equi factor neutralization, however, showed thatantibody is widespread in the horse population (46, 71, 151,152, 162, 180). The intestinal tract is probably the majorsource of antigenic stimulation (30, 131, 160, 163). ELISAantibody (immunoglobulin A [IgA], IgG, IgM) titers rose

following oral inoculation with the avirulent ATCC 6939 or aclinical isolate (163, 164).

In foals, maternal antibody from colostrum declines to itslowest levels by about 8 weeks of age, after which timeantibody is actively produced by the foal. Foals with lowlevels of maternal antibody detected by ELISA are particu-larly susceptible to naturally induced R. equi pneumonia (4,71), emphasizing the importance of antibody in enhanceduptake and killing by macrophages and neutrophils (69, 70,103, 170). The clinical severity of R. equi pneumonia may berelated to the amount of specific antibody present, sincefoals with higher amounts show milder disease (71).

Recently, the immunoprophylactic capacity of specificimmune plasma in foals with experimentally induced R. equipneumonia has been elegantly established (101), supportingearlier suggestions that plasma or serum might be used totreat R. equi infections (15, 68, 97). The clinical course of thedisease in passively immunized foals was dramatically lesssevere than that in foals treated with plasma from nonimmu-nized donors (101). Plasma donor horses for this experimentwere immunized with live R. equi. Immunized foals hadsignificantly higher ELISA values than control foals admin-istered plasma from nonimmunized horses. Nevertheless, inthis passive immunization study one foal with very lowELISA values did not develop clinical disease, whereas asecond with high ELISA values died (101). Direct evidencefor a protective function capacity of R. equi antibody de-tected by ELISA is lacking. Indeed, levels of ELISA anti-body correlate poorly with measures of opsonizing activitysuch as bactericidal and chemiluminescence assays (102).Part of the protective effect of immune plasma may thereforeresult from nonspecific factors in plasma, including lym-phokines and interferons (101). Trials are now being con-ducted in the field to confirm the value of passive immuni-zation in the control of R. equi pneumonia in foals.However, further work is required to define the basis of theprotection observed experimentally.The nature of the antigens that give rise to the opsonizing

and possible protective effect of specific antiserum to R. equiis not established. Antigens used in ELISA procedures havebeen largely derived from the bacterial surface or fromculture supernatants and therefore would likely include bothcapsular polysaccharide and the exoenzyme (equi factors)antigens in major amounts (46, 71, 162, 163). Althoughantibodies to equi factors might be involved in protectiveimmunity, it is more likely that protective antigens arebacterial surface components. Immunoblotting procedureswith sera from healthy as well as pneumonic foals, however,showed greater antibody responses to supernatant productsthan to whole cells of R. equi (29).

Celular Immunity in Horses

Cell-mediated mechanisms are thought to be the majormeans of immunity to R. equi and other intracellular patho-gens such as Mycobacterium tuberculosis (87, 109). Protec-tion against such intracellular pathogens depends on subtleand complex coordination of the cell-mediated immuneresponse, which is poorly understood. A primary event inintracellular killing is activation of macrophages by gammainterferon, a lymphokine secreted primarily by CD4+ lym-phocytes (87). Some strains of mycobacteria, however, areresistant to killing by macrophages stimulated by gammainterferon, suggesting that other cells may be involved, forexample, CD8+ cytotoxic cells or macrophages activated byother cytokines such as tumor necrosis factor (87). Thus,

VOL. 4, 1991

on June 29, 2020 by guesthttp://cm

r.asm.org/

Dow

nloaded from

CLIN. MICROBIOL. REV.

while gamma interferon plays a central role in protectinganimals against intracellular pathogens by activating macro-phages, an apparently coordinated but poorly understoodinterplay with other T cells is also needed (87).The molecular mechanisms that culminate in the genera-

tion of antigen-stimulated gamma interferon are complex andalso poorly understood. Theoretically, any interference withone or more of the key steps leading to macrophage activa-tion (accessory cell function, interleukin 1 and 2 secretion,interleukin 2 receptor suppression) could lead to a failure ofgamma interferon production and consequently impairedimmunity to an intracellular pathogen. There is no evidencethat R. equi specifically affects any of these steps in foals. Inhuman patients with certain severe manifestations of leish-maniasis, leprosy, and tuberculosis, in vitro defects in anti-gen-induced gamma interferon production are limited tofailure to respond to the infecting pathogen alone (109). Thisrestricted and reversible antigen-specific defect correlateswith a lack of generalized susceptibility to opportunistpathogens in such patients. There is no evidence that such anevent occurs in foals, though it seems likely that componentsof R. equi may induce immunosuppression of some type, as,for example, does the mycobacterial cell wall constituent,arabinomannan (65). Support for immunosuppressive activ-ity of R. equi components comes from a study in whichinjection of mice with a water-soluble extract of R. equisuppressed production of reaginic antibodies against a hap-ten carrier (55).

Less is known about the role and importance of cell-mediated immunity to R. equi than humoral immunity.Because of the intracellular nature of the organism, cell-mediated immunity is assumed to be of major importance inR. equi infections (45, 131). Delayed-type hypersensitivityskin reactions have been demonstrated in apparently healthyor experimentally infected horses (45, 180, 182) and indicatewidespread exposure to R. equi. Such hypersensitivity de-velops with age but does not reflect the extent of environ-mental contamination (180).The best descriptions of the important interactions of

cell-mediated immunity with humoral immunity come fromin vitro studies of macrophage killing of R. equi (69).Alveolar macrophages from foals that had been experimen-tally exposed to R. equi phagocytized and killed both nonop-sonized and opsonized R. equi more efficiently than alveolarmacrophages from nonexposed control foals. The activity ofmacrophages from exposed foals was similar to that ofmacrophages from adult horses, although the rate of killingof opsonized R. equi was slightly less in foals. In all alveolarmacrophages tested, opsonization of R. equi markedly in-creased phagosome-lysosome fusion compared with nonop-sonized bacteria. Supernatants from foal lymphocyte cul-tures stimulated with R. equi antigens significantly enhancedkilling of R. equi by alveolar macrophages isolated from bothexposed and nonexposed foals. The combination of humoral(opsonized bacteria) and cell-mediated (T-cell-activatedmacrophages) immunity resulted in complete killing by in-cubated macrophages in vitro (69). These studies underlinethe importance of both humoral and cellular immunity inprotection of horses against R. equi pneumonia and reinforcethe conclusion from reports of R. equi pneumonia in patientswith AIDS that cell-mediated immunity is of major impor-tance in resistance to infection. Understanding of cell-mediated aspects of immunity of foals to R. equi is clearlyrudimentary.Although not well documented, anecdotal evidence sug-

gests that individual mares may consistently produce foals

that are particularly susceptible to infection (180, 182). Suchsusceptibility might be explained by low colostral antibodylevels, the functional immaturity of neutrophils in certainfoals (99), or a genetically controlled predisposition to infec-tion. Reports of horse breed predisposition to infection areconflicting, with suggestions that Arabians (49) or Standard-breds (199) may be at increased risk. Resistance in mice tointracellular pathogens is genetically determined, occurringat the level of the macrophage independently of T-cellfunction (21). The single gene or gene complex determiningresistance to intracellular pathogens such as M. lepraemu-rium, Salmonella typhimurium, Leishmania donovani, andM. bovis (BCG) is the Bcg gene (or complex) on chromo-some 1 (21). Other species such as cattle appear to haveinnate macrophage-based resistance to other facultative in-tracellular pathogens (134).

Immunity to Infection in Humans

Impairment of cell-mediated immune mechanisms pre-disposes human patients to R. equi infections. Immuno-suppressive drugs (cyclosporin A, corticosteroids, and anti-metabolites) suppress mitogen-induced gamma interferonproduction by CD4+ cells in many renal and bone marrowtransplant recipients and in patients treated for lymphaticleukemias (Table 1). Therefore, these patients are suscepti-ble to an array of opportunistic infections caused by intra-cellular pathogens (109). In adult AIDS patients, the char-acteristic destruction of CD4+ cells by the virus impairscell-mediated immune mechanisms and predisposes to infec-tion with a variety of intracellular organisms that require anintact T-cell-macrophage system (40). These infections mayrepresent reactivated latent infections (toxoplasmosis, tu-berculosis, herpes simplex virus, or cytomegalovirus infec-tions), be caused by agents such as Pneumocystis cariniiwhich do not cause disease in healthy hosts, or be acquiredopportunists such as M. avium-M. intracellulare complex,Salmonella enteritidis, Cryptococcus neoformans, or Cryp-tosporidium parvum (14), or, rarely, R. equi. M. avium-M.intracellulare complex infection is a late occurrence in AIDSpatients causing widely disseminated systemic disease and isobserved in 25 to 30% of adult AIDS patients at the time ofdeath (27, 194). The relative infrequency of infections withother environmental opportunistic intracellular pathogenssuch as N. asteroides or Listeria monocytogenes (or even R.equi) compared with the frequency of M. avium-M. intracel-lulare complex infection (94) suggests that the AIDS virushas a special preference for T cells involved in immunityagainst mycobacterial cells (37).

Immunotherapy in Horses

Field studies by Magnusson in 1923 and 1924 failed toshow the value of vaccination with "killed broth cultures" ofR. equi (97). Vaccination of foals with a Formalin-killed R.equi bacterin did not protect them against a heavy intratra-cheal challenge (129). The excellent protection against aero-sol challenge which followed oral immunization of youngfoals (immunized on four occasions over 5 weeks with anonattenuated clinical isolate of R. equi [30]) supports theidea that intestinal exposure of foals to R. equi results innatural vaccination. The basis of this protection was notestablished, although others have shown that specific IgGlevels rise in foals after oral administration of R. equi (162,166).

Effective immunization against intracellular bacterial

26 PRESCOTT

on June 29, 2020 by guesthttp://cm

r.asm.org/

Dow

nloaded from

RHODOCOCCUS EQUI 27

pathogens such as mycobacteria requires the use of liveorganisms rather than killed vaccines to promote cellularimmunity. The reasons are unclear but may relate to arequirement for antigen persistence, the activation of dif-ferent pathways within phagocytic cells, or, least likely, thesecretion of specific protective antigens by live organisms(32). Although the relative importance of each arm (cellmediated, humoral) of the immune system in resistance to R.equi is unknown, it seems likely that vaccination with liveorganisms, which promotes cell-mediated immunity, will berequired to protect foals fully against R. equi pneumonia.The value of both opsonizing antibody in promoting uptakeand killing by macrophages and passive immunization inprotecting against pneumonia has been established and sup-ports continued attempts to define the antigens required topromote protective antibodies.Whereas the role of a protective antigen is well established

in antibody-mediated immunity, it is highly speculative incell-mediated immunity, in which protection may involve acomplex balance requiring recognition of many antigens aswell as a variety of T-cell subsets (92, 193). Indeed, despiteconsiderable efforts, immunodominant determinants of M.tuberculosis and M. leprae, including those which induceexpansion of T cells with selective function, have not yetbeen isolated and characterized (92, 143, 193).

Active immunization is not an option in human patientswith AIDS, and the value of passive immunization withhyperimmune anti-R. equi serum in treating foals has notbeen fully determined. Another approach to immunotherapymight involve lymphokine therapy. Trials in humans indicatethat gamma interferon activates human macrophages in vivoand induces other immunoregulatory activities, includingnatural killer cell number and cytotoxicity, lymphocyteproliferation and monocyte Fc receptor expression (109).These trials do not suggest that gamma interferon activatesmacrophages against all intracellular pathogens, but dosupport the value of further work to define those diseases inwhich such lymphokine therapy might be effective (109). R.equi pneumonia of foals or other experimental animals mighttherefore be a useful model to determine the value ofimmunotherapy in AIDS patients with R. equi infections.

LABORATORY DIAGNOSIS

Isolation and Identification of the OrganismIsolation and colonial morphology. R. equi grows readily

when incubated aerobically at 37°C on nonselective mediaroutinely used in clinical microbiology laboratories. At 24 hof incubation, colonies are 1 to 2 mm in diameter and are notdistinctive. By 48 h of incubation on nonselective mediasuch as Trypticase soy blood agar, they have developedtheir characteristic appearance (Fig. 2): irregularly round,smooth, semitransparent, glistening, coalescing, mucoid,teardrop colonies with entire edges. The colonies vary insize from 2 to 4 mm, although coalesced colonies may appearlarger. The organism does not grow on most types ofMacConkey agar. Rare isolates of R. equi may grow poorlyat 37°C (108).Colony variation is present in fresh cultures. The classical

viscous-mucoid coalescing colony is usually the predomi-nant type present, but less mucoid forms may also be seen.A small proportion of small, 1 mm or less, nonmucoidcolonies are also present (Fig. 2). In addition to the colonyvariation within a strain, four stable colony types of R. equihave been described (classical mucoid, less mucoid, disso-

FIG. 2. Culture ofR. equi on Trypticase soy agar with 5% bovineblood after 48 h at 37°C. Photograph illustrates the typical mucoidteardrop appearance of the organism at that time and shows thetypical colonial variation.

ciating, and small nonmucoid) (111). Cultures may have aslightly earthy smell.Pigment production is rarely marked in cultures <4 days

old (6) and may surprise those expecting all rhodococci(red-pigmented cocci) to show obviously pink or red colonialpigmentation immediately on isolation. After 4 to 7 days onnonselective medium, such as Trypticase soy blood agar,colonies may develop a delicate shade of salmon pink,although they may be nonpigmented or appear slightlyyellow (6). Perhaps the best description of the typical colonypigment on blood agar is as light fawn colored. Cultures kepton slopes for prolonged periods without subculture com-monly become rough, dry, and orange-red but revert toclassical colonies on subculture (86, 181). The macroscopiccolony morphology of the type culture, ATCC 6939 (96), issmaller than usual, smooth, and dry (rather than mucoid)and appears more pigmented than typical R. equi.

Microscopic morphology and staining properties. R. equi isa gram-positive pleomorphic coccobacillus, varying fromdistinctly coccoid to bacillary depending on growth condi-tions. It is usually coccoid, both on solid media and inpurulent material from patients, but in liquid media, partic-ularly in young cultures, it forms long rods or short filamentswhich may show rudimentary branching.The occasional reports that R. equi is acid fast in the

Ziehl-Neelsen stain appear to depend on staining technique,the older age of the cultures, and the growth medium (6, 82,177). Most report a failure to demonstrate acid fastness ofR.equi. Metachromatic granules have occasionally been de-scribed (86, 96, 97) but, like acid-fast staining, are aninconsistent feature of the organism that are medium andstain dependent.

Identifying biochemical characteristics. Characteristics thatcan be used in routine clinical microbiology laboratories toidentify R. equi are shown in Table 3.The organism is generally biochemically unreactive. It

fails to oxidize or ferment any carbohydrates or alcohols

VOL. 4, 1991

on June 29, 2020 by guesthttp://cm

r.asm.org/

Dow

nloaded from

CLIN. MICROBIOL. REV.

TABLE 3. Biochemical characteristics for identifying R. equi

Characteristic Reaction % positive Reference(s)

Catalase + 100Cytochrome c oxidase - 1-5 128Carbohydrate fermentation - 100Alcohol fermentation - 100Gelatin hydrolysis - 100Indole - 100H2S Variable 32-62 111, 128Urease + 95 111, 128Hippurate hydrolysis - 1 128Esculin hydrolysis - 4 128Nitrate reduction + 88 111equi factors + 100 53, 128Lipase + 100 114Phosphatase + 100 114

tested and is nonproteolytic. No acid metabolic productsfrom glucose were detected by gas-liquid chromatography(140). It is catalase positive, almost invariably urease posi-tive, and oxidase negative. R. equi produces equi factorswhich interact with the phospholipase D of Corynebacteriumpseudotuberculosis, the beta-toxin of S. aureus, and anuncharacterized partial hemolysin of L. monocytogenes togive an area of complete hemolysis with sheep or cattleerythrocytes (Fig. 3) (53, 128). While this characteristic isnot unique for R. equi, it is distinctive for this organism andshould always be used in identification since no equi factor-negative isolates of R. equi have been described. R. equiproduces lipase and phosphatase, but not DNase, elastase,lecithinase, or protease (114). The API ZYM (AnalytabProducts Inc., Plainview, N.Y.) gave distinctive enzymeprofiles of value in identifying R. equi (113). All isolates

FIG. 3. Demonstration of the production of equi factors, whichis a consistent and useful identifying characteristic of R. equi. R.equi has been streaked vertically on the left- and right-hand sides ofthe plates. The upper horizontal streak culture is Corynebacteriumpseudotuberculosis and the lower is a beta-toxin-producing S.aureus. Left-hand side is a 48-h growth of the isolates; right-handside is a 24-h growth.

tested were positive for leucine arylamidase, acid phos-phatase, and phosphamidase; >90% were positive for valineamylamidase, esterase lipase, and alpha-glucosidase.Other identifying characteristics. The organism grows over

a wide range of temperature, from 10 to 40°C. Althoughdisputed, the optimal temperature of growth appears to be30°C, but growth rate is only marginally less at 37°C (78).Cultures grow as well at room temperature as at 37°C (97). R.equi is an obligate aerobe with simple growth requirements(61). It can utilize carbon from a wide variety of sole carbonsources (63), including acetic, pyruvic, butyric, and propi-onic acids (190) and nitrogen from ammonium sulfate orpotassium nitrate (123). The organism is nonmotile andnonflagellated and may produce small numbers of pili (192).Organisms grow readily in nonselective broth media,

generally producing moderate turbidity, sometimes with aslight salmon-pink sediment after 48 h (86). Some strainsmay produce a thin pellicle which is readily disrupted (6,174). Antimicrobial drug susceptibility, discussed below, isconsistent and might be a useful adjunct in identification.The electrophoresis pattern of whole-cell preparations inpolyacrylamide gels has been described before (28). Furthercharacteristics of R. equi, which cannot be readily deter-mined in a routine clinical microbiology laboratory but areused in taxonomic classification, have been discussed previ-ously (see section, Taxonomic Status).

Capsular serotypes. R. equi possesses a distinct, antigen-ically variable, lamellar polysaccharide capsule (185), whichhas provided the basis of a capsular serotyping system (118,124). At least 27 different capsular serotypes have beenidentified (118), of which Prescott capsular serotype 1 (or itsequivalent in other typing schemes) is the most commonworldwide (88, 110, 118, 124). No relationship betweenserotype and virulence is apparent.

Diagnostic Procedures

Isolation of R. equi from the infection site is the bestmethod of diagnosing infection. In humans with pneumonicdisease, sputum is often but not always a useful specimen(60, 98). Diagnosis is more reliably made by bronchialbrushings, percutaneous thoracic aspiration, or, more dras-tically, open lung biopsy during lobectomy (24, 66, 148). Onoccasion, bronchial brushings in human patients have failedto yield R. equi (148). This failure supports the findings ofMartens et al. that culture of bronchial secretions obtainedby transtracheal aspiration from experimentally infectedfoals does not consistently yield R. equi (100). They attrib-uted this inconsistency to either periodic shedding intoairways or the inability of organisms inside phagocytes to beisolated by standard culture techniques. Although mostculture-positive foals with R. equi pneumonia yield purecultures of organisms from lung aspirates, up to one-thirdmay also yield a variety of opportunist pathogens (49). Thus,R. equi may be present in mixed culture and may be ignoredas "diphtheroids" (121), in part because it does not show itsdistinctive colony size and mucoid appearance until after 48h of incubation. Blood cultures from febrile human patientswith pneumonic disease have yielded R. equi (sometimesrepeatedly) in about one-third of cases reported (12, 51).A number of workers have investigated alternative, immu-

nological approaches to diagnosis in foals but none can berecommended as replacing cultural approaches. In immuno-suppressed human patients, including those with AIDS,immunological tests cannot be depended on for diagnosis(41). In foals aged 2 months or less, a lymphocyte stimula-

28 PRESCOTT

on June 29, 2020 by guesthttp://cm

r.asm.org/

Dow

nloaded from

RHODOCOCCUS EQUI 29

tion assay with R. equi antigens and peripheral blood lym-phocytes clearly distinguished animals with R. equi pneumo-nia from healthy foals (131). The test was not useful inanimals older than 2 months because the response of lym-phocytes from healthy foals or adult horses sometimesexceeded that of sick foals.A synergistic hemolysis-inhibition as ay for antibody to

equi factors distinguished foals with naturally occurring R.equi pneumonia from healthy foals, although it failed toidentify foals in the early stages of experimentally induceddisease (127). An agar gel immunoprecipitation test forantibody to equi factors was also shown to identify foals withR. equi pneumonia, although a small proportion of appar-ently healthy foals as also positive in the test (117). Usingmore sensitive synergistic-hemolysis inhibition and immuno-precipitation assays than described previously, Skalka dem-onstrated the widespread nature of subclinical infection orexposure to R. equi in foals (151, 152).

In general, ELISA has been used to distinguish infectedfrom healthy foals (162). In one study, a Tween 20 extractfrom the type strain, ATCC 6939, gave the broadest cross-reactivity of the serotypes tested (162). The mean values ofanti-R. equi IgG antibody in foals known or suspected to beill with R. equi pneumonia were strikingly higher than innormal foals in the limited number of animals tested, butthere were no differences in mean values of anti-R. equi IgMor IgA antibodies (162, 163). This ELISA was more sensitivethan agar gel diffusion or indirect haemagglutination tests(162). In two unrelated studies, using different antigens, nocorrelation was found between ELISA antibody levels andclinical disease (46, 71). The antigens used in ELISA are

important in determining their value for serological diagnosis(162). Supernatants of autoclaved cells or of homogenizedcells were relatively poor antigens compared with Tween20-extracted or sonicated materials (162).Takai et al. (161) quantified fecal R. equi to determine

which foals would develop fever, cough, and diarrhea. Usingselective medium, they found that as foals became ill fecal R.equi numbers rose from base levels of about 104 per g toi107 (161).

TREATMENT AND CONTROL

Antimicrobial Susceptibility

The usual antimicrobial drug susceptibility of R. equi isshown in Table 4. The organism is particularly susceptible toerythromycin and clindamycin; the aminoglycosides amika-cin, gentamicin, neomycin, and tobramycin; rifampin; andvancomycin. It is only moderately susceptible to penicillinG, ampicillin, and tetracyclines and is usually moderatelysusceptible or resistant to first- and second-generation ceph-alosporins.

Successful treatment of R. equi infections depends on theuse of lipophilic antimicrobial drugs that can penetrate themacrophages or neutrophils in which the organisms survive(74, 75, 132). The standard treatment in pneumonic foals isoral administration of a combination of erythromycin esto-late and rifampin (75). Not only are these drugs highlyeffective in vitro, they also penetrate macrophages well andhave been shown to have additive and often synergisticactivity in vitro (130). Use of this combination has consid-erably reduced foal mortality due to R. equi infection (74,159). These drugs have also been used alone or in combina-tion, usually with good clinical response, in many humaninfections (73, 85, 90, 121, 146, 147). Although the combina-

TABLE 4. Antimicrobial drug susceptibility of R. equia

MIC (,ug/ml)Drug

50% 90% Range reported

Amikacin '1 2 -14Ampicillin 4 4 0.12-16Cephalothin 32 >64 1->64Cefoxitin 12.5 12.5 12.5Chloramphenicol 16 16 1.6-50Ciprofloxacin 1.0Clindamycin 2 2 2-4Doxycycline 3.1 3.1 1.6-12.5Erythromycin -0.25 -0.25 s0.25-1Gentamicin s0.25 0.5 s0.25-8Lincomycin 1.6 3.2 1.6-3.2Kanamycin 4 8 2-16Methicillin >16 >16 0.25->100Neomycin s0.2 <0.2Penicillin 2 >4 0.12-64Rifampin 0.06 0.06 0.008-0.25Spectinomycin 6.3 12.5 6.3-12.5Streptomycin 3.1 3.1 <0.2-12.5Tetracycline 2 4 0.4-25Tobramycin 0.5 1 0.5-2Trimethoprim- 8/152 32/608 0.5/9.5 ->32/608

sulfamethoxazoleVancomycin 0.5 <.25-<1

a Table is a composite of data from references 12, 40, 49, 60, 89-91, 98, 121,125, 130, 146, 176, and 186, with most data derived from references 89, 125,and 186.

tion of gentamicin with penicillin gave additive and some-times synergistic activity in vitro (130), in one study, use ofthis combination to treat foals with R. equi pneumonia wasinvariably associated with their death (74), possibly becausepenetration of macrophages or neutrophils by these drugs isso poor. Others, however, have used this combination infoals with greater success (155). Successful treatment of R.equi pneumonia in foals depends not only on the choice oflipophilic antimicrobial drugs but also on prolonged treat-ment, usually until the animals appear radiographically andhematologically normal (74, 159). Return of serum fibrinogenconcentrations to normal has been used as an indicator ofsuccessful treatment (74). Duration of therapy is often from4 to 9 weeks and usually results in complete resolution oflesions (74, 75). This time is consistent with the prolongedtreatment that may be required in humans for effectivecontrol of pneumonic R. equi infection (12, 24). Two humanswere treated for R. equi endophthalmitis following a trau-matic injury. One received gentamicin alone, and the otherreceived gentamicin and cefazolin; both had a rapid, un-eventful recovery (40, 76).Although erythromycin-rifampin combination is the treat-

ment of choice for foals, other drugs have been used withapparent success. For example, clinical success has beenreported with the use of large doses of orally administeredtrimethoprim-sulfamethoxazole in foals with early pneumo-nia or in those with extensive muscle soreness due toinjection of other drugs (132). However, resistance to thisdrug combination has been described in vitro. Clindamycinor lincomycin is not administered to horses because of theirtendency to cause fatal colitis. Since these drugs are lipo-philic, they might have a place in the treatment of humanswith R. equi infections, but one report does not support thissuggestion (178). R. equi is usually susceptible to vancomy-cin, but this drug does not seem to be notably useful

VOL. 4, 1991

on June 29, 2020 by guesthttp://cm

r.asm.org/

Dow

nloaded from

CLIN. MICROBIOL. REV.

clinically (51, 146). Penicillin G administered intravenouslyin high doses has been used successfully to treat foals withR. equi pneumonia (58) but cannot be recommended becauseof the hydrophilic nature of penicillin, the organism's rela-tive resistance, and the difficulty of administering drug bythis route. Some human infections have not responded tobenzyl penicillin (24), although response to ampicillin oramoxicillin has been described (98, 148). Resistance tobeta-lactam antibiotics has also developed during therapy inhuman patients (146, 178). For these reasons, beta-lactamdrugs should probably be avoided or administered only withother drugs. In vitro, the combination of penicillin anderythromycin showed synergistic activity (130). The combi-nation of gentamicin with erythromycin or rifampin in vitrogave antagonistic activity compared with either drug alone(130) and should probably be avoided in treatment.

Besides the use of antimicrobial drugs, the approach usedin treatment of human infection involves drainage of suppu-rative lesions, surgical resection of granulomatous tissue,and control of predisposing factors such as decreasing thedose of concurrent immunosuppressive drugs or control ofunderlying malignancies. More information is needed on thevalue of gamma interferon (see subsection, "Immunother-apy in Horses") because its use has been contraindicated inhuman infections involving certain intracellular parasites.

CONCLUSIONS

Much has been learned in the last decade about R. equiinfections in foals that can be applied to the diagnosis andtreatment of infection in humans. Important questions aboutthe disease in foals remain to be resolved to improve controlof infection in both horses and humans. The development ofa suitable mouse model for infection, for example, usingmice with genetically defined immunological defects, wouldreduce the expense and other difficulties of using foals as anexperimental animal. Such a mouse model could be used todefine the antigen(s) of importance in humoral immunity andthe value of antibody to them in the prevention and treat-ment of infection, the antigens of importance in and the basisof cellular immunity to infection, the value of gamma inter-feron or other lymphokines in the treatment of infection, andthe optimal antimicrobial treatment regimen. Also, such amodel could be used to compare the virulence of differentstrains, including the nonpathogenic ATCC 6939, and torelate virulence differences to cell wall components. Thisinformation might be useful in the development of effectivevaccines. A model is needed to screen candidate live vac-cines for potential use in foals. Availability of a suitablemodel would also allow investigation of the role of T-cellsubsets in the pathogenesis of infection.

ACKNOWLEDGMENTS

Support for work in my laboratory in this area of research hascome from the Natural Sciences and Engineering Research Councilof Canada, the Ontario Ministry of Agriculture and Food, theOntario Racing Commission, and the Canadian Veterinary ResearchTrust Fund.

REFERENCES1. Addo, P. B., and S. M. Dennis. 1977. Corynebacteria associ-

ated with diseases of cattle, sheep and goats in northernNigeria. Br. Vet. J. 133:334-339.

2. Aihara, H., K. Watanabe, and R. Nakamura. 1986. Character-ization of production of cholesterol oxidases in three Rhodo-coccus strains. J. Appl. Bacteriol. 61:269-274.

3. Apajalahti, J. H. A., P. Karpanoja, and M. S. Salkinoja-Salonen. 1986. Rhodococcus chlorophenolicus sp. nov., achlorophenol-mineralizing actinomycete. Int. J. Syst. Bacte-riol. 36:246-251.

4. Ardans, A. A., S. K. Hietala, M. S. Spensley, and A. Sansome.1986. Studies of naturally occurring and experimental Rhodo-coccus equi (Corynebacterium equi) pneumonia in foals. Proc.Am. Assoc. Equine Pract. 32:129-144.

5. Bain, A. M. 1963. Corynebacterium equi infection in theequine. Aust. Vet. J. 39:116-121.

6. Barton, M. D., and K. L. Hughes. 1980. Corynebacterium equi:a review. Vet. Bull. 50:65-80.

7. Barton, M. D., and K. L. Hughes. 1984. Ecology of Rhodococ-cus equi. Vet. Microbiol. 9:65-76.

8. Bauwens, L., E. Van Dyck, W. De Meurichy, and P. Piot. 1987.Corynebacterium equi pneumonia in three Baikal seals (Pusasibirica). Aquat. Mammals 13:17-22.

9. Beaman, B. L., and C. Black. 1985. Interaction of Nocardiaasteroides in BALB/c mice: modulation of macrophage func-tion, enzyme activity and induction of immunologically spe-cific T-cell bactericidal activity. Curr. Top. Microbiol. Immu-nol. 122:138-147.

10. Beaman, B. L., and S. E. Moring. 1988. Relationship amongcell wall composition, stage of growth, and virulence of No-cardia asteroides GUH-2. Infect. Immun. 56:557-563.

11. Bernheimer, A. W., R. Linder, and L. S. Avigad. 1980. Step-wise degradation of membrane sphingomyelin by corynebac-terial phospholipases. Infect. Immun. 29:123-131.

12. Berg, R., H. Chmel, J. Mayo, and D. Armstrong. 1977. Cory-nebacterium equi infection complicating neoplastic disease.Am. J. Clin. Pathol. 68:73-77.

13. Bishopric, G. A., M. F. d'Agay, B. Schlemmer, E. Sarfati, andE. Brocheriou. 1988. Pulmonary pseudotumour due to Coryne-bacterium equi in a patient with acquired immunodeficiencysyndrome. Thorax 43:486-487.

14. Blaser, M. J., and D. L. Cohn. 1986. Opportunistic infections inpatients with AIDS: clues to the epidemiology ofAIDS and therelative virulence of pathogens. Rev. Infect. Dis. 8:21-30.

15. Boulay, P., and G. Bouley. 1958. Etude bacteriologique de deuxsouches normandes de Corynebacterium equi (C. magnus-soni). Rec. Med. Vet. 84:723-730.

16. Bowles, P. M., J. B. Woolcock, and M. D. Mutimer. 1987.Experimental infection of mice with Rhodococcus equi: differ-ences in virulence between mice. Vet. Microbiol. 14:259-268.

17. Bowles, P. M., J. B. Woolcock, and M. D. Mutimer. 1989. Earlyevents associated with experimental infection of the murinelung with Rhodococcus equi. J. Comp. Pathol. 101:411-420.

18. Bruner, D. W., W. W. Dimock, and P. R. Edwards. 1939. Theserological classification of Corynebacterium equi. J. Infect.Dis. 65:92-96.

19. Bruner, D. W., and P. R. Edwards. 1941. Classification ofCorynebacterium equi. Ky. Agric. Exp. Stn. Bull. 414:91-107.

20. Bull, L. B. 1924. Corynebacterial pyaemia of foals. J. Comp.Pathol. 37:294-298.

21. Buschman, E., A. A. Apt, B. V. Nickonenko, A. M. Moroz,M. K. Averbakh, and E. Skamene. 1988. Genetic aspects ofinnate resistance and acquired immunity to mycobacteria ininbred mice. Springer Semin. Immunopathol. 10:319-336.

22. Butler, W. R., D. G. Ahearn, and J. 0. Kilburn. 1986.High-performance liquid chromatography of mycolic acids as atool in the identification of Corynebacterium, Nocardia,Rhodococcus, and Mycobacterium species. J. Clin. Microbiol.23:182-185.

23. Carman, M. G., and R. T. Hodges. 1987. Distribution ofRhodococcus equi in animals, birds and from the environment.N. Z. Vet. J. 35:114-115.

24. Carpenter, J. L., and J. Blom. 1976. Corynebacterium equipneumonia in a patient with Hodgkin's disease. Am. Rev.Respir. Dis. 114:235-239.

25. Carrigan, M. J., I. J. Links, and A. G. Morton. 1988. Rhodo-coccus equi infection in goats. Aust. Vet. J. 65:331-332.

26. Carter, G. R., and G. A. Hylton. 1974. An indirect hemagglu-tination test for antibodies to Corynebacterium equi. Am. J.

30 PRESCOTT

on June 29, 2020 by guesthttp://cm

r.asm.org/

Dow

nloaded from

RHODOCOCCUS EQUI 31

Vet. Res. 35:1393-1395.27. Chaisson, R. E., and P. C. Hopewell. 1988. Editorial. Myco-

bacteria and AIDS mortality. Am. Rev. Respir. Dis. 139:1-3.28. Chirino-Trejo, J. M., and J. F. Prescott. 1987. Polyacrylamide

gel electrophoresis of whole-cell preparations of Rhodococcusequi. Can. J. Vet. Res. 51:297-300.

29. Chirino-Trejo, J. M., and J. F. Prescott. 1987. Antibodyresponse of horses to Rhodococcus equi antigens. Can. J. Vet.Res. 51:301-305.

30. Chirino-Trejo, J. M., J. F. Prescott, and J. A. Yager. 1987.Protection of foals against experimental Rhodococcus equipneumonia by oral immunization. Can. J. Vet. Res. 51:444447.

31. Cimprich, R. E., and J. R. Rooney. 1977. Corynebacteriumequi enteritis in foals. Vet. Pathol. 14:95-102.

32. Collins, F. M. 1988. AIDS-related mycobacterial disease.Springer Semin. Immunopathol. 10:375-391.

33. Collins, M. D., J. Smida, M. Dorsch, and E. Stackebrandt.1988. Isukumurella gen. nov. harboring Corynebacterium pau-rometabolum and Rhodococcus aurianticus. Int. J. Syst. Bac-teriol. 38:385-391.

34. Cotchin, E. 1943. Corynebacterium equi in the submaxillarylymph nodes of swine. J. Comp. Pathol. 53:298-309.

35. Craig, J. F., and G. 0. Davies. 1940. Corynebacterium equi inbovine pyometra. Br. Vet. J. 96:417-419.

36. Dennis, S. M., and V. W. Bamford. 1966. The role of coryne-bacteria in perinatal lamb mortality. Vet. Rec. 79:105-107.

37. Deo, M. G. 1988. Editorial. Immunological approach for con-trol of Mycobacterium avium-intracellulare infections inAIDS-an hypothesis. Int. J. Lepr. Mycobact. Dis. 56:455-463.

38. Dewes, H. F. 1972. Strongyloides westeri and Corynebacteriumequi in foals. N. Z. Vet. J. 20:82.

39. Dewes, H. F. 1989. The association between weather, frenziedbehavior, percutaneous invasion by Strongyloides westeri lar-vae and Rhodococcus equi disease in foals. N. Z. Vet. J.37:69-73.

40. Eales, L.-J., and J. M. Parkin. 1988. Current concepts in theimmunopathogenesis of AIDS and HIV infection. Br. Med.Bull. 44:38-55.

41. Ebersole, L. L., and J. L. Paturzo. 1988. Endophthalmitiscaused by Rhodococcus equi Prescott serotype 4. J. Clin.Microbiol. 26:1221-1222.

42. Elissalde, G. S., and H. W. Renshaw. 1980. Corynebacteriumequi: an interhost review with emphasis on the foal. Comp.Immunol. Microbiol. Infect. Dis. 3:433-445.

43. Elienberger, M. A., and R. P. Genetzky. 1986. Rhodococcusequi infections: literature review. Comp. Cont. Educ. Pract.Vet. 8:S414-S424.

44. Ellenberger, M. A., M. L. Kabaerlc, and J. A. Roth. 1984.Effect of Rhodococcus equi on equine polymorphonuclearleukocyte function. Vet. Immunol. Immunopathol. 7:315-324.

45. Elienberger, M. A., M. L. Kaeberle, and J. A. Roth. 1984.Equine cell-mediated immune response to Rhodococcus (Co-rynebacterium) equi. Am. J. Vet. Res. 45:2424-2427.

46. Elienberger, M. A., M. L. Kaeberle, and J. A. Roth. 1984.Equine humoral immune response to Rhodococcus (Coryne-bacterium) equi. Am. J. Vet. Res. 45:2428-2430.

47. Elliot, G., G. H. K. Lawson, and C. P. Mackenzie. 1986.Rhodococcus equi infection in cats. Vet. Rec. 118:693-694.

48. Etherington, W. G., and J. F. Prescott. 1980. Corynebacteriumequi cellulitis associated with Strongyloides penetration in afoal. J. Am. Vet. Med. Assoc. 177:1025-1027.

49. Falcon, J., B. P. Smith, T. R. O'Brien, G. P. Carlson, and E.Biberstein. 1985. Clinical and radiographic findings in Coryne-bacterium equi pneumonia of foals. J. Am. Vet. Med. Assoc.186:593-599.

50. Feldman, W. H., H. E. Moses, and A. G. Karlson. 1940.Corynebacterium equi as a possible cause of tuberculosis-likelesions of swine. Cornell Vet. 30:465-481.

51. Fierer, J., P. Wolf, L. Seed, T. Gay, K. Noonan, and P.Haghighi. 1986. Non-pulmonary Rhodococcus equi infectionsin patients with acquired immune deficiency syndrome (AIDS).

J. Clin. Pathol. 40:556-558.52. Flatla, J. L. 1942. Infeksjon med Corynebacterium equi hos

foll. Norsk Vet. Tidsskr. 54:249-276, 322-336.53. Fraser, G. 1964. The effect on animal erythrocytes of combi-

nations of diffusible substances produced by bacteria. J.Pathol. Bacteriol. 88:43-53.

54. Freestone, J. F., S. Hietala, J. Moulton, and S. Vivrette. 1987.Acquired immunodeficiency in a seven-year-old horse. J. Am.Vet. Med. Assoc. 190:689-691.

55. Furuichi, K., H. Ezoe, H. Katoh, H. Adachi, and T. Obara.1981. Regulation of allergic reaction by aerobic Corynebacte-rium equi extract, CEF. Int. Arch. Allergy Appl. Immunol.64:345-352.

56. Gainsford, S. A., and E. Frater. 1986. Two cases of infectioninvolving Rhodococcus equi. N. Z. J. Med. Lab. Technol.40:100-101.

57. Gardner, S. E., T. Pearson, and W. T. Hughes. 1976. Pneumo-nitis due to Corynebacterium equi. Chest 70:92-94.

58. Gay, C. C., V. Sloss, R. H. Wrigley, and R. Horsey. 1981. Thetreatment of pneumonia in foals caused by Rhodococcus(Corynebacterium) equi. Aust. Vet. J. 57:150-151.

59. Genetsky, R. M., M. P. Bettcher, L. H. Arp, and P. L. White.1982. Corynebacterium equi infection in a mare. Mod. Vet.Pract. 63:876-879.

60. Golub, B., G. Falk, and W. W. Spink. 1967. Lung abscess dueto Corynebacterium equi. Report of first human infection.Ann. Intern. Med. 66:1174-1177.

61. GoodfelHow, M. 1986. Genus Rhodococcus, p. 1472. InP. H. A. Sneath, N. S. Mair, M. E. Sharpe, and J. G. Holt(ed.), Bergey's manual of systematic bacteriology, vol. 2. TheWilliams & Wilkins Co., Baltimore.

62. Goodfellow, M. 1987. The taxonomic status of Rhodococcusequi. Vet. Microbiol. 14:205-209.

63. Goodfellow, M., A. R. Beckham, and M. D. Barton. 1982.Numerical classification of Rhodococcus equi and relatedactinomycetes. J. Appl. Bacteriol. 53:199-207.

64. Goodfellow, M., and T. Cross. 1984. Classification, p. 7. In M.Goodfellow, M. Mordaski, and S. T. Williams (ed.), Thebiology of the actinomycetes. Academic Press, Inc. (London),Ltd., London.

65. Goren, M. B. 1982. Immunoreactive substances of mycobac-teria. Am. Rev. Respir. Dis. 125:50-69.

66. Haglund, L. A., J. A. Trotter, L. N. Slater, S. L. Harris, P. J.Rettig, and J. R. Harkess. 1989. Case 9-1989: AIDS and acavitary pulmonary lesion. N. Engl. J. Med. 321:395.

67. Hall, R. M., and C. Ratledge. 1986. Distribution and applica-tion of mycobactins for the characterization of species withinthe genus Rhodococcus. J. Gen. Microbiol. 132:853-856.

68. Henry, J. I. 1979. Corynebacterium equi septicaemia in foals:control and therapy, no. 786. Post-graduate Committee inVeterinary Science, University of Sydney, Sydney, Australia.

69. Hietala, S. K., and A. A. Ardans. 1987. Interaction of Rhodo-coccus equi with phagocytic cells from R. equi-exposed andnon-exposed foals. Vet. Microbiol. 14:307-320.

70. Hietala, S. K., and A. A. Ardans. 1987. Neutrophil phagocyticand serum opsonic response of the foal to Corynebacteriumequi. Vet. Immunol. Immunopathol. 14:279-294.

71. Hietala, S. K., A. A. Ardans, and A. Sansome. 1985. Detectionof Corynebacterium equi-specific antibody in horses by en-zyme-linked immunosorbent assay. Am. J. Vet. Res. 46:13-15.

72. Higgins, R., and M. Paradis. 1980. Abscess caused by Cory-nebacterium equi in a cat. Can. Vet. J. 21:63-64.

73. Hillerdal, G., I. Riesenfeldt-Orn, A. Pedersen, and E. Ivani-cova. 1988. Infection with Rhodococcus equi in a patient withsarcoidosis treated with corticosteroids. Scand. J. Infect. Dis.20:673-677.

74. Hillidge, C. J. 1986. Review of Corynebacterium (Rhodococ-cus) equi lung abscesses in foals: pathogenesis, diagnosis andtreatment. Vet. Rec. 119:261-264.

75. Hillidge, C. J. 1987. Use of erythromycin-rifampin combina-tion in treatment of Rhodococcus equi pneumonia. Vet. Mi-crobiol. 14:337-342.

76. Hillman, D., B. Garretson, and R. Fiscella. 1989. Rhodococcus

VOL. 4, 1991

on June 29, 2020 by guesthttp://cm

r.asm.org/

Dow

nloaded from

CLIN. MICROBIOL. REV.

equi endophthalmitis. Arch. Ophthamol. 107:20.77. Holtman, D. F. 1945. Corynebacterium equi in chronic pneu-

monia of the calf. J. Bacteriol. 49:159-162.78. Hughes, K. L., and I. Sulaiman. 1987. The ecology of Rhodo-

coccus equi and physicochemical influences on growth. Vet.Microbiol. 14:241-250.

79. Ishino, S., M. Nakazawa, and I. Matsuda. 1987. Pathologicalfindings of guinea-pigs infected intratracheally with Rhodococ-cus (Corynebacterium) equi. Jpn. J. Vet. Sci. 49:395-402.

80. Jang, S. S., A. Lock, and E. L. Biberstein. 1975. A cat withCorynebacterium equi lymphadenitis clinically simulating lym-phosarcoma. Cornell Vet. 65:232-239.

81. Jasmin, A. M., J. M. Carroll, and J. N. Baucom. 1969.Corynebacterium equi infection in the American alligator (Al-ligator missippiensis) and the American crocodile (Crocodilusacutus). J. Comp. Lab. Med. 3:71-72.

82. Jensen, H. L. 1934. Studies on saprophytic mycobacteria andcorynebacteria. Proc. Linn. Soc. N. S. W. 59:19-61.

83. Johnson, J. A., J. F. Prescott, and R. J. F. Markham. 1983. Thepathology of experimental Corynebacterium equi infection infoals following intrabronchial challenge. Vet. Pathol. 20:440-449.

84. Johnson, J. A., J. F. Prescott, and R. J. F. Markham. 1983. Thepathology of experimental Corynebacterium equi infection infoals following intragastric challenge. Vet. Pathol. 20:450-459.

85. Jones, M. R., P. J. Say, T. J. Neale, and J. G. Horne. 1989.Rhodococcus equi: an emerging opportunistic pathogen? Aust.N. Z. J. Med. 19:103-107.

86. Karlson, A. G., H. E. Moses, and W. H. Feldman. 1940.Corynebacterium equi (Magnusson, 1923) in the submaxillarylymph nodes of swine. J. Infect. Dis. 67:243-251.

87. Kaufmann, S. H. E., and I. E. A. Flesch. 1988. The role of theT cell-macrophage interactions in tuberculosis. SpringerSemin. Immunopathol. 10:337-358.

88. Kaura, Y. K., and M. D. Mutimer. 1987. Biochemical andserological characteristics of Indian strains of Rhodococcusequi (Corynebacterium equi). Indian J. Microbiol. 27:39-42.

89. Knight, H. D., and S. Hietala. 1978. Antimicrobic susceptibilitypatterns in horses, p. 63-68. In J. D. Powers and T. E. Powers(ed.), Proceedings, 2nd Equine Pharmacology Symposium.American Association of Equine Practitioners, Golden Colo.

90. Kunke, P. J. 1987. Serious infection in an AIDS patient due toRhodococcus equi. Clin. Microbiol. Newsl. 9:163-164.

91. LeBar, W. D., and M. I. Pensler. 1986. Pleural effusion due toRhodococcus equi. J. Infect. Dis. 154:919-920.

92. Lamb, J. R., and A. D. M. Rees. 1988. Antigen specificity andfunction of human T lymphocyte clones reactive with myco-bacteria. Br. Med. Bull. 4:600-610.

93. Linder, R., and A. W. Bernheimer. 1982. Enzymatic oxidationof membrane cholesterol oxidase in relation to lysis of sheeperythrocytes by corynebacterial enzymes. Arch. Biochem.Biophys. 213:395-404.

94. Lynn, W., M. Whyte, and J. Weber. 1989. Nocardia, Myco-bacteria, and AIDS. AIDS 3:766-767.

95. MacGregor, J. H., W. M. Samuelson, D. C. Sane, and J. D.Godwin. 1986. Opportunistic lung infection caused by Rhodo-coccus (Corynebacterium) equi. Radiology 160:83-84.

96. Magnusson, H. 1923. Spezifische infektioese Pneumonie beimFohlen. Ein Neuer Eitererreger beim Pferd. Arch. Wiss. Prakt.Tierheilkd. 50:22-37.

97. Magnusson, H. 1938. Pyaemia in foals caused by Corynebac-terium equi. Vet. Rec. 50:1459-1468.

98. Marsch, J. C., and A. von Graevenitz. 1973. Recurrent Cory-nebacterium equi infection with lymphoma. Cancer 32:147-149.

99. Martens, J. G., R. J. Martens, and H. W. Renshaw. 1988.Rhodococcus (Corynebacterium) equi: bactericidal capacity ofneutrophils from neonatal and adult horses. Am. J. Vet. Res.49:295-299.

100. Martens, R. J., R. A. Fiske, and H. W. Renshaw. 1982.Experimental subacute foal pneumonia induced by aerosoladministration of Corynebacterium equi. Equine Vet. J. 14:111-116.

101. Martens, R. J., J. G. Martens, R. A. Fiske, and S. K. Hietala.1989. Rhodococcus equi foal pneumonia: protective effects ofimmune plasma in experimentally infected foals. Equine Vet.J. 21:249-255.

102. Martens, R. J., J. G. Martens, H. W. Renshaw, and S. K.Hietala. 1987. Rhodococcus equi: equine neutrophil chemilu-minescent and bactericidal responses to opsonizing antibody.Vet. Microbiol. 14:277-286.

103. Martens, R. J., H. W. Renshaw, and R. A. Fiske. 1983.Corynebacterium equi: experimental production of chronicpneumonia, p. 11. In Third Veterinary Respiratory Sympo-sium. Comparative Respiratory Society, Urbana, Ill.

104. Mbawuike, I. N., and H. B. Herscowitz. 1988. Role of activa-tion in alveolar macrophage-mediated suppression of theplaque-forming cell response. Infect. Immun. 56:577-581.

105. McKenzie, R. A., and B. A. Donald. 1979. Lymphadenitis incattle associated with Corynebacterium equi: a problem inbovine tuberculosis diagnosis. J. Comp. Pathol. 89:31-38.

106. Minnikin, D. E., M. Goodfellow, and M. D. Collins. 1978. Lipidcomposition in the classification and identification of coryne-form and related taxa, p. 84-160. In I. J. Bousfield and A. G.Calley (ed.), Coryneform bacteria. Academic Press, Inc. (Lon-don), Ltd., London.

107. Moitra, A. K. 1972. Incidence of Corynebacterium equi inbovine pneumonic lungs. Indian Vet. J. 49:973-974.

108. Muller, F., K. P. Schaal, A. von Graevenitz, L. von Moos, J. B.Woolcock, J. Wust, and A. F. Yassin. 1988. Characterization ofRhodococcus equi-like bacterium isolated from a wound infec-tion in a noncompromised host. J. Clin. Microbiol. 26:618-620.

109. Murray, H. W. 1988. Interferon-gamma, the activated macro-phage, and host defense against microbial challenge. Ann.Intern. Med. 108:595-608.

110. Mutimer, M. D., J. F. Prescott, and J. B. Woolcock. 1982.Capsular serotypes of Rhodococcus equi. Aust. Vet. J. 58:67-69.

111. Mutimer, M. D., and J. B. Woolcock. 1981. Some problemsassociated with the identification of Corynebacterium equi.Vet. Microbiol. 6:331-338.

112. Mutimer, M. D., and J. B. Woolcock. 1982. ExperimentalCorynebacterium equi infection in mice. J. Reprod. Fertil.Suppl. 32:469-472.

113. Mutimer, M. D., and J. B. Woolcock. 1982. API ZYM foridentification of Corynebacterium equi. Zentralbl. Bakteriol.Parasitenkd. Infektionskr. Hyg. Abt. 1 Orig. Reihe C 3:410-415.

114. Mutimer, M. D., and J. B. Woolcock. 1983. A note on hydro-lytic enzymes of Corynebacterium equi. J. Appl. Bacteriol.55:367-369.

115. Mutimer, M. D, J. B. Woolcock, and B. R. Sturgess. 1979.Corynebacterium equi in human faeces. Med. J. Aust. 2:422.

116. Nakazawa, M., M. Haritani, C. Sugimoto, and Y. Isayama.1983. Virulence of Rhodococcus equi for mice. Jpn. J. Vet.Sci. 45:679-682.

117. Nakazawa, M., Y. Isayama, and M. Kashiwazaki. 1987. Diag-nosis of Rhodococcus equi infection in foals by the agar geldiffusion test with protein antigen. Vet. Microbiol. 15:105-113.

118. Nakazawa, M., M. Kubo, C. Sugimoto, and Y. Isayama. 1983.Serogrouping of Rhodococcus equi. Microbiol. Immunol. 27:837-846.

119. Nakazawa, M., C. Sugimoto, and Y. Isayama. 1983. Quantita-tive culture of Rhodococcus equi from the feces of horse. Natl.Inst. Anim. Health Q. Jpn. 23:67-68.

120. Neave, R. M. S. 1951. An outbreak of ulcerative lymphangitisin young heifers in Kenya. Vet. Rec. 63:185.

121. Novak, R. M., E. L. Polisky, W. M. Janda, and C. R. Libertin.1988. Osteomyelitis caused by Rhodococcus equi in a renaltransplant patient. Infection 16:186-188.

122. Perdrizet, J. A., and D. W. Scott. 1987. Cellulitis and subcuta-neous abscesses caused by Rhodococcus equi infection in afoal. J. Am. Vet. Med. Assoc. 190:1559-1561.

123. Pradip, I. S., A. D. Larson, and C. S. McClesky. 1966.Nutritional factors affecting growth and pigmentation of Cory-nebacterium equi. Bacteriol. Proc. 1:20.

32 PRESCOTT

on June 29, 2020 by guesthttp://cm

r.asm.org/

Dow

nloaded from

RHODOCOCCUS EQUI 33

124. Prescott, J. F. 1981. Capsular serotypes of Corynebacteriumequi. Can. J. Comp. Med. 45:130-134.

125. Prescott, J. F. 1981. The susceptibility of isolates of Coryne-bacterium equi to antimicrobial drugs. J. Vet. Pharmacol.Ther. 4:27-31.

126. Prescott, J. F. 1987. Epidemiology of Rhodococcus equi infec-tion in horses. Vet. Microbiol. 14:211-214.

127. Prescott, J. F., R. Coshan-Gauthier, and L. Barksdale. 1984.Antibody to equi factor(s) in the diagnosis of Corynebacteriumequi pneumonia of foals. Can. J. Comp. Med. 48:370-373.

128. Prescott, J. F., M. Lastra, and L. Barksdale. 1982. Equi factorsin the identification of Corynebacterium equi Magnusson. J.Clin. Microbiol. 16:988-990.

129. Prescott, J. F., R. J. F. Markham, and J. A. Johnson. 1979.Cellular and humoral immune response of foals to vaccinationwith Corynebacterium equi. Can. J. Comp. Med. 43:356-364.

130. Prescott, J. F., and V. M. Nicholson. 1984. The effects ofcombinations of selected antibiotics on the growth of Coryne-bacterium equi. J. Vet. Pharmacol. Ther. 7:61-64.

131. Prescott, J. F., T. H. Ogilvie, and R. J. F. Markham. 1980.Lymphocyte immunostimulation in the diagnosis of Coryne-bacterium equi pneumonia of foals. Am. J. Vet. Res. 41:2073-2075.

132. Prescott, J. F., and C. R. Sweeney. 1985. Treatment of Cory-nebacterium equi pneumonia of foals: a review. J. Am. Vet.Med. Assoc. 187:725-728.

133. Prescott, J. F., M. Travers, and J. A. Yager-Johnson. 1984.Epidemiological survey of Corynebacterium equi infections onfive Ontario horse farms. Can. J. Comp. Med. 48:10-13.

134. Price, R. E., J. W. Templeton, R. Smith Ill, and L. G. Adams.1990. Ability of mononuclear phagocytes from cattle naturallyresistant or susceptible to brucellosis to control in vitro intra-cellular survival of Brucella abortus. Infect. Immun. 58:879-886.

135. Rahman, A. 1957. The sensitivity of various bacteria to che-motherapeutic agents. Br. Vet. J. 113:175-188.

136. Rahman, H., and K. K. Baxi. 1983. Corynebacterium equi inmastitis in a buffalo (Bubalus bubalis). Vet. Rec. 112:208-209.

137. Rajagopalan, V. R. 1938. Pneumonia in foals due to Coryne-bacterium equi. Indian J. Vet. Sci. Anim. Husb. 7:38-53.

138. Rajagopalan, V. R., and V. R. Rajagopalan. 1938. The occur-rence of Corynebacterium equi in a she-buffalo. Indian J. Vet.Sci. Anim. Husb. 8:225-234.

139. Rao, M. S., S. Zaki, B. S. Keshavamurthy, and K. C. Singh.1982. An outbreak of acute Corynebacterium equi infection inpiglets. Indian Vet. J. 59:487-488.

140. Reddy, C. A., and M. Kao. 1978. Value of acid metabolicproducts in identification of certain corynebacteria. J. Clin.Microbiol. 7:428-433.

141. Roberts, D. S. 1957. Corynebacterium equi infection in asheep. Aust. Vet. J. 33:21.

142. Roberts, M. C., D. R. Hodgson, and W. R. Kelly. 1980.Corynebacterium equi infection in an adult horse. Aust. Vet. J.56:96.

143. Rook, G. W. A. 1987. Progress in the immunology of themycobacterioses. Clin. Exp. Immunol. 69:1-9.

144. Rooney, R. J. 1966. Corynebacterial infections in foals. Mod.Vet. Pract. 47:43-45.

145. Rubin, R. H. 1982. Pneumonia in the compromised host, p.1-25. In A. P. Fishman (ed.), Update: pulmonary disease anddisorders. McGraw-Hill Book Co., New York.

146. Samies, J. H., B. N. Hathaway, R. M. Echols, J. M. Veazey,and V. A. Pilom. 1986. Lung abscess due to Corynebacteriumequi. Report of the first case in a patient with acquiredimmunodeficiency syndrome. Am. J. Med. 80:685-688.

147. Sane, D. C., and D. T. Durack. 1986. Infection with Rhodo-coccus equi in AIDS. N. Engl. J. Med. 314:56-57.

148. Savdie, E., P. Pigott, and F. Jennis. 1977. Lung abscess due toCorynebacterium equi in a renal transplant recipient. Med. J.Aust. 1:817-819.

149. Sawai, H., Y. Sumi, S. Kurano, S. Gondaira, Y. Kato, I.Tomiyasu, S. Imaizumi, K. Kaneda, and I. Yano. 1987. Gran-uloma formation by the glycolipids containing mycolic acid in

Nocardia, Rhodococcus and related actinomycetes and theirstructure analysis. Yakugaku Zasshi 107:37-45.

150. Simpson, R. 1964. Corynebacterium equi in adult horses inKenya. Bull. Epizoot. Dis. Africa 12:303-306.

151. Skalka, B. 1987. Dynamics of equi-factor antibodies in sera offoals kept on farms with differing histories of Rhodococcusequi pneumonia. Vet. Microbiol. 14:269-276.

152. Skalka, B., and A. Svastova. 1984/1985. Two techniques fordetection of antibodies against Corynebacterium (Rhodococ-cus) equi in horse sera. Vet. Microbiol. 10:293-300.

153. Skerman, V. D. B., V. McGowan, and P. H. A. Sneath. 1980.Approved list of bacterial names. Int. J. Syst. Bacteriol.30:225-400.

154. Smith, B. P., and S. S. Jang. 1980. Isolation of Corynebacte-rium equi from a foal with an ulcerated leg wound and apectoral abscess. J. Am. Vet. Med. Assoc. 177:623-624.

155. Smith, B. P., and R. C. Robinson. 1981. Studies of an outbreakof Corynebacterium equi pneumonia in foals. Equine Vet. J.13:223-228.

156. Sonnet, J., G. Wauters, F. Zech, and J. Gigi. 1987. Opportu-nistic Rhodococcus equi infection in an African AIDS case(1976-1981). Acta Clin. Belg. 42:215-216.

157. Stackebrandt, E., J. Smida, and M. D. Collins. 1988. Evidenceof phylogenetic heterogeneity within the genus Rhodococcus:revival of the genus Gordona (Tsukumura). J. Gen. Appl.Microbiol. 34:341-348.

158. Stein, F. J., and G. Stott. 1979. Corynebacterium equi in thecottontop marmoset (Saguinus oedipus): a case report. Lab.Anim. Sci. 29:519-520.

159. Sweeney, C. R., R. W. Sweeney, and T. J. Divers. 1987.Rhodococcus equi pneumonia in 48 foals: response to antimi-crobial agents. Vet. Microbiol. 14:329-336.

160. Takai, S., T. Fujimori, K. Katsuzaki, and S. Tsubaki. 1987.Ecology of Rhodococcus equi in horses and their environmenton horse-breeding farms. Vet. Microbiol. 14:233-239.

161. Takai, S., S. limori, and S. Tsubaki. 1986. Quantitative fecalculture for early diagnosis of Corynebacterium (Rhodococcus)equi enteritis in foals. Can. J. Vet. Res. 50:479-484.

162. Takai, S., S. Kawazu, and S. Tsubaki. 1985. Enzyme-linkedimmunosorbent assay for diagnosis of Corynebacterium(Rhodococcus) equi infection in foals. Am. J. Vet. Res. 46:2166-2170.

163. Takai, S., S. Kawazu, and S. Tsubaki. 1986. Immunoglobulinand specific antibody responses to Rhodococcus (Corynebac-terium) equi infection in foals as measured by enzyme-linkedimmunosorbent assay. J. Clin. Microbiol. 23:943-947.

164. Takai, S., S. Kawazu, and S. Tsubaki. 1987. Humoral immuneresponse of foals to experimental infection with Rhodococcusequi. Vet. Microbiol. 14:321-328.

165. Takai, S., T. Michizoe, K. Matsumura, M. Nagai, H. Sato, andS. Tsubaki. 1985. Correlation of in vitro properties of Rhodo-coccus (Corynebacterium) equi with virulence for mice. Mi-crobiol. Immunol. 29:1175-1184.

166. Takai, S., K. Narita, K. Ando, and S. Tsubaki. 1986. Ecologyof Rhodococcus (Corynebacterium) equi in soil on a horse-breeding farm. Vet. Microbiol. 12:169-177.

167. Takai, S., H. Ohkura, Y. Watanabe, and S. Tsubaki. 1986.Quantitative aspects of fecal Rhodococcus (Corynebacterium)equi in foals. J. Clin. Microbiol. 23:794-796.

168. Takai, S., T. Takeuchi, and S. Tsubaki. 1986. Isolation ofRhodococcus (Corynebacterium) equi and atypical mycobac-teria from the lymph nodes of healthy pigs. Jpn. J. Vet. Sci.48:445-448.

169. Takai, S., and S. Tsubaki. 1985. The incidence ofRhodococcus(Corynebacterium) equi in domestic animals and soil. Jpn. J.Vet. Sci. 47:493-496.

170. Takai, S., T. Yamagata, and S. Tsubaki. 1989. Serum immu-noglobulin concentrations in foals infected with Rhodococcusequi. Jpn. J. Vet. Sci. 51:1291-1293.

171. Thal, E., and L. Rutqvist. 1959. The pathogenicity of Coryne-bacterium equi for pigs and small laboratory animals. Nord.Vet. Med. 11:298-304.

172. Thomsen, V. F., U. Henriques, and M. Magnusson. 1968.

VOL. 4, 1991

on June 29, 2020 by guesthttp://cm

r.asm.org/

Dow

nloaded from

CLIN. MICROBIOL. REV.

Corynebacterium equi Magnusson isolated from a tuberculoidlesion in a child with adenitis coli. Dan. Med. Bull. 15:135-138.

173. Tsukumura, M. 1971. Proposal of a new genus, Gordona, forslightly acid-fast organisms occurring in the sputa of patientswith pulmonary disease and in soil. J. Gen. Microbiol. 68:15-26.

174. Tsukumura, M. 1982. Numerical analysis of the taxonomy ofnocardiae and rhodococci. Division of Nocardia asteroidessensu stricto into two species and descriptions of Nocardiaparatuberculosis sp. nov. Tsukamura (formerly the Kyoto-1group of Tsukumura), Nocardia nova sp. nov. Tsukumura,Rhodococcus aichiensis sp. nov. Tsukumura, Rhodococcuschubuensis sp. nov. Tsukumura, and Rhodococcus obuensissp. nov. Tsukumura. Microbiol. Immunol. 26:1101-1119.

175. Tsukumura, M. 1988. Genus Rhodococcus and Rhodococcusinfection. Kekkaku 63:779-783.

176. Van Etta, L. L., G. A. Filice, R. M. Ferguson, and D. N.Gerding. 1983. Corynebacterium equi: a review of 12 cases ofhuman infection. Rev. Infect. Dis. 5:1012-1018.

177. Verge, J., and F. Senthille. 1942. Caracteres differentiels deCorynebacterium equi, agent de l'adenites pseudotuberculeusedu porc. C. R. Seance Soc. Biol. (Paris) 136:295-296.

178. Weingarten, J. S., Y. Huang, and J. D. Jackman. 1988.Rhodococcus equi pneumonia. An unusual early manifestationof the acquired immunodeficiency syndrome. Chest 94:195-196.

179. Whitford, H. W., and L. P. Jones. 1974. Corynebacterium equiinfection in the goat. Southwest. Vet. 27:261-262.

180. Wilks, C. R., M. D. Barton, and J. F. Allison. 1982. Immunityto and immunotherapy for Rhodococcus equi. J. Reprod.Fertil. Suppl. 32:497-505.

181. Williams, G. D., W. J. Flanigan, and G. S. Campbell. 1971.Surgical management of localized thoracic infections in immu-nosuppressed patients. Ann. Thor. Surg. 12:471-482.

182. Wilson, M. M. 1955. A study of Corynebacterium equi infec-tion in a stud of thoroughbred horses in Victoria. Aust. Vet. J.31:175-181.

183. Woodroofe, G. M. 1950. Studies on strains of Corynebacteriumequi isolated from pigs. Aust. J. Exp. Biol. Med. Sci. 28:399-409.

184. Woolcock, J. B., A.-M. T. Farmer, and M. D. Mutimer. 1979.Selective medium for Corynebacterium equi isolation. J. Clin.Microbiol. 9:640-642.

185. Woolcock, J. B., and M. D. Mutimer. 1978. The capsules of

Corynebacterium equi and Streptococcus equi. J. Gen. Micro-biol. 109:127-130.

186. Woolcock, J. B., and M. D. Mutimer. 1980. Corynebacteriumequi: in vitro susceptibility to twenty-six antimicrobial agents.Antimicrob. Agents Chemother. 18:976-977.

187. Woolcock, J. B., M. D. Mutimer, and P. M. Bowles. 1987. Theimmunological response of foals to Rhodococcus equi: a re-view. Vet. Microbiol. 14:215-224.

188. Woolcock, J. B., M. D. Mutimer, and A.-M. T. Farmer. 1980.Epidemiology of Corynebacterium equi in horses. Res. Vet.Sci. 28:87-90.

189. Woolcock, J. B., and H. B. Ruddock. 1973. Corynebacteriumequi in cattle. Aust. Vet. J. 49:319.

190. Yager, J. A. 1987. The pathogenesis of Rhodococcus equipneumonia in foals. Vet. Microbiol. 14:225-232.

191. Yager, J. A., S. F. Foster, M. C. Zink, J. F. Prescott, and J. H.Lumsden. 1986. In vitro bactericidal efficacy of equine poly-morphonuclear leukocytes against Corynebacterium equi. Am.J. Vet. Res. 47:438-440.

192. Yanagawa, R., and E. Honda. 1976. Presence of pili in speciesof human and animal parasites and pathogens of the genusCorynebacterium. Infect. Immun. 13:1293-1295.

193. Young, D. B. 1988. Structure of mycobacterial antigens. Br.Med. Bull. 44:562-583.

194. Young, L. S. 1988. Mycobacterium avium complex infection. J.Infect. Dis. 157:863-867.

195. Zakrzewska-Czerwinska, J., M. Mordarski, and M. Goodfel-low. 1988. DNA base composition and homology values in theclassification of some Rhodococcus species. J. Gen. Microbiol.134:2807-2813.

196. Zink, M. C., and J. A. Yager. 1987. Experimental infection ofpiglets by aerosols of Rhodococcus equi. Can. J. Vet. Res.51:290-296.

197. Zink, M. C., J. A. Yager, J. F. Prescott, and M. A. Fernando.1987. Electron microscopic investigation of intracellular eventsafter ingestion of Rhodococcus equi by foal alveolar macro-phages. Vet. Microbiol. 14:295-305.

198. Zink, M. C., J. A. Yager, J. F. Prescott, and B. N. Wilkie. 1985.In vitro phagocytosis and killing of Corynebacterium equi byalveolar macrophages of foals. Am. J. Vet. Res. 46:2171-2174.

199. Zink, M. C., J. A. Yager, and N. L. Smart. 1986. Corynebac-terium equi infections in horses, 1958-1984: a review of 131cases. Can. Vet. J. 27:213-217.

34 PRESCOTT

on June 29, 2020 by guesthttp://cm

r.asm.org/

Dow

nloaded from