Ecology of Rhodococcus equi

Veterinary Microbiology, 9 (1984) 65--76 Elsevier Science Publishers B.V., Amsterdam --Printed in The Netherlands

65

E C O L O G Y OF RHODOCOCCUS EQUI

M.D. BARTON'

Department of Agriculture, Veterinary Research Institute, Park Drive, Parkville, Victoria 3052 (Australia) K.L. HUGHES

School of Veterinary Science, University of Melbourne, Park Drive, Parkville, Victoria 3052 (Australia) ' Author to whom correspondence should be addressed. (Accepted 30 June 1983)

ABSTRACT Barton, M.D. and Hughes, K.L., 1984. Ecology of Rhodococcus equi. Vet. Microbiol., 9" 65--76. A selective broth enrichment technique was used to study the distribution of Rhodococcus equi in soil and grazing animals. Rhodococcus equi was isolated from 54% of soils examined and from the gut contents, rectal faeces and dung of all grazing herbivorous species examined. Rhodococcus equi was not isolated from the faeces or dung of penned animals which did not have access to grazing. The isolation rate from dung was much higher than from other samples and this was found to be due to the ability of R. equi to multiply more readily in dung. Delayed hypersensitivity tests were carried out on horses, sheep and cattle, but only horses reacted significantly. The physiological characteristics of R. equi and the nature of its distribution in the environment suggested that R. equi is a soil organism.

INTRODUCTION T a x o n o m i c studies ( B a r t o n a n d Hughes, 1 9 8 2 ) indicate t h a t the organism c o m m o n l y k n o w n as Corynebacterium equi is best classified in the genus R h o d o c o c c u s , t h e m e m b e r s o f w h i c h can be regarded as a c t i n o m y c e t e s . This b a c t e r i u m , a p p r o p r i a t e l y described as R h o d o c o c c u s equi (Magnusson, 1 9 2 3 ; G o o d f e l l o w a n d A l d e r s o n , 1 9 7 7 ; B a r t o n a n d Hughes, 1 9 8 2 ) , is recognised w o r l d w i d e as an i m p o r t a n t cause o f b r o n c h o p n e u m o n i a , ulcerative enteritis a n d l y m p h a d e n i t i s in foals, a n d as t h e causative a g e n t o f sporadic i n f e c t i o n s in m a n y animal species ( B a r t o n and Hughes, 1 9 8 0 ) . Despite r e c o g n i t i o n o f its i m p o r t a n c e as a cause o f m o r t a l i t y in foals ( A n o n . , 1 9 7 8 ) a n d t h e u n r e s o l v e d a r g u m e n t s a b o u t its role in causing s u b m a n d i b u l a r l y m p h a d e n i t i s in swine ( B a r t o n a n d Hughes, 1 9 8 0 ) little a t t e n t i o n has been paid t o t h e n o r m a l h a b i t a t o f R. equi.

0378-1135/84/$03.00

© 1984 Elsevier Science Publishers B.V.

66 While Magnusson (1938) was of the view that R. equi is present in soil there is no record of him having isolated it from that site. Some workers have reported isolating R. equi from soil where horses or pigs have been present (Bendixen and Jepson, 1940; Flatla, 1942; Wilson, 1955; Bain, 1963; Woolcock et al., 1980) or absent (Jensen, 1934; Bendixen and Jepson, 1940). It is n o t clear whether these authors held the opinion that R. equi is normally present in soil or whether it can be found there only after contamination from the excreta and discharges from carrier or infected animals. Mahaffey {1962) noted that there was insufficient evidence to indicate that the primary habitat of R. equi is soil, but he did n o t dispute that it survived there for long periods. Cow dung (Ottosen, 1945; Woolcock et al., 1980), cow, horse and pig faeces (Woolcock et al., 1979; Mutimer and Woolcock, 1980) and gut contents (Woolcock and Mutiner, 1981), human sputum (Ottosen, 1945) and human faeces (Mutimer et al., 1979) have been reported to yield R. equi on culture. Woolcock et al. {1980) challenged the concept that R. equi is a soil organism and suggested that the gastrointestinal tract of horses is the natural habitat of R. equi. However, they restricted their investigations to a small number of horse farms, and did n o t distinguish clearly between samples of rectal faeces and dung. A preliminary search for the presence of R. equi in softs and in the faeces and large bowel contents of a number of grazing species by means of a broth enrichment technique followed by sub-culture onto selective plates (Barton and Hughes, 1981) suggested that R. equi was widely distributed in the environment. It was therefore decided to pursue these findings with a more extensive investigation to obtain further information on the ecology of R. equi. MATERIALS AND METHODS

Selective media Selective media for the isolation of R. equi have been developed (Barton and Hughes, 1981). TANP broth contained Trypticase Soy Broth (BBL) 30 g 1-1, cycloheximide 50 ~g m1-1, penicillin 10 iu m1-1 and 0.005% potassium teUurite. Selective plates used were the M3 medium of R o w b o t h a m and Cross (1977b) modified by the addition of 0.005% potassium tellurite (hereinafter referred to as M3T medium) and Tinsdale's medium (Oxoid) modified with the addition of 50 pg m1-1 cycloheximide.

Survey o f soils, bowel contents, rectal faeces and dung Two-hundred-and-seventy-seven soil samples were collected from a number of widely dispersed sites in Victoria and New South Wales. The overlying debris was scraped away and approximately 50 g of the top few centi-

67 metres of soil were collected in a 100 ml sterile disposable container (Johns Professional Products, Cheltenham, South Australia). Large bowel contents from 105 sheep, 209 cattle, 121 pigs and 26 goats were collected during routine slaughter at a number of abattoirs in Melbourne. No more than 25 samples were collected on one day and sampling t o o k place over several hours of killing to reduce bias in the source of the specimens. Rectal faeces were collected from 57 sheep, 40 cattle, 54 horses, 10 dogs, 1 cat, 10 chickens and 1 kangaroo during routine visits to farms and from carcasses examined at post-mortem. Samples of dung from 50 horses, 60 cattle, 5 sheep, 5 kangaroos, 10 deer, 8 w o m b a t s and 8 rabbits were collected at irregular intervals throughout the year from a range of localities, similar to those surveyed for soil samples. Approximately 1 g of each specimen was added to 10 ml TANP broth which was incubated at 30°C for six to seven days before being subcultured o n t o M3T and Tinsdale's plates which were incubated at 30°C for four to five days. Colonies resembling R. equi were picked o f f and presumptively identified on the basis of their colonial morphology on Trypticase Soy Agar (BBL) plates and their microscopic appearance after staining by Gram's m e t h o d (Gram-positive, cocco-bacillus). Seventy-nine of the isolates were included in a retrospective numerical taxonomic study (Barton and Hughes, 1982) and all b u t eight (all highly pigmented strains, four of which were isolated from soil and four from dung) fell into the major cluster representing R. equi or were closely related (similarity level 80% or greater) to that cluster. Soil pH and texture were determined for a small number of soil samples by officers of the Department of Agriculture, Victoria.

Multiplication of R. equi in dung Immediately after faeces were voided, the top portions of dung uncontaminated with soil were collected from seven grazing horses. This dung was divided into two: one half was incubated on the lawn or garden bed of a suburban garden and the other half on the r o o f of a building in the University of Melbourne where they were exposed to natural weather conditions. Some of the latter samples were placed on new aluminium foil trays and some on trays containing soil and old dung pats. Counts of R. equi were carried o u t by suspending 1 g of faecal material in 10 ml of 0.1% Peptone Water (Oxoid) with thorough mixing on a SuperMixer (Lab-Line Instruments Inc., Illinois, U.S.A.). The samples were centrifuged at low speed to sediment large solids and 0.1 ml o f the supernatant fluid was spread over each of three M3T plates which were incubated at 30°C for 4--5 days. Counts of bacterial colonies were carried o u t at regular intervals over 35 days.

68 At the start of the experiment fresh dung samples from these horses were cultured by the TANP broth enrichment technique. The experiment was carried out in spring and rainfall was greater than normal. The daytime temperatures fluctuated markedly ( 1 5 - - 3 7 ° C ) a n d night temperatures were between 5 and 9 ° C. The experiment was repeated using freshly voided faeces from four horses in pens with impervious floors which were cleansed daily by hosing with water. These horses had no access to grazing but were fed oaten chaff and pasture hay which were cultured for R. equi by the TANP broth enrichment technique.

Response o f animals to intradermal administration o f culture filtrates o f R. equi Five isolates from lesions in foals were each grown in Trypticase Soy Broth (BBL) at 30°C for four days. The cultures were treated with 0.5% phenol for 24 h and then centrifuged at 3000 g for 20 min. The supernatant fluids were passed through a 22 t~m membrane filter (Millipore Corporation, Bedford, MA, U.S.A.) and pooled. Pooled filtrates of five cattle and five soil isolates were prepared in a similar manner. Animals were inoculated intradermally in the neck (cattle, horses) or groin (sheep) with 0.1 ml of antigen prepared from foal isolates of R. equi. Initially 0.1 ml of sterile broth was also injected as a control, but this was discontinued when it became evident t h a t there were no reactions to the broth control. The sites were inspected 48--72 h after injection and thickening of the skin or more general swelling was regarded as a positive reaction. Small biopsies were taken from some of the reactions to confirm the response was a type IV delayed hypersensitivity reaction (DHT). Antigens prepared from isolates derived from soil and cattle were injected into the skin of 25 cattle and the reactions compared with those resulting from inoculation of the antigen prepared from R. equi from foals. RESULTS

Survey o f soils, large bowel contents, rectal faeces and dung pats for It. equi Rhodococcus equi was isolated from 150 of 277 (54%) of softs cultured. There was no apparent geographical restriction on the soil distribution of R. equi. One observation made was t h a t it was often difficult to isolate R. equi from wet soils. Occasionally it was possible to re-sample these sites when they became dry and R. equi was then frequently recovered. Ninety-six soil samples were examined from 19 farms where the principal activity was horse breeding and R. equi was isolated from 58 samples (60%) representing 18 of the 19 farms.

69 It was difficult to categorise the remaining samples. Some came from arable land where animals were n o t husbanded. However, apart from the widely distributed native animals such as macropods, or introduced species such as rabbits which would have been present in all areas sampled, most areas would have been grazed by domestic livestock at some time, albeit not in recent memory. There was no significant difference (P ~< 0.25) between the isolation of R. equi from highly acidic soils {59% of samples in range pH 4.8--5.4} compared with soils of pH ~> 5.4 (49% of 63 samples for which pH determined). Half of the samples with pH i> 7.0 yielded R. equi, but as only four samples fell in this range no conclusion could be drawn. Seventy-nine percent of 34 samples of loamy sand, sandy loam or sand yielded R. equi compared with 48% of 20 samples of clay, clay loam or sandy clay loam (P ~< 0.025). Cattle, horses, sheep, pigs and goats were found to have R. equi in their gastro-intestinal tract. Isolations from the dung of grazing animals averaged 68%. Seventy-four percent of dung specimens from horses yielded R. equi compared with 13% of equine rectal faeces (P ~< 0.001). Similarly, more dung samples from cattle and sheep yielded R. equi than did rectal faeces or large bowel contents from those animals (P ~< 0.001). Dung samples were collected all the year-round and isolation rates were not affected by seasonal or other climatic changes. Rhodococcus equi was n o t recovered from the rectal faeces or dung of 16 horses, 10 cattle, 10 sheep, 3 goats and one pig housed in pens witho u t access to grazing.

Multiplication of R. equi in dung pats (Fig. 1) Use of TANP enrichment broth resulted in the isolation of R. equi from all b u t one of the fresh dung samples examined. Counts of R. equi per gram dry weight of dung increased ten thousandfold between one and two weeks after being freshly deposited. The counts were comparable regardless of whether dung was left on bare aluminium foil trays, trays containing soil or placed on garden beds or lawn or whether R. equi was isolated from the underlying material or not. When dung samples from penned animals were studied, results similar to those described above were obtained although counts of R. equi at 10 days were slightly lower than those obtained with the dung of grazing animals. TANP broth cultures did n o t yield R. equi from any of the four dung samples examined at the start of the experiment whereas those from oaten chaff being fed to the animals did.

Response of animals to intradermal injection of culture filtrates of R. equi Dermal delayed hypersensitivity (DHT) reactions were not seen in any

70

76-

.:4

54 -

2

I 0 DAYS

1 10 AFTER

1 20 COLLECTION

I 30

i

71

".

.,~t....

•

....

.._.:.•.~

6

/: /."

./.. 9

•

..-

4./

5

e--- .--eR5

.......

/

~ ....

4

3

~G5

b-

~O tM

2 .~. , . . . ~

"m . ~

,~

,..,~ ~

--41

y •

6

50 4

."~[

e,--

---eR6

~ ....

ne~

z e~ i

~G6

3

2! ..... ' - ..ji t

6

u:

.'/

..'e/ :'/ .

./

e-- --eR7

. . . . . .

~ . . . . .~G7 I

I

I

I

O

10

20

30

DAYS

AFTER

4

o

3

O -J

2

COLLECTION

Fig. 1. Multiplication of Rhodococcus equi in dung pats placed on the roof of a building (R) or on the ground in a suburban garden (G): colony forming units of R. equi per gram dry weight of dung pat material during incubation under various conditions for 35 days. R1--R4: fresh dung pats placed on trays containing soil and weathered dung from which R. equi had been isolated previously; R5--R7: fresh dung pats placed on new bare aluminium foil trays; G1, GS, G6: fresh dung pats placed on garden lawn, R. equi isolated previously from underlying soil; G2, G3: fresh dung pats placed on garden bed from which R. equi had not been isolated previously; G4, G7: fresh dung pats placed on garden bed from which R. equi had been isolated previously. (-.....- ..... -) Indicates lower limit o f sensitivity of plate count technique.

72 of the 150 sheep tested and only one bull reacted from among 143 cattle tested with the antigen prepared from isolates from foals. There was no correlation between breed, history of a R. equi problem or isolation of R. equi from soil or dung and the proportion of horses that gave positive DHT reactions. Histological examination of biopsy samples confirmed that swellings seen were DHT reactions. None of the antigens derived from soil and cattle isolates of R. equi elicited any response when inoculated intradermally in cattle; horses were n o t tested with these preparations. DISCUSSION The proportion of softs which yielded R. equi is higher than that reported by Woolcock et al. (1980) who isolated R. equi from 7/14 soil samples from 7/9 horse farms. The isolation rate is also higher than that reported previously for " R h o d o c h r o u s " strains (Schaal and Bickenbach, 1978) or nocardioform actinomycetes (Cross et al., 1976; Orchard et al., 1977). This may reflect the greater sensitivity of a broth enrichment technique over other methods such as paraffin baiting or selective plating technique using media which favour the isolation of streptomycetes rather than nocardioform actinomycetes (Cross et al., 1976). One serious disadvantage of the TANP enrichment technique is that it is qualitative rather than quantitative and does n o t permit enumeration of the number of R. equi present in a sample. Perhaps statistical methods incorporating a technique using a series of dilutions of TANP broth can be developed to overcome this. Attempts to carry o u t counts for R. equi using M3T and Tinsdale's medium and plate dilution techniques indicated that the counts per gram dry weight of soil were less than 3000 c.f.u, per gram which is below the limit of sensitivity of the technique used (M.D. Barton, 1982, unpublished results). Unlike most actinomycetes which prefer neutral or alkaline soil (Lacey, 1973) the isolation of R. equi was n o t inhibited from acid soils, in keeping with its known resistance to treatment with acid (Karlson et al., 1940). Acidophilic actinomycetes will grow only on acidified medium (Khan and Williams, 1975) unlike R. equi which also grows well on neutral media. Limited studies suggest a possible correlation between sandy softs and the isolation of R. equi. Studies with actinomycetes have suggested previously that different genera are favoured by different soils (Lacey, 1973). Rhodococcus equi was isolated from soil from sites with variable land use including a fauna reserve, native forest and arable land, from wheat farms, grazing enterprises and irrigated dairy farms. Some isolates were made from soil on farms where horses were n o t known to have grazed. However, as faeces and dung pats of all herbivores (domestic livestock and native and introduced wild animals) contained R. equi, any attempt to correlate the presence of R. equi with grazing by horses is unlikely to

73 be profitable. (The converse is also true in that it would be very difficult to find virgin soil unpolluted by animals' excreta to examine for the presence of R. equi.) Highest numbers of actinomycetes can be isolated from soils with high levels of organic matter (Lacey, 1973). The difficulty experienced in isolating R. equi from w e t soils is in keeping with the knowledge that actinomycetes are more tolerant of dry conditions (Ishizawa et al., 1958). Inhibition of growth occurs when soils are waterlogged hence gaseous diffusion is reduced (Williams et al., 1972). However, some nocardioform actinomycetes, including Rhodococcus, have been isolated from aquatic habitats ( R o w b o t h a m and Cross, 1977a, b). The extent of gastro-intestinal tract carriage of R. equi by grazing herbivores suggests that all herbivores could carry R. equi at least while grazing. The failure to isolate R. equi from the faeces or dung of penned animals suggests that its presence in the gut is transient and depends upon ingestion of pasture or feedstuffs contaminated with R. equi. There have been few studies of the gastro-intestinal tract flora of horses (Trum, 1949; Becker, 1964; Nurmio et al., 1973) and none of these reported isolating R. equi even though their methods were directed towards the isolation of aerobes. On the other hand the bacterial microflora of the intestinal tract of swine has been extensively studied, but no workers have reported isolating R. equi as a c o m p o n e n t of the normal flora. It is n o t unusual for soil bacteria for example, saprophytic mycobacteria and nocardioform actinomycetes to be isolated from the gut contents or faeces of animals (M.D. Barton, 1974, unpublished observation). There is no ready explanation for the differences in isolation rates from faeces between horses and cattle, particularly when there was no significant difference in isolation rates from dung. The differences in isolation rates between gut contents and rectal faeces probably reflect the higher water c o n t e n t in the former specimens for which no correction was made when the samples were cultured. Reduction in moisture content could account partly for the difference in isolation rates between faeces and dung pats. However, the capacity of R. equi to multiply ten-thousandfold in dung no d o u b t accounts principally for these differences. The high isolation rate from dung of other herbivores suggests that this phenomenon is n o t restricted to horse dung. If this coprophilic characteristic is widespread among actinomycetes it may help account for the higher isolation rate of actinomycetes from grazing land (Lacey, 1973). The ability of R. coprophilus to multiply in cow dung has been d o c u m e n t e d previously ( R o w b o t h a m and Cross, 1977a). The maximum colony counts of R. equi were the same regardless of where the dung was incubated. As R. equi was detected in the original fresh dung (albeit in low numbers below the sensitivity of the counting technique) this suggests that multiplication was primarily from R. equi already in the dung rather than by invasion from soil or old dung nearby, at least for the duration of the experiment. The fact that dung from penned

74

animals (from whose faeces R. equi could n o t be isolated) would show the same multiplication pattern of R. equi was explained when R. equi was isolated in low numbers from oaten chaff being fed to them. Thus, R. equi is probably ingested by animals in their f o o d although, being an aerobe, little or no multiplication occurs in the gut. By comparison, voided dung provides a more favourable environment for its augmentation outside the animal host. The capacity of R. equi to multiply to high numbers in dung has implications for the study of diseases induced by R. equi in foals. As coprophagia is a normal part of a foal's development (Francis-Smith and WoodGush, 1977), crowded foaling paddocks would provide an environment conducive to massive challenge of foals at a time when they are apparently most susceptible to developing infection with R. equi; alternatively dried faecal material could be inhaled. Considering the ubiquitous nature of R. equi and the high proportion of horses giving DHT reactions it seems likely that most horses m o u n t a cell-mediated immune response against it. This p h e n o m e n o n was first described by Wilson (1955). More recently, development of cell-mediated immune responses after vaccination and experimental infection (Prescott et al., 1979) have been described. There is no ready explanation as to why the sheep and cattle did n o t respond in this way in the face of similar environmental exposure to R. equi. It may be that horses generally or thoroughbreds in particular are more susceptible than other species. The physiological characteristics of R. equi (Goodfellow et al., 1982a, b; Barton and Hughes, 1982), its widespread distribution in soil and grazing herbivores and the apparent dependence of gut carriage on the presence of R. equi in the diet support the concept that R. equi is a soil organism. ACKNOWLEDGEMENTS

The authors gratefully acknowledge the assistance of B. Cole, L.J. Fulton, W.P. Howey, J. Rodger, R. Slarke, P. Schroder and B. Curnow with the collection of soil, gut contents and faecal samples. Thanks are also due to K. Peverill and H. Jones for carrying out soil pH and texture analyses.

REFERENCES Anon., 1978. Report of foal pneumonia panel. J. Eq. Med. Surg., 2: 400--412. Bain, A.M., 1963. Coryneybacterium equi infections in the equine. Aust. Vet. J., 39: 116--121. Barton, M.D. and Hughes, K.L., 1980. Corynebacterium equi: a review. Vet. Bull., 50: 65--80. Barton, M.D. and Hughes, K.L., 1981. Comparison of three techniques for isolation of Rhodococcus equi from contaminated sources. J. Clin. Microbiol., 13: 219--221. Barton, M.D. and Hughes, K.L., 1982. Is Rhodococcus equi a soil organism? Proceedings of the Third International Symposium on Equine Reproduction, J. Reprod. Fertil. Suppl., 32 : 481--489.

75 Becker, C.R., 1964. A study of the intestinal microflora in the equine. Mich. State Univ. Vet., 24: 123--126. Bendixen, H.C. and Jepson, A., 1940. Fortgesetze Untersuchung ~ Corynebakterien equi mit besonderer Beriicksichtigung gewissen morphologisher und biologischer Verhaltnisse sowie der Pathogenn~tvsvehaltnisse Schweinen geniiber. Z. Infektionskr. Parasit. Kr. Hyg., Berlin, 57: 9--36. Cross, T., Rowbotham, T.J., Mishustin, E.N., Tepper, E.Z., Antoineportaels, F., Schaal, K.P. and Bickenbach, H., 1976. The ecology of nocardioform actinomycetes. In: M. Goodfellow, G.H. Brownell and J.A. Serrano (Editors), The Biology of the Nocardiae. Academic Press, London, pp. 337--371. Flatla, J.L., 1942. Infeksjon reed Corynebacterium equi hos f~ll. Nor. Vet. Tidsskr., 5 4 : 2 4 9 - - 2 7 6 and 322--377. Francis-Smith, K. and Wood-Gush, D.G.M., 1977. Coprophagia as seen in thoroughbred foals. Eq. Vet. J., 9: 155--157. Goodfellow, M. and Alderson, G., 1977. The actinomycete-genus Rhodococcus: a home for the " r h o d o c h r o u s " complex. J. Gen. Microbiol., 100: 99--122. Goodfellow, M., Beckham, A.R. and Barton, M.D., 1982a. Numerical classification of Rhodococcus equi and related actinomycetes. J. Appl. Bacteriol., 53: 199--207. Goodfellow, M., Weaver, G.R. and Minnikin. D.E., 1982b. Numerical classification of some rhodococci, corynebacteria and related organisms. J. Gen. Microbiol., 128: 731--745. Ishizawa, S., Suzuki, T., Koda, T. and Sato, I., 1958. Studies on the micro-organism, and their activities in soil. Bull. Natl. Inst. Agric. Sci., B8: 167--211. Jensen, H.L., 1934. Studies on the saprophytic mycobacteria and corynebacteria. Proc. Linn. Soc. N. S. W., 59: 19--61. Karlson, A.G., Moses, H.E. and Feldman, W.H., 1940. Corynebacterium equi in the submaxillary lymph nodes of swine. J. Infect. Dis., 67 : 243--251. Khan, M.R. and Williams, S.T., 1975. Studies on the ecology of actinomycetes in soil. VIII. Distribution and characteristics of acidophilic actinomycetes. Soil Biol. Biochem., 7: 345--348. Lacey, J., 1973. Actinomycetes in soil, composts and fodder. In: G. Sykes and F.A. Skinner (Editors), Actinomycetes Characteristics and Practical Importance. Academic Press, London, pp. 231--251. Magnusson, H., 1923. Specifische infektiose Pneumonie beim Fohlen. Ein neuer Eiterreger beim Pferd. Arch. Wissenschaftl. Prakt. Tierheilk., 50: 22--38. Magnusson, H., 1938. Pyaemia in foals caused by Corynebacterium equi. Vet. Rec., 50: 1459--1468. Mahaffey, L.W., 1962. Respiratory conditions in horses. Vet. Rec., 74: 1295--1314. Mutimer, M.D. and Woolcock, J.B., 1980. Corynebacterium equi in cattle and in pigs. Vet. Q., 2: 2 5 - 2 7 . Mutimer, M.D., Woolcock, J.B. and Sturgess, B.R., 1979. Corynebactcrium equi in human faeces. Med. J. Aust., ii: 422. Nurmio, P., Koiranen, L. and Tupamaki, A., 1973. The faecal microflora of horses. Suom. Elainlaakarileki, 79: 668---681. Orchard, V., Goodfellow, M. and Williams, S.T,, 1977. Selective isolation and occurrence of nocardiae in soil. Soil Biol. Biochem., 9: 233--238. Ottosen, H.E., 1945. Unders~bgelser over Corynebacterium Magnusson-Holth: specielt reed henblik paa des serologiske forhold. Carl Fr. Martensen: Copenhagen. Prescott, J.F., Markham, R.J.F. and Johnson, J.A., 1979. Cellular and humoral immune response of foals to vaccinations with Corynebacterium equi. Can. J. Comp. Med., 43: 356--364. Rowbotham, T.J. and Cross, T., 1977a. Rhodococcus coprophilus sp. nov.: an aerobic nocardioform actinomycete belonging to the "rhodochrous" complex. J. Gen. Microbiol., 100: 123--138.

76 Rowbotham, T.J. and Cross, T., 1977b. Ecology of Rhodococcus coprophilus and associated actinomycetes in fresh water and agricultural habitats. J. Gen. Microbiol., 100: 231--240. Schaal, K.P. and Bickenbach, H., 1978. Soil occurrence of pathogenic nocardia. In: M. Mordarski, W. Kurylowicz and J. Jeljaszewicz (Editors), Nocardia and Streptomyces. Proceedings of the International Symposium on Nocardia and Streptomyces, Warsaw, 1976. Zentralbl. Bakteriol. Parasitenkd. Infektionskr. Hyg. Abt. 1, Suppl. 6. Fischer Verlag, Stuttgart, pp. 429--434. Trum, B.F., 1949. Diseases of the foal. J. Am. Vet. Med. Assoc., 114: 218--223. Williams, S.T., Shameemullah, M., Watson, E.T. and Mayfield, C.I., 1972. Studies on the ecology of actinomycetes in soil. VI. The influence of moisture tension on growth and survival. Soil Biol. Biochem., 4: 215--225. Wilson, M.M., 1955. A study of Corynebacterium equi infection in a stud of thoroughbred horses in Victoria. Aust. Vet. J., 31: 175--181. Woolcock, J.B. and Mutimer, M.D., 1981. Corynebacterium equi in the gastro-intestinal tract of ruminants. Vet. Res. Commun., 4: 291--294. Woolcock, J.B., Farmer, A.M.T. and Mutimer, M.D., 1979, Selective medium for Corynebacterium equi. J. Clin. Microbiol., 9: 640---642. Woolcock, J.B., Mutimer, M.D. and Farmer, A.M.T., 1980. Epidemiology of Corynebacterium equi in horses. Res. Vet. Sci., 28: 87--90.