IN-DEPTH: EQUI

Clinical Manifestations, Pathogenesis, and Diagnosis of Caused by Rhodococcus equi in Foals

Steeve Gigue`re, DVM, PhD, Diplomate ACVIM

Rhodococcus equi is a common cause of in foals, but extrapulmonary manifestations are common. The of R. equi for foals depends on the presence of a large expressing the genes for multiple virulence-associated proteins (Vap). The plasmid-encoded VapA is essential for virulence, but expression of VapA alone without any other plasmid-encoded genes is not sufficient to restore virulence. Bacteriologic culture and polymerase chain reaction (PCR) amplification of the vapA gene combined with cytological examination of a tracheobronchial aspirate are necessary to make a definitive diagnosis of pneumonia caused by R. equi. Author’s address: College of Veteri- nary Medicine, University of Georgia, Athens, Georgia 30602; e-mail: [email protected]. © 2010 AAEP.

1. Introduction markable ability of foals to compensate for the pro- Rhodococcus equi, a gram-positive facultative intra- gressive loss of functional lung makes early clinical cellular pathogen, is one of the most common causes diagnosis difficult. Early clinical signs often only of pneumonia in foals between 3 wk and 5 mo of age. consist of a mild fever or a slight increase in respi- Although R. equi can be cultured from the environ- ratory rate that may not be apparent unless foals ment of virtually all horse farms, the clinical disease are exercised or stressed by handling. As the dis- in foals is endemic and devastating on some farms, ease progresses, clinical signs may include decreased sporadic on others, and unrecognized on most. appetite, lethargy, fever, tachypnea, and labored On farms where the disease is endemic, costs asso- breathing. Cough and bilateral nasal discharge are ciated with veterinary care, long-term therapy, and inconsistent findings. mortality of some foals may be very high. This text Extrapulmonary manifestations of rhodococcal in- reviews the clinical manifestations, pathogenesis, fections are common. In a retrospective study of and diagnosis of R. equi infections in foals. 150 foals with R. equi infections, 111 (74%) had at least 1 of 39 extrapulmonary disorders.1 Survival 2. Clinical Manifestations was significantly higher among foals without ex- The most common manifestation of R. equi infec- trapulmonary disorders (32/39; 82%) than among tions in foals is a chronic suppurative bronchopneu- foals with extrapulmonary disorders (48/111; 43%), monia with extensive abscessation. The slow but many such disorders were only recognized after spread of the lung combined with the re- death.1 Intestinal lesions are present in ϳ50% of

NOTES

AAEP PROCEEDINGS ր Vol. 56 ր 2010 121 IN-DEPTH: RHODOCOCCUS EQUI foals with R. equi pneumonia presented for necrop- and contains the genes for a sy.2 However, the majority of foals with R. equi family of eight closely related virulence-associated pneumonia do not show clinical signs of intestinal proteins designated VapA and VapC–VapI. VapA disease. Abdominal lesions may include ulcerative is expressed on the bacterial surface, and its expres- enterocolitis and typhlitis over the area of the Peyer’s sion is temperature regulated, occurring between patches, granulomatous or suppurative inflammation 34°C and 41°C.16 VapC, -D, and -E are secreted of the mesenteric and/or colonic lymph nodes, or, in proteins concomitantly regulated by temperature some cases, a single large abdominal abscess may be with VapA.17 In a recent study, a R. equi mutant the only lesion.2 Polysynovitis is present in ϳ25– lacking a 7.9-kb DNA region spanning 6 vap genes 30% of cases with R. equi infections.1,3 The degree of (vapA, -C, -D, -E, -F, -I) was attenuated for virulence joint effusion is variable and, in most cases, lame- in mice and failed to replicate in .18 ness is mild or absent. Cytological examination of Complementation with vapA alone could restore full the synovial fluid usually shows a non-septic mono- virulence, whereas complementation with vapC, nuclear pleocytosis, and bacteriologic culture of the vapD, or vapE could not. Conversely, a recombi- synovial fluid is negative.3 Histologic examination nant plasmid-cured derivative expressing wild-type of the synovial membrane of a few affected foals levels of VapA failed to survive and replicate in showed lymphoplasmacytic synovitis, suggesting an macrophages and remained avirulent for foals, immune-mediated process.4,5 Immune-mediated showing that expression of VapA alone is not suffi- processes may also contribute to the development of cient to restore the virulence phenotype.13 These uveitis, anemia, and thrombocytopenia in some foals findings show that, although VapA is essential for infected with R. equi.1 virulence, other plasmid-encoded products also con- Bacteremic spread of the organism from the lungs tribute to the ability of R. equi to cause disease. or gastrointestinal tract may occasionally result in All vap genes, as well as five other ORFs within the septic arthritis and, more commonly, osteomyelitis. pathogenicity island, are upregulated when R. equi However, foals can occasionally develop R. equi sep- is grown in monolayers.19 Of all the tic arthritis or osteomyelitis without apparent lung vap genes, vapA, vapD, and vapG are preferentially involvement or other source of infection. R. equi expressed in the lungs of infected foals, and expres- vertebral osteomyelitis or disk spondylitis resulting sion of these genes in the lungs is significantly in spinal cord compression has also been reported.6–8 higher than that achieved during in vitro growth.20 Other rare extrapulmonary manifestations of R. equi Regulation of expression of the genes of the patho- infections in foals include panophthalmitis, guttural genicity island is complex and depends on at least pouch empyema, sinusitis, pericarditis, nephritis, and five environmental signals including temperature, hepatic, renal, and intracranial abscessation.1,9,10 pH, oxidative stress, magnesium, and iron.16,19,21 The precise role of each of these genes in the patho- 3. Virulence and Pathogenesis genesis of R. equi infections remains to be The ability of R. equi to induce disease in foals likely determined. depends on both host and microbial factors. R. equi R. equi isolates are often classified as virulent, is a facultative intracellular pathogen, and its abil- avirulent, or intermediately virulent based on their ity to persist in, and eventually destroy, alveolar ability to induce disease or death in mice. Virulent macrophages seems to be the basis of its pathoge- R. equi isolates contain the large plasmid described nicity. Knowledge of the virulence mechanisms of above and express VapA. Intermediately virulent R. equi were largely speculative until the discovery R. equi isolates contain one of four distinct large of a virulence plasmid.11,12 Unlike most environ- encoding a 20-kDa antigen (VapB) that is mental R. equi, isolates from pneumonic foals typi- related to, but distinct from, VapA. In contrast, cally contain an 80- to 90-kb plasmid. Plasmid- avirulent R. equi isolates do not express Vap anti- cured derivatives of virulent R. equi strains lose gens. All three categories of R. equi have the abil- their ability to replicate and survive in macro- ity to cause disease in immunosuppressed . phages.13 Plasmid-cured derivatives also fail to in- Analysis of R. equi isolates from immunocompro- duce pneumonia and are completely cleared from mised patients with and without AIDS the lungs of foals within 2 wk after heavy intrabron- showed that only ϳ20% of isolates contain an 80- to chial challenge, confirming the absolute necessity of 90-kb plasmid and express VapA.22,23 Therefore, the large plasmid for the virulence of R. equi.13,14 the pathogenesis of R. equi infection in immunocom- Nucleotide sequencing of the large plasmid of promised human patients seems to be different from two foal isolates showed the presence of 69 open the pathogenesis in foals, in which the virulence reading frames (ORFs).15 Comparisons of the plas- plasmid is always found. To the author’s knowl- mid sequence with genes previously identified in edge, intermediately virulent isolates (expressing other microorganisms has identified three func- VapB) have never been isolated from foals with nat- tional regions. Two of these regions contain genes urally acquired R. equi infections. Experimentally, encoding proteins involved in conjugation and in heavy intrabronchial challenge of foals with inter- plasmid replication stability and segregation.15 mediately virulent R. equi results in pneumonia but The third region of 27.5 kb bears the hallmark of a at a dose much higher than that required for induc-

122 2010 ր Vol. 56 ր AAEP PROCEEDINGS IN-DEPTH: RHODOCOCCUS EQUI tion of pneumonia with VapA-expressing strains.24 PCR amplification of the vapA gene may be done in Almost all isolates from the submaxillary lymph association with, but should not replace, bacterial nodes of produce VapB and are intermediately culture. virulent to mice, suggesting that pigs or their envi- On endemic farms, many foals without clinical ronment may be the source of infection for some disease have R. equi in their trachea as a result of human cases.25 Most isolates from , , inhalation of contaminated dust or as a result of a and dogs are avirulent and do not contain plasmids subclinical infection.40 For this reason, culture or encoding vapA or vapB.26–28 In contrast, most iso- PCR amplification of R. equi from a TBA should be lates from contain a large plasmid and express interpreted in the context of cytological evaluation VapA.27 and clinical examination. A light growth of R. equi Although it has been firmly established that the from a foal with no clinical signs of respiratory dis- virulence plasmid is essential for infection of foals, it ease, no cytological evidence of airway inflamma- is also clear that chromosomally encoded factors, tion, and no ultrasonographic or radiographic such as regulatory genes and metabolic pathways, evidence of pulmonary lesions is likely an incidental are also important in allowing the pathogen to finding. thrive within the host.29–32 R. equi is closely related Several independent studies have recently evalu- to , as evidenced by a ated the performance of available serological tests partial genome sequence of R. equi that showed that for diagnosis of infection caused by R. equi on en- the majority of R. equi genes are largely homologous demic farms.41–44 The serological tests evaluated with M. tuberculosis.33 were found to either have low sensitivity, low spec- Inhalation of virulent R. equi is the major route of ificity, or both. Improving either sensitivity or pneumonic infection. The incubation period after specificity of enzyme-linked immunosorbent assay experimental intrabronchial challenge varies from (ELISA) assays by changing the cut-off value of the ϳ9 days after administration of a heavy inoculum to tests could only be done at the detriment of the ϳ2–4 wk when a lower inoculum is adminis- other. The performance of each assay was not im- tered.13,34 Lung consolidation can be detected as proved by sequential sampling of each foal at 2-wk early as 3 days after heavy intrabronchial chal- intervals over time.44 Reliance on serology as the 13 lenge. The incubation period under field condi- sole diagnostic test for R. equi infections results in tions is unknown but likely depends on the number overdiagnosis of the disease and in missing infec- of virulent inhaled and various host-related tions in the early stages. factors. Ingestion of the organism is a significant route of exposure, and likely also of immunization, but rarely leads to hematogenously acquired pneu- References and Footnotes monia unless the foal has multiple exposures to 1. Reuss SM, Chaffin MK, Cohen ND. Extrapulmonary disor- large numbers of bacteria.35 ders associated with Rhodococcus equi infection in foals: a retrospective study of 150 cases (1987–2007). J Am Vet Med 4. Diagnosis Assoc 2009;235:855–863. 2. Zink MC, Yager JA, Smart NL. equi infec- The distinction between lower respiratory tract in- tions in horses, 1958–1984: a review of 131 cases. Can J fections caused by R. equi and that caused by other Vet Res 1986;27:213–217. pathogens is problematic, especially on farms with 3. Sweeney CR, Sweeney RW, Divers TJ. Rhodococcus equi pneumonia in 48 foals: response to antimicrobial therapy. no previous history of R. equi infections. Many di- Vet Microbiol 1987;14:329–336. agnostic tests including complete blood count (CBC), 4. Kenney DG, Robbins SC, Prescott JF, et al. Development of measurement of fibrinogen concentrations, ultra- reactive arthritis and resistance to erythromycin and ri- sonography, radiographs, and serology may help dis- fampin in a foal during treatment for Rhodococcus equi pneu- monia. Equine Vet J 1994;26:246–248. tinguish pneumonia caused by R. equi from that 5. Madison JB, Scarratt WK. Immune-mediated polysynovitis caused by other pathogens. In one study, white cell in four foals. J Am Vet Med Assoc 1988;192:1581–1584. counts Ͼ20,000 cells/␮l, fibrinogen concentrations 6. Chaffin MK, Honnas CM, Crabill MR, et al. Cauda equina Ͼ700 mg/dl, and evidence of thoracic abscessation syndrome, diskospondylitis, and a paravertebral abscess were more likely to be found in foals with pneumo- caused by Rhodococcus equi in a foal. J Am Vet Med Assoc 1995;206:215–220. nia caused by R. equi than in foals with pneumonia 36 7. Gigue`re S, Lavoie JP. Rhodococcus equi vertebral osteomy- caused by other bacterial pathogens. However, elitis in 3 Quarter Horse colts . Equine Vet J 1994;26:74–77. bacteriologic culture and/or polymerase chain reac- 8. Olchowy TW. Vertebral body osteomyelitis due to Rhodococ- tion (PCR) amplification of the vapA gene and cyto- cus equi in two Arabian foals. Equine Vet J 1994;26:79–82. logical examination of a tracheobronchial aspirate 9. Chaffin MK, Martens RJ. Extrapulmonary disorders asso- ciated with Rhodococcus equi pneumonia in foals: retrospective (TBA) are necessary to make a definitive diagnosis study of 61 cases (1988–1996), in Proceedings. 43rd American of pneumonia caused by R. equi. Association of Equine Practioners 1997;79–80. PCR amplification of the vapA gene is more sen- 10. Janicek JC, Kramer J, Coates JR, et al. Intracranial ab- sitive than bacterial culture.37–39 However, culture scess caused by Rhodococcus equi infection in a foal. JAm Vet Med Assoc 2006;228:251–253. offers the advantage of detecting other bacterial 11. Takai S, Sekizaki T, Ozawa T, et al. Association between a pathogens present and permits in vitro susceptibil- large plasmid and 15- to 17-kilodalton antigens in virulent ity testing of the recovered pathogens. As a result, Rhodococcus equi. Infect Immun 1991;59:4056–4060.

AAEP PROCEEDINGS ր Vol. 56 ր 2010 123 IN-DEPTH: RHODOCOCCUS EQUI

12. Tkachuk-Saad O, Prescott J. Rhodococcus equi plasmids: 30. Linder R, Bernheimer AW. Oxidation of macrophage mem- isolation and partial characterization. J Clin Microbiol brane cholesterol by intracellular Rhodococcus equi. Vet 1991;29:2696–2700. Microbiol 1997;56:269–276. 13. Gigue`re S, Hondalus MK, Yager JA, et al. Role of the 85- 31. Navas J, Gonzalez-Zorn B, Ladron N, et al. Identification kilobase plasmid and plasmid-encoded virulence-associated and mutagenesis by allelic exchange of choE, encoding a protein A in intracellular survival and virulence of Rhodococ- cholesterol oxidase from the intracellular pathogen Rhodo- cus equi. Infect Immun 1999;67:3548–3557. coccus equi. J Bacteriol 2001;183:4796–4805. 14. Wada R, Kamada M, Anzai T, et al. Pathogenicity and vir- 32. Wall DM, Duffy PS, Dupont C, et al. Isocitrate lyase activity ulence of Rhodococcus equi in foals following intratracheal is required for virulence of the intracellular pathogen Rhodo- challenge. Vet Microbiol 1997;56:301–312. coccus equi. Infect Immun 2005;73:6736–6741. 15. Takai S, Hines SA, Sekizaki T, et al. DNA Sequence and comparison of virulence plasmids from Rhodococcus equi 33. Rahman MT, Herron LL, Kapur V, et al. Partial genome ATCC 33701 and 103. Infect Immun 2000;68:6840–6847. sequencing of Rhodococcus equi ATCC 33701. Vet Microbiol 16. Takai S, Fukunaga N, Kamisawa K, et al. Expression of 2003;94:143–158. virulence-associated antigens of Rhodococcus equi is regu- 34. Barton MD, Embury DH. Studies of the pathogenesis of lated by temperature and pH. Microbiol Immunol 1996;40: Rhodococcus equi infection in foals. Aust Vet J 1987;64:332– 591–594. 339. 17. Byrne BA, Prescott JF, Palmer GH, et al. Virulence plasmid 35. Johnson JA, Prescott JF, Markham RJ. The pathology of of Rhodococcus equi contains inducible gene family encoding experimental Corynebacterium equi infection in foals follow- secreted proteins. Infect Immun 2001;69:650–656. ing intragastric challenge. Vet Pathol 1983;20:450–459. 18. Jain S, Bloom BR, Hondalus MK. Deletion of vapA encoding 36. Leclere M, Magdesian KG, Kass PH, et al. Comparison of virulence associated protein A attenuates the intracellular the clinical, microbiological, radiological and haematological actinomycete Rhodococcus equi. Mol Microbiol 50:115–128, features of foals with pneumonia caused by Rhodococcus equi 2003. and other bacteria. Vet J 2009. 19. Ren J, Prescott JF. Analysis of virulence plasmid gene ex- 37. Pusterla N, Wilson WD, Mapes S, et al. Diagnostic evalua- pression of intra-macrophage and in vitro grown Rhodococcus tion of real-time PCR in the detection of Rhodococcus equi in equi ATCC 33701. Vet Microbiol 2003;94:167–182. faeces and nasopharyngeal swabs from foals with pneumo- 20. Jacks S, Gigue`re S, Prescott JF. In vivo expression of nia. Vet Rec 2007;161:272–275. and cell-mediated immune responses to the plasmid-encoded 38. Rodriguez-Lazaro D, Lewis DA, Ocampo-Sosa AA, et al. virulence-associated proteins of Rhodococcus equi in foals. Clin Vaccine Immunol 2007;14:369–374. Internally controlled real-time PCR method for quantitative 21. Benoit S, Benachour A, Taouji S, et al. H(2)O(2), which species-specific detection and vapA genotyping of Rhodo- causes macrophage-related stress, triggers induction of ex- coccus equi. Appl Environ Microbiol 2006;72:4256–4263. pression of virulence-associated plasmid determinants in 39. Sellon DC, Besser TE, Vivrette SL, et al. Comparison of Rhodococcus equi. Infect Immun 2002;70:3768–3776. nucleic acid amplification, serology, and microbiologic culture 22. Takai S, Sasaki Y, Ikeda T, et al. Virulence of Rhodococcus for diagnosis of Rhodococcus equi pneumonia in foals. J Clin equi isolates from patients with and without AIDS. J Clin Microbiol 2001;39:1289–1293. Microbiol 1994;32:457–460. 40. Ardans AA, Hietala SK, Spensley MS, et al. Studies of nat- 23. Takai S, Imai Y, Fukunaga N, et al. Identification of viru- urally occuring and experimental Rhodococcus equi,inPro- lence-associated antigens and plasmids in Rhodococcus equi ceedings. 32nd Annual Convention of the American from patients with AIDS. J Infect Dis 1995;172:1306–1311. Association of Equine Practioners 1986;129–144. 24. Takai S, Anzai T, Fujita Y, et al. Pathogenicity of Rhodo- 41. Gigue`re S, Hernandez J, Gaskin JM, et al. Evaluation of coccus equi expressing a virulence-associated 20 kDa protein WBC concentration, plasma fibrinogen concentration, and an (VapB) in foals. Vet Microbiol 2000;76:71–80. agar gel immunodiffusion test for early identification of foals 25. Takai S, Fukunaga N, Ochiai S, et al. Identification of in- with Rhodococcus equi pneumonia. J Am Vet Med Assoc termediately virulent Rhodococcus equi isolates from pigs. 2003;222:775–781. J Clin Microbiol 1996;34:1034–1037. 42. Gigue`re S, Hernandez J, Gaskin J, et al. Performance of five 26. Flynn O, Quigley F, Costello E, et al. Virulence-associated serological assays for diagnosis of Rhodococcus equi pneumo- protein characterisation of Rhodococcus equi isolated from bovine lymph nodes. Vet Microbiol 2001;78:221–228. nia in foals. Clin Diagn Lab Immunol 2003;10:241–245. 27. Takai S, Martens RJ, Julian A, et al. Virulence of Rhodo- 43. Phumoonna T, Muscatello G, Chicken C, et al. Clinical eval- coccus equi isolated from cats and dogs. J Clin Microbiol uation of a peptide-ELISA based upon N-terminal B-cell 2003;41:4468–4470. epitope of the VapA protein for diagnosis of Rhodococcus equi 28. Tkachuk-Saad O, Lusis P, Welsh RD, et al. Rhodococcus pneumonia in foals. J Vet Med B Infect Dis Vet Public equi infections in goats. Vet Rec 1998;143:311–312. Health 2006;53:126–132. 29. Boland CA, Meijer WG. The iron dependent regulatory pro- 44. Martens RJ, Cohen ND, Chaffin MK, et al. Evaluation of tein IdeR (DtxR) of Rhodococcus equi. FEMS Microbiol Lett 5 serologic assays to detect Rhodococcus equi pneumonia 2000;191:1–5. in foals. J Am Vet Med Assoc 2002;221:825–83.

124 2010 ր Vol. 56 ր AAEP PROCEEDINGS