Haemotropic Mycoplasmas of Cats and Dogs: Transmission, Diagnosis, Prevalence and Importance in Europe
B. Willi; M. Novacco; M.L. Meli; G.A. Wolf-Jäckel; F.S. Boretti; N. Wengi; H. Lutz; R. Hofmann-Lehmann
Read the German translation: Hämotrope Mykoplasmen bei Katze Und Hund: Übertragung, Diagnose, Prävalenz und Bedeutung in Europa
Haemotropic mycoplasmas (aka haemoplasmas) are small (< 1 µm), cell wall-less, discoid-shaped bacteria, closely attached to red blood cells of infected animals. They were formerly known as Haemobartonella and Eperythrozoon species and are the causative agents of infectious anaemia in a variety of mammalian species. Increasing knowledge on the bacterial genomic sequences led to the reclassification of these organisms within the genus Mycoplasma as haemotropic mycoplasmas.1 Infections with haemoplasmas can induce acute haemolysis, associated with anorexia, lethargy, dehydration, weight loss and sudden death of infected animals. The inability to culture these agents in vitro has limited the possibilities to investigate haemotropic mycoplasmas. However, species-specific PCR assays developed in recent years allow assessing the pathogenesis and epidemiology of haemoplasma infections in cats and dogs in more detail.
Haemotropic Mycoplasmas in Cats
Early studies described two distinct feline haemoplasma species: the Ohio isolate (large form) and the California isolate (small form) of Haemobartonella felis.2-4 These isolates were subsequently renamed as Mycoplasma haemofelis (Mhf)1 and 'Candidatus Mycoplasma haemominutum' (CMhm).5 In 2002, a third haemotropic Mycoplasma species was identified in a privately owned Swiss cat that presented with haemolytic anaemia; this third species was designated 'Candidatus Mycoplasma turicensis' (CMt).6,7 Recently, a fourth species similar to the canine haemoplasma 'Candidatus M. haematoparvum' (CMhp) has been detected in cats in California.8 By applying PCR-based methods, haemoplasma infections have been diagnosed in domestic cats and wild felids worldwide.9 Co-infections with several feline haemoplasmas were reported. In the West and South of Switzerland, feline haemoplasma infections were more prevalent than in the remainder of the country.7
The pathogenic potential of the different feline haemoplasma species varies, and co-factors, such as immunosuppression or pre-existing retroviral infections, may increase the disease severity. In general, Mhf was found to be more pathogenic than CMhm.3,10 CMt may induce mild to moderate anaemia in experimentally infected domestic cats.7,11,12 In naturally infected cats, risk factors for feline haemoplasma infections include male gender, old age, cat bite abscesses, retroviral infection, non-pedigree lineage and outdoor access. The clinical presentation may depend on the stage of the haemoplasma infection.13 Generally, however, it is not well understood why some cats show life-threatening anaemia. Doxycycline, enrofloxacin and marbofloxacin treatment have been shown to reduce feline haemoplasma blood loads and clinical signs in infected cats.14-17 Since some cats treated with high or intravenous doses of enrofloxacin may develop acute blindness from retinotoxicity, doxycycline (10 mg/kg/day PO) represents the antibiotic of choice to treat feline haemoplasma infection. Esophagitis and esophageal strictures may occur subsequent to oral doxycycline treatment; thus, it is recommended to provide water or food following the tablet. To date, it is assumed unlikely that antibiotic treatments completely eliminate haemoplasma infections and infected cats may become asymptomatic carriers for months or years. However, reactivation of clinical disease seems rare.3
Haemotropic Mycoplasmas in Dogs
Haemobartonella canis infections have been sporadically reported in dogs with anaemia. Subsequently, the agent was reclassified as Mycoplasma haemocanis (Mhc). In addition, CMhp has been discovered in an anaemic splenectomised dog.18,19 Both agents seem to exhibit worldwide distribution, but only limited prevalence data based on molecular detection methods are yet available. Using specific PCR assays, a higher prevalence was recently reported in European countries with Mediterranean and sub-Mediterranean climates when compared to Switzerland.20-22 Additionally, in some populations young animals and male dogs seemed more susceptible to canine haemoplasma infections than adult and female dogs, respectively.20 Severe haemolytic anaemia has only occasionally been described in haemoplasma-infected dogs, mainly in immune-compromised or splenectomized animals. Furthermore, mange infection was associated with canine haemoplasma infections.20 Recent PCR-based investigations in Europe also support the low pathogenic potential of these agents.20,21 Most haemoplasma infected dogs present with chronic, asymptomatic infections. These animals seem unable to clear the infection. In a recent study using quantitative real-time PCR, all samples collected from three infected dogs throughout a follow-up period of up to 13 months tested positive.21 As described for other haemoplasma infections, antibiotic treatment may be unable to eliminate canine haemoplasma infections completely, but was found to reduce clinical signs of infection.
The natural mode of transmission of feline and canine haemoplasmas has not been definitely elucidated. Blood transfusions have been reported as a source of Mhf and CMhm infections.13,23 Furthermore, blood-sucking arthropods may be involved in the transmission of feline and canine haemoplasmas. Mhf and CMhm DNA was detected in the cat flea, Ctenocephalides felis, and in flea faeces, but a recent experimental transmission study for Mhf and CMhm via C. felis was not conclusive.24-29 In Switzerland, no haemoplasma DNA was detected in Ixodes ticks collected from the vegetation.27,30 In contrast, the brown dog tick, Rhipicephalus sanguineus, is likely to play a role as a vector and reservoir for canine haemoplasmas.31 In Europe, this tick species is commonly encountered in areas with Mediterranean and sub-Mediterranean climates, which parallels the high prevalence of canine haemoplasma infections found in these countries.20,22 Some feline haemoplasmas were detected in saliva and faeces of infected cats12,27,32 indicating that direct transmission of haemoplasmas may be important. However, a recent in vivo study that modeled CMt transmission via social contact among cats was unable to infect cats by transmission of CMt PCR-positive saliva.12 In contrast, subcutaneous inoculation of PCR-positive blood resulted in infection, which may indicate that aggressive interaction is necessary to transmit CMt between cats.12 This parallels observations for Mhc, for which the prevalence was higher in Japanese fighting dogs compared to other breeds.33
Specific conventional and quantitative real-time TaqMan PCR systems have been introduced and are now considered the gold standard for the detection and differentiation of feline and canine haemoplasma species.9,21,30,34-36 No in vitro culture system has been established to date to propagate feline and canine haemoplasmas outside their hosts. The light microscopic investigation of Giemsa-stained blood smears from infected animals was shown unreliable. In particular, light microscopy is unfit to diagnose CMt infection because of the frequently low CMt blood loads. Most recently, we were able for the first time to demonstrate CMt using electron microscopy. Serological assays are not yet routinely available.
Prevalence and Importance in Europe
Feline haemoplasma infections were found in all investigated populations in Europe. In Switzerland, canine haemoplasma infections are less prevalent particularly than in countries with a Mediterranean climate.20,21 The infected dogs in Switzerland had either been imported from or visited regions where R. sanguineus is indigenous.21 These observations support the hypothesis that canine haemoplasmas may be indirectly transmitted by blood-sucking arthropods, in particular those that rely on a warm climate for their survival. In a recent study on the situation in Europe, we identified several risk factors for canine haemoplasma infections, e.g., living in kennels, young age, and non-pedigree lineage.20 A higher prevalence of canine haemoplasma infections in kennel-kept dogs may occur due to the fact that dogs in kennels are often group-housed, which could increase the risk of a direct haemoplasma transmission among dogs and their risk of exposure to fleas and ticks.
Haemoplasma infections may lead to severe haemolytic anaemia in infected individuals; the disease may become fatal if the infection is not recognized in time and treated adequately. The close relationship between CMt and rodent haemoplasmas as well as among certain feline and canine haemoplasmas suggests a potential interspecies transmission of these agents.7,18 While CMt has not yet been detected in rodents,27 CMhm infection has been reported in a dog in China.37 Moreover, Mhf has recently been detected by PCR in the blood of a human AIDS patient,38 and PCR-positive results for porcine haemoplasma-like organisms have been obtained from blood of Chinese farm workers and swine veterinarians.39-41 This substantiates the risk of an interspecies transmission of haemoplasmas. Nevertheless, so far the zoonotic potential of haemoplasma infections has been largely neglected.
Feline haemoplasma infections are commonly encountered in cats in Switzerland and other European countries. As important differences exist in the pathogenic potential of the feline haemoplasma species, a species differentiation by molecular methods is pivotal. However, molecular methods are unable to distinguish between acute and chronic infections. Canine haemoplasma infections are usually only encountered in Switzerland in dogs with a travel history, but were common in dogs from countries with Mediterranean climate. The climate and living conditions (i.e., living in kennels) seem to influence the canine haemoplasma prevalence. The agents usually exhibit a low pathogenic potential, and most infected dogs are asymptomatic carriers. However, severe immunosuppression or splenectomy has been shown to precipitate disease and can result in life-threatening haemolytic anaemia.
To date, diagnosis of haemoplasma infections relies on the molecular detection of the agents in the blood. As antibiotic therapy can result in false-negative results, blood samples for PCR analyses must be collected before antibiotic therapy is initiated. Doxycycline, enrofloxacin and marbofloxacin have been reported to reduce clinical sings and bacterial loads in feline haemoplasma infections, although a clearance of infection by antibiotic therapy seems unlikely. No experimental studies on the efficacy of antibiotic treatment for canine haemoplasma infections have been published, but in analogy to cats, doxycycline may be recommended to treat dogs with signs of hemoplasmosis. Transmission seems to rely on contact with infected blood and blood-sucking arthropods, and aggressive interactions between cats or dogs are thought to play a role. Therefore, strict flea and tick control and avoidance of aggressive interactions may be important to prevent the spread of the agents. Moreover, blood donors should be tested by PCR for haemoplasma infections prior to blood transfusion. Finally, according to PCR-based reports of the occurrence of Mhf and porcine haemoplasma-like organisms in humans, the zoonotic potential of haemotropic mycoplasmas needs to be further addressed.
1. Neimark, et al. Int J Syst Evol Microbiol 2001;51:891.
2. Berent, et al. Am J Vet Res 1998;59:1215.
3. Foley, et al. Am J Vet Res 1998;59:1581.
4. Messick, et al. J Clin Microbiol 1998;36:462.
5. Foley, Pedersen. Int J Syst Evol Microbiol 2001;51:815.
6. Willi, et al. Schweiz Arch Tierheilkd 2006;148:139.
7. Willi, et al. J Clin Microbiol 2005;43:2581.
8. Sykes, et al. J Vet Intern Med 2007;21:685.
9. Willi, et al. J Clin Microbiol 2007;45:1159.
10. Westfall, et al. Am J Vet Res 2001;62:687.
11. Tasker, et al. Vet Microbiol 2009;139:323.
12. Museux, et al. Vet Res 2009;40:45.
13. Willi, et al. J Clin Microbiol 2006;44:961.
14. Dowers, et al. J Am Vet Med Assoc 2002;221:250.
15. Tasker, et al. Microbes Infect 2006.
16. Tasker, et al. Vet Microbiol 2006;117:169.
17. Tasker, et al. J Microbiol Methods 2004;56:63.
18. Sykes, et al. J Am Vet Med Assoc 2004;224:1946.
19. Messick, et al. Int J Syst Evol Microbiol 2002;52:693.
20. Novacco, et al. Vet Microbiol 2009;[Epub ahead of print].
21. Wengi, et al. Vet Microbiol 2008;126:132.
22. Kenny, et al. J Clin Microbiol 2004;42:5397.
23. Gary, et al. J Feline Med Surg 2006;8:321.
24. Kamrani, et al. Can J Vet Res 2008;72:411.
25. Lappin, et al. J Feline Med Surg 2006;8:85.
26. Shaw, et al. Vet Microbiol 2004;102:183.
27. Willi, et al. Appl Environ Microbiol 2007;73:3798.
28. Woods, et al. Am J Vet Res 2005;66:1008.
29. Woods, et al. Am J Vet Res 2006;67:494.
30. Willi, et al. J Clin Microbiol 2009;47:4049.
31. Seneviratna, et al. Res Vet Sci 1973;14:112.
32. Dean, et al. J Feline Med Surg 2008;10:413.
33. Sasaki, et al. Journal of Protozoology Research 2008:57.
34. Peters, et al. Vet Microbiol 2008;126:142.
35. Barker, et al. Vet Microbiol 2010;140:167.
36. Tasker, et al. J Clin Microbiol 2003;41:439.
37. Zhuang, et al. Trop Anim Health Prod 2009;41:689.
38. dos Santos, et al. Emerg Infect Dis 2008;14:1922.
39. Duarte, et al. J Infect Dis 1992;165:976.
40. Yuan, et al. Am J Vet Res 2009;70:890.
41. Hu, et al. Emerg Infect Dis 2009;15:1139.