Leptospirosis and Tularaemia in Raccoons (Procyon lotor) of Larimer Country, Colorado
Summary
Raccoons (Procyon lotor) are commonly implicated as carriers of many zoonotic pathogens. The purpose of this cross-sectional study was to look for Leptospira interrogans and Francisella tularensis in opportunistically sampled, free-ranging raccoons of Larimer Country, Colorado, USA. Sixty-five animals were included in the study and testing consisted of gross post-mortem examination, histopathology, and both immunohistochemistry and PCR for L. interrogans and F. tularensis. No significant gross lesions were identified and the most common histological lesions were lymphoplasmacytic interstitial nephritis and pulmonary silicosis; rare periportal hepatitis, splenic lymphoid hyperplasia and small pulmonary granulomas were also identified. Of 65 animals, 20 (30%) were positive for Leptospira on IHC but only one by PCR. Animals with inflammation in their kidneys were seven times more likely to be positive for Leptospira than animals without inflammation. The severity of inflammation was variable but often mild with minimal associated renal pathology. One animal was positive for Francisella on both IHC and PCR; IHC staining was localized to histiocytic cells within a pulmonary granuloma. In Colorado the significance and epidemiology of Leptospira is poorly understood. The high prevalence of infection in raccoons in this study population suggests that this species may be important in the regional epidemiology or could be used to estimate risk to domestic animals and humans. Identification of a single Francisella positive animal is significant as this is an uncommon disease in terrestrial animals within the state; the apparently higher prevalence in this peridomestic species implies that raccoons may be good indicators of the pathogen in the region. The results of this study suggest that raccoons may serve as effective sentinels for both Leptospira and Francisella in the state of Colorado. Further studies are needed to better characterize the prevalence and epidemiology of both organisms within the region.
Impacts
- •
There is a high prevalence of Leptospira interrogans and low prevalence of Francisella tularensis infection in this group of opportunistically tested Colorado raccoons.
- •
Immunohistochemistry may be a more sensitive test for leptospirosis in raccoons compared with PCR.
- •
Raccoons may be good sentinels for zoonotic pathogens in the state of Colorado.
Introduction
Wild animals are often implicated in the maintenance and spread of infectious organisms; unfortunately however, their role in disease is often poorly understood due to a lack of data. With growing awareness of emerging and zoonotic disease, there has been increased attention on the role of free-ranging animals in the epidemiology of infectious diseases affecting both man and domestic animal species (Kruse et al., 2004).
Leptospirosis, caused by the spirochete Leptospira interrogans, is widespread and considered to be one of the most important zoonoses in the world; outbreaks and endemic disease have been identified in urban and rural populations of both industrialized and developing countries (Vinetz, 2001; Bharti et al., 2003). The epidemiology of leptospirosis is not completely understood owing to its complex ecology involving many different serovars and reservoir host species. This complexity precludes the use of universal control measures and dictates the need to evaluate the epidemiology of leptospirosis at a smaller regional level.
Wild animals are widely accepted to be important reservoirs of Leptospira (Levett, 1999; Ward et al., 2002). Serosurveys of numerous North American wild mammals suggest a history of exposure to the organism with subsequent seroconversion; based on prevalence of infection, small and intermediate sized mammals including skunks (Ferguson and Heidt, 1981), raccoons (Alexander et al., 1972; Bigler et al., 1975), rats (Thiermann, 1981), opossums and foxes (Kingscote, 1986) are often implicated as important maintenance hosts.
Tularaemia is a zoonotic disease caused by the bacterium Francisella tularensis. In recent years the organism has been identified in new locations and new hosts to the extent that it has been classified as emerging or re-emerging (Petersen and Schriefer, 2005). Owing to its pathogenicity, ease of dissemination and potential to cause widespread public health problems it has been listed as a ‘category A’ bioterrorism agent by the Centers for Disease Control and Prevention.1 While two patterns of disease are most commonly described, a terrestrial cycle with type A Francisella and an aquatic cycle with type B Francisella, the organism has a broad list of hosts and many wild animal species are important in its ecology (Morner, 1992).
Raccoons are well adapted to urban areas and are present in high numbers close to human residences. Given their abundance, wide geographic range and ability to harbour numerous zoonotic pathogens, raccoons have been proposed as a valuable indicator of regional zoonoses including rabies, tularaemia, arboviruses, parasites and enteric bacteria (Bigler et al., 1975; Sorvillo et al., 2002; Dietrich et al., 2005; Krebs et al., 2005; Berrada et al., 2006). In Colorado, raccoons are often trapped and humanely euthanized when they are an urban nuisance or exhibit signs of disease; as such carcasses are available for disease surveillance activities. The objective of this study was to identify the presence and frequency of L. interrogans and F. tularensis infection within a sample of opportunistically tested raccoons in Larimer country, Colorado.
Materials and Methods
Tissues from raccoons in Larimer Country presented to the Colorado Division of Wildlife for diagnostic testing, the Larimer Humane Society for disposal following vehicle associated trauma or USDA-APHIS-WS for research between September 2006 and December 2007 were included in the study. Post-mortem examinations were conducted on all animals. From each animal, fresh and formalin fixed kidney, spleen, lung and liver were collected; fresh tissues were stored at −20°C until sample collection was complete and diagnostic testing was conducted. Descriptive and comparative statistics (odds ratios) were conducted using commercially available software.2
Histopathology and immunohistochemistry
Tissues were fixed in 10% buffered formalin, paraffin-embedded and tissue blocks were sectioned at 5 μm and stained with haematoxylin and eosin for histopathological evaluation. Pathologists were blinded to IHC and PCR results at the time of slide review.
Immunohistochemical staining was performed using standard bench-top techniques. Briefly, 5 μm sections were cut and mounted on positively charged slides. The sections were deparaffinized then rehydrated with descending alcohol concentrations to buffer. Sections were incubated with primary antibody for 1 h at room temperature. For Leptospira the primary antibody was a cocktail of six antisera: canicola–Hond Utrecht IV, grippotyphosa-Andaman, hardjo-Hardojoprajtino, copenhageni-M20A, Pomona-Pomona, and Bratislava-Jez Bratislava (NVSL; Ames, IA, USA) diluted 1 : 1000. For Francisella the primary antibody was a rabbit antisera (240939; BD Diagnostic systems, Sparks, MD, USA) diluted 1 : 100. A pre-diluted dual-link HRP-conjugated secondary antibody (K4061; Dako, Carpinteria, CA, USA) and DAB+ (K3467; Dako) was utilized to detect the immunoreactive complexes. The slides were then counterstained with Mayer’s haematoxylin and mounted using xylene based permanent media. Leptospira positive canine kidney slides and tularaemia positive beaver lung slides were used for positive tissue controls. Primary antibody was replaced with antibody dilutant for negative antibody controls.
Polymerase chain reaction
The Leptospira PCR protocol used to detect six pathogenic serovars of L. interrogans has been previously described (Woo et al., 1997; Harkin et al., 2003). Briefly, 5 μl of DNA extract, prepared using the Qiagen DNA mini kit3 per manufacturer’s instructions, was utilized as template in a PCR reaction mixture containing 2.5 units of heat-activated taq polymerase, 1.5 mm MgCl2, and 1 μm forward and reverse primers. Cycling conditions include one cycle of 95°C for 5 min, 35 cycles of 94°C for 45 s, 52°C for 45 s and 72°C for 45 s followed by a final extension at 72°C for 5 min. Amplicons were analyzed by ethidium bromide staining on a 1.5% agarose gel. PCR was done on kidney only.
The F. tularensis real-time PCR assays were performed using a Cepheid Smart Cycler4 real-time PCR platform adapted using previously described primer and probe sequences designed with a dual-labelled FRET probe chemistry (Versage et al., 2003). Primers and corresponding probes targeting three specific genetic sequences of F. tularensis, ISFtu2, 23 kDa and tul4, were all optimized for use on the Smart Cycler platform. Quantitative standards were designed for all three target genes and revealed a strong analytical sensitivity and equivocal minimum detection limit of 10 copies in all three targets. Approximately 1 g of pooled spleen, liver and kidney were homogenized and DNA was isolated using the Qiagen DNA mini kit5 following manufacturer’s instructions. Real-time PCR was carried out using the Qiagen QuantiTect Probe kit6 with 5 μl of DNA extract, heat-activated taq polymerase, 4 mm MgCl2, 0.4 μm of tul4 forward and reverse primers and 0.1 μm of tul4 probe. Cycling conditions were one cycle at 95°C for 15 min followed by 45 cycles at 95°C for 10 s and 60°C for 30 s.
Results
Tissues were collected from 65 raccoons during the period of the study. None of the animals included were noted to be clinically diseased and the only significant gross lesions noted on post-mortem examination were related to vehicular trauma. The most common histological changes were observed in the lungs. Pulmonary silicosis with occasional foci of mineralization was identified in 25 (38%) of the animals. Four animals had small, multifocal to coalescing pulmonary graunulomas that contained numerous eosinophils and were often associated with blood vessels. In the liver, seven (11%) animals had bile stasis and mild periportal, lymphoplasmacytic hepatitis; small microgranulomas were observed in the livers of 6 (10%) animals. The most common changes observed in the spleen were mild lymphoid hyperplasia (seven animals, 11%) and amyloid deposition within follicles (five animals, 8%). Inflammation was observed in the kidney of 21 (32%) animals. Inflammation was characterized as interstitial, lymphoplasmacytic and ranged from mild to severe. In three animals there was also mineral present within renal tubules and a single animal had an associated tubulitis with tubular degeneration.
Leptospirosis
Of all raccoon kidney sections, 20 (30%) were positive for Leptospira by IHC. Animals with inflammation in their kidneys were seven times more likely to be IHC positive than those without inflammation (OR 7, 95% CI 2–22). Only a single animal was positive for L. interrogans by PCR, this animal was also positive on IHC with moderate staining but devoid of significant nephritis.
Tularaemia
A single animal had strong IHC staining for F. tularensis in the lung that was associated with small, multifocal to coalescing eosinophillic and histiocytic pulmonary granulomas. Staining was localized to histiocytic cells but not other inflammatory cells or pneumocytes. This animal was the only individual positive by PCR.
Discussion
Results of this study provide useful information regarding the presence of L. interrogans and F. tularensis in Colorado raccoons. Leptospires have been isolated from the kidneys of raccoons with a variable amount of renal pathology (Mc et al., 1958; Reilly, 1970; Mikaelian et al., 1997; Hamir et al., 2001; Richardson and Gauthier, 2003). The most commonly reported pathological lesions are multifocal, chronic interstitial nephritis consistent with the pattern of inflammation observed in this study. Serosurveys of raccoons indicate a high prevalence of infection ranging from 11% to 66% (Alexander et al., 1972; Mikaelian et al., 1997; Mitchell et al., 1999; Richardson and Gauthier, 2003; Bischof and Rogers, 2005). Leptospires are proposed to persist within a region due to infection of maintenance or reservoir host species. Maintenance hosts are defined as species in which infection is endemic and the bacteria persist for long periods, most commonly in renal tubules of the kidney and are shed in urine (Babudieri, 1958). Infection of maintenance hosts is commonly acquired at an early age, usually by direct transmission, resulting in high prevalence of infection within the local species population; disease in these hosts is often chronic and/or asymptomatic (Levett, 2001). Incidental or accidental hosts are those humans or animals that do not classically or reliably maintain the organism in the environment but are susceptible and become diseased, usually more acutely and severely than the associated reservoir host. Like maintenance hosts, incidental hosts shed leptospires in urine and can transmit the disease to other animals or humans, however, the incidence of infection and urinary shedding is less than that seen in maintenance hosts (Richardson and Gauthier, 2003). The lack of significant renal pathology observed in this study suggests that Colorado raccoons are most likely a maintenance host for the bacteria.
The distribution of serovars is complex and widely varied (Adler and de la Pena Moctezuma, 2010). There are universal serovars including L. interrogans icterohaemorrhagiae and L. interrogans canicola that can be isolated regularly from all parts of the world and many animal species. In contrast, regional serovars are dictated largely by available maintenance hosts. Worldwide, there is marked variation in maintenance hosts and the serovars that they carry. In order to understand the epidemiology of leptospirosis within a region one must be aware of the prevalent regional serovars and the maintenance hosts that carry them. While both the PCR and IHC techniques employed in this study were capable of detecting a combination of serovars to increase the sensitivity of the assay, specific serovars were not elucidated. The high prevalence of infection suggests that raccoons may be useful sentinels for follow-up studies focused on the identification of specific serovars within Colorado.
In this study, inflammation in the kidney was significantly associated with IHC immunoreactivity for Leptospira. While lymphoplasmacytic inflammation in the kidney is a relatively non-specific finding, this information is useful for diagnosticians who should be aware that follow-up testing for leptospirosis may be warranted. In this study the PCR failed to identify most animals classified as positive by IHC. Based upon the pattern of IHC staining which is morphologically consistent with colonization of tubules by spirochetes, the discordant results between IHC and PCR are interpreted as PCR false negatives. Molecular diagnostics are often impacted by the presence of inhibitory substances resulting in lower diagnostic sensitivity and specificity. It has been previously shown that the presence of high urea content in urine and presumably kidney tissue, may result in inhibition of the PCR amplification process (Khan et al., 1991). It is possible that the presence of high urea salts in tissue specimens may have impacted PCR results in this study. Alternatively only a small piece of tissue is used for molecular diagnostics and given the patchy distribution of organism in some samples the lesion may be missed in the utilized sample.
The identification of a single animal with pulmonary pathology that was positive for tularaemia by IHC and PCR is interesting. In Colorado, tularaemia infections are reported but uncommon; less than one case per million people are reported annually (E.W. Lawaczeck, 2009 personal communication). While sample sizes are low, the higher apparent prevalence in raccoons of this study relative to humans suggest that using these animals as sentinels may be a useful way to characterize the distribution of the bacterium in Colorado and potentially estimate regional risk to humans. In Martha’s Vineyard, Massachusetts, where tularaemia is endemic in humans and animals, raccoons were proposed as a key element of the regional surveillance system given their high seroprevalence (Berrada et al., 2006). The host range of F. tularensis is broad and transmission may occur through biting insect vectors or contact with the organism through infected animals or contaminated environments (Morner, 1992; Petersen and Schriefer, 2005). In animals, the spectrum of clinical disease is broad and multiple systems are infected (Feldman, 2003). Foci of pulmonary inflammation and necrosis with positive IHC staining for the organism have also been described in aborted sheep (O’Toole et al., 2008). Given that all animals in the study were euthanized or killed by vehicle-associated trauma and that in the single positive raccoon there were no gross lesions, the infection was likely clinically insignificant at the time of the animal’s death.
Other lesions identified histologically in these animals included follicular amyloid in the spleen, eosinophilic pulmonary granulomas and pulmonary anthracosis. Splenic amyloid is a non-specific finding secondary to immune stimulation, a change that has been previously described in raccoons (Hamir et al., 2007). Eosinophilic pulmonary granulomas are commonly associated with parasitic infection and/or migration in raccoons. Pulmonary anthracosis or ‘dust granulomas’ are small foci of inflammation containing phagocytosed particles that are commonly observed in the lungs of humans and animals exposed to airborne organic material and is not surprising given the peridomestic nature of raccoons.
The results of this study lend further support to the concept of using wildlife, specifically raccoons, as sentinels for zoonotic pathogens. Such sentinels may be helpful for defining the regional distribution and prevalence of these organisms such that human or domestic animal risks could be mitigated. In the case of leptospirosis, further investigation into the identification of specific serovars may be useful for implementation of appropriate control strategies. In this study sampling was opportunistic and specific geographic locations where dead raccoons were found were not available for individual animals; as such, extrapolation of results beyond Larimer Country should be made with caution. Future studies characterizing infection prevalence and distribution on a larger scale as well as correlation of wildlife data with information on domestic animal and human cases is required to better understand the role of raccoons in the epidemiology of these zoonotic pathogens within the state.
Footnotes
Acknowledgements
The authors would like to acknowledge the assistance of Larimer Humane Society and Tricia Fry of the USDA-APHIS-WS for assistance in the procurement of raccoon tissues. Funding for this study was provided by the Rocky Mountain Regional Center for Excellence.
