Mycoplasmosis and Upper Respiratory
Tract Disease of Tortoises
Upper Respiratory Tract Disease (URTD): Anatomic Components, Definition,
and Possible Causes:
The major component of the upper respiratory tract of tortoises is a large nasal cavity located cranial to the eye (Jacobson et al. 1991; Figure 1; Figure 2; Figure 3). The external nares open into
a short vestibule that is continuous with a recess ventrally and with olfactory chambers dorsally.
The ventral recess continues into the nasal passageway that opens into the roof of the mouth as the choanae. By definition, disease is “any deviation from or interruption of the normal structure or function of any part, organ, or
system (or combination thereof) of the body that is manifested by characteristic set of symptoms and signs and whose etiology, pathology, and prognosis may be known or unknown” (Dorland’s Illustrated Medical Dictionary 1985).
While hypovitaminosis A was considered a predisposing problem in certain tortoises, Fowler (1980) discussed the differences between URTD and hypovitaminosis A. Several agents have been hypothesized to cause respiratory tract disease in tortoises, including viruses (Jackson and Needham 1983), Mycoplasma spp. (Fowler 1980; Lawrence and Needham 1985), and Pasteurella testudinis (Snipes et al. 1980; Snipes and Biberstein 1995; Snipes et al. 1995). In Europe, Sendai virus was considered a possible cause of rhinitis in captive spur-thighed (Testudo graeca) and Hermann's tortoises (T. hermanni) (Jackson and Needham 1983), but a later study (Lawrence and Needham 1985) found no increase in antibody titers against Sendai virus over a 3-month period.
Mycoplasmosis of Desert and Gopher Tortoises: Rhinitis and chronic URTD have been reported in a variety of species of wild and captive tortoises in the United States (Jacobson et al. 1991) and Europe (Jackson 1980). In the 1980s, major declines (33-76% over 10 yr) of desert tortoises (Gopherus agassizii) were documented at several sites in the western Mojave Desert of California, USA, and at one site in the eastern Mojave (Corn 1994). Tortoises with clinical signs of URTD were observed among affected populations at several sites (Knowles 1989; Berry 1990). As a result of the declines, desert tortoises in the Mojave Desert north and west of the Colorado River were declared threatened (U.S. Fish and Wildlife Service 1990). Beginning in 1989, efforts were undertaken to determine the etiology of URTD in desert tortoises (Jacobson et al, 1991). By electron microscopy, pleomorphic organisms resembling Mycoplasma sp. were seen on cell surfaces and tightly adhered to cell membranes of ill tortoises (Figure 4; Figure 5). The Mycoplasma was isolated and determined to be a new species, named Mycoplasma agassizii (Brown et al. 1995). They are quite pleomorphic, ranging from 300 to 900 nm (Figure 6). A monoclonal antibody was produced against desert tortoise light chain of IgY and IgM. An enzyme-linked immunosorbent assay (ELISA) to detect antibodies against the Mycoplasma in plasma and serum samples was developed (Schumacher et al. 1993), and experiments were undertaken to determine if M. agassizii caused URTD. The most stringent requirement for definitive proof of a causative relationship between an infectious agent and a disease is the fulfillment of the Henle-Koch-Evans postulates (Evans 1976a, 1976b). Thus, three controlled experimental infection cohort studies were performed to fulfill the Henle-Koch-Evans postulates and additionally, to determine sensitivity, specificity, and predictive values of diagnostic assays for mycoplasmal infections of tortoises (Brown et al. 1994; Brown et al. 1999; McLaughlin 1997; and summarized in Brown et al. 2002). The disease was induced by inoculation of tortoises with pure cultures of the Mycoplasma, but not Pasteurella testudinus (Brown et al. 1994). Histologically, the lesions were consistent with those seen in the previously examined naturally infected tortoises. A polymerase chain reaction (PCR) test was developed to detect nucleotide sequences of the 16s rRNA gene of the bacteria in nasal flush and swab samples (Brown et al. 1995). Although URTD has been seen in captive gopher tortoises (ERJ, unpublished data), the first documentation of the disease in wild gopher tortoises was in 1989, when an epizootic of URTD was documented on Sanibel Island, Lee County, Florida, USA (G.S. McLaughlin and M.S. Elie, unpublished data), during the course of an ecological study (McLaughlin 1990). When tested by ELISA, >80% of the adult tortoises from Sanibel Island were seropositive for antibodies against M. agassizii (Beyer 1993).
Due to the 1979 listing by the then Florida Game and Fresh Water Fish Commission of the gopher tortoise as a species of special concern, and the subsequent permitting of over 450 relocations involving more than 8000 tortoises (J.E. Berish, personal communication), particular attention was focused on the dynamics and persistence of both natural and relocated populations during the late 1980’s (Cox 1989). The observation of URTD on Sanibel Island and the association of mycoplasmosis with decline of certain desert tortoise populations in the Mojave Desert elicited concerns regarding declining and isolated populations of gopher tortoises. Because an understanding of the effects of URTD on both individuals and populations is essential for proper management of remaining populations, a study was begun in 1993 at the University of Florida on the etiology, pathology, and diagnosis of URTD in gopher tortoises. The gopher tortoise is currently listed as a Threatened species (since 2007) by the Florida Fish and Wildlife Conservation Commission, and estimates of legally relocated tortoises over the last several decades exceed 70,000 (J.E. Berish, personal communication).
Contrary to what was recently reported elsewhere (Sandmeier et al., 2009), only two mycoplasmas have been isolated and named from the desert tortoise. An original isolate from a desert tortoise with URTD was named Mycoplasma agassizii (Brown et al. 1994, 1995, 2001). Experimental transmission studies have confirmed this Mycoplasma as a cause of URTD in desert and gopher tortoises (Brown et al. 1994; Brown et al. 1999b). A second, genetically distinct Mycoplasma has also been isolated from desert tortoises and from gopher tortoises in northeastern Florida and has been named M. testudineum (Brown et al, 2002). Transmission studies with this Mycoplasma in gopher tortoises are ongoing (Wendland, personal communication). Contrary to what was presented in areview of URTD in desert tortoises (Sandmeier et al, 2009), a third Mycoplasma , M. testudinis, was not isolated from a desert tortoise, but instead was isolated from the cloaca of a Greek tortoise in England (Hill, 1985). M. testudinis has not been associated with URTD. Other mycoplasmas from tortoises will probably be described in the future.
Hosts for Chelonian Mycoplasmosis: To date, mycoplasmas have been recovered by culture or detected by PCR in the following species of chelonians: desert tortoise (Gopherus agassizii), gopher tortoise (G. polyphemus), Texas tortoise (G. berlandieri), Chaco (Argentine) tortoise (Geochelone chilensis), leopard tortoise (G. pardalis), Indian star tortoise (G. elegans), African spurred tortoise (G. sulcata), radiated tortoise (G. radiata), red footed tortoise (G. carbonaria), Travancore tortoise (Indotestudo forstenii), spider tortoise (Pyxis arachnoides), flat-tailed tortoise (P. planicauda), spurred-thighed tortoise (Testudo graeca), marginated tortoise (T. marginata), Egyptian tortoise (T. kleinmanni), Russian tortoise (T. horsfieldii), Hermann's tortoise (T. hermanni), and eastern box turtles (Terrapene carolina carolina) in the United States (Brown et al. 2002; Wendland et al. 2006) and captive Testudo sp. in the United Kingdom (Soares et al. 2004).
Distribution in the Wild: Known to occur in wild gopher tortoises in Florida, USA, and in wild desert tortoises in California, Nevada, Utah, Arizona, USA and Testudo graeca in France.
Ages Affected: Primarily seen in adult tortoises in the wild, but under experimental conditions all age groups are susceptible.
Clinical Signs vs. Symptoms; Clinical vs Subclinical Infection: Symptoms are often confused with clinical signs in non-medical journals (see Sandmeier et al. 2009). Symptoms are sensations experienced by a human. Clinical signs are abnormalities we observe in an animal. Regarding URTD, major clinical signs include palpebral edema, conjunctivitis, and rhinitis. Clear serous to tenacious mucous may be seen bubbling from nares (Figure 7). Mycoplasmosis can occur as a subclinical infection (Jacobson et al. 1995). A subclinical infection is one in which lesions may be present in a tissue or organ without the animal manifesting any clinical signs of disease. In other words, the animal, while infected, may appear clinically normal. Special diagnostic tests may be needed to distinguish between a healthy animal and one with subclinical disease. We have seen an annual cycle of convalescence and recrudescence of clinical signs in some captive desert and gopher tortoises. Mycoplasmal diseases in other hosts also can exist as chronic, subclinical infections, with recurrence of clinical signs and increases in transmission potential when the host is stressed (Simecka et al. 1992). Thus, this pattern is not unique to mycoplasmosis in tortoises.
Mycoplasmosis and Modern Concepts of Infectious Disease: Mycoplasmosis refers to any disease caused by infection with Mycoplasma spp. Previous concepts of microbial pathogenicity do not take into account that both the microorganism and the host contribute to microbial pathogenesis. Recently, a damage-response framework of microbial pathogenesis has been proposed as a new theoretical approach to understand microbial pathogenesis (Casadevall and Pirofski, 2003). The three tenets of this framework are: “1. Microbial pathogenesis is an outcome of an interaction between a host and a microorganism. 2. The host-relevant outcome of the host-microorganism interaction is determined by the amount of damage to the host; and 3) Host damage can result from microbial factors and/or the host response”. In chelonian mycoplasmosis, the host response to pathogenic Mycoplasma accounts for much of the pathological change seen in the nasal cavity of the host. Damage is enhanced by strong immune responses of the host. Also, as discussed below, predisposing and contributing factors are probably involved in epizootics of mycoplasmosis in tortoises and other animals.
Predisposing and Contributing Factors: Although data are lacking, we suspect that various extrinsic and predisposing factors are involved in outbreaks of URTD (Jacobson et al, 1991; Brown et al, 2002; Sandmeier et al, 2009). Clinically silent infections may become exacerbated by environmental stress (Brown et al, 2002). Environmental pertuberations may influence the periodicity of outbreaks in populations of tortoises known to have mycoplasmosis. Annual fluctuations in temperature, rainfall, and forage availability may be sufficient to cause detectable outbreaks in an infected population. Increased morbidity and mortality may occur in times of unusually severe environmental stress, such as prolonged drought, hurricanes, excessive rainfall with flooding of burrows, or very cold winters. The data available for desert tortoises indicate that drought is a natural part of the desert tortoise’s environment, but when combined with disease or habitat loss, may contribute to additional disease problems and mortality (Peterson 1996). Clinical signs of URTD, lower leukocyte counts, positive nasal cultures of M. agassizii, and mild to moderate azotemia were more commonly seen in 1993–94, a year of below-normal annual and winter precipitation (Christopher et al 2003). Heteropenia has been associated with drought and starvation (Berry et al. 2002). Tortoises entering hibernation in a drought year may be physiologically compromised, because clinical signs of URTD and heteropenia were noted at the time of emergence from hibernation in 1990–91 and 1994–95, years following a period of drought (Christopher et al, 2003). Most dehydrated tortoises and most deaths at the Desert Tortoise Natural Area and Ivanpah also occurred in 1990–91 and 1994–95, years following dry winters. Human impacts on tortoises and their habitat, whether through disruption of normal behavior patterns, degradation of habitat through agriculture, silviculture, mining or development operations, or pollution, may cause sufficient physiological stress to trigger proliferation of the Mycoplasma and recurrence of signs. Capturing and transporting of tortoises during relocation, restocking and repatriation efforts also may be significant sources of stress that result in overt disease. The release of ill captive tortoises may be a significant factor accounting for the presence of URTD in certain populations. For instance, the release onto Sanibel Island of gopher tortoises originating in northern Florida and southern Georgia following tortoise races has been documented. Many of these tortoises were kept under very poor husbandry conditions that would have allowed transmission of various pathogens (Dietlein and Smith, 1979). A Brochure with photographs with clinical signs has been developed for the gopher tortoise and a link to this document will be available in the near future. Use Acrobat Reader to view this brochure.
Pathologic Diagnosis: On a light microscopic level, the ventral recess consists of a mucous and ciliated epithelial mucosa (Figure 8), while the olfactory chambers consist of a multilayered olfactory epithelium (Figure 9). Disease is a change in structure and/or function of an organ and in tortoises with URTD, there is focal to diffuse, minimal to severe, inflammatory changes in the nasal cavity (Figure 10; Figure 11). Basal cell hyperplasia, infiltrates of heterophils and histiocytes, and lymphoid hyperplasia in the submucosa all may be seen. By electron microscopy, mycoplasmas can be seen closely associated with the nasal cavity mucosa. Mycoplasmas are more readily seen in the ventral recess compared to other areas of the upper respiratory tract. The lower respiratory tract consists of the glottis, trachea, trachea, bronchi, and paired lungs. Rarely is the lower respiratory tract affected in tortoises with URTD.
ELISA Diagnostics: An ELISA was developed at the University of Florida to measure plasma/serum antibodies that are specific for Mycoplasma (Schumacher et al, 1993). Monoclonal antibodies (MAbs) specific for desert tortoise immunoglobulins were developed to ensure the long-term availability of highly specific secondary antibodies for testing of plasma from desert tortoise populations by ELISA. Figure 1 in Schumacher et al (1993) shows the Western blot reactivity on desert tortoise plasma of the IgY(L)-specific MAb HL673 and the IgY(H)-specific MAbHL665. Lanes 2, 4, and 6 were negative controls in which PBS/A was substituted for primary antibody. The blot (lane 7) shows that MAb HL673 reacted with a single band at approximately 27,000 Da corresponding to desert tortoise IgY(L). Mab HL665 reacted with a single band at approximately 65,000 Da (lane 5) corresponding to desert tortoise IgY(H) (1). Polyclonal anti-Testudo horsfieldii IgY(H) and polyclonal anti-T. horsfieldii IgM(H) were tested on the same immunoblot for their cross-reactivities with desert tortoise plasma to determine whether these antibodies, although available in very limited supply, could serve as polyclonal reference reagents for the newly developed MAbs. Polyclonal anti-T. horsfieldii IgM(H) (lane 1) reacted with one major dark band at approximately 74,000 Da which corresponds to desert tortoise IgM(H). Polyclonal anti-T. horsfieldii IgY(H) (lane 3) reacted with a single band of approximately 65,000 Da which corresponds to desert tortoise IgY(H) (1). MAb HL673 also reacted with desert tortoise IgM(L), as determined by ELISA on IgM-rich fractions of desert tortoise immunoglobulins (Schumacher et al, 1993). The monoclonal based ELISA was validated using experimental transmission studies (see below) with Mycoplasma agassizii in desert tortoises (Gopherus agassizii, Brown et al, 1994) and gopher tortoises (G. polyphemus; Brown et al, 1999b). Gold standards for confirmation of mycoplasmosis used in these studies were clinical signs, Mycoplasma culture, Mycoplasma PCR, and histopathology (Brown et al, 2002). In a recent publication (Sandmeier et al, 2009) the authors claim that past infection studies have selected tortoises with the lowest ‘‘background” levels of antibodies as the ‘‘best” negative control specimens (Schumacher et al., 1993; Brown et al., 1994), which may have introduced bias into the research design”. This assessment is incorrect since in Schumacher et al (1993), pre-challenge plasma samples were used to verify post-challenge seroconversion, and in Brown et al (1994), the control group (see Figure 3 in Brown et al, 1994) did not have the lowest “background” levels.) Since its first development at the University of Florida, the ELISA has been refined, drawing on the accumulation of an immense database of ELISA results from more than 20,000 plasma/serum samples (Wendland et al, 2007). Results of the original ELISA were reported as an enzyme immunoassay ratio (EIA) ratio. An EIA ratio >3 was considered to be a positive result. The ELISA was refined to include more stringent quality control measures and has been converted to a clinically more meaningful titer reporting system, consistent with other diagnostic serologic tests. The ELISA results for 5,954 desert and gopher tortoises were plotted, and a subset of these serum samples (n = 90) was used to determine end-point titers, to establish an optimum serum dilution for analyzing samples, and to construct a standard curve. The relationship between titer and A405was validated using 77 serum samples from known positive (n = 48) and negative (n =29) control tortoises from prior transmission studies. The Youden index, J, and the optimal cut point, c, were estimated using ELISA results from the 77 control sera. Based on this evaluation, the refinement has substantially improved the performance of the assay (sensitivity of 0.98, specificity of 0.99, and J of 0.98), thus providing a clinically more reliable diagnostic test for this important infection of tortoises.
PCR Diagnostics: Limitations associated with detection of tortoise mycoplasmas and a theoretical low limit of detection by culture necessitated the development of an alternative diagnostic test for the presence of Mycoplasma in nasal lavage samples from tortoises. The principle of the PCR test is to synthesize an easily detectable number of DNA copies of a segment of the mycoplasmal chromosome by using Mycoplasma-specific primers. The segment of the M. agassizii chromosome selected for analysis, the 16S ribosomal RNA gene, contains conserved (genus-specific) DNA sequences and intervening variable (species-specific) DNA sequences. Thus a primer pair consisting of 1 genus-specific primer and 1 species-specific primer can very reliably distinguish among organisms. Alternatively, the presence of other Mycoplasma species can be determined by using generic genus-specific PCR primers. However, species identification for such samples usually requires further nucleotide sequence analyses. Nasal lavage samples or cultures can be analyzed for presence of Mycoplasma by PCR. The sensitivity of the assay can be increased by culturing nasal flush samples before analysis. Contamination by other sources of DNA does not interfere with PCR. However, blood, calcium alginate swabs and other unidentified agents can inhibit the reaction. The advantages of diagnosing infections by PCR are noninvasive sampling, direct proof of infection at the time the sample was taken, the reaction is not inhibited by sample contamination with other microbes, the potential short sample turnaround time if the sample is not cultured before analysis, accurate identification of M. agassizii and other species of Mycoplasma, and a theoretical low limit of detection. Disadvantages of PCR include the need for specialized laboratory equipment and sophisticated and meticulous technique, the high cost of special reagents, the potential for false positive results caused by cross-contamination, uninformative samples caused by inhibitory substances in the reaction, and consumption of a portion of the sample during the reaction.
Additional Diagnostic Tests: Recently a polyclonal ELISA was developed for determining exposure of desert tortoise to Mycoplasma agassizii (Hunter et al, 2008). However, in this report, no transmission studies were performed to demonstrate that seroconversion could be detected and no indices of performance such as sensitivity and specificity were provided.
Natural Antibody and Western Blots: In a recent publication (Hunter et al, 2008), natural antibody was reported to occur in desert tortoises. The desert tortoises used in this study originated from a university colony that purportedly was free of Mycoplasma infection. However, little information was presented on the history of this colony, specifically, the Mycoplasma serological status of this colony and the culture and PCR status of nasal lavage specimens from this colony. It is unclear if necropsies were ever performed to assess the histological status of the upper respiratory tract to confirm that the tortoises in the colony were free of lesions in the nasal cavities. While natural antibodies probably exist in chelonians, the well-established polyreactive nature of such antibodies for multiple microbial epitopes seriously confounds the interpretation of the Western Blot data described in Hunter et al (2008). Hunter et al (2008 did not demonstrate that natural antibodies were specific for M. agassizii. It is impossible to distinguish IgM natural antibody in an individual (theoretically developed in the absence of specific extrinsic antigen exposure) from IgM antibodies produced very early in the immune response following specific antigen exposure. We have monitored the IgM antibody response to M .agassizii using IgM specific monoclonal antibodies. Mycoplasma infection of tortoises elicits an IgM antibody response approximately 4 weeks after exposure and shifts to a long-lasting, predominantly IgY-like antigen-specific antibody response approximately 10 weeks after exposure. Natural antibody is polyreactive, produced by B-1 cells and reacts with many epitopes on multiple, mostly unrelated antigens that are often found on multiple potential pathogenic microbes. Thus, such “natural” antibodies in desert tortoises are not M. agassizii-specific.In Sandmeier et al (2009), the authors state: “a Western blot may be used as a confirmatory test to an ELISA”. The authors appear to be unaware of the limitations of Western blots. Extensive literature supports the need for multiple Mycoplasma strains as antigens in Western blot analysis of naturally infected animals. Most mycoplasmas exhibit extensive intraspecies genotypic and phenotypic variability that can be manifested as antigenic variation in the contexts of immune recognition. Further, this occurs even in individual animals infected with a defined isolate. This heterogeneity can confound analysis of mycoplasmal immunogen recognition, especially in Western blots, when only a single isolate is used as antigen (Kittelberger et al, 2006). Hunter et al (2008) hypothesized that differences observed in Western blot patterns were a result of natural antibodies. Unfortunately, lack of experience with mycoplasmas and unfamiliarity with the published literature on mycoplasmal diagnostics has likely led to well-intentioned, but erroneous conclusions. The data presented in Hunter et al. (2008) does not show that any of the bands revealed on Western Blots are indeed due to M. agassizii specific antibodies. The reliance of Hunter et al (2008) on a single strain as antigen in the Western blot is likely to result in false negatives. Importantly, detection of specific antibodies by ELISA is strain-independent, whereas other assays such as Western blot, metabolic inhibition, and complement fixation assays are documented to be strain-dependent. These differences are explained by the location of the antigens (surface exposed, membrane or cytosolic), degree of surface variation, biofunctional assays, and in vivo expression of antigens. Individual variation in the immune response among animals, even to the same strain of M. agassizii, is common in Western blots. An individual tortoise may produce antibody that recognizes all or only a few of the proteins of a strain, and yet another animal may have a very different profile to the same antigen preparation. This observation also is consistent with mycoplasmal infections in other species.
Monoclonal vs. Polyclonal Antibodies: The decision regarding whether to use polyclonal antibodies (PAb) or a monoclonal antibody (MAb) depends on a number of factors, the most important of which are its intended use and whether the antibody is readily available from commercial suppliers or researchers (Lipman et al, 2005). Hunter et al. (2008) indicated that “PAbs can be generated much more rapidly, at less expense, and with less technical skill than is required to produce MAbs. One can reasonably expect to obtain PAbs within several months of initiating immunizations, whereas the generation of hybridomas and subsequent production of MAbs is more costly and can take up to a year or longer, to produce”. However, the principal advantages of MAbs are their homogeneity and consistency and, once the desired hybridoma has been generated, MAbs can be produced as a constant and renewable resource for the ELISA. In contrast, PAbs generated to the same antigen using multiple animals will differ among immunized animals, and their avidity may change as they are harvested over time. The quantity of PAbs obtained is limited by the size of the animal and its lifespan. Importantly, the concentration and levels of specific antibody are higher in MAbs. The concentration of specific antibody in polyclonal sera is typically 50 to 200 µg/mL, and the range of total Ig concentration in sera is between 5 and 20 mg/mL. In comparison, MAbs generated as ascites or in specialized cell culture vessels are frequently 10-fold higher in concentration and of much higher purity. The HL-673 monoclonal (Mab) antibody-based ELISA, developed at the University of Florida, has been validated over the past 17 years using rigorous statistical methods and comparisons with highly purified polyclonal anti-tortoise antibodies as well as selective diagnostic necropsies to confirm the absence of histopathological lesions, all for the development of reliable negative and positive control reagents (Shumacher et al., 1993, Brown et al., 1994, Brown et al., 1999, Wendland et al., 2007). A principal advantage of using this highly specific and potent Mab is its unlimited supply that has insured high year-to-year assay reliability. This is not the case for polyclonal antibodies. The HL673 monoclonal antibody has never failed to bind to the light chains of any desert or gopher tortoise immunoglobulins during more than 10,000 assays and titrations on control wells coated with dilutions of individual tortoise’s plasma. In addition we have shown (Wendland, et al, submitted for publication) that Mab HL673 reacts strongly with the immunoglobulins of 28 other Chelonian species, including Chelonia mydas and Caretta caretta. This compares exceptionally well with our most highly reactive affinity purified polyclonal which reacted with the immunoglobulins of all 30 tested Chelonian species (Wendland, et al, submitted for publication). In contrast, Hunter et al (2008) developed a polyclonal antibody-based ELISA test , but did not disclose any details concerning the methods used to validate their “new ELISA” (Sandmeir et al. 2009). As a result it is impossible to compare results in these recent reports with the well-established and validated ELISA developed by the University of Florida.
Pathogen Transmission: The results of the experimental challenge studies support the hypothesis that M. agassizii was horizontally transmitted between adult tortoises, and between adult and hatchling tortoises(Brown et al. 1994; Brown et al, 1999b; Brown et al, 2002). Preliminary observations indicate that seronegative hatchlings are at least as susceptible to infection as adults, and that the disease progresses more rapidly, with high morbidity in the first 6 months post infection (McLaughlin, 1997). Direct contact is the most likely route of transmission between tortoises and can easily occur during combat or courtship. If aerosol transmission occurs, it is probably over short distances (<0.5 m). Sentinel and control animals housed in pen sections adjacent to ill tortoises did not become ill or seroconvert, which indicated that aerosolized bacteria did not travel even relatively short distances over low (0.7 m) barriers. Based on the environmental transmission study, fomites are unlikely to play a major role in Mycoplasma transmission. The hypothesis that transmission was more likely when the infected tortoise was symptomatic was supported. However, tortoises without clinical signs may be infected and able to transmit the pathogen under appropriate conditions (Jacobson et al, 1995). After rainstorms, it is not uncommon to find two or more tortoises drinking from the same puddle. When tortoises drink, they often get water in their noses, which they then blow or "sneeze" out. An asymptomatic tortoise harboring bacteria may shed enough in this manner to infect a nearby conspecific via the aerosol route. “The current evidence supports horizontal transmission of Mycoplasma in desert and gopher tortoises (Brown et al. 2002)., At this time there is no evidence of vertical transmission (McLaughlin, 1997). Eggs from gopher tortoises with mycoplasmosis were collected and no maternal cloacal swab samples, or egg yolk, albumin, membrane, or chorioallantoic/amniotic fluid samples were positive by culture or PCR for Mycoplasma. None of 200 hatchling gopher tortoises hatched from eggs collected from females with mycoplasmosis showed evidence of infection with M. agassizii. Therefore, tortoises with clinical signs of URTD can produce M. agassizii-free eggs. However, the sample size in this study was small and vertical transmission cannot be ruled out (Brown et al, 2002). Further, maternal antibodies were transferred via the egg yolk, and were detected by Effects of repeated exposure of clinically healthy seropositive adult gopher tortoises in experimental studies indicated that the clinical response of these tortoises was more rapid and severe than naive tortoises, suggesting no protection was conferred by previous exposure to the organism (McLaughlin, 1997). This is consistent with some other mycoplasmal diseases in which the immune response confers limited or no protection (Brown et al, 2002), or contributes to disease pathogenesis, such as arthritis in fowl caused by M. synoviae, conjunctivitis in cattle caused by M. bovoculi, and pneumonia in humans caused by M. pneumoniae. In addition, some mycoplasmal surface proteins share sequence and structural homologies with vertebrate proteins; these may play a role in eliciting autoimmune responses. Repeated exposure to mycoplasmal proteins that resemble a host's proteins may sensitize the host and induce an autoimmune response, leading to chronic manifestations of disease even if the primary etiologic agent is cleared.
Consequences of Infection: The effects of mycoplasmosis on the long-term health and viability of affected chelonian populations is poorly understood. Based on data collected at Sanibel Island, Florida (McLaughlin 1990), at least 30% and possibly up to 50% of the adults on one site died with signs of URTD. Given the low recruitment rate of gopher tortoise populations (Cox 1989), it is unknown if the population on that site will recover to its previous levels without intensive management. While M. agassizii was associated with declines in certain population of desert tortoises in the Mojave Desert, the exact causes of death of those tortoises were not determined. However, changes in the hormonal profiles of some infected desert tortoises (Rostal et al, 1996) could lead to altered foraging and reproductive behavior and decreased reproductive potential. In nutritional studies with captive desert tortoises, deaths of juvenile tortoises were attributed to URTD (Oftedal et al, 1996). The chronic inflammatory changes seen in the nasal cavity tissues of affected tortoises could have adverse effects on foraging and reproductive behavior due to disruption of olfactory function. Both direct and indirect pathogenic mechanisms need to be studied to better predict the effects of URTD on tortoise populations.
Maternal Antibody: Desert tortoise females transfer specific IgG and IgM antibodies to their offspring that are still detectable after 1 year (Schumacher et al, 1999). The IgG and IgM antibodies were transferred. but M agassizii-specific antibodies were of the IgG class.Yolk and hatchling plasma had significantly lower amounts of specific antibodies than did plasma from adult females. Hatchlings were not infected with mycoplasmas. Offspring of sick females had significantly higher specific antibody titers than did offspring of healthy females. Titers were still significantly different in 1-year-old hatchlings. Since maternal antibody in neonates potentially confounds interpreting the ELISA test in tortoises of this age group, infection with M agassizii may be misdiagnosed in hatchlings with persistent maternal antibodies.
Control and Treatment: Ill tortoises need to be isolated from healthy tortoises. Do not release ill tortoises back to the wild. Healthy tortoises need to be thoroughly evaluated if they are to be released to the wild. In a seroprevalence study of captive desert tortoises from the greater Barstow area, Mojave Desert, USA, of 179 tortoises sampled, anti-mycoplasmal antibodies were present in 82.7 % of the tortoises. What was also interesting was the presence of anti-herpesvirus antibodies in 26.6 % of the sampled tortoises. Herpesvirus is known to infect the desert tortoise (Johnson et al, 2005) and the souce of this virus is unknown. Herpesvirus infections have been reported in a variety of exotic tortoises (Jacobson et al., 1985; Heldstab and Bestetti, 1989; Muro et al., 1998; Drury et al., 1998) and 15.5 % of the owners of desert tortoises in the greater Barstow area also had exotic chelonians in their collection. Thus desert tortoise potentially can become infected with an exotic herpesvirus and escape or release of these animals into wild populations may result in a new emerging disease. In affected tortoises, mycoplasmosis is a typically a chronic upper respiratory disease. While various antibiotics have been utilized for treatment of mycoplasmosis, it is unknown if the organism will be completely eliminated. Clinical signs of illness may abate following treatment, but relapses commonly occur.
Consensus Statement: This consensus statement reviews the pertinent accumulated information on URTD in gopher tortoises as studied by the University of Florida group and attempts to provide a scientifically sound perspective on the known effects of the disease.
It is certain that:
- Mycoplasma agassizii (strains PS6 and 723) is a cause of URTD in desert and gopher tortoises.The pathology of mycoplasmosis involves hyperplastic and dysplastic lesions in the upper respiratory tract and eyes.Clinical signs of URTD vary in onset, duration, and severity
- Mycoplasmosis is chronic and may be clinically silent (subclinical) in adult tortoises. Infection with Mycoplasma agassizii elicits specific antibody responses that can be detected by ELISA.
- The antibody responses to Mycoplasma agassizii are reliably detectable by ELISA beginning 8 weeks after experimental infection. Under experimental conditions, gopher tortoises become ill quicker after repeated exposure to Mycoplasma agassizii. Colonization of the upper respiratory tract with Mycoplasma agassizii may be detected by culture and PCR, but assay sensitivity is not as high as the ELISA.
- Mycoplasmosis is a horizontally transmissible disease.
It is probable, but not clearly established, that:
- Pathogenic and nonpathogenic tortoise mycoplasmas exist. There is variation among strains of Mycoplasma agassizii in their ability to cause URTD.Other species of Mycoplasma (such as M. testudineum) also may cause URTD.
- Mycoplasma can be transmitted by some forms of indirect contact.
Areas of uncertainty:
- If vertical egg transmission of M. agassizii occurs.
- The effect of mycoplasmosis on survival and reproduction of individual tortoises.
- The effect of mycoplasmosis on desert and gopher tortoise population dynamics and viability.
- The relationship among infection rates, transmission rates, population size, and clinical disease expression.
- The probability of transmission via burrows, fomites or vectors.
- If tortoises can clear M. agassizii and develop protective immunity.
- Potential systemic effects of mycoplasmosis and the association between M. agassizii infection and hemosiderosis in the liver.
M. agassizii ELISA and euthanasia of tortoises. ELISA testing has been used as a convenient method for determining the ultimate disposition of gopher and desert tortoises in certain parts of their range. It is important to note that the Jacobson et al. (1995) paper did not recommend indiscriminate euthanasia of M. agassizii ELISA-positive desert tortoises that were otherwise clinically healthy. The policy of euthanasia of such tortoises can be found in the 1996 “Protocol for the Prevention of the Transmission of Disease among Desert Tortoises (Gopherus agassizii) at the Desert Tortoise Conservation Center and Transfer and Holding Facility”, which was developed by Southern Nevada Environmental Incorporated and the Bureau of Land Management.This policy was only adopted in Nevada and not in any other state where desert tortoises are found. In Brown et al. (2002) the following is stated: “There are inadequate scientific data to provide definitive guidelines for the disposition of seropositive tortoises.” This paper was partially responsible for termination of the ELISA-based euthanasia program in Nevada in June 2007 (Roy Averill-Murray, USFWS, personal communication). Despite continued research over the past two decades, that statement still holds true. Euthanasia of seropostive tortoises eliminates animals that might otherwise have provided valuable reproductive and genetic contributions to wild populations. However, relocation of seropositive tortoises may result in spread of mycoplasmosis to susceptible animals and could have detrimental impacts on recipient populations. Thus, when making management decisions on the basis of the M. agassizii ELISA, it is critical to establish clear goals for the tortoise population of interest, to determine a necessary sample size to meet the goals for detection, and finally, upon receipt of results, to consider the predictive values of the test before implementing any policy. The establishment of clear goals will greatly facilitate the decision-making process.
It is important to stress that the M. agassizii ELISA alone should not be used as the sole means for evaluating the health of an individual animal; however, it is one of a very small battery of non-subjective tools available that can be used as part of a comprehensive health assessment.
Other causes of URTD: In 1985 a herpesvirus-like infection was identified in a die-off of Argentine tortoises (Geochelone chilensis) imported into south Florida that had rhinitis, and anorexia (Jacobson et al, 1985). Herpesvirus infections were first described in tortoises in the early 1980s (Harper et al., 1982). They have since been reported in many species of tortoises, including the desert tortoise, with varying clinical signs and degrees of severity (Heldstab and Bestetti, 1989; Drury et al., 1998; Muro et al., 1998). Infection is most often associated with the oral cavity and respiratory tract, with necrotizing stomatitis, glossitis, tracheitis, pharyngitis, and rhinitis having been repeatedly described (Jacobson et al., 1985; Muro et al., 1998; Drury et al., 1998; Johnson et al, 2005). Encephalitis (Heldstab and Bestetti, 1989) and hepatitis (Herva´s et al., 2002) have also been observed.
An iridovirus (Ranavirus) was identified in a gopher tortoise in Florida with clinical signs of URTD (Westhouse et al, 1996), and then several years later five instances of Ranavirus infection were documented in chelonians between 2003 and 2005 in Georgia, Florida, New York, and Pennsylvania, USA (Johnson et al, 2008). Affected species included captive Burmese star tortoises (Geochelone platynota), a free-ranging gopher tortoise (Gopherus polyphemus), free-ranging eastern box turtles (Terrapene carolina carolina), and a Florida box turtle (Terrepene carolina bauri). Results suggested that certain amphibians and chelonians are infected with a similar virus and that different viruses exist among different chelonians. Amphibians may serve as a reservoir host for susceptible chelonians. This paper also demonstrated that significant disease associated with Ranavirus infections are likely more widespread in chelonians than previously suspected.
Collaborators: The work on Mycoplasmosis of tortoises has been a multidisciplinary effort at the University of Florida and elsewhere involving faculty, graduate students, research scientists and laboratory technicians. The following present and former UF personnel were involved in this project:
Elliott R. Jacobson, D.V.M, Ph.D,Professor
Mary B. Brown, Ph.D., Professor
Paul A. Klein, Ph.D., Professor
Lori Wendland, D.V.M., Ph.D., Research Assistant Professor
Daniel R. Brown Ph.D., Associate Professor
Isabella M. Schumacher, D.V.M., (Research scientist, BEECS Program)
Bruce L. Homer, D.V.M, Ph,D, D.A.C.Z.M.
Catherine McKenna (Biological scientist)
Grace S. McLaughlin, Ph.D. (Research coordinator)
Barbara C. Crenshaw (Laboratory manager)
Linda Green, Scientific Director, UF Hybridoma Core Laboratory
Diane Duke Laboratory technician, UF Hybridoma Core Laboratory
John Hutchison (Technical assistant)
Alyssa Whitemarsh (Technical assistant)
Michael Lao (Technical assistant)
David Bunger (Technical assistant)
Sylvia J. Tucker (Senior laboratory technician)
Noelle T. Chmielewski, D.V.M. (Ophthalmology resident)
Nicole Gottdenker, D.V.M. (Graduate assistant)
The following non-UFL scientists and faculty were also involved:
Kristin H. Berry, Ph.D., USGS, Riverside, CA.
Mary M. Christopher, D.V.M., Ph.D., University of California at Davis
Joan Berish, MS, FWS Gainesville, Florida
Contact:
Dr. Mary Brown
Department of Pathobiology
Box 110926
College of Veterinary Medicine
University of Florida
Gainesville, FL 32610
E-Mail: mbbrown@nervm.nerdc.ufl.edu
For General Information Contact:
Dr. Elliott Jacobson
Box 100126
Department of Small Animal Clinical Sciences
College of Veterinary Medicine
University of Florida
Gainesville, FL 32610-0126
E-Mail: JacobsonE@vetmed.ufl.edu