The California sea lion (Zalophus californianus), a permanent inhabitant of the Gulf of California in Mexico, is susceptible to pathogenic Leptospira spp. infection, which can result in hepatic and renal damage and may lead to renal failure and death. During summer 2013, we used the microscopic agglutination test (MAT) to investigate the prevalence of anti-Leptospira antibodies in blood of clinically healthy sea lion pups from seven rookery islands on the Pacific Coast of Baja California (Pacific Ocean) and in the Gulf of California. We also used PCR to examine blood for Leptospira DNA. Isolation of Leptospira in liquid media was unsuccessful. We found higher antibody prevalence in sea lions from the rookery islands in the gulf than in those from the Pacific Coast. Antibodies against 11 serovars were identified in the Gulf of California population; the most frequent reactions were against serovars Bataviae (90%), Pyrogenes (86%), Wolffi (86%), Celledoni (71%), and Pomona (65%). In the Pacific Ocean population, MAT was positive against eight serovars, where Wolffi (88%), Pomona (75%), and Bataviae (70%) were the most frequent. Serum samples agglutinated with more than one Leptospira serovar. The maximum titer was 3,200. Each island had a different serology profile, and islands combined showed a distinct profile for each region. We detected pathogenic Leptospira DNA in 63% of blood samples, but we found no saprophytic Leptospira. Positive PCR results were obtained in blood samples with high and low MAT titers. Together, these two methods enhance the diagnosis and interpretation of sea lion leptospirosis. Our results may be related to human activities or the presence of other reservoirs with which sea lions interact, and they may also be related to sea lion stranding.
The California sea lion (Zalophus californianus) is a protected marine mammal worldwide. The genus Zalophus contains three species of sea lions: Zalophus japonicas, which has been extinct since 1950 but once inhabited the Japanese archipelago; Zalophus wallebaeki, which inhabits the Galapagos Islands; and Z. californianus, which is distributed in the western part of the Northern Hemisphere. Zalophus californianus is the most abundant pinniped species in Mexico and a permanent inhabitant of the Gulf of California (Aurioles-Gamboa 1993). Its distribution extends from British Columbia, Canada (51°N), to the Mary Islands (19°N), and the Pacific Coast and Gulf of California, Mexico (King 1983). These mammals are under special protection by Mexican federal regulation NOM-059-ECOL-2010 (SEMARNAT 2010). Nevertheless, the population has decreased in recent decades (Hernández-Camacho et al. 2008). The genus Leptospira (phylum Spirochaetes) causes leptospirosis worldwide, and it is a cause of sea lion deaths in the northern Pacific Ocean (Gulland et al. 1996; Greig 2005). In mammals, the disease is transmitted through direct or indirect contact with infected urine or with water contaminated by the urine of animal reservoirs (Monahan et al. 2009). However, in sea lions, the mode of transmission is unknown.
Leptospira infection in sea lions affects all ages and is characterized by bacterial colonization of the liver and kidneys, causing acute renal failure and death. In adult sea lions, Leptospira is one of the most frequent causes of stranding (Mancia et al. 2012). Its prevalence differs among populations: In Southern California, leptospirosis was formerly considered a rare disease in sea lions stranded between 1970 and 1981 (Trillmich et al. 1991), but its prevalence has increased and is now considered epizootic (Gulland et al. 1996). The only serovar that has been isolated from sea lions is Leptospira interrogans serovar Pomona (Zuerner and Alt 2009; Prager et al. 2013). In contrast, in New Zealand, leptospirosis is not considered a health threat due to its low prevalence (Roe et al. 2010). It is not known whether recurring epidemics in sea lion populations are due to external sources of infection, such as contact with wildlife or other carrier animal species, or due to internal epidemic cycles related to the population's changing immunity. The latter cause may involve the protection of individuals due to previous infections, or offspring and individual migration between groups, resulting in a higher percentage of susceptible individuals leading to outbreaks (Lloyd-Smith et al. 2007).
Sea lions are distributed along the Pacific Ocean coasts and the Gulf of California, and their populations have been threatened by several Leptospira serovars (Gulland et al. 1996; Godinez et al. 1999; Acevedo-Whitehouse et al. 2003). We assessed the antibody prevalence and the presence of Leptospira in sea lion pups from seven reproductive islands in this area.
MATERIALS AND METHODS
Animals and blood samples
Under the Mexican government authorization SGPA/2897/12, we sampled sea lion pups during June 2013 in the Pacific Ocean islands of Asuncion (27°06′N, 114°17′W), Natividad (27°52′N, 115°11′W), and Cedros (28°11′N, 115°13′W), and during July 2013 in the Gulf of California islands Granito (29°33′N, 113°32′W), Coloradito (24°18′N, 110°21′W), Los Cantiles (29°32′N, 113°29′W), El Partido (28°54′N, 113°02′W), and Roca Consag (31°06′N, 114°27′W) (Fig. 1).
We collected 91 blood samples using 18-gauge, 38-mm needles from the jugular vein of manually restrained California sea lion pups. If needed, inhaled anesthesia was applied with induction of 5% isoflurane and maintenance at 3% (Haulena and Heath 2001). Blood samples were taken in a minimally invasive manner from 51 pups from rookeries of five islands in the Gulf of California and from 40 pups from three rookeries in the Pacific Ocean. The samples were deposited in sterile tubes for serum separation and into tubes with sodium citrate. Blood samples were refrigerated and transported to the laboratory, and serum aliquots were frozen at −20 C until used. Additionally, each animal was examined for clinical signs of disease, and weight, measurements, and body temperature were recorded (Luque-Flores and Aurioles-Gamboa 2001).
Three drops of each blood sample were placed into semisolid Fletcher medium at the moment of sampling and transported to the laboratory. Each sample was subcultured in Ellinghausen-McCullough-Johnson-Harris (EMJH) liquid medium and maintained at 30 C. Cultures were periodically observed by dark field microscopy to detect spirochaetal forms. Cultures were maintained for 6 mo before being considered negative.
We extracted DNA from blood samples with the DNeasy Blood & Tissue Kit (Qiagen, Valencia, California, USA), according to the manufacturer's recommendations and suspended in 50 μL of nuclease-free water. The DNA was quantified using an Epoch microplate spectrophotometer (Biotek, Winooski, Vermont, USA) and stored at 4 C until used for PCR.
The PCR targets were the 23S rDNA, which identifies DNA from the genus Leptospira, the IS1500 insertion sequence, which identifies only L. interrogans (sensu lato), and the 16S rRNA, which identifies only saprophytic Leptospira. The Leptospira genus-specific primers L737 (5′-GACCCGAAGCCTGTCGAG-3′) and L1218 (5′-GCCATGCTTAGTCCCGATTAC-3′), based on the 23S rDNA (Woo et al. 1998), were used to amplify a 482-base-pair (bp) fragment in a final volume of 50 μL with 0.5 μg of DNA, 1.2 μM of each primer, 0.2 mM of each deoxyribonucleotides (dNTP), 10 μL of 5X GoTaq Flexi Buffer, 3 mM MgCl2, 0.5 mg/mL of bovine serum albumin (BSA), and 2.5 U GoTaq (Promega, Madison, Wisconsin, USA). The PCR protocol was an initial denaturation at 94 C for 5 min, followed by 45 cycles of denaturation at 94 C for 15 s, annealing at 59 C for 40 s, and extension at 74 C for 1 min and 20 s, and a final extension at 74 C for 10 min. To specifically amplify the IS1500 insertion sequence of L. interrogans (sensu lato) (Zuerner and Bolin 1997), the primers P1 (5′-TCGCTGAAATRGGWGTTCGT-3′) and M16 (5′-CGCCTGGYTCMCCGATT-3′) were used in a final volume of 50 μL with 0.5 μg of DNA, 1 μM of each primer, 0.2 mM of each dNTP, 10 μL of 5X GoTaq Flexi Buffer, 3 mM MgCl2, 0.5 mg/mL of BSA, 1% Triton X-100, and 2.5 U GoTaq (Promega). The same PCR conditions were used to identify saprophytic Leptospira species, with the primers (5′-AGAAATTTGTGCTAATACCGAATGT-3′) and (5′-GGCGTCGCTGCTTCAGGCTTTCG-3′), based on the rrs gene of the 16S rRNA (Murgia et al. 1997), which amplified a 240-bp fragment. Reactions with a final volume of 50 μL contained 0.5 μg of DNA, 0.6 μM of each primer, 0.2 mM of each dNTP, 10 μL of 5X GoTaq Flexi Buffer, 2 mM MgCl2, and 1 U GoTaq (Promega). The PCR protocol was an initial denaturation at 94 C for 5 min, followed by 40 cycles of denaturation at 94 C for 30 s, annealing at 60 C for 45 s, and extension at 74 C for 1 min with an increment of 5 s/cycle and a final extension at 74 C for 10 min. Control amplification templates included water as a negative control and genomic DNA of serovars Autumnalis, Bataviae, Bratislava, Canicola, Celledoni, Grippothyphosa, Hardjoprajitno, Icterohaemorrhagiae, Pomona, Pyrogenes, Tarassovi, Wolffi, and Biflexa serovar Patoc strain Patoc. Amplified products were visualized by electrophoresis on 1.6% agarose gels and staining with ethidium bromide.
Microscopic agglutination test
The microscopic agglutination test (MAT) was performed as described by the Panamerican Health Organization (Myers 1985). Four to 7-d cultures of 12 Leptospira serovars—Autumnalis, Bataviae, Bratislava, Canicola, Celledoni, Grippotyphosa, Hardjoprajitno, Icterohaemorrhagiae, Pomona, Pyrogenes, Tarassovi, and Wolffi, grown in EMJH—were used as antigens. These serovars are part of the collection of the Microbiology and Immunology Department, Facultad de Medicina Veterinaria y Zootecnia, Universidad Nacional Autónoma de México. Serum samples were diluted 1:50 for screening, and 50-μL aliquots of the 12 serovars were added to Nunc 96-well flat-bottom microtiter plates (Nalge Nunc International, Rochester, New York, USA). A negative control was included for each serovar. The plate was gently stirred and incubated at room temperature for 1 h. Plates were read by dark field microscopy (Carl Zeiss, Oberkochen, Germany). A serial dilution of each serum sample (1:50 to 1:3,200) was used to determine the titer for each serovar, with the final titer representing the reciprocal of the maximum dilution at which agglutination was observed.
Descriptive statistics were applied to the MAT results, and statistical analysis of the PCR results was performed using Graphpad Prism software (GraphPad Software, La Jolla, California, USA). Spearman's rank correlation was used to assess the correlation among the three PCRs. Analysis of variance and Tukey's tests were applied to each serovar, and Student's t-test was used to compare between regions (Pacific vs. Gulf of California).
No spirochaetal forms were observed, and all blood samples were considered negative after 6 mo.
Serology of Leptospira spp.
For the interpretation of MAT results, a cutoff titer was not established given that sea lions are not vaccinated. Serum antibodies against at least one Leptospira serovar were detected in all serum samples.
In the Pacific Ocean samples, positive MAT reactions were observed against eight serovars. No reactions occurred against four serovars (Table 1). In the Gulf of California samples, there were MAT reactions against 11 of the 12 serovars (Table 1). Each island showed a unique MAT pattern to the serovars tested (Table 2).
There was a statistically significant difference in frequency among serovars (P<0.0001). A significant difference in frequency was found between samples from the Gulf of California and the Pacific Coasts for serovars Bataviae (P<0.0001), Canicola (P=0.0009), Pomona (P=0.0336), Pyrogenes (P<0.0001), Tarassovi (P<0.0001), and Autumnalis (P=0.0128). The only serovar with prevalences that were not significantly different between sampling areas was Wolffi (P=0.9045). A statistical analysis of the remaining serovars were not conducted because they were not present in the Pacific Ocean samples.
We analyzed 91 DNA samples by PCR. Amplification of the 23S rRNA identified Leptospira in 47 samples (52%). The PCR based on the IS1500 insertion sequence detected DNA of pathogenic L. interrogans (sensu lato) in 28 samples (31%). The IS1500 PCR identified 10 blood samples positive for Leptospira that had previously been negative to the 23S rRNA PCR. Despite the low correlation between them (rs=0.117), we assumed that the identification with IS1500, which is only present in pathogenic Leptospira, indicated the general presence of the genus Leptospira. Therefore, the combined result of both PCRs indicated that 63% (n=57) of samples were positive, and 37% (n=34) of samples were negative. By the 16S-based PCR, none of the samples tested was identified as a carrier of saprophytic serovars.
In comparison to the MAT results, 48 PCR-positive samples had their highest titers to L. interrogans (sensu stricto), including serovars Autumnalis, Bataviae, Canicola, Wolffi, Icterohaemorrhagiae, Pomona, and Pyrogenes. Ten PCR-positive samples had maximum titers to Leptospira borgpetersenii, namely, serovar Tarassovi, and seven PCR-positive samples had their maximum titer to Leptospira kirschneri, represented by serovar Grippotyphosa. Of those samples with highest titer to more than one serovar, seven had titers corresponding to more than one Leptospira species (Table 3). Samples with high (1,600) and low (50) titers were positive by PCR.
Our MAT results indicate that the Gulf of California has a higher prevalence of pathogenic Leptospira serovars than the Pacific Ocean. In the serum samples from the Gulf of California, positive agglutination was observed to 11 serovars from the 12 tested in MAT, while in the Pacific Ocean samples, agglutination was observed to only eight serovars. Additionally, the serum samples tested by MAT from the Pacific Ocean were positive to at least one serovar, and up to five serovars, while samples from the Gulf of California were positive to at least two, and up to nine serovars. However, MAT results cannot rule out cross-reactive antibodies. Previous studies report asymptomatic carriage of Leptospira in sea lions (Cameron et al. 2008; Prager et al. 2013), consistent with our observations of no signs of leptospirosis during sampling. Regarding the maximum titers in sea lions, the rookeries in the Gulf of California had higher titers compared to those of the Pacific islands, and only one sample had titers of 3,200. A similar titer was reported previously for serovars Wolffi, Pomona, and Bataviae (Pedernera-Romano 2004). However, animals that developed disease had higher titers (Prager et al. 2013).
The serovars we report differ from those reported by Godinez et al. (1999); the main difference is that we report serovar Celledoni in the area. Also, Godinez et al. (1999) reported a low frequency of serovar Icterohaemorrhagiae, while in our study, the frequency of serovar Icterohaemorrhagiae was higher in four islands. The Leptospira serovars varied among sea lion rookeries (Table 2), highlighting the changes in Leptospira prevalence, and the relevance of epidemiologic surveillance (SEMARNAT 2014).
Serovars Pyrogenes, Ballum, Wolffi, Celledoni, and Pyrogenes had been previously reported in the Gulf of California (Pedernera-Romano 2004), and we found a high prevalence of these serovars, in addition to serovars Pomona and Bataviae, with Bataviae being the most frequent. This high prevalence was observed in geographically distant rookeries with different characteristics. For example, Coloradito and El Partido are small islands rarely visited by humans. El Coloradito, El Partido, and Consag are occasionally used by fishermen as a refuge during unfavorable weather or as a journey station, but this is not frequent or for prolonged periods, making it difficult to invoke anthropogenic contamination as a source of Leptospira.
The islands in the Pacific Ocean were highly uniform with respect to antibody prevalence. Serovars found with high frequency were Wolffi, Pomona, and Bataviae, and prevalences for serovars Celledoni and Pyrogenes were low. Four serovars were not identified in the Pacific Ocean islands, and most of the serum samples had low antibody titers (Table 2). Asuncion and Natividad islands do not have recent reports of feral mice, dogs, or rats. In contrast, Cedros is populated by humans, and sea lion contact with feral dogs, cats, mice, and rats is more likely. In fact, feral dogs are common predators of sea lions (Gallo-Reynoso and García-Aguilar 2008). On Cedros, we found titers of 1,600 and 800 in two and four sea lions, respectively. On Asunción and Natividad, we found low titers (100). These differences may be explained by contact with carrier animals in Cedros and in two more islands, and due to migration of juveniles and males among these three nearby islands.
The serovars frequently found in the Pacific Ocean samples were Pomona, Wolffi, and Bataviae. This agrees with work on the California coast by Vedros et al. (1971), who identified serovar Pomona as the most frequent L. interrogans serovar, which was later isolated by Prager et al. (2013). This observation might be also related to the sea lions' migration behavior, which may influence the distribution of Leptospira among islands, even in the absence of a classic carrier of this pathogen.
Our PCR protocols were chosen to identify the genus Leptospira and L. interrogans (sensu lato), mainly because, to our knowledge, only L. interrogans serovar Pomona has been isolated from sea lions (Prager et al. 2013). We detected Leptospira DNA in 63% of the sea lions and measured antibody prevalence in these samples. A positive PCR did not always correspond with higher titers, which were as low as 50. Leptospira DNA was present in the blood of healthy sea lion pups in which antibody titers varied from 50 to 1,600 without clinical signs of disease. However, bacterial viability cannot be assessed by PCR.
These results, and the fact that the sampling was during the breeding season, suggest that leptospires were likely transmitted between adults and pups by close contact. It is possible that some environmental factors, such as “El Niño,” cause behavior changes in adults and pups. In addition, changes in host population susceptibility rather than pathogen availability may cause outbreaks, such as the northern elephant seal (Mirounga angustirostris) strandings on the coast of California (Colegrove et al. 2005) and the epidemic outbreak in California sea lions on the California coast (Gulland et al. 1996).
Antibodies against the saprophytic serovar, Leptospira Patoc, were previously reported by MAT in 53.8% of sea lion pups on 11 breeding islands in the Gulf of California (Pedernera-Romano 2004). However, saprophytic serovars were not included in our MAT, and we found no positive samples using the 16S PCR.
Our data suggest a threat to the health of these protected sea lions. We encourage further study of human activities and other conditions that promote the presence of Leptospira serovars in sea lions, as well as how their asymptomatic carrier status can change into a health condition leading to stranding.
This study was supported by Programa de Apoyo a Proyectos de Investigación e Innovación Tecnológica (PAPIIT) grants IN221314 and IN229111, Universidad Nacional Autónoma de México. We thank Hugo Moreno Prado, Rito Vale Navarro, Eduardo Guillén Díaz, and Joel Prieto Ceseña, rangers in the Área de Protección de Flora y Fauna Islas del Golfo de California, Baja California, National Commission for Natural Protected Areas, DVM; Osvaldo Martinez Rey from the Africam Safari Zoo; and personnel of Exportadora de Sal Sociedad Anónima de Capital Variable for their logistical support. Rosalía Avalos received Consejo Nacional de Ciencia y Tecnología scholarship 201527. We thank Eloísa Reyes for statistical assistance and Michael F. Dunn for critical review of the manuscript.
These authors contributed equally to this manuscript.