Abstract

Emergent hypermucoviscous (HMV) strains of Klebsiella pneumoniae have been reported in multiple marine mammal species; however, there is limited information regarding the epidemiology and pathogenesis of this infection in these species. We determined the prevalence of HMV K. pneumoniae in wild-caught and stranded marine mammal populations on the US Pacific Coast. Samples were collected from 270 free-ranging California sea lions (CSLs; Zalophus californianus) captured at three discrete sampling sites and from 336 stranded marine mammals of various species. We recovered HMV K. pneumoniae only from CSLs, with a prevalence of 1.5% (4 of 275) in stranded animals, compared with 1.1% (3 of 270) in wild-caught animals. We assessed the phenotypic and genotypic variability of recovered HMV K. pneumoniae isolates recovered from CSLs (n=11) and of archival HMV and non-HMV isolates from stranded marine mammals (n=19). All but two HMV isolates were of the K2 serotype, whereas none of the non-HMV isolates belonged to this serotype. Of the HMV isolates, 96% (24 of 25) were PCR positive for the HMV-associated gene p-rmpA, whereas 92% (23 of 25) were PCR positive for p-rmpA2. Genetic fingerprinting by repetitive extragenic palindromic PCR showed four discrete clusters, demonstrating genotypic variability that loosely correlated with phenotype. Antimicrobial susceptibility testing revealed all isolates from stranded CSLs were susceptible to ceftiofur, indicating this antimicrobial agent is an appropriate choice for treatment of HMV K. pneumoniae infections in stranded CSLs. Our culture assay could reliably detect HMV K. pneumoniae from concentrations as low as 102 colony-forming units per milligram of feces. We identified the presence of HMV K. pneumoniae in both wild-caught and stranded CSLs from the US Pacific Coast and highlight the need for further studies to evaluate the potential impact of this pathogen on marine mammal health.

INTRODUCTION

Klebsiella pneumoniae is a Gram-negative facultative anaerobic bacterium belonging to the family Enterobacteriaceae. Typically found on mammalian mucosal surfaces and in the environment, K. pneumoniae has been associated with a wide range of infections in humans and animals (Podschun and Ullmann 1998; Doud et al. 2009). In marine mammals, this bacterium has been isolated from lesions associated with polyarthritis, meningoencephalitis, and peritonitis (Castinel et al. 2007a). Because it can infect multiple hosts and adopt rapidly evolving antimicrobial resistance patterns, this bacterium is recognized as an emerging pathogen in human and veterinary medicine (Lederman and Crum 2005; Jang et al. 2010).

Hypermucoviscous (HMV) K. pneumoniae, considered a hypervirulent form due to its ability to cause life-threatening, community-acquired infection in healthy individuals; its ability to cause metastatic infections; and its greater virulence in animal models (Kong et al. 2012; Shon et al. 2013; Lin et al. 2014), has been associated with pleuritis and suppurative pneumonia in California sea lions (CSLs; Zalophus californianus) and with septicemia and meningitis in New Zealand sea lion (NZSL; Phocarctos hookeri) pups (Jang et al. 2010; Roe et al. 2015). Little is known about the pathogenesis and epidemiology of this infection in marine hosts, and there is limited information regarding its diagnosis and treatment in both rehabilitated and wild marine mammals.

We aimed to evaluate the prevalence of HMV K. pneumoniae in stranded marine mammals and in wild-caught CSLs from the US Pacific Coast. We assessed the phenotypic and genotypic variability of all recovered HMV K. pneumoniae isolates as well as of archived isolates collected from marine mammals. In addition, the limit of HMV K. pneumoniae detection by culture techniques was evaluated to assess current standards for identification of HMV K. pneumoniae in clinical settings.

MATERIALS AND METHODS

Prevalence of HMV K. pneumoniae in marine mammals

Between August 2015 and May 2016, 275 stranded CSLs, 53 stranded pacific harbor seals (PHSs; Phoca vitulina), three stranded northern elephant seals (NESs; Mirounga angustirostris), and five stranded northern fur seals (NFSs; Callorhinus ursinus) were sampled after admittance to the Marine Mammal Center (TMMC) in Sausalito, California, for rehabilitation. Stranded animals originated predominantly from the central and northern California coast. Animals of all ages and both sexes were included in this study. Upon admittance, animals were examined by a veterinarian and when possible, mucosal site swabs (oral, rectal, and nasal) were obtained with a sterile Amies bacteriologic swab (Becton, Dickinson and Co. Scientific, Sparks, Maryland, USA). Animals that died during rehabilitation or that were dead upon presentation were necropsied at TMMC, and mucosal sites, lung, liver, kidney, and grossly visible abscesses were swabbed.

Between August 2015 and May 2016, predominantly yearling and juvenile CSLs of both sexes were sampled from Año Nuevo Island, California (n=68) and San Miguel Island, California (n=119), and adult and subadult male CSLs were sampled from Astoria, Oregon (n=83). All wild-caught animals were apparently healthy at the time of capture and sampling, and samples were obtained while animals were under general anesthesia. Swabs of oral, rectal, and nasal sites were obtained similarly those from stranded animals by using the Amies bacteriologic swabs. All wild animal sampling was performed under the National Oceanic Atmospheric Administration National Marine Fisheries Service permits 17115-03 and 16087-01 and with the approval of the University of California–Los Angeles Chancellor's Animal Research Committee in the Office of Animal Research Oversight (ARC 2012-035-11). Hypermucoviscous K. pneumoniae isolation prevalence was determined for each of the following groups: stranded animals including CSLs, PHSs, NESs, and NFSs and wild-caught CSLs. Wild-caught CSLs were further categorized by geographic sampling location.

Bacterial isolation

Swabs collected from stranded and wild-caught animals were stored at room temperature and used to inoculate 5% sheep blood agar (SBA) plates (Biological Media Services, University of California, Davis, California, USA) within 7 d of collection. In previous studies, HMV K. pneumoniae remained viable for more than 7 d when maintained at these conditions. Inoculated plates were incubated for 48 h at 37 C. Colonies consistent with the appearance of HMV K. pneumoniae (grayish, mucoid colonies) were tested for hypermucoviscosity by the string test (Twenhafel et al. 2008). Individual colonies from string test–positive isolates were replated on SBA plates and incubated for 48 h at 37 C, Gram stained, and tested for production of cytochrome oxidase. All Gram-negative, cytochrome oxidase–negative, string test–positive isolates that grew on SBA plates were expanded in Luria broth overnight at 37 C and cryogenically stored (−80 C; 20% [v/v] glycerol) for further characterization.

Archived marine mammal isolates collected from stranded animals at TMMC as well as samples collected from stranded marine mammals in northern California and archived at the University of California–Davis Veterinary Medical Teaching Hospital were evaluated for genotypic and phenotypic comparisons. Archived isolates were revived on SBA plates as previously described and subjected to the same identification protocols. String test–negative samples identified molecularly as K. pneumoniae were categorized as non-HMV K. pneumoniae isolates. In total, 30 isolates, including isolates collected during the prevalence survey, in addition to archival HMV K. pneumoniae and non-HMV K. pneumoniae were included in this study. A Klebsiella quasipneumoniae subsp. similipneumoniae (formerly K. pneumoniae K6; ATCC 700603) was included in the analysis as an outlier for genetic characterization.

Molecular characterization

Genomic DNA was isolated using the DNEasy Blood and Tissue kit (Qiagen, Valencia, California, USA). Aliquots were then stored at −80 C until further use. Isolated DNA was used in endpoint PCR amplification of the K. pneumoniae capsular fucose synthesis gene wcaG and the K2 capsular type gene wzyK2 (Table 1; Turton et al. 2010). Real-time PCR detection of the Klebsiella hemolysin gene khe and the K1 capsular type mucoviscosity-associated gene magA (Hartman et al. 2009) was also performed (Table 1). The K1 and K2 capsular types were targeted, as these are most frequently associated with HMV K. pneumoniae in other studies (Fang et al. 2007). The wcaG primer was used, as this gene is a putative HMV K. pneumoniae virulence factor (Turton et al. 2010) and the khe gene was targeted, as it is specific for K. pneumoniae (Roe et al. 2015; Table 1). Positive endpoint PCR results were defined as possessing banding patterns similar to positive controls, whereas real-time PCR results were deemed positive when the threshold cycle value was less than <38. Based on previously published evidence indicating that the plasmid-encoded genes rmpA and rmpA2 are highly associated with hypermucoviscosity and virulence (Hsu et al. 2011), real-time PCR assays to detect p-rmpA and p-rmpA2 were developed by the University of California–Davis Real-time PCR Research and Diagnostics Core Facility (Table 1). Both p-rmpA2 (AB289643) and p-rampA (AB289642) sequences from the GenBank were collected using Sequence Analysis and Molecular Biology Data Management software Vector NTI AdvanceTM11 (Thermo Fisher Scientific, Carlsbad, California, USA). A BLASTn (National Center for Biotechnology Information 2017) search of the resulting sequence nonredundant nucleotide database was used to confirm sequence specificity.

Table 1

Probe/primer sets for each of the PCR targets, including nucleotide sequence and published reference for characterization of hypermucoviscous strains of Klebsiella pneumoniae isolated from marine mammals that were stranded or wild caught along the US Pacific Coast in 2015–16.

Probe/primer sets for each of the PCR targets, including nucleotide sequence and published reference for characterization of hypermucoviscous strains of Klebsiella pneumoniae isolated from marine mammals that were stranded or wild caught along the US Pacific Coast in 2015–16.
Probe/primer sets for each of the PCR targets, including nucleotide sequence and published reference for characterization of hypermucoviscous strains of Klebsiella pneumoniae isolated from marine mammals that were stranded or wild caught along the US Pacific Coast in 2015–16.

Genomic characterization

We designed PCR primers targeting fragments of the β subunit of the bacterial RNA polymerase (rpoB) and the B subunit of topoisomerase II (gyrB) based on K. pneumoniae genomes by using the Primer3 plugin in Geneious® 10.0 (Newark, New Jersey, USA; Table 1). Amplicons were purified using the QIAquick PCR Purification kit (Qiagen), and single Sanger sequencing reads were generated commercially (Eurofins Genomics, Louisville, Kentucky, USA) by using the forward primer for each target. Ambiguous base calls were annotated manually based on corresponding chromatograms and sequences were trimmed to a uniform length in Geneious 10.0. rpoB and gyrB sequence fragments of Klebsiella spp. isolates obtained from aquatic mammals were compared with K. pneumoniae genomes identified by a BLASTn search (National Center for Biotechnology Information 2017).

Phylogenetic inferences based on concatenated rpoB>gyrB gene sequences of Klebsiella spp. isolates from marine mammals were made in MEGA 6.0 (Tamura et al. 2013). In brief, all concatenated sequences were aligned with MUSCLE (Edgar 2004), and Bayesian inference criterion identified the Tamura three-parameter model with gamma distribution (five categories +G, parameter=0.2623) as the best-fit nucleotide substitution model for maximum likelihood analysis (Nei and Kumar 2000; Tamura et al. 2013). All positions containing gaps and missing data were eliminated, and the final tree was constructed from 1,000 bootstrap replicates. The concatenated rpoB>gyrB sequences of K. quasipneumoniae ATCC 700603 were included as an outlier.

Repetitive extragenic palindromic PCR (REP-PCR) fingerprinting was performed on case isolates by using the BOX primer (Table 1) following protocols for bacterial genotyping from previous studies (Versalovic et al. 1994; Griffin et al. 2013). Genetic fingerprints generated by the BOX primer were analyzed using Quantity One 4.6.9 software (Bio-Rad Laboratories, Inc., Hercules, California, USA). Band sizes were estimated by comparison with concurrently run standards, and distinct bands were manually annotated to calculate Dice coefficients and generate a dendrogram based on the unweighted pair group method using arithmetic averages. As mentioned, genomic DNA from K. quasipneumoniae (ATCC 700603) was included as an outlier.

Antimicrobial susceptibility

The minimal inhibitory concentrations (MICs) of antimicrobial agents to all the K. pneumoniae isolates included in this study were tested using the Sensititre AVIANF plate system (Trek Diagnostic System Inc., Cleveland, Ohio, USA). The MIC was determined based on the manufacturer's suggested protocol and the Clinical Laboratory Standards Institute–published protocol (Watts and Clinical and Laboratory Standards Institute 2008). The MIC50 was defined as the antimicrobial concentration that inhibited growth of 50% of the isolates.

Detection limit of HMV K. pneumoniae by fecal culture assay

A spiking experiment was used to determine the limit of detection of HMV K. pneumoniae in feces and in nuclease-free water based on the culture methods we used. Of the three mucosal sampling sites, fecal swabs were selected for this experiment as they were the most likely to be contaminated by organisms that could potentially inhibit growth of HMV K. pneumoniae. A fresh stool sample was collected opportunistically from a stranded CSL of unknown K. pneumoniae status being rehabilitated at TMMC. The sample was confirmed negative for HMV K. pneumoniae by SBA plating for 48 h and summarily performing the string test on any suspect colonies. The sample was then used to create a solution comprised of 10 mg of feces/mL in sterile phosphate-buffered saline. We chose this concentration because it approximated the amount of stool visually present on fecal swabs. Tenfold serial dilutions of isolate HMV K. pneumoniae University of California Fish Pathology Lab ID number (UCD) 18 were created and added to 100-μL aliquots of the stool solution to create concentrations ranging from 101 to 106 colony-forming units (CFU)/mg of feces. Each spiked sample was analyzed in triplicate. To serve as positive controls, 100-μL aliquots of nuclease-free water were similarly spiked in triplicate with the same serial dilutions of HMV K. pneumoniae UCD 18. For each group, a triplicate set of nonspiked samples served as the negative control. Using a sterile swab, samples were streaked onto SBA plates and incubated at 37 C for 48 h, after which all plates were examined visually and string tests performed to confirm growth of HMV K. pneumoniae.

Statistical methods

The Fisher's exact test was used to identify significant differences between prevalence in wild-caught CSLs vs. stranded CSLs, with an α level set at 0.05. The Wilcoxon rank-sum test was used to evaluate significant differences between MICs for HMV vs. non-HMV isolates and for HMV stranded vs. wild-caught CSLs, with α set at 0.05. All calculations were performed using R statistical software (R Development Core Team 2017).

RESULTS

Prevalence of HMV K. pneumoniae in marine mammals

Hypermucoviscous K. pneumoniae was recovered from 1.2% (4 of 336) stranded marine mammals sampled at TMMC. All isolates were recovered from CSL. The four animals that were positive either died or were euthanized, two due to likely HMV K. pneumoniae infection and two due to other causes. Hypermucoviscous K. pneumoniae was recovered from 1.1% (3 of 270) wild-caught CSLs, 1.4% (1 of 68) at Año Nuevo Island, 1.2% (1 of 83) at Astoria, and 0.8% (1 of 119) at San Miguel Island. Wild-caught CSLs were apparently healthy. When just stranded CSLs were considered, isolates were obtained from 1.5% (4 of 275) animals; however, the difference in prevalence between stranded and wild-caught CSLs was not significant (P=1.000).

Bacterial identification

Of the 30 K. pneumoniae isolates included in this study, 11 isolates recovered during the prevalence sampling period, seven CSL isolates collected in 2015 before the study period, and seven marine mammal isolates (five CSLs, one harbor porpoise [Phocoena phocoena], and one PHS) archived at the University of California–Davis Veterinary Medical Teaching Hospital were HMV. Eighty-four percent (21 of 25) of the HMV isolates were recovered from males. Five archived K. pneumoniae isolates (two CSLs and three PHSs) were nonhypermucoviscous with 60% (2 of 3) recovered from males.

Molecular characterization

All 30 recovered isolates and the K. quasipneumoniae ATCC 700603 outlier were khe positive and magA negative. All but two of the HMV isolates were capsular type K2 serotype, whereas all non-HMV isolates did not belong to this serotype. Only one isolate, a non-HMV isolate from a CSL, tested positive for the wcaG gene. Of the analyzed HMV isolates, 96% (24 of 25) were positive for HMV associated gene p-rmpA and 92% (23 of 25) were positive for p-rmpA2 (Table 2). One non-HMV isolate tested positive for both p-rmpA and p-rmpA2 (Table 2).

Table 2

String test results and molecular characterization for hypermucoviscous isolates of Klebsiella pneumoniae isolated from marine mammals that were stranded or wild caught along the US Pacific Coast in 2015–16 evaluated in this study.a

String test results and molecular characterization for hypermucoviscous isolates of Klebsiella pneumoniae isolated from marine mammals that were stranded or wild caught along the US Pacific Coast in 2015–16 evaluated in this study.a
String test results and molecular characterization for hypermucoviscous isolates of Klebsiella pneumoniae isolated from marine mammals that were stranded or wild caught along the US Pacific Coast in 2015–16 evaluated in this study.a

Genomic characterization

All Klebsiella spp. isolates recovered from marine mammals were identified by rpoB and gyrB sequence as K. pneumoniae. At rpoB, K. pneumoniae isolates in this study averaged 99.9% identity (99.8–100%) among them and 99.8–100% similarity to selected K. pneumoniae genomes. Similarly, gyrB sequences of these isolates averaged 99.8% (99.5–100%) identity among them and 99.5–100.0% identity to K. pneumoniae genomes. Comparatively, K. pneumoniae spp. isolates recovered from marine mammals shared 97.3–97.4% similarity at rpoB and 94.8–94.9% similarity at gyrB to K. quasipneumoniae ATCC 700603 genome (GenBank no. CP014696). Phylogenetic analysis of these isolates based on concatenated rpoB>gyrB sequence was consistent with sequence identity patterns above, suggesting the K. pneumoniae isolates from marine mammals described here formed a discrete conspecific group, distinct from K. quasipneumoniae. Although variability at these two gene targets was limited, a majority (84.6%, 22 of 26) of the HMV K. pneumoniae from marine mammals shared the same rpoB>gyrB sequence type (Fig. 1).

Figure 1

Maximum likelihood tree containing Klebsiella pneumoniae isolates from marine mammals that were stranded or wild caught in 2015–16 from along the US Pacific Coast. Values at the nodes represent bootstrap confidence values >80, based on 1,000 pseudoreplicates. Scale bar indicates the number of substitutions per site. KATCC=Klebsiella quasipneumoniae subsp. similipneumoniae (included as an outlier); UCD=University of California Fish Pathology Lab ID number; HS=Pacific harbor seal (Phoca vitulina); CSL=California sea lion (Zalophus californianus); HP=harbor porpoise (Phocoena phocoena); +=hypermucoviscous; −=nonhypermucoviscous; stranded vs. wild=location.

Figure 1

Maximum likelihood tree containing Klebsiella pneumoniae isolates from marine mammals that were stranded or wild caught in 2015–16 from along the US Pacific Coast. Values at the nodes represent bootstrap confidence values >80, based on 1,000 pseudoreplicates. Scale bar indicates the number of substitutions per site. KATCC=Klebsiella quasipneumoniae subsp. similipneumoniae (included as an outlier); UCD=University of California Fish Pathology Lab ID number; HS=Pacific harbor seal (Phoca vitulina); CSL=California sea lion (Zalophus californianus); HP=harbor porpoise (Phocoena phocoena); +=hypermucoviscous; −=nonhypermucoviscous; stranded vs. wild=location.

Genetic fingerprints obtained from REP-PCR formed four discrete patterns (Fig. 2), demonstrating genotypic variability among K. pneumoniae from marine mammals from the US Pacific Coast, which loosely correlated with phenotype. Similar to the rpoB>gyrB concatenated sequences, REP-PCR data suggested the majority of HMV K. pneumonia from CSLs (82.6%, 19 of 25) originated from a single clonal phylogroup (Fig. 3). Although a majority (80%, 4 of 5) of non-HMV isolates also clustered together, a greater sample size is necessary to draw more robust conclusions regarding the relative genetic homogeneity of non-HMV isolates from marine mammals.

Figure 2

Repetitive extragenic palindromic PCR amplification of Klebsiella pneumoniae recovered from marine mammals that were stranded or wild caught in 2015–16 from along the US Pacific Coast. Lane 1: University of California Fish Pathology Lab (UCD) 8; Lane 2: UCD 5; Lane 3: UCD 21; Lane 4: UCD 16; Lane 5: UCD 6b; Lane 6: UCD 6a; Lane 7: UCD 14; Lane 8: UCD 7; Lane 9: UCD 17; Lane 10: UCD 18; Lane 11: UCD 10; Lane 12: UCD 15; Lane 13: UCD 13; Lane 14: UCD 22; Lane 15: UCD 12; Lane 16: UCD 9b; Lane 17: UCD 9a; Lane 18: UCD 9c; Lane 19: UCD 19; Lane 20: UCD 11; Lane 21: UCD 20; Lane 22: UCD 25; Lane 23: UCD 24a; Lane 24: UCD 24b; Lane 25: UCD 23; Lane 26: UCD 4; Lane 27: UCD 3; Lane 28: UCD 1; Lane 29: UCD 2; Lane 30: UCD 26; Lane 31: Klebsiella quasipneumoniae subsp. similipneumoniae outlier. Genetic profiles were generated using the BOX primer. L=hyperladder 50 bp; N=no template control.

Figure 2

Repetitive extragenic palindromic PCR amplification of Klebsiella pneumoniae recovered from marine mammals that were stranded or wild caught in 2015–16 from along the US Pacific Coast. Lane 1: University of California Fish Pathology Lab (UCD) 8; Lane 2: UCD 5; Lane 3: UCD 21; Lane 4: UCD 16; Lane 5: UCD 6b; Lane 6: UCD 6a; Lane 7: UCD 14; Lane 8: UCD 7; Lane 9: UCD 17; Lane 10: UCD 18; Lane 11: UCD 10; Lane 12: UCD 15; Lane 13: UCD 13; Lane 14: UCD 22; Lane 15: UCD 12; Lane 16: UCD 9b; Lane 17: UCD 9a; Lane 18: UCD 9c; Lane 19: UCD 19; Lane 20: UCD 11; Lane 21: UCD 20; Lane 22: UCD 25; Lane 23: UCD 24a; Lane 24: UCD 24b; Lane 25: UCD 23; Lane 26: UCD 4; Lane 27: UCD 3; Lane 28: UCD 1; Lane 29: UCD 2; Lane 30: UCD 26; Lane 31: Klebsiella quasipneumoniae subsp. similipneumoniae outlier. Genetic profiles were generated using the BOX primer. L=hyperladder 50 bp; N=no template control.

Figure 3

Dendrogram generated with Dice coefficient and the unweighted pair group method with arithmetic mean clustering method, showing the genetic similarity among Klebsiella pneumoniae isolates by repetitive extragenic palindromic PCR genotyping using the BOX primer (Versalovic et al. 1994). KATCC=Klebsiella quasipneumoniae subsp. similipneumoniae; UCD=University of California Fish Pathology Lab ID number; CSL=California sea lion (Zalophus californianus); HS=Pacific harbor seal (Phoca vitulina); HP=harbor porpoise (Phocoena phocoena); −=nonhypermucoviscous; +=hypermucoviscous; wild vs. stranded=location.

Figure 3

Dendrogram generated with Dice coefficient and the unweighted pair group method with arithmetic mean clustering method, showing the genetic similarity among Klebsiella pneumoniae isolates by repetitive extragenic palindromic PCR genotyping using the BOX primer (Versalovic et al. 1994). KATCC=Klebsiella quasipneumoniae subsp. similipneumoniae; UCD=University of California Fish Pathology Lab ID number; CSL=California sea lion (Zalophus californianus); HS=Pacific harbor seal (Phoca vitulina); HP=harbor porpoise (Phocoena phocoena); −=nonhypermucoviscous; +=hypermucoviscous; wild vs. stranded=location.

Antimicrobial susceptibility testing

Range and MIC for each antimicrobial agent were tested using the broth-microdilution method (Table 3). When tested with the Wilcoxon rank-sum test, the MICs for the non-HMV isolates were found to be significantly (P≤0.05) higher than the MICs for HMV isolates for enrofloxacin (P=0.036), ceftiofur (P<0.001), oxytetracycline (P=0.004), tetracycline (P=0.003), amoxicillin (P=0.021), spectinomycin (P=0.009), trimethoprim/sulfamethoxazole (P<0.001), florfenicol (P<0.001), and penicillin (P=0.003). When HMV isolates from stranded and wild-caught CSLs were compared using the Wilcoxon rank-sum test, the MICs for gentamicin (P=0.039), ceftiofur (P<0.001), tetracycline (P=0.028), amoxicillin (P=0.018), florfenicol (P=0.002), and penicillin (P=0.032) were significantly (P≤0.05) higher than those in the wild-caught group.

Table 3

Range and antimicrobial concentration that inhibited growth of 50% of the isolates (MIC50, μg/mL) of 18 different antimicrobial agents to hypermucoviscous (HMV) and non-HMV Klebsiella pneumoniae isolates from marine mammals that were stranded or wild caught along the US Pacific Coast in 2015–16 as determined by broth microdilution.a

Range and antimicrobial concentration that inhibited growth of 50% of the isolates (MIC50, μg/mL) of 18 different antimicrobial agents to hypermucoviscous (HMV) and non-HMV Klebsiella pneumoniae isolates from marine mammals that were stranded or wild caught along the US Pacific Coast in 2015–16 as determined by broth microdilution.a
Range and antimicrobial concentration that inhibited growth of 50% of the isolates (MIC50, μg/mL) of 18 different antimicrobial agents to hypermucoviscous (HMV) and non-HMV Klebsiella pneumoniae isolates from marine mammals that were stranded or wild caught along the US Pacific Coast in 2015–16 as determined by broth microdilution.a

Detection limit of HMV K. pneumoniae by fecal culture assay

Swab inocula from 10-mg/mL suspended fecal solutions reliably detected HMV K. pneumoniae from concentrations as low as 102 CFU/mg feces, with three of three replicate cultures all yielding positive results. Even at concentrations of 101 CFU/mg feces, two of three replicates grew string test–positive colonies. All inocula from spiked nuclease-free water samples were culture positive. All unspiked stool solution and nuclease-free water inocula were HMV K. pneumoniae culture negative.

DISCUSSION

We investigated the baseline prevalence of HMV K. pneumoniae in stranded marine mammals and in apparently healthy wild-caught CSL. The prevalence of HMV K. penumoniae was 1.2% among 336 stranded marine mammals tested. No isolates were recovered from PHSs, NESs, or NFSs, likely due to the small number of animals from these species that had stranded during our sampling period. Future studies evaluating larger sample sizes may give a more representative overview of prevalence of HMV K. pneumoniae in other marine mammals.

Previous epidemics of K. pneumoniae have been identified in NZSLs in New Zealand. The bacterium was found as the etiologic agent of mortality events in this species in 2001–03. Although not evaluated for the presence of HMV K. pneumoniae phenotype, these epidemics increased mortality in NZSL pups by three times the mean in nonepidemic years (Castinel et al. 2007b). Evaluation of similar epidemics in NZSLs in 2006–07 and 2009–10 demonstrated HMV K. pneumoniae infection as the cause of death in 58% of the pups presented for necropsy (Roe et al. 2015). The baseline prevalence of HMV K. pneumoniae in CSLs that we reported was much lower than the reported prevalence of K. pneumoniae and HMV K. pneumoniae for NZSLs during these epidemics (Castinel et al. 2007a; Roe et al. 2015), which is likely a function of our samples being collected at a time at which there was no active epizootic event, the inclusion of multiple age classes, and sampling of live and dead animals. Larger sample sizes, particularly among species other than CSLs, and extended sampling times over periods of several years targeting at-risk age classes such as young pups, are needed to evaluate the dynamics of HMV K. pneumoniae emergence in marine mammals of the US Pacific Coast.

Based on complications HMV K. pneumoniae can cause in human nosocomial settings (Podschun and Ullmann 1998; Shon and Russo 2012), the prevalence of this organism in stranded CSLs and in wild-caught CSLs was compared. No significant difference was found between the two. Stranded animals originated from a wild population of CSLs and were sampled only at admission to TMMC. Therefore, it is likely the two populations would have similar prevalence of HMV K. pneumoniae. Evaluation of the incidence of this infection and its ability to cause clinical disease in marine mammal hospital populations can help elucidate the importance of HMV K. pneumoniae as a nosocomial agent.

Recent studies using isolates recovered from human patients have demonstrated the rmpA and rmpA2 genes, either chromosomally or plasmid encoded, are closely associated with hypermucoviscosity (Hsu et al. 2011). All but one of the HMV isolates recovered from marine mammals possessed the plasmid form of both rmpA and rmpA2. Interestingly, one of the five non-HMV isolates possessed both p-rmpA and p-rmpA2. Similar results regarding the hypermucoviscosity phenotype have been found when comparing K1/K2 and non-K1/K2 isolates recovered from liver abscesses in human patients (Yu et al. 2008).

Capsular serotypes K1 and K2 have been most frequently associated with HMV K. pneumoniae infection in humans (Fang et al. 2007; Yu et al. 2008). However, of the 77 capsular types identified for K. pneumoniae, there are several associated with hypermucoviscosity in addition to types K1 and K2 (Turton et al. 2010). Roe et al. (2015) reported all HMV K. pneumoniae isolates recovered from NZSLs belonged to the capsular K2 serotype. Similar findings were reported by Jang et al. (2010) in CSLs. In the current study, most of the HMV isolates belonged to the K2 type. However, two HMV isolates were negative for K1 and K2, suggesting other capsular types may be associated with the HMV phenotype in marine mammals.

In addition to capsular types, the wcaG gene is a putative K. pneumoniae virulence factor thought to be involved in macrophage evasion (Turton et al. 2010). The use of wcaG as a potential marker for HMV K. pneumoniae virulence in marine mammals has not been evaluated. All but one of the isolates recovered from marine mammals in our study was negative for wcaG, suggesting wcaG is not likely a reliable indicator of HMV K. pneumoniae type in the species we tested.

There was limited genotypic and phenotypic variability among K. pneumoniae isolates based on the markers that we used. Based on rpoB>gyrB concatenated sequences and REP-PCR, the majority of HMV K. pneumoniae from CSLs seem to originate from a single clonal genetic phylogroup, with some geographic exceptions. Similarly, a majority of non-HMV isolates also cluster together. Although promising, more representative isolates are required to make inferences regarding the relative genetic homogeneity of HMV and non-HMV isolates and utility of REP-PCR in virulence screening.

In humans, there has been increasing concern that many of the current strains have developed antimicrobial resistance (Elemam et al. 2009) including recent infections with resistance to all antibiotics tested (Chen et al. 2017). Most of the hypervirulent strains, however, have been susceptible to many common antimicrobial agents, with the exception of ampicillin (Lin et al. 2010). This is consistent with current findings. When the MICs for non-HMV isolates was compared with those of HMV K. pneumoniae isolates, non-HMV isolates had higher MICs for many of the antimicrobial agents evaluated. One consideration however is the relatively small sample size of non-HMV isolates available for comparison and the potential for outliers to influence analysis and give the appearance of greater resistance. Sample source must also be taken into consideration, as microenvironments can affect the antimicrobial resistance pattern of many bacteria. Evaluation of more isolates from representative sample sources is needed to determine whether HMV strains in the marine environment are developing resistance profiles similar to previously studied human strains. The presence of this emerging pathogen in marine mammals and the potential acquisition of antimicrobial resistance, however, are causes for concern (Li et al. 2014), given the ability of HMV K. pneumoniae to combine hypervirulence with multidrug resistance (Shon and Russo 2012).

In our study, all stranded CSL HMV K. pneumoniae isolates exhibited a ceftiofur MIC (≤0.5 μg/mL). This was consistent with doses of ceftiofur typically administered to CSLs at TMMC to treat suspected cases of bacterial pneumonia (Meegan et al. 2013). Therefore, ceftiofur may be an appropriate initial choice for treatment of suspected HMV K. pneumoniae infections in CSLs originating from the US Pacific Coast. However, confirmation of clinical effectiveness of this antimicrobial agent needs to be established with the collection of individual case susceptibility data. Although the sample size was very small, the MICs for HMV wild-caught CSL isolates were significantly higher than those of HMV stranded CSL isolates for several antibiotics including ceftiofur, amoxicillin, tetracycline, gentamicin, florfenicol, and penicillin.

No studies have tested the limit of detection of HMV K. pneumoniae in marine mammal stool samples by using the methods used in this study. Cultivation methods can consistently detect 101 CFU/g feces of Salmonella in CSL feces (Berardi et al. 2014). In our analysis, a 10-mg/mL fecal solution was used for spiking to closely approximate the amount of fecal material that would be obtained on a fecal swab. Based on our results, detection with cultivation techniques is quite sensitive, as it could reliably detect HMV K. pneumoniae at concentrations as low as 102 CFU/mg. This seems to be a reliable method for detecting the presence of this bacteria in CSL. This is particularly important as most of the HMV K. pneumoniae recovered from marine mammals in our laboratory do not grow well on the MacConkey agar typically used to recover Enterobacteriaceae from contaminated sites (Jang et al. 2010). Biochemical comparisons between HMV K. pneumoniae isolated from different species of marine mammals, including isolates from NZSL (Roe et al. 2015) are warranted to clarify the nutrient requirements of these isolates and to generate better diagnostic methods for this elusive pathogen.

Historically, HMV K. pneumoniae has been of minimal concern to the health of aquatic mammals; however, the recent emergence of this pathogen as a cause of clinical infection in marine mammal rehabilitation centers is alarming (Jang et al. 2010). Further detailed phenotypic and genotypic characterization of the isolates examined in this study, and surveys of the prevalence of HMV K. pneumoniae in stranded and wild-caught populations, has advanced our understanding of HMV K. pneumoniae dynamics in marine mammals of the US Pacific Coast. In addition, these data suggest the rmpA gene has potential as a molecular marker for HMV K. pneumoniae and molecular diagnostic method for this phenotype. The prevalence of HMV K. pneumoniae in other aquatic animals and its persistence in aquatic environments is still unknown.

ACKNOWLEDGMENTS

We thank the members of the University of California–Davis Aquatic Animal Health Laboratory, the Department of Medicine and Epidemiology in the University of California–Davis School of Veterinary Medicine, and the Marine Mammal Center. Funding was provided by the National Science Foundation (OCE-1335657). We thank Irina Udaltsova, Ashley Hill, and Janet Foley (School of Veterinary Medicine, University of California–Davis) and Maris Brenn-White, Ketki Raje, and Yehonatan Berkowic for support.

LITERATURE CITED

Berardi
T,
Shapiro
K,
Byrne
BA,
Miller
W.
2014
.
Prevalence and characterization of Salmonella shed by captive and free-range California sea lions (Zalophus californianus) from a rehabilitation center and three state reserves along the California coast
.
J Zoo Wildl Med
45
:
527
533
.
Castinel
A,
Duignan
PJ,
Pomroy
WE,
Lopez-Villalobos
N,
Gibbs
NJ,
Chilvers
BL,
Wilkinson
IS.
2007
a
.
Neonatal mortality in New Zealand sea lions (Phocarctos hookeri) at Sandy Bay, Enderby Island, Auckland Islands from 1998 to 2005
.
J Wildl Dis
43
:
461
474
.
Castinel
A,
Grinberg
A,
Pattison
R,
Duignan
P,
Pomroy
B,
Rogers
L,
Wilkinson
I.
2007
b
.
Characterization of Klebsiella pneumoniae isolates from New Zealand sea lion (Phocarctos hookeri) pups during and after the epidemics on Enderby Island, Auckland Islands
.
Vet Microbiol
122
:
178
184
.
Chen
L,
Todd
R,
Kiehlbauch
J,
Walters
M,
Kallen
A.
2017
.
Notes from the field: Pan-resistant New Delhi metallo-beta-lactamase-producing Klebsiella pneumoniae—Washoe County, Nevada, 2016
.
MMWR Morb Mortal Wkly Rep
66
:
33
.
Doud
MS,
Grimes-Zeppegno
R,
Molina
E,
Miller
N,
Balachandar
D,
Schneper
L,
Poppiti
R,
Mathee
K.
2009
.
A k2A-positive Klebsiella pneumoniae causes liver and brain abscess in a Saint Kitt's man
.
Int J Med Sci
6
:
301
304
.
Edgar
RC.
2004
.
MUSCLE: Multiple sequence alignment with high accuracy and high throughput
.
Nucleic Acids Res
32
:
1792
1797
.
Elemam
A,
Rahimian
J,
Mandell
W.
2009
.
Infection with panresistant Klebsiella pneumoniae: A report of 2 cases and a brief review of the literature
.
Clin Infect Dis
49
:
271
274
.
Fang
CT,
Lai
SY,
Yi
WC,
Hsueh
PR,
Liu
KL,
Chang
SC.
2007
.
Klebsiella pneumoniae genotype K1: An emerging pathogen that causes septic ocular or central nervous system complications from pyogenic liver abscess
.
Clin Infect Dis
45
:
284
293
.
Fenollar
F,
Roux
V,
Stein
A,
Drancourt
M,
Raoult
D.
2006
.
Analysis of 525 samples to determine the usefulness of PCR amplification and sequencing of the 16S rRNA gene for diagnosis of bone and joint infections
.
J Clin Microbiol
44
:
1018
1028
.
Griffin
MJ,
Quiniou
SM,
Cody
T,
Tabuchi
M,
Ware
C,
Cipriano
RC,
Mauel
MJ,
Soto
E.
2013
.
Comparative analysis of Edwardsiella isolates from fish in the eastern United States identifies two distinct genetic taxa amongst organisms phenotypically classified as E. tarda
.
Vet Microbiol
165
:
358
372
.
Hartman
LJ,
Selby
EB,
Whitehouse
CA,
Coyne
SR,
Jaissle
JG,
Twenhafel
NA,
Burke
RL,
Kulesh
DA.
2009
.
Rapid real-time PCR assays for detection of Klebsiella pneumoniae with the rmpA or magA genes associated with the hypermucoviscosity phenotype: Screening of nonhuman primates
.
J Mol Diagn
11
:
464
471
.
Hsu
CR,
Lin
TL,
Chen
YC,
Chou
HC,
Wang
JT.
2011
.
The role of Klebsiella pneumoniae rmpA in capsular polysaccharide synthesis and virulence revisited
.
Microbiology
157
:
3446
3457
.
Jang
S,
Wheeler
L,
Carey
RB,
Jensen
B,
Crandall
CM,
Schrader
KN,
Jessup
D,
Colegrove
K,
Gulland
FMD.
2010
.
Pleuritis and suppurative pneumonia associated with a hypermucoviscosity phenotype of Klebsiella pneumoniae in California sea lions (Zalophus californianus)
.
Vet Microbiol
141
:
174
177
.
Kong
Q,
Beanan
JM,
Olson
R,
MacDonald
U,
Shon
AS,
Metzger
DJ,
Pomakov
AO,
Russo
TA.
2012
.
Biofilm formed by a hypervirulent (hypermucoviscous) variant of Klebsiella pneumoniae does not enhance serum resistance or survival in an in vivo abscess model
.
Virulence
3
:
309
318
.
Lederman
ER,
Crum
NF.
2005
.
Klebsiella liver abscess: a coast-to-coast phenomenon
.
Clin Infect Dis
41
:
273
.
Li
W,
Sun
G,
Yu
Y,
Li
N,
Chen
M,
Jin
R,
Jiao
Y,
Wu
H.
2014
.
Increasing occurrence of antimicrobial-resistant hypervirulent (hypermucoviscous) Klebsiella pneumoniae isolates in China
.
Clin Infect Dis
58
:
225
232
.
Lin
JC,
Koh
TH,
Lee
N,
Fung
CP,
Chang
FY,
Tsai
YK,
Ip
M,
Siu
LK.
2014
.
Genotypes and virulence in serotype K2 Klebsiella pneumoniae from liver abscesses and non-infectious carriers in Hong Kong, Singapore and Taiwan
.
Gut Pathog
6
:
21
.
Lin
YT,
Jeng
YY,
Chen
TL,
Fung
CP.
2010
.
Bacteremic community-acquired pneumonia due to Klebsiella pneumoniae: Clinical and microbiological characteristics in Taiwan, 2001–2008
.
BMC Infect Dis
10
:
307
.
Meegan
J,
Collard
WT,
Grover
GS,
Pussini
N,
Van Bonn
WG,
Gulland
FM.
2013
.
Pharmacokinetics of ceftiofur crystalline-free acid (EXCEDE sterile suspension®) administered via intramuscular injection in wild California sea lions (Zalophus californianus)
.
J Zoo Wildl Med
44
:
714
720
.
National Center for Biotechnology Information
.
2017
.
Nucleotide basic local alignment search tool (BLASTn)
.
US National Library of Medicine
,
Bethesda, Maryland
. .
Nei
M,
Kumar
S.
2000
.
Molecular evolution and phylogenetics
.
Oxford University Press
,
Oxford, UK
,
352
pp
.
Podschun
R,
Ullmann
U.
1998
.
Klebsiella spp. as nosocomial pathogens: Epidemiology, taxonomy, typing methods, and pathogenicity factors
.
Clin Microbiol Rev
11
:
589
603
.
R Development Core Team
.
2017
.
R: A language and environment for statistical compound. R Foundation for Statistical Computing
.
Vienna
,
Austria
.
http://www.R-project.org. Accessed February 2017
.
Roe
WD,
Rogers
L,
Pinpimai
K,
Dittmer
K,
Marshall
J,
Chilvers
BL.
2015
.
Septicaemia and meningitis caused by infection of New Zealand sea lion pups with a hypermucoviscous strain of Klebsiella pneumoniae
.
Vet Microbiol
176
:
301
308
.
Shon
AS,
Bajwa
RP,
Russo
TA.
2013
.
Hypervirulent (hypermucoviscous) Klebsiella pneumoniae: A new and dangerous breed
.
Virulence
4
:
107
118
.
Shon
AS,
Russo
TA.
2012
.
Hypervirulent Klebsiella pneumoniae: The next superbug?
Future Microbiol
7
:
669
671
.
Tamura
K,
Stecher
G,
Peterson
D,
Filipski
A,
Kumar
S.
2013
.
MEGA6: Molecular evolutionary genetics analysis version 6.0
.
Mol Biol Evol
30
:
2725
2729
.
Turton
JF,
Perry
C,
Elgohari
S,
Hampton
CV.
2010
.
PCR characterization and typing of Klebsiella pneumoniae using capsular type-specific, variable number tandem repeat and virulence gene targets
.
J Med Microbiol
59
:
541
547
.
Twenhafel
NA,
Whitehouse
CA,
Stevens
EL,
Hottel
HE,
Foster
CD,
Gamble
S,
Abbott
S,
Janda
JM,
Kreiselmeier
N,
Steele
KE.
2008
.
Multisystemic abscesses in African green monkeys (Chlorocebus aethiops) with invasive Klebsiella pneumoniae--Identification of the hypermucoviscosity phenotype
.
Vet Pathol
45
:
226
231
.
Versalovic
J,
Schneider
M,
De Bruijn
FJ,
Lupski
JR.
1994
.
Genomic fingerprinting of bacteria with repetitive sequence-based polymerase chain reaction
.
Methods Mol Cell Biol
5
:
25
40
.
Watts
JL,
Clinical and Laboratory Standards Institute
.
2008
.
Performance standards for antimicrobial disk and dilution susceptibility tests for bacteria isolated from animals: Approved standard, third edition
.
CLSI document VET01-A3
.
National Committee for Clinical Laboratory Standards
,
Wayne, Pennsylvania
,
pp
.
32
35
.
Yu
WL,
Ko
WC,
Cheng
KC,
Lee
CC,
Lai
CC,
Chuang
YC.
2008
.
Comparison of prevalence of virulence factors for Klebsiella pneumoniae liver abscesses between isolates with capsular K1/K2 and non-K1/K2 serotypes
.
Diagn Microbiol Infect Dis
62
:
1
6
.