Characterization of a Brucella sp. Strain as a Marine-Mammal Type despite Isolation from a Patient with Spinal Osteomyelitis in New Zealand

Transcription

1 JOURNAL OF CLINICAL MICROBIOLOGY, Dec. 2006, p Vol. 44, No /06/$ doi: /jcm Copyright 2006, American Society for Microbiology. All Rights Reserved. Characterization of a Brucella sp. Strain as a Marine-Mammal Type despite Isolation from a Patient with Spinal Osteomyelitis in New Zealand W. L. McDonald, 1 * R. Jamaludin, 1 G. Mackereth, 1 M. Hansen, 1 S. Humphrey, 1 P. Short, 2 T. Taylor, 3 J. Swingler, 3 C. E. Dawson, 4 A. M. Whatmore, 4 E. Stubberfield, 4 L. L. Perrett, 4 and G. Simmons 5 Investigation and Diagnostic Centre, Biosecurity New Zealand, Ministry of Agriculture and Forestry, Ward St., Wallaceville, New Zealand 1 ; Institute of Environmental Science and Research, Ltd., Kenepuru Science Centre, 34 Kenepuru Dr., Porirua, New Zealand 2 ; Australian Animal Health Laboratory, Geelong, Victoria, Australia 3 ; Veterinary Laboratories Agency, New Haw, Addlestone, Surrey KT15 3NB, United Kingdom 4 ; and Auckland Regional Public Health Service, Auckland, New Zealand 5 Received 30 March 2006/Returned for modification 11 June 2006/Accepted 17 September 2006 Naturally acquired infection of humans with a marine mammal-associated Brucella sp. has only been reported once previously in a study describing infections of two patients from Peru. We report the isolation and characterization of a strain of Brucella from a New Zealand patient that appears most closely related to strains previously identified from marine mammals. The isolate was preliminarily identified as Brucella suis using conventional bacteriological tests in our laboratory. However, the results profile was not an exact match, and the isolate was forwarded to four international reference laboratories for further identification. The reference laboratories identified the isolate as either B. suis or B. melitensis by traditional bacteriological methods in three laboratories and by a molecular test in the fourth laboratory. Molecular characterization by PCR, PCR-restriction fragment length polymorphism, and DNA sequencing of the bp26 gene; IS711; the omp genes omp25, omp31, omp2a, and omp2b; IRS-PCR fragments I, III, and IV; and five housekeeping gene fragments was conducted to resolve the discrepant identification of the isolate. The isolate was identified to be closely related to a Brucella sp. originating from a United States bottlenose dolphin (Tursiops truncatus) and common seals (Phoca vitulina). Brucellosis is an important zoonotic disease of humans, causing a variety of vague symptoms including undulant fever, fatigue, malaise, joint pain, myalgia, depression, and anorexia (22). Chronic sequelae and recrudescence decades after initial infection also occur. Brucella may be transmitted from animals to humans by direct contact with infected animals, ingestion of infected food products, and inhalation of aerosols. Four species of Brucella are the primary causes of infection in humans. Brucella melitensis is highly infectious and is transmitted from sheep and goats, B. abortus is transmitted from cattle, B. suis is transmitted from pigs and, infrequently, B. canis is transmitted from dogs. Other species of Brucella have been rarely or not reported to infect humans. There are only two reports in the literature of humans infected with marine mammal strains of Brucella. One report was of a laboratory worker who displayed symptoms consistent with brucellosis (4). The infection was confirmed by a positive serological response, isolation, and PCR-restriction fragment length polymorphism (RFLP) identification of a marine Brucella strain. Two patients originating from Peru and diagnosed with neurobrucellosis were also confirmed to be infected with * Corresponding author. Mailing address: Investigation and Diagnostic Centre, Biosecurity New Zealand, Ministry of Agriculture and Forestry, Ward St., Wallaceville, New Zealand. Phone: 61(0) Fax: 61(0) wendy.mcdonald@maf.govt.nz. Present address: Tropical and Aquatic Animal Health Laboratory, River Blvd., Townsville, Queensland, Australia. Published ahead of print on 11 October marine mammal strains of Brucella by isolation, PCR, and DNA sequencing (41). The two Peruvian patients were not laboratory workers, and the infection was naturally acquired. Serological evidence and isolation of brucellae have been reported from a variety of marine mammals on numerous occasions from locations in the northern hemisphere. The serological prevalence ranges from 0 to 38% for cetaceans, pinnipeds, and mustelids (6, 24, 26, 29, 32, 34, 43). The largest studies of 1,855 pinnipeds from North America and 1,386 pinnipeds and cetaceans from the North Atlantic Ocean revealed 3.1 and 8.2%, respectively, to have serological evidence of brucella (33, 43). Brucella have also been isolated from 31% (54/175) of marine mammals sampled under a variety of circumstances (8, 14, 17, 19, 23, 29, 31, 40, 43, 44). There are few reports of serological testing of pinnipeds and cetaceans from the Southern Hemisphere. In the Antarctic, positive reactions were detected in 3.5% (6/17) of pinnipeds (36). A high prevalence of serological reactions (55.2% [32/58]) was reported in cetaceans off the coast of central Peru (45). In pinnipeds of Australia, serological reactions were reported in three species, including 75% (9/12) of sea lions (Neophoca cinerea) (13). In New Zealand, no serological reactions were detected in 101 New Zealand sea lions (Arctocephalus hookeri) (28). Brucella species isolated in New Zealand include B. ovis from sheep and deer, B. abortus (eradicated from livestock in 1989), and B. melitensis and B. suis (both from human patients). Infection of humans with the latter two species was confirmed as having been acquired overseas (7, 39). Hitherto, 4363

2 4364 MCDONALD ET AL. J. CLIN. MICROBIOL. TABLE 1. Results of brucella serology tests performed on samples from the patient at two New Zealand medical laboratories (W and M) Result (days from onset) Test Screen a SAT titer Coombs titer W M W b M W M First (29) (36) 1:1,280 (29)* 1:20 (36) 1:2,560 (29) 1:640 (36) Second NT (153) 1:320 (40)* 1:40 (153) 1:1,280 (40) 1:160 (153) a NT, not tested., Positive;, negative. b *, SAT titers using 2-mercaptoethanol were 1:160 and 1:80 in the first and second tests, respectively. marine mammal Brucella strains had not been isolated in New Zealand. However, serum and tissue samples of stranded Hector s dolphins (Cephalorhynchus hectori) tested at our laboratory, the Investigation and Diagnostic Centre (IDC), have revealed serological and molecular evidence of brucella infection (unpublished data). The present study describes the phenotypic and molecular characterization of a strain of Brucella isolated from a New Zealand patient that is most closely related to isolates previously identified from marine mammals. Case report. In February 2002, a 43-year-old male patient from South Auckland in the North Island of New Zealand presented with 2 weeks of symptoms of spinal osteomyelitis fever, rigors, and lumbar spinal tenderness. Blood samples were tested at two New Zealand medical laboratories (M and W) by a screening agglutination test using B. abortus antigen (Immunostics, Inc.; Remel), a serum agglutination test (SAT) using B. abortus antigen (Murex) with or without the addition of 2-mercaptoethanol to the diluent, and a Coombs anti-brucella test using anti-human globulin (CSL Epiclone). The patient tested positive with all three serological tests within 36 days of onset of illness and tested negative in the SAT and with a declining titer in the Coombs test 153 days from onset of illness (Table 1). Magnetic resonance imaging revealed multifocal inflammation involving lumbar vertebrae 1 and 4 and also the right ilium. Aspiration of the right iliac lesion failed to isolate an organism, but after 3 days blood cultures grew a gram-negative bacillus. Treatment had consisted of intravenous flucloxacillin, which was changed to intravenous ceftriaxone after 3 days based on culture sensitivities. The patient was treated with intravenous ceftriaxone for a total of 6 weeks and made a complete recovery. The organism (02/611) was misidentified using traditional bacteriological methods and reported in April 2002 as B. suis (3). Acquisition. The patient had last traveled overseas 10 years prior to illness. There was no previous history of a similar illness. The patient had no exposure to unpasteurized milk or cheese and had no occupational exposure to meat or meat products. The patient had dressed two freshly killed pigs in December 2001, which led to the belief that the patient acquired infection while dressing the pigs. Clinical and serological surveys of the associated pig population and pig farmers were undertaken. No other cases were confirmed. After molecular characterization of the organism (02/611), the patient was reinterviewed to determine a possible source of the marine mammal source of the Brucella sp. The patient had not been exposed to marine mammals. The patient had been exposed to uncooked fish bait, including pilchards (Sardina and Sardinops spp.), bonito (Sarda spp.), squid (Nototodarus sp.), and mullet (Mugilidae) and regularly caught snapper (Chrysophrys auratus), kahawai (Arripis trutta), trevally (Pseudocaranx dentex), and gurnard (Chelidonichthys kumu). The patient consumed raw freshly caught snapper with lemon juice and coconut cream. MATERIALS AND METHODS Identification strategy. A range of phenotypic tests in addition to DNA sequencing of a fragment of the omp25 gene were undertaken in our laboratory in order to identify the isolate (02/611). After we obtained the results of the in-house testing (IDC), the isolate was forwarded for confirmation to four reference laboratories, designated A, B, C, and D. Upon receipt of the results from the reference laboratories, further characterization of the isolate was undertaken using molecular techniques in our laboratory (IDC) and the Veterinary Laboratories Agency (United Kingdom) (VLA). Brucella isolates. The Brucella species examined in the present study were B. abortus bv. 2 (Australian Animal Health Laboratory), B. abortus bv. 3 (Central Animal Health Laboratory, New Zealand strain Tulya), B. abortus bv. 9 (CAHL strain C68), B. ovis (AgResearch; NZ00/1614), B. canis (ATCC 23365), B. suis bv. 3 (AAHL), and B. melitensis bv. 2 (Environmental Science and Research, Ltd.; NZ02/820). Eight Brucella spp. of marine origin (UK1/03 and UK39/94 isolated from common seals [Phoca vitulina]; UK61/94 isolated from a gray seal [Halichoerus grypus]; UK9/02 and 36/94, both isolated from harbor porpoises [Phocoena phocoena]; F5/99 isolated from a bottlenose dolphin [Tursiops truncatus United States], UK22/00 isolated from a white-sided dolphin [Lagenorphynchus acutus] and 59/94 isolated from a striped dolphin [Stenella coeruleoalba]) were all supplied by the VLA. Culture. All brucella isolates were stored at 70 C on cryobeads. The isolates were resuscitated using brucella enrichment broth (Farrells medium) and incubated at 37 C for 3 to 7 days. The broths were subcultured onto serum dextrose agar and chocolate blood agar and incubated at 37 C in O 2 and CO 2 (1, 28). Phenotypic testing. Phenotypic testing was performed in our laboratory (IDC) using standard conventional tests for brucella (1, 30). This included growth requirements, biochemical tests, dye sensitivity, serotyping, and phage typing. PCR, RFLP, and DNA sequencing. DNA was extracted from the brucella isolates by using the QIAMP DNA minikit (QIAGEN, Hilden, Germany), according to the manufacturer s instructions. The genetic targets for molecular characterization were the omp31 gene (differentiation of B. abortus); the IS711 element downstream of bp26 (differentiation of terrestrial and marine strains); the genes omp25, -2a and -2b; IS711; housekeeping genes; and specific IRS-PCR DNA fragments (differentiation of species). The omp25 gene fragment was initially amplified and sequenced using oligonucleotides designed in-house (IDC). A 50- l PCR mixture contained 1 PCR buffer (Roche, Basal, Switzerland), 2.5 U of Taq polymerase (Roche, Basal, Switzerland), 0.2 mm deoxynucleoside triphosphate, 0.5 mm MgCl 2, 500 ng of forward primer BC1-GTTGAAGTAGCTCCCCAGTA (nucleotide position 109 of GenBank submission BMU33003) and 500 ng of reverse primer BC2-ACTG GGTGTAACGGTACTCA (position 431 BMU33003), and 5 l of genomic DNA. The PCR was conducted in a Mastercycler gradient (Eppendorf, Hamburg, Germany) or an MJ Research Minicycler (Bio-Rad, Waltham, MA) with a denaturation step of 96 C for 60 s; followed by 30 cycles of 96 C for 40 s, 60 C for 60 s, and 72 C for 60 s; and with a final extension step of 72 C for 5 min. The PCR product was purified and sent to the University of Waikato (New Zealand) for sequencing. Chromatograms were checked for miscalls, and the identity of

3 VOL. 44, 2006 CHARACTERIZATION OF BRUCELLA SP TABLE 2. Comparison of biochemical, serotyping, phage typing, growth in the presence of dyes, and other growth requirements of the brucella isolate 02/611 from four laboratories with results reported in the literature for marine isolates of brucella a Test or test laboratory location Result for marine mammal brucella the DNA sequences was analyzed by comparison to DNA sequences within GenBank using BLASTN (2; The omp25 gene was later amplified using the primers 25A and 25B that encompassed the in-house designed primers and provided additional single nucleotide polymorphisms for RFLP testing and DNA sequencing (11). A 50- l PCR mixture contained 1 HotStar master mix (QIAGEN), 0.1 M concentrations of each primer, and 5 l of genomic DNA. The PCR was conducted with a denaturation step at 95 C for 15 min; followed by 35 cycles of 95 C for 1 min, 58 C for 2 min, and 70 C for 3 min; followed by a final extension step of 70 C for 10 min. The amplified fragment was digested with EcoRV. A 20- l digestion reaction contained 5 l of DNA, 10 U of enzyme (New England Biolabs, Beverly, MA; Promega Corp., Madison, WI), 100 g of bovine serum albumin, and the corresponding 1 buffer for at least 3hat37 C in a water bath. Digests were electrophoresed in a 3% agarose gel in 1 TAE (40 mm Tris acetate 2 mm EDTA). The bp26 gene and downstream IS711 element was amplified using the primers 26A and 26B (9). The reaction mixture was as described for the second omp25 PCR. The PCR was conducted with a denaturation step of 95 C for 15 min; followed by 35 cycles of denaturation at 94 C for 30 s, annealing at 58 C for 60 s, and extension at 72 C for 60 s; followed by a hold at 72 C for 5 min. The omp31 gene was amplified using the primers 31sd and 31ter (46). The reaction mixture and cycling parameters were as described for the bp26 PCR. Result for bottlenose dolphin (F5/99) Result at laboratory: IDC (repeat) A (repeat) B C CO 2 dependency (38/59) b c H 2 S production (29/29) w( ) Oxidase (13/13) Catalase (19/19) Urease production (40/40) d Growth on: Thionin (1:25,000) (9/9) Thionin (1:50,000) (34/34) Thionin (1:100,000) (4/4) Basic fuchsin (1:50,000) (34/34) Basic fuchsin (1:100,000) / (3/3) Safranin O (1:5,000) (5/5) w Safranin O (1:10,000) (33/33) Thionin blue (1 mg/ml) (8/8) I-erythritol (1 mg/ml) (2/2) (5/5, 2 g/ml) Penicillin (5 IU/ml) (8/8) (10 m disk) DL-Ornithine decarboxylase (22/22) e ( ) L-Arginine dihydrolase (22/22) e ( ) L-Lysine decarboxylase (22/22) e ( ) D-Galactose (11/22) f Agglutination with: Monospecific sera A (22/41) Monospecific sera M (2/41) (? rough) Sera A and M (17/41) Weybridge, United Kingdom g NL (7/28) CL PL (NL) NL NL PL (17/28) CL (4/28) Tbilisi, Georgia g NL (27/30) NL NL NL NL PL (3/30) Berkeley, CA g PL (5/22) CL (NL) CL CL (17/22) Firenze, Italy g NL (13/27) NL (NL) NL PL (14/27) R/C gh NL (28/28) CL NL (plq) NL a, Positive;, negative. NL, no lysis; PL, partial lysis; CL, complete lysis; plq, evident plaques. b Brucella isolates from pinnipeds require CO 2. c Growth was enhanced by CO 2. d Positive within 10 min. e Tetrazolium reduction. f Brucella isolates from cetaceans utilize D-galactose. g Phages were tested at the routine test dilution. h R/C, is a derivative of phage R; phage R was obtained by simultaneous incubation of a number of brucella phages with B. abortus. The omp2a gene was amplified using the primers 2aA and 2aB (11). The reaction mixture and cycling parameters were as described above for the second omp25 PCR. The amplified DNA was digested by using four restriction endonucleases: HinfI, PstI, AluI, and BanI. The digestion mixture and incubation was as described for the omp25 digestion. The omp2b gene was amplified using the primers 2bA and 2bB and the conditions described previously (11). The amplified DNA was digested by using six restriction endonucleases: BglII, EcoRI, HaeIII, HinfI, KpnI, and TaqI. The omp2a and -2b PCR products from 02/611 and F5/99 were sent to Monash University (Australia) for sequencing. After editing and compilation of the contigs, the omp2a and -2b DNA sequences were phylogenetically analyzed by creating a phylogram from a neighbor-joined consensus tree of Kimura distance matrices created from 500 bootstraps of the CLUSTAL W aligned sequences (16, 42) (BioManager by ANGIS [ TreeView [Win32]; For the IS711 RFLP analysis, genomic DNA was digested with EcoRI (5). Digested DNA was separated on 0.8% agarose gels and Southern blotted by using a vacuum method. The membrane-bound DNA was probed using a digoxigenin (DIG)-labeled IS711 probe, and the hybridized probe was detected using anti-dig monoclonal antibodies. The membrane was immersed in CSPD chemiluminescence substrate to allow visualization of the probe hybridization by exposure to X-ray film. The resulting images were analyzed by using Bionumerics

4 4366 MCDONALD ET AL. J. CLIN. MICROBIOL. software (Applied Maths, Belgium), and a dendrogram analysis of the profiles was produced using the coefficient of Jaccard to calculate similarities between the fingerprint patterns. Isolate 02/611 was also characterized by using a multilocus sequence typing approach currently under development at the VLA (A. M. Whatmore, L. L. Perrett, and A. P. Macmillan, submitted for publication). Fragments of six genes (dnak, gyrb, glk, trpe, cobq, and omp25) were amplified by PCR using a proofreading polymerase and sequenced from either end. The sequences obtained were compared to a database of equivalent sequences from 200 strains representing all known Brucella species and biovars. The IRS-PCR fragments were amplified using the primer pairs I1/I2 (PCR I), III1/III2 (PCR III), and IV1/IV2 (PCR IV) (10). The reaction mixture was the same as described for the second omp25 PCR. The PCR was conducted with a hold step at 95 C for 15 min; followed by 40 cycles of 95 C for 1 min, 62 C for 2 min, and 72 C for 2 min; followed by a final hold at 72 C for 10 min. RESULTS Phenotypic characterization. Isolate 02/611 is a gram-negative coccobacillus that is nonmotile, grows aerobically, and is oxidase, catalase, and urease positive, features typical of the genus Brucella (Table 2). The rapid metabolism of urea was suggestive of B. suis and with positive results for L-arginine dihydrolase, L-lysine decarboxylase, DL-ornithine decarboxylase, and growth on I-erythritol, the isolate was tentatively identified as B. suis, although it was atypical in that it grew on basic fuchsin at 1:25,000. To confirm the identity of the brucella isolate, it was forwarded to four reference laboratories, and the results of their analyses are presented in Table 1. Reference laboratory A identified the isolate as B. suis biovar 3, and laboratories B and C identified it as B. melitensis biovar 1. Molecular characterization. At the time of the submission of isolate 02/611, brucella isolates were routinely tested in our laboratory (IDC) using the in-house designed omp25 PCR, and sequencing BlastN analysis conducted in April 2002 of the partial DNA sequence from the forward primer (BC1) revealed that the isolate was most closely related to the following Brucella spp. (no. of matching nucleotides): B. suis (398/403), B. melitensis (398/403), B. abortus (398/403), B. neotomae (398/ 403), B. ovis (397/403), and B. canis (396/404). The E value was 0 for each Brucella spp. The next lowest value was 1e-126 for Ochrobacterum anthropi. Reference laboratory D returned an identification of B. suis biovar 5 using molecular tests; however, details on the methodology were not forthcoming from this laboratory. Further molecular characterization was undertaken in our laboratory (IDC), including an updated BLASTN analysis conducted in June omp25 sequences from marine isolates of brucella were submitted to GenBank in September The latter BLASTN analysis of the compiled sequence obtained with the forward and reverse primers (25A and 25B) now revealed 02/611 to be a 100% match (642/642 nucleotides) to B. cetaceae accession numbers AY and AY and B. pinnipediae accession number AY The omp25 PCR amplified a 700-bp DNA fragment from all Brucella spp., and upon digestion with EcoRV, only the B. melitensis isolate remained uncut. Isolate 02/611 produced a P1 pattern (11). The bp26 PCR amplified a 1,900-bp fragment from brucella isolates from marine mammals, including pinnipeds and cetaceans, and a smaller fragment of 1,029 bp was amplified from FIG. 1. Restriction digestion of omp2a PCR product with HinfI (A), PstI (B), AluI (C), and BanI (D). Lanes: 1, B. abortus bv. 2; 2, B. abortus bv. 9; 3, B. ovis; 4,B. canis; 5,B. suis bv. 3; 6, B. melitensis; 7, 02/611; 8, marine mammal brucella (seal UK1/03); 9, marine mammal brucella (porpoise UK9/02); 10, marine mammal brucella (United States bottlenose dolphin F5/99); 11, marine mammal brucella (striped dolphin UK59/94); 12, B. ovis; 13, VI marker (Boehringer-Mannheim).

5 VOL. 44, 2006 CHARACTERIZATION OF BRUCELLA SP FIG. 2. Phylogram comparing the omp2a and -2b genes from 02/611 with the omp2a and -2b genes from marine and terrestrial isolates of brucella. Internal nodes are labeled with the percent bootstrap values. Downloaded from brucella isolates from terrestrial animals. The amplified fragment from isolate 02/611 corresponded in size to the marine mammal isolates of brucella. The omp31 PCR amplified 900 bp from all Brucella spp. tested (including isolate 02/611) with the exception of B. abortus. Products of 1,230 bp were produced upon amplification of the omp2a and omp2b genes of all species of Brucella. The RFLP patterns produced after restriction digestion of omp2a are shown in Fig. 1. The omp2a RFLP pattern produced by the human isolate 02/611 matched the RFLP patterns produced by the isolate F5/99 originating from a bottlenose dolphin and the isolate UK1/03 originating from a common seal. The combined omp2a and -2b profile of RFLP patterns produced by 02/611 was identical to that produced by F5/99. All other Brucella spp. differed by at least one RFLP pattern. DNA sequencing of the omp2 genes of isolate 02/611 revealed that the isolate contained one omp2a and one omp2b and that both genes were 100% identical in sequence to the omp2a and -2b genes, respectively, of F5/99. The omp2a gene from 02/611 was also 100% identical to the omp2a genes from two common seal isolates (GenBank accessions AF and DQ059380) but not to the omp2b genes from these two isolates. Phylogenetic analysis showed 02/611 and F5/99 to be identical and closely related to isolates from common seals (Fig. 2). IS711 RFLP molecular characterization of isolate 02/611 produced a fingerprint profile identical to that of the F5/99 United States bottlenose dolphin isolate (Fig. 3). Each genome contained 25 copies of the IS711 element sized between 20,000 and 1,520 bp. The profiles were similar to those described for brucella isolated from marine mammals (5). Sequencing of fragments from the five housekeeping genes and omp25 revealed that the most closely related sequences were those of marine mammal isolates and B. suis. A report fully describing this technique is in preparation, but isolate 02/611 differed from each of the B. suis biovar type strains by a minimum of four base differences. Of the five distinct marine mammal genotypes seen to date, isolate 02/611 varied from four of these by two to four nucleotides but was an exact match with F5/99 (United States bottlenose dolphin) (Table 3). Despite our having examined some 50 marine mammal isolates by this approach, this genotype has only been previously seen in this single isolate. PCR I specific for B. pinnipediae amplified a 300-bp fragment from 02/611, UK1/03 (common seal), and F5/99 (United States bottlenose dolphin). PCR III and IV specific for B. cetaceae did not amplify DNA fragments from 02/611, F5/99, or UK1/03. A 400-bp fragment was amplified from UK9/02 (porpoise) using PCR III, and a 300-bp fragment was amplified from UK59/94 (striped dolphin) using PCR IV. Phenotypic reevaluation. Following the conflicting identification and results from follow-up molecular testing, reevalua- on May 4, 2018 by guest

6 4368 MCDONALD ET AL. J. CLIN. MICROBIOL. FIG. 3. IS711 fingerprint patterns of Brucella species tested in the present study. tion of the phenotypic profile for isolate 02/611 with repeat testing of L-arginine dihydrolase, L-lysine decarboxylase, and DL-ornithine decarboxylase revealed that it was consistent with the phenotypic results for marine brucella isolates reported in the literature (Table 1) (8, 14, 18, 19, 21, 25, 31, 44). In addition, isolate 02/611 produced the same phenotypic results as those reported for the brucella isolate from the United States bottlenose dolphin (13). DISCUSSION This is the second published report of humans with naturally acquired infection originating from a marine mammal strain of Brucella and the first report of a male presenting with spinal brucellosis due to a marine mammal strain. The earlier report of two males with naturally acquired infection with a marine Brucella spp. were described as having neurological symptoms (41). Marine mammal strains of Brucella have been reported to cause a range of clinical symptoms with varying severity in pinnipeds and cetaceans ranging from asymptomatic carriage to orchitis, abortions, and meningeoencephalitis (16, 23, 31, TABLE 3. Comparison of 02/611 sequences with B. suis and marine mammal brucella sequences showing all bases variable between the isolates Brucella isolate Gene polymorphism a glk dnak gyrb trpe cobq omp25 Human isolate 02/611 AGG GACC GGC AA G C B. suis bv G.T A.T GG.. B. suis bv. 2 G...G.T..T GG.. B. suis bv G.T AAT GG.. B. suis bv G.T AAT GG.. B. suis bv GA.... GG.. Marine mammal brucella 36/94..A AG..... G... Marine mammal brucella 39/94.A..G..... G... Marine mammal brucella 59/94....G..... G. A T Marine mammal brucella 61/94....G..... G... Marine mammal brucella F5/ a Residues identical to 02/611 are indicated as a dot. 34). Similarly, there is evidence to suggest that marine mammal strains of Brucella can cause a range of symptoms in humans, including headaches, malaise, severe sinusitis, seizures (4, 41), and spinal osteomyelitis. It is also possible that marine mammal strains of Brucella may cause only mild symptoms in humans, and thus the disease goes undiagnosed or misdiagnosed. In this report the patient was symptomatic and serologically positive since testing was within the early phase of infection. However, in the previous report only one of the males had a single positive titer to brucella; both males had a longer history of symptoms when tested (41). Zoonotic disease of humans with brucella has been commonly associated with B. melitensis, B. abortus, B. suis, and B. canis. Given the reported rarity of infection of humans with marine mammal strains of Brucella, and the high-risk exposure to pigs within the incubation period of disease, it is not surprising that initial interpretation of the phenotypic results led to a preliminary identification of B. suis. After results were obtained from the molecular tests, a review of the phenotypic results found these consistent with identification of the isolate as a marine mammal strain of Brucella. Classification of marine mammal strains of Brucella as B. pinnipediae and B. cetaceae has been proposed as species originating from pinnipeds and cetaceans, respectively (12). These two species can generally be differentiated on the basis of CO 2 dependency (B. pinnipediae) and metabolism of D-galactose (B. cetaceae) (20). The human isolate most closely resembled the isolate from the United States bottlenose dolphin, and both would be classified as B. cetaceae based on the phenotypic results. The phenotypic test results do not correlate with the results from the specific PCR test (I) and sequencing of the omp2a gene since both genotypic test results indicate that the human and bottlenose dolphin isolates are more closely related to B. pinnipediae than B. cetaceae. In addition, the variation in results obtained from repeat biochemical testing indicates that identification to the species level using phenotypic traits may not be reliable. Restriction patterns of the PCR-amplified brucella omp2a

7 VOL. 44, 2006 CHARACTERIZATION OF BRUCELLA SP gene of isolate (02/611) produced an overall pattern classification of I as described previously (12). Pattern I has been associated with isolates from a hooded seal (Cystophora cristata), a gray seal (Halichoerus grypus), 10 common seals, and an otter (Lutra lutra), whereas cetacean isolates have produced overall patterns classified as J or K (12). Restriction patterns of the PCR-amplified brucella omp2b gene of isolate 02/611 produced an overall pattern classification similar to O and P; however, an EcoRI pattern 3 was observed which has not been previously noted. Pattern O has been associated with two common seals, and pattern P has been associated with a hooded seal. Overall, patterns L, M, and N have been associated with cetacean isolates, including a bottlenose dolphin with a complete omp2a, opm2b profile as M-J. It is interesting that the overall patterns of the bottlenose dolphin (United States F5/ 99) tested in the present study also matched the I (O/P) profile. The omp2a and -2b genes are 85% homologous, and both the human isolate (02/611) and the bottlenose dolphin isolate (F5/ 99) were determined to have one omp2a gene and one omp2b gene. The majority of isolates from pinnipeds have also been found to contain one omp2a and one omp2b gene compared to the majority of isolates from cetaceans that have two omp2b genes (12). The close relationship of isolate 02/611 and the marine mammal isolate F5/99 was supported by two other approaches. First, the two isolates shared an identical profile as determined by IS711 fingerprinting. Second, the isolates were identical to each other based on housekeeping gene sequence data, and the most closely related organisms are other marine mammal brucella. Thus, evidence from a variety of approaches supports the observation that the closest known relative of isolate 02/611 is an isolate originating from a marine mammal. The 02/611- F5/99 genotype has never been reported elsewhere, so we cannot be certain about its natural distribution and prevalence. The results from the I, III, and IV PCR tests indicate that both the human isolate and the United States bottlenose dolphin are closely related to B. pinnipediae. The brucella isolates from the two Peruvian patients were found to be identical to a Brucella sp. originating from a common seal (41). The source of the brucella infection of the Peruvians was not identified. The origin of infection of the New Zealand patient is under investigation. In the absence of evidence of contact with marine mammals, the investigation has focused on other possible infection pathways. Brucellae have been named for their pathogenicity and host preference, but it is possible for them to cross the species barrier (15, 17, 27, 38). Cattle and sheep have been experimentally infected with marine mammal strains of Brucella. An isolate from a Pacific harbor seal (Phoca vitulina richardsi) caused abortion in one-third of the infected cows (37). Diagnosis was based on histopathological changes, the isolation of brucella from the cows that aborted, and 100% seroconversion, albeit transient (37). In seropositive cows that did not abort, brucella was not isolated from a wide range of tissues, indicating that the infection was transient (37). Sheep inoculated with an isolate from a seal produced only low antibody levels that were also transient, with isolation of the Brucella strain from 1 of 16 aborted ewes and her fetus (35). Similarly, piglets inoculated with isolate 02/611 produced low and transient antibody levels in 3 of 10 piglets, with the other 7 not seroconverting (J. Bingham, unpublished data). The organism was isolated post mortem from the retropharyngeal and mandibular lymph nodes of three piglets. On initial investigation, pigs were implicated as the likely source of infection based on the presumptive identification of the isolate as B. suis and the patient s contact with pigs in the timeframe of infection. There was no serological evidence of infected pigs in trace herds (3). However, a low-prevalence transient antibody response probably would not have been detected. Further testing of pigs for Brucella spp. is being conducted. Lymph nodes (retropharyngeal, submaxillary, gastrohepatic, internal iliac, and inguinal) collected from 74 chopper pigs sent to an abattoir from 22 pig farms were pooled and tested in our laboratory (IDC) by the nested omp25 PCR for Brucella sp. All tested negative. The New Zealand patient was known to have eaten raw snapper. The two Peruvian patients were reported to have consumed unpasteurized cheese and raw shellfish, and it was speculated that the source of their infection with brucella may have been other than marine mammals (41). This case highlights the need for follow-up testing of unusual phenotypic results of brucella isolates and proves that the molecular characterization techniques described in the literature are very useful tools for differentiating Brucella species. This case also reinforces earlier warnings that marine mammal strains of Brucella may be an emerging zoonotic disease, and consideration needs to be given to marine sources when brucella infections are being diagnosed. ACKNOWLEDGMENTS We thank Susan Taylor and Christopher Mansell for providing information on the serological testing of the patient. REFERENCES 1. Alton, G. G., L. M. Jones, R. D. Angus, and J. M. Verger Bacteriological methods, p In Techniques for the brucellosis laboratory. Insititut National de la Recherche Agronomique, Paris, France. 2. Altschul, S. F., T. L. Madden, A. A. Schäffer, J. Zhang, Z. Zhang, W. Miller, and D. J. Lipman Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25: Bramley, D., G. Simmons, S. Taylor, K. Godfrey, and G. Mackereth The public health response to a case of brucellosis in Auckland. N. Z. Public Health Rep. 9: Brew, S. D., L. L. Perrett, J. A. Stack, A. P. MacMillan, and N. J. Staunton Human exposure to Brucella recovered from a sea mammal. Vet. Rec. 144: Bricker, B. J., D. R. Ewalt, A. P. MacMillan, G. Foster, and S. Brew Molecular characterization of Brucella strains isolated from marine mammals. J. Clin. Microbiol. 38: Calle, P. P., D. J. Seagars, C. McClave, D. Senne, C. House, and J. A. House Viral and bacterial serology of free-ranging Pacific walrus. J. Wildl. Dis. 38: Carman, M Brucella abortus biovars isolated in New Zealand. Surveillance 19: Clavareau, C., V. Wellemans, K. Walravens, M. Tryland, J. M. Verger, M. Grayon, A. Cloeckaert, J. J. Letesson, and J. Godfroid Phenotypic and molecular characterization of a Brucella strain isolated from a minke whale (Balaenoptera acutorostrata). Microbiology 144: Cloeckaert, A., M. Grayon, and O. Grépinet An IS711 element downstream of the bp26 gene is a specific marker of Brucella spp. isolated from marine mammals. Clin. Diagn. Lab. Immunol. 7: Cloeckaert, A., M. Grayon, O. Grépinet, and K. S. Boumedine Classification of Brucella strains isolated from marine mammals by infrequent restriction site-pcr and development of specific PCR identification tests. Microb. Infect. 5: Cloeckaert, A., J. Verger, M. Grayon, and O. Grépinet Restriction site polymorphism of the genes encoding the major 25 kda and 36 kda outermembrane proteins of brucella. Microbiology 141: Cloeckaert, A., J. M. Verger, M. Grayon, J. Y. Paquet, B. Garin Bastuji, G. Foster, and J. Godfroid Classification of Brucella spp. isolated from marine mammals by DNA polymorphism at the omp2 locus. Microb. Infect. 3:

8 4370 MCDONALD ET AL. J. CLIN. MICROBIOL. 13. Dawson, C. E Anti-brucella antibodies in pinnipeds of Australia. Microbiol. Aust. 26: Ewalt, D., J. Payeur, B. Martin, D. Cummins, and W. Miller Characteristics of a Brucella species from a bottlenose dolphin (Tursiops truncatus). J. Vet. Diagn. Investig. 6: Ewalt, D., J. Payeur, J. Rhyan, and P. Geer Brucella suis biovar 1 in naturally infected cattle: a bacteriological, serological, and histological study. J. Vet. Diagn. Investig. 9: Felsenstein, J PHYLIP phylogeny inference package (version 3.2). Cladistics 5: Forbes, L., and S. Tessaro Transmission of brucellosis from reindeer to cattle. J. Am. Vet. Med. Assoc. 203: Forbes, L. B., O. Nielsen, L. Measures, and D. R. Ewalt Brucellosis in ringed seals and harp seals from Canada. J. Wildl. Dis. 36: Foster, G., K. L. Jahans, R. J. Reid, and H. M. Ross Isolation of Brucella species from cetaceans, seals and an otter. Vet. Rec. 138: Foster, G., A. P. MacMillan, J. Godfroid, F. Howie, H. M. Ross, A. Cloeckaert, R. J. Reid, S. Brew, and I. A. P. Patterson A review of Brucella sp. infection of sea mammals with particular emphasis on isolates from Scotland. Vet. Microbiol. 90: Garner, M., D. Lambourn, S. Jeffries, P. Hall, J. Rhyan, D. Ewalt, L. Polzin, and N. Cheville Evidence of brucella infection in Parafilaroides lungworms in a Pacific harbor seal (Phoca vitulina richardsi). J. Vet. Diagn. Investig. 9: Godfroid, J., A. Cloeckaert, J. Liautard, S. Kohler, D. Fretin, K. Walravens, B. Garin-Bastuji, and J. Letesson From the discovery of the Malta fever s agent to the discovery of a marine mammal reservoir, brucellosis has continuously been a re-emerging zoonosis. Vet. Res. 36: Gonzalez, L., I. A. Patterson, R. J. Reid, G. Foster, M. Barberan, J. M. Blasco, S. Kennedy, F. E. Howie, J. Godfroid, A. P. MacMillan, A. Schock, and D. Buxton Chronic meningoencephalitis associated with Brucella sp. infection in live-stranded striped dolphins (Stenella coeruleoalba). J. Comp. Pathol. 126: Hanni, K. D., J. A. K. Mazet, F. M. D. Gulland, J. Estes, M. Staedler, M. J. Murray, M. Miller, and D. A. Jessup Clinical pathology and assessment of pathogen exposure in Southern and Alaskan sea otters. J. Wildl. Dis. 39: Jahans, K. L., G. Foster, and E. S. Broughton The characterisation of Brucella strains isolated from marine mammals. Vet. Microbiol. 57: Jepson, P. D., S. Brew, A. P. MacMillan, J. R. Baker, J. Barnett, J. K. Kirkwood, T. Kuiken, I. R. Robinson, and V. R. Simpson Antibodies to brucella in marine mammals around the coast of England and Wales. Vet. Rec. 141: Kahler, S Brucella melitensis infection discovered in cattle for first time, goats also infected. J. Am. Vet. Med. Assoc. 216: Mackereth, G. F., K. M. Webb, J. S. O Keefe, P. J. Duignan, and R. Kittleberger Serological survey of pre-weaned New Zealand fur seals (Arctocephalus forsteri) for brucellosis and leptospirosis. N. Z. Vet. J. 53: Maratea, J., D. R. Ewalt, S. Frasca, Jr., J. L. Dunn, S. de Guise, L. Szkudlarek, D. J. St. Aubin, R. A. French, and S. de Guise Evidence of Brucella sp. infection in marine mammals stranded along the coast of Southern. N. Engl. J. Zoo Wildl. Med. 34: Mayfield, J. E., J. A. Bantle, D. R. Ewalt, V. P. Meador, and L. B. Tabatabi Detection of brucella cells and cell components, p In K. Nielsen and J. R. Duncan (ed.), Animal brucellosis. CRC Press, Inc., Boston, Mass. 31. Miller, W., L. Adams, T. Ficht, N. Cheville, J. Payeur, D. Harley, C. House, and S. Ridgway Brucella-induced abortions and infection in bottlenose dolphins (Tursiops truncatus). J. Zoo Wildl. Med. 30: Nielsen, O., K. Nielsen, R. Braun, and L. Kelly A comparison of four serologic assays in screening for brucella exposure in Hawaiian monk seals. J. Wildl. Dis. 41: Nielsen, O., R. E. A. Stewart, K. Nielsen, L. Measures, and P. Duignan Serologic survey of Brucella spp. antibodies in some marine mammals of North America. J. Wildl. Dis. 37: Ohishi, K., R. Zenitani, T. Bando, Y. Goto, K. Uchida, T. Maruyama, S. Yamamoto, N. Miyazaki, and Y. Fujise Pathological and serological evidence of brucella-infection in baleen whales (Mysticeti) in the western North Pacific. Comp. Immunol. Microbiol. Infect. Dis. 26: Perrett, L. L., S. D. Brew, J. A. Stack, A. P. MacMillan, and J. B. Bashiruddin Experimental assessment of the pathogenicity of Brucella strains from marine mammals for pregnant sheep. Sm. Rum. Res. 51: Retamal, P., O. Blank, P. Abalos, and D. Torres Detection of antibrucella antibodies in pinnipeds from the Antarctic territory. Vet. Rec. 146: Rhyan, J. C., T. Gidlewski, D. R. Ewalt, S. G. Hennager, D. M. Lambourne, and S. C. Olsen Seroconversion and abortion in cattle experimentally infected with Brucella sp. isolated from a Pacific harbor seal (Phoca vitulina richardsi). J. Vet. Diagn. Investig. 13: Ridler, A., D. West, K. Stafford, P. Wilson, and S. Fenwick Transmission of Brucella ovis from rams to red deer stags. N. Z. Vet. J. 48: Ridler, A. L An overview of Brucella ovis infection in New Zealand. N. Z. Vet. J. 50: Ross, H. M., G. Foster, R. J. Reid, K. L. Jahans, and A. P. MacMillan Brucella species infection in sea mammals. Vet. Rec. 134: Sohn, A., W. Probert, C. Glaser, N. Gupta, A. Bollen, J. Wong, E. Grace, and W. McDonald Human neurobrucellosis with intracerebral granuloma caused by a marine mammal Brucella spp. Emerg. Infect. Dis. 9: Thompson, J. D., D. G. Higgins, and T. J. Gibson CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22: Tryland, M., L. Kleivane, A. Alfredsson, M. Kjeld, A. Arnason, S. Stuen, and J. Godfroid Evidence of brucella infection in marine mammals in the North Atlantic Ocean. Vet. Rec. 144: Tryland, M., K. Sørensen, and J. Godfroid Prevalence of Brucella pinnipediae in healthy hooded seals (Cystophora cristata) from the North Atlantic Ocean and ringed seals (Phoca hispida) from Svalbard. Vet. Microbiol. 105: Van Bressem, M., K. Van Waerebeek, J. Raga, J. Godfroid, S. Brew, and A. MacMillan Serological evidence of Brucella species infection in odontocetes from the south Pacific and the Mediterranean. Vet. Rec. 148: Vizcaíno, N., J. Verger, M. Grayon, M. Zygmunt, and A. Cloeckaert DNA polymorphism at the omp-31 locus of Brucella spp.: evidence for a large deletion in Brucella abortus, and other species-specific markers. Microbiology 143: