Identification and prevalence of an enterotoxin-related gene, se-int, in Staphylococcus intermedius isolates from dogs and pigeons


Keiko Futagawa-Saito, Department of Animal Health 2, School of Veterinary Medicine, Azabu University, 1-17-71 Fuchinobe, Sagamihara, Kanagawa 229-8501, Japan (e-mail:


Aims:  To determine the prevalence of enterotoxin-producing Staphylococcus intermedius in dogs and pigeons.

Methods and Results:  A total of 106 S. intermedius isolates from 44 dogs and 62 pigeons were tested for the production of enterotoxins A, B, C and D by reverse passive latex agglutination (RPLA) and for sec-canine by PCR. Only one isolate from dog was positive for SEC and sec-canine. Screening of sec-canine-negative strains by nested PCR led to the identification of a novel enterotoxin-related gene, se-int. SE-int showed a significant homology (59–61% identity) with SEC and (56·6% identity) SEB. All 44 isolates from dogs and five isolates (8·1%) from pigeons were se-int positive.

Conclusions:  While S. intermedius was isolated more frequently from pigeons than from dogs, se-int was more prevalent among the S. intermedius isolates from dogs, compared with the pigeon isolates.

Significance and Impact of the Study:  Further characterization of the se-int-positive S. intermedius strains should clarify their pathogenic potential including enterotoxigenicity and zoonotic transmissibility to human beings.


Staphylococci are one of the most common bacteria isolated from human beings and animals. Staphylococcus aureus is a primary pathogen of Staphylococcal infections in human beings, whereas S. intermedius is predominantly associated with animals. Similar to S. aureus, S. intermedius is known to carry many virulence factors such as coagulase, haemolysin, leucocidin and enterotoxins.

Staphylococcal enterotoxins (SEs) are well known as pyogenic toxin superantigens (PTSAgs). They are classified as SEA to SEE by their antigenic properties and other SEs (SEG, SEH, SEI, SEJ, SEK, SEL, SEM, SEN, SEO, SEP, SEQ, SER and SEU) are identified on the basis of their homologies with the well-recognized enterotoxins (Betley et al. 1990; Su and Wong 1995; Munson et al. 1998; Zhang et al. 1998; Fitzgerald et al. 2001; Jarraud et al. 2001; Kuroda et al. 2001; Orwin et al. 2001, 2002; Letertre et al. 2003; Omoe et al. 2003). All these staphylococcal enterotoxins were identified in S. aureus, except SEC-canine (SEC subtype) that was identified in S. intermedius from dog (Edwards et al. 1997).

Staphylococcus intermedius is found in a wide range of animal species including free-living birds (Hajek 1976; Biberstein et al. 1984; Hajek et al. 1991; Lilenbaum et al. 1998). It has been isolated from the skin, the hair, the ear and the gingiva of healthy domestic dogs. Staphylococcus intermedius is considered as a primary pathogen of canine skin infections including pyoderma and otitis externa (Hajek 1976; Raus and Love 1983). The occurrence of S. intermedius infections in human beings is very rare (Talan et al. 1989b), but S. intermedius is occasionally recovered from human nasopharyngeal flora after dog bites (Talan et al. 1989a; Lee 1994). In recent reports, S. intermedius strains recovered from dog owners have been shown to be identical to those isolated from their dogs (Tanner et al. 2000; Guardabassi et al. 2004). Moreover, enterotoxin-producing S. intermedius has also been implicated in an outbreak of food poisoning (Khambaty et al. 1994). Enterotoxin-producing S. intermedius is therefore considered as a zoonotic as well as an opportunistic pathogen.

Dogs seem to be an important host of S. intermedius. Depending upon the region where dogs are bred and the method of enterotoxin assay, the isolation frequency of enterotoxin-producing S. intermedius from dogs ranges from 10 to 40% (Burkett and Frank 1998; Becker et al. 2001; Hendricks et al. 2002). Moreover, 12·6% of S. intermedius isolates from dogs carried sec-canine as detected by multiplex PCR DNA-EIA (Becker et al. 2001).

Pigeons also harbour S. intermedius as their nasopharyngeal flora (Hajek 1976; Hajek et al. 1991). These birds live in parks, stations, harbours and marketplaces, and co-inhabit with human beings and therefore serving as an important vehicle for the spread of opportunistic or zoonotic pathogens. The prevalence and characterization of S. intermedius in pigeons have not been reported yet.

In this study, S. intermedius isolates from dogs and pigeons obtained from four prefectures (Chiba, Kanagawa, Saitama and Tokyo) in Japan were tested for their production of enterotoxins A, B, C and D by reverse passive latex agglutination assay (RPLA) and for sec-canine by PCR. Screening of sec-canine-negative strains by nested PCR led to the identification of a novel enterotoxin-related gene from S. intermedius. Here, we present its prevalence, in comparison with the prevalence of sec-canine, in S. intermedius isolates from dogs and pigeons.

Materials and methods

Animals and bacterial strains

Swabs (one per each animal) were taken from healthy skin or infected site of 107 dogs (40 healthy dogs, 27 dogs with pyoderma and 40 dogs with otitis externa) and from nostrils of 83 pigeons (five pigeons from a zoo, 20 domesticated pigeons and 58 wild pigeons). The swabs were inoculated onto mannitol salt agar (Nissui Co., Tokyo, Japan). At least 10 colonies from each culture plate were randomly selected for screening. Staphylococci were identified on the basis of colony characteristics, Gram-stained appearance, catalase activity and lysostaphin susceptibility. All staphylococcal isolates recovered from each animal were further tested for coagulase, β-galactosidase and acetoin activity (Roberson et al. 1992) to identify S. intermedius. The results were confirmed by using the API ID 32 Staph System (bioMérieux, I'Etoile, France). Staphylococcus intermedius BM10902, BM10904, BM10905 and JCM2422T from pigeons, and BM10355 and BM10356 from dogs were used as quality control strains. The BM strains were purchased from the Institute of Pasteur (Paris).

Enterotoxin assay

Staphylococcus intermedius isolates were cultured in brain–heart infusion broth for 18 h at 37°C. The culture supernatants were tested for the activity of enterotoxins A, B, C and D, using a Staphylococcal Enterotoxin Reverse Passive Latex Agglutination (SET-RPLA) kit (Denka Seiken Co., Tokyo, Japan).

PCR and nested PCR

Genomic DNA was prepared from S. intermedius isolates as described by Matsuhashi et al. (1986). Detection of sec-canine by PCR was performed as previously described (Edwards et al. 1997). Nested PCR was performed using PCR products as template and internal (forward and reverse) primers as described previously (Monday and Bohach 1999). DNA amplification was performed for 35 cycles of 94°C for 1 min, 55°C for 1 min and 72°C for 1 min. Nested PCR products were resolved by electrophoresis in 2·0% agarose gels.


The nested PCR products were purified from agarose gels using a Gene Clean Kit (Bio 101, Carlsbad, CA, USA) and were cloned into pCR2·1 TOPO Vector (Invitrogen, Carlsbad, CA, USA). Upstream and downstream regions of the cloned DNA fragment were subsequently cloned from the genomic DNA by a TaKaRa LA PCRTMin vitro cloning kit (TaKaRa, Shiga, Japan) using 5′-agaaaagtatgaacaagaaacga-3′ and 5′-caaatccattagccatattattgt-3′ primers.

DNA sequencing and analysis

DNA sequences on both strands were determined at the Gene Analysis Center TaKaRa (Shiga). Signal peptide prediction analyses were carried out using a program available online ( The BLAST and CLUSTAL W programs were used for sequence similarity search and multisequence alignment. Nucleotide sequence data reported here are available in the DDBJ/EMBL/GenBank databases under the accession number AB116378.

Transcription analysis by RT-PCR

Total RNA was extracted from an overnight-grown broth culture by using an E.Z.N.A. Bacterial RNA kit (Omega Bio-tek, Doraville, GA, USA). Genomic DNA was removed by digestion with DNase I (TaKaRa). RT-PCR was carried out with total RNA (0·05 μg) and THERMO-RT using the display RT-PCR system (Display Systems Biotech, Vista, CA, USA). PCR amplification was carried with gene-specific primers, 5′-gcaag catatcattacatttg-3′ and 5′-acttgatataccctgtttcgt-3′ (0·2 mm each) for 30 cycles of 94°C for 1 min, 55°C for 1 min and 72°C for 1 min. RT-PCR products were resolved by electrophoresis in 2% agarose gels.

Dot blot analysis

Dot blot hybridization was carried out with genomic DNA and PCR-amplified probes (834-bp se-int and 777-bp sec-canine probes) using a DIG DNA labeling and detection kit (Roche, Mannheim, Germany).


Prevalence of SEs in staphylococci and S. intermedius isolates from dogs and pigeons

The isolation rates of staphylococci and S. intermedius were 57% (61/107) and 41·1% (44/107) for dogs and 96·4% (80/83) and 74·7% (62/83) for pigeons, respectively (Table 1). Forty-four S. intermedius isolates from dogs (eight healthy dogs, 23 dogs with pyoderma and 13 dogs with otitis externa) and 62 isolates from pigeons (five pigeons from a zoo, 10 domesticated pigeons and 47 wild pigeons) were screened for sec-canine by PCR and for SEA, SEB, SEC and SED production by SET-RPLA. Only one isolate from a dog with pyoderma was positive for both sec-canine and SEC. The other 105 isolates were negative for sec-canine, SEA, SEB, SEC and SED.

Table 1.  Isolation frequency of staphylococci and Staphylococcus intermedius from dogs and pigeons
SourcenNo. of isolates (%)
StaphylococciS. intermedius
 Healthy4017 (42·5)8 (20·0)
 Pyoderma2723 (85·2)23 (85·2)
 Otitis externa4021 (52·5)13 (32·5)
 Total10761 (57·0)44 (41·1)
 Zoo55 (100)5 (100)
 Domesticated2019 (95·0)10 (50·0)
 Wild5856 (96·6)47 (81·0)
 Total8380 (96·4)62 (74·7)

Identification of a novel enterotoxin-related gene

When the PCR-negative strains were further examined by nested PCR, one strain, AV8004, exhibited a single 280-bp band (data not shown). Because the nucleic acid sequence of the 280-bp fragment showed a truncated ORF and its deduced amino acid sequence had a strong homology with SEC, a genomic fragment (834 bp) was subsequently cloned by PCR. The cloned fragment contained a 792-bp complete ORF encoding 264 deduced amino acid residues with the N-terminal sequence of DHTLDPTPDQ. The mature protein would have 237 amino acid residues with the predicted molecular weight of 27 kDa and pI of 5·82. The deduced mature protein exhibited a significant homology with the known enterotoxins, particularly with SEB (56·6% indentity) and SECs (59·4–60·7% identity) (Table 2; Fig. 1).

Table 2.  Amino acid sequence identity between the mature form of SE-int and other SEs
SE% Identity to SE-int
  1. Enterotoxin sequences were obtained from the following GenBank accession numbers: SEA (M18970), SEB (M11118), SEC1 (X05815), SEC2 (P34071), SEC3 (X51661), SEC-canine (U91526), SED (M28521), SEE (M21319), SEG (AF064773), SEH (AAA19777), SEI (AF064774), SEJ (AF053140), SEK (AF410775), SEL (AF217235), SEM (AF285760), SEN (AF285760), SEO (AF285760), SEP (AP00363), SEQ (AF410775).

  2. *Identified from S. intermedius.

Figure 1.

The alignment of the predicted primary sequences of SE-int with the sequences of SEB and SECs. Identical amino acid residues are shaded and s-s bonds between the two cystine residues are boxed. Asterisks indicate the toxic site represented by seven residues. Amino acid sequences are taken from the GenBank entries whose accession numbers are shown in Table 1

Transcriptional analysis

RT-PCR with se-int-specific primers identified a single 150-bp transcript in AV8004 (Fig. 2), indicating that se-int was expressed at the transcriptional level.

Figure 2.

Expression of se-int in Staphylococcus intermedius AV8004 as detected by RT-PCR. Lane 1, cDNA synthesized from total RNA of AV8004; lane 2, a negative control (minus RT); M, molecular size maker (Φ 174-Hal I digest)

Screening of se-int and sec-canine by dot blot hybridaization

All S. intermedius isolates from dogs (44/44, 100%) and pigeons from the zoo (five of 62, 8%) were se-int positive, but none of the isolates from domesticated and wild pigeons were se-int positive. One S. intermedius isolate was positive for both sec-canine and se-int (Fig. 3). Dot blot results were confirmed by PCR (data not shown).

Figure 3.

Dot blot analysis of genomic DNA from Staphylococcus intermedius strains using se-int probe (a) and sec-canine probe (b). Spots a1–a4, dogs with pyoderma; a5–b2, dogs with otitis exterma; b3–b6, healthy dogs; c1–c3, pigeons from zoo; c4–c6 domesticated pigeons; d1–d6, wild pigeons


The enterotoxins are simple single-chained proteins with the molecular mass of 24–29 kDa (Betley et al. 1990; Su and Wong 1995; Munson et al. 1998; Zhang et al. 1998; Fitzgerald et al. 2001; Jarraud et al. 2001; Kuroda et al. 2001; Orwin et al. 2001, 2002; Letertre et al. 2003; Omoe et al. 2003). In this study, we identified a gene encoding for a novel enterotoxin-like protein in an S. intermedius isolate from dog. As a BLAST search demonstrated that the encoded protein shared a significant homology with SECs (59–61% identity) and SEB (56·6% identity), the ORF and its encoded protein were designated se-int and SE-int, respectively (S. intermedius enterotoxin). Analysis of SE-int in comparison with the well-characterized SEB revealed a striking resemblance between the two proteins. In SEB, an S–S bond (Betley et al. 1990) joins two cysteines. Two half-cystine residues create a cystine loop, and the sequence motif (Cys-Met-Tyr-Gly-Gly-Val-Thr) adjacent to the cystine loop is within the toxic site of the molecule (Bergdoll et al. 1974). The identical motif was found in SE-int. Residues Asn23, Tyr175 and Asn179 surrounding the T-cell receptor-binding cavity on the surface of SEB are important for TCR binding (Briggs et al. 1997). With the exception of Asn179, these residues were conserved in SE-int. In addition, SE-int is proposed to possess superantigenicity because Thr112 residue, which is located in the TCR binding cavity and thus required for T-cell specificity (Baker et al. 2002) was also conserved in SE-int. Taken together, all these observations suggest, but not prove, that SE-int is an enterotoxin-related putative toxin. With the previously identified SEC-canine, SE-int would be the second known enterotoxin in S. intermedius. However, additional testing for emetic activity and superantigenicity is required to prove the enterotoxigenicity in S. intermedius.

A considerably high isolation rate (41·1%) of S. intermedius from dogs in this study is in agreement with previous reports in which S. intermedius was the most frequently isolated staphylococci from dogs (Cox et al. 1988; Becker et al. 2001). Our results demonstrated for the first time that S. intermedius was the predominant staphylococci isolatable from pigeons as well and that its carriage in pigeons (74·7%) was remarkably higher than that in dogs (41·1%) (P < 0·005).

Additionally, using the same experimental methods and animals, the isolation rate (one of 44, 2·3%) of enterotoxin-producing S. intermedius from dogs in this study was obviously lower compared with other studies from the United Kingdom (26%) (Hendricks et al. 2002) and Germany (11·7% by immunoassay and 12·6% by multiplex PCR DNA-EIA) (Becker et al. 2001). This observed wide range of isolation rate is considered to be dependent on the geographical confines. With the exception of the isolates from pigeons in the zoo, which were se-int positive, all S. intermedius isolates from pigeons did not produce any enterotoxins tested. However, all S. intermedius isolates from dogs in this study harboured se-int, suggesting that the possession of se-int is advantageous to S. intermedius in adaptation to its host, particularly dogs.

In line with the reports that S. intermedius isolates from different host species such as pigeons, dogs, minks and horses are distinct as determined by rRNA gene restriction site polymorphism analysis (Chesneau et al. 2000), pulsed-field gel electrophoresis (PFGE) typing (Wakita et al. 2002) and 16S–23S intergenic ribosomal DNA spacer polymorphism analysis (Bes et al. 2002), we also observed the apparent difference between S. intermedius strains from pigeons and dogs with regard to their occurrence and carriage of se-int.

Transmission of S. intermedius between dogs and their owners has been documented recently (Guardabassi et al. 2004). In addition, pigeons are known to transmit various diseases such as Newcastle disease, toxoplasmosis and salmonella food poisoning to human beings. Therefore, it is of interest to further examine these se-int-positive S. intermedius strains for the production of other enterotoxins not tested herein and for their pathogenic potential including zoonotic transmissibility to human beings.