A Novel Gene, fudoh, in the SCCmec Region Suppresses the Colony Spreading Ability and Virulence of Staphylococcus aureus

Staphylococcus aureus colonies can spread on soft agar plates. We compared colony spreading of clinically isolated methicillin-sensitive S. aureus (MSSA) and methicillin-resistant S. aureus (MRSA). All MSSA strains showed colony spreading, but most MRSA strains (73%) carrying SCCmec type-II showed little colony spreading. Deletion of the entire SCCmec type-II region from these MRSA strains restored colony spreading. Introduction of a novel gene, fudoh, carried by SCCmec type-II into Newman strain suppressed colony spreading. MRSA strains with high spreading ability (27%) had no fudoh or a point-mutated fudoh that did not suppress colony spreading. The fudoh-transformed Newman strain had decreased exotoxin production and attenuated virulence in mice. Most community-acquired MRSA strains carried SCCmec type-IV, which does not include fudoh, and showed high colony spreading ability. These findings suggest that fudoh in the SCCmec type-II region suppresses colony spreading and exotoxin production, and is involved in S. aureus pathogenesis.


Introduction
Staphylococcus aureus is a pathogenic bacterium that causes various diseases in humans. In the last few decades, the emergence of methicillin-resistant S. aureus (MRSA), vancomycin-intermediate S. aureus, vancomycin-resistant S. aureus, and community acquired methicillin-resistant S. aureus (CA-MRSA) has caused serious clinical problems [1,2]. A better understanding of the virulence mechanism of the MRSA strains is important for establishing effective therapeutic strategies. Whether the virulence of MRSA is different from that of MSSA, however, is controversial [3,4]. The virulence of CA-MRSA is higher than that of MRSA [5][6][7], a difference thought to be related to the production of Panton-Valentine leukocidin [8,9] and phenol-soluble modulin (PSM) [10], but the regulatory mechanisms causing enhanced expressions of these toxins has not been identified.
S. aureus is a non-flagellated gram-positive bacterium that until recently was believed not to have the ability to translocate. We previously reported that S. aureus has the ability to spread on soft agar surfaces, which we termed ''colony spreading'' [11]. Because bacterial translocation, such as flagella-driven swimming, is considered a bacterial virulence mechanism [12][13][14], the spreading ability of S. aureus is suspected to have a role in its virulence. In this study, we examined the spreading ability of clinically isolated MSSA, MRSA, and CA-MRSA strains, and report a novel gene that affects both the colony spreading ability and virulence of S. aureus.

Most MRSA strains have less colony spreading ability than MSSA strains
We examined the colony spreading of 10 MSSA strains and 40 MRSA strains on soft agar plates. All MSSA strains showed colony spreading, and the average diameter of the spread after 10 h incubation was 68 mm. In contrast, the average diameter of the spread of the MRSA strains was 32 mm, which was significantly smaller than that of MSSA. The diameter of the spread of 73% of the MRSA strains was under 35 mm. The remaining MRSA strains showed colony spreading comparable to that of MSSA ( Figure 1A). Therefore, we examined the reason for the decreased colony spreading of most of the MRSA strains, and the high colony spreading of some MRSA strains and all MSSA strains.
Deletion of the SCCmec region restores the colony spreading ability of MRSA strains One difference between MRSA and MSSA is the SCCmec region, which harbors the mecA gene encoding PBP2' (or PBP2a) that confers resistance against methicillin. We hypothesized that the SCCmec region was involved in the decreased colony spreading of most MRSA strains. To evaluate this possibility, we deleted the SCCmec region from the MRSA strains and examined colony spreading. MRSA strains NI-3, NI-4, and NI-5, which showed decreased colony spreading, were transformed with a plasmid carrying ccrAB genes, which encode the recombinases that catalyze precise excision of SCCmec from the chromosome. The deletion of SCCmec in the ccrAB-transformants was verified by a decrease in the methicillin minimum inhibitory concentration (MIC) values to 2 mg/ml, and by the absence of the mecA gene based on the results of polymerase chain reaction (PCR) experiments ( Figure 1B). The SCCmec-deleted mutants showed higher colony spreading than their respective parent strains ( Figure 1C and 1D). Expression of the ccrAB genes itself did not alter colony spreading of the RN4220 strain, a laboratory strain of S. aureus ( Figure S1A and S1D). Therefore, we concluded that the colony spreading of MRSA strains is suppressed by a factor (factors) encoded by SCCmec.
A newly identified gene, fudoh, which locates close to the mecI gene, suppresses the colony spreading of S. aureus To identify the gene in the SCCmec region that is responsible for suppressing the colony spreading of S. aureus, we integrated various flanking regions of the mecA gene of the SCCmec region (Figure 2A) into the chromosome of the Newman strain, a laboratory MSSA strain, using an integration vector by a single homologous recombination to examine the colony spreading of the recombinant strains. The Newman strain integrated with the mecA-R1-I gene complex and its surrounding regions (pInt-mecAR1I-fudoh) showed less colony spreading than the parent strain integrated with an empty vector (pInt) ( Figure 2B and 2C). Thus, a gene (genes) in this region suppresses the colony spreading of S. aureus. To examine which subregion of the pInt-mecAR1I-fudoh was responsible for the suppression, we constructed integration plasmids harboring each subregion of the plasmid. The subregion containing the mecA gene (pInt-mecA) had little effect on colony spreading, but the subregion containing the mecR1I genes (pInt-mecR1I-fudoh) suppressed colony spreading. The vectors (pInt-mecI-fudoh, pInt-mecR1-fudoh, pInt-dmecR1I-fudoh) containing deletions of mecR1, mecI, or both from pInt-mecR1I-fudoh suppressed the colony spreading ability of the Newman strain ( Figure 2B and 2C). Thus, the gene that suppresses colony spreading is neither the mecR1 gene nor the mecI gene. We searched for open reading frames (ORFs) in the residual region and found an ORF encoding 70 amino acids downstream of the mecI gene (position 49366-49578 of the S. aureus N315 genome database [15], Figure 2A). The integration plasmid harboring this ORF (pIntfudoh) suppressed colony spreading of the Newman strain ( Figure 2B, 2C, and Movie S1), indicating that a single copy of the ORF suppresses colony spreading of S. aureus. The ORF has not been registered in the S. aureus N315 genome database, and its function is unknown. We named the gene fudoh (GenBank accession number AB442164), which means 'non-motile' in Japanese. A motif search against the fudoh gene product identified a transmembrane domain in amino acid residues 5 to 24, but no other motifs.
We searched the fudoh gene from SCCmec type-I to type-V sequences deposited on GenBank (type-I, AB033763; type-II, D86934; type-III, AB037671; type-IV, AJ810121; type-V, AB121219), and found that fudoh exists in SCCmec types-II and -III, but not in types-I, -IV, and -V in S. aureus. To verify whether the SCCmec type-III including the fudoh gene suppresses the colony spreading, we examined the colony spreading of the type-III SCCmec-deleted mutants of four clinical isolates 86/560, 86/961, 86/2652, and 85/3566 [16]. Deletion of the entire SCCmec type-III region from two of four strains, 86/961 and 85/3566, increased colony spreading ( Figure S2), suggesting that fudoh in type-III SCCmec also has the function to suppress the colony spreading. Deletion of SCCmec from other two strains did not increase the colony spreading (data not shown), indicating that there are other mechanisms to suppress the colony spreading.
To exclude the possibility that integration of the vector into the Newman chromosome altered the gene expression and contributed to suppress colony spreading, we examined whether fudoh expression from a multicopy plasmid suppressed the colony spreading ability of S. aureus. Plasmid-induced expression of fudoh suppressed colony spreading of the Newman strain ( Figure S1B). Therefore, the suppression of colony spreading by fudoh was not an artifact caused by integration of the vector into the chromosome. The fudoh-induced suppression of colony spreading was also observed in RN4220 strain ( Figure S1C and S1D), suggesting that fudoh-induced suppression of colony spreading ability is a common mechanism in S. aureus. To address whether the differences of the colony spreading ability could be attributed to the differences of growth rates, we examined the growth curves of the fudohtransformed Newman and the SCCmec-deleted mutants of NI-3, NI-4, and NI-5. There were no differences of growth curves between NI-3, NI-4, and the SCCmec-deleted mutants ( Figure S3), indicating that the differences of the colony spreading cannot be explained by their growth rates. We found that the fudohtransformed Newman showed slight longer lag period than the empty vector-transformed Newman, although the doubling time was both 30 min. The doubling time of the SCCmec-deleted mutant of NI-5 was 27 min, which was slight shorter than that of the parent strain, 33 min. These differences of growth could contribute to the differences of colony spreading ability.
fudoh is mutated in highly spreading MRSA isolates Some MRSA strains showed high colony spreading ability that was comparable to that of the MSSA strains ( Figure 1A). We hypothesized that fudoh may be mutated in the high-spreading MRSA strains. To test this notion, we sequenced the fudoh gene in all 40 MRSA strains. Ten MRSA strains that showed high colony spreading had a point mutation in the fudoh gene inducing a K29R amino acid substitution ( Figure 3A). Another high-spreading MRSA strain, NI-15, did not contain the fudoh gene. There was no mutation in the fudoh genes of the remaining 29 MRSA strains (73%) with decreased colony spreading ability (spread diameters ,35 mm/10 h) ( Figure 3A). Multiplex-PCR identified the SCCmec of NI-15 as SCCmec type-IV [17], which lacks the fudoh gene, whereas most MRSA strains (37 strains) contained SCCmec type-II.
To determine whether the point-mutated fudoh gene loses the suppression activity against colony spreading, we examined colony spreading of Newman transformed with fudoh with a K29R point mutation. The K29R-mutated fudoh did not suppress colony spreading of the Newman strain ( Figure 3B, 3C, and S1B). The loss of the suppression activity of the K29R-mutated fudoh was also observed in the RN4220 strain ( Figure S1C and S1D). Thus, the K29R-mutated fudoh loses the ability to suppress colony spreading of S. aureus. The increased colony spreading of 11 MRSA strains can thus be explained by a mutation in, or absence of, the fudoh gene.
The Newman strain transformed with the fudoh gene produces less exotoxin and has attenuated virulence in mice To determine whether the fudoh gene affects the virulence of S. aureus, we examined the exotoxin production of the Newman strain transformed with the fudoh gene. The Newman strain transformed with the fudoh gene produced less hemolysin and nuclease than the strain transformed with an empty vector ( Figure 4A and 4B).
These results indicate that the fudoh gene suppresses exotoxin production by S. aureus.
To examine the effect of fudoh on the virulence of S. aureus in animals, we studied the virulence of the Newman strain transformed with the fudoh gene in a murine model of systemic infection. Mice injected with the fudoh-transformed Newman strain lived longer than mice injected with the empty vector-transformed Newman ( Figure 4C). Therefore, the fudoh gene suppressed S. aureus virulence in a mouse model.

The colony spreading ability of CA-MRSA strains
To gain further insight into the relation between the fudoh gene, colony spreading ability, and virulence, we examined colony spreading ability and the presence of the fudoh gene in CA-MRSA strains, which have high virulence and have been the recent cause of serious clinical problems. MW2 (USA400) was isolated in North Dakota in 1998, USA300 was isolated in San Francisco in 2000, and the others were isolated in Chicago between 1996 and 1999 [6,18,19]. The average diameter of the spread of these CA-MRSAs was 60 mm ( Figure 5A), which was significantly greater than that of MRSAs (P = 0.0001). Amplification of the fudoh gene by PCR revealed that 13 of 14 tested strains (except for CA07) did not have the fudoh gene ( Figure 5B). The result was consistent with reports that these strains (other than CA07) have SCCmec type IV [6,18,19], the type that does not contain the fudoh gene. Thus, most CA-MRSAs have increased colony spreading ability because they lack the fudoh gene. CA07 possessed the fudoh gene, and had decreased colony spreading, consistent with the idea that the fudoh gene suppresses S. aureus colony spreading. Strains 4/16-6N and 5/6-8N showed decreased colony spreading, despite their lack of the fudoh gene. These strains may have an unidentified mechanism that decreases colony spreading ability in the absence of the fudoh gene.

Discussion
The findings of the present study revealed that the colony spreading ability of MSSAs is greater than that of most MRSAs. Furthermore, our data indicate that the fudoh gene in the SCCmec region decreases the colony spreading ability of MRSAs. S. aureus containing the fudoh gene also had decreased exotoxin production and decreased virulence in mice. This is the first observation that SCCmec, which carries the methicillin-resistance determinant, is involved in suppressing the colony spreading and virulence of S. aureus. Most MRSAs isolated in this study had SCCmec type-II, which is typically found in MRSA strains associated with hospitals or other health-care institutions (HA-MRSAs). Thus, it is considered that the MRSAs that acquired SCCmec type-II have reduced colony spreading and virulence, although they have also acquired resistance against all beta-lactam antibiotics. We further demonstrated that colony spreading was restored in MRSAs harboring the mutated fudoh gene in SCCmec type-II. The result indicates that the fudoh gene is not a constant suppressor of colony spreading and virulence of S. aureus, but also can function in restoring these abilities by its mutation.
CA-MRSA strains carrying SCCmec type-IV that includes no fudoh showed greater colony spreading than MRSAs carrying SCCmec type-II. CA-MRSAs cause more severe infections and produce higher amounts of exotoxins than HA-MRSAs [10]. Introduction of fudoh into the Newman strain decreased exotoxin production and colony spreading. These results suggest that the absence of fudoh in SCCmec type-IV is a plausible cause of the increased colony spreading, exotoxin production, and virulence of CA-MRSAs. Examination of SCCmec type-I to type-V sequences obtained from the DNA database suggest that fudoh exists in SCCmec types-II and -III, but not in types-I, -IV, and -V. Investigation of the relation between the existence of fudoh or the fudoh mutation and the clinical observations will contribute to a better understanding of biologic role of fudoh in SCCmec types-II and -III and to the development of more effective detection methods of highly virulent MRSAs and therapies against MRSA infections. SCCmec is not only present in S. aureus, but also in other staphylococcal species. Staphylococcus epidermidis RP62A has fudoh (ORF number, SERP2518) in its SCCmec region. S. haemolyticus, S. hominis, and S. saprophyticus are deduced to carry fudoh from the typing of their SCCmec [20,21]. The fudoh gene has not been found in other chromosomal regions or in other organisms sequenced to date. As SCCmec is a foreign DNA introduced into Staphylococcus species [22], the coexistence of the fudoh and methicillin resistance genes in the same SCCmec might be a survival strategy of this foreign DNA to modulate virulence as well as methicillin resistance to increase the fitness of the host bacteria in the infected animal body. Such positive effect of the fudoh gene onto the fitness of bacteria has not been demonstrated yet in our study. It should be noted that the suppressing effect of fudoh onto virulence might be meaningful itself if the bacteria will not kill host animals by modulating their virulence to coexist with host animals. This possibility should be examined using other infection models. Our preliminary results from a microarray analysis suggested that fudoh affects the expression of many other genes. Elucidating the mechanisms underlying how the fudoh gene product, presumably a membrane-associated 70-aminoacid peptide, decreases S. aureus  colony spreading on soft agar plates, and exotoxin production and virulence in mice will enhance our understanding of the interrelationship between foreign DNA, host bacteria, and an infected animal.

Bacterial strains and growth conditions
The JM109 strain of E. coli was used as a host for pND50, pInt, and their derivatives. E. coli strains transformed with the plasmids were cultured at 37uC in Luria-Bertani broth containing 25 mg/ml chloramphenicol. S. aureus strains were aerobically cultured in tryptic soy broth at 37uC, and 12.5 mg/ml chloramphenicol was added to the medium to maintain the plasmids. Clinical isolates of MSSA and MRSA strains were obtained from Nippon Medical School. These strains were streaked on mannitol sodium chloride plates (Eiken Chemical Inc., Tokyo, Japan) and their utilization of mannitol and high-salt resistance were confirmed. For MRSA strains, MIC values against methicillin were examined and resistance to methicillin was confirmed. Details of the bacterial strains and plasmids used in this study are shown in Table S1.

Colony spreading assay
Tryptic soy broth (Becton, Sparks, MD) supplemented with 0.24% agar (Nacalai Tesque Inc., Kyoto, Japan) was autoclaved. Sterile medium (50 ml) was poured onto a plate (150 mm diameter, FALCON 351058, Becton Dickinson Labware, NJ). Plates were dried in a safety cabinet for 20 min before inoculation with bacteria. Overnight cultures of S. aureus (2 ml) were spotted onto the center of the plates using a P-20 Pipetteman (Gilson S. A. S., Villiers-le-Bel, France). After inoculation, the plates were dried in a safety cabinet for 15 min and incubated at 37uC for 10 h. Photographs were taken using a FinePix S9000 digital camera (Fuji Photo Film Co. Ltd., Tokyo, Japan).

DNA manipulation
Transformation of E. coli, extraction of plasmid DNA from E. coli, PCR, and Southern blot analyses were performed as previously reported [23]. Extraction of genomic DNA from S. aureus was performed using a QIAamp DNA Blood Kit (Qiagen Sciences, Germantown, MD). S. aureus was transformed with plasmid DNA by electroporation [24]. Primers used in this study are listed in Table S2.

Construction of SCCmec deletion mutants
The deletion of the SCCmec region was performed according to the method of Katayama et. al. [25]. The ccrAB genes were amplified by PCR using oligonucleotide primers FccrAB and RccrAB and the N315 genome as the template. The amplified fragments were inserted into pND50, resulting in pccrAB. The RN4220 strain was transformed with pccrAB by electroporation and the plasmids were extracted from colonies resistant to chloramphenicol. NI-3, NI-4, and NI-5 strains were transformed with the plasmid and strains resistant to chloramphenicol were obtained. Chloramphenicol-resistant colonies were inoculated into tryptic soy broth and cultured overnight. A 5-ml aliquot of overnight culture was inoculated into 5 ml of tryptic soy broth and cultured overnight. The inoculation was repeated. The overnight culture was plated onto tryptic soy agar and a single colony was isolated. The methicillin MIC of each strain was examined and methicillin-sensitive strains (2 mg/ml) were obtained. Frequencies of methicillin-sensitive strains from NI-3, NI-4, and NI-5 harboring pccrAB were 100%, 71%, and 100%, respectively.
The mecA gene was amplified by PCR using genomes of each strain and oligonucleotide primers FmecA and RmecA.

Construction of the Newman strain harboring the SCCmec region
The 717-bp of genomic region of RN4220 (41282-41998 in the NCTC8325 genome database) was amplified by PCR using primers Int-F and Int-R and inserted into the XbaI and HindIII sites of pCK20 [26], resulting in pInt, which is able to integrate into the S. aureus chromosome by a single homologous recombination. SCCmec regions were amplified by PCR using N315 genome as the template and inserted into the SmaI site of pInt, resulting in various plasmids as illustrated in Figure 2A. To construct pInt-mecR1-fudoh, pInt-mecI-fudoh, and pInt-dme-cR1I-fudoh, the DNA fragments were amplified by PCR using pInt-mecR1I-fudoh as the template, and self-ligated. RN4220 strains were transformed with plasmids and colonies resistant to chloramphenicol were isolated. The chromosomal region harboring the plasmid was transferred to the Newman strain by phage 80 alpha transduction [27]. Integration of the SCCmec region into the desired chromosomal locus was confirmed by Southern blot analysis using the PCR-amplified region of SCCmec and pInt as probes. To express the fudoh gene from a multicopy plasmid, SCCmec regions were inserted into the SmaI site of pND50.

Typing of the SCCmec region of clinical isolates
Multiplex PCRs were performed to identify the SCCmec types according to the established method [28]. Primer sets M-PCR1 and M-PCR2 were used. NI-7, NI-22, NI-36, and NI-38 were ccrC positive. NI-13 and NI-14 were not typed because DNA fragments were not amplified using M-PCR1 and M-PCR2.

Mouse infection experiment
Bacterial suspensions (100 ml) were injected into the tail vein of 8-wk-old female CD-1 mice (Charles River Laboratories, Kanagawa, Japan) [29]. After the injection, mouse survival was monitored. All mouse protocols were reviewed by the Animal Use Committee at the Graduate School of Pharmaceutical Science at the University of Tokyo.

Supporting Information
Movie S1 Time-lapse imaging of the colony spreading of the fudoh-transformed Newman. The colony spreading ability of the Newman strain transformed with an empty vector (pInt, left panel) or the integration plasmid harboring the intact fudoh gene (pIntfudoh, right panel) was examined using time-lapse imaging. Ten hours observation was shortened to 20 seconds. Found at: doi:10.1371/journal.pone.0003921.s001 (3.31 MB MOV) Figure S1 Plasmid-induced expression of fudoh suppresses colony spreading in Newman and RN4220. (A) Overnight culture of RN4220 harboring pND50 or pccrAB was spotted onto soft agar plates and incubated for 10 h. (B) The Newman strain was transformed with the plasmids described in Table S1, and colony spreading was examined. The means6standard deviations of the halo diameters of at least two independent experiments are presented. Statistical analysis was performed with Student's t test.