Streptococcal collagen-like surface protein 1 promotes adhesion to the respiratory epithelial cell

Background Collagen-like surface proteins Scl1 and Scl2 on Streptococcus pyogenes contain contiguous Gly-X-X triplet amino acid motifs, the characteristic structure of human collagen. Although the potential role of Scl1 in adhesion has been studied, the conclusions may be affected by the use of different S. pyogenes strains and their carriages of various adhesins. To explore the bona fide nature of Scl1 in adherence to human epithelial cells without the potential interference of other streptococcal surface factors, we constructed a scl1 isogenic mutant from the Scl2-defective S. pyogenes strain and a Scl1-expressed Escherichia coli. Results Loss of Scl1 in a Scl2-defective S. pyogenes strain dramatically decreased the adhesion of bacteria to HEp-2 human epithelial cells. Expression of Scl1 on the surface of the heterologous bacteria E. coli significantly increased adhesion to HEp-2. The increase in adhesion was nullified when Scl1-expressed E. coli was pre-incubated with proteases or antibodies against recombinant Scl1 (rScl1) protein. Treatment of HEp-2 cells with rScl protein or pronase drastically reduced the binding capability of Scl1-expressed E. coli. These findings suggest that the adhesion is mediated through Scl1 on bacterial surface and protein receptor(s) on epithelial cells. Further blocking of potential integrins revealed significant contributions of α2 and β1 integrins in Scl1-mediated binding to epithelial cells. Conclusions Together, these results underscore the importance of Scl1 in the virulence of S. pyogenes and implicate Scl1 as an adhesin during pathogenesis of streptococcal infection.


Background
Streptococcus pyogenes causes heterogeneous disease types, including pharyngitis, cellulitis, and bacteremia [1]. The pathogenesis of S. pyogenes infection involves an intriguing host-pathogen interplay in which the biological activity of several bacterial virulence products are modulated by host factors [2]. The details of the molecular interaction between the bacterium and the host, as well as their influences on the prognosis and severity of streptococcal infection, remain poorly understood. S. pyogenes has been reported to produce a number of surface-associated and extracellular products contributing to the pathogenesis. In particular, several cell surface proteins have been documented as being involved in adherence and colonization during infection [3].
Many cell surface proteins of gram-positive bacteria share similar structural characteristics that include a variable amino terminus, a central region with repeated sequences, and a cell-associated region with a LPXTGX cell wall anchored motif [4]. A new S. pyogenes cell surface protein family, streptococcal collagen-like (Scl) protein, has been identified recently [5][6][7][8][9][10]. Scl1 (SclA) and Scl2 (SclB), two Scl protein family members, share a similar structure motif, including the LPXTGX motif and a central region composed of variable numbers of Gly-X-X (GXX) collagen-like motifs. Collagen exhibits a triple-helical, elongated protein structure that is the structural component of the extracellular matrix in multicellular organisms. As eukaryotic cells are known to bind to collagen through receptors expressed on cell surfaces [11], it is reasonable to speculate that the Scl protein family may participate in the colonization/binding of S. pyogenes to receptors on the host cell. Although the potential role of Scl1 in adhesion has been demonstrated by disrupting the scl1 gene in different S. pyogenes strains [5,6], the conclusions may be affected by the use of different S. pyogenes strains and their carriages of various adhesins. In addition, the presence of other Scl family proteins, as well as other streptococcal surface proteins, which may mask the potential role of Scl1 in adhesion, was not taken into consideration in these studies.
Recent studies have demonstrated that collagen receptor, α2β1 and α11β1 integrins [9,12,13], low density lipoprotein [14], thrombin-activatable fibrinolysis inhibitor [15], cellular fibronectin and laminin [16] and human complement regulatory plasma glycoprotein FH [17] may serve as ligands for Scl proteins. While the scl1 gene has been found in all S. pyogenes isolates tested, the scl2 gene sequence was only detected in some strains [7,10,18]. To determine the bona fide nature of Scl1 in colonization and adherence of S. pyogenes to human epithelial cells without the potential interference of other streptococcal surface factors, we generated a scl1 mutant from a Scl2defective S. pyogenes M29588 strain, and expressed Scl1 in the heterologous bacteria Escherichia coli. The adhesion to human epithelial cells was greatly impaired upon the loss of Scl1 in S. pyogenes and was markedly increased upon expression of Scl1 on E. coli.

Results
Identification and analysis of scl1 and scl2 genes in S. pyogenes M29588 strain To identify genes encoding streptococcal collagen-like surface protein 1 and 2 (scl1 and scl2) in S. pyogenes M29588 strain, full lengths of scl1 and scl2 genes were amplified by PCR and sequenced. The scl1 ORF of S. pyogenes M29588 is 1,287 bp, which encodes a protein with 428 amino acid residues ( Figure 1A). The Ala38 was the predicted signal peptidase cleavage site. The length of variable (V) region is 71 amino acids. The collagen-like (CL) region is composed of 46 GXX triplet repeats, followed by a gram-positive bacteria cell wall anchor motif (LPATGE) in the cell wall membrane (WM) region. The CL region and cell wall anchor motif are connected by 6 repeats with a PGEKAPEKS core sequence in the linker (L) region.
It has been shown that the expression of Scl2 is controlled by slipped-strand mispairing at sites containing pentanucleotide coding repeats [7,10,18]. In this study, S. pyogenes M29588 strain contained the scl2 gene, which is predicted to be translated from a putative GTG (Val) start codon ( Figure 1B). We identified 13 AACAA pentanucleotide sequence repeats adjacent to the presumed GTG start codon in S. pyogenes M29588, followed by a premature translation termination at the 89th amino acid residue upon production of Scl2 protein ( Figure 1B). However, the prematurely translated Scl2 protein contains neither CL region nor the anchor motif, suggesting it is not functional and not anchored on the bacteria. These observations show that the S. pyogenes M29588 strain appears to express Scl1 protein consisting of 46 GXX triplet repeats and premature non-functional Scl2 protein.
Loss of adherence to human epithelial cells in S. pyogenes mutant deficient in both Scl1 and Scl2 To determine the role of Scl1 in the adherence of S. pyogenes to human epithelial cells in the absence of Scl2, we generated a scl1 mutant from the Scl2-defective S. pyogenes M29588 strain. A kanamycin-resistant mutant (ST2) was identified after electroporation of S. pyogenes M29588 with the non-replicating plasmid pPJT8, which contains the internal fragment of the scl1 coding region. PCR and Southern blot analysis confirmed the site of mutation, and indicated that the integration occurred through a Campbell-like mechanism (data not shown). No difference in growth rates between the mutant and wild-type strains in TSBY was identified (data not shown), suggesting that the disruption of scl1 did not affect major metabolic pathways under a nutrient-enriched condition, and the integration of pPJT8 did not affect the neighboring genes of scl1. To further clarify if the mutagenesis strategy affected other surface factors, we determined the expression of fibronectin binding proteins, sfb and prtF1, and another known adhesin, oppA, as well as an exotoxin speB as the internal control (Figure 2A). Expression of these four genes was not affected in the scl1 mutant ST2. These results suggest that the mutagenesis strategy did not influence other surface factors, and the scl1 mutant has not compensated for the loss of this adhesin by altering expression profiles for other potential surface binding proteins we tested. In addition, DNA sequence and the number of pentanucleotide repeats of scl2 were not altered in ST2 (data not shown).
To further investigate the role of Scl1 in mediating the adherence of S. pyogenes to human epithelial cells, wildtype and scl1-mutated S. pyogenes ST2, in the exponential phase, were examined for adhesion to human HEp-2 epithelial cells. Adhesion of ST2, was decreased about 70% compared with that of the wild-type (P < 0.01, Figure 2B), suggesting that Scl1 is critical in the adherence of S. pyogenes to human epithelial cells.

Ectopic expression of Scl1 on E. coli
To exclude the interference of other streptococcal surface factors during the adhesion, and to test whether A.

G T T A C A A A A C G C A A A G A A A A C A A C T A A
B. Figure 1 Nucleotide and inferred amino acid sequences of scl1 and scl2 genes in S. pyogenes M29588 strain (M92 type). (A) scl1 coding sequence consists of 1,287 bp which encodes a protein with 428 amino acids. Scl1 protein is composed of signal sequence (SS) followed by a predicted cleavage site (arrowhead), 71 amino acids in V region, 46 GXX triplet motifs (boxed) in CL region, and 6 PGEKAPEKS repeats (underlined) in L region, and the LPATGE cell wall anchor motif (shaded) in WM region. (B) Scl2 protein is translated from the predicted GTG start codon (Val). Thirteen AACAA coding repeats (boxed), located immediately after the GTG start codon, are followed by a premature translation termination at the 89th amino acid residue (asteriated).  Figure  3A). Consistent with this observation, the negative staining of electron microscopy revealed hairy structures in ET3, but these structures were not identified in either E. coli DH5α or ET2 ( Figure 3B). To further demonstrate that Scl1 was ectopically expressed on E. coli, outer membrane fraction of proteins was isolated from ET2 and ET3. Western blot analysis with anti-Scl1 antibodies identified Scl1 in the outer membrane fraction of ET3 but not in that of ET2 (Left panel, Figure 3C). Consistently, a molecular weight shift was revealed by anti-OmpA antibodies in the outer membrane fraction of ET3 (Right panel, Figure 3C). Thus, our data confirmed that Scl1 protein was ectopically expressed on E. coli and can be detected by anti-Scl1 antibodies.

Adherence of Scl1-expressed E. coli to human epithelial cells
Adhesion analysis demonstrated that Scl1-expressed E. coli ET3 dramatically increased its adherence to HEp-2, compared with that of vector-expressed E. coli ET2 and E. coli DH5α ( Figure 4A). Pre-incubation of E. coli ET3 with proteinase K significantly attenuated the Scl1mediated increase in adhesion, suggesting that Scl1 proteins on E. coli are critical for this binding. Thus, the addition of Scl1 on the surface of the heterologous bacteria greatly enhances their adherence to human epithelial cells.
To directly address the role of Scl1 in the binding process, we performed competition studies using anti-Scl1 antibodies and recombinant Scl1 (rScl1) protein.
Polyclonal anti-Scl1 antibodies were generated in 4-week-old BALB/c mice. The full-length rScl1 protein containing sequences shown in Figure 1A was generated and confirmed by SDS-PAGE as a single band of approximately 48 kDa (Lane 1, Figure 4B) and by western blot analysis with anti-Scl1 antibodies (Lane 2, Figure 4B). Both pre-incubation of HEp-2 cells with rScl1 and pre-incubation of ET3 bacteria with anti-Scl1 antibodies significantly blocked the adherence of E. coli ET3 to human epithelial cells ( Figure 4C). The adherence of E. coli ET3 to HEp-2 cells was not affected by pre-incubation of ET3 bacteria with non-specific mouse IgG. These results reveal both the importance and sufficiency of Scl1 in mediating the adherence of bacteria to human epithelial cells.  Adherence through protein receptor(s) on epithelial cells Our previous data showed that the adhesion was affected when Scl1-expressed E. coli was pre-incubated with proteinase K, suggesting that the adhesion is mediated through a protein-like molecule on the bacteria. To further determine the corresponding side of surface molecules on epithelial cells mediating this binding process, HEp-2 cells were treated with pronase and phospholipase A2 to modify the protein and lipid contents on the cell membrane, respectively [19]. Treatment of pronase significantly inhibited the binding of ET3 to epithelial cells in a dose-dependent manner ( Figure 5A). In contrast, treatment of phospholipase A2 did not affect the binding of ET3 to epithelial cells ( Figure 5A). These results suggest that a protein receptor for Scl1 on epithelial cells is likely to mediate this binding event.
It has been proposed that α2β1 and α11β1 integrins might serve as receptors in mediating the Scl1 adherence to epithelial cells [9,12,13]. To determine the role of integrins in the Scl1-mediated binding process, we used monoclonal antibodies against α2, β1, and α2β1 integrins, and performed a competition assay. Pretreatment of monoclonal antibodies against α2, β1, and α2β1 integrins to HEp-2 cells did not affect Scl1-mediated increase in the adhesion of E. coli to human epithelial cells ( Figure  5B). However, we observed a trend, although not significant, toward reduction in the adhesion of E. coli to HEp-2 cells in the presence of monoclonal α2β1 antibodies, suggesting that α2β1 integrin is involved to some extent in the Scl1-mediated binding process. To avoid the lack of interference of the abovementioned monoclonal antibodies in the binding interaction, we employed polyclonal antibodies against α2 and β1 integrins. Polyclonal antibodies against α2 and β1 integrins significantly decreased Scl1-mediated adhesion of E. coli to human epithelial cells ( Figure 5C). These results suggest that protein receptors α2 and β1 integrins underlie the Scl1dependent binding to human epithelial cells.

A.
B. (C) Inhibition of binding by rScl1 protein and anti-Scl1 antibody. Prior to the adhesion assay, HEp-2 cells were pre-treated with rScl1 protein and ET3 were pre-treated with anti-Scl1 antibody and mouse IgG, respectively. **, P < 0.01 and ***, P < 0.001.

C.
To further examine the Scl1-mediated adhesion of E. coli to other eukaryotic cell types known for expression of collagen receptors, we employed two types of cell lines, C2C12 myoblast and human umbilical vein endothelial cell (HUVEC) for the adhesion assay. C2C12 cells are known to express β1 integrins [20], whereas primary HUVECs express α2β1 integrins [21]. Our results show that Scl1-expressed E. coli ET3 exhibited significantly increased adherence to both C2C12 and HUVEC cells, compared to control ET2 ( Figure 5D). Thus multiple eukaryotic cell types may bind and adhere to Scl1-expressed E. coli.

Discussion
The Scl1 protein in the S. pyogenes M29588 strain (M92 type) contains a predicted signal peptidase cleavage site on Ala38, 71 amino acids in V region, 46 GXX repeats in CL region, 6 conserved repeats (PGEKAPEKS) in L region, and followed by a cell wall anchor motif (LPATGE). It has been proposed that the V-region primary sequence in Scl1 is M type associated [7]. Based on the previous study in characterization of the scl1 gene among 21 different M type strains [6], the length of V region in M92 strain is identical to those in M49 and M56 strains. The number of GXX repeats in CL region in M92 strain is equal to those in M2 and M49 strains. The number of PGEKAPEKS repeats in L region in M92 strain is the same with those in M4 and M9 strains. These findings demonstrate significant and extensive genetic variations among clinical isolates of S. pyogenes.
Rasmussen et al. demonstrated that an isogenic Scl1deficient M1 strain (AP1) with 57 GXX repeats did not alter its adhesion ability to Detroit 562 pharyngeal cells

A.
B. [5]. In contrast, Lukomski et al. demonstrated that two independent isogenic Scl1-deficient M1 strains (MGAS 6708 and 5005) with 50 GXX repeats had significantly reduced adherence to human A549 epithelial cells [6]. Although the differences on the surface of various host epithelial cells cannot be excluded, this inconsistency may stem from the carriage of various group A streptococcal adhesins and potential interference of another Scl family member, Scl2. The role of Scl2 in adhesion has been directly addressed in another study by Rasmussen et al. showing that Scl2-deficient isogenic mutants had decreased adherence to human fibroblast cells, but no influence on adherence to pharyngeal cells [18]. Thus, Scl2 appears to be involved in the adhesion process, and the presence of Scl2 could therefore potentially influence and mask the effect of Scl1 in the adhesion. However, Scl2 production in all M1-type strains investigated so far is early terminated at the level of translation [7,18]. In our study, we also demonstrated that the S. pyogenes M29588 strain expresses a pre-terminated Scl2, which contains neither CL region nor anchor motif, according to our sequence analysis. These findings suggest that Scl2 in this particular strain is not functional due to the absence of CL region, and is not anchored on the cell membrane because of the lack of an anchor motif. Our adherence results based on this Scl2defective S. pyogenes M29588 strain provide evidence for the contribution of Scl1 on the binding to host epithelial cells. While Rasmussen et al. used a Scl2-defective AP1 strain to demonstrate that Scl1 mutation does not affect adherence of bacteria to pharyngeal cells [5], their study may have utilized a background where the Scl1 mutation was compensated for by other adhesins, such as protein H [22], C5a peptidase [23]. In our study, we also identified the expression of some surface proteins in this M29588 strain. To exclude the interference of other streptococcal surface factors during Scl1-mediated adhesion, the heterologous expression of Scl1 on E. coli would be an alternative. The outer membrane of Gramnegative bacteria presents an effective barrier that restricts the release of proteins from the bacteria [24]. Many peptides have been inserted within external loops of various outer membrane proteins and have been shown to be exposed on the surface of intact E. coli by immunochemical techniques [24][25][26]. In addition, studies have demonstrated that OmpA chimeric proteins were stably anchored on the external side of the bacteria [24,25,27]. Here we demonstrated that truncated Scl1 fused with OmpA was directed to the outer membrane fraction of E. coli by western blot analysis, and likely exposed on the surface of E. coli by FACS analysis. While ectopic expression of Scl1 on the heterologous bacteria E. coli is an alternative approach to reduce the potential interference of other factors on the surface of S. pyogenes, there are some limitations in our study. For example, it can not be ruled out that Scl1 protein was secreted to the periplasmic space, because Scl1 was constructed after the OmpA signal sequence. To avoid this problem, we performed FACS analysis on whole bacteria using Scl1 antibodies to detect the location of Scl1 in/on E. coli. FACS analysis has been widely used in identification of cell surface molecules in many immunologic and hematologic studies. Furthermore, we isolated proteins from the outer membrane fraction and confirmed the existence of Scl1 by western blot analysis with antibodies against Scl1 and its fusion protein OmpA. However, the proper folding of ectopically expressed Scl1 and the integrity of the outer membrane of E. coli account for other issues influencing our interpretation of Scl1 in adhesion. Nevertheless, our findings concerning the adherence of Scl1-expressed E. coli to human epithelial cells unequivocally show that Scl1 contributes significantly to the adhesion of bacteria to human epithelial cells.

C. D.
Collagen is a triple-helical, elongated protein structure that is the main structural component of the extra-cellular matrix in all multicellular organisms. Collagen-like sequences are found not only in proteins of multicellular organisms but also in proteins of microorganisms, such as a pullulanase in Klebsiella pneuminiae [28] and a platelet aggregation-associated protein in S. sanguis [29,30]. Moreover, collagens interact with several macromolecules in a specific manner, suggesting that the collagen-like repeat sequences not only play a basic structural role, but also have a functional significance. Many eukaryotic cells bind collagen through integrins expressed on their surface [11]. Studies have demonstrated that the recombinant Scl1.41 protein interacted with α2β1 and α11β1 integrins, induced intracellular signaling in host cells, and promoted the internalization of S. pyogenes [9,12,13]. While the hypothesized region mediating the binding to α2β1 and α11β1 integrins in the recombinant Scl1.41 is in a motif called the GLPGER motif [9,12,13], Scl1 protein of S. pyogenes M29588 strain in our study does not contain the GLPGER motif. The novel aspect of this study is the observation that, in this Scl1 sequence type, the GLPGER motif is absent, yet adherence is maintained. Nevertheless, our results indicate that protein receptors, α2 and β1 integrins, contribute to Scl1-dependent binding to the surface of human epithelial cells. Consistently, Scl1-mediated adhesion was also demonstrated in other eukaryotic cell types known for expression of collagen receptors.

Conclusions
In summary, we demonstrated that loss of Scl1 in a Scl2-defective S. pyogenes strain decreased the adhesion of bacteria to human epithelial cells. Ectopic expression of Scl1 in the heterologous Gram-negative bacteria E. coli promoted the adhesion of bacteria to epithelial cells. The increase in adhesion was nullified by proteinase K, rScl1 protein and anti-Scl1 antibody. This binding event appears to be mediated through protein receptors, α2 and β1 integrins, instead of a lipid component, on the surface of epithelial cells. Our results underscore the importance of Scl1 in the adherence of S. pyogenes to human epithelial cells. Understanding the mechanisms by which S. pyogenes adheres to nasal epithelial cells may lead to alternative therapeutic methods of decolonization and decrease the dependence on antibiotics.

Bacterial strains and plasmids
S. pyogenes strain M29588 (emm sequence type 92) was recovered from a patient with necrotizing fasciitis at the Tzu-Chi General Hospital. S. pyogenes cultures were grown in tryptic soy broth supplemented with 0.5% yeast extract (TSBY). E. coli DH5α was grown in Luria broth (LB). Plasmid pSF151 was kindly provided by Dr. Tao of the University of Missouri, Kansas City, USA [31]. Plasmid pST1, which contains the truncated OmpA fusion protein derived from pCR2.1-TOPO (Invitrogen), was kindly provided by Dr. C. Y. Chen of National Taiwan University, Taipei, Taiwan. ET2 and ET3 are E. coli DH5a containing plasmids pST1 and pPJT9, respectively. E. coli was transformed according to the method of Sambrook et al. [32]. S. pyogenes was electroporated according to the method of Schalen et al. [31].

Adhesion assay
Human epidermoid carcinoma epithelial cells (HEp-2; ATCC CCL-23) and C2C12 mouse myoblasts (ATCC CRL-1772) were cultured in DMEM supplemented with 10% FCS. HUVECs were cultured on 0.04% gelatin-coated (Sigma) plates in M199 supplemented with 2 mM L-glutamine (Invitrogen), 10% FBS, and 25% EGM. Adhesion of FITC-conjugated bacteria to cells was measured using a previously described method with slight modifications [34,35]. Bacteria were suspended in cell culture medium to a density of 4 × 10 8 cells/ml. FITC-conjugated bacterial suspension was added to the confluence cells at a M.O.I. of 100 and incubated for 2 hrs at 37°C. The fluorescence of each well was measured by a CytoFluor II flourescence reader (Millipore) with excitation and detection wavelength of 485 nm and 530 nm, respectively. Compared to the results from the conventional plating experiment, the FITC conjugation did not affect the adherence of bacteria.

Blocking assay
For the proteolytic treatment of bacteria, the bacterial suspension (10 8 CFU/ml) was incubated with proteinase K (10 μg/ml) for 1 h at 37°C. The suspension was washed and re-suspended in 1 ml of PBS for the subsequent FITC-conjugation and adhesion assay. In the antibody blocking assay, FITC-conjugated bacteria was incubated with anti-Scl1 antibody (10 μg/ml) for 30 min at room temperature. In the recombinant protein blocking assay, HEp-2 cells were pre-incubated with recombinant Scl1 protein (10 μg/ml), and subsequently incubated with FITC-conjugated bacteria for the adhesion assay. To enzymatically modify cell surface proteins, confluence HEp-2 cells were incubated with pronase (500 and 50 μg/ml; Sigma) or phospholipase A2 (200 μg/ml; Sigma) in FCS-free DMEM for 30 min at 37°C, washed twice with PBS, and then incubated with FITC-conjugated bacteria for the adhesion assay. For the integrin blocking assay, confluence HEp-2 cells were incubated with antibodies (10 μg/ml) against α2 (P1E6, monoclonal, Chemicon International; P17301, polyclonal, Millipore), β1 (P4G1, monoclonal, Chemicon International; P05556, polyclonal, Millipore), α2β1 (BHA2.1, monoclonal, Chemicon International) integrins and mouse IgG (Sigma) for 30 min before the incubation with FITC-conjugated bacteria for the adhesion assay.

Electron microscopy
Drops of bacterial suspension fixed with 2.5% glutaraldehyde were concentrated and placed on formvar-coated copper grids for 1 min. After removal of excess fluid by placing on filter paper, the wet residues were immediately covered with the stain for 30 sec. The grid was airdried before examination for negative staining electron microscopy.

Surface protein isolation
Outer membrane proteins were isolated from bacteria cultures according to a protocol by Fountoulakis and Gasser [36]. Briefly, the overnight E. coli culture was pelleted and the bacteria were resuspended. After shacking and a centrifugation, the new pellet was resuspended and disrupted 3 times by sonication. To remove unbroken cells and debris, sonicated bacteria were centrifuged at 3,000 rpm and subsequently the supernatants were centrifuged at 90,000 rpm. To solubilize the inner membrane protein, the pellet was incubated with 2 ml 2% sodium N-laung sarcosinate and subsequently the supernatants were centrifuged at 90,000 rpm. The pelleted outer membrane proteins were resuspended. OmpA expression pattern performed by western blot using anti-OmpA antibody was represented as an internal control.

Recombinant protein and preparation of antibody
The 1.3-kb full-length sc1l gene was cloned into plasmid pQE30 to construct plasmid pPJ10. The recombinant protein was expressed after isopropyl-β-D-thiogalactopyranoside induction. The expressed protein containing the His 6 tag was separated in a Ni-chelated column (Amersham Biosciences) and eluted by a 0 to 50 mM imidazole gradient. The purified protein was verified by SDS-PAGE and western blot analysis with anti-His monoclonal antibody (Invitrogen). Antibody against purified rScl1 was raised in 4-week-old BALB/c mice. One hundred microgram of rScl1 was applied in the initial immunization of BALB/c mice, with succeeding injections 2 and 4 wks thereafter. Anti-Scl1 mouse IgGs were enriched and purified by affinity chromatography with a column made of Sepharose conjugated to protein A.

Data analysis
Values were reported as mean ± SEM. Statistical analysis was conducted using JMP software (SAS Institute). Student's t-test was used for comparisons between groups and differences were considered to be statistically significant with P value less than 0.05.