JB Email Content Delivery
Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Other Versions of this Article:
JB.00748-07v1
189/21/7610    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Chen, Q.
Right arrow Articles by Fives-Taylor, P. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chen, Q.
Right arrow Articles by Fives-Taylor, P. M.
Journal of Bacteriology, November 2007, p. 7610-7617, Vol. 189, No. 21
0021-9193/07/$08.00+0     doi:10.1128/JB.00748-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Differential Roles of Individual Domains in Selection of Secretion Route of a Streptococcus parasanguinis Serine-Rich Adhesin, Fap1{triangledown}

Qiang Chen,1,{dagger} Baiming Sun,1 Hui Wu,2 Zhixiang Peng,1,3 and Paula M. Fives-Taylor1*

Department of Microbiology and Molecular Genetics, University of Vermont, Burlington, Vermont 05405,1 Departments of Pediatric Dentistry and Microbiology, University of Alabama at Birmingham, Birmingham, Alabama 35294,2 Center of Stomatology, Tongji Hospital, Tongji Medical Collage, Huazhong University of Science and Technology, Wuhan 430030, China3

Received 11 May 2007/ Accepted 21 August 2007


   ABSTRACT
Top
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Fimbria-associated protein 1 (Fap1) is a high-molecular-mass glycosylated surface adhesin required for fimbria biogenesis and biofilm formation in Streptococcus parasanguinis. The secretion of mature Fap1 is dependent on the presence of SecA2, a protein with some homology to, but with a different role from, SecA. The signals that direct the secretion of Fap1 to the SecA2-dependent secretion pathway rather than the SecA-dependent secretion pathway have not yet been identified. In this study, Fap1 variants containing different domains were expressed in both secA2 wild-type and mutant backgrounds and were tested for their ability to be secreted by the SecA- or SecA2-dependent pathway. The presence or absence of the cell wall anchor domain (residues 2531 to 2570) at the C terminus did not alter the selection of the Fap1 secretion route. The Fap1 signal peptide (residues 1 to 68) was sufficient to support the secretion of a heterologous protein via the SecA-dependent pathway, suggesting that the signal peptide was sufficient for recognition by the SecA-dependent pathway. The minimal sequences of Fap1 required for the SecA2-dependent pathway included the N-terminal signal peptide, nonrepetitive region I (residues 69 to 102), and part of nonrepetitive region II (residues 169 to 342). The two serine-rich repeat regions (residues 103 to 168 and 505 to 2530) were not required for Fap1 secretion. However, they were both involved in the specific inhibition of Fap1 secretion via the SecA-dependent pathway.


   INTRODUCTION
Top
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Oral biofilm formation is initiated by the adhesion of primary colonizers such as Streptococcus and Actinomyces species to the tooth surface (14). One such primary colonizer, Streptococcus parasanguinis FW213, uses long fimbriae as its major adhesin for the tooth surface (8, 10). The structural subunit of the fimbriae, fimbria-associated protein 1 (Fap1), is involved in fimbria biogenesis, adhesion, and biofilm formation (11, 34, 35). Fap1 is a high-molecular-mass surface glycoprotein, and its glycosylation mechanism has not been totally elucidated (28, 33, 35).

The polypeptide of Fap1 is composed of an unusually long signal peptide (SP), two nonrepetitive regions (NRI and NRII), two serine-rich repetitive regions (RI and RII), and a cell wall anchor domain (CWA) (Fig. 1). The SP comprises 68 residues and is absent in Fap1 secreted into the culture media (CM) (34, 35). It is longer than a canonical signal peptide, which usually has 18 to 30 residues (32). The serine-rich repetitive regions contain putative glycosylation sites, as these regions have amino acid compositions similar to those of the glycopeptides purified from pronase-digested Fap1 (28). A cell wall anchor domain, a hallmark of gram-positive bacterial surface adhesins (18), is present in the C terminus of Fap1.


Figure 1
View larger version (21K):
[in this window]
[in a new window]

  		FIG. 1. Schematic diagram of Fap1 variants expressed by S. parasanguinis. Numbers of the beginning residues of each domain of full-length Fap1 are shown, as are the predicted sizes of the proteins.

 
Other high-molecular-mass serine-rich proteins in gram-positive bacteria have also been identified, including Hsa of Streptococcus gordonii DL1 (29), GspB of S. gordonii M99 (2), SraP of Staphylococcus aureus (26), SrpA of Streptococcus sanguinis (20), SrpA of Streptococcus cristatus (12), and Srr-2 of Streptococcus agalactiae (25). They share the following common characteristics with Fap1: (i) all are involved in adhesion or virulence; (ii) all are composed of an unusually long signal peptide, two serine-rich repeats, and a cell wall anchor domain; (iii) their genes are all linked to a locus containing secA2 and other genes that collectively constitute an accessory secretion pathway (30, 33).

S. parasanguinis SecA2 has a protein sequence similar to that of SecA, but the two proteins differs significantly in their substrate specificity, subcellular distribution, and other physical-biochemical characteristics (5, 6). The secretion of Fap1 in S. parasanguinis (5) and GspB in S. gordonii (2) is dependent solely on the species-specific SecA2 proteins. It is not clear what signal directs the secretion of Fap1 and other Fap1-like proteins to the SecA2-dependent accessory secretion pathway, in lieu of the SecA-dependent canonical secretion pathway. Gram-positive bacteria route most of their secretome in an unfolded conformation to the SecA-dependent pathway via the recognition of the canonical signal peptide (31). However, the canonical signal peptide is absent in a large number of proteins that are secreted by the SecA2 pathway in Mycobacterium tuberculosis and Listeria monocytogenes (4, 15). The signal peptide of S. gordonii GspB is not sufficient for the secretion of heterologous protein by any secretion pathway (3). The twin-arginine translocation pathway has been identified in an increasing number of gram-positive bacteria in the past few years (21, 22, 24). However, its recognition motif, R/K-R-X-{varphi}-{varphi} (where {varphi} is a hydrophobic residue) (7), is not present in the Fap1 sequence.

It is possible that some common structural characteristics of these serine-rich proteins can be recognized by the SecA2 pathway. A truncated nonglycosylated GspB variant can be secreted by the SecA pathway (3), indicating that the SecA pathway and the SecA2 pathway can be used alternately depending on the presence or absence of their corresponding signals. The mutagenesis of the GspB N-terminal region results in decreased secretion of GspB variants, suggesting that this region is important for SecA2-dependent secretion (3). However, the domains within this region were not deleted individually and the mutagenesis has not been performed with a secA2 mutant to assess the impact of the mutagenesis on both the SecA2 and SecA pathways. Thus, the roles of individual domains in promoting or inhibiting one secretion pathway versus the other remain to be unequivocally determined.

The carbohydrate moiety is not likely to be a major requirement for SecA2-mediated secretion, as Fap1 species that are not fully glycosylated have been detected on the cell surfaces of some mutants (5). In S. gordonii M99, nonglycosylated GspB is still secreted by the SecA2 pathway and by the SecA pathway as well. However, the carbohydrates of GspB interfere with secretion by the SecA pathway, thus indirectly promoting secretion by the SecA2 pathway (3). The main purpose of this study is to evaluate the contribution of various polypeptide regions of Fap1 to the selection of the SecA2 pathway versus the SecA pathway.


   MATERIALS AND METHODS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Growth conditions. Escherichia coli strains were grown with aeration at 37°C in Luria-Bertani broth or agar supplemented with erythromycin (500 µg/ml) or kanamycin (50 µg/ml) when necessary. S. parasanguinis strains were grown statically in the presence of 5% CO2 at 37°C in Todd-Hewitt broth or agar supplemented with erythromycin (10 µg/ml) or kanamycin (125 µg/ml) when necessary.

DNA methods. Routine DNA manipulations were performed as described by Sambrook et al. (23). Genomic DNA of S. parasanguinis was prepared by using the Puregene DNA isolation kit (Gentra System). The PCR primers used are listed in Table 1. The transformation of S. parasanguinis was performed as described previously (9). All constructs generated in this study were confirmed by partial sequencing to ensure that no mutation was introduced into the fap1 coding region by PCR.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Primers used in this study

 
Construction of a fap1 and secA2 double mutant. A suicidal plasmid containing an aphA3 cassette flanked by fap1 sequences corresponding to the N terminus of Fap1, pVT1392 (35), was electroporated into a secA2 mutant, VT1574 (5). The transformant, VT1703, was selected by testing for kanamycin resistance and used as a fap1 secA2 double mutant in the following experiments.

Construction of S. parasanguinis strains expressing different Fap1 proteins. (i) FW213 and VT1574. Full-length Fap1 was expressed from the chromosomes of FW213 (wild type) and the secA2 mutant VT1574 (5).

(ii) VT1428 and VT1716. A previously constructed pVT1427 plasmid contained a fap1 fragment with an erythromycin resistance cassette inserted at the junction of the regions coding for RII and the CWA. The plasmid was electroporated into FW213, and transformants were selected by using erythromycin. In the resulting strain, VT1428, a truncated form of Fap1 without the CWA was expressed from the chromosome (34). The same strategy was used in this study to inactivate the CWA in the secA2 mutant VT1574, resulting in strain VT1716.

(iii) VT1708 and VT1709. An E. coli plasmid carrying fap1, pVT1175 (35), was digested with HindIII and ligated with a HindIII fragment of an E. coli-streptococcal shuttle plasmid, pVA838 (16). The resultant fap1-containing plasmid, pVT1704, was used as a template in an inverse PCR using KpnI-linked primers fap1F1 and fap1R1 going outward from the fap1 region coding for RII. The PCR product with the deletion of RII was then digested with KpnI and self-ligated. The resultant plasmid, pVT1707, was electroporated into a fap1 mutant (VT1393) (35) and a fap1 secA2 double mutant (VT1703), resulting in strains VT1708 and VT1709, respectively. The truncated Fap1 was expressed from the plasmid.

(iv) VT2005 and VT2006. DNA coding for Fap1 residues 1 to 342 (SP-NRI-RI-NRIIa, where NRIIa is the part of NRII from residues 169 to 342) was PCR amplified from FW213 by using SalI-linked forward primer fap1F2 and BamHI-linked reverse primer fap1R2. The PCR product was cloned into a shuttle plasmid, pVT1664 (6), downstream of a maltose promoter at SalI and BamHI sites. The resultant plasmid, pVT2000, was electroporated into a fap1 mutant (VT1393) (35) and a fap1 secA2 double mutant (VT1703), resulting in strains VT2005 and VT2006, respectively. The truncated Fap1 was expressed from the plasmid.

(v) VT2101 and VT2102. pVT2000 was used as a template in an inverse PCR with KpnI-linked primers fap1F3 and pVT1664R1 going outward from the fap1 region coding for the SP. The PCR product with the deletion of the SP was then digested with KpnI and self-ligated. The resultant plasmid was electroporated into a fap1 mutant (VT1393) (35) and a fap1 secA2 double mutant (VT1703), resulting in strains VT2101 and VT2102, respectively. The truncated Fap1 was expressed from the plasmid.

(vi) VT2103 and VT2104. pVT2000 was used as a template in an inverse PCR using KpnI-linked primers fap1F4 and fap1R3 going outward from the fap1 region coding for NRI. The PCR product with the deletion of NRI was then digested with KpnI and self-ligated. The resultant plasmid was electroporated into a fap1 mutant (VT1393) (35) and a fap1 secA2 double mutant (VT1703), resulting in strains VT2103 and VT2104, respectively. The truncated Fap1 was expressed from the plasmid.

(vii) VT2105 and VT2106. pVT2000 was used as a template in an inverse PCR using KpnI-linked primers fap1F5 and fap1R4 going outward from the fap1 region coding for RI. The PCR product with the deletion of RI was then digested with KpnI and self-ligated. The resultant plasmid was electroporated into a fap1 mutant (VT1393) (35) and a fap1 secA2 double mutant (VT1703), resulting in strains VT2105 and VT2106, respectively. The truncated Fap1 was expressed from the plasmid.

(viii) VT2107 and VT2108. pVT2000 was used as a template in an inverse PCR using BglII-linked primers pVT1664F1 and fap1R5 going outward from the fap1 region coding for NRIIa. The PCR product with the deletion of NRIIa was then digested with BglII and self-ligated. The resultant plasmid was electroporated into a fap1 mutant (VT1393) (35) and a fap1 secA2 double mutant (VT1703), resulting in strains VT2107 and VT2108, respectively. The truncated Fap1 was expressed from the plasmid.

(iv) VT2003 and VT2004. The sequence of fap1 corresponding to the SP was PCR amplified from FW213 by using SalI-linked forward primer fap1F2 and KpnI-linked reverse primer fap1R6. The PCR product was cloned into a shuttle plasmid, pVT1666 (6), between a maltose promoter and gfp, coding for green fluorescent protein (GFP), at SalI and KpnI sites. The resultant plasmid was electroporated into a fap1 mutant (VT1393) (35) and a fap1 secA2 double mutant (VT1703), resulting in strains VT2003 and VT2004, respectively. The Fap1 SP-GFP fusion protein was expressed from the plasmid.

(v) VT2001 and VT2002. The shuttle plasmid expressing GFP under the control of a maltose promoter, pVT1666 (6), was electroporated into a fap1 mutant (VT1393) (35) and a fap1 secA2 double mutant (VT1703), resulting in strains VT2001 and VT2002, respectively.

Secretion analysis. S. parasanguinis strains grown to mid-exponential phase (an optical density at 470 nm of 0.6) in 100 ml of Todd-Hewitt broth were centrifuged at 10,000 x g for 10 min. All the spent CM were separated from cells by filtration through a 0.2-µm-pore-size membrane. Proteins from the CM were precipitated with ethanol as described previously (13). Lysate was prepared by passing the bacterial cell suspension through a French pressure cell as described previously (5). Cell wall-associated proteins (CW) were extracted using a cell wall-hydrolyzing enzyme, mutanolysin (Sigma-Aldrich), as described previously (5). Proteins from 108 CFU from the original culture were separated by 4 to 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (with gel from Cambrex) and subjected to Western blotting using SuperSignal West substrates (Pierce) for signal detection. Goat anti-mouse and goat anti-rabbit immunoglobulin G antibodies conjugated with horseradish peroxidase (Jackson) were used as secondary antibodies.

Azide inhibition assay. S. parasanguinis strains grown to early exponential phase (an optical density at 470 nm of 0.4) were centrifuged at 10,000 x g for 10 min. Spent CM were decanted. Bacterial pellets were washed twice with 40 ml of Tris-buffered saline (pH 7.4), resuspended in warm Todd-Hewitt broth supplemented with 30 mM sodium azide, and incubated for 30 min at 37°C. Proteins from the CM and lysates were prepared as described above.


   RESULTS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Fap1 requires SecA2 but not SecA for secretion. Fap1 secreted by the wild-type strain FW213 was found in the CM and among the CW (Fig. 2A, top panel, lanes 1 to 3), as detected by a Fap1 peptide-specific monoclonal antibody, E42 (28). A cytoplasmic protein, Tpx (27), was present only in the lysate and was absent from the CM and CW fractions (Fig. 2A, bottom), suggesting that the CM and CW fractions were not contaminated from the lysate. The secretion was not mediated by the canonical SecA-dependent secretion pathway, as mature Fap1 was not secreted from the secA2 mutant VT1574 (Fig. 2A, top panel, lanes 7 to 9), even though SecA was intact. Fap1 secretion after transient sodium azide treatment, an established method for the specific inhibition of SecA in both gram-negative and gram-positive bacteria (17, 19), including S. gordonii (3), was also analyzed. Mature Fap1, represented by a band beyond 220 kDa in the CM and CW fractions, was still secreted after azide treatment (Fig. 2B, lanes 4 and 5). Besides the high-molecular-mass band, other low-molecular-mass bands and smears were occasionally detected by the peptide-specific antibody E42, a Fap1 monoclonal antibody with the epitope previously mapped to the Fap1 peptide region from residues 69 to 102 or 169 to 342 (28). These unstable Fap1 species were either intermediate or degradation products (5, 35). Some of them were highly sensitive to azide treatment, suggesting that they were secreted by the canonical pathway. However, the nature of these species was unknown.


Figure 2
View larger version (56K):
[in this window]
[in a new window]

  		FIG. 2. Secretion of full-length Fap1 and Fap1 without CWA. (A) Proteins prepared from spent CM, CW, and lysates (L) from FW213 (wild type), VT1428 (CWA mutant), VT1574 (secA2 mutant), and VT1716 (secA2-CWA double mutant) were analyzed by Western blotting and detected by monoclonal antibody E42 (top) or anti-Tpx antibody ({alpha}-Tpx) (bottom). Each CM and CW fraction contained proteins from 108 CFU. Each lysate fraction contained proteins from 2.5 x 107 CFU. +, present; –, absent. (B) FW213 (wild type) and VT1428 (CWA mutant) were mocked treated (–) or treated with sodium azide (+) for 30 min prior to protein extraction from spent CM, CW, and lysates. Proteins were analyzed by Western blotting and detected by monoclonal antibody E42. VT1393, a fap1 null mutant, was used as a negative control. Each CM and CW fraction contained proteins from 108 CFU. Each lysate fraction contained proteins from 2.5 x 107 CFU.

 
Fap1 SP alone can be recognized by the SecA-dependent canonical pathway. The SP of Fap1 (68 residues) is longer than the usual signal peptides in gram-positive bacteria (18 to 30 residues). To test if the Fap1 SP can function as a regular signal peptide recognized by the canonical pathway, a Fap1 SP and GFP chimeric protein was constructed and tested for its secretion. This chimeric protein was secreted both in the presence (VT2003) and in the absence (VT2004) of SecA2 (Fig. 3A), suggesting that the Fap1 SP was sufficient for the SecA2-independent secretion of a heterologous protein. As a control, GFP alone was not secreted either in the presence (VT2001) or in the absence (VT2002) of SecA2. The secretion of SP-GFP was inhibited by azide both in the presence (VT2003) and in the absence (VT2004) of SecA2 (Fig. 3B), suggesting that the secretion was mediated by the SecA pathway but not the SecA2 pathway. Therefore, the Fap1 SP alone was recognized by the SecA pathway.


Figure 3
View larger version (58K):
[in this window]
[in a new window]

  		FIG. 3. Secretion of SP-GFP fusion protein. (A) Proteins from two strains expressing GFP in a secA2 wild-type (VT2001) or mutant (VT2002) background and two strains expressing SP-GFP in a secA2 wild-type (VT2003) or mutant (VT2004) background were prepared from spent CM and lysates (L), analyzed by Western blotting, and detected by monoclonal antibodies against GFP. Each fraction contained proteins from 108 CFU. +, present; –, absent. (B) VT2003 and VT2004 were mocked treated (–) or treated with sodium azide (+) for 30 min prior to the protein extraction from spent CM and lysates. Proteins from 108 CFU were analyzed by Western blotting and detected by monoclonal antibodies against GFP.

 
It was not determined from the above-reported data if the Fap1 SP is necessary for the SecA2 pathway or not. Nevertheless, Fap1 SP alone was not sufficient to support the secretion of GFP via the SecA2 pathway, as shown by the inhibition of secretion by azide even in the presence of SecA2 (VT2003). The structural elements that inhibit the SecA pathway and promote the SecA2 pathway are in regions other than Fap1 SP. To test this new hypothesis, various regions of Fap1 were deleted to determine if some deletions could eliminate the dependence on SecA2 and reestablish the secretion by the SecA pathway. E42 reacts with a nonglycosylated recombinant protein, drFap1, which contains two Fap1 regions, residues 69 to 102 and 169 to 342 (28). It was not determined if the E42 epitope was in Fap1 residues 69 to 102 or 169 to 342. The E42 peptide-specific antibody was used to detect all the truncated proteins bearing the same peptide epitope, regardless of their complex glycosylation statuses associated with the presence or absence of some regions.

Fap1 CWA does not affect the secretion route of Fap1. To test if the Fap1 CWA has any role in determining the secretion route, the CWA was mutagenized in a secA2 mutant background and its secretion was compared to that from the wild type and a secA2 mutant (Fig. 2A). Full-length Fap1 was secreted by the wild type FW213 but not by the secA2 mutant VT1574. Anchorless Fap1 was secreted completely into the CM by VT1428, which had an intact form of SecA2, but not by VT1716, which lacked SecA2. These data suggested that the presence or absence of the CWA does not affect the dependence of Fap1 secretion on SecA2.

To confirm that Fap1 secretion by the CWA mutant strain VT1428 was also mediated by the accessory secretion pathway but not by the canonical secretion pathway, secretion in the presence of sodium azide was analyzed as described above for secretion by FW213 (Fig. 2B, lane 7 to 12). Azide treatment did not abolish the secretion of the high-molecular-mass anchorless Fap1, represented by the smear beyond 220 kDa, by the CWA mutant. Although azide treatment diminished the amounts of Fap1 in all fractions to some extent, the secretion patterns of FW213 and the CWA mutant VT1428 were not dramatically altered (Fig. 2B, lanes 1 to 12). The negative control, VT1393, a fap1 null mutant (35), did not produce any detectable Fap1 (Fig. 2B, lanes 13 to 15).

Fap1 RII partially inhibits secretion by the SecA-dependent canonical pathway. RII was deleted in both secA2 wild-type and mutant backgrounds, and the secretion of the truncated Fap1 was tested. A large amount of truncated Fap1 was present in the CM from the secA2 wild-type strain VT1708 (Fig. 4A, top panel, lane 1). A small fraction of the truncated protein was anchored to CW (Fig. 4A, top panel, lane 2). The CM and CW fractions were not contaminated from the lysate, as the cytoplasmic protein, Tpx, was detected only in the lysate (Fig. 4A, bottom panel). The apparent molecular masses of the intracellular Fap1 species (Fig. 4A, top panel, lane 3) and the CW species were notably different from that of the species found in the CM, which we have noted with some other Fap1 constructs. The most likely explanation is that the species released into the CM was more mature in terms of glycosylation and/or processing after it was dissociated from CW by an unknown mechanism. The same truncated Fap1 construct was secreted by the isogenic secA2 mutant strain VT1709, as shown by the presence of a low-molecular-mass band in the CM and a high-molecular-mass band in CW (Fig. 4A, top panel, lanes 7 to 8). The efficiency was low, as the majority of the truncated Fap1 was retained in the lysate (Fig. 4A, top panel, lane 9).


Figure 4
View larger version (43K):
[in this window]
[in a new window]

  		FIG. 4. Secretion of Fap1 without RII. S. parasanguinis strains were mocked treated (–) or treated with sodium azide (+) for 30 min prior to the protein extraction. Proteins from 108 CFU were analyzed by Western blotting and detected by monoclonal antibody E42 or anti-Tpx antibody ({alpha}-Tpx). (A) Results for strains expressing Fap1{Delta}RII in a secA2 wild-type (VT1708) or mutant (VT1709) background. CM, CW, and lysates (L) were analyzed. (B) Results for strains expressing Fap1 with the deletion of NRII residues 342 to 504, RII, and the CWA in a secA2 wild-type (VT2005) or mutant (VT2006) background. CM and lysates were analyzed.

 
The above-reported data suggested that in the presence of RII, such as in the wild type and the CWA mutant, the secretion of Fap1 via the canonical secretion pathway was completely inhibited (Fig. 2A). In the absence of RII, the Fap1 truncated species was also secreted, albeit inefficiently, by the canonical secretion pathway, as in the absence of SecA2 (Fig. 4A). In short, the presence of RII was necessary for the complete inhibition of the canonical pathway but unnecessary for the SecA2-dependent accessory pathway.

To confirm that the minimal secretion by the RII-secA2 double mutant VT1709 was indeed mediated by the canonical pathway, the secretion of the truncated Fap1 in the presence of azide was analyzed. As predicted, SecA-dependent secretion was completely obliterated by azide treatment (Fig. 4A, top panel, lanes 10 and 11). The strain with RII deleted in a secA2 wild-type background, VT1708, was still able to secrete the truncated Fap1 proteins after azide treatment. However, secretion was affected by azide to some extent, as the CW species and some of the CM species were sensitive to azide treatment (Fig. 4A, top panel, lanes 4 and 5). Although Fap1 secretion by the canonical pathway was derepressed to some extent by the deletion of RII, the secretion was inefficient compared to the efficient secretion that occurs via the accessory pathway (Fig. 4A, top panel, lanes 1 to 3 and 7 to 9).

A truncated form of Fap1 (designated SP-NRI-RI-NRIIa) with a deletion of part of NRII (residues 342 to 504), RII, and the CWA was expressed in both secA2 wild-type and mutant backgrounds and tested for secretion (Fig. 4B). The apparent molecular masses of the intracellular Fap1 species was notably different from those of the secreted species, possibly because the intracellular species were less mature in terms of their glycosylation and/or processing compared to their extracellular counterparts. The truncated Fap1 was efficiently secreted in the secA2 wild-type background of strain VT2005. The secretion was not mediated by the canonical pathway, as it was not abolished by azide treatment (Fig. 4B, lanes 1 to 4). In the secA2 mutant background of strain VT2006, this truncated Fap1 was inefficiently secreted by the canonical pathway and this secretion was abolished by azide treatment (Fig. 4B, lanes 5 to 8). These data confirmed that RII partially inhibited the canonical pathway and suggested that the additional deletion of the CWA and residues 342 to 504 of NRII did not alter the relative contributions of the SecA2 pathway and the SecA pathway to Fap1 secretion.

Fap1 SP, NRI, and NRIIa are minimal requirements for secretion by the SecA2-dependent accessory pathway. To determine which specific region in SP-NRI-RI-NRIIa inhibits the SecA-dependent canonical pathway and/or promotes secretion by the SecA2-dependent accessory pathway, the SP, NRI, RI, and NRIIa were deleted individually. Mutants with a single deletion of the SP, NRI, or NRIIa exhibited similar phenotypes: all the corresponding Fap1 species were not secreted even when SecA2 was present (Fig. 5A). SP-NRI-RI in VT2107 and VT2108 was not detected by E42 (data not shown), suggesting that the E42 epitope was within the region from residues 169 to 342, which was deleted in this species. However, this species was still detected by a polyclonal antibody raised against recombinant Fap1, a nonglycosylated recombinant protein containing Fap1 residues 69 to 342 (NRI, RI, and NRIIa) expressed in E. coli (28).


Figure 5
View larger version (53K):
[in this window]
[in a new window]

  		FIG. 5. Effects of individual N-terminal domains on Fap1 secretion. (A) SP, NRI, RI, and NRIIa were deleted individually from SP-NRI-RI-NRIIa in both secA2 wild-type and mutant backgrounds. Proteins prepared from spent CM and lysates (L) from the deletion strains were analyzed by Western blotting. The first three deletion constructs were detected by monoclonal antibody E42. SP-NRI-RI (bottom panel) was detected by a polyclonal antibody against recombinant Fap1. The fap1 null mutant, VT1393, was used as a negative control for Fap1 antibodies. Each fraction contained proteins from 108 CFU. (B) Two strains expressing SP-NRI-NRIIa, VT2105 (secA2 wild type) and VT2106 (secA2 mutant), were mocked treated (–) or treated with sodium azide (+) for 30 min prior to protein extraction from spent CM and lysates. Proteins from 108 CFU were analyzed by Western blotting and detected by monoclonal antibody E42.

 
These data strongly suggested that the SP, NRI, and NRIIa are all required for the secretion of SP-NRI-RI-NRIIa. It is of interest if they are required for both the canonical and accessory pathways or for the accessory pathway specifically. Because the SP alone was sufficient for SecA-dependent secretion of SP-GFP (Fig. 3), it was predicted that NRI and NRIIa were (i) not necessary for the canonical pathway and (ii) required for the SecA2-dependent accessory pathway. The first prediction was confirmed by the secretion of SP-NRI-NRIIa in the absence of SecA2 (VT2106). In the absence of azide, SP-NRI-NRIIa was secreted efficiently by the canonical pathway; in the presence of azide, the secretion was completely inhibited (Fig. 5B, lanes 5 to 8). The second prediction was confirmed by the secretion of SP-NRI-NRIIa in the presence of SecA2 (VT2105). SP-NRI-NRIIa was secreted efficiently by the accessory pathway, as the secretion was not abolished by azide treatment (Fig. 5B, lanes 1 to 4). The second prediction was also supported by the fact that any Fap1 constructs harboring SP, NRI, and NRIIa could be secreted by the SecA2-dependent accessory pathway. Therefore, these three regions are minimal requirements for secretion by the accessory pathway.

RI inhibits secretion by the SecA-dependent canonical pathway. Although SP-RI-NRIIa and SP-NRI-RI did not meet the minimal requirements for SecA2-dependent secretion, they both met the minimal requirement for SecA-dependent secretion, which is the presence of the Fap1 SP. The question then arose as to why they were not secreted by the canonical pathway either (Fig. 5A). One pronounced common feature of these two Fap1 truncated proteins as well as NRI-RI-NRIIa was that they all possessed RI. It was possible that RI inhibited the secretion by the canonical pathway, just as RII did. When RI was deleted from SP-NRI-RI-NRIIa, the resulting SP-NRI-NRIIa was able to be completely secreted in the absence (VT2106) of SecA2 (Fig. 5A). This secretion was SecA dependent, as it was abolished by azide treatment (Fig. 5B, lanes 5 to 8). The derepression by the RI (and RII) deletion did not automatically reestablish SecA-dependent secretion in the presence of SecA2, possibly due to the presence of other regions, such as the SP, NRI, and NRIIa, which favors secretion by the SecA2-dependent accessory pathway.


   DISCUSSION
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
The roles of the individual domains of Fap1 in the selection of a secretion route were determined by the deletion of various polypeptide regions. To distinguish the contributions of the canonical pathway from those of the accessory pathway, the mutagenesis was performed in wild-type as well as in secA2 mutant backgrounds and the secretion was also analyzed in the presence of azide, which inhibits secretion by the canonical pathway. The data revealed that SecA2-mediated Fap1 secretion required the SP, NRI, and NRIIa. Furthermore, RI and RII both selectively inhibited secretion by the canonical pathway. In the absence of the inhibition, the canonical pathway was able to function normally via the Fap1 SP.

The ability of the Fap1 SP to mediate the SecA-dependent secretion of SP-GFP (Fig. 3) initially seemed difficult to reconcile with the previous finding that the N-terminal signal peptide of GspB (a Fap1 homolog) is not sufficient for the secretion of heterologous proteins (3). The very recent work from Bensing et al. (1) reveals that the main reason the GspB signal peptide is not efficiently recognized by the canonical pathway is the presence of three glycine residues in the hydrophobic domain of the signal peptide. These residues potentially inhibit the hydrophobic domain from adopting an {alpha}-helical conformation, which is essential for the function of the signal peptide in GspB. Another element essential for SecA-dependent secretion is a KSGKXW motif in the N-terminal region of the signal peptide. Among all the GspB homologs, Fap1 has the least homology in these two regions that are important for signal peptide function in GspB (see Fig. 2 in reference 1): Fap1 carries the sequences mhKSGKnWvR and avAlGAltGatVVs, and GspB contains rfkliKSGKhWlR and iiAtGAvlGgaVVt (different residues are indicated in lowercase). Perhaps the replacement of the glycine residues in the Fap1 SP would further enhance SecA-dependent secretion or inhibit SecA2-dependent secretion. However, the present data from Fap1 secretion suggest that the mechanisms of secretion routing are not the same for Fap1 and GspB. At least, the effect of the two glycine residues of the Fap1 SP on the inhibition of the canonical pathway, if any, was not apparent. This lack of effect is also supported by the highly efficient secretion of SP-NRI-NRIIa by VT2106 (Fig. 5). Nevertheless, it is consistent that the signal peptides of Fap1 and GspB are both essential for SecA2-dependent secretion (Fig. 5A, VT2101) (3).

Another consensus opinion arising from the deletion analysis of RI and RII is that these regions are not absolutely required for the SecA2-dependent secretion of Fap1 or GspB (Fig. 4) (3). The differential effects of individual domains on secretion by the accessory pathway and the canonical pathway were dissected by the deletion of defined regions in a secA2 mutant background and by azide treatment. These steps enabled us to clearly show that the single deletion of either RI or RII derepressed SecA2-dependent secretion, suggesting that they both inhibit the SecA-dependent pathway (Fig. 4 and 5). It was not determined which serine-rich repeat region contributes more to the inhibition because the single deletion of RI from full-length Fap1 (or from the RII deletion mutant) was not successful after repeated attempts. The complete secretion of SP-NRI-NRIIa by the canonical pathway in a secA2 mutant background (Fig. 5, VT2106) suggested that the inhibition effects of RI and RII may be additive. One possible explanation for the inhibition is that the carbohydrate associated with these two regions inhibits the canonical pathway, consistent with the previous finding for GspB (3). When RII was absent or both RI and RII were absent, the corresponding Fap1 species were not detected by monoclonal antibody F51 (data not shown), which detects a carbohydrate epitope on the fully glycosylated Fap1 (28, 33). These data suggested that these deletion constructs were not as fully glycosylated as full-length Fap1, which in turn released the inhibition imposed by the carbohydrate. Although the inhibition of the canonical pathway by carbohydrate was suggested based on previous studies of S. gordonii (3), it has not been experimentally proved and it is difficult to examine in the poorly transformable S. parasanguinis, as a fap1 secA2 gtf triple mutant is not available yet for similar studies.

The deletion of NRI and NRIIa individually revealed the requirements for these two regions for secretion by the SecA2-dependent accessory pathway (Fig. 5A). This finding suggested that the accessory pathway may need additional sequences besides the SP for recognition. One possibility is that these regions bear motifs specifically recognized by components of the accessory pathway. Alternatively, these regions may be important for Fap1 to adopt a conformation permissive for the accessory pathway. It is well known that the canonical pathway mediates the secretion of proteins in an unfolded conformation. However, it is unknown whether the accessory pathway mediates secretion in a conformation-dependent manner.

The mutagenesis of the CWA did not alter the secretion route of Fap1 (Fig. 2). More Fap1 was found in the CM of the CWA mutant than in that of the wild type detected with the same peptide-specific monoclonal antibody recognizing residues 169 to 342. It may be that the Fap1 from the CWA mutant cannot be anchored to CW, as shown by previous results (34). Alternatively, it is possible that CW-anchored Fap1 is or triggers a signal that no more Fap1 needs to be secreted. In the absence of the CWA, Fap1 secretion would then be deregulated. Although this hypothesis has not yet been tested in a more quantitative manner, it is possible, as the secretion of adhesin was found to be regulated in other bacteria (3).

One interesting phenomenon of the Fap1 variants without RII (Fig. 4) or without both RI and RII (Fig. 5B) was that the species in the lysate and cell wall fractions migrated more slowly than the species in the CM. This is not too surprising considering that Fap1 has an extremely complex stepwise glycosylation-processing pathway (28, 33). It is unknown where each glycosylation and processing step occurs. The Fap1 species in different locations may be glycosylated and processed to different extents. For S. gordonii, it has been shown that the apparent molecular sizes of one GspB variant can be different due to different glycosylation statuses in different locations (3). The Fap1-associated carbohydrate structure and linkage in S. parasanguinis are less well characterized. Thus far, we have not found any antibodies or lectins that can be used to differentiate the glycosylation statuses of these Fap1 species with different sizes or locations.

In summary, multiple regions, including the SP, NRI, and NRIIa, were found to be essential for recognition by the SecA2 pathway. In contrast, the SP alone was sufficient for recognition by the SecA pathway. However, the presence of RI and RII inhibited the secretion via the SecA pathway. Whether the SecA2 recognition signal and/or the SecA inhibition signal is of a peptide or carbohydrate nature remains to be clarified with a mutant that is completely defective in Fap1 glycosylation.


   ACKNOWLEDGMENTS
 
This work was supported by grants NIH/NIDCR R01-DE11000 to P. M. Fives-Taylor and NIH/NIDCR K22-DE14726 and R01-017954 to H. Wu.


   FOOTNOTES
 
* Corresponding author. Mailing address: Department of Microbiology and Molecular Genetics, College of Medicine and College of Agriculture and Life Sciences, 116 Stafford Hall, 95 Carrigan Dr., University of Vermont, Burlington, VT 05405. Phone: (802) 656-1121. Fax: (802) 656-8749. E-mail: pfivesta{at}uvm.edu Back

{triangledown} Published ahead of print on 31 August 2007. Back

{dagger} Present address: Department of Molecular Genetics and Microbiology, University of Massachusetts Medical School, 55 Lake Ave., North, Worcester, MA 01655. Back


   REFERENCES
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Bensing, B. A., I. R. Siboo, and P. M. Sullam. 2007. Glycine residues in the hydrophobic core of the GspB signal sequence route export towards the accessory Sec pathway. J. Bacteriol. 189:3846-3854.[Abstract/Free Full Text]
  2. Bensing, B. A., and P. M. Sullam. 2002. An accessory sec locus of Streptococcus gordonii is required for export of the surface protein GspB and for normal levels of binding to human platelets. Mol. Microbiol. 44:1081-1094.[CrossRef][Medline]
  3. Bensing, B. A., D. Takamatsu, and P. M. Sullam. 2005. Determinants of the streptococcal surface glycoprotein GspB that facilitate export by the accessory Sec system. Mol. Microbiol. 58:1468-1481.[Medline]
  4. Braunstein, M., B. J. Espinosa, J. Chan, J. T. Belisle, and W. R. Jacobs, Jr. 2003. SecA2 functions in the secretion of superoxide dismutase A and in the virulence of Mycobacterium tuberculosis. Mol. Microbiol. 48:453-464.[CrossRef][Medline]
  5. Chen, Q., H. Wu, and P. M. Fives-Taylor. 2004. Investigating the role of secA2 in secretion and glycosylation of a fimbrial adhesin in Streptococcus parasanguis FW213. Mol. Microbiol. 53:843-856.[CrossRef][Medline]
  6. Chen, Q., H. Wu, R. Kumar, Z. Peng, and P. M. Fives-Taylor. 2006. SecA2 is distinct from SecA in immunogenic specificity, subcellular distribution and requirement for membrane anchoring in Streptococcus parasanguis. FEMS Microbiol. Lett. 264:174-181.[CrossRef][Medline]
  7. Cristobal, S., J. W. de Gier, H. Nielsen, and G. von Heijne. 1999. Competition between Sec- and TAT-dependent protein translocation in Escherichia coli. EMBO J. 18:2982-2990.[CrossRef][Medline]
  8. Elder, B. L., and P. Fives-Taylor. 1986. Characterization of monoclonal antibodies specific for adhesion: isolation of an adhesin of Streptococcus sanguis FW213. Infect. Immun. 54:421-427.[Abstract/Free Full Text]
  9. Fenno, J. C., A. Shaikh, and P. Fives-Taylor. 1993. Characterization of allelic replacement in Streptococcus parasanguis: transformation and homologous recombination in a ‘nontransformable’ streptococcus. Gene 130:81-90.[CrossRef][Medline]
  10. Fives-Taylor, P. M., and D. W. Thompson. 1985. Surface properties of Streptococcus sanguis FW213 mutants nonadherent to saliva-coated hydroxyapatite. Infect. Immun. 47:752-759.[Abstract/Free Full Text]
  11. Froeliger, E. H., and P. Fives-Taylor. 2001. Streptococcus parasanguis fimbriae-associated adhesin Fap1 is required for biofilm formation. Infect. Immun. 69:2512-2519.[Abstract/Free Full Text]
  12. Handley, P. S., F. F. Correia, K. Russell, B. Rosan, and J. M. DiRienzo. 2005. Association of a novel high molecular mass, serine-rich protein (SrpA) with fibril-mediated adhesion of the oral biofilm bacterium Streptococcus cristatus. Oral Microbiol. Immunol. 20:131-140.[CrossRef][Medline]
  13. Kachlany, S. C., D. H. Fine, and D. H. Figurski. 2000. Secretion of RTX leukotoxin by Actinobacillus actinomycetemcomitans. Infect. Immun. 68:6094-6100.[Abstract/Free Full Text]
  14. Kolenbrander, P. E. 2000. Oral microbial communities: biofilms, interactions, and genetic systems. Annu. Rev. Microbiol. 54:413-437.[CrossRef][Medline]
  15. Lenz, L. L., S. Mohammadi, A. Geissler, and D. A. Portnoy. 2003. SecA2-dependent secretion of autolytic enzymes promotes Listeria monocytogenes pathogenesis. Proc. Natl. Acad. Sci. USA 100:12432-12437.[Abstract/Free Full Text]
  16. Macrina, F. L., J. A. Tobian, K. R. Jones, R. P. Evans, and D. B. Clewell. 1982. A cloning vector able to replicate in Escherichia coli and Streptococcus sanguis. Gene 19:345-353.[CrossRef][Medline]
  17. Meens, J., E. Frings, M. Klose, and R. Freudl. 1993. An outer membrane protein (OmpA) of Escherichia coli can be translocated across the cytoplasmic membrane of Bacillus subtilis. Mol. Microbiol. 9:847-855.[Medline]
  18. Navarre, W. W., and O. Schneewind. 1999. Surface proteins of gram-positive bacteria and mechanisms of their targeting to the cell wall envelope. Microbiol. Mol. Biol. Rev. 63:174-229.[Abstract/Free Full Text]
  19. Oliver, D. B., R. J. Cabelli, K. M. Dolan, and G. P. Jarosik. 1990. Azide-resistant mutants of Escherichia coli alter the SecA protein, an azide-sensitive component of the protein export machinery. Proc. Natl. Acad. Sci. USA 87:8227-8231.[Abstract/Free Full Text]
  20. Plummer, C., H. Wu, S. W. Kerrigan, G. Meade, D. Cox, and C. W. Ian Douglas. 2005. A serine-rich glycoprotein of Streptococcus sanguis mediates adhesion to platelets via GPIb. Br. J. Haematol. 129:101-109.[CrossRef][Medline]
  21. Posey, J. E., T. M. Shinnick, and F. D. Quinn. 2006. Characterization of the twin-arginine translocase secretion system of Mycobacterium smegmatis. J. Bacteriol. 188:1332-1340.[Abstract/Free Full Text]
  22. Saint-Joanis, B., C. Demangel, M. Jackson, P. Brodin, L. Marsollier, H. Boshoff, and S. T. Cole. 2006. Inactivation of Rv2525c, a substrate of the twin arginine translocation (Tat) system of Mycobacterium tuberculosis, increases ß-lactam susceptibility and virulence. J. Bacteriol. 188:6669-6679.[Abstract/Free Full Text]
  23. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.
  24. Schreiber, S., R. Stengel, M. Westermann, R. Volkmer-Engert, O. I. Pop, and J. P. Muller. 2006. Affinity of TatCd for TatAd elucidates its receptor function in the Bacillus subtilis twin arginine translocation (Tat) translocase system. J. Biol. Chem. 281:19977-19984.[Abstract/Free Full Text]
  25. Seifert, K. N., E. E. Adderson, A. A. Whiting, J. F. Bohnsack, P. J. Crowley, and L. J. Brady. 2006. A unique serine-rich repeat protein (Srr-2) and novel surface antigen (epsilon) associated with a virulent lineage of serotype III Streptococcus agalactiae. Microbiology 152:1029-1040.[Abstract/Free Full Text]
  26. Siboo, I. R., H. F. Chambers, and P. M. Sullam. 2005. Role of SraP, a serine-rich surface protein of Staphylococcus aureus, in binding to human platelets. Infect. Immun. 73:2273-2280.[Abstract/Free Full Text]
  27. Spatafora, G., N. Van Hoeven, K. Wagner, and P. Fives-Taylor. 2002. Evidence that ORF3 at the Streptococcus parasanguis fimA locus encodes a thiol-specific antioxidant. Microbiology 148:755-762.[Abstract/Free Full Text]
  28. Stephenson, A. E., H. Wu, J. Novak, M. Tomana, K. Mintz, and P. Fives-Taylor. 2002. The Fap1 fimbrial adhesin is a glycoprotein: antibodies specific for the glycan moiety block the adhesion of Streptococcus parasanguis in an in vitro tooth model. Mol. Microbiol. 43:147-157.[CrossRef][Medline]
  29. Takahashi, Y., K. Konishi, J. O. Cisar, and M. Yoshikawa. 2002. Identification and characterization of hsa, the gene encoding the sialic acid-binding adhesin of Streptococcus gordonii DL1. Infect. Immun. 70:1209-1218.[Abstract/Free Full Text]
  30. Takamatsu, D., B. A. Bensing, and P. M. Sullam. 2004. Genes in the accessory sec locus of Streptococcus gordonii have three functionally distinct effects on the expression of the platelet-binding protein GspB. Mol. Microbiol. 52:189-203.[CrossRef][Medline]
  31. van Wely, K. H., J. Swaving, R. Freudl, and A. J. Driessen. 2001. Translocation of proteins across the cell envelope of Gram-positive bacteria. FEMS Microbiol. Rev. 25:437-454.[Medline]
  32. von Heijne, G. 1998. Life and death of a signal peptide. Nature 396:111-113.[CrossRef][Medline]
  33. Wu, H., S. Bu, P. Newell, Q. Chen, and P. Fives-Taylor. 2007. Two gene determinants are differentially involved in the biogenesis of Fap1 precursors in Streptococcus parasanguis. J. Bacteriol. 189:1390-1398.[Abstract/Free Full Text]
  34. Wu, H., and P. M. Fives-Taylor. 1999. Identification of dipeptide repeats and a cell wall sorting signal in the fimbriae-associated adhesin, Fap1, of Streptococcus parasanguis. Mol. Microbiol. 34:1070-1081.[CrossRef][Medline]
  35. Wu, H., K. P. Mintz, M. Ladha, and P. M. Fives-Taylor. 1998. Isolation and characterization of Fap1, a fimbriae-associated adhesin of Streptococcus parasanguis FW213. Mol. Microbiol. 28:487-500.[CrossRef][Medline]

Journal of Bacteriology, November 2007, p. 7610-7617, Vol. 189, No. 21
0021-9193/07/$08.00+0     doi:10.1128/JB.00748-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Other Versions of this Article:
JB.00748-07v1
189/21/7610    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Chen, Q.
Right arrow Articles by Fives-Taylor, P. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chen, Q.
Right arrow Articles by Fives-Taylor, P. M.

Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
Appl. Environ. Microbiol. Infect. Immun. Eukaryot. Cell
Mol. Cell. Biol. J. Virol. Microbiol. Mol. Biol. Rev.
ALL ASM JOURNALS