The Tra Domain of the Lactococcal CluA Surface Protein Is a Unique Domain That Contributes to Sex Factor DNA Transfer
http://www.100md.com
《细菌学杂志》
Institute of Food Research, Norwich Research Park, Norwich NR4 7UA, United Kingdom
ABSTRACT
CluA is a cell surface-presented protein that causes cell aggregation and is essential for a high-efficiency conjugation process in Lactococcus lactis. We know from previous work that in addition to promoting cell-to-cell contact, CluA is involved in sex factor DNA transfer. To define the CluA domains involved in aggregation and in transfer, we first performed random mutagenesis of the cluA gene using a modified mini-Tn7 element which generated five amino acid insertions located throughout the encoded protein. Thirty independent cluA insertion mutants expressing modified CluA proteins at the cell surface were isolated and characterized further. The level of aggregation of each mutant was determined. The cell binding capacity of CluA was affected strongly when the protein had a mutation in its N-terminal region, which defined an aggregation domain extending from amino acid 153 to amino acid 483. Of the cluA mutants that still exhibited aggregation, eight showed an attenuated ability to conjugate, and six mutations were located in a 300-amino-acid C-terminal region of the protein defining a transfer domain (Tra). This result was confirmed by a phenotypic analysis of an additional five mutants obtained using site-directed mutagenesis in which charged amino acids of the Tra domain were replaced by alanine residues. Two distinct functional domains of the CluA protein were defined in this work; the first domain is involved in cell binding specificity, and the Tra domain is probably involved in the formation of the DNA transport machinery. This is the first report of a protein involved in conjugation that actively contributes to DNA transfer and mediates contact between donor and recipient strains.
INTRODUCTION
CluA is a 136-kDa surface-bound protein encoded by the chromosomally located sex factor of Lactococcus lactis MG1363 (16); it is associated with cell aggregation linked to high-frequency transfer of the sex factor (13, 14). The amino terminus of CluA has a signal sequence typical of secreted proteins, and the carboxy terminus has a cell wall binding motif that includes a conserved LPXTGE motif (11). CluA exhibits significant homology with two groups of cell surface proteins found in other gram-positive bacteria (16): two enterococcal proteins involved in plasmid conjugation, Asa1 (12) and Asc10 (22), and two streptococcal proteins involved in binding to salivary glycoproteins, SpaP (23) and Ssp5 (8). A comparison of the amino acid sequences of CluA with the sequences of these four other proteins led to identification of six distinct domains in CluA (16), including four conserved domains (domains I, II, IV, and V), a variable domain (domain III), and an additional domain (designated domain IVB in this study) that is specific to CluA and is located at the C-terminal end between domains IV and V. In previous work, we investigated the role of CluA in aggregation and conjugation by controlling the level of protein expression using the lactococcal nisA promoter (29). In a sex factor-negative MG1363 derivative, overexpression of CluA resulted in aggregation of the cells, indicating that CluA is the only sex factor component required for aggregation. We clearly showed that CluA enhances sex factor transfer when it is expressed in a donor strain having a cluA gene deletion on the chromosome. However, expression of CluA in the recipient strain could not complement the lack of CluA in a cluA donor strain. This led us to the conclusion that CluA, in addition to its capacity to trigger cell-to-cell contact, has a direct role in DNA transport. This was surprising since the Enterococcus faecalis pheromone-inducible aggregation substance Asc10 provides efficient conjugation when the protein is expressed in either the donor or recipient cells (26). The functional domains of Asc10 were characterized using the phage-based TnlacZ/in and TnphoA/in transposition systems to construct a library of 22 in-frame insertions encoding 31 amino acids located throughout Asc10 (36). The plasmid conjugation efficiencies of the different insertion mutants correlated with aggregation in the different mutants, as expected. In our further characterization of CluA we used the well-developed pentapeptide scanning mutagenesis (PSM) technique (17) to insert randomly a variable 5-amino-acid cassette throughout CluA in order to determine the functional domains of the protein involved in aggregation and also the domain(s) that plays a role in DNA transfer.
PSM is a transposon-based strategy that generates a small fingerprint (15 bp) in the gene of interest, leading to a random 5-amino-acid insertion into the encoded protein. PSM can be performed after in vivo transposition using the Tn4430 transposon originating from Bacillus thuringiensis that transposes efficiently in Escherichia coli (17, 18) or by using an in vitro transposition method based on the use of a Tn7 derivative (2). The Tn7-based technique uses the TnsAB transposase associated with the ATP-dependent DNA binding protein TnsCA225V (28) to insert in vitro a mini-Tn7 element containing the Tn7 terminal cis-acting transposition sequences flanking an antibiotic resistance gene at random locations in the target DNA (for a review see reference 6). The mini-Tn7 PmeI element is altered at both ends to generate PmeI restriction endonuclease sites (2). Removal of the mini-Tn7 element using the PmeI restriction endonuclease sites leads to a 15-bp insertion that includes a unique PmeI site. An in-frame stop codon is generated in two of the six reading frames due to the UAA stop codon included in the PmeI sequence. The generation of these truncated proteins is the major limitation of this system for use in analyses of surface proteins anchored to the cell wall. Out-of-frame insertions occurring in the N-terminal region of the LPXTGE motif lead to mutated proteins that lack the membrane-associated domain and the cell wall anchor motif and could cause missorting or release of the proteins into the culture medium. Another limitation of this system is the use of the mini-Tn7 PmeI selection markers, which are suitable only in gram-negative bacteria.
In this study, we developed a mutagenesis system suitable for gram-positive bacteria that produces only full-length proteins. This system was used successfully to generate a collection of derivatives with pentapeptide insertions located throughout CluA. These derivatives were employed for identification of regions of the protein associated with stability, aggregation, and conjugal transfer. Individual amino acid substitutions were created to further confirm the definition of the transfer domain.
MATERIALS AND METHODS
Media, growth conditions, and transformations. E. coli was grown at 37°C in Luria-Bertani medium (27) supplemented with 15 μg/ml of chloramphenicol or 150 μg/ml of erythromycin if necessary. L. lactis strains were grown at 30°C in M17 medium (30) with 0.5% glucose (GM17) or 0.5% sucrose (SM17) for selection of the nisin-sucrose transposon. Antibiotic resistance markers in L. lactis were selected using 5 μg/ml chloramphenicol or 5 μg/ml erythromycin. E. coli and L. lactis electrocompetent cells were prepared and transformed by the methods of Dower et al. (10) and Holo and Nes (19), respectively.
Bacterial strains. E. coli strain TG1 (15) was used for cloning experiments and plasmid propagation. L. lactis rifampin-resistant strain FI9979 (29) is a sex factor-negative strain that has a chromosomal copy of the complete nisin operon with a 20-bp insertion into nisA (9). This strain was used as a cluA background strain to monitor the aggregation of the different CluA mutants. Strains FI9980 and FI9981 (29) are FI9979 derivatives that harbor pFI2209, an expression vector containing the PnisA promoter, and pFI2213, the same expression vector containing the wild-type cluA gene, respectively. These two strains were used as negative and positive controls to assess aggregation. FI9980 was also used as a recipient strain in conjugation experiments. L. lactis strain FI9983 (29) is a cluA derivative of FI8164 (containing a complete sex factor with the tetM gene from Tn1545 [4] inserted into the tellurium resistance genes [C. J Pillidge, personal communication]). FI9983 also carries a chromosomal copy of the complete nisin operon with a 20-bp insertion into nisA. This strain was used in a complementation study to monitor the contributions of the different variants of CluA to conjugation. FI9984 (FI9983 containing the pFI2209 vector) and FI9985 (FI9983 carrying pFI2213) (29) were used as negative and positive controls, respectively, to assess conjugation efficiency.
Plasmid constructs. Phagemid pBluescript SK(–) (Stratagene) and plasmid pBR322 (3) were used for cloning experiments with E. coli TG1 (15). Plasmid pFI2453 was used as a target for insertion mutagenesis. This plasmid is a derivative of pFI2213 (expression vector pFI2209 containing the wild-type cluA gene under control of the PnisA promoter) (29) in which the unique BsrGI site was removed by blunt ending the 5' overhangs using the T4 DNA polymerase (Promega) under the conditions described by the manufacturer. All the relevant plasmids resulting from cluA insertional mutagenesis are listed in Table 1. Donor plasmid pFI2452 carrying the Eryr mini-Tn7 BsrGI element was constructed as follows. A 1,432-bp fragment of the pGPS4 plasmid (New England Biolabs) containing the Cmrr mini-Tn7 PmeI element was isolated after digestion with KpnI and SacI and cloned into the SacI/KpnI restriction endonuclease sites of pBSK–, generating pFI2449. The chloramphenicol resistance gene (903 bp) was then removed after BamHI digestion and religation of the vector, generating pFI2450. To amplify the Tn7 recombination sequences, PCR was performed using primers LmaB and RmaB (see below) and pFI2450 as the template. The 508-bp PCR product was then digested with EcoRI and AvaI and inserted into the EcoRI/AvaI restriction endonuclease sites of pBR322, yielding pFI2451. The erythromycin resistance gene (ermC) of plasmid pE194 (20) was excised from pUC7E (Claire Shearman, personal communication) after digestion with BamHI and inserted into the BamHI restriction endonuclease site of pFI2451, which yielded pFI2452.
DNA manipulation, PCR, and sequencing. General molecular cloning techniques and electrophoresis of DNA on agarose gels were carried out essentially as described by Sambrook et al. (27). The PCRs were performed using the BIO-X-ACT proofreading polymerase as described by the manufacturer (Bioline). Primers LmaB (GATGAATTCTTTAGTGTGTACACACAATAAAGTTGGGAACTG) having an EcoRI site at its 5' end and RmaB (GATCTCGGGCTAAATGTGTACACACAATAAAGTCTTAAACTG) having an AvaI site at its 5' end were used to replace the PmeI sites of pGSP4 by PCR-directed mutagenesis (BsrGI restriction endonuclease sites are underlined in both primers). Nucleotide sequencing was performed by the dideoxy chain termination method, using fluorescent dye terminators and cycle sequencing reactions. Samples were analyzed with an Applied Biosystems 373A DNA sequencer.
Site-directed mutagenesis of cluA. PCR mutagenesis was performed with the cluA gene by using the BIO-X-ACT proofreading polymerase (Bioline) to replace charged amino acids with alanine residues in CluA. Two sets of primers were used to amplify two cluA fragments from MG1363 chromosomal DNA. A primer upstream of the charged amino acid cluster was paired with a mutagenic oligonucleotide to hybridize at the region within cluA at which the mutation occurred. The mutagenic primer contained the sequence specifying the bases for the alanine substitutions. A second PCR was performed by using a mutagenic oligonucleotide that was complementary to the oligonucleotide used in the first amplification paired with a primer downstream from the desired substitution. The two PCR products were mixed and used as a template in crossover PCR (24), using the two external primers to generate a recombined PCR fragment containing the alanine substitutions. The resulting PCR fragment was then digested using restriction enzymes whose sites were incorporated into the external primers. The fragment was cloned into the cluA gene carried by the pFI956 plasmid, a pBluescript SK(–) vector containing the cluA gene (29), to replace the wild-type cluA internal fragment. The fragments were sequenced in order to verify that no misincorporation had occurred in the DNA sequence during PCR amplification. The cluA gene containing the point mutations was excised with SmaI and XhoI and cloned into the pFI2209 expression vector (29). The five resulting plasmids are listed in Table 2.
Transposition reactions. The in vitro transposition reaction was performed with pFI2452 using the conditions described for the GPS-LS linker-scanning system (New England Biolabs) by the manufacturer. One microliter of the transposition reaction was used to transform E. coli TG1. The pBR322-based pFI2452 donor plasmid was digested prior to the transposition reaction with both the PvuI and PvuII restriction enzymes, each of which cut once in the plasmid outside the mini-Tn7 BsrGI site to prevent replication in E. coli. A 0.02-μg portion of the pFI2452 digested donor DNA and 0.08 μg of the pFI2453 target DNA were used in the transposition reaction. After transformation of E. coli with the transposition reaction mixture DNA, transposition events were selected on plates containing erythromycin. A pool of plasmids extracted from 400 colonies was transformed into L. lactis FI9979, and the transformants were selected with erythromycin.
Nisin induction and quantification of the aggregation level. Appropriate dilutions of nisin (Aplin & Barret, Ltd.) were prepared for induction of the bacterial strains. One milliliter of an overnight cell culture grown in the presence of 50 ng/ml nisin was loaded into a 48-well microtitration plate to screen for L. lactis transformants containing mutated variants of pFI2213 that did not exhibit aggregation. The plate was then inverted several times until aggregation was visible. For a quantitative comparison of aggregation by the 30 different insertion mutants, a 1.5-ml overnight culture of the strains in SM17 containing 50 ng/ml of nisin was vortexed for 1 min to trigger aggregation and left at room temperature for 1 h, which allowed the aggregates to sediment. A 100-μl sample was then taken from the top (containing culture media with nonaggregated cells) and transferred to a 96-well microtitration plate. A 100-μl sample of the same culture that had not been vortexed was transferred to the same plate. The optical density at 600 nm (OD600) of each sample was determined simultaneously using a THERMOmax microplate reader (Molecular Devices). The optical density of each mutant in both conditions was determined three times independently. The level of aggregation was determined using the following formula: A% = 100 – (ODv/ODnv x 100), where A% is the percentage of aggregation, ODv is the OD600 of the culture that had been vortexed, and ODnv is the OD600 of the same culture that had not been vortexed.
Surface extraction and Western immunoblotting. A lysozyme surface extract of each strain was obtained as previously described (29). Briefly, the pellet from a 10-ml culture in SM17 was resuspended in a solution containing 50 mM Tris-HCl (pH 8), 0.1x protease inhibitor cocktail set II (Calbiochem), 10% sucrose, and 25 mg/ml lysozyme. The OD600 of the suspension was adjusted to 10, and the suspension was shaken gently at room temperature for 1 h and centrifuged at 3,500 rpm. Twenty microliters of the supernatant was loaded on a NuPage (Invitrogen) sodium dodecyl sulfate-polyacrylamide gel for electrophoresis. The amount of protein for each sample was checked by quantifying the signals using the TotalLab software (Nonlinear Dynamics) after staining of the gel with colloidal blue stain (Invitrogen). Western blot analyses were performed using a Western Breeze chromogenic kit by following the manufacturer's instructions (Invitrogen). Polyclonal antibodies raised against CluA (29) were used as primary antibodies. Quantification of the signals detected was performed using the TotalLab software.
Sex factor transfer. The FI9983 donor strain (tetracycline resistant) harboring plasmids carrying wild-type cluA, the different cluA insertion mutations, or the cluA point mutations or the strain carrying only the vector was grown overnight and diluted 100-fold in SM17 containing 5 μg/ml of chloramphenicol and 50 ng/ml of nisin. Recipient strain FI9980 (rifampin resistant) was grown under the same conditions but in the absence of nisin. Cell mating was triggered by centrifugation of a mixture of donor and recipient strains grown to the same optical density and not by vortexing as previously described (29). The conjugation efficiencies of the different strains were determined for donor and recipient cells grown to an OD600 of 0.8. Then 0.1 ml of donor cells and 0.9 ml of recipient cells were mixed, and the mixture was centrifuged at 5,000 rpm for 5 min. The pellet was resuspended in 1 ml of GM17 for all donor strains. Each mixture was incubated for 1 h at 30°C. Cells were then diluted and plated on GM17 agar containing rifampin (for recipient and transconjugant selection) or tetracycline (for donor and transconjugant selection) or both antibiotics (for transconjugant selection). The conjugation frequency for each donor was determined at least three times independently.
Statistical analysis. Statistical significance was calculated by determining confidence intervals for the differences of two population means when population variances were known. Cluster analysis was carried out to classify the strains on the basis of sex factor transfer level and aggregation. Hierarchical clustering (35) was used to determine the optimum number of clusters. The mean values for these clusters were used as seeds for a nonhierarchical cluster analysis (k-means) in order to optimize the classification of the strains. Statistical differences (P < 0.05) in the sex factor transfer level and/or level of aggregation between the three clusters were assessed by the Student-Newman-Keuls test for comparison of multiple means.
RESULTS
Construction of a CluA insertion library using an in vitro transposon mutagenesis system producing only full-length proteins. A transposon-based technique was used to define the regions of the CluA protein that are important for aggregation and the regions that may contribute to DNA transfer. To do this, we first engineered the mini-Tn7 PmeI element included in the GPS-LS linker-scanning system (New England Biolabs) to make it suitable for the study of CluA. This minitransposon is flanked by two PmeI restriction endonuclease sites that result in one-third of the proteins being truncated after mutagenesis. We therefore decided to replace this restriction endonuclease site with a BsrGI site using primers LmaB and RmaB (see Materials and Methods). Introduction of a BsrGI site involved changing the R4 and R7 residues of the terminal inverted repeats of the mini-Tn7 element, and these nucleotides could be altered without affecting the transposition activity (2). Furthermore, the resistance genes used to select the transposition events in the mini-Tn7 PmeI system are suitable for use only in gram-negative bacteria. Consequently, we used a gram-positive erythromycin resistance gene that was also functional in E. coli to replace the chloramphenicol resistance cassette flanked by the Tn7 terminal recombination sequences in mini-Tn7 PmeI. Construction of the Eryr mini-Tn7 BsrGI system carried by donor plasmid pFI2452 is described in detail in Materials and Methods. The new Eryr mini-Tn7 BsrGI system led, after excision of the mini-Tn7 element using BsrGI, to insertion of five amino acids in all six frames (Fig. 1), thus avoiding production of truncated proteins.
Independent 15-bp in-frame insertions located throughout cluA were obtained after use of the Eryr mini-Tn7 BsrGI element in a PSM experiment. For this, the new donor plasmid pFI2452 was used in an in vitro transposition reaction with the target plasmid pFI2453, a derivative of the pFI2213 plasmid carrying the wild-type cluA gene under control of the PnisA nisin promoter (29) in which the unique BsrGI site was removed. The transposition events were selected in E. coli using erythromycin, and a pool of plasmids extracted from 400 colonies was transformed into L. lactis FI9979, a suitable strain for nisin induction and CluA aggregation assessment. To select for transposon insertions into the cluA gene, the aggregation abilities of 400 independent clones were tested, and 152 transformants were unable to clump. The mini-Tn7 element was excised from a pool of the 152 corresponding plasmids using BsrGI, and the resulting plasmids were transformed back into FI9979. Of 130 nonaggregating transformants, 58 mutants carried a smaller plasmid with at least the 3' half of cluA missing, resulting from DNA rearrangement events. In previous work, it has been observed that the full-length cluA gene is unstable in E. coli (29). The plasmids of the other 72 transformants were analyzed by restriction mapping and nucleotide sequencing in the vicinity of each insertion. Mapping of each plasmid revealed that 10 of the insertions had occurred between the nisin promoter and the ATG start codon of cluA. Surprisingly, seven insertions were located downstream and close to the cluA terminator. Fifty-five insertions were located in the cluA gene coding sequence, and there were 31 independent insertions. The designations of these insertional mutations indicate the amino acid residue encoded by the codon that was disrupted by the insertion or that preceded this codon (Table 1). These insertions were equally distributed along the cluA open reading frame (Fig. 2A), showing clearly that the mini-Tn7 BsrGI element constructed in this work provided random insertions located throughout a target gene.
Consequences of insertions for the stability of CluA and determination of an N-terminal region responsible for cell aggregation. Previous to this study no structural data about CluA were available. We performed a Consensus Secondary Structure Prediction analysis of CluA using the Network Protein Sequence Analysis server (http://npsa-pbil.ibcp.fr) (5), and the results for the predicted alpha-helices are shown in Fig. 2B. In predicted domain I, the 14-amino-acid hydrophobic region (T4 to L17) is likely to be the region of the signal peptide that organizes into an alpha-helical conformation when it is in contact with the membrane lipid phase (34). Domain II consists mainly of alpha-helical structures, and a major structure is located between S75 and I154. Another important helical region is located in domain IV (T583 to Y600), and a final predicted alpha-helix is located downstream from the LPKTGE motif in domain V (A1207 to R1227), corresponding to the membrane-anchoring domain typical of cell wall-anchored proteins (11). Some pentapeptide insertions that occurred in predicted alpha-helical domains of CluA and led to protein instability are described below.
The presence of CluA on the cell surface of the 31 mutants was assessed by immunodetection of CluA following cell surface extraction, as shown in Fig. 3A. The controls were the negative control FI9980, carrying the pFI2209 empty vector, and FI9981, carrying pFI2213 that contained the wild-type cluA gene expressed under control of the PnisA nisin promoter. As expected, the cell surface extract of FI9980 contained no reactive CluA protein, whereas wild-type CluA expressed in FI9981 was detected. For most of the insertion mutants, CluA was detected on the Western blot (Fig. 3A). The mutant carrying the P485 mutation contained no CluA and was not used in further experiments. A 136-kDa signal corresponding to the size of CluA was observed for all mutants except the strains carrying the Y119 and T125 mutations. For most of the mutants the total amount of detected signals comprising full-size CluA and its degradation products was similar to the amount of full-size CluA detected in FI9981 (data not shown). Since the extraction protocol released C-terminal cell wall-anchored CluA variants, it is likely that the 30 mutants expressed a full-size protein on the cell surface at a level comparable to that of wild-type CluA in FI9981. The effects of the 30 different insertions on CluA aggregation were quantified (see Materials and Methods), and the results are shown in Fig. 3B.
Aggregation was completely prevented for Y119 and T125 insertions, and this result agrees with the strong protein degradation observed after cell surface protein extraction (Fig. 3A). No signal above 70 kDa was detected for Y119 and T125, and a smear of strong bands migrating between 50 and 70 kDa corresponding to degraded CluA products was observed. This indicates that insertions located at positions Y119 and T125 affected the stability of CluA either in vivo or during the extraction process. According to the secondary structure prediction, these two insertions are located within the major predicted alpha-helical structure of predicted domain II (Fig. 2B), suggesting that this structure is important for protein stability.
For the different CluA mutants in which the 136-kDa product was observed, the insertions affecting aggregation were located mainly in the N-terminal half of the protein in a region bordered by the amino acids D153 and I483. Although the D153, Q360, H361, and I483 mutants showed little or no protein degradation after surface protein extraction (Fig. 3A), they were unable to aggregate (Fig. 3B). Likewise, no protein degradation was observed for the S400 and W410 mutants, whereas their aggregation abilities were significantly reduced (P < 0.01). This indicates that the insertions in these six mutants occurred in regions that are important for aggregation and likely to be involved in the control of CluA binding specificity. Although the M281 mutant showed low aggregation ability, the faint signals observed for the 136-kDa product and the degradation products suggest that an insertion at this location led to lower expression of the protein on the cell surface and/or decreased its stability.
The low levels of aggregation observed for the four mutants carrying the insertions located at positions L519 to E605 can be explained by the detection of a very weak signal at 136 kDa, whereas strong degradation products were observed around 60 kDa. The three insertions other than L519 are located within (T583 and N583) or close to (E605) the T583-Y600 predicted alpha-helical structure, suggesting that this region is also important for CluA stability. Although the mutants with mutations located in the D838-to-A1157 region exhibited reduced aggregation, all the corresponding proteins exhibited some degradation. In particular, the major product observed for the T858, T1083, and T1141 insertions was a 120-kDa CluA degradation product (Fig. 3A). This suggests that part of the C-terminal region is involved in aggregation and/or that the stability of the protein is affected by these insertions. In summary, the analysis of the insertions revealed a CluA domain that is responsible for cell-to-cell binding, including amino acids D153 to I483.
Definition of a CluA domain involved in transfer of the sex factor. In L. lactis, expression of CluA in the recipient strain was unable to complement DNA transfer of the sex factor from a donor strain lacking a cluA functional gene, whereas complementation took place when CluA was expressed in a cluA donor strain (29). This observation led to the conclusion that in addition to promoting cell-to-cell contact, CluA is involved in the sex factor DNA transfer machinery and that an association of CluA with one component or several components is necessary to enable DNA transfer.
It was anticipated that the region of the CluA protein associated with DNA transfer was different from the region involved directly in aggregation. The conjugation efficiencies of the mutants that were still able to aggregate were assessed to discriminate between potential insertions that affect this functional domain. The 17 plasmids carrying CluA mutations that were still able to promote aggregation (aggregation, >35%) were transformed into the FI9983 donor background strain containing a chromosomal cluA deletion. The sex factor transfer efficiencies were determined for the strain containing the empty expression vector (FI9984), the expression vector harboring the wild-type cluA gene (FI9985), and the 17 mutated plasmids. Since enhanced DNA transfer seems to be dependent on aggregation ability (12, 13, 29, 33) (see above) (i.e., on the ability of the donor strain to promote cell-to-cell contact), transfer efficiency data are presented in relation to aggregation efficiency in Fig. 4. We performed a cluster analysis using Ward's method (35) and identified three clusters of insertion mutants (Fig. 4). The first cluster (cluster 1) comprised the insertion mutants in which there was little effect on aggregation or sex factor transfer and which acted like the FI9985 CluA control. Mutants in this cluster exhibited the highest level of aggregation (76%) and the highest sex factor transfer value (80.4 x 10–4 transconjugant/donor cell on average), and both values were significantly different from the values for clusters 2 and 3. The second cluster (cluster 2) contained the CluA variants that exhibited a lower level of aggregation, which led to a lower sex factor transfer value. This was the result predicted if sex factor transfer relies solely on the level of aggregation. The third cluster (cluster 3) contained the insertion mutants that had intermediate levels of aggregation but sex factor transfer values that were lower than those that were predicted. There were no significant differences in the aggregation values between clusters 2 and 3 (ca. 48 to 49%), whereas the strains belonging to cluster 3 exhibited the lowest sex factor transfer efficiency, 14 x 10–4 transconjugant/donor cell and this value was significantly different from the transfer efficiency detected in cluster 2 (60.2 x 10–4 transconjugant/donor cell). The low transfer frequency observed in cluster 3 strains shows that these insertion mutants were affected in regions of CluA that play an important role in CluA's DNA transfer capacity. In this cluster, six of eight mutations affecting CluA were located in a region covering the 40 amino acids of the C-terminal end of predicted domain IV and in domain IVB (Fig. 2) specific for CluA. The M266 and W410 mutants had mutations that were located in the predicted variable domain III, and these two areas of CluA are also likely to be involved in DNA transfer.
Site-directed mutagenesis of the CluA Tra domain. To investigate further and refine the region of CluA involved in DNA transfer, we introduced point mutations in this part of the protein. In their study of RbsT's activity, a protein involved in stress response in Bacillus subtilis (38), Woodbury and coworkers replaced clusters of charged amino acids located throughout the length of the protein with alanine residues. Charged amino acids were chosen because they are residues that are likely to be involved in protein-ligand interactions (39) with a high probability that they are surface exposed and, consequently, less likely to affect the protein's general stability (38).
An analysis of charged amino acids groups in the entire CluA protein showed that these residues are the residues that are most frequently found in the transfer domain. Five clusters of charged amino acid residues located along the length of the CluA transfer region were found (Fig. 5). The charged amino acids at the sites were replaced by the uncharged amino acid alanine by using a PCR-based mutagenesis technique (see Materials and Methods for details). The five plasmids containing the five different sets of mutations are listed in Table 2. The alanine clusters in the transfer region of the five CluA mutants were designated Tra1 to Tra5 according to the orientation from the N terminus to the C terminus. The plasmids were used to transform FI9979, and the aggregation abilities of the resulting strains were assessed. All the strains exhibited levels of aggregation similar to that of the strain carrying the wild-type cluA plasmid (data not shown). The conjugation efficiency of each cluA variant expressed in the FI9983 donor strain was determined (Table 2), and in parallel, Western blotting was performed with cell surface extracts of the different donors (data not shown). The signals detected showed that the protein abundance and apparent size were similar to those of the wild type. Compared to the donor strain expressing wild-type CluA, the donor strains containing the Tra alleles were all affected in the DNA transfer capacity. The strain containing the mutant allele Tra2 had a 13-fold-lower DNA efficiency than the wild type, strains containing Tra3 and Tra5 had a 6-fold-lower DNA efficiency, and strains containing Tra1 and Tra4 had a 3-fold-lower transfer efficiency. The results for noninduced cells are also shown in Table 2, and the conjugation frequencies were similar to the conjugation frequency obtained for the same donor strain lacking cluA. Overall, the results showed clearly that charged amino acids belonging to the newly defined CluA transfer (Tra) domain bordered by amino acids K784 and K1056 are involved in the ability of CluA to transfer DNA, in addition to triggering cell-to-cell contact during conjugation.
DISCUSSION
The functional analysis of 30 cluA insertion mutants generated in this study identified two regions of the CluA protein with distinct functions: the D153-I483 region of CluA that mediates bacterial aggregation and the K784-K1056 Tra domain that is involved in DNA transfer and is responsible for a high conjugation frequency.
To define the regions of CluA important in promoting cell-to-cell binding, we determined and compared the levels of aggregation of the 30 cluA insertion mutants. Six of the CluA variants showed complete or partial inhibition of aggregation compared to the wild-type CluA strain (Fig. 3). These mutants defined a region extending from amino acid D153 to amino acid I483. This D153-I483 region includes predicted variable domain III of CluA, and this is in agreement with the two aggregation domains identified in Asc10 of E. faecalis, the K156-A358 region (37) that covers the N-terminal half of the Asc10 variable domain and the Q473-L683 region (36) that includes its C-terminal third. Furthermore, a 92-amino-acid region in the variable domain of the E. faecalis Asa1 aggregation substance has also been shown to play a dominant role in cell aggregation (25).
The CluA D153-I483 aggregation domain might extend farther into predicted domain IV as downstream from I483, four insertions (L519, T583, N593, and E605) were also affected in aggregation. However, the variant CluA proteins exhibited a degraded protein pattern, and it is not clear whether this occurred during extraction or in vivo, leaving the question of whether these insertions have a direct effect on aggregation open.
Five insertions in the C-terminal part of CluA (D838 to A1157) partially impaired CluA's aggregation capacity, and this region is likely to play a role in aggregation. In a similar way, the C-terminal domain of Asc10 has been shown to play an essential role in aggregation, and specific regions that bring the N-terminal functional domain into the correct conformation to mediate aggregation have been identified (36).
It is not known how the CluA protein interacts with the lactococcal cell surface to cause aggregation. In E. faecalis aggregation is mediated by a carbohydrate recognition process in which the lipoteichoic acid, a surface molecule that contains sugar residues, is the potential bacterial receptor (32, 37). Interestingly, the L238-V486 region of CluA exhibits significant (E-value, 2 x 10–4) homology (24% sequence identity) with the L554-I750 region of the central domain (21) of the SspB protein (7), a member of the streptococcal antigen I/II family of protein adhesins that recognize salivary agglutinin glycoprotein. The structure of this domain in Streptococcus mutans antigen I/II has been determined, and the results revealed its potential to bind sugars (31). Overall, these results suggest that CluA might also recognize and bind specific sugar motifs presented on the surface of some L. lactis strains (33), including MG1363.
Aggregating CluA mutants were tested for the ability to transfer the sex factor. All insertions located in the Tra domain, unique to CluA compared to other gram-positive cell surface protein homologues (16), were affected in sex factor transfer. The importance of this region was confirmed by replacement of charged amino acids located along the length of the Tra domain, which showed that these amino acids are essential for optimized DNA transfer. It is known that electrostatic interactions provided by charged amino acids at the binding interface play an important role in protein-ligand interactions (39). Sex factor transfer could be the result of a protein-ligand association involving the Tra domain of CluA and another component of the conjugation machinery, the nature of which still needs to be determined. A sequence database search with BLAST (1) using the Tra domain as a probe gave no significant homologies with other proteins.
Additional areas of CluA are essential for sex factor transfer. A decreased level of sex factor transfer was observed with the M266 and W410 insertions. These two positions in the CluA N terminus might play a role in DNA transfer by interacting with a component of the DNA transfer machinery or by directly or indirectly interacting with the CluA Tra domain to optimize DNA transfer.
Surprisingly, in mutants exhibiting similar levels of aggregation (Fig. 3), the sex factor transfer can vary, as observed for the S400 mutant, whose sex factor transfer was four times higher than that of the W410 mutant. The difference occurred despite the closeness of the two insertions, as S400 and W410 are only 9 amino acids apart. It is likely that the amino acid sequences SCVHR inserted into S400 and VVYTT inserted into W410 (Table 1) disrupted the protein differently, thus leading to noticeable differences in the capacity of the mutated CluA to transfer DNA.
This is the first report of a protein involved in conjugation that actively contributes to DNA transfer and also mediates contact between donor and recipient strains. To achieve this, we developed a mutagenesis tool adapted to the structural and functional analysis of CluA. The newly developed miniTn7 BsrGI transposon has potential for use in diverse protein analyses in which full-length mutated proteins are required. Furthermore, the use of the erythromycin resistance gene to select miniTn7 insertions makes it an appropriate tool for protein studies in diverse species of gram-positive bacteria.
ACKNOWLEDGMENTS
We thank Carmen Pin and Yvan Lemarc for their help with statistical analysis and John Payne for help with DNA sequencing.
R.S. was funded by BBSRC contract 99123941.
REFERENCES
Altschul, S. F., T. L. Madden, A. A. Schffer, J. Zhang, Z. Zhang, W. Miller, and D. J. Lipman. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25:3389-3402.
Biery, M. C., F. J. Stewart, A. E. Stellwagen, E. A. Raleigh, and N. L. Craig. 2000. A simple in vitro Tn7-based transposition system with low target site selectivity for genome and gene analysis. Nucleic Acids Res. 28:1067-1077.
Bolivar, F., R. L. Rodriguez, P. J. Greene, M. C. Betlach, H. L. Heyneker, and H. W. Boyer. 1977. Construction and characterization of new cloning vehicles. II. A multipurpose cloning system. Gene 2:93-113.
Caillot, F., C. Carlier, and P. Courvalin. 1987. Physical analysis of the conjugative shuttle transposon Tn1545. Plasmid 17:58-60.
Combet, C., C. Blanchet, C. Geourjon, and G. Deleage. 2000. NPS@: Network Protein Sequence Analysis. Trends Biochem. Sci. 25:147-150.
Craig, N. L. 1996. Transposon Tn7. Curr. Top. Microbiol. Immunol. 204:27-48.
Demuth, D. R., Y. Duan, W. Brooks, A. R. Holmes, R. McNab, and H. F. Jenkinson. 1996. Tandem genes encode cell-surface polypeptides SspA and SspB which mediate adhesion of the oral bacterium Streptococcus gordonii to human and bacterial receptors. Mol. Microbiol. 20:403-413.
Demuth, D. R., E. E. Golug, and D. Malamud. 1990. Streptococcal-host interactions. Structural and functional analysis of a Streptococcus sanguis receptor for a human salivary glycoprotein. J. Biol. Chem. 265:7120-7126.
Dodd, H. M., N. Horn, W. C. Chan, C. J. Giffard., B. W. Bycroft, G. C. Roberts, and M. J. Gasson. 1996. Molecular analysis of the regulation of nisin immunity. Microbiology 142:2385-2392.
Dower, W. J., J. F. Miller, and C. W. Ragsdale. 1988. High efficiency transformation of Escherichia coli by high voltage electroporation. Nucleic Acids Res. 16:6127-6145.
Fischetti, V. A., V. Pancholi, and O. Schneewind. 1990. Conservation of a hexapeptide sequence in the anchor region surface proteins of Gram-positive cocci. Mol. Microbiol. 4:1603-1605.
Galli, D., F. Lottspeich, and R. Wirth. 1990. Sequence analysis of Enterococcus faecalis aggregation substance encoded by the sex-pheromone plasmid pAD1. Mol. Microbiol. 4:895-904.
Gasson, M. J., and F. L. Davies. 1980. High-frequency conjugation associated with Streptococcus lactis donor cell aggregation. J. Bacteriol. 143:1260-1264.
Gasson, M. J., S. Swindell, S. Maeda, and H. M. Dodd. 1992. Molecular rearrangement of lactose plasmid DNA associated with high-frequency transfer and cell aggregation in Lactococcus lactis 712. Mol. Microbiol. 6:3213-3223.
Gibson, T. J. 1984. Studies on the Epstein-Barr virus genome. Ph.D. thesis. Cambridge University, Cambridge, United Kingdom.
Godon, J. J., K. Jury, C. A. Shearman, and M. J. Gasson. 1994. The Lactococcus lactis sex-factor aggregation gene cluA. Mol. Microbiol. 12:655-663.
Hallet, B., D. J. Sherratt, and F. Hayes. 1997. Pentapeptide scanning mutagenesis: random insertion of a variable five amino acid cassette in a target protein. Nucleic Acids Res. 25:1866-1867.
Hayes, F., and B. Hallet. 2000. Pentapeptide scanning mutagenesis: encouraging old proteins to execute unusual tricks. Trends Microbiol. 8:571-577.
Holo, H., and I. F. Nes. 1995. Transformation of Lactococcus by electroporation. Methods Mol. Biol. 47:195-199.
Horinouchi, S., and B. Weisblum. 1984. Nucleotide sequence and functional map of pE194, a plasmid that specifies inducible resistance to macrolide, lincosamide, and streptogramin type B antibiotics. J. Bacteriol. 150:804-814.
Jenkinson, H. F., and D. R. Demuth. 1997. Structure, function and immunogenicity of streptococcal antigen I/II polypeptides. Mol. Microbiol. 23:183-190.
Kao, S. M., S. B. Olmsted, A. S. Vilksnins, J. C. Gallo, and G. M. Dunny. 1991. Molecular and genetic analysis of a region of plasmid pCF10 containing positive control genes encoding surface proteins involved in pheromone-inducible conjugation in Enterococcus faecalis. J. Bacteriol. 173:7650-7664.
Kelly, C., P. Evans, L. Bermeier, S. F. Lee, A. Progulske-Fox, A. C. Harris, A. Aitken, A. S. Bleiweis, and T. Lehner. 1989. Sequence analysis of the cloned streptococcal cell surface antigen I/II. FEBS Lett. 1:127-132.
Link, A. J., D. Phillips, and G. M. Church. 1997. Methods for generating precise deletions and insertions in the genome of wild-type Escherichia coli: application to open reading frame characterization. J. Bacteriol. 179:6228-6237.
Muscholl-Silberhorn, A. 1998. Analysis of the clumping-mediating domain(s) of sex pheromone plasmid pAD1-encoded aggregation substance. Eur. J. Biochem. 258:515-520.
Olmsted, S. B., S. M. Kao, L. J. van Putte, J. C. Gallo, and G. M. Dunny. 1991. Role of the pheromone inducible surface protein Asc10 in mating aggregate formation and conjugal transfer of the Enterococcus faecalis plasmid pCF10. J. Bacteriol. 173:7665-7672.
Sambrook, J., and D. W. Russel. 2001. Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
Stellwagen, A. E., and N. L. Craig. 1997. Gain-of-function mutations in TnsC, an ATP-dependent transposition protein that activates the bacterial transposon Tn7. Genetics 145:573-585.
Stentz, R., K. Jury, T. Eaton, M. Parker, A. Narbad, M. Gasson, and C. Shearman. 2004. Controlled expression of CluA in Lactococcus lactis and its role in conjugation. Microbiology 150:2503-2512.
Terzaghi, B. E., and W. E. Sandine. 1975. Improved medium for lactic streptococci. Curr. Microbiol. 7:245-250.
Troffer-Charlier, N., J. Ogier, D. Moras, and J. Cavarelli. 2002. Crystal structure of the V-region of Streptococcus mutans antigen I/II at 2.4 resolution suggests a sugar preformed binding site. J. Mol. Biol. 318:179-188.
Trotter, K. M., and G. M. Dunny. 1990. Mutants of Enterococcus faecalis deficient as recipients in mating with donors carrying pheromone-inducible plasmids. Plasmid 24:57-67.
van der Lelie, D., F. Chavarri, G. Venema, and M. J. Gasson. 1991. Identification of a new genetic determinant for cell aggregation associated with lactose plasmid transfer in Lactococcus lactis. Appl. Environ. Microbiol. 57:201-206.
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.
Ward, J. H. 1963. Hierarchical grouping to optimize an objective function. J. Am. Stat. Assoc. 58:236-244.
Waters, C. M., and G. M. Dunny. 2001. Analysis of functional domains of the Enterococcus faecalis pheromone-induced surface protein aggregation substance. J. Bacteriol. 183:5659-5667.
Waters, C. M., H. Hirt, J. K. McCormick, P. M. Schlievert, C. L. Wells, and G. M. Dunny. 2004. An amino-terminal domain of Enterococcus faecalis aggregation substance is required for aggregation, bacterial internalization by epithelial cells and binding to lipoteichoic acid. Mol. Microbiol. 52:1159-1171.
Woodbury, R. L., T. Luo, L. Grant, and W. G. Haldenwang. 2004. Mutational analysis of RbsT, an activator of the Bacillus subtilis stress response transcription factor B. J. Bacteriol. 186:2789-2797.
Xu, D., S. L. Lin, and R. Nussinov. 1997. Protein binding versus protein folding: the role of hydrophilic bridges in protein associations. J. Mol. Biol. 265:68-84.(Regis Stentz, Mike Gasson)
ABSTRACT
CluA is a cell surface-presented protein that causes cell aggregation and is essential for a high-efficiency conjugation process in Lactococcus lactis. We know from previous work that in addition to promoting cell-to-cell contact, CluA is involved in sex factor DNA transfer. To define the CluA domains involved in aggregation and in transfer, we first performed random mutagenesis of the cluA gene using a modified mini-Tn7 element which generated five amino acid insertions located throughout the encoded protein. Thirty independent cluA insertion mutants expressing modified CluA proteins at the cell surface were isolated and characterized further. The level of aggregation of each mutant was determined. The cell binding capacity of CluA was affected strongly when the protein had a mutation in its N-terminal region, which defined an aggregation domain extending from amino acid 153 to amino acid 483. Of the cluA mutants that still exhibited aggregation, eight showed an attenuated ability to conjugate, and six mutations were located in a 300-amino-acid C-terminal region of the protein defining a transfer domain (Tra). This result was confirmed by a phenotypic analysis of an additional five mutants obtained using site-directed mutagenesis in which charged amino acids of the Tra domain were replaced by alanine residues. Two distinct functional domains of the CluA protein were defined in this work; the first domain is involved in cell binding specificity, and the Tra domain is probably involved in the formation of the DNA transport machinery. This is the first report of a protein involved in conjugation that actively contributes to DNA transfer and mediates contact between donor and recipient strains.
INTRODUCTION
CluA is a 136-kDa surface-bound protein encoded by the chromosomally located sex factor of Lactococcus lactis MG1363 (16); it is associated with cell aggregation linked to high-frequency transfer of the sex factor (13, 14). The amino terminus of CluA has a signal sequence typical of secreted proteins, and the carboxy terminus has a cell wall binding motif that includes a conserved LPXTGE motif (11). CluA exhibits significant homology with two groups of cell surface proteins found in other gram-positive bacteria (16): two enterococcal proteins involved in plasmid conjugation, Asa1 (12) and Asc10 (22), and two streptococcal proteins involved in binding to salivary glycoproteins, SpaP (23) and Ssp5 (8). A comparison of the amino acid sequences of CluA with the sequences of these four other proteins led to identification of six distinct domains in CluA (16), including four conserved domains (domains I, II, IV, and V), a variable domain (domain III), and an additional domain (designated domain IVB in this study) that is specific to CluA and is located at the C-terminal end between domains IV and V. In previous work, we investigated the role of CluA in aggregation and conjugation by controlling the level of protein expression using the lactococcal nisA promoter (29). In a sex factor-negative MG1363 derivative, overexpression of CluA resulted in aggregation of the cells, indicating that CluA is the only sex factor component required for aggregation. We clearly showed that CluA enhances sex factor transfer when it is expressed in a donor strain having a cluA gene deletion on the chromosome. However, expression of CluA in the recipient strain could not complement the lack of CluA in a cluA donor strain. This led us to the conclusion that CluA, in addition to its capacity to trigger cell-to-cell contact, has a direct role in DNA transport. This was surprising since the Enterococcus faecalis pheromone-inducible aggregation substance Asc10 provides efficient conjugation when the protein is expressed in either the donor or recipient cells (26). The functional domains of Asc10 were characterized using the phage-based TnlacZ/in and TnphoA/in transposition systems to construct a library of 22 in-frame insertions encoding 31 amino acids located throughout Asc10 (36). The plasmid conjugation efficiencies of the different insertion mutants correlated with aggregation in the different mutants, as expected. In our further characterization of CluA we used the well-developed pentapeptide scanning mutagenesis (PSM) technique (17) to insert randomly a variable 5-amino-acid cassette throughout CluA in order to determine the functional domains of the protein involved in aggregation and also the domain(s) that plays a role in DNA transfer.
PSM is a transposon-based strategy that generates a small fingerprint (15 bp) in the gene of interest, leading to a random 5-amino-acid insertion into the encoded protein. PSM can be performed after in vivo transposition using the Tn4430 transposon originating from Bacillus thuringiensis that transposes efficiently in Escherichia coli (17, 18) or by using an in vitro transposition method based on the use of a Tn7 derivative (2). The Tn7-based technique uses the TnsAB transposase associated with the ATP-dependent DNA binding protein TnsCA225V (28) to insert in vitro a mini-Tn7 element containing the Tn7 terminal cis-acting transposition sequences flanking an antibiotic resistance gene at random locations in the target DNA (for a review see reference 6). The mini-Tn7 PmeI element is altered at both ends to generate PmeI restriction endonuclease sites (2). Removal of the mini-Tn7 element using the PmeI restriction endonuclease sites leads to a 15-bp insertion that includes a unique PmeI site. An in-frame stop codon is generated in two of the six reading frames due to the UAA stop codon included in the PmeI sequence. The generation of these truncated proteins is the major limitation of this system for use in analyses of surface proteins anchored to the cell wall. Out-of-frame insertions occurring in the N-terminal region of the LPXTGE motif lead to mutated proteins that lack the membrane-associated domain and the cell wall anchor motif and could cause missorting or release of the proteins into the culture medium. Another limitation of this system is the use of the mini-Tn7 PmeI selection markers, which are suitable only in gram-negative bacteria.
In this study, we developed a mutagenesis system suitable for gram-positive bacteria that produces only full-length proteins. This system was used successfully to generate a collection of derivatives with pentapeptide insertions located throughout CluA. These derivatives were employed for identification of regions of the protein associated with stability, aggregation, and conjugal transfer. Individual amino acid substitutions were created to further confirm the definition of the transfer domain.
MATERIALS AND METHODS
Media, growth conditions, and transformations. E. coli was grown at 37°C in Luria-Bertani medium (27) supplemented with 15 μg/ml of chloramphenicol or 150 μg/ml of erythromycin if necessary. L. lactis strains were grown at 30°C in M17 medium (30) with 0.5% glucose (GM17) or 0.5% sucrose (SM17) for selection of the nisin-sucrose transposon. Antibiotic resistance markers in L. lactis were selected using 5 μg/ml chloramphenicol or 5 μg/ml erythromycin. E. coli and L. lactis electrocompetent cells were prepared and transformed by the methods of Dower et al. (10) and Holo and Nes (19), respectively.
Bacterial strains. E. coli strain TG1 (15) was used for cloning experiments and plasmid propagation. L. lactis rifampin-resistant strain FI9979 (29) is a sex factor-negative strain that has a chromosomal copy of the complete nisin operon with a 20-bp insertion into nisA (9). This strain was used as a cluA background strain to monitor the aggregation of the different CluA mutants. Strains FI9980 and FI9981 (29) are FI9979 derivatives that harbor pFI2209, an expression vector containing the PnisA promoter, and pFI2213, the same expression vector containing the wild-type cluA gene, respectively. These two strains were used as negative and positive controls to assess aggregation. FI9980 was also used as a recipient strain in conjugation experiments. L. lactis strain FI9983 (29) is a cluA derivative of FI8164 (containing a complete sex factor with the tetM gene from Tn1545 [4] inserted into the tellurium resistance genes [C. J Pillidge, personal communication]). FI9983 also carries a chromosomal copy of the complete nisin operon with a 20-bp insertion into nisA. This strain was used in a complementation study to monitor the contributions of the different variants of CluA to conjugation. FI9984 (FI9983 containing the pFI2209 vector) and FI9985 (FI9983 carrying pFI2213) (29) were used as negative and positive controls, respectively, to assess conjugation efficiency.
Plasmid constructs. Phagemid pBluescript SK(–) (Stratagene) and plasmid pBR322 (3) were used for cloning experiments with E. coli TG1 (15). Plasmid pFI2453 was used as a target for insertion mutagenesis. This plasmid is a derivative of pFI2213 (expression vector pFI2209 containing the wild-type cluA gene under control of the PnisA promoter) (29) in which the unique BsrGI site was removed by blunt ending the 5' overhangs using the T4 DNA polymerase (Promega) under the conditions described by the manufacturer. All the relevant plasmids resulting from cluA insertional mutagenesis are listed in Table 1. Donor plasmid pFI2452 carrying the Eryr mini-Tn7 BsrGI element was constructed as follows. A 1,432-bp fragment of the pGPS4 plasmid (New England Biolabs) containing the Cmrr mini-Tn7 PmeI element was isolated after digestion with KpnI and SacI and cloned into the SacI/KpnI restriction endonuclease sites of pBSK–, generating pFI2449. The chloramphenicol resistance gene (903 bp) was then removed after BamHI digestion and religation of the vector, generating pFI2450. To amplify the Tn7 recombination sequences, PCR was performed using primers LmaB and RmaB (see below) and pFI2450 as the template. The 508-bp PCR product was then digested with EcoRI and AvaI and inserted into the EcoRI/AvaI restriction endonuclease sites of pBR322, yielding pFI2451. The erythromycin resistance gene (ermC) of plasmid pE194 (20) was excised from pUC7E (Claire Shearman, personal communication) after digestion with BamHI and inserted into the BamHI restriction endonuclease site of pFI2451, which yielded pFI2452.
DNA manipulation, PCR, and sequencing. General molecular cloning techniques and electrophoresis of DNA on agarose gels were carried out essentially as described by Sambrook et al. (27). The PCRs were performed using the BIO-X-ACT proofreading polymerase as described by the manufacturer (Bioline). Primers LmaB (GATGAATTCTTTAGTGTGTACACACAATAAAGTTGGGAACTG) having an EcoRI site at its 5' end and RmaB (GATCTCGGGCTAAATGTGTACACACAATAAAGTCTTAAACTG) having an AvaI site at its 5' end were used to replace the PmeI sites of pGSP4 by PCR-directed mutagenesis (BsrGI restriction endonuclease sites are underlined in both primers). Nucleotide sequencing was performed by the dideoxy chain termination method, using fluorescent dye terminators and cycle sequencing reactions. Samples were analyzed with an Applied Biosystems 373A DNA sequencer.
Site-directed mutagenesis of cluA. PCR mutagenesis was performed with the cluA gene by using the BIO-X-ACT proofreading polymerase (Bioline) to replace charged amino acids with alanine residues in CluA. Two sets of primers were used to amplify two cluA fragments from MG1363 chromosomal DNA. A primer upstream of the charged amino acid cluster was paired with a mutagenic oligonucleotide to hybridize at the region within cluA at which the mutation occurred. The mutagenic primer contained the sequence specifying the bases for the alanine substitutions. A second PCR was performed by using a mutagenic oligonucleotide that was complementary to the oligonucleotide used in the first amplification paired with a primer downstream from the desired substitution. The two PCR products were mixed and used as a template in crossover PCR (24), using the two external primers to generate a recombined PCR fragment containing the alanine substitutions. The resulting PCR fragment was then digested using restriction enzymes whose sites were incorporated into the external primers. The fragment was cloned into the cluA gene carried by the pFI956 plasmid, a pBluescript SK(–) vector containing the cluA gene (29), to replace the wild-type cluA internal fragment. The fragments were sequenced in order to verify that no misincorporation had occurred in the DNA sequence during PCR amplification. The cluA gene containing the point mutations was excised with SmaI and XhoI and cloned into the pFI2209 expression vector (29). The five resulting plasmids are listed in Table 2.
Transposition reactions. The in vitro transposition reaction was performed with pFI2452 using the conditions described for the GPS-LS linker-scanning system (New England Biolabs) by the manufacturer. One microliter of the transposition reaction was used to transform E. coli TG1. The pBR322-based pFI2452 donor plasmid was digested prior to the transposition reaction with both the PvuI and PvuII restriction enzymes, each of which cut once in the plasmid outside the mini-Tn7 BsrGI site to prevent replication in E. coli. A 0.02-μg portion of the pFI2452 digested donor DNA and 0.08 μg of the pFI2453 target DNA were used in the transposition reaction. After transformation of E. coli with the transposition reaction mixture DNA, transposition events were selected on plates containing erythromycin. A pool of plasmids extracted from 400 colonies was transformed into L. lactis FI9979, and the transformants were selected with erythromycin.
Nisin induction and quantification of the aggregation level. Appropriate dilutions of nisin (Aplin & Barret, Ltd.) were prepared for induction of the bacterial strains. One milliliter of an overnight cell culture grown in the presence of 50 ng/ml nisin was loaded into a 48-well microtitration plate to screen for L. lactis transformants containing mutated variants of pFI2213 that did not exhibit aggregation. The plate was then inverted several times until aggregation was visible. For a quantitative comparison of aggregation by the 30 different insertion mutants, a 1.5-ml overnight culture of the strains in SM17 containing 50 ng/ml of nisin was vortexed for 1 min to trigger aggregation and left at room temperature for 1 h, which allowed the aggregates to sediment. A 100-μl sample was then taken from the top (containing culture media with nonaggregated cells) and transferred to a 96-well microtitration plate. A 100-μl sample of the same culture that had not been vortexed was transferred to the same plate. The optical density at 600 nm (OD600) of each sample was determined simultaneously using a THERMOmax microplate reader (Molecular Devices). The optical density of each mutant in both conditions was determined three times independently. The level of aggregation was determined using the following formula: A% = 100 – (ODv/ODnv x 100), where A% is the percentage of aggregation, ODv is the OD600 of the culture that had been vortexed, and ODnv is the OD600 of the same culture that had not been vortexed.
Surface extraction and Western immunoblotting. A lysozyme surface extract of each strain was obtained as previously described (29). Briefly, the pellet from a 10-ml culture in SM17 was resuspended in a solution containing 50 mM Tris-HCl (pH 8), 0.1x protease inhibitor cocktail set II (Calbiochem), 10% sucrose, and 25 mg/ml lysozyme. The OD600 of the suspension was adjusted to 10, and the suspension was shaken gently at room temperature for 1 h and centrifuged at 3,500 rpm. Twenty microliters of the supernatant was loaded on a NuPage (Invitrogen) sodium dodecyl sulfate-polyacrylamide gel for electrophoresis. The amount of protein for each sample was checked by quantifying the signals using the TotalLab software (Nonlinear Dynamics) after staining of the gel with colloidal blue stain (Invitrogen). Western blot analyses were performed using a Western Breeze chromogenic kit by following the manufacturer's instructions (Invitrogen). Polyclonal antibodies raised against CluA (29) were used as primary antibodies. Quantification of the signals detected was performed using the TotalLab software.
Sex factor transfer. The FI9983 donor strain (tetracycline resistant) harboring plasmids carrying wild-type cluA, the different cluA insertion mutations, or the cluA point mutations or the strain carrying only the vector was grown overnight and diluted 100-fold in SM17 containing 5 μg/ml of chloramphenicol and 50 ng/ml of nisin. Recipient strain FI9980 (rifampin resistant) was grown under the same conditions but in the absence of nisin. Cell mating was triggered by centrifugation of a mixture of donor and recipient strains grown to the same optical density and not by vortexing as previously described (29). The conjugation efficiencies of the different strains were determined for donor and recipient cells grown to an OD600 of 0.8. Then 0.1 ml of donor cells and 0.9 ml of recipient cells were mixed, and the mixture was centrifuged at 5,000 rpm for 5 min. The pellet was resuspended in 1 ml of GM17 for all donor strains. Each mixture was incubated for 1 h at 30°C. Cells were then diluted and plated on GM17 agar containing rifampin (for recipient and transconjugant selection) or tetracycline (for donor and transconjugant selection) or both antibiotics (for transconjugant selection). The conjugation frequency for each donor was determined at least three times independently.
Statistical analysis. Statistical significance was calculated by determining confidence intervals for the differences of two population means when population variances were known. Cluster analysis was carried out to classify the strains on the basis of sex factor transfer level and aggregation. Hierarchical clustering (35) was used to determine the optimum number of clusters. The mean values for these clusters were used as seeds for a nonhierarchical cluster analysis (k-means) in order to optimize the classification of the strains. Statistical differences (P < 0.05) in the sex factor transfer level and/or level of aggregation between the three clusters were assessed by the Student-Newman-Keuls test for comparison of multiple means.
RESULTS
Construction of a CluA insertion library using an in vitro transposon mutagenesis system producing only full-length proteins. A transposon-based technique was used to define the regions of the CluA protein that are important for aggregation and the regions that may contribute to DNA transfer. To do this, we first engineered the mini-Tn7 PmeI element included in the GPS-LS linker-scanning system (New England Biolabs) to make it suitable for the study of CluA. This minitransposon is flanked by two PmeI restriction endonuclease sites that result in one-third of the proteins being truncated after mutagenesis. We therefore decided to replace this restriction endonuclease site with a BsrGI site using primers LmaB and RmaB (see Materials and Methods). Introduction of a BsrGI site involved changing the R4 and R7 residues of the terminal inverted repeats of the mini-Tn7 element, and these nucleotides could be altered without affecting the transposition activity (2). Furthermore, the resistance genes used to select the transposition events in the mini-Tn7 PmeI system are suitable for use only in gram-negative bacteria. Consequently, we used a gram-positive erythromycin resistance gene that was also functional in E. coli to replace the chloramphenicol resistance cassette flanked by the Tn7 terminal recombination sequences in mini-Tn7 PmeI. Construction of the Eryr mini-Tn7 BsrGI system carried by donor plasmid pFI2452 is described in detail in Materials and Methods. The new Eryr mini-Tn7 BsrGI system led, after excision of the mini-Tn7 element using BsrGI, to insertion of five amino acids in all six frames (Fig. 1), thus avoiding production of truncated proteins.
Independent 15-bp in-frame insertions located throughout cluA were obtained after use of the Eryr mini-Tn7 BsrGI element in a PSM experiment. For this, the new donor plasmid pFI2452 was used in an in vitro transposition reaction with the target plasmid pFI2453, a derivative of the pFI2213 plasmid carrying the wild-type cluA gene under control of the PnisA nisin promoter (29) in which the unique BsrGI site was removed. The transposition events were selected in E. coli using erythromycin, and a pool of plasmids extracted from 400 colonies was transformed into L. lactis FI9979, a suitable strain for nisin induction and CluA aggregation assessment. To select for transposon insertions into the cluA gene, the aggregation abilities of 400 independent clones were tested, and 152 transformants were unable to clump. The mini-Tn7 element was excised from a pool of the 152 corresponding plasmids using BsrGI, and the resulting plasmids were transformed back into FI9979. Of 130 nonaggregating transformants, 58 mutants carried a smaller plasmid with at least the 3' half of cluA missing, resulting from DNA rearrangement events. In previous work, it has been observed that the full-length cluA gene is unstable in E. coli (29). The plasmids of the other 72 transformants were analyzed by restriction mapping and nucleotide sequencing in the vicinity of each insertion. Mapping of each plasmid revealed that 10 of the insertions had occurred between the nisin promoter and the ATG start codon of cluA. Surprisingly, seven insertions were located downstream and close to the cluA terminator. Fifty-five insertions were located in the cluA gene coding sequence, and there were 31 independent insertions. The designations of these insertional mutations indicate the amino acid residue encoded by the codon that was disrupted by the insertion or that preceded this codon (Table 1). These insertions were equally distributed along the cluA open reading frame (Fig. 2A), showing clearly that the mini-Tn7 BsrGI element constructed in this work provided random insertions located throughout a target gene.
Consequences of insertions for the stability of CluA and determination of an N-terminal region responsible for cell aggregation. Previous to this study no structural data about CluA were available. We performed a Consensus Secondary Structure Prediction analysis of CluA using the Network Protein Sequence Analysis server (http://npsa-pbil.ibcp.fr) (5), and the results for the predicted alpha-helices are shown in Fig. 2B. In predicted domain I, the 14-amino-acid hydrophobic region (T4 to L17) is likely to be the region of the signal peptide that organizes into an alpha-helical conformation when it is in contact with the membrane lipid phase (34). Domain II consists mainly of alpha-helical structures, and a major structure is located between S75 and I154. Another important helical region is located in domain IV (T583 to Y600), and a final predicted alpha-helix is located downstream from the LPKTGE motif in domain V (A1207 to R1227), corresponding to the membrane-anchoring domain typical of cell wall-anchored proteins (11). Some pentapeptide insertions that occurred in predicted alpha-helical domains of CluA and led to protein instability are described below.
The presence of CluA on the cell surface of the 31 mutants was assessed by immunodetection of CluA following cell surface extraction, as shown in Fig. 3A. The controls were the negative control FI9980, carrying the pFI2209 empty vector, and FI9981, carrying pFI2213 that contained the wild-type cluA gene expressed under control of the PnisA nisin promoter. As expected, the cell surface extract of FI9980 contained no reactive CluA protein, whereas wild-type CluA expressed in FI9981 was detected. For most of the insertion mutants, CluA was detected on the Western blot (Fig. 3A). The mutant carrying the P485 mutation contained no CluA and was not used in further experiments. A 136-kDa signal corresponding to the size of CluA was observed for all mutants except the strains carrying the Y119 and T125 mutations. For most of the mutants the total amount of detected signals comprising full-size CluA and its degradation products was similar to the amount of full-size CluA detected in FI9981 (data not shown). Since the extraction protocol released C-terminal cell wall-anchored CluA variants, it is likely that the 30 mutants expressed a full-size protein on the cell surface at a level comparable to that of wild-type CluA in FI9981. The effects of the 30 different insertions on CluA aggregation were quantified (see Materials and Methods), and the results are shown in Fig. 3B.
Aggregation was completely prevented for Y119 and T125 insertions, and this result agrees with the strong protein degradation observed after cell surface protein extraction (Fig. 3A). No signal above 70 kDa was detected for Y119 and T125, and a smear of strong bands migrating between 50 and 70 kDa corresponding to degraded CluA products was observed. This indicates that insertions located at positions Y119 and T125 affected the stability of CluA either in vivo or during the extraction process. According to the secondary structure prediction, these two insertions are located within the major predicted alpha-helical structure of predicted domain II (Fig. 2B), suggesting that this structure is important for protein stability.
For the different CluA mutants in which the 136-kDa product was observed, the insertions affecting aggregation were located mainly in the N-terminal half of the protein in a region bordered by the amino acids D153 and I483. Although the D153, Q360, H361, and I483 mutants showed little or no protein degradation after surface protein extraction (Fig. 3A), they were unable to aggregate (Fig. 3B). Likewise, no protein degradation was observed for the S400 and W410 mutants, whereas their aggregation abilities were significantly reduced (P < 0.01). This indicates that the insertions in these six mutants occurred in regions that are important for aggregation and likely to be involved in the control of CluA binding specificity. Although the M281 mutant showed low aggregation ability, the faint signals observed for the 136-kDa product and the degradation products suggest that an insertion at this location led to lower expression of the protein on the cell surface and/or decreased its stability.
The low levels of aggregation observed for the four mutants carrying the insertions located at positions L519 to E605 can be explained by the detection of a very weak signal at 136 kDa, whereas strong degradation products were observed around 60 kDa. The three insertions other than L519 are located within (T583 and N583) or close to (E605) the T583-Y600 predicted alpha-helical structure, suggesting that this region is also important for CluA stability. Although the mutants with mutations located in the D838-to-A1157 region exhibited reduced aggregation, all the corresponding proteins exhibited some degradation. In particular, the major product observed for the T858, T1083, and T1141 insertions was a 120-kDa CluA degradation product (Fig. 3A). This suggests that part of the C-terminal region is involved in aggregation and/or that the stability of the protein is affected by these insertions. In summary, the analysis of the insertions revealed a CluA domain that is responsible for cell-to-cell binding, including amino acids D153 to I483.
Definition of a CluA domain involved in transfer of the sex factor. In L. lactis, expression of CluA in the recipient strain was unable to complement DNA transfer of the sex factor from a donor strain lacking a cluA functional gene, whereas complementation took place when CluA was expressed in a cluA donor strain (29). This observation led to the conclusion that in addition to promoting cell-to-cell contact, CluA is involved in the sex factor DNA transfer machinery and that an association of CluA with one component or several components is necessary to enable DNA transfer.
It was anticipated that the region of the CluA protein associated with DNA transfer was different from the region involved directly in aggregation. The conjugation efficiencies of the mutants that were still able to aggregate were assessed to discriminate between potential insertions that affect this functional domain. The 17 plasmids carrying CluA mutations that were still able to promote aggregation (aggregation, >35%) were transformed into the FI9983 donor background strain containing a chromosomal cluA deletion. The sex factor transfer efficiencies were determined for the strain containing the empty expression vector (FI9984), the expression vector harboring the wild-type cluA gene (FI9985), and the 17 mutated plasmids. Since enhanced DNA transfer seems to be dependent on aggregation ability (12, 13, 29, 33) (see above) (i.e., on the ability of the donor strain to promote cell-to-cell contact), transfer efficiency data are presented in relation to aggregation efficiency in Fig. 4. We performed a cluster analysis using Ward's method (35) and identified three clusters of insertion mutants (Fig. 4). The first cluster (cluster 1) comprised the insertion mutants in which there was little effect on aggregation or sex factor transfer and which acted like the FI9985 CluA control. Mutants in this cluster exhibited the highest level of aggregation (76%) and the highest sex factor transfer value (80.4 x 10–4 transconjugant/donor cell on average), and both values were significantly different from the values for clusters 2 and 3. The second cluster (cluster 2) contained the CluA variants that exhibited a lower level of aggregation, which led to a lower sex factor transfer value. This was the result predicted if sex factor transfer relies solely on the level of aggregation. The third cluster (cluster 3) contained the insertion mutants that had intermediate levels of aggregation but sex factor transfer values that were lower than those that were predicted. There were no significant differences in the aggregation values between clusters 2 and 3 (ca. 48 to 49%), whereas the strains belonging to cluster 3 exhibited the lowest sex factor transfer efficiency, 14 x 10–4 transconjugant/donor cell and this value was significantly different from the transfer efficiency detected in cluster 2 (60.2 x 10–4 transconjugant/donor cell). The low transfer frequency observed in cluster 3 strains shows that these insertion mutants were affected in regions of CluA that play an important role in CluA's DNA transfer capacity. In this cluster, six of eight mutations affecting CluA were located in a region covering the 40 amino acids of the C-terminal end of predicted domain IV and in domain IVB (Fig. 2) specific for CluA. The M266 and W410 mutants had mutations that were located in the predicted variable domain III, and these two areas of CluA are also likely to be involved in DNA transfer.
Site-directed mutagenesis of the CluA Tra domain. To investigate further and refine the region of CluA involved in DNA transfer, we introduced point mutations in this part of the protein. In their study of RbsT's activity, a protein involved in stress response in Bacillus subtilis (38), Woodbury and coworkers replaced clusters of charged amino acids located throughout the length of the protein with alanine residues. Charged amino acids were chosen because they are residues that are likely to be involved in protein-ligand interactions (39) with a high probability that they are surface exposed and, consequently, less likely to affect the protein's general stability (38).
An analysis of charged amino acids groups in the entire CluA protein showed that these residues are the residues that are most frequently found in the transfer domain. Five clusters of charged amino acid residues located along the length of the CluA transfer region were found (Fig. 5). The charged amino acids at the sites were replaced by the uncharged amino acid alanine by using a PCR-based mutagenesis technique (see Materials and Methods for details). The five plasmids containing the five different sets of mutations are listed in Table 2. The alanine clusters in the transfer region of the five CluA mutants were designated Tra1 to Tra5 according to the orientation from the N terminus to the C terminus. The plasmids were used to transform FI9979, and the aggregation abilities of the resulting strains were assessed. All the strains exhibited levels of aggregation similar to that of the strain carrying the wild-type cluA plasmid (data not shown). The conjugation efficiency of each cluA variant expressed in the FI9983 donor strain was determined (Table 2), and in parallel, Western blotting was performed with cell surface extracts of the different donors (data not shown). The signals detected showed that the protein abundance and apparent size were similar to those of the wild type. Compared to the donor strain expressing wild-type CluA, the donor strains containing the Tra alleles were all affected in the DNA transfer capacity. The strain containing the mutant allele Tra2 had a 13-fold-lower DNA efficiency than the wild type, strains containing Tra3 and Tra5 had a 6-fold-lower DNA efficiency, and strains containing Tra1 and Tra4 had a 3-fold-lower transfer efficiency. The results for noninduced cells are also shown in Table 2, and the conjugation frequencies were similar to the conjugation frequency obtained for the same donor strain lacking cluA. Overall, the results showed clearly that charged amino acids belonging to the newly defined CluA transfer (Tra) domain bordered by amino acids K784 and K1056 are involved in the ability of CluA to transfer DNA, in addition to triggering cell-to-cell contact during conjugation.
DISCUSSION
The functional analysis of 30 cluA insertion mutants generated in this study identified two regions of the CluA protein with distinct functions: the D153-I483 region of CluA that mediates bacterial aggregation and the K784-K1056 Tra domain that is involved in DNA transfer and is responsible for a high conjugation frequency.
To define the regions of CluA important in promoting cell-to-cell binding, we determined and compared the levels of aggregation of the 30 cluA insertion mutants. Six of the CluA variants showed complete or partial inhibition of aggregation compared to the wild-type CluA strain (Fig. 3). These mutants defined a region extending from amino acid D153 to amino acid I483. This D153-I483 region includes predicted variable domain III of CluA, and this is in agreement with the two aggregation domains identified in Asc10 of E. faecalis, the K156-A358 region (37) that covers the N-terminal half of the Asc10 variable domain and the Q473-L683 region (36) that includes its C-terminal third. Furthermore, a 92-amino-acid region in the variable domain of the E. faecalis Asa1 aggregation substance has also been shown to play a dominant role in cell aggregation (25).
The CluA D153-I483 aggregation domain might extend farther into predicted domain IV as downstream from I483, four insertions (L519, T583, N593, and E605) were also affected in aggregation. However, the variant CluA proteins exhibited a degraded protein pattern, and it is not clear whether this occurred during extraction or in vivo, leaving the question of whether these insertions have a direct effect on aggregation open.
Five insertions in the C-terminal part of CluA (D838 to A1157) partially impaired CluA's aggregation capacity, and this region is likely to play a role in aggregation. In a similar way, the C-terminal domain of Asc10 has been shown to play an essential role in aggregation, and specific regions that bring the N-terminal functional domain into the correct conformation to mediate aggregation have been identified (36).
It is not known how the CluA protein interacts with the lactococcal cell surface to cause aggregation. In E. faecalis aggregation is mediated by a carbohydrate recognition process in which the lipoteichoic acid, a surface molecule that contains sugar residues, is the potential bacterial receptor (32, 37). Interestingly, the L238-V486 region of CluA exhibits significant (E-value, 2 x 10–4) homology (24% sequence identity) with the L554-I750 region of the central domain (21) of the SspB protein (7), a member of the streptococcal antigen I/II family of protein adhesins that recognize salivary agglutinin glycoprotein. The structure of this domain in Streptococcus mutans antigen I/II has been determined, and the results revealed its potential to bind sugars (31). Overall, these results suggest that CluA might also recognize and bind specific sugar motifs presented on the surface of some L. lactis strains (33), including MG1363.
Aggregating CluA mutants were tested for the ability to transfer the sex factor. All insertions located in the Tra domain, unique to CluA compared to other gram-positive cell surface protein homologues (16), were affected in sex factor transfer. The importance of this region was confirmed by replacement of charged amino acids located along the length of the Tra domain, which showed that these amino acids are essential for optimized DNA transfer. It is known that electrostatic interactions provided by charged amino acids at the binding interface play an important role in protein-ligand interactions (39). Sex factor transfer could be the result of a protein-ligand association involving the Tra domain of CluA and another component of the conjugation machinery, the nature of which still needs to be determined. A sequence database search with BLAST (1) using the Tra domain as a probe gave no significant homologies with other proteins.
Additional areas of CluA are essential for sex factor transfer. A decreased level of sex factor transfer was observed with the M266 and W410 insertions. These two positions in the CluA N terminus might play a role in DNA transfer by interacting with a component of the DNA transfer machinery or by directly or indirectly interacting with the CluA Tra domain to optimize DNA transfer.
Surprisingly, in mutants exhibiting similar levels of aggregation (Fig. 3), the sex factor transfer can vary, as observed for the S400 mutant, whose sex factor transfer was four times higher than that of the W410 mutant. The difference occurred despite the closeness of the two insertions, as S400 and W410 are only 9 amino acids apart. It is likely that the amino acid sequences SCVHR inserted into S400 and VVYTT inserted into W410 (Table 1) disrupted the protein differently, thus leading to noticeable differences in the capacity of the mutated CluA to transfer DNA.
This is the first report of a protein involved in conjugation that actively contributes to DNA transfer and also mediates contact between donor and recipient strains. To achieve this, we developed a mutagenesis tool adapted to the structural and functional analysis of CluA. The newly developed miniTn7 BsrGI transposon has potential for use in diverse protein analyses in which full-length mutated proteins are required. Furthermore, the use of the erythromycin resistance gene to select miniTn7 insertions makes it an appropriate tool for protein studies in diverse species of gram-positive bacteria.
ACKNOWLEDGMENTS
We thank Carmen Pin and Yvan Lemarc for their help with statistical analysis and John Payne for help with DNA sequencing.
R.S. was funded by BBSRC contract 99123941.
REFERENCES
Altschul, S. F., T. L. Madden, A. A. Schffer, J. Zhang, Z. Zhang, W. Miller, and D. J. Lipman. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25:3389-3402.
Biery, M. C., F. J. Stewart, A. E. Stellwagen, E. A. Raleigh, and N. L. Craig. 2000. A simple in vitro Tn7-based transposition system with low target site selectivity for genome and gene analysis. Nucleic Acids Res. 28:1067-1077.
Bolivar, F., R. L. Rodriguez, P. J. Greene, M. C. Betlach, H. L. Heyneker, and H. W. Boyer. 1977. Construction and characterization of new cloning vehicles. II. A multipurpose cloning system. Gene 2:93-113.
Caillot, F., C. Carlier, and P. Courvalin. 1987. Physical analysis of the conjugative shuttle transposon Tn1545. Plasmid 17:58-60.
Combet, C., C. Blanchet, C. Geourjon, and G. Deleage. 2000. NPS@: Network Protein Sequence Analysis. Trends Biochem. Sci. 25:147-150.
Craig, N. L. 1996. Transposon Tn7. Curr. Top. Microbiol. Immunol. 204:27-48.
Demuth, D. R., Y. Duan, W. Brooks, A. R. Holmes, R. McNab, and H. F. Jenkinson. 1996. Tandem genes encode cell-surface polypeptides SspA and SspB which mediate adhesion of the oral bacterium Streptococcus gordonii to human and bacterial receptors. Mol. Microbiol. 20:403-413.
Demuth, D. R., E. E. Golug, and D. Malamud. 1990. Streptococcal-host interactions. Structural and functional analysis of a Streptococcus sanguis receptor for a human salivary glycoprotein. J. Biol. Chem. 265:7120-7126.
Dodd, H. M., N. Horn, W. C. Chan, C. J. Giffard., B. W. Bycroft, G. C. Roberts, and M. J. Gasson. 1996. Molecular analysis of the regulation of nisin immunity. Microbiology 142:2385-2392.
Dower, W. J., J. F. Miller, and C. W. Ragsdale. 1988. High efficiency transformation of Escherichia coli by high voltage electroporation. Nucleic Acids Res. 16:6127-6145.
Fischetti, V. A., V. Pancholi, and O. Schneewind. 1990. Conservation of a hexapeptide sequence in the anchor region surface proteins of Gram-positive cocci. Mol. Microbiol. 4:1603-1605.
Galli, D., F. Lottspeich, and R. Wirth. 1990. Sequence analysis of Enterococcus faecalis aggregation substance encoded by the sex-pheromone plasmid pAD1. Mol. Microbiol. 4:895-904.
Gasson, M. J., and F. L. Davies. 1980. High-frequency conjugation associated with Streptococcus lactis donor cell aggregation. J. Bacteriol. 143:1260-1264.
Gasson, M. J., S. Swindell, S. Maeda, and H. M. Dodd. 1992. Molecular rearrangement of lactose plasmid DNA associated with high-frequency transfer and cell aggregation in Lactococcus lactis 712. Mol. Microbiol. 6:3213-3223.
Gibson, T. J. 1984. Studies on the Epstein-Barr virus genome. Ph.D. thesis. Cambridge University, Cambridge, United Kingdom.
Godon, J. J., K. Jury, C. A. Shearman, and M. J. Gasson. 1994. The Lactococcus lactis sex-factor aggregation gene cluA. Mol. Microbiol. 12:655-663.
Hallet, B., D. J. Sherratt, and F. Hayes. 1997. Pentapeptide scanning mutagenesis: random insertion of a variable five amino acid cassette in a target protein. Nucleic Acids Res. 25:1866-1867.
Hayes, F., and B. Hallet. 2000. Pentapeptide scanning mutagenesis: encouraging old proteins to execute unusual tricks. Trends Microbiol. 8:571-577.
Holo, H., and I. F. Nes. 1995. Transformation of Lactococcus by electroporation. Methods Mol. Biol. 47:195-199.
Horinouchi, S., and B. Weisblum. 1984. Nucleotide sequence and functional map of pE194, a plasmid that specifies inducible resistance to macrolide, lincosamide, and streptogramin type B antibiotics. J. Bacteriol. 150:804-814.
Jenkinson, H. F., and D. R. Demuth. 1997. Structure, function and immunogenicity of streptococcal antigen I/II polypeptides. Mol. Microbiol. 23:183-190.
Kao, S. M., S. B. Olmsted, A. S. Vilksnins, J. C. Gallo, and G. M. Dunny. 1991. Molecular and genetic analysis of a region of plasmid pCF10 containing positive control genes encoding surface proteins involved in pheromone-inducible conjugation in Enterococcus faecalis. J. Bacteriol. 173:7650-7664.
Kelly, C., P. Evans, L. Bermeier, S. F. Lee, A. Progulske-Fox, A. C. Harris, A. Aitken, A. S. Bleiweis, and T. Lehner. 1989. Sequence analysis of the cloned streptococcal cell surface antigen I/II. FEBS Lett. 1:127-132.
Link, A. J., D. Phillips, and G. M. Church. 1997. Methods for generating precise deletions and insertions in the genome of wild-type Escherichia coli: application to open reading frame characterization. J. Bacteriol. 179:6228-6237.
Muscholl-Silberhorn, A. 1998. Analysis of the clumping-mediating domain(s) of sex pheromone plasmid pAD1-encoded aggregation substance. Eur. J. Biochem. 258:515-520.
Olmsted, S. B., S. M. Kao, L. J. van Putte, J. C. Gallo, and G. M. Dunny. 1991. Role of the pheromone inducible surface protein Asc10 in mating aggregate formation and conjugal transfer of the Enterococcus faecalis plasmid pCF10. J. Bacteriol. 173:7665-7672.
Sambrook, J., and D. W. Russel. 2001. Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
Stellwagen, A. E., and N. L. Craig. 1997. Gain-of-function mutations in TnsC, an ATP-dependent transposition protein that activates the bacterial transposon Tn7. Genetics 145:573-585.
Stentz, R., K. Jury, T. Eaton, M. Parker, A. Narbad, M. Gasson, and C. Shearman. 2004. Controlled expression of CluA in Lactococcus lactis and its role in conjugation. Microbiology 150:2503-2512.
Terzaghi, B. E., and W. E. Sandine. 1975. Improved medium for lactic streptococci. Curr. Microbiol. 7:245-250.
Troffer-Charlier, N., J. Ogier, D. Moras, and J. Cavarelli. 2002. Crystal structure of the V-region of Streptococcus mutans antigen I/II at 2.4 resolution suggests a sugar preformed binding site. J. Mol. Biol. 318:179-188.
Trotter, K. M., and G. M. Dunny. 1990. Mutants of Enterococcus faecalis deficient as recipients in mating with donors carrying pheromone-inducible plasmids. Plasmid 24:57-67.
van der Lelie, D., F. Chavarri, G. Venema, and M. J. Gasson. 1991. Identification of a new genetic determinant for cell aggregation associated with lactose plasmid transfer in Lactococcus lactis. Appl. Environ. Microbiol. 57:201-206.
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.
Ward, J. H. 1963. Hierarchical grouping to optimize an objective function. J. Am. Stat. Assoc. 58:236-244.
Waters, C. M., and G. M. Dunny. 2001. Analysis of functional domains of the Enterococcus faecalis pheromone-induced surface protein aggregation substance. J. Bacteriol. 183:5659-5667.
Waters, C. M., H. Hirt, J. K. McCormick, P. M. Schlievert, C. L. Wells, and G. M. Dunny. 2004. An amino-terminal domain of Enterococcus faecalis aggregation substance is required for aggregation, bacterial internalization by epithelial cells and binding to lipoteichoic acid. Mol. Microbiol. 52:1159-1171.
Woodbury, R. L., T. Luo, L. Grant, and W. G. Haldenwang. 2004. Mutational analysis of RbsT, an activator of the Bacillus subtilis stress response transcription factor B. J. Bacteriol. 186:2789-2797.
Xu, D., S. L. Lin, and R. Nussinov. 1997. Protein binding versus protein folding: the role of hydrophilic bridges in protein associations. J. Mol. Biol. 265:68-84.(Regis Stentz, Mike Gasson)