The Ruminococcus albus pilA1pilA2 locus: expression and putative role of two adjacent pil genes in pilus formation and bacterial adhesion to cellulose

Harivony Rakotoarivonina1, Marilynn A. Larson2, Mark Morrison3, Jean-Pierre Girardeau1, Brigitte Gaillard-Martinie1, Evelyne Forano1 and Pascale Mosoni1

1 Unité de Microbiologie, INRA, Centre de Recherches de Clermont-Ferrand-Theix, 63122 Saint-Genès-Champanelle, France
2 Department of Pathology and Microbiology, University of Nebraska Medical Center, Omaha, NE 68198, USA
3 The MAPLE Research Program, Department of Animal Sciences, The Ohio State University, 2027 Coffey Road, Columbus, OH 43210, USA

Correspondence
Pascale Mosoni
pmosoni{at}clermont.inra.fr


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Ruminococcus albus produces fimbria-like structures that are involved with the bacterium's adhesion to cellulose. The subunit protein has been identified in strain 8 (CbpC) and strain 20 (GP25) and both are type IV fimbrial (Pil) proteins. The presence of a pil locus that is organized similarly in both strains is reported here together with the results of an initial examination of a second Pil protein. Downstream of the cbpC/gp25 gene (hereafter referred to as pilA1) is a second pilin gene (pilA2). Northern blot analysis of pilA1 and pilA2 transcripts showed that the pilA1 transcript is much more abundant in R. albus 8, and real-time PCR was used to measure pilA1 and pilA2 transcript abundance in R. albus 20 and its adhesion-defective mutant D5. Similar to the findings with R. albus 8, the relative expression of pilA1 in the wild-type strain was 73-fold higher than that of pilA2 following growth with cellobiose, and there were only slight differences between the wild-type and mutant strain in pilA1 and pilA2 transcript abundances, indicating that neither pilA1 nor pilA2 transcription is adversely affected in the mutant strain. Western immunoblots showed that the PilA2 protein is localized primarily to the membrane fraction, and the anti-PilA2 antiserum does not inhibit bacterial adhesion to cellulose. These results suggest that the PilA2 protein plays a role in the synthesis and assembly of type IV fimbriae-like structures by R. albus, but its role is restricted to cell-associated functions, rather than as part of the externalized fimbrial structure.


Abbreviations: CBP, cellulose-binding protein

The GenBank/EMBL/DDBJ accession numbers for the pilA2 gene sequences of R. albus strains 8 and 20 are respectively AY880680 and AJ853471.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The initial step in the degradation of cellulosic biomass by many anaerobic micro-organisms requires the adhesion of the organism to this substrate. Adhesion facilitates maximal enzyme–substrate interaction and gives the adherent cells an advantage for substrate uptake, and can be a rate-limiting step in polysaccharide hydrolysis. In the gastrointestinal tracts of herbivores and ruminants, Ruminococcus albus, Ruminococcus flavefaciens and Fibrobacter succinogenes are widely believed to coordinate degradation of the cellulosic component of plant biomass. Our understanding of the adhesion mechanism(s) employed by these three bacteria has been limited by the lack of genetic tools available to dissect the process(es), but recent studies have identified a variety of structures including novel cellulose-binding modules present in cellulosomes (Ding et al., 2001; Rincon et al., 2001, 2003), cell-surface-associated cellulases (Devillard et al., 2004; Mitsumori & Minato, 2000; Malburg et al., 1997), cell-surface exopolysaccharides (Mosoni & Gaillard-Martinie, 2001) and other uncharacterized cellulose-binding proteins (CBPs) (Gong et al., 1996; Miron & Forsberg, 1999). These different adhesion mechanisms are described in recent reviews (Morrison & Miron, 2000; Miron et al., 2001).

Recent work in our laboratory has established that type IV fimbria-like structures are involved in the adhesion of R. albus 20 to cellulose (Rakotoarivonina et al., 2002). The fimbrial structural protein was first identified as a CBP called CbpC in R. albus 8 (Pegden et al., 1998). By comparing R. albus strain 20 with its spontaneous adhesion-defective mutant (strain D5), we recently showed the presence of pili on the surface of R. albus 20 and identified a glycoprotein, initially called GP25, as the major pilus subunit (Rakotoarivonina et al., 2002). Pil protein homologues have also been shown by Southern and Western blot analyses to be encoded in the genomes of other strains of R. albus (Pegden et al., 1998).

In an effort to understand these adhesion structures better, we looked for the presence of genes flanking cbpC and gp25 likely to encode functions involved with the secretion and assembly of pili in R. albus. We report here the presence of a second pil gene present in both R. albus strains 20 and 8 that is located adjacent to cbpC/gp25. We examined the expression of these tandemly arranged pil genes by quantitative RT-PCR in strain 20 and mutant D5, and by Northern blot analysis in R. albus strain 8. Antibodies specific to the second pil-gene product in R. albus strain 20 were prepared and used to show that this second gene product is not part of the externalized fimbrial structure but resides mainly in the membrane of the bacterium.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains and culture media.
R. albus 8 was kindly provided by M. A. Cotta (USDA-NCAUR, Peoria, IL, USA) and was routinely cultured in EM-cellobiose medium (Reveneau et al., 2003) alone or supplemented with 5 % (v/v) clarified rumen fluid or 25 µM of both phenylacetic acid (PAA) and phenylpropionic acid (PPA) (Sigma). R. albus 20 (=ATCC 27211) and its adhesion-defective mutant R. albus D5 were cultivated in a rumen fluid medium with 0·3 % (w/v) cellobiose (RF cellobiose medium; Mosoni & Gaillard-Martinie, 2001). Growth was monitored spectrophotometrically by the increase in culture OD600 in 18x150 mm Balch tubes (Bellco Glass). Escherichia coli BL21(DE3) (Stratagene) was routinely grown on Luria–Bertani (LB) medium with appropriate antibiotic selection at 37 °C; conditions used for expression of cloned products in the vector pET28a (Novagen) are described below.

Cloning and nucleotide sequence analysis of the pilA1pilA2 locus in R. albus strains 8 and 20.
Genomic DNA isolation and the cloning and sequencing of the gp25 and cbpC genes has been described previously (Rakotoarivonina et al., 2002; Pegden et al., 1998). For both strains, genome-walking procedures were used to obtain nucleotide sequence information downstream of the genes encoding GP25 and CbpC, in R. albus strains 20 and 8, respectively (Universal Genome walker kit; Clontech). Oligonucleotide primers (Table 1) were synthesized by either MWG Biotech or Sigma-Genosys, and the primary and secondary gp25/cbpC gene-specific primers were designed according to the manufacturer's recommendations. For R. albus 20, a PCR product encoding 1·5 kb of sequence downstream of the gp25 gene was amplified using GW1-F and the adaptor primer AP2 and PvuII-digested genomic DNA. The PCR product was purified from agarose gel slices using the Strataprep PCR purification kit (Stratagene) and sequenced by primer-walking methodology with an ABI Prism BigDye Terminator cycle sequencing ready reaction kit (Perkin Elmer).


View this table:
[in this window]
[in a new window]
 
Table 1. Oligonucleotide primers used in this study

The following abbreviations are used in primer names: GW, genome walking; F, forward; R, reverse; LC, LightCycler. Residues shown in italics were added to the 5' ends of primers to generate restriction enzyme cleavage sites (underlined).

 
For R. albus 8, genomic walking procedures produced a 4·9 kb PCR product that was initially end-sequenced. Next, primers 3F2 and 3R3 were used with R. albus 8 genomic DNA as the template. The PCR product was sequenced by the DNA sequencing facility at Iowa State University using primer-walking methodology.

The DNA sequences obtained from strains 20 and 8 were analysed with the BLASTX program (Altschul et al., 1990) and the Wisconsin Genetics Computer Group (GCG) package version 9.1 to determine sequence similarities to those available in the GenBank and EMBL databases.

Isolation of the pilA2 gene from the adhesion-defective mutant D5.
To amplify a DNA fragment encoding pilA2 and its upstream region (i.e. putative promoter) in R. albus D5, a PCR was carried out with genomic DNA using primers PilA2-5'-F and PilA2-3'-R (Table 1) and 1 U AmpliTaq DNA polymerase (Appligene). The PCR parameters were an initial denaturation step (94 °C for 5 min), 25 cycles of amplification (94 °C for 90 s, 55 °C for 60 s and 72 °C for 90 s) and a final elongation step (72 °C for 7 min). The PCR product was sequenced in both directions with the same primers as those used for amplification.

RNA extraction and Northern blot analyses.
Cultures of R. albus 8 were harvested during the mid-exponential phase of growth (OD600 ~0·5) by anaerobic centrifugation (10 000 g, 10 min at 4 °C). The pelleted cells were washed twice with 10 ml volumes of 25 mM sodium phosphate buffer (pH 7·0) containing 25 mM EDTA. The cells were then suspended in 10 ml of the same buffer and cell lysis was accomplished by the addition of mutanolysin (200 U ml–1), proteinase K (150 µg ml–1) and 0·5 % (w/v) SDS for 5 min at 55 °C. The RNA was recovered from the lysis mixture by using the RNeasy purification system from Qiagen according to the manufacturer's recommendations. Total RNA was DNase-treated according to standard procedures (Sambrook et al., 1989) and RNA integrity was checked by electrophoresis in a 1 % (w/v) agarose gel and ethidium bromide staining.

For Northern blot analyses, 10 µg total RNA for each growth condition was fractionated in 1 % (w/v) agarose gels in the presence of glyoxal and DMSO according to standard procedures (Sambrook et al., 1989). The RNA was transferred to a Zeta-Probe GT membrane (Bio-Rad Laboratories) by vacuum blotting and immobilized by UV cross-linking. The pilA2-specific probe was radiolabelled with [{alpha}-32P]dCTP by random primer labelling (Pharmacia Biotech) according to the manufacturer's recommendations and blot hybridizations were performed overnight at 43 °C. After hybridization, the membranes were washed twice at 43 °C for 30 min in 40 mM disodium phosphate buffer (pH 7·0) containing 1 mM EDTA and 5 % (w/v) SDS and then washed twice at 43 °C for 30 min in 40 mM disodium phosphate buffer (pH 7·0) containing 1 mM EDTA and 1 % (w/v) SDS. Differences in total RNA loading were determined by stripping the blots and hybridizing with a 32P-labelled oligonucleotide that is complementary to virtually all known 16S rRNA sequences (5'-TACCGCGGCTGCTGGCAC-3'; Zheng et al., 1996). A PhosphorImager (Molecular Dynamics) was used to quantify differences in transcript abundance. Transcript size was estimated by using a standard curve produced from an RNA standard ladder (Bio-Rad) that was loaded on the same gels used for Northern analysis.

Isolation of a pilA1pilA2 cotranscript.
Amplification of a pilA1pilA2 cotranscript was carried out using cDNA obtained from R. albus 20 as template and a forward primer located in the 5' region of pilA1 (PilA1-F) and a reverse primer located in the 3' region of pilA2 (PilA2-R) (Table 1). Positive controls were also performed using either forward and reverse pilA1 primers or forward and reverse pilA2 primers. A negative control was included, using RNA not reverse-transcribed (see below) as template. PCRs were carried out with 1 U AmpliTaq DNA polymerase and 100 ng cDNA. The PCR parameters were an initial denaturation step (94 °C for 5 min), 30 cycles of amplification (94 °C for 90 s, 60 °C for 60 s and 72 °C for 90 s) and a final elongation step (72 °C for 7 min).

Analysis of gene expression using real-time RT-PCR.
Total RNA was isolated from R. albus 20 and D5 cultures grown in RF cellobiose medium. Cells (10 ml culture) were harvested in the stationary phase of growth by centrifugation and washed three times in 1 ml DEPC-treated TE buffer (10 mM Tris/HCl, 1 mM EDTA, pH 8). Initial cell lysis was accomplished in 200 µl of the same buffer containing lysozyme (3 mg ml–1) and proteinase K (100 µg ml–1) for 1 h at 37 °C. Cells were then disrupted by bead beating for 1 min with 1 g zirconia beads. Subsequent RNA extraction was performed using the RNeasy Mini kit (Qiagen) according to the manufacturer's recommendations. The RNA eluted in 40 µl DEPC-treated water was treated with 10 U FPLC-pure DNase I (Amersham) for 30 min at 37 °C. RNA concentration was estimated by absorbance at 260 nm. RNA integrity was verified by electrophoresis of an ethidium-bromide-stained 1·4 % (w/v) agarose gel.

Total RNA (1 µg) was reverse-transcribed into cDNA using random hexamer primers (Invitrogen) and 200 U SuperscriptII Rnase H reverse transcriptase (Invitrogen) according to the procedure supplied with the enzyme. For each RNA, a negative RT reaction (no addition of reverse transcriptase) was performed to be used as a negative control in subsequent PCRs.

Real-time PCR was carried out using a LightCycler system (Roche Molecular Biochemicals). Primers for pilA1 and pilA2 yielded PCR products of 200 bp (Table 1). To control for RNA quality and cDNA synthesis, 16S rRNA was also amplified using primers that also yielded a 200 bp product. For LightCycler PCRs, a master mixture containing MgCl2 (4 mM), forward and reverse primers (0·5 µM each) and LightCycler-FastStart DNA Master SYBR Green I (2 µl per reaction; Roche Molecular Biochemicals) was prepared. Aliquots of the master mix (18 µl) were dispensed in the LightCycler glass capillaries and then various amounts of cDNA (0·1, 0·25, 0·5 or 1 ng reverse-transcribed total RNA) were added as PCR template to give a final volume of 20 µl. After denaturation at 95 °C for 10 min, the amplification and quantification program was repeated 45 times (95 °C for 15 s, 62 °C for 5 s, 72 °C for 8 s with a single fluorescence measurement), followed by the melting curve program (65–95 °C with a heating rate of 0·1 °C s–1 and a continuous fluorescence measurement) and finally a cooling step to 45 °C. Negative controls without cDNA template or with RNA not reverse-transcribed were run with every assay to assess the overall specificity.

Quantification involved the use of standard curves that had been prepared with PCR products corresponding to almost the entire sequence of each gene (pilA1, 520 bp; pilA2, 450 bp; rrs, 1484 bp). The pilA1 and pilA2 genes were amplified from R. albus 20 genomic DNA, using the same primers and PCR parameters as those given for the isolation of transcripts. For the rrs gene, the forward bacterial domain primer F8 (5'-AGAGTTTGATCMTGGCTC-3'; Tm 52 °C) and the reverse universal domain primer 1492R (5'-GNTACCTTGTTACGACTT-3'; Tm 52 °C) were used. For the construction of standard curves for pilA1, pilA2 and rrs genes, PCR products were prepared as a 10-fold serial dilution in water, from 102 to 108 copies. LightCycler quantification software (version 3.5) was used to compare amplification in experimental samples during the log-linear phase to the standard curve from the dilution series of control PCR products. Relative results are reported in terms of the copy number of pilA1 or pilA2 mRNA per 106 copies of 16S rRNA. Each experiment was repeated three times to confirm the reproducibility of the results.

Cloning and expression of the pilA2 gene.
The pilA2 gene was PCR-amplified using primers that removed the first 18 nucleotides encoding the signal sequence of the predicted protein and also contained NcoI and XhoI sites at their 5' and 3' ends, respectively (Table 1). The resulting PCR product was cloned into an NcoI/XhoI-digested pET28a(+) vector such that a His6 tag was fused with the C terminus of the recombinant protein. Clones were expressed in E. coli BL21(DE3) as follows: single colonies were transferred into 10 ml LB medium with 25 µg kanamycin ml–1. These cultures were then used to inoculate 1 l volumes of the same medium and were allowed to grow at 37 °C with agitation to an OD600 of 0·3–0·5; 1 mM IPTG was then added and the culture was incubated for a further 6 h at 37 °C.

Purification of His-tagged PilA2 and preparation of anti-PilA2-His6 serum.
The PilA2-His6 recombinant protein was purified from E. coli cell extracts by NTA-resin affinity chromatography (Qiaexpressionist System; Qiagen). Briefly, cultures of recombinant E. coli were centrifuged (10 000 g, 4 °C, 15 min) and the cell pellet was resuspended in 25 ml lysis buffer (50 mM NaH2PO4, 300 mM NaCl, pH 8) and incubated with lysozyme (1 mg ml–1) at 4 °C for 30 min. The cell suspension was then broken by two passages through a French pressure cell set at 1250 bar (125 MPa). Unbroken cells and large debris were removed by centrifugation (10 000 g, 4 °C, 30 min) and the supernatant containing cytoplasmic proteins was loaded onto 4 ml Ni-NTA resin equilibrated with lysis buffer. Adsorbed proteins were eluted by adding 12 ml aliquots of 20, 50, 100, 150, 200 and 250 mM imidazole in lysis buffer. The PilA2-His6 recombinant protein appeared in fractions eluted with 100 and 150 mM imidazole that were pooled and dialysed against water using a regenerated cellulose membrane (Spectra/Por; Spectrum) with a molecular mass cut-off between 6 and 8 kDa. The dialysate was concentrated by centrifugation (5000 g, 4 °C, 90 min) on a Microsep concentrator with a molecular mass cut-off of 10 kDa (Pall Filtron). It was finally purified by a gel extraction procedure (Hames, 1981). The anti-PilA2-His6 serum was prepared by injecting adult New Zealand white rabbits subcutaneously with 50 µg protein (in acrylamide) that had been suspended in 1 ml sterile NaCl (0·9 % w/v) and emulsified with Freund's incomplete adjuvant. Injections were performed three times, at 3-week intervals, and the rabbits were bled 2 weeks after the last immunization. The blood was incubated without agitation for 3 h at 37 °C and centrifuged at 3000 g for 25 min. The upper, serum phase was stored at –20 °C. Blood was also collected from the rabbits prior to immunization and treated in the same manner to provide pre-immune serum.

Preparation of anti-peptides targeting PilA2.
Two peptides of 15 amino acids designated the internal peptide (MIYNSIAANAKNSG; positions 90–104) or C-terminal peptide (NSVHNVPTWKAYKAD; positions 141–155) were chosen to prepare specific antibodies against PilA2. These peptides were chosen to ensure that the extremely hydrophobic and highly conserved N-terminal region of type IV pilins was excluded as an epitope(s), while the C-terminal region of the protein should be quite antigenic (Hazes et al., 2000). The internal peptide was chosen by using hydrophobic cluster analysis (Gaboriaud et al., 1987) and was predicted to reside in a loop at the boundary of {beta}2–{beta}3 strands of type IV pilins. This sensitive two-dimensional method of sequence analysis is able to detect structural relationships between protein sequences sharing low levels of sequence identity (typically below 20 % identity).

Pre-immune serum, the antiserum obtained against the two (C-terminal and internal) PilA2 peptides hereafter referred to as anti-peptide antiserum and affinity-purified antibodies against either the C-terminal peptide (anti-PilA2-C antiserum) or the internal peptide of PilA2 (anti-PilA2-I antiserum) were prepared and provided by Eurogentec.

Western blot analysis of R. albus proteins.
After growth on RF cellobiose medium to an OD600 of approx. 1·2, R. albus 20 and the adhesion-defective mutant D5 were harvested by centrifugation (10 000 g, 4 °C, 20 min). The resulting cell pellets were washed three times in 50 mM sodium phosphate buffer (pH 6·9) and then resuspended in one-tenth of their original volume of the same buffer and frozen at –20 °C. The culture fluid, containing extracellular proteins, was also aliquotted and frozen. CBPs from the extracellular culture fluid were recovered as described previously (Rakotoarivonina et al., 2002) and membrane and cytoplasmic proteins plus cell-associated CBPs were prepared following already-described procedures (Mosoni & Gaillard-Martinie, 2001). Western immunoblot analysis of these different protein preparations was performed as described by Rakotoarivonina et al. (2002) except that the antibodies used were the anti-PilA2-C antiserum (dilution 1/2000).

Inhibition of bacterial adhesion.
The effect of the anti-peptide antiserum on the adhesion of R. albus 20 to cellulose was determined as previously described (Rakotoarivonina et al., 2002).


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Gene organization at the cbpC/gp25 loci of R. albus strains 8 and 20
Located 115 bp downstream of the transcriptional terminator for the cbpC gene in R. albus 8 was another open reading frame (ORF) that would encode a 155 residue protein with a calculated molecular mass of 16 995 Da. The N-terminal third of this ORF showed significant sequence identity with the N-terminal sequences of CbpC and other Pil proteins. This second Pil protein (hereafter referred to as PilA2) possesses a six-amino-acid positively charged leader sequence with the Gly–Phe dipeptide proteolytic cleavage site, the canonical Glu residue at position 5 of the mature protein and the highly conserved hydrophobic sequence found in the N terminus of type IV pilin subunits and homologues (Dalrymple & Mattick, 1987). A Gly residue was present at position 58 of the mature protein, as was also found in the mature CbpC protein. This residue represents the border between the conserved, N-terminal third and variable, C-terminal two-thirds of Pil-like proteins (Dalrymple & Mattick, 1987). In addition to the type IV fimbria-like signal peptide encoded by pilA2, a {sigma}70 promoter, Shine–Dalgarno sequence and Rho-independent transcriptional terminator were also present.

The two ORFs directly downstream of pilA2 in R. albus 8 have significant amino acid identity to SecD and SecF. The SecD-like protein has 34 % (84 of 245 residues) and 29 % (127 of 437 residues) amino acid identity to the Synechocystis sp. protein-export membrane protein SecD and to the SecD component of the Bacillus subtilis SecDF polypeptide, respectively. The SecF-like protein of R. albus has 31 % (91 of 291 residues) amino acid identity to SecF of the B. subtilis SecDF polypeptide. In B. subtilis, the SecD and SecF proteins are fused and, as an integral membrane protein, contribute to the efficiency of protein secretion via the SecYEG translocation channel (Bolhuis et al., 1998). The 468 residue SecD-like and 324 residue SecF-like proteins of R. albus have predicted molecular masses of 50 516 Da and 35 382 Da, respectively. The presence of a {sigma}70 promoter sequence upstream of the secD-like gene, but the lack of a prominent promoter sequence upstream of the secF-like gene, suggests that the two genes are cotranscribed. Considering that the translational start codon of the SecF-like protein overlaps the translational stop codon of the SecD-like protein, the SecD- and SecF-like proteins appear to be translationally coupled.

Gene organization downstream of the gp25 gene in R. albus 20 was the same as that in R. albus 8 (Fig. 1). The 1·5 kb of sequenced DNA was found to contain two ORFs in the same orientation as the gp25 gene. The first ORF (468 bp) was preceded by a {sigma}70-like promoter sequence and ribosome-binding site and ended with a putative Rho-independent terminator. It encoded a 155 amino acid protein with a calculated molecular mass of 16 844 kDa that shared 98 % identity with PilA2 of R. albus 8 and 41·1 and 41·8 % identity with GP25 (R. albus 20) and CbpC (R. albus 8), respectively. The N terminus of the protein also possessed all the characteristics of Pil proteins, as described above. On the basis of these identities, the gp25 and cbpC genes were renamed pilA1 and the novel downstream gene in both strains was termed pilA2. Although the second ORF (396 bp) is incomplete, it shared 100 % identity with the SecD-like protein identified in R. albus 8, supporting the contention that both strains possess secD/F-like genes in this region of their genomes.



View larger version (7K):
[in this window]
[in a new window]
 
Fig. 1. Organization of the pil/sec locus in R. albus strains 8 and 20. The 3·7 kb region sequenced in strain 8 contains two pil genes, cbpC and pilA2, and two presumably linked ORFs (secDF). The 1·8 kb region sequenced in strain 20 contains two pil genes, gp25 and pilA2, and the beginning of a third ORF (secD). In agreement with the pil gene nomenclature, genes cbpC and gp25 were renamed: the new names are given in parentheses. Genes are represented by boxes with leader sequences in black. Putative promoters are indicated by arrowheads.

 
The pilA2 gene and the region immediately upstream (containing a putative promoter) was sequenced from mutant D5 and the nucleotide sequence was found to be identical to that obtained from strain 20, showing that no mutation occurred in this gene in the adhesion-defective mutant.

The pilA1 and pilA2 genes are expressed as both mono- and bicistronic transcripts
We have previously shown that the pilA1 (cbpC) transcript is predominantly monocistronic in R. albus 8 cultured with cellobiose (Pegden et al., 1998). Northern blot analysis of pilA2 expression in R. albus 8 following growth with cellobiose is shown in Fig. 2. The predominant pilA2-encoding transcript was also monocistronic, but a second, larger transcript was also detected. The estimated size of this second transcript (1·35 kb) is large enough to encode both the PilA1 and PilA2 coding sequences and the RT-PCR results for strain 20 (Fig. 3) verified the existence of a bicistronic transcript encoding both PilA1 and PilA2 in R. albus 20. When normalized relative to the amount of 16S rRNA present in each sample by phosphorimaging analysis, there were no differences in the abundance of either the mono- or bicistronic transcripts encoding pilA2 following growth of R. albus 8 in EM-cellobiose medium prepared to contain either rumen fluid, PAA/PPA or neither addition (Fig. 2). We therefore conclude that the pilA2 gene is not conditionally expressed in response to PAA/PPA.



View larger version (60K):
[in this window]
[in a new window]
 
Fig. 2. Northern blot analysis of pilA2 expression in R. albus 8. Total RNA was extracted from actively growing R. albus 8 cells cultured in EM-cellobiose medium prepared to contain either ruminal fluid (lane 1), no additions (lane 2) or ruminal fluid replaced with 25 µM each of PAA and PPA (lane 3). The RNA was subjected to agarose gel electrophoresis, transferred to a nitrocellulose membrane and probed with a pilA2-specific probe, as described in Methods. The mono- and bicistronic messages identified with the pilA2-specific probe are marked, as are the migration distances of RNA standards used to estimate transcript size. Phosphorimager analysis of the same blots probed with a radiolabelled oligonucleotide specific for 16S rRNA showed that there were no differences in total RNA loading (data not shown), verifying that pilA2 transcript abundance is similar under the growth conditions tested.

 


View larger version (51K):
[in this window]
[in a new window]
 
Fig. 3. Cotranscription of pilA1 and pilA2. RT-PCR product analysis obtained from R. albus 20 cDNAs using primer pairs specific to pilA1 (lane 2), pilA2 (lane 3) and the pilA1pilA2 region (lane 4). Molecular size DNA marker and a negative control performed with pilA1 primers and RNA not reverse-transcribed as template are shown in lanes 1 and 5, respectively. cDNAs were obtained by reverse transcription with random primers of total RNAs extracted from R. albus 20 cultivated on cellobiose.

 
Comparative analysis of pilA1 and pilA2 transcription in R. albus strains 8 and 20
Phosphorimaging analysis of Northern blots indicated that the pilA1 monocistronic transcript was 7·5- to 8-fold more abundant than the pilA2 transcript, suggesting that pilA1 gene expression is much higher than that of pilA2 (data not shown).

Transcript abundance for the pilA1 and pilA2 genes was therefore measured by quantitative RT-PCR and calibrated with that of the rrs gene in R. albus 20 and in the mutant D5, following growth to stationary phase on cellobiose (Fig. 4). There were only limited differences between strains 20 and D5 in terms of pilA1 and pilA2 transcript abundance when the strains were cultured with cellobiose, but levels tended to be slightly higher for the mutant strain. Similar to the results obtained with R. albus 8, the amount of pilA1 transcript was 73- and 77-fold higher than that of pilA2 in R. albus 20 and the mutant D5, respectively. Collectively, the results show that the mutant produces similar amounts of pilA1/A2 transcripts as the parent and that the pilA1 transcript is much more abundant than pilA2, although there is evidence for the production of a transcript encoding both the PilA1 and PilA2 ORFs.



View larger version (35K):
[in this window]
[in a new window]
 
Fig. 4. Quantification of pilA1 and pilA2 transcripts using real-time PCR (LightCycler system). 16S rRNAs were quantified to standardize and compare the expression levels of pilA1 and pilA2 according to the strain (R. albus 20 and adhesion-defective mutant D5), following growth to stationary phase on RF cellobiose medium. Purified PCR products of pilA1, pilA2 and rrs genes were used as templates (102–108 copies) to establish the external standard curves necessary for absolute quantification (see Methods).

 
Western immunoblot of the PilA2 protein
The antiserum raised against the PilA2-His6 recombinant protein showed a high degree of cross-reactivity with PilA1 even after several adsorptions of this serum with the PilA1 protein, and antibodies raised against the internal PilA2 antipeptide scarcely reacted with PilA2-His6 (not shown). However, the antiserum produced against the C-terminal PilA2 antipeptide reacted strongly with the recombinant protein (Fig. 5; lanes 1 and 9) and this antibody preparation (referred to as anti-PilA2-C) was used for all subsequent studies. The anti-PilA2-C antiserum reacted specifically with a protein of approximately 17 kDa that was present in similar amounts in total cellular protein preparations from mutant D5 and the parent strain 20 (Fig. 5; lanes 10 and 11). In addition, a larger amount of immunoreactive protein was present in the membrane than in the cytoplasmic protein fractions of both strains (Fig. 5; lanes 5–8). Furthermore, the anti-PilA2-C antiserum did not cross-react with cell-associated or extracellular forms of CBPs prepared from both R. albus 20 and D5 (Fig. 5; lanes 2–4). We previously showed that the extracellular fluid of R. albus 20 contained large amounts of broken pili and that these pili bound to cellulose (Rakotoarivonina et al., 2002). Consequently, the absence of antibody cross-reactivity with extracellular CBPs indicates that PilA2 is not secreted as a pilus component. Overall, these Western immunoblot results show that PilA2 is not part of the externalized structure of the pilus and is probably not involved in cellulose binding. It is probably a membrane-associated protein and the mutant strain D5 produces similar amounts of this protein as the wild-type strain.



View larger version (19K):
[in this window]
[in a new window]
 
Fig. 5. Western blot analysis of bacterial and extracellular proteins of R. albus 20 and D5 grown on cellobiose with the anti-PilA2-C antiserum. His-tagged PilA2 purified from recombinant E. coli was used as a control (lanes 1 and 9). The PilA2 protein was present in the cytoplasmic, membrane and total cellular proteins of R. albus 20 (lanes 5, 6 and 10, respectively) and D5 (lanes 7, 8 and 11, respectively), while it was not detectable in CBPs prepared from the extracellular fluid of R. albus 20 or D5 (lanes 2 and 3, respectively), nor in CBPs prepared from the cytoplasmic protein fraction of R. albus 20 (lane 4). For R. albus 20 and D5, a 10 µg protein sample was loaded on the gel (lanes 2–8), except for total cellular proteins (20 µg protein sample loaded; lanes 10 and 11).

 
Anti-PilA2 antibodies do not block the adhesion of R. albus 20 to cellulose
The anti-peptide antiserum was also specific to PilA2 (data not shown) and was tested for its effect on the adhesion of R. albus 20. The adhesion of R. albus 20 to cellulose was not reduced when either the anti-peptide antiserum or pre-immune serum were added to the adhesion assays, at concentrations of 2, 5 and 10 % (v/v) (data not shown). This is in contrast to the results obtained with the anti-Adh serum (i.e. anti-PilA1 antiserum) reported by Rakotoarivonina et al. (2002) and we conclude that the PilA2 protein is not directly involved with the adhesion process per se in R. albus.


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The structural subunit of type IV fimbriae has been characterized in a variety of pathogenic bacteria (Dalrymple & Mattick, 1987) and the biogenesis and function of type IV pili in Pseudomonas aeruginosa and in Neisseria gonorrhoeae is controlled by a large number of genes (Alm & Mattick, 1997; Wolfgang et al., 2000). For example, in P. aeruginosa, the fimbrial subunit gene (pilA) is clustered with the pilB, pilC and pilD genes (Nunn et al., 1990) and a two-component sensor–regulator pair (pilSpilR) controls the transcription of pilA (Hobbs et al., 1993). The PilB protein is associated with the cytoplasmic face of the inner membrane and possesses a conserved phosphate-binding loop motif (Walker box A) common to many nucleotide-binding proteins. The PilC protein is predicted to represent a polytopic inner-membrane protein that might provide an assembly platform for the fimbrial strand and PilD has been characterized as the prepilin peptidase. A model of type IV pili assembly in P. aeruginosa, involving PilA, PilB, PilC and PilD but also many other proteins (PilQ, P, E, V, W, X etc.), has recently been proposed (Mattick, 2002).

We previously showed that R. albus elaborates type IV pili on its surface and that these pili mediate bacterial adhesion to cellulose (Pegden et al., 1998; Rakotoarivonina et al., 2002). Such findings are indeed novel for Gram-positive bacteria, and the genetics and molecular biology underpinning the biogenesis and assembly of these structures is virtually undefined. In R. albus 20, the major pilus subunit is the 25 kDa glycoprotein now referred to as PilA1, and, in R. albus 8, the CbpC protein (Pegden et al., 1998), which is 73 % identical to PilA1, presumably plays the same role. Accordingly, we propose that CbpC now be referred to as PilA1. Both R. albus 8 and 20 also possess a second gene immediately downstream of pilA1 that encodes a Pil protein (pilA2), followed by genes homologous to secD and secF. This tandem arrangement of pil genes has been described before in several Gram-negative bacteria and in a strain of Synechocystis sp. In Pseudomonas stutzeri, the pilAI gene encodes the major fimbrial subunit, and inactivation of the pilAII gene does not affect fimbrial production but, instead, increases the transformability of P. stutzeri cells, suggesting that this second gene is involved in repression of DNA uptake (Graupner & Wackernagel, 2001). In Eikenella corrodens, the major fimbrial subunit is encoded by the pilA1 gene, and inactivation of the pilA2 gene does not change the phenotype, indicating that the pilA2 gene in this bacterium is also not essential for the biosynthesis of functionally normal fimbriae. However, when the pilA1 gene was deleted and the pilA2 gene was placed under the control of the pilA1 promoter, E. corrodens produced fimbriae composed of the PilA2 protein (Villar et al., 1999, 2001). The cyanobacterium Synechocystis sp. is surrounded by thin and thick pili structures and, while pilA2 gene inactivation does not have any effect on the presence of either type of fimbriae, the thick fimbriae disappear when the pilA1 gene is inactivated (Bhaya et al., 2000). Using antibodies specific to PilA2, we were able to show that PilA2 is not part of the externalized fimbrial structure but instead is an internalized, membrane-associated protein, as revealed by the presence of PilA2 in both membrane and cytoplasmic fractions of R. albus. In addition, the fact that the anti-PilA2 serum does not inhibit the adhesion of the bacterium to cellulose suggests that PilA2 is not externalized or directly involved in adhesion coordinated by PilA1, similar to the findings outlined above. Considering the cellular localization of the R. albus PilA2 protein and the relative abundance of the pilA2 transcript (and protein) compared with our previous findings for PilA1 (Rakotoarivonina et al., 2002), the PilA2 protein may be a part of the assembly complex, without being an integral part of the fimbrial structure itself. Several minor, membrane-associated Pil proteins are proposed to form the base structure of the fimbriae in P. aeruginosa (Mattick, 2002), in a manner similar to the inner-membrane Pil proteins that are a part of the type II (general) protein-secretion pathway in many Gram-negative bacteria (Hobbs & Mattick, 1993; Pugsley, 1993). Inactivation of the pilA2 gene in R. albus would permit a determination of its role in fimbrial assembly and function, but, until these bacteria are rendered amenable to experimental techniques in bacterial genetics, this is technically not possible.

We showed the presence of a pilA1pilA2 cotranscript, as well as a monocistronic pilA2 transcript. Our previous study confirmed that pilA1 is also expressed as a monocistronic transcript (Pegden et al., 1998). In this study, pilA1 mRNAs were approximately 70-fold more abundant than pilA2 mRNAs. To explain these results, we can offer two alternative hypotheses: (i) pilA1 and pilA2 genes are cotranscribed and the attenuation of pilA2 expression relative to pilA1 is due to the presence of a predicted hairpin structure in the pilA1pilA2 intergenic region or (ii) pilA2 constitutes a single gene transcriptional unit, with the pilA1 promoter being much stronger than the pilA2 promoter, and transcription starting from pilA1 promoter sometimes goes through the pilA1pilA2 intergenic region, giving rise to a polycistronic message.

The pilA1 and pilA2 gene sequences are identical in R. albus strain 20 and the D5 mutant, indicating that the mutant phenotype cannot be explained by an alteration in these genes. In addition, the fact that the amount of PilA2 protein is similar in the mutant and the parent following growth on cellobiose is in agreement with the pilA2 transcript abundances measured in both strains. On the contrary, although the transcript levels of the pilA1 gene in the parent and mutant strains are similar, PilA1 protein abundance is reduced in the mutant strain D5 (Mosoni & Gaillard-Martinie, 2001). Because of the complex phenotype of the mutant (lack of pili, reduced cellulase activity and altered glycocalyx), we previously assumed that the mutation in strain D5 is pleiotropic in its effects, perhaps within a gene encoding a regulatory function associated with the synthesis or the secretion of surface and/or extracellular molecules. The fact that PilA1 protein is less abundant in the mutant than in the parent but there are only limited differences between the two strains in terms of pilA1 transcript abundance suggests that the mutation exerts its effects at the level of post-translation or secretion.

In conclusion, our initial hypothesis that the pilA1 gene is part of a larger collection of genes involved with the secretion and assembly of type IV pili was supported by the identification of a second Pil-like protein, PilA2, followed by genes encoding accessory functions for protein secretion. Now that the genome sequence for R. albus strain 8 is almost complete, it should be possible to assemble a more holistic picture of the candidate genes involved in the synthesis and assembly of type IV fimbrial proteins in R. albus and to develop a comparative understanding of this mechanism of adhesion for Gram-positive bacteria.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Alm, R. A. & Mattick, J. S. (1997). Genes involved in the biogenesis and function of type-4 fimbriae in Pseudomonas aeruginosa. Gene 192, 89–98.[CrossRef][Medline]

Altschul, S. F., Gish, W., Miller, W., Myers, E. W. & Lipman, D. J. (1990). Basic local alignment search tool. J Mol Biol 215, 403–410.[CrossRef][Medline]

Bhaya, D., Bianco, N. R., Bryant, D. & Grossman, A. (2000). Type IV pilus biogenesis and motility in the cyanobacterium Synechocystis sp. PCC6803. Mol Microbiol 37, 941–951.[CrossRef][Medline]

Bolhuis, A., Broekhuizen, C. P., Sorokin, A., van Roosmalen, M. L., Venema, G., Bron, S., Quax, W. J. & van Dijl, J. M. (1998). SecDF of Bacillus subtilis, a molecular Siamese twin required for the efficient secretion of proteins. J Biol Chem 273, 21217–21224.[Abstract/Free Full Text]

Dalrymple, B. & Mattick, J. S. (1987). An analysis of the organization and evolution of type 4 fimbrial (MePhe) subunit proteins. J Mol Evol 25, 261–269.[Medline]

Devillard, E., Goodheart, D. B., Karnati, S. K., Bayer, E. A., Lamed, R., Miron, J., Nelson, K. E. & Morrison, M. (2004). Ruminococcus albus 8 mutants defective in cellulose degradation are deficient in two processive endocellulases, Cel48A and Cel9B, both of which possess a novel modular architecture. J Bacteriol 186, 136–145.[Abstract/Free Full Text]

Ding, S. Y., Rincon, M. T., Lamed, R., Martin, J. C., McCrae, S. I., Aurilia, V., Shoham, Y., Bayer, E. A. & Flint, H. J. (2001). Cellulosomal scaffoldin-like proteins from Ruminococcus flavefaciens. J Bacteriol 183, 1945–1953.[Abstract/Free Full Text]

Gaboriaud, C., Bissery, V., Benchetrit, T. & Mornon, J. P. (1987). Hydrophobic cluster analysis: an efficient new way to compare and analyse amino acid sequences. FEBS Lett 224, 149–155.[CrossRef][Medline]

Gong, V., Egbosimba, E. E. & Forsberg, C. W. (1996). Cellulose-binding proteins of Fibrobacter succinogenes and the possible role of a 180-kDa cellulose-binding glycoprotein in adhesion to cellulose. Can J Microbiol 42, 453–460.

Graupner, S. & Wackernagel, W. (2001). Pseudomonas stutzeri has two closely related pilA genes (type IV pilus structural protein) with opposite influences on natural genetic transformation. J Bacteriol 183, 2359–2366.[Abstract/Free Full Text]

Hames, B. D. (1981). Recovery of separated proteins. In Gel Electrophoresis of Proteins: a Practical Approach, pp. 61–64. Edited by B. D. Hames & D. Rickwood. Oxford: IRL Press.

Hazes, B., Sastry, P. A., Hayakawa, K., Read, R. J. & Irvin, R. T. (2000). Crystal structure of Pseudomonas aeruginosa PAK pilin suggests a main-chain-dominated mode of receptor binding. J Mol Biol 299, 1005–1017.[CrossRef][Medline]

Hobbs, M. & Mattick, J. S. (1993). Common components in the assembly of type 4 fimbriae, DNA transfer systems, filamentous phage and protein-secretion apparatus: a general system for the formation of surface-associated protein complexes. Mol Microbiol 10, 233–243.[Medline]

Hobbs, M., Collie, E. S., Free, P. D., Livingston, S. P. & Mattick, J. S. (1993). PilS and PilR, a two-component transcriptional regulatory system controlling expression of type 4 fimbriae in Pseudomonas aeruginosa. Mol Microbiol 7, 669–682.[Medline]

Malburg, S. R., Malburg, L. M., Jr, Liu, T., Iyo, A. H. & Forsberg, C. W. (1997). Catalytic properties of the cellulose-binding endoglucanase F from Fibrobacter succinogenes S85. Appl Environ Microbiol 63, 2449–2453.[Abstract]

Mattick, J. S. (2002). Type IV pili and twitching motility. Annu Rev Microbiol 56, 289–314.[CrossRef][Medline]

Miron, J. & Forsberg, C. W. (1999). Characterisation of cellulose-binding proteins that are involved in the adhesion mechanism of Fibrobacter intestinalis DR7. Appl Microbiol Biotechnol 51, 491–497.[CrossRef][Medline]

Miron, J., Ben-Ghedalia, D. & Morrison, M. (2001). Invited review: adhesion mechanisms of rumen cellulolytic bacteria. J Dairy Sci 84, 1294–1309.[Abstract/Free Full Text]

Mitsumori, M. & Minato, H. (2000). Identification of the cellulose-binding domain of Fibrobacter succinogenes endoglucanase F. FEMS Microbiol Lett 183, 99–103.[CrossRef][Medline]

Morrison, M. & Miron, J. (2000). Adhesion to cellulose by Ruminococcus albus: a combination of cellulosomes and Pil-proteins? FEMS Microbiol Lett 185, 109–115.[CrossRef][Medline]

Mosoni, P. & Gaillard-Martinie, B. (2001). Characterization of a spontaneous adhesion-defective mutant of Ruminococcus albus strain 20. Arch Microbiol 176, 52–61.[CrossRef][Medline]

Nunn, D., Bergman, S. & Lory, S. (1990). Products of three accessory genes, pilB, pilC, and pilD, are required for biogenesis of Pseudomonas aeruginosa pili. J Bacteriol 172, 2911–2919.[Medline]

Pegden, R. S., Larson, M. A., Grant, R. J. & Morrison, M. (1998). Adherence of the gram-positive bacterium Ruminococcus albus to cellulose and identification of a novel form of cellulose-binding protein which belongs to the Pil family of proteins. J Bacteriol 180, 5921–5927.[Abstract/Free Full Text]

Pugsley, A. P. (1993). The complete general secretory pathway in gram-negative bacteria. Microbiol Rev 57, 50–108.[Medline]

Rakotoarivonina, H., Jubelin, G., Hebraud, M., Gaillard-Martinie, B., Forano, E. & Mosoni, P. (2002). Adhesion to cellulose of the Gram-positive bacterium Ruminococcus albus involves type IV pili. Microbiology 148, 1871–1880.[Medline]

Reveneau, C., Adams, S. E., Cotta, M. A. & Morrison, M. (2003). Phenylacetic and phenylpropionic acids do not affect xylan degradation by Ruminococcus albus. Appl Environ Microbiol 69, 6954–6958.[Abstract/Free Full Text]

Rincon, M. T., McCrae, S. I., Kirby, J., Scott, K. P. & Flint, H. J. (2001). EndB, a multidomain family 44 cellulase from Ruminococcus flavefaciens 17, binds to cellulose via a novel cellulose-binding module and to another R. flavefaciens protein via a dockerin domain. Appl Environ Microbiol 67, 4426–4431.[Abstract/Free Full Text]

Rincon, M. T., Ding, S. Y., McCrae, S. I., Martin, J. C., Aurilia, V., Lamed, R., Shoham, Y., Bayer, E. A. & Flint, H. J. (2003). Novel organization and divergent dockerin specificities in the cellulosome system of Ruminococcus flavefaciens. J Bacteriol 185, 703–713.[Abstract/Free Full Text]

Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Villar, M. T., Helber, J. T., Hood, B., Schaefer, M. R. & Hirschberg, R. L. (1999). Eikenella corrodens phase variation involves a posttranslational event in pilus formation. J Bacteriol 181, 4154–4160.[Abstract/Free Full Text]

Villar, M. T., Hirschberg, R. L. & Schaefer, M. R. (2001). Role of the Eikenella corrodens pilA locus in pilus function and phase variation. J Bacteriol 183, 55–62.[Abstract/Free Full Text]

Wolfgang, M., van Putten, J. P., Hayes, S. F., Dorward, D. & Koomey, M. (2000). Components and dynamics of fiber formation define a ubiquitous biogenesis pathway for bacterial pili. EMBO J 19, 6408–6418.[Abstract/Free Full Text]

Zheng, D., Alm, E. W., Stahl, D. A. & Raskin, L. (1996). Characterization of universal small-subunit rRNA hybridization probes for quantitative molecular microbial ecology studies. Appl Environ Microbiol 62, 4504–4513.[Abstract]

Received 29 October 2004; revised 11 January 2005; accepted 12 January 2005.



This Article
Abstract
Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Citation Map
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Google Scholar
Articles by Rakotoarivonina, H.
Articles by Mosoni, P.
Articles citing this Article
PubMed
PubMed Citation
Articles by Rakotoarivonina, H.
Articles by Mosoni, P.
Agricola
Articles by Rakotoarivonina, H.
Articles by Mosoni, P.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
INT J SYST EVOL MICROBIOL MICROBIOLOGY J GEN VIROL
J MED MICROBIOL ALL SGM JOURNALS
Copyright © 2005 Society for General Microbiology.