This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental material
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Qi, M.
Right arrow Articles by Forsberg, C. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Qi, M.
Right arrow Articles by Forsberg, C. W.

 Previous Article  |  Next Article 

Journal of Bacteriology, June 2005, p. 3739-3751, Vol. 187, No. 11
0021-9193/05/$08.00+0     doi:10.1128/JB.187.11.3739-3751.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.

Novel Molecular Features of the Fibrolytic Intestinal Bacterium Fibrobacter intestinalis Not Shared with Fibrobacter succinogenes as Determined by Suppressive Subtractive Hybridization{dagger}

Meng Qi,1 Karen E. Nelson,2 Sean C. Daugherty,2 William C. Nelson,2 Ioana R. Hance,2 Mark Morrison,3 and Cecil W. Forsberg1*

Department of Molecular and Cellular Biology, University of Guelph, Guelph, Ontario, Canada,1 The Institute for Genomic Research, Rockville, Maryland,2 Department of Animal Science, Ohio State University, Columbus, Ohio3

Received 18 February 2005/ Accepted 2 March 2005


arrow
ABSTRACT
 
Suppressive subtractive hybridization was conducted to identify unique genes coding for plant cell wall hydrolytic enzymes and other properties of the gastrointestinal bacterium Fibrobacter intestinalis DR7 not shared by Fibrobacter succinogenes S85. Subtractive clones from F. intestinalis were sequenced and assembled to form 712 nonredundant contigs with an average length of 525 bp. Of these, 55 sequences were unique to F. intestinalis. The remaining contigs contained 764 genes with BLASTX similarities to other proteins; of these, 80% had the highest similarities to proteins in F. succinogenes, including 30 that coded for carbohydrate active enzymes. The expression of 17 of these genes was verified by Northern dot blot analysis. Of genes not exhibiting BLASTX similarity to F. succinogenes, 30 encoded putative transposases, 6 encoded restriction modification genes, and 45% had highest similarities to proteins in other species of gastrointestinal bacteria, a finding suggestive of either horizontal gene transfer to F. intestinalis or gene loss from F. succinogenes. Analysis of contigs containing segments of two or more adjacent genes revealed that only 35% exhibited BLASTX similarity and were in the same orientation as those of F. succinogenes, indicating extensive chromosomal rearrangement. The expression of eight transposases, and three restriction-modification genes was confirmed by Northern dot blot analysis. These data clearly document the maintenance of carbohydrate active enzymes in F. intestinalis necessitated by the preponderance of polysaccharide substrates available in the ruminal environment. It also documents substantive changes in the genome from that of F. succinogenes, which may be related to the introduction of the array of transposase and restriction-modification genes.


arrow
INTRODUCTION
 
The genus Fibrobacter belongs to the poorly defined phylum Fibrobacteres (34) and is currently positioned between the Spirochetes and Bacteroidetes. The high cellulolytic activity and ability to degrade refractory plant structural polysaccharides of these organisms has given them a prominent role in gastrointestinal energy metabolism. Two species, Fibrobacter intestinalis and Fibrobacter succinogenes have been characterized based on 16S rRNA sequences and phenotypic properties. Both F. intestinalis and F. succinogenes are present in the rumen and ceca of cattle, pigs, and rats (3, 19, 30, 42). Of these two species F. succinogenes has received the most attention because of its substantially higher fibrolytic activity (28, 42). Genetic studies have identified seven cellulases and four xylanases, and have also shown that cellulose-binding proteins are important for the cellulolytic activity of this strain (9). The complexity of the cellulase system was further corroborated by sequence analysis of the F. succinogenes genome (31). Genes coding for over one hundred carbohydrate active enzymes, including cellulases, xylanases, polysaccharide lyases, and esterases, have been identified in the genome of F. succinogenes.

The most-studied strain of the F. intestinalis species, DR7, was shown to have an endoglucanase and a xylanase with high similarity to the respective enzymes of F. succinogenes S85 (18). In addition, multiple cellulose-binding proteins were identified (29). Furthermore, in vivo 13C nuclear magnetic resonance study of glucose and cellobiose metabolism by F. intestinalis and F. succinogenes has revealed marked homogeneity in their carbon metabolism, with glucose being used for glycogen storage and energy production, while part of the cellobiose was diverted from glycolysis for synthesis of cellodextrins and the accumulation of glucose-6-phosphate (23). Despite these genotypic and phenotypic similarities between the two species the 16S rRNA similarity is only 92% (30), suggesting that there are substantial, yet undefined genomic differences between the two strains.

In the present study we have applied the technique of suppressive subtractive hybridization (SSH) (2) to explore the genomic differences between F. intestinalis and F. succinogenes by identifying genes present in F. intestinalis that either have low or no similarity to those of F. succinogenes.


arrow
MATERIALS AND METHODS
 
Bacterial strains and growth conditions. F. succinogenes S85 (ATCC 19169) and F. intestinalis DR7 (ATCC 43855) strains were grown in a chemically defined medium prepared as described by Scott and Dehority (39), with either 0.3% (wt/vol) glucose or 0.3% (wt/vol) ball-milled cellulose as a carbon and energy source and with carbon dioxide as the gas phase. Cultures (10 ml) were grown at 37°C with reciprocating shaking at 150 rpm. To prepare genomic DNA, the bacteria were grown in medium containing glucose for 12 h. For total RNA extraction, F. intestinalis DR7 was grown from a 5% inoculum for either 12 h in medium containing glucose or 24 h in medium containing ball-milled cellulose. Cells isolated after these incubation periods were in the mid to late exponential phase of growth. Ball-milled cellulose was prepared by mixing 200 ml of a 6% (wt/vol) aqueous suspension of Sigmacell-20 cellulose with flint balls in an 800-ml Mill Jar (Norton, Chemical Process Products Division, Ohio) at 70 rpm for 48 h.

Genomic DNA extraction. Genomic DNA was isolated from F. intestinalis DR7 and F. succinogenes S85 by using the cetyltrimethylammonium bromide procedure described by Ausubel et al. (5).

Suppression subtractive hybridization. SSH was performed by using the PCR-Select bacterial genome subtraction kit (Clontech), with F. intestinalis DR7 as the tester and F. succinogenes S85 as the driver. Briefly, genomic DNA (2 µg) from the tester (F. intestinalis DR7) was digested with RsaI and separated into two portions, each of which was subjected to a ligation reaction to attach a different set of PCR adaptors. The two portions were then separately hybridized to excess RsaI-digested F. succinogenes S85 driver DNA at 63°C for 1.5 h to allow all sequences common between the two strains to hybridize, leaving enriched tester-specific single-stranded DNA. The two preparations were then mixed together and hybridized at 63°C for 20 h. Only sequences specific for the tester strain that had different adaptors on each strand were amplified in the subsequent PCRs as describe in the Clontech subtraction protocol.

The mixture of PCR products was purified by phenol-chloroform extraction and ethyl alcohol precipitation, ligated to the T-A cloning vector pGEM-T (Promega, Madison, WI), and transformed into Escherichia coli JM109. The E. coli strains were grown in LB medium supplemented with ampicillin at 100 µg/ml and screened for clones by using X-Gal (5-bromo-4-chloro-3-indolyl-ß-D-galactopyranoside)-IPTG (isopropyl-ß-D-thiogalactopyranoside) plates containing 1.6% agar, 0.2 mM IPTG, and 40 µg of X-Gal/ml. A total of 1056 white colonies with inserts were recovered; they were restreaked for purity, and the plasmid inserts were sequenced at The Institute for Genomic Research (TIGR; Rockville, MD) as described by Nesbø et al. (32).

Sequence analysis. Sequences were examined by using the TIGR BLAST program (http://tigrblast.tigr.org/ufmg/) and the National Center for Biotechnology Information (NCBI) BLAST program (http://www.ncbi.nlm.nih.gov/BLAST/). The expectation value of 0.001 was chosen as the cutoff, i.e., those with expectation values of >0.001 were considered to be lacking similarity. Similar criteria were used in several recent studies (32, 41). Genes with protein sequences lacking similarity according to this criterion to proteins encoded by either the F. succinogenes genome or gene in GenBank were further analyzed by position-specific iterative BLAST (PSI-BLAST) and by searches against the conserved domain databases. The genome of F. succinogenes S85 which is available at http://www.tigr.org/ was used for BLASTN and TBLASTX searches. The database used for BLASTX and PSI-BLAST searches was a combined database that included both the GenBank nonredundant amino acid database (nraa) and the translated coding sequences of F. succinogenes (http://www.tigr.org/tdb/ufmg/) to enable similarity comparisons by expectation value without the influence of database size. The PSI-BLAST was iterated up to three times for each sequence. Reverse-position-specific BLAST (RPS-BLAST) was used to search the conserved domain databases, including the NCBI's Conserved Domain Database, the Protein families database of alignments and HMMs (Pfam), Clusters of Orthologous Groups (COG and KOG), and the Simple Modular Architecture Research Tool (SMART), which are available at ftp://ftp.ncbi.nih.gov/pub/mmdb/cdd. The BLAST results were parsed and analyzed by using Microsoft Excel 2000 and Visual Basic for Application, version 6.3. Transposase genes from other organisms were downloaded from the UniProt database (http://www.uniprot.org/). A dendrogram showing the relationships of these genes were constructed by using CLUSTAL W (version 1.83) and visualized by using TreeView (version 1.6.6; http://taxonomy.zoology.gla.ac.uk/rod/treeview.html).

RNA extraction, cDNA synthesis, and digoxigenin labeling. Total RNA samples were prepared from a 10-ml culture of F. intestinalis DR7 grown in medium containing either glucose or cellulose by using the RNeasy minikit (Qiagen, Valencia, CA). Total RNA (10 µg) prepared from each culture was treated with 2 U of RNase-free DNase I (Amp Grade; Invitrogen) for 15 min at 22°C. The DNase I was then inactivated by the addition of 1 µl of 25 mM EDTA and heating for 10 min at 65°C. rRNA was removed from RNA preparations by using the MICROBExpress bacterial mRNA enrichment kit (Ambion). The enriched mRNA was treated a second time with 0.5 U of RNase-free DNase I for 15 min to ensure absence of DNA contamination. Reverse transcription reactions were performed by using Superscript II reverse transcriptase (Invitrogen) according to the recommended protocol. After 50 min, 1 U of RNase H was added to both samples, followed by incubation at 37°C for a further 20 min. cDNA was then purified by using a QIAquick PCR purification kit (Qiagen) and denatured at 96°C for 10 min. Denatured cDNA samples were chilled on ice and labeled with a digoxigenin (DIG) High-Prime labeling mixture (Roche), followed by incubation at 37°C for 20 h, and then stored at –20°C until used. The amounts of the DIG-labeled probe were determined as described in the DIG High-Prime labeling mixture manual.

Dot blot analysis to assess expression of selected genes. To assess differential mRNA synthesis of selected genes by F. intestinalis, inserts in selected SSH clones (see Table 4) were amplified by PCR with nested primers 1 and 2R (Clontech protocol) and Platinum Taq DNA Polymerase High Fidelity (Invitrogen) by using 1-µl quantities of 16-h E. coli clone cultures as the templates. The reactions were performed in 30-µl volumes for 20 cycles with an annealing temperature of 52°C. To further dilute the cellular rRNA in the PCR, 0.3 µl of PCR product from each clone was amplified a further 20 cycles under the same conditions. The concentration of PCR product was determined by using PicoGreen reagent (Molecular Probes) as described by Ahn et al. (1). A 120-ng quantity of amplified insert from each of the clones was mixed with 0.4 M NaOH and heated at 100°C and then spotted on duplicate Hybond N+ nylon membranes (Amersham) by using a Bio-Dot SF Microfiltration Apparatus (Bio-Rad, Hercules, CA). The membranes were then baked for 2 h at 80°C. The hybridization and detection procedure that followed was carried out by using 400 ng of the DIG-labeled cDNA prepared with mRNA extracted from cells grown in medium containing either glucose or cellulose as an energy source as described by Ross et al. (37). The membranes were scanned, and the band profiles were plotted by using Scion Image software (version 4.0.2; Scion, Frederick, MD). The intensity of each band was represented by the area under the corresponding peak. A standard curve was plotted by spotting a range of 0.01 to 0.3 pg of standard DIG-labeled DNA onto a membrane and, after color development, the intensities of the bands were detected by the same procedure. The intensity increased in proportion to the amount of DIG-labeled probe up to a value of 900 arbitrary units with 0.3 pg of probe, which encompassed the intensity of the test sample blots. To determine the extent to which the genes were either up- or downregulated in cells during growth on glucose or on cellulose two constitutive house-keeping genes, RSGA_692, a phosphoglycerate kinase, and RSGA_128, a glucose-6-phosphate isomerase, were included on each blot to correct for differences in expression. The band intensities of the two housekeeping genes for cells grown on cellulose were divided by the values for cells grown on glucose. The average of the two ratios was 1.075, showing that there was slightly higher expression of the housekeeping genes on cellulose. To correct for this higher level of expression, the band intensity for each of the genes of cellulose grown cells was divided by the 1.075. Each gene selected was tested at least twice.


View this table:
[in this window]
[in a new window]
  		TABLE 4. Differential expression of putative glycosyl hydrolases, transposases, and restriction modification genes in F. intestinalis

Codon usage analysis. The codon usage was analyzed by using the General Codon Usage Analysis program as described by McInerney (26). Pearson correlation coefficients between all pairs of cellulase genes and transposase genes were calculated and transformed into distances and clustered by using the unweighted pair group method with the arithmetic mean algorithm as previously described (10). The dendrogram was generated by the TreeView (version 1.6.6) program. Since short DNA fragments were cloned, complete open reading frames (ORFs) were not often obtained. The partial ORF sequences were identified by BLASTX matches to known proteins in the combined nraa databases of F. succinogenes and GenBank. All ORFs of the F. succinogenes genome were analyzed for codon usage.

Nucleotide sequence accession numbers. The sequences determined in the present study have been submitted to the GenBank nucleotide sequence database under accession numbers CZ 443203 to CZ 443914. The nucleotide sequences with annotations can also be accessed on the FibRumBa website (http://www.tigr.org/tdb/rumenomics/).


arrow
RESULTS
 
A total of 1,056 clones were sequenced in both directions, the sequences were assembled, and redundant sequences were removed, resulting in 712 contigs with an average length of 525 bp (Fig. 1). As determined by a BLASTN search of the genome of F. succinogenes, of the 712 nonredundant DNA contigs, 437 contained genes different from those in F. succinogenes (with an expectation value of >0.001). Another 206 had expectation values of ≤0.001, but with the DNA similarity to F. succinogenes of <80%, and 69 had sequence similarities of >80% over a ≥100-nucleotide region to the genes in F. succinogenes. None of the SSH contigs had a DNA similarity to F. succinogenes genome of >90% along the whole length of the sequence. The average G+C% of the unique clones was 49.4%, which is similar to the 48.1% reported for F. succinogenes (31). Using the estimate of redundancy as described by Nesbø et al. (32), the theoretical library size would be ca. 3,400 clones, with ca. 2,400 unique clones each containing 525 bp, which is equivalent to 1,260 kb of DNA. Therefore, assuming the genome size of F. intestinalis is similar to that of F. succinogenes, i.e., 3843 kb, the difference in genome content between the two strains is ca. 33%. A total of 145 contigs had no similarity to the genome of F. succinogenes, while among these contigs, BLASTX searches against the GenBank nraa showed similarity to 79 of them, and PSI-BLAST and conserved domain database searches found 11 more contigs that may share conserved domains with other proteins; the remaining 55 contigs were unique with no similarity to proteins in the nraa databases as of September 2004 (Fig. 1B). These 145 contigs demonstrate that at least 66 kb of DNA in the F. intestinalis chromosome is totally different from that in F. succinogenes.



View larger version (21K):
[in this window]
[in a new window]
  		FIG. 1. Summary of the similarities of F. intestinalis contigs to genes present in the GenBank nonredundant amino acid database and the F. succinogenes S85 genome translated amino acid sequence database. The distribution of contigs is based on the gene in each contig with the lowest E-value. A possible second gene in a contig was not taken into account. The DNA inserts from clones were assembled into contigs, and redundant sequences were eliminated prior to BLAST analysis. (A) Similarity based on BLASTN search; (B) similarity based on BLASTX, tBLASTX, conserved domain database search, and PSI-BLAST search. Abbreviations: nraa, nonredundant amino acid database (which contains GenBank CDS translations, RefSeq Proteins, PDB, SwissProt, PIR and PRF); GFS, Genome of F. succinogenes database; GFS-aa, GFS amino acid sequence database; GFS-nraa, combined database contains both nraa and GFS-aa. Superscript a, total nucleic acid base pairs.

The results of BLASTX analysis of the contigs, and the assignment of coding sequences to general functional roles, are shown in Table 1. The majority of contigs coded for proteins with partial rather than full-length similarities to proteins in the respective databases, a reflection of the digestion pattern resulting from the enzymes used in the SSH experiments. However, 137 (19%) of the contigs contained at least two genes and the intergenic region, i.e., each end of the contig encoding a different protein (see below). In all, 764 genes encoding different proteins were identified by BLAST searches and, among these, 109 genes coded for proteins that were unrelated to those of F. succinogenes, of which 45% had the highest similarities to the proteins of other gastrointestinal bacteria (Table 2). The remaining 655 genes coded for proteins that showed similarity to those in F. succinogenes (Table S1), and ca. 93% of these showed greatest similarity to proteins in F. succinogenes.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Roles of the unique F. intestinalis genes identified by SSH that exhibit similarity to protein sequences in GenBank and the genome of F. succinogenesa


View this table:
[in this window]
[in a new window]
  		TABLE 2. F. intestinalis SSH contigs that contain genes encoding putative proteins with no similarity to proteins in F. succinogenesa

Among the 137 contigs that contained at least two genes, 118 coded for proteins that showed similarities to two or three different proteins encoded by the F. succinogenes genome (Tables S1 and S2 in the supplemental material), 15 had one or two matches to the proteins from F. succinogenes and one match to the GenBank nraa excluding F. succinogenes proteins, and 4 contained two genes with no matches to proteins from F. succinogenes (Table 2). Of the 118 contigs encoding two different proteins with similarity to proteins from F. succinogenes, only 48 were encoded by genes that had the same gene order on the chromosome as those in F. succinogenes (Table S2 in the supplemental material), indicating substantial differences between the two species in terms of gene organization.

Carbohydrate-active enzymes. All of the cellulases and xylanases identified in F. intestinalis DR7 exhibited highest amino acid similarity to F. succinogenes S85 glycosyl hydrolases, although the percent identities did not exceed 89%. The glucanases corresponded to four family 5, two family 8, four family 9, one family 16, one family 18, one family 51 and one family 74 glycoside hydrolases. Two family 43 and one family 10 xylanases were also present in the library, and there was one family 26 mannanase gene and three polysaccharide lyases (Table 3).


View this table:
[in this window]
[in a new window]
  		TABLE 3. Putative carbohydrate active enzymes of F. intestinalis identified by SSH

Two clones contained gene fragments that encoded proteins exhibiting high similarity to the endoglucanase 3 encoded by the cel-3 that was previously cloned from F. intestinalis DR7 (18) (accession L39840), and they accounted for ca. 61% of the coverage of the gene. The cel-3 gene, which encodes a family 5 catalytic domain, had 68% DNA sequence similarity to its counterpart in F. succinogenes, and the similarity ranged from 30 to 80% in different regions.

Four genes showed highest similarities to the F. succinogenes endoglucanase gene celF (22). CelF, formerly known as endoglucanase 2 (24), is a major endoglucanase in the periplasm and extracellular culture fluid fraction of F. succinogenes. A putative celF gene fragment was cloned from F. intestinalis strain NR9 by Bera-Maillet et al. (6), which had 90% DNA similarity to the celF genes in F. intestinalis DR7, thereby documenting the universality of this gene in Fibrobacter sp.

Although all of the cellulase and xylanases genes identified encoded proteins showing the highest similarity to their counterparts in F. succinogenes, it is remarkable that 47% of these gene products showed the highest similarities (identities ranged from 28 to 44%) to translation products in the Cytophaga species when a BLASTX search was done against the GenBank nraa database, indicating the close relationship between the cellulase systems of the Fibrobacter and Cytophaga genera.

Genes involved in energy metabolism. Most of the genes identified with a role of energy metabolism coded for proteins with the highest similarity to those from F. succinogenes. However, two arsenate reductase genes were identified, which coded for proteins with 76% similarity over a 52-amino-acid sequence to the same enzyme in Clostridium perfringens, and 58% similarity over a 70-amino-acid sequence to the same enzyme in Chlorobium tepidum but had no counterpart in the F. succinogenes genome (Table 2). Arsenate is reduced to arsenite by some bacteria that grow anaerobically using the nonrespiratory substrate lactate as the electron donor (16). The existence of arsenate reductases may indicate the existence of arsenate respiration in F. intestinalis, although the possibility that it has a role limited to detoxification, or indeed, an unrecognized role, cannot be excluded.

Another gene that coded for a protein in this category that has no homolog with the F. succinogenes genome is a phosphoglycerate mutase (2,3-bisphosphoglycerate dependent) with 67% similarity over 86 amino acid residues to the C terminus of the 2,3-bisphosphoglycerate-dependent phosphoglycerate mutase gene in Ralstonia solanacearum (38). In contrast to F. intestinalis, the F. succinogenes genome contains a putative phosphoglycerate mutase, which is a 2,3-bisphosphoglycerate-independent enzyme. In nature two types of phosphoglycerate mutases catalyze the conversion of 3-phosphoglycerate to 2-phosphoglycerate. One uses 2,3-bisphosphoglycerate as a cofactor and catalyzes an intermolecular reaction with a phosphorylated histidine residue intermediate (13). The other type is the cofactor-independent phosphoglycerate mutase. Animals mainly contain the first type, while the second type is often found in plants. However, both types are found in bacteria. Whether F. intestinalis DR7 has a functional 2,3-bisphosphoglycerate-dependent phosphoglycerate mutase remains to be determined.

Genes involved in transport and binding. Thirteen of the genes identified coded for transport and binding proteins that showed no similarity to F. succinogenes. These included three proteins involved in ferrous or ferric ion transport, a phosphate-binding protein, three ABC transporter proteins that may be responsible for nitrate/sulfonate/bicarbonate and polysaccharide/polyol phosphate transport, two Na+/H+ antiporters, one chromate transport protein, one periplasmic substrate-binding protein, one putative permease, and a transporter of unknown function (Table 2). The differences in transport capabilities are most likely a reflection of the different environmental niches that the two species occupy.

Genes involved in DNA metabolism. Nine of the genes in this functional group coded for proteins that showed greatest similarity to those in bacteria other than F. succinogenes. These included an anaerobic ribonucleoside-triphosphate reductase involved in DNA replication, a DNA damage-inducible protein, a nuclease subunit of the exonuclease complex, three type I, one type II, and one type IIS restriction modification enzymes and a cytosine-specific modification methyltransferase. The type IIS restriction-modification enzyme identified was most similar to a putative type IIS restriction-modification enzyme in Bacteroides thetaiotaomicron (41), whereas this kind of restriction-modification system was absent from the genome of F. succinogenes. Previous studies have shown the existence of a type II restriction-modification system in F. succinogenes (17), and type I and III restriction-modification systems have also been identified from the genome of F. succinogenes (31).

Genes involved in cell envelope metabolism. The putative proteins encoded by 11 genes with roles in cell envelope metabolism showed no similarity to proteins in F. succinogenes. Most of these proteins were involved in the biosynthesis of cell surface polysaccharides and included three glycosyl transferases, two glycerophosphotransferases, one capsule biosynthesis protein, one GDP-D-mannose dehydratase, one cell-wall-associated protein, two outer membrane proteins, and one putative flippase. Since no research has been conducted on the cell surface structure of F. intestinalis, the possible function of each protein is speculative. In addition, because of the observed differences in the genes between F. intestinalis and F. succinogenes, differences in the organization and structure of the cell surfaces would be expected.

Transposases of F. intestinalis. In the present study, 30 different transposase gene segments were identified in F. intestinalis (Table 2). Ten of the putative proteins had high similarity to transposases from various bacteria, while twenty of them were homologous to a transposase, encoded by ISPg7, from Porphyromonas gingivalis. Since the genes identified from the SSH contigs were short, none of the encoded proteins matched the whole ISPg7 transposase. However, 11 of the 20 protein products, which had >95% DNA identity within the group, were homologous to some 110 amino acids at the N terminus of the P. gingivalis genes with 58 to 65% similarities, indicating the existence of multiple copies of the same insertion sequence (IS) elements. Thus, altogether it would appear that at least 11 different types of IS elements exist in F. intestinalis. TBLASTX searches of the F. succinogenes genome with each of the F. intestinalis transposase gene fragments did not reveal homologous putative transposases (31).

BLAST searches indicated that the 30 transposase gene fragments belong to five IS families (http://www-is.biotoul.fr/). Three belong to the ISL3 family, three belong to the IS4 family, one belongs to the IS256 family, and three belong to the IS605 family. The remaining 20 transposases, which were homologous to ISPg7 from P. gingivalis, belong to the IS5 family (Fig. 2).



View larger version (51K):
[in this window]
[in a new window]
  		FIG. 2. Dendrogram of putative transposases and their closest relatives identified from the Swiss-Prot database (http://www.expasy.org/), showing that they belong to different insertion sequence families. The genes from F. intestinalis were highlighted as RSGA plus the number. The genes closest to the F. intestinalis clones were selected by using a BLASTP search against the Swiss-Prot database. The accession number and name of organisms are shown in the figure. The tree was constructed by using CLUSTAL W (version 1.83) and TreeView (version 1.6.6). The dashed line separates the different transposase families. The genes are clustered in different branches based on their sequence similarities. However, conclusive phylogenetic relationships between the genes cannot be accurately determined because the SSH transposase genes are incomplete sequences.

Transcription of glycosyl hydrolases, transposases, and restriction-modification enzymes in F. intestinalis DR7 that have low DNA similarity to F. succinogenes S85. To examine whether the genes identified in F. intestinalis were functional or simply vestigial DNA, mRNA synthesis for 23 glycosyl hydrolases, 9 transposases, and 7 restriction-modification enzymes were determined by dot hybridization of mRNA from cells grown on glucose and cellulose as an energy source, respectively (Table 4). Of the 23 glycosyl hydrolase genes, 16 were expressed and, of these, 8 were expressed at higher levels and 2 were expressed at a lower level during growth on cellulose compared to growth on glucose. Six genes exhibited no detectable differences in expression during growth with either substrate. The eight genes with higher levels of expression during growth on cellulose included six endoglucanases, one rhamnogalacturonan lyase, and one lichenase. An endoglucanase E homolog (21, 25), a endoglucanase 3 homolog (21, 25), and an uncharacterized gene (RSGA_190) were expressed at a much higher level (Table 4).

The celF gene homolog (RSGA_183), which exhibited the highest expression level, was not differentially expressed on either glucose or cellulose. A family 9 glycosyl hydrolase (RSGA_675) and a family 18 chitinase (RSGA_360) were also expressed at a relative high level. A 1,4-{alpha}-glucan branching enzyme (RSGA_395) was expressed at a low level and perhaps was slightly downregulated during growth on cellulose.

Among the transposases RSGA_532 and RSGA_150 were substantially upregulated by growth on cellulose, whereas the remaining transposases were expressed constitutively, except for RSGA_113 that was not expressed. The type I restriction enzyme gene RSGA_009 was substantially upregulated by growth on cellulose, whereas the type III restriction enzyme gene RSGA_245 was downregulated. The housekeeping genes phosphoglycerate kinase and glucose-6-phosphate isomerase used as the reference for assessing expression were expressed constitutively.


arrow
DISCUSSION
 
Suppressive subtractive hybridization was used to study the genomic differences between two species in the genus of Fibrobacter by determining the genes present in F. intestinalis that exhibit low similarity to those present in F. succinogenes, as well as those that are completely absent from the genome of F. succinogenes. As detemined by a BLASTN search of the F. intestinalis contigs, 61% of the genes that were identified by SSH had no DNA similarity to the genome of F. succinogenes, and only 9.7% had more than 80% DNA identity, indicating the enrichment of F. intestinalis specific genes by SSH and, as well, the level of genomic diversity between these two species. However, at the protein level, genes having low DNA identity still showed high amino acid sequence similarity with 80% of the genes having a higher BLASTX similarity to the F. succinogenes proteins than to other organisms in the GenBank nraa database (Table 1 and Table S1 in the supplemental material).

In the present study as in previous SSH studies with other microbial species (4, 32), many of the F. intestinalis genes identified by SSH were more closely related to the genome that was used in the subtraction process, in this case F. succinogenes. Consequently, any gene detected by SSH as being present only in F. intestinalis should be treated as putatively unique, and other supporting evidence is essential for conclusive identification. In our case, the availability of the complete genome sequence of F. succinogenes (the driver) has greatly facilitated the interpretation of the SSH data, since the amino acid identities could be found by similarity searches.

SSH showed that, although divergent, F. intestinalis and F. succinogenes are common in several aspects. In F. succinogenes, 113 putative enzymes have been identified to have a role in plant cell wall degradation, including 40 cellulases and 29 xylanases (31). All of the subtractive cellulases and xylanases identified in F. intestinalis exhibit highest similarity to the comparable proteins in F. succinogenes. The celF gene, which codes for a predominant cellulase in F. succinogenes and belongs to family 51 glycosyl hydrolases, was present in F. intestinalis and was highly expressed in cells grown with either glucose or cellulose as the sole source of carbohydrate, just as it is in F. succinogenes. Most members in family 51 are arabinofuranosidases, which do not exhibit endoglucanase activity (7); however, CelF of F. succinogenes S85 had no activity on p-nitrophenol arabinofuranoside but had endoglucanase activity (22). Since the amino acid sequence identity between CelF and the homologous protein from F. intestinalis was described above 75%, whereas the identity between CelF and other family 51 arabinofuranosidases were <25%, it is very likely that the celF counterpart is also cellulolytic rather than exhibiting arabinase activity, although further experiments are needed to confirm this suggestion.

Conservation of the cellulases between the two species indicates the unique nature of the cellulase system in the Fibrobacter genus. Distinct strategies are exploited by different bacteria for the complete degradation of cellulosic materials. Generally, in anaerobic bacteria the cellulases are cell associated and are an intricate part of a cellulosome structure (20). However, none of the genes in the F. succinogenes genome or the F. intestinalis SSH clones exhibit similarity to the classical cohesin/dockerin domain signatures or to the scaffolding protein of a cellulosome complex. It has been reported that some of the glycosyl hydrolases in F. succinogenes have a basic terminal domain but lack high amino acid sequence similarity (21). One F. intestinalis contig (contig RSG_705) encodes a protein that has high similarity to the C terminus of ORF03727in F. succinogenes that encodes a xylanase with a basic terminal domain. The basic terminal domain was proposed to interact with the negatively charged cell surface rather than interact with a scaffolding protein (21). These data are insufficient to develop a model for cellulose digestion, but it does indicate that Fibrobacter species have a cellulase organization different from the typical cellulosomal structure present in many other anaerobic bacteria.

Besides conservation of the cellulase system, ca. 80% of the remaining genes that were identified had the highest similarity to genes in F. succinogenes (Table 1 and Table S1 in the supplemental material). Functions of many of the genes were assigned by similarity search, but some of them remain unknown and were annotated as hypothetical proteins. Among them are several hypothetical proteins that exist only within the two Fibrobacter species. As the genetics and physiology of this genus become more fully characterized, the true distribution and roles of these hypothetical and conserved hypothetical proteins will be determined.

SSH also showed clear differences between F. intestinalis and F. succinogenes. About 33% of the genome of F. intestinalis DR7 was absent or diverged greatly from the genome of F. succinogenes S85. Initial analyses reveal that there is at least 66 kb of DNA in the F. intestinalis genome that lacks similarity to F. succinogenes at the protein level.

Organization of the F. intestinalis DR7 genome. There is evidence of substantive genome rearrangements in F. intestinalis compared to F. succinogenes. If the organization of genes in F. intestinalis and F. succinogenes were identical, adjacent gene pairs should be the same in the two species. However, 89 of the 137 contigs or 65% of the contigs with segments of two adjacent genes did not contain the same gene order in F. succinogenes (Table 2 and Tables S1 and S2 in the supplemental material). Homologs of genes adjacent in F. intestinalis are 1.8 to 1,865 kbp apart in the F. succinogenes chromosome (Fig. S1 in the supplemental material). Furthermore, the fact that the difference in distribution of genes approached randomness suggests that the reorganization has been random rather than changes in restricted areas of the chromosomes of the species.

A fascinating question is the mechanism underlying the difference in genome organization between F. intestinalis and F. succinogenes. RecA-dependent recombination can result in rearrangements in the order of genes on the chromosome which may include duplication, translocation or inversion (12). IS elements with long repeats are often involved in chromosomal rearrangements. There is a positive correlation between the total number of IS elements and degree of rearrangement that is consistent with a role of IS elements in transposition and inversion events of DNA in a genome (27). Transposases are integral components of the IS elements. Since at least five transposase families have been identified in the genome, the repeat sequences of IS elements are very likely present within the F. intestinalis genome.

In some bacterial genomes, the restriction-modification elements were shown to be polymorphic (36). Similar polymorphisms were observed in F. succinogenes (31) and F. intestinalis. In an SSH study of Ruminococcus flavefaciens strain FD-1, Antonopoulos et al. (4) reported on several transposases and a restriction enzyme that were absent from strain JM1. These authors presented evidence of apparent domain swapping among the glycosyl hydrolases, which is probably related to the presence of transposases and the restriction enzyme. Similarly, in F. intestinalis the restriction-modification systems may have contributed to the genome rearrangement, since restriction-modification systems participate in generating bacterial diversity by promoting homologous recombination (11). Furthermore, a variety of mobile genetic units were found to be linked to restriction-modification gene complexes (15).

The difference in the gene complement of the F. intestinalis and F. succinogenes could also be due to horizontal gene transfer (HGT). HGT is commonly assumed to be an important force in shaping the prokaryote genome (33). Since limited information is known on the strains in the genus, HGT can only be inferred from the deviant base composition (including G+C content and codon usage) and abnormal sequence similarity. Since 80% of the genes identified are most closely related to the F. succinogenes genome, it is likely that most of these genes originate from the same ancestor and have evolved by internal shuffling, duplication, and/or accumulation of point mutation. The codon usage distance among the cellulases from F. succinogenes and F. intestinalis and transposases from F. intestinalis was calculated, and a dendrogram generated (Fig. 3). The similar codon usage, especially for the cellulases, indicates that most of the genes from both species originated from the same ancestor. In contrast, codon usage of the transposase genes is very different from that of the cellulase genes (Fig. 3). The clustering of transposases in three branches distinct from most of the cellulases suggests that they were recently introduced by HGT in contrast to practically all of the cellulases that have evolved slowly and are presumably restricted to the Fibrobacter genus. Differences in the G+C composition (G+C%) and G+C composition at the third codon (G+C3%) of genes compared to the host genome are also used as evidence for HGT (10). The mean and standard deviation values for G+C% and G+C3% for the transposase gene segments were 49.6% ± 0.1% and 65.7% ± 0.1%, respectively, whereas the values for the carbohydrate active enzyme gene segments were 49.6% ± 0.3% and 61.6% ± 0.1%, respectively. As a basis for these comparisons the mean values of the total G+C% and G+C3% of the SSH genes were 48.6% ± 3.7% and 56.9% ± 8.4%, respectively. From these calculations it can be seen that the G+C% value for the transposases was similar to that of the carbohydrate active enzymes and all of the SSH genes, although the variation of the latter was greater, as indicated by the greater standard deviation. Similarly, the G+C3% values for the transposases were higher and the variations were greater than for both the carbohydrate active enzymes and the SSH genes. However, at least half of the gene segments of the carbohydrate active enzymes showed as great a variation as that of the transposases, which indicates that these criteria are of limited value, at least for the F. intestinalis genome. However, the fact that BLAST searches showed that the transposases and restriction-modification enzymes have the highest similarities to proteins in other unrelated organisms and no similarity to proteins of F. succinogenes (Table 2) is supportive evidence of HGT. However, due to the lack of genetic information within the Fibrobacter genus, we cannot tell whether the differences among the genes are due to HGT into F. intestinalis or deletion from the genome of F. succinogenes. Further study is needed to assess the distribution of the specific genes among other phylogenetically related species. This conclusion is consistent with the complexity observed in previous studies to determine whether evolving differences in bacteria are due to HGT or gene loss (14, 35).



View larger version (43K):
[in this window]
[in a new window]
  		FIG. 3. Dendrogram of the relationships between the codon usage of F. succinogenes cellulolytic enzymes and F. intestinalis cellulolytic enzymes and transposases. Clones of SSH from F. intestinalis DR7 were named as RSGA plus the number for cellulolytic enzymes and Tra plus the number for transposases. The transposases showing the highest similarity to the ISPg7 transposase of P. gingivalis are represented by Tra201 and Tra532 because of their nearly identical DNA sequences. Genes from F. succinogenes were labeled with ORF numbers. Most of the genes from F. succinogenes were clustered in regions A and B as shown in the figures. Region A contains 33 F. succinogenes genes and includes ORFs 106, 267, 924, 1077, 1107, 1635, 1685, 1708, 1766, 2210, 2280, 2282, 2283, 2284, 2321, 2325, 2552, 2754, 2861, 2862, 3023, 3555, 3598, 3766, 3767, 3815, 3983, 4138, 4190, 4208, 4500, 4614, and A00451. Region B contains 41 F. succinogenes genes and includes ORFs 65, 143, 241, 279, 369, 601, 925, 990, 1079, 1086, 1088, 1258, 1538, 1785, 1918, 2003, 2115, 2519, 2520, 2686, 2715, 2748, 3025, 3444, 3674, 3685, 3687, 3690, 3711, 3713, 3720, 3726, 3727, 4081, 4082, 4341, 4367, 4534, 4628, A00235, and A00267.

Vast and diverse communities of microbes colonize the gastrointestinal tracts of animals and are essential to the nutrition of the animal. Despite the difficulty in resolving the roles of HGT and gene loss in the evolution of genomes of bacteria in the intestinal environment, HGT among these organisms would seem very likely due to the high bacteriophage numbers (40) and close proximity of organisms in these densely populated ecosystems (8). As a result, it may not be a surprise that 45% of the genes identified by SSH had highest similarities to the genes of other gastrointestinal bacteria. Among them, 15 genes (Table 2 and unpublished data) that exist in F. intestinalis showed greatest similarity to genes from the ubiquitous intestinal bacterium B. thetaiotaomicron (41).

F. succinogenes strain S85 has been the model organism for studying the mechanism of efficient cellulose degradation within the genus. SSH between the genomes of F. intestinalis and F. succinogenes has demonstrated that, although most of the divergent genes including those coding for cellulolytic enzymes still have a high degree of conservation, there is extensive genome rearrangement in F. intestinalis relative to F. succinogenes. Furthermore, we have identified novel genes in F. intestinalis whose presence may be attributed to the numerous IS elements, different restriction-modification systems in F. intestinalis and perhaps HGT. These data provide a fascinating snapshot of the evolving interrelationships of F. intestinalis with F. succinogenes, and with other microorganisms in the gastrointestinal tract, that provides a solid foundation for further studies of the genomics and physiology of fibrolytic organisms in this unique ecosystem.


arrow
ACKNOWLEDGMENTS
 
This research was supported by U.S. Department of Agriculture grants for the North American Consortium for Genomics of Fibrolytic Ruminal Bacteria (grant 00-52100-9618), CSREES, IFAFS, and the Natural Science and Engineering Research Council of Canada.

We thank Lewis Lukens for advice on Bioinformatics and Reggie Lo for proofreading the manuscript.


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: Department of Molecular and Cellular Biology, University of Guelph, Guelph, N1G 2W1 Ontario, Canada. Phone: (519) 824-4120, ext. 53433. Fax: (519) 837-1802. E-mail: cforsber{at}uoguelph.ca. Back

{dagger} Supplemental material for this article may be found at http://jb.asm.org/. Back


arrow
REFERENCES
 
    1
  1. Ahn, S. J., J. Costa, and J. R. Emanuel. 1996. PicoGreen quantitation of DNA: effective evaluation of samples pre- or post-PCR. Nucleic Acids Res. 24:2623-2625.[Free Full Text]
    2. 2
  2. Akopyants, N. S., A. Fradkov, L. Diatchenko, J. E. Hill, P. D. Siebert, S. A. Lukyanov, E. D. Sverdlov, and D. E. Berg. 1998. PCR-based subtractive hybridization and differences in gene content among strains of Helicobacter pylori. Proc. Natl. Acad. Sci. USA 95:13108-13113.[Abstract/Free Full Text]
    3. 3
  3. Amann, R. I., C. H. Lin, R. Key, L. Montgomery, and D. A. Stahl. 1992. Diversity among Fibrobacter isolates: toward a phylogenetic classification. Syst. Appl. Microbiol. 15:23-31.
    4. 4
  4. Antonopoulos, D. A., K. E. Nelson, M. Morrison, and B. A. White. 2004. Strain-specific genomic regions of Ruminococcus flavefaciens FD-1 as revealed by combinatorial random-phase genome sequencing and suppressive subtractive hybridization. Environ. Microbiol. 6:335-346.[CrossRef][Medline]
    5. 5
  5. Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Sidman, J. A. Smith, and K. Struhl. 1994. Current protocols in molecular biology, p. 2.4.1-2.4.5. Greene Publishing Associates/John Wiley & Sons, Inc., New York, N.Y.
    6. 6
  6. Bera-Maillet, C., Y. Ribot, and E. Forano. 2004. Fiber-degrading systems of different strains of the genus Fibrobacter. Appl. Environ. Microbiol. 70:2172-2179.[Abstract/Free Full Text]
    7. 7
  7. Beylot, M. H., V. A. McKie, A. G. Voragen, C. H. Doeswijk-Voragen, and H. J. Gilbert. 2001. The Pseudomonas cellulosa glycoside hydrolase family 51 arabinofuranosidase exhibits wide substrate specificity. Biochem. J. 358:607-614.[CrossRef][Medline]
    8. 8
  8. Flint, H. J. 1994. Molecular genetics of obligate anaerobes from the rumen. FEMS Microbiol. Lett. 121:259-267.[CrossRef][Medline]
    9. 9
  9. Forsberg, C. W., E. Forano, and A. Chesson. 2000. Microbial adherence to plant cell wall and enzymatic hydrolysis, p. 79-88. In P. B. Cronje (ed.), Ruminant physiology digestion, metabolism, growth, and reproduction. CABI Publishing, Wallingford, Oxon, United Kingdom.
    10. 10
  10. Garcia-Vallve, S., A. Romeu, and J. Palau. 2000. Horizontal gene transfer of glycosyl hydrolases of the rumen fungi. Mol. Biol. Evol. 17:352-361.[Abstract/Free Full Text]
    11. 11
  11. Handa, N., Y. Nakayama, M. Sadykov, and I. Kobayashi. 2001. Experimental genome evolution: large-scale genome rearrangements associated with resistance to replacement of a chromosomal restriction-modification gene complex. Mol. Microbiol. 40:932-940.[CrossRef][Medline]
    12. 12
  12. Hughes, D. 2000. Evaluating genome dynamics: the constraints on rearrangements within bacterial genomes. Genome Biol. 1:REVIEWS0006.[Medline]
    13. 13
  13. Jedrzejas, M. J. 2000. Structure, function, and evolution of phosphoglycerate mutases: comparison with fructose-2,6-bisphosphatase, acid phosphatase, and alkaline phosphatase. Prog. Biophys. Mol. Biol. 73:263-287.[CrossRef][Medline]
    14. 14
  14. Jeltsch, A. 2003. Maintenance of species identity and controlling speciation of bacteria: a new function for restriction/modification systems? Gene 317:13-16.[CrossRef][Medline]
    15. 15
  15. Kobayashi, I. 2001. Behavior of restriction-modification systems as selfish mobile elements and their impact on genome evolution. Nucleic Acids Res. 29:3742-3756.[Abstract/Free Full Text]
    16. 16
  16. Krafft, T., and J. M. Macy. 1998. Purification and characterization of the respiratory arsenate reductase of Chrysiogenes arsenatis. Eur. J. Biochem. 255:647-653.[Medline]
    17. 17
  17. Lee, S. F., C. W. Forsberg, and A. M. Gibbins. 1992. Type II DNA restriction-modification system and an endonuclease from the ruminal bacterium Fibrobacter succinogenes S85. J. Bacteriol. 174:5275-5283.[Abstract/Free Full Text]
    18. 18
  18. Lin, C., and D. A. Stahl. 1995. Comparative analyses reveal a highly conserved endoglucanase in the cellulolytic genus Fibrobacter. J. Bacteriol. 177:2543-2549.[Abstract/Free Full Text]
    19. 19
  19. Lin, C. Z., B. Flesher, W. C. Capman, R. I. Amann, and D. A. Stahl. 1994. Taxon specific hybridization probes for fiber-digesting bacteria suggest novel gut-associated Fibrobacter. Syst. Appl. Microbiol. 17:418-424.
    20. 20
  20. Lynd, L. R., P. J. Weimer, W. H. van Zyl, and I. S. Pretorius. 2002. Microbial cellulose utilization: fundamentals and biotechnology. Microbiol. Mol. Biol. Rev. 66:506-577.[Abstract/Free Full Text]
    21. 21
  21. Malburg, L. M., Jr., A. H. Iyo, and C. W. Forsberg. 1996. A novel family 9 endoglucanase gene (celD), whose product cleaves substrates mainly to glucose, and its adjacent upstream homolog (celE) from Fibrobacter succinogenes S85. Appl. Environ. Microbiol. 62:898-906.[Abstract]
    22. 22
  22. Malburg, S. R., L. M. Malburg, Jr., T. Liu, A. H. Iyo, and C. W. Forsberg. 1997. Catalytic properties of the cellulose-binding endoglucanase F from Fibrobacter succinogenes S85. Appl. Environ. Microbiol. 63:2449-2453.[Abstract]
    23. 23
  23. Matheron, C., A. M. Delort, G. Gaudet, and E. Forano. 1998. In vivo 13C NMR study of glucose and cellobiose metabolism by four cellulolytic strains of the genus Fibrobacter. Biodegradation 9:451-461.[CrossRef][Medline]
    24. 24
  24. McGavin, M., and C. W. Forsberg. 1988. Isolation and characterization of endoglucanases 1 and 2 from Bacteroides succinogenes S85. J. Bacteriol. 170:2914-2922.[Abstract/Free Full Text]
    25. 25
  25. McGavin, M. J., C. W. Forsberg, B. Crosby, A. W. Bell, D. Dignard, and D. Y. Thomas. 1989. Structure of the cel-3 gene from Fibrobacter succinogenes S85 and characteristics of the encoded gene product, endoglucanase 3. J. Bacteriol. 171:5587-5595.[Abstract/Free Full Text]
    26. 26
  26. McInerney, J. O. 1998. GCUA: general codon usage analysis. Bioinformatics 14:372-373.[Abstract/Free Full Text]
    27. 27
  27. Mira, A., L. Klasson, and S. G. Andersson. 2002. Microbial genome evolution: sources of variability. Curr. Opin. Microbiol. 5:506-512.[CrossRef][Medline]
    28. 28
  28. Miron, J., and D. Ben Ghedalia. 1993. Digestion of cell-wall monosaccharides of ryegrass and alfalfa hays by the ruminal bacteria Fibrobacter succinogenes and Butyrivibrio fibrisolvens. Can. J. Microbiol. 39:780-786.[Medline]
    29. 29
  29. Miron, J., and C. W. Forsberg. 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]
    30. 30
  30. Montgomery, L., B. Flesher, and D. Stahl. 1988. Transfer of Bacteroides succinogenes (Hungate) to Fibrobacter gen. nov. as Fibrobacter succinogenes comb. nov. and description of Fibrobacter intestinalis sp. nov. Int. J. Syst. Bacteriol. 38:430-435.[Abstract/Free Full Text]
    31. 31
  31. Morrison, M., K. E. Nelson, I. K. O. Cann, C. W. Forsberg, R. I. Mackie, J. B. Russell, B. A. White, K. Amava, B. Cheng, M. Qi, H. Jun, S. Mulligan, K. Tran, H. A. Carty, H. Khouri, W. Nelson, S. Daugherty, and C. M. Fraser. 2003. The Fibrobacter succinogenes strain S85 genome sequencing project. Abstr. 3rd ASM-TIGR Conference of Microbial Genomes, p. 33.
    32. 32
  32. Nesbø, C. L., K. E. Nelson, and W. F. Doolittle. 2002. Suppressive subtractive hybridization detects extensive genomic diversity in Thermotoga maritima. J. Bacteriol. 184:4475-4488.[Abstract/Free Full Text]
    33. 33
  33. Ochman, H., J. G. Lawrence, and E. A. Groisman. 2000. Lateral gene transfer and the nature of bacterial innovation. Nature 405:299-304.[CrossRef][Medline]
    34. 34
  34. Olsen, G. J., C. R. Woese, and R. Overbeek. 1994. The winds of (evolutionary) change: breathing new life into microbiology. J. Bacteriol. 176:1-6.[Free Full Text]
    35. 35
  35. Philippe, H., and C. J. Douady. 2003. Horizontal gene transfer and phylogenetics. Curr. Opin. Microbiol. 6:498-505.[CrossRef][Medline]
    36. 36
  36. Roberts, R. J., and D. Macelis. 2001. REBASE-restriction enzymes and methylases. Nucleic Acids Res. 29:268-269.[Abstract/Free Full Text]
    37. 37
  37. Ross, R., X. L. Ross, B. Krueger, T. Lenin, and A. B. Reeks-Kunz. 1999. No radioactive detection of differentially expressed genes using complex RNA or DNA hybridization probes. BioTechniques 26:150-155.[Medline]
    38. 38
  38. Salanoubat, M., S. Genin, F. Artiguenave, J. Gouzy, S. Mangenot, M. Arlat, A. Billault, P. Brottier, J. C. Camus, L. Cattolico, M. Chandler, N. Choisne, C. Claudel-Renard, S. Cunnac, N. Demange, C. Gaspin, M. Lavie, A. Moisan, C. Robert, W. Saurin, T. Schiex, P. Siguier, P. Thebault, M. Whalen, P. Wincker, M. Levy, J. Weissenbach, and C. A. Boucher. 2002. Genome sequence of the plant pathogen Ralstonia solanacearum. Nature 415:497-502.[CrossRef][Medline]
    39. 39
  39. Scott, H. W., and B. A. Dehority. 1965. Vitamin requirements of several cellulolytic rumen bacteria. J. Bacteriol. 89:1169-1175.[Abstract/Free Full Text]
    40. 40
  40. Swain, R. A., J. V. Nolan, and A. V. Klieve. 1996. Natural variability and diurnal fluctuations within the bacteriophage population of the rumen. Appl. Environ. Microbiol. 62:994-997.[Abstract]
    41. 41
  41. Xu, J., M. K. Bjursell, J. Himrod, S. Deng, L. K. Carmichael, H. C. Chiang, L. V. Hooper, and J. I. Gordon. 2003. A genomic view of the human-Bacteroides thetaiotaomicron symbiosis. Science 299:2074-2076.[Abstract/Free Full Text]
    42. 42
  42. Ziemer, C. J., R. Sharp, M. D. Stern, M. A. Cotta, T. R. Whitehead, and D. A. Stahl. 2000. Comparison of microbial populations in model and natural rumens using 16S rRNA-targeted probes. Environ. Microbiol. 2:632-643.[CrossRef][Medline]

Journal of Bacteriology, June 2005, p. 3739-3751, Vol. 187, No. 11
0021-9193/05/$08.00+0     doi:10.1128/JB.187.11.3739-3751.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.



This article has been cited by other articles:

  • Toyoda, A., Iio, W., Mitsumori, M., Minato, H. (2009). Isolation and Identification of Cellulose-Binding Proteins from Sheep Rumen Contents. Appl. Environ. Microbiol. 75: 1667-1673 [Abstract] [Full Text]  
  • Alavi, S. M., Sanjari, S., Durand, F., Brin, C., Manceau, C., Poussier, S. (2008). Assessment of the Genetic Diversity of Xanthomonas axonopodis pv. phaseoli and Xanthomonas fuscans subsp. fuscans as a Basis To Identify Putative Pathogenicity Genes and a Type III Secretion System of the SPI-1 Family by Multiple Suppression Subtractive Hybridizations. Appl. Environ. Microbiol. 74: 3295-3301 [Abstract] [Full Text]  
  • Qi, M., Nelson, K. E., Daugherty, S. C., Nelson, W. C., Hance, I. R., Morrison, M., Forsberg, C. W. (2008). Genomic Differences between Fibrobacter succinogenes S85 and Fibrobacter intestinalis DR7, Identified by Suppression Subtractive Hybridization. Appl. Environ. Microbiol. 74: 987-993 [Abstract] [Full Text]  
  • Jun, H.-S., Qi, M., Gong, J., Egbosimba, E. E., Forsberg, C. W. (2007). Outer Membrane Proteins of Fibrobacter succinogenes with Potential Roles in Adhesion to Cellulose and in Cellulose Digestion. J. Bacteriol. 189: 6806-6815 [Abstract] [Full Text]  
  • Qi, M., Jun, H.-S., Forsberg, C. W. (2007). Characterization and Synergistic Interactions of Fibrobacter succinogenes Glycoside Hydrolases. Appl. Environ. Microbiol. 73: 6098-6105 [Abstract] [Full Text]  

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental material
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Qi, M.
Right arrow Articles by Forsberg, C. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Qi, M.
Right arrow Articles by Forsberg, C. W.

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»