Transcription of Clostridium cellulovorans Cellulosomal Cellulase and Hemicellulase Genes

Bacterial strains and growth conditions. C. cellulovorans ATCC 35296 was used as the source of genomic DNA and total RNA. The organism was grown under strictly anaerobic conditions at 37°C in 500- or 3,000-ml round-bottom flasks containing a previously described medium (21, 23), which included 0.5% (wt/vol) of several carbon sources (cellulose [Avicel], cellobiose, locust bean gum, pectin, and xylan [oat spelt or birchwood]).

Nucleic acid isolation.Chromosomal DNA of C. cellulovorans was isolated by using the Promega genomic DNA purification kit according to the manufacturers' instructions. Total RNA was extracted from C. cellulovorans broth cultures with the Qiagen RNeasy kit (Qiagen), with an additional step of treatment with RNAlater solution (Ambion) and RNase-free DNase (Promega) according to the manufacturers' instructions.

Northern hybridization.RNA samples (up to 20 μg) were denatured in RNA sample buffer, which consisted of 150 μl of formamide, 83 μl of 37% (wt/vol) formaldehyde, 83 μl of Promega blue/orange loading dye, and 50 μl of 10× MOPS (morpholinepropanesulfonic acid) buffer (20 mM MOPS, 5 mM sodium acetate, 1 mM EDTA [pH 7.0], 34 μl of distilled water) at 65°C for 10 min and separated through 1% agarose gels in MOPS buffer with 17% (vol/vol) formaldehyde. DNA probes were synthesized by PCR with specific oligonucleotides derived from the C. cellulovorans sequence as a template (Table 1). The probes were nonradioactively labeled by random priming with digoxigenin (DIG)-High Prime (Roche). RNA was transferred overnight to a positively charged nylon membrane (Roche) by capillary transfer with 20× SSC (3 M NaCl, 300 mM trisodium citrate [pH 7.0]). Hybridization was carried out for 16 to 20 h at 50°C in DIG Eazy Hyb buffer solution (Roche). Washing of the membrane and detection of specific transcripts on the blots were carried out with the Roche DIG luminescent detection kit according to the kit's protocol.

RT-PCR and primer extension analysis.Reverse transcriptase (RT) reactions were performed on total RNA with a commercially available reverse transcription system (Promega) with slight modifications to the recommended protocol. RT reactions were performed in a final volume of 20 μl, which contained 5 mM MgCl₂, RT buffer (10 mM Tris-HCl [pH 9.0], 50 mM KCl, 0.1% Triton X-100), 1 mM (each) deoxynucleoside triphosphates, 1 U of recombinant RNasin RNase inhibitor, 15 U of avian myeloblastosis virus (AMV) RT, 0.25 μM oligonucleotide primer, and 10 μg of substrate RNA. Reaction mixtures were incubated at 42°C for 60 min, and reactions were terminated by heating the mixtures at 95°C for 5 min, followed by incubation on ice for 5 min. The cDNA products were then amplified in 25-μl PCR mixtures with 2.5 μl of the RT reaction mixture as the template.

Primer extension reactions were carried out with AMV RT according to the manufacturer's instructions (Promega). IRDye700 or IRDye800-labeled oligonucleotides were obtained from Li-cor and were used to determine a transcription start point (Table 1). The annealing of 1 pmol of each primer was performed with approximately 10 to 100 μg of total C. cellulovorans RNA.

S1 nuclease protection analysis.The labeled fragments were prepared by PCR amplification with the template DNA and reverse oligonucleotide (complementary to the region about 100 bp downstream of the 5′ end of the proposed coding region) labeled at the 5′ end by IRDye700 or IRDye800 (Li-cor), and unlabeled direct oligonucleotide about 500 bp upstream of the 5′ end of the proposed coding region (Table 1). Equal amounts of RNA (50 μg) were hybridized with aliquots of the single-stranded labeled DNA probe. The aliquots of single-stranded labeled DNA probe were annealed to mRNA in 50 μl of a hybridization reaction (0.3 M NaCl, 1 mM EDTA, 38 mM HEPES [pH 7.0], 0.01% Triton X-100) with 50 μg of total RNA for 18 h at 55°C. Hybridization products were digested for 30 min at 37°C by the addition of 50 U of S1 nuclease (Invitrogen) and 450 ml of buffer containing 330 mM NaCl, 66 mM sodium acetate (pH 4.6), 2.2 mM ZnSO₄, and 0.01% Triton X-100. S1 nuclease digestions were terminated by adding 30 μl of stop mixture (2.5 M sodium acetate, 40 mM EDTA, 3 μg of tRNA).

Nonradioactive determination of the transcription start point.The primer extension reaction and S1 nuclease digestion products were ethanol precipitated and analyzed on a 4% polyacrylamide sequencing gel running on a Li-cor model 4200L DNA sequencer (Li-cor) as instructed by the manufacturer. The length of each of the loading samples was determined by running sequence reactions with the same primers on the same gel. The sequencing reactions were performed with the SequiTherm cycle sequencing kit for Li-cor (Epicentre) by using a thermocycler.

Transcriptional analysis of the cellulase gene cluster.The tandem organization of the cellulase and hemicellulase genes in the cbpA gene cluster (Fig. 1A) allowed us to predict that the genes would be transcribed as mono- or polycistronic mRNAs. To examine this prediction and to find the lengths of the transcripts, the biosynthesis of mRNA under various conditions was measured by Northern hybridization and RT-PCR analyses. We isolated total RNA from C. cellulovorans cells grown in medium containing cellulose, cellobiose, locust bean gum, pectin, or xylan as the sole energy source. Although usually the same concentration of RNA was used for comparative purposes in these experiments, the equal labeling of probes was problematic, since the primer sequences were different, and therefore the labeling of the probes was not identical. These experiments primarily demonstrated the sizes of the transcripts and the relative locations of the genes.

• Open in new tab
• Download powerpoint

FIG. 1.

Northern hybridization analyses of cellulase and hemicellulase mRNA. (A) The cbpA gene cluster. Total RNA was isolated from C. cellulovorans grown in the presence of cellulose (lanes 1 and 3 to 7), cellobiose (lane 2), xylan (lane 8), or pectin (lane 9) as the sole carbon source. RNA (10 μg) was subjected to electrophoresis through 1.5% formaldehyde gels and transferred to nylon membranes, which were subsequently probed with the DIG-labeled probes specific for cbpA (lanes 1 and 2), engH (lane 3), hbpA (lane 4), manA (lane 5), engM (lane 6), engE (lane 7), xynA (lane 8), and pelA (lane 9). The oblong circles represent full-length specific transcripts. The size of RNA markers (M) is indicated to the left in bases.

In Northern blot analyses, membranes containing RNA were hybridized with intragenic probes derived from the cellulase and hemicellulase genes. In these experiments, we used the same amount of RNA during the tests for comparative purposes. With the cbpA or exgS probes, a prominent band corresponding to an 8-kb transcript was detected (Fig. 1B, lanes 1 and 2). Also a weak additional hybridizing band 12 kb in length was detected under all conditions. The levels of both the 8- and 12-kb transcripts were enhanced when cellulose (Avicel) was included in the medium (Fig. 1B, lanes 1 and 2). In addition, the 12-kb transcript was detected when total RNA was hybridized with an engH internal probe (Fig. 1B, lane 3). The prominent shorter 8-kb band may represent mRNA processing of the full-length 12-kb cbpA-exgS-engH-engK transcript and not an independent transcript, since no transcription termination signal was evident. No additional promoters could be detected downstream of the cbpA, exgS, and engH promoter regions (see next section).

To determine whether 12-kb RNA processing occurred, RNAs prepared at different time points after the addition of rifampin were analyzed by Northern hybridization. If processing occurred following the termination of transcription by addition of rifampin, then the amount of the 12-kb transcript should decrease, and the amount of the 8-kb transcript should increase over time. However, both the 8- and 12-kb transcripts decreased together after the addition of rifampin (Fig. 2). This indicated strongly that the 8-kb transcript was not a processed product of the longer 12-kb message and that transcription termination could yield either an 8-kb product or a longer run-through product of 12 kb. The 8-kb mRNA was large enough to code for CbpA and ExgS. The 12-kb mRNA coded for CbpA, ExgS, EngH, and EngK (see the summary of these results in Fig. 6).

• Open in new tab
• Download powerpoint

FIG. 2.

Northern hybridization analysis showing decay of cbpA-specific transcripts. Total RNA was isolated from C. cellulovorans grown to the middle of the exponential phase in the presence of cellulose (Avicel) as the sole carbon source at 0, 2, 4, 6, 8, and 10 min after the addition of rifampin (100 μg/ml). The two cbpA-specific transcripts (8 and 12 kb) were examined by hybridizing 8 μg of each sample with an exgS-specific probe.

The RT-PCR analysis was also performed with the primers EngH-S1MP-R and EngH-DIG-R (Table 1), which were specific for identifying the putative exgS-engH transcript. The mRNA products from the RT-PCR analysis were found 112 and 1,044 nucleotides (nt) downstream of the engH start codon, respectively (Fig. 3A and B). The RT-dependent products indicated that the exgS and engH genes were transcriptionally coupled (Fig. 3B, lanes 1 and 2). The amplified products were similar to that seen in the control reaction with chromosomal DNA as a template with the same PCR primers. A negative control using the RNA that was not reverse transcribed was included to ensure that there was no contaminating DNA present in the RNA preparation (data not shown). These results, summarized in Fig. 3A suggested that the cbpA and exgS genes were in an operon and that the transcription of this operon could continue through engH to the engK gene (see the summary in Fig. 6).

• Open in new tab
• Download powerpoint

FIG. 3.

RT-PCR analysis of engH, hbpA, engM, and xynA RNA. (A) Schematic showing of PCR-amplified transcripts. The locations and extents of the exgS, engH, engK, hbpA, engL, manA, engM, and xynA genes are shown. The oligonucleotide primers used in RT-PCR analysis are numbered as follows: 1, EngH-S1MP-F2 and EngH-S1MP-R; 2, EngH-DIG-F and EngH-DIG-R; 3, HbpA-S1MP-F2 and HbpA-S1MP-R; 4, HbpA-DIG-F and HbpA-DIG-R; 5, EngM-S1MP-F2 and EngM-S1MP-R; 6, EngM-DIG-F and EngM-DIG-R; 7, XynA-S1MP-F2 and XynA-S1MP-R; and 8, XynA-DIG-F and XynA-DIG-R. (For more details, see Table 1.) The locations and sizes of the PCR-amplified transcripts are indicated by the bars. The black and white bars represent the transcription status shown in the presence and absence of PCR amplification, respectively. (B) Agarose gel electrophoresis of RT-PCR products. RT-PCR analysis was performed with total RNA isolated from C. cellulovorans in the medium containing cellulose (Avicel). In negative controls, the reactions were performed in the absence of RT or RNA template (data not shown).

Probes that consisted of the regions of hbpA and engM hybridized to 6.0- and 2.7-kb mRNA species, respectively (Fig. 1B, lanes 4 and 6). However, it was difficult by Northern blot analyses to probe for the hbpA and engM genes, since the apparent low level of transcription of these genes was sometimes below the level of detection by the Northern blot assay and/or the RNA transcripts were unstable. Therefore, the alternative method of RT-PCR analysis was performed with specific primers such as HbpA-S1MP-R and EngM-S1MP-R for the detection of the putative hbpA and engM transcripts.

The resultant cDNA molecules were then amplified by PCR with primers HbpA-S1MP-F2 and HbpA-S1MP-R, which produced a product of 560 bp (Fig. 3B, lane 3). With primers HbpA-DIG-F and HbpA-DIG-R, the cDNA molecules were also amplified and produced a product of 648 bp (Fig. 3B, lane 4). The oligonucleotide HbpA-S1MP-F2 binds within engK, 498 bp upstream of the hbpA start codon (Fig. 3A, lane 3). These RT-dependent PCR products were observed, indicating that the engK and hbpA genes were linked (see summary in Fig. 6). To determine whether a promoter was present upstream of the hbpA gene, primer extension analysis was carried out with RNA from cellulose and cellobiose cultures by using primer HbpA-S1MP-R, which was complementary to a region 62 nt downstream of the initiation codon of hbpA. Contrary to the RT-PCR results, a transcriptional start point was detected (as given in more detail in the next section). These results indicated that the engK and hbpA genes formed an operon with a promoter in front of the engK gene as well as an internal promoter proximal to the hbpA gene (see summary in Fig. 6).

In confirmation of the Northern blot analysis (Fig. 1B, lane 6), RT-PCR detected the engM transcripts (Fig. 3B, lane 6). Upon analysis of the upstream region of EngM by RT-PCR, no cDNA products were detected (Fig. 3B, lane 5). This suggested that engM was transcribed as a monocistronic gene.

Northern hybridizations of the RNA with a manA probe showed a single transcript of 1.8 kb (Fig. 1B, lane 5). This mRNA was in good agreement with the size of the manA gene (1,275 nt) and indicated that manA was a monocistronic gene within the cbpA gene cluster.

Transcriptional analysis of cellulosomal genes unlinked to the cbpA cluster.Transcriptional analyses were carried out with cellulosomal genes that were not linked with the cbpA gene cluster. This included engE, xynA, and pelA. The results of Northern analysis with an intragenic probe derived from engE showed that engE formed a transcript of about 3 kb under all conditions (Fig. 1B, lane 7). This mRNA size was in good agreement with the size of the engE gene (3,039 nt) and indicated that engE was a monocistronic gene.

The transcript that included the xynA gene was 2.7 kb in length and was found to be expressed constitutively (Fig. 1B, lane 8). Two additional hybridizing bands 2.3 and 1.6 kb in length for xynA were also detected. These shorter bands probably represented mRNA processing of the full-length xynA transcript or transcription termination events (i.e., putative terminators were found downstream of xynA at 1,693 and 2,335 nt), since no cDNA could be detected upstream of the xynA region (Fig. 3B, lane 7); in addition, xynA mRNA could be detected with an internal probe (Fig. 3B, lane 8). The 1.6-kb mRNA corresponded well in size with the xynA gene (1,560 nt) (Fig. 1B, lane 8).

When C. cellulovorans was grown with pectin as a carbon source, a unique transcript of about 4 kb was detected with a probe internal to the pelA gene (Fig. 1B, lane 9). The longest size of the transcript (4 kb) indicated that it may be part of an operon, because the size of the pelA gene is 2,745 kb.

Determination of transcription start points and promoters for cellulase genes.In order to localize the promoters, the transcription start sites of selected cellulase genes were determined by primer extension analysis and high-resolution S1 nuclease protection analysis. Our initial studies focused on the promoter location for cbpA, the first gene in the cbpA gene cluster. An 866-bp PCR fragment (CbpA-S1MP-F/CbpA-S1MP-R1) was generated for use as a probe with 512 bp complementary to the upstream and potential promoter region of cbpA. This probe was annealed to total RNA isolated from C. cellulovorans, cultivated with cellulose or cellobiose as the sole carbon source. One clear signal in position bp −233, relative to the cbpA start codon, was obtained by both S1 nuclease protection analysis and primer extension analysis (Fig. 4). In addition, this transcription start point was confirmed by S1 nuclease protection analysis and primer extension analysis with two additional oligonucleotides (CbpA-S1MP-F/CbpA-S1MP-R2 and CbpA-S1MP-F/CbpA-S1MP-R3) (Fig. 5). The cbpA mRNA start point suggested a putative promoter region, TAAAAAN₁₇TATTAT, with high similarity to the σ^A consensus promoter sequences of gram-negative (σ⁷⁰) and -positive (σ^A) bacteria (Table 2). The relatively long leader sequence for the cbpA operon suggests that some type of control may be exerted during translation of his operon. Similar experiments were carried out with cellulosomal genes engH, hbpA, manA, engM, engE, and xynA (data not shown). The results are summarized in Fig. 6.

• Open in new tab
• Download powerpoint

FIG. 4.

S1 nuclease mapping (S1) and primer extension (PE) analysis of cbpA. Shown are S1 and primer extension reactions of total RNA isolated from C. cellulovorans separated in 4% polyacrylamide sequencing gels alongside sequencing reactions to determine the transcription start point of the cbpA gene. The transcription start point is indicated by the bent black arrow.

• Open in new tab
• Download powerpoint

FIG. 5.

Location of putative promoters in the cbpA upstream region. The transcription start point (TSP; +1) is indicated by the bent black arrow. The putative promoter regions (consensus −10 and −35 regions) are underlined. A putative ribosome binding site (RBS) and the first codon (Met) are designated. The positions of the oligonucleotide primers CbpA-S1MP-R1, CbpA-S1MP-R2, and CbpA-S1MP-R3, which were used for primer extension analyses and S1 nuclease mapping analyses, are shown in boxes. A palindrome is indicated by arrows facing each other above the sequence. Coordinates refer to the published sequence in reference 22 (GenBank accession no. M73817 ).

• Open in new tab
• Download powerpoint

FIG. 6.

Locations of mRNA among the cellulase and hemicellulase genes. The black arrows correspond to the transcripts of the genes. The bent black arrows represent the transcription start points and also indicate that the transcription start points were found by S1 nuclease mapping and/or primer extension analysis. The bent white arrows indicate the putative transcription start point according to Northern blot analysis and DNA sequence analysis. The stem and loops indicate transcription termination points.

The transcript of manA was initiated at a guanine residue 64 bp upstream of the translational start site. Two major transcription start sites, corresponding to positions 61 (more intense band) and 63 bp upstream of the hbpA start codon, respectively, were detected, indicating that hbpA could be transcribed independently of engK. The engE transcriptional start point was detected at a single guanine located 97 bases upstream of the ATG start codon.

The results in Fig. 4 and 5 typify the results obtained for the cbpA promoter, and similar results were obtained for all of the genes analyzed. The results for the other genes are summarized in Fig. 6 and Table 2.

Comparison of the promoter sequences.The C. cellulovorans promoter sequences at −10 (5′-TAGAAA-3′) and −35 (5′-TATAAT-3) present in the transcription start sites of the gene cluster in C. cellulovorans are homologous to elements recognized by Bacillus subtilis (σ^A) and Escherichia coli (σ⁷⁰) RNA polymerases (33) (Table 2). However, when primer extension and S1 nuclease analyses were performed to determine the presence of engH, engM, and xynA promoters, none could be detected by both types of analyses. Hybridizing bands by Northern blot analysis were observed irregularly for both the engM and xynA probes (data not shown), suggesting that the RNA transcripts were unstable or present in very small amounts. Although primer extension and S1 nuclease analyses did not detect a transcript start point, RT-PCR analysis indicated that both the engM (Fig. 3B, lanes 5 and 6) and xynA (Fig. 3B, lanes 7 and 8) genes were transcribed from promoters adjacent to their 5′ ends.