ABSTRACT
The spo0B gene, which exists as an operon with the obg gene, is required to initiate sporulation (stage 0) of Bacillus subtilis. This gene encodes a phosphotransferase in the multicomponent phosphorelay system. We here report the novel finding that a spo0B 5'-terminal SLR (stem-loop structure sequestering ribosome binding sequence; ACUCCUAA-X16-UUGGGAGU, [Delta]G = -8.71 kcal/mol) attenuated spo0B translation. The spo0B gene was efficiently transcribed but Spo0B protein was not overproduced in Escherichia coli when spo0B was induced using expression vectors carrying the SLR-spo0B region under control of the tac promoter. Deletion of the SLR from the vectors resulted in overexpression of spo0B. Therefore, to characterize expression of spo0B with a SLR in B.subtilits we constructed transcriptional and translational lacZ fusions combined with the spo0B 5'-terminal region with a deleted or mutagenized SLR. These constructs were subsequently introduced into B.subtilis as multiple and single copies, then [beta]-galactosidase activities were measured. The possible SLR also functioned as a negative cis element in B.subtilis. Furthermore, B.subtilis strain 1S16 (spo0B136) lysogenized [Phi]CD0B-S and -W, harboring spo0B with mutagenized SLRs that were more ([Delta]G = -14.0 kcal/mol) and less-stable ([Delta]G = -1.31 kcal/mol) compared with the wild-type, exhibited null and wild-type sporulation respectively. These results indicate that the spo0B 5'-SLR affects spo0B gene expression for sporulation but that low expression of spo0B through the wild-type SLR was sufficient to initiate sporulation in B.subtilis.
Gene expression is generally controlled at the transcriptional, post-transcriptional, translational and post-translational levels. In prokaryotes initiation of translation involves recognition by 16S rRNA in the ribosome 30S subunit at the Shine-Dalgarno (SD) sequence (1), which lies near the start codon of a coding sequence. The values of free energy ([Delta]G) for an SD sequence and secondary structure in mRNA can be calculated by the method of Tinoco et al. (2). Translational efficiency mainly depends on: (i) SD strength; (ii) spacing between the SD sequence and the initiation codon; (iii) the type of initiation codon; (iv) codon usage; (v) secondary structure of the mRNA (3,4). The secondary structure of mRNA is different in different mRNAs and can be classified by location within the mRNA: (i) in the 5'-terminal region, as for the UTR of the ompA gene (5); (ii) in the structural gene, as for the stacking region of the nprM gene (6); (iii) in the 3'-terminal region, as for the REP sequence of the gdhA gene (7). These secondary structures may affect mRNA stability or translational efficiency, which may be important for gene expression as well as regulation at the transcriptional level.
The morphological and biochemical events of sporulation in Bacillus subtilis are induced by deprivation of nutrients in the medium. A large number of sporulation genes (spo) are sequentially expressed during sporulation (8). Initiation of sporulation is controlled by at least nine spo0 genes, termed spo0A, 0B, 0E, 0F, 0H, 0J, 0K, 0L (= rapA)and 0P (= rapB) (9). Among them, the spo0B gene has been cloned (10) and sequenced (11). The spo0B gene is preferentially expressed during exponential growth from a single promoter; it is not induced by exhaustion of the growth medium nor repressed by glucose. The spo0B-obg locus consists of an operon and the spo0B transcript extends through the obg gene and ends at a terminator located in the 3'-downstream region of the obg gene (12). The obg gene is essential for cell growth and encodes a GTP binding protein, which is homologous to those of the RAS family, and Obg function may be involved in sensing intracellular GTP levels prior to sporulation (13). In sporulation initiation in B.subtilis, the multicomponent phosphorelay system (14) may receive environmental signals as follows. At first a sensor protein KinA (= SpoIIJ) autophosphorylates at a histidine residue and transfers the phosphate to an aspartate residue in Spo0F. Phosphorylated Spo0F then delivers the phosphate to Spo0B and phosphorylated Spo0B passes it on to the Spo0A protein. Phosphorylated Spo0A (15,16) binds effectively to a specific DNA sequence (0A box) and activates or represses transcription from relevant genes (17-19). In this study a cis sequence was found in the 5'-terminal stem-loop structure of spo0B and spo0B gene expression on deletion or mutagenization of the structure was investigated in E.coli and B.subtilis.
Escherichia coli strains were cultivated at 37°C in 2× TY medium (20). Bacillus subtilis strains were cultured at 37°C in L Broth (21) or the sporulation medium of Schaeffer et al. (22).
An 857 bp region of the BstNI-Sau96I fragment carrying the entire spo0B coding sequence with its 5'-upstream region (Fig. 1) was isolated from the genomic DNA of phage [rho]11pspo0B-1 (10), filled in with Klenow enzyme and joined to EcoRI linkers. This EcoRI fragment was inserted into the EcoRI site located downstream of the tac promoter in E.coli expression vector pKK223-3 (23) to yield pKB27. The spo0B promoter region was deleted with exonuclease III (24) as follows. pUBX7 DNA (25) carrying the spo0B gene and its 5'-upstream region was digested with BamHI and KpnI, then incubated with exonuclease III. The deleted plasmids were filled-in and self-ligated with Klenow enzyme and T4 DNA ligase, respectively, to create a series of pDB plasmids. Each EcoRI-XbaI small fragment was isolated from the respective pDB plasmid and replaced the EcoRI-XbaI fragment in pKBH3 (25), derived from pKB27, to yield the new expression vectors pEXB252, 224, 205, 183, 170, 152 and 139, respectively. For studies of spo0B-lacZ fusions (Fig. 4) plasmids P21-pSZB152 and 170 were constructed from pSZ-1 with an ~550 bp BamHI-HindIII fragment of pEXB170 and 152, respectively (25). Plasmids P21-pKZB152 and 170 were also created from pKZ701, as for P21-pSZB152 and 170 (25). An 870 bp XbaI fragment carrying SLR-spo0B was isolated from pKB27 and inserted into the XbaI site of pUC119 (26) to create pUB87R. The spo0B-lacZ fusions pKLB122, 122-S and 122-W were obtained from pKZ724 with an ~270 bp XbaI-PstI fragment from pUB87R, 87-S and 87-W, respectively (see Site-directed mutagenesis below). To construct plasmids for prophage transformation (Fig. 5) an ~870 bp XbaI fragment was isolated from pUB87R, 87-S and 87-W, then inserted into the XbaI site of pCD301 (27) to create pCD0B, 0B-S and 0B-W, respectively.
To overproduce and detect the Spo0B protein in E.coli the host JM105 strain harboring expression vector pEXB was cultivated in 100 ml 2× TY liquid medium containing ampicillin (50 µg/ml) at 37°C. When the turbidity of cell cultures reached 50 Klett units (OD600 ~ 0.5) isopropyl-[beta]-d-thiogalactopyranoside (IPTG) was added to the culture at a final concentration of 1 mM. Five hours after adding IPTG, 1 ml cell culture was harvested (another 10 ml were collected for RNA preparation; see below) and proteins were resolved by 14% SDS-PAGE (28) to determine if the 24 kDa protein (Spo0B) was overexpressed or not.
Total RNA was prepared from E.coli JM105 harboring each pEXB plasmid as described by Gilman and Chamberlin (29). RNA was spotted onto a Hybond-N membrane (Amersham), then dot blot hybridization was performed at 42°C as described by Okamoto et al. (30). A 323 bp PvuII fragment (Fig. 1) derived from pUBX7, which contains part of the spo0B gene, was labeled with digoxigenin-dUTP using a random primed DNA labeling kit (Boehringer), then used as a spo0B-specific probe.
Site-directed mutagenesis was performed according to Kunkel (31) using a kit purchased from TaKaRa Shuzo (Japan). Oligonucleotides 5'-TACAGACTCCCAAATAAGAAA-3' (21mer) for a strong (stable) SLR and 5'-TATACAGACTGCTAAATAAGA-3' (21mer) for a weak (unstable) SLR were synthesized and annealed to a single-stranded DNA generated from pUB87R (25) carrying the wild-type SLR-spo0B. After producing homoduplex strands carrying each site-directed mutation we obtained plasmids pUB87R-S and pUB87R-W. The mutation on each plasmid was verified by DNA sequencing.
Bacillus subtilis PSL1 (or JH642) harboring each lacZ fusion plasmid was grown in L Broth (or Schaeffer sporulation) liquid medium supplemented with kanamycin at a final concentration of 5 µg/ml and growth was monitored by measuring absorbance at 600 nm. One milliliter of cell culture was withdrawn at 2 (or 1) h intervals during growth (Fig. 5). These samples were centrifuged and the pellets suspended in 0.7 ml Z buffer (32) containing a final concentration of 1 mM phenylmethylfluorosulfate, pH 7.0, and 0.25 mg/ml lysozyme. Seven microliters of 10% (v/v) Triton X-100 was added to the suspended cells and vigorously vortex mixed for 10 s. Orthonitrophenylgalactoside solution (0.2 ml) (32) was immediately added, then the mixture was incubated at 28°C for 20 min. The reaction was stopped by adding 0.4 ml 1 M Na2CO3 with stirring. The clear supernatant was collected by centrifugation for 2 min and absorbance was measured at 420 nm. [beta]-Galactosidase units are expressed according to the following equation (33): 1 [beta]-galactosidase unit = 1000 × A420 × 1/t (min) × 1/V (ml) × 1/A600, where t represents the time of enzymatic reaction (min), V is the volume of the cell culture (ml) and A600 reflects the cell density just before assay.
Prophage transformation proceeded according to Iijima et al. (34), with a slight modification (15). Briefly, each plasmid pCD0B, 0B-S and 0B-W was digested with restriction enzyme ScaI and the linearized DNA individually introduced into B.subtilis JH642 (wild-type) or 1S16 (deficient of spo0B) cells lysogenized with phage [Phi]ED411 (27). Chloramphenicol (5 µg/ml)-resistant and erythromycin (5 µg/ml)-sensitive transformants integrated as phages [Phi]CD0B, 0B-S, and 0B-W were selected respectively (Fig. 5).
Recombinant DNA techniques were performed as described by Sambrook et al. (26). DNA was sequenced according to Sanger et al. (35). The stem-loop structure was analyzed using a program (2,36) of the GENETYX software suite from SDC Co. Ltd (Japan).
It is known that the spo0B gene is preferentially expressed during exponential growth of B.subtilis (11) and its gene product is required for sporulation. Because of its unique and essential role for spo0B overexpression in E.coli, a spo0B 5'-upstream cis element has recently been the focus of intensive research (25). To overproduce and purify the Spo0B protein from E.coli an expression vector, pKB27, which carries the entire spo0B coding sequence with the 5'-upstream region was constructed. However, the spo0B gene did not overexpress in host E.coli JM105 (37) harboring pKB27 (Fig. 2). We therefore considered that the 5'-upstream region of spo0B possesses a negative cis element for overexpression. Consequently, the spo0B 5'-upstream region in pKB27 was deleted and a series of new expression vectors, pEXB252, 224, 205, 183, 170, 152 and 139, was reconstructed (Fig. 1; Materials and Methods). These pEXB plasmids were introduced into E.coli JM105 cells and we again examined whether or not the spo0B gene was overexpressed in the host strains. A 24 kDa (Spo0B) protein was overproduced in host E.coli cells harboring pEXB152 and pEXB139 (Fig. 2A). In contrast, overproduction of Spo0B was not observed in cells carrying other pEXB vectors and the gel profiles from pEXB252, 224, 205, 183 and 170 seem to exhibit the same pattern as that from vector pKK223-3 (no insert). The copy numbers and stabilities of these pEXB plasmids in host cells were the same (data not shown). We also assessed the quantities of spo0B transcript synthesized from pEXB and pKK223-3, respectively, by dot blot hybridization (Fig. 2B). When we used total RNA prepared from cells carrying pKK223-3, no signal corresponding to spo0B mRNA appeared in the hybridization. In contrast, positive signals were observed from cells harboring pEXB224, 170 and 152 (as well as 252, 205, 183 and 139; data not shown) and their signal intensities were almost the same (Fig. 2B). These results indicated that spo0B was transcribed in the tac promoter-operator system (23) from each expression vector, but that translation was hindered when there was a negative cis sequence in the untranslated leader region of spo0B. Therefore, overexpression of spo0B from pEXB152 and 139 mightoccur on deletion of the negative sequence.
To investigate the factor(s) affecting spo0B gene expression at the translational (or post-transcriptional) level in E.coli we analyzed the secondary structure on the spo0B untranslated leader region. A possible stem-loop structure (ACUCCUAA-X16-UUGGGAGU, [Delta]G = -8.71 kcal/mol) appeared in the mRNA at the spo0B 5'-terminal region from +15 to +46 (Fig. 3). This base paired region includes the ribosome binding sequence (GGAG) in the stem of the stem-loop structure. Therefore, this potential mRNA secondary structure was termed a SLR (stem-loop structure sequestering ribosome binding sequence). spo0B overexpression (Fig. 2) suggested that the SLR reduced the efficiency of spo0B translation in E.coli. To clarify whether or not the SLR also attenuates spo0B translation in B.subtilis we analyzed spo0B expression through a deleted or mutagenized SLR in B.subtilis using lacZ fusion vectors as follows.
The upper panel of Figure 4A shows a schematic representation of the lacZ fusion vectors pSZ-1 and pKZ701 and their derivatives carrying the 5'-region of the spo0B gene. These are shuttle vectors (25,27) with which gene expression at the transcriptional and translational levels respectively can be analyzed. For example, P21-pSZB152 and P21-pSZB170 are transcriptional fusions and have a ribosome binding site (SD) for the alkaline serine protease gene (aprE) of B.subtilis (38). Since these vectors also have the P21-promoter of the 21K gene (39,40) which is constitutively expressed in B.subtilis, transcription of spo0B from the P21 promoter causes synthesis of lacZ fusion protein, which uses an SD of the aprE gene that is independent of the SLR effect. Therefore, [beta]-galactosidase activity and spo0B transcriptional expression from P21-pSZB152 and P21-pSZB170 must be at the same level. On the other hand, P21-pKZB170 and P21-pKZB152 as translational fusions have only the SD of the spo0B gene and the 5'-terminal region of the spo0B gene carrying the P21 promoter (without the spo0B promoter) fused to the lacZ gene in-frame. Thus [beta]-galactosidase activity and Spo0B translational expression from P21-pKZB170 may be lower than that of P21-pKZB152, since synthesis of the fused protein (Spo0B-LacZ) from P21-pKZB170 will be reduced by the SLR. The spo0B gene is preferentially expressed during the exponential phase of cell growth in B.subtilis and is neither induced by exhaustion of the growth medium nor repressed by glucose (11). To analyze regulation of expression of the spo0B gene under exponential growth conditions we cultivated host B.subtilis PSL-1 harboring these vectors in L Broth liquid medium, which is abundant in nutrients. The lower panel of Figure 4A shows that the [beta]-galactosidase activities of P21-pSZB170 and P21-pSZB152 are similar, but that of P21-pKZB170 is lower than that of P21-pKZB152. These results support the notion that spo0B expression also involves translational attenuation by the SLR in B.subtilis.
For further studies of spo0B gene expression with the SLR we constructed spo0B-lacZ fusions pKLB122-S and pKLB122-W, carrying a mutagenized SLR (Fig. 4B, upper panel). These plasmids were created from shuttle vector pKZ724 for the translational lacZ fusion. Plasmid pKLB122 contains an insert carrying a part of the spo0B 5'-region and its upstream region with its own promoter. Thus the N-terminal 37 amino acids of Spo0B were fused in-frame with [beta]-galactosidase. Plasmid pKLB122 has a wild-type SLR ([Delta]G = -8.71 kcal/mol), whereas pKLB122-W and pKLB122-S plasmids have mutagenized SLRs in which the sequence is changed from C (+18) to G (for a weak SLR, [Delta]G = -1.31 kcal/mol) and from U (+20) to C (for a strong SLR, [Delta]G = -14.0 kcal/mol), respectively, in the stem region (Fig. 3). The wild-type strain of B.subtilis JH642 (spo0B+) harboring these plasmids was cultivated in Schaeffer's sporulation liquid medium and [beta]-galactosidase activity measured in the cells after various incubation periods (stages of cell growth from T-1 to T3; Fig. 4B, lower panel). The spo0B gene was maximally translated from each plasmid from the exponentional growth to the onset of sporulation phase (t0). [beta]-Galactosidase activity from pKLB122-S was lower than that from pKLB122. In contrast, the level of [beta]-galactisidase activity from pKLB112-W was significantly higher than that from pKLB122. We therefore concluded that the potential spo0B 5'-SLR directly affects spo0B translation efficiency. When the free energy of the SLR increased, the secondary structure of the SLR might have become more stable, resulting in decreased translational efficiency.
In the above lacZ fusion studies using transcriptional and translational multicopy plasmids, translational attenuation by the spo0B 5'-SLR was demonstrated in B.subtilis. Generally in bacteria there are occasional titration effects of gene dosage upon gene expression in trans when a target gene on a multicopy vector is introduced into a host cell. To avoid this and to better understand the effect of the spo0B 5'-SLR on B.subtilis chromosomal DNA we performed prophage transformations, which transported the target gene to the genome as a single copy (Fig. 5). At first we constructed plasmids pCD0B, 0B-S and 0B-W, which bear spo0B and its 5'-upstream regions carrying the wild-type, strong and weak SLRs, respectively. The plasmid DNA was then introduced into B.subtilis JH642 (wild-type, spo0B+) or 1S16 (deficient of spo0B, spo0B136) cells lysogenized with phage [Phi]ED411 by prophage transformation (Fig. 5). The sporulation efficiencies of these lysogens are shown in Table 1. The sporulation abilities of JH642 (spo0B+) lysogenized with [Phi]CD0B-W and [Phi]CD0B-S were normal, suggesting that a mutagenized SLR has no effect upon sporulation in the wild-type strain (spo0B+). Although 1S16 (spo0B136) lysozenized with [Phi]CD0B or [Phi]CDB-W recovered normal sporulation ability, 1S16 (spo0B136) lysogenized with [Phi]CDB-S remained sporulation deficient. These results indicate that spo0B expression is reduced to a level which is not sufficient to support sporulation by the strong SLR. However, the 1S16 (spo0B136, [Phi]CD0B-W) and JH642 (spo0B+, [Phi]CD0B-W) strains exhibited wild-type sporulation, demonstrating that the level of spo0B expression through the unstable (weak) SLR did not hinder spoluration. Naturally, the low level of spo0B expression through the wild-type SLR was sufficient for sporulation initiation in B.subtilis.
The level of expression of spo0B is relatively low during exponential cell growth in B.subilits (11). Bouvier et al. speculated that low expression is caused by regulation as follows: (i) the spo0B promoter has a consensus sequence at -10 (TATAAT) but an unusual sequence at -35 (TTTTCT) and this sequence, which shows little homology, results in insufficient interaction between RNA polymerase holoenzyme and the DNA strand during transcription initiation; (ii) an insufficient interaction of spo0B mRNA with the ribosome.
This study revealed a possible cis element (SLR) associated with translational attenuation located in the untranslated leader region of the spo0B-obg operon transcript. The [Delta]G value for the mRNA secondary structure of the SLR is -8.71 kcal/mol (Fig. 3). When the [Delta]G value for the mRNA secondary structure located in an SD sequence and its flanking regions is higher (more negative) than ~-5.0 kcal/mol the efficiency of translation initiation drops markedly to 30% or less of full expression (3,41). Furthermore, B.subtilis is less able to tolerate a secondary structure in the SD sequence than E.coli (4). According to the rules of Tinoco et al. the free energy ([Delta]G value) of binding of the spo0B SD sequence (GGAG) to B.subtilis 16S rRNA (5'-UCACCUCCUUUCU-3') is -9.0 kcal/mol, which is not high compared with average values (-14 to -16 kcal/mol) in Gram-positive bacteria like B.subtilis (42). Based on these free energy values the strength is strong SLR (-14.0 kcal/mol) > SD (-9.0) > wild-type SLR (-8.71) > weak SLR (-1.31). If the whole structure of the SLR is more stable than the GGAG pairing of the SD sequence, as is the case for the strong SLR, translation will be strongly hindered by attenuation. In the case of the wild-type SLR this value is slightly lower than that of the SD sequence, allowing low expression of spo0B through attenuation. On the other hand, in vivo data for B.subtilis showed the relative order of translational efficiency with the three initiation codons to be: AUG > UUG > GUG (4) and the general codon AUG is also used in the spo0B gene (Fig. 1). Therefore, spo0B attenuation might eventually correlate with the relationship of the positions and sequence spacing between the potential stem-loop and SD structures in the 5'-SLR.
From the gene positioning in the spo0B-obg operon (Fig. 1) it appears that obg translation is coupled to that of the upstream spo0B gene. If so, both genes will be repressed simultaneously by the 5'-SLR in B.subtilis. The mutagenized more stable spo0B 5'-SLR structure (strong) resulted in less spo0B translational expression than that with the wild-type SLR and resulted in a null phenotype of sporulation in B.subtilis (Table 1). This result revealed the biological importance of the possible 5'-SLR for spo0B gene expression. In this study only the mutagenized SLR-spo0B region without obg was introduced into the [Phi]105 prophage locus in the B.subtilis genome (Fig. 5), the site of which differs from that of the spo0B-obg locus. Therefore, whether expression from the spo0B-obg operon, which is essential for cell growth and encodes a GTP binding protein, with the mutagenized SLR affects cell growth or sporulation remains unresolved.
Table 1.
To further characterize the spo0B 5'-SLR the structure can be compared with those of other stem-loops located in untranslated leader transcript or leader peptide coding sequences. Regarding lamB gene expression in E.coli, the 5'-terminal stem-loop structure contains a SD sequence and the start codon and a mutagenized more stable stem structure causes low expression of the gene (43). This phenomenon implies a functional role for the stem-loop structure and reinforces the idea that the presence of a stem-loop structure containing an SD sequence might result in translational attenuation. We also found a potential stem-loop structure somewhat similar to the spo0B 5'-SLR located in the 5'-terminal mRNA of the spoIIIG gene, which encodes a forespore-specific [sigma] factor termed [sigma]G in B.subtilis (44). This transcript appears at ~T0.5-1 of sporulation (0.5-1 h from the end of exponential cell growth), but synthesis of the [sigma]G protein is undetectable until T2-3, due to possible translational attenuation by the stem-loop structure (45,46). This fact suggests that there is another translational attenuation of gene expression during the late stage of sporulation in B.subtilis. Two other types of stem-loop structure affect mRNA stability and mRNA binding regulatory protein in B.subtilis. In the former, ribosome stalling in the ermC leader peptide coding sequence causes a 10- to 15-fold increase in ermC mRNA half-life (47,48). In the latter, the tryptophan operon (trpEDCFBA) is regulated by transcription attenuation involving two stem-loop structures in the leader transcript, of which the sequence is a target for TRAP protein (mtrB gene product), which can induce transcription termination (49). In the spo0B 5'-SLR the untranslated leader sequence (52 nt, Fig. 4), including the SLR, is much shorter than those of ermC and trpEDCFBA and no typical leader peptide and direct repeat sequences were found.
The loop region (+23 to +38) in the spo0B 5'-SLR is AU-rich and only two guanines appear in 16 nt (Fig. 3). Interestingly, spo0B overexpression from pEXB152 carrying a half length AU-rich sequence was much higher than that from pEXB139 without this sequence (Fig. 2). Because the deleted first G in the SD sequence (Figs 1 and 3) is replaced by the G from the vector sequence at the junction site in pEXB139, the complete SD sequence, GGAG, of spo0B is maintained in expression vector pEXB139. The region from +30 to +41, containing part of the AU-rich sequence, may therefore be an effective cis element for spo0B overexpression from pEXB152. A U-rich sequence in the 5'-terminal region is known to be required for translation of prokaryotic mRNA (50) and a specific hairpin sequence is associated with RNA hairpin stability (51).
This work was supported by a Grant-in-Aid for Scientific Research on Priority Areas from the Ministry of Education, Science, Sports and Culture of Japan.
Nucleic Acids Research
Pages
Introduction
Materials And Methods
Culture and media
Plasmid construction
Overexpression of spo0B
RNA preparation and dot blot hybridization
Site-directed mutagenesis
[beta]-Galactosidase assay
Prophage transformation
Other techniques
Results
spo0B expression in E.coli
Secondary structure of the SLR in spo0B mRNA
Transcriptional and translational spo0B expression without a SLR in B.subtilis
spo0B expression through a mutagenized SLR in B.subtilis
Effects of modulation of spo0B expression by the SLR upon sporulation
Discussion
Translational attenuation by the spo0B 5'-SLR in B.subtilis
Implications for the spo0B 5'-SLR
Acknowledgement
References
Strain
Cells/ml
Spores/ml
JH642 ([Phi]CD301)
3.6 × 108
4.1 × 108
JH642 ([Phi]CD0B)
6.8 × 108
2.2 × 108
JH642 ([Phi]CD0B-S)
3.5 × 108
3.3 × 108
JH642 ([Phi]CD0B-W)
8.1 × 108
6.0 × 108
1S16 ([Phi]CD301)
5.3 × 108
<10
1S16 ([Phi]CD0B)
6.8 × 108
4.4 × 108
1S16 ([Phi]CD0B-S)
2.4 × 108
<10
1S16 ([Phi]CD0B-W)
5.6 × 108
3.9 × 108
REFERENCES
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 23 Jan 1998
Copyright© Oxford University Press, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  H. Geng, S. Nakano, and M. M. Nakano Transcriptional Activation by Bacillus subtilis ResD: Tandem Binding to Target Elements and Phosphorylation-Dependent and -Independent Transcriptional Activation J. Bacteriol., April 1, 2004; 186(7): 2028 - 2037. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||