INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

The Rhodobacter capsulatus photosynthetic apparatus is composed of three membrane-spanning photopigment protein complexes known as the reaction center and the B875 light-harvesting-I (LH-I) and B800-850 light-harvesting-II (LH-II) antenna complexes (15, 56). The light-harvesting complexes are responsible for absorption of most of the light energy that irradiates the cells. The absorbed energy is subsequently passed to the reaction center, which donates an electron from bacteriochlorophyll to ubiquinone. Structural studies indicate that the reaction center is surrounded by a pigmented ring composed of 16 LH-I  and  polypeptide pairs (18). The reaction center-LH-I core complex is, in turn, associated with a variable number of LH-II rings composed of nine  and  polypeptide pairs (28, 31, 56).

The  and  structural proteins of the LH-II antenna complex are encoded by the pucBA gene pair that has been found in all photosynthetic purple nonsulfur bacteria that have been examined (56). Organization of this gene pair on the chromosome, the size of the intergenic region between the pucB and pucA genes, the proteins they encode, and their regulation are highly conserved among purple nonsulfur bacteria (2, 4, 10, 19, 24, 36, 56). The R. capsulatus puc operon contains three additional open reading frames, pucC, pucD, and pucE (43, 45). PucC is thought to be involved in assembly of the LH-II ring (23, 57), but the roles of PucD and PucE have not yet been established (23, 45).

Northern blot analysis has demonstrated the presence of several puc transcripts including a 2.4-kb segment coding for pucBACDE, a 1.0-kb segment composed of a mixture of pucC and pucDE transcripts that are thought to be generated as processing products from the larger pucBACDE transcript, and a 0.55-kb pucBA segment (5, 23, 44). A predicted secondary structure is located in the pucA-pucC intercistronic region, and it has been suggested that the short pucBA transcript may be generated by differential decay of the larger pucBACDE transcript (24, 44, 57). This presumably occurs in a manner analogous to the well-characterized mRNA processing events that generate a stable pufBA mRNA segment that codes for the LH-I structural polypeptides (6, 20, 21, 23, 54, 55).

Given the numerous studies that have been undertaken on the regulation of puc expression, it is surprising that there is relatively little information regarding the R. capsulatus puc promoter. To date, neither high-resolution mRNA mapping or detailed promoter deletion studies have been undertaken on the R. capsulatus puc promoter. In contrast, detailed studies on the puc promoter from Rhodobacter sphaeroides have indicated that oxygen and light control of promoter activity involve a complex set of cis-acting regulatory sites, some of which are located at extended distances (up to 629 bases) from the start of transcription (25-27). Clearly, comparable studies on the R. capsulatus puc promoter are warranted, and indeed necessary, now that several transcription factors responsible for controlling puc expression in response to oxygen have been identified and isolated from this species (4, 32, 35, 41).

As a prelude to performing DNA binding studies of transcription factors to the puc promoter region, we need to have a good understanding of the location of transcription initiation as well as of the functional length of the puc promoter region. Consequently, this study describes the high-resolution mapping of the start site of puc expression coupled with detailed in vitro and in vivo mapping of the puc promoter. Our results indicate that the R. capsulatus puc operon has a single transcriptional start site located 116 bp upstream from the initiation codon of pucB and that the putative puc promoter exhibits a ⁷⁰-type sequence motif that is similar to that of the carotenoid (crt) and bacteriochlorophyll (bch) promoters (1, 3, 30, 50). Functional length of the puc promoter region was also analyzed by both deletion analysis and gel mobility shift analysis using crude extracts derived from various regulatory mutants that are known to affect puc expression.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Bacterial strains and cultures. All bacterial strains used in this study are listed in Table 1. R. capsulatus strains were routinely grown at 34°C in peptone-yeast extract (PY) broth or agar medium (52), whereas Escherichia coli strains were routinely grown at 37°C in Luria broth (LB) or agar medium (38). For maintenance of plasmids, spectinomycin or kanamycin was added to media at a concentration of 10 µg/ml for R. capsulatus. For E. coli, spectinomycin, kanamycin, and rifampin were used at a concentration of 100 µg/ml, whereas ampicillin was used at a concentration of 200 µg/ml.

DNA manipulations and promoter probe analysis. Descriptions of plasmids used in this study are listed in Table 2. Standard recombinant DNA techniques were performed with E. coli DH5 as a host (38). Restriction enzymes and T4 DNA ligase were purchased from New England BioLabs, Inc. The promoter probe vector pZM400 was used to make transcriptional fusions of the puc upstream regulatory sequence (pucURS) to a promoterless lacZ gene (30). Details are given in Table 2.

Strain constructions. Mobilization of plasmids from E. coli DH5 to R. capsulatus was accomplished by triparental mating with E. coli HB101 harboring the helper plasmid pRK2013, which conjugates plasmids containing an RK2 origin of replication (14, 30). Donor and helper strains were grown with antibiotic selection overnight, centrifuged, resuspended in fresh LB with a 1:10 dilution, and incubated for 2 h before mating, while recipient strains were grown overnight in PY medium. Spot matings were performed on the surface of plates of slightly dry RCV medium (49) for 2 h of incubation followed by soft agar overlay containing appropriate antibiotics. After a 48-h incubation, transconjugants were purified by repeated streaking on RCV medium supplemented with 10 µg of spectinomycin/ml.

-Galactosidase assays. Cell extracts were prepared and assayed for -galactosidase activity as described by Young et al. (52). Protein concentrations were determined by Bradford assay (9). Final results are reported as the amount of o-nitrophenyl--galactoside (ONPG) hydrolyzed per minute per milligram of total protein. Reported -galactosidase values all had standard deviations of 6.2%.

Dideoxynucleic acid sequencing. puc operon subclones in M13mp18 and M13mp19 were used to generate single-stranded DNA (ssDNA) templates in both orientations which were used for DNA sequence analysis (38) (Table 2). Plasmid pRPSLHIIKAN was digested with ClaI and then blunt ended with DNA polymerase I Klenow fragment, followed by digestion with EcoRI, releasing a 2-kb EcoRI-ClaI fragment that includes the pucURS (53). The EcoRI-ClaI fragment was subcloned into the EcoRI-AccI sites of the M13 phage vectors. DNA sequence was obtained by using the Sanger dideoxynucleotide chain termination method with modified T7 DNA polymerase (Sequenase; U.S. Biochemical Corp.) in combination with the nucleotide analog 7-deaza-guanidine added to reaction mixes. DNA sequence data were analyzed and assembled using programs from the GCG Sequence Analysis Package of the University of Wisconsin Genetics Computer Group. All oligonucleotides used for DNA sequencing are shown in Fig. 1.

View larger version (63K): [in this window] [in a new window]   FIG. 1.   Primers used to analyze the puc promoter region. Arrows located above the strand indicate primers identical to the strand shown, whereas primers located below the strand are complementary. Dotted lines indicate nonhomologous sequences added for cloning with the following scheme: PCR1, GCGGTAACC; PCR2, GCGCATGC; PCR3, GCTACGTA; PCR4, GCGGGCC; PCR5, GCGCATG; PCR6, GCGGTAACC; PCR7, GCGCATGC; PCR8, GCGCA. Sequence numbers indicate location relative to the start of the pucB gene, which is numbered +1. Primers Puc1, Puc2, Puc3, and Puc4 were used for sequence, Northern blot, and primer extension analyses. Primer Puc5 was used for sequence and primer extension analyses, and Puc6 was used for sequence analysis. Primers Puc7, PucRev1, PucRev2, PucRev3, PucRev4, PucRev5, PucRev6, and PucRev7 were used for sequence and Northern blot analyses. Primers PCR1, PCR2, PCR3, PCR4, PCR5, PCR6, PCR7, and PCR8 were used as PCR primers.

Northern blot mRNA transcript analysis. Total RNA was isolated from R. capsulatus cells grown photosynthetically in 500-ml tissue culture flasks completely filled with PY medium by the guanidinium thiocyanate method of Chomczynski and Sacchi (12). Northern blot analyses were performed by size separating 30 µg of RNA on a 1.5% agarose gel containing 8.4% (vol/vol) formaldehyde. RNA was then blotted onto a nylon membrane (Nytran) by capillary transfer in 20× SSC buffer (1× SSC is 0.15 M NaCl plus 0.015 M sodium citrate) (38) and then UV cross-linked to the filter by exposure to 120 J of UV irradiation. Filters were hybridized for 16 h with the ³²P-labeled Puc1 oligonucleotide (Fig. 1) in a hybridization solution composed of 0.5 M NaH₂PO₄ (pH 7.2), 1 mM EDTA, 6.8% (wt/vol) sodium dodecyl sulfate (SDS), 10 mg of bovine serum albumin/ml, and 100 µg of sheared salmon sperm DNA/ml at 52°C (38). Following hybridization, filters were washed in 2× SSC-0.1% SDS at 52°C and exposed to X-ray film with an intensifying screen at 80°C. Lengths of puc mRNA fragments were estimated by ethidium bromide staining of Brome mosaic virus RNA ladders of 3.2, 2.9, 2.1, and 0.8 kb in size (obtained as a gift from Cheng Kao).

Primer extension analysis. Primer extension was performed as described in reference 13 with slight modifications. ³²P-end-labelled Puc1 oligonucleotide was hybridized with 125 µg of total RNA by heating for 10 min at 80°C and slowly cooling to 52°C. Samples were precipitated, resuspended in reverse transcriptase reaction buffer with 5 U of avian myeloblastosis virus-reverse transcriptase (Boehringer Mannheim), and incubated at 42°C for 2 h. After transcription the samples were precipitated and resuspended in standard denaturing gel electrophoresis buffer containing 10 mM NaOH, heated to 75°C for 2 min, and separated on urea denaturing 6% polyacrylamide gels. Dideoxynucleotide sequencing ladders were prepared with the ³²P-end-labelled Puc1 oligonucleotide, using the sequencing reactions as described above. Sequencing gels were dried and exposed to film with an intensifying screen at 80°C.

Gel mobility shift assays. Crude cell extracts were prepared as described by Ma et al. except that cells were lysed by sonication (30). Probes for gel mobility shift assays were prepared by PCR amplification with plasmid pRPSLH2KAN digested with EcoRI as a template (53) using 5'-³²P-end-labelled oligonucleotide primers. Probe I was amplified with the Puc4-PucRev3 oligonucleotide pair (Fig. 1), which amplifies a 242-bp fragment that includes the distal pucURS. Probe II was amplified with the PCR2-PucRev4 oligonucleotide pair (Fig. 1), which amplifies a 285-bp fragment that includes the proximal pucURS. Probe III was amplified with the Puc1-PucRev2 oligonucleotide pair (Fig. 1), which amplifies a 146-bp fragment that includes the pucP1 promoter and 68 bp upstream of the transcriptional start site. Cell extracts were diluted with the buffer used to prepare the extracts to a final concentration of 15 µg of protein/20 µl of reaction mix. A total of 1 µl of labelled DNA probe (2 to 4 fmol, ca. 5,000 cpm) and 2 µl (1 µg/µl) of poly(dI-dC) nonspecific competitor DNA at a 500-fold weight excess relative to the probe were added. Reaction mixtures were then incubated at 28°C for 30 min, loaded onto 4% polyacrylamide gels composed of 50 mM Tris-HCl (pH 8.3)-380 mM glycine-2 mM EDTA, and electrophoresed at 115 V for 2.5 h. Gels were dried and exposed to film with an intensifying screen at 80°C. Retention factors (R_f) were calculated for shifted probes according to the following formula: distance traveled by the shifted probe/distance traveled by the free probe.

Nucleotide sequence accession number. The sequence shown in Fig. 1 was submitted to the GenBank and EMBL databases under accession no. AF031407.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Sequencing of the pucURS. As a prelude to promoter mapping studies, we performed DNA sequence analysis of a 1,543-bp segment upstream of the pucB translational start codon (Fig. 1). Inspection of this sequence (Fig. 2) indicated the presence of (i) a potential ⁷⁰ promoter sequence motif spanning positions 123 to 152 relative to the start of pucB, (ii) two palindromes centered at positions 162 and 403 that contain a consensus DNA binding sequence (TGT-N₁₂-ACA) for the aerobic repressor CrtJ, and (iii) potential integration host factor (IHF) binding sites centered at positions 182, 491, 504, 517, 696, and 683 that exhibit sequence similarities to the R. capsulatus IHF consensus sequence (46-48). The results below address several of these features.

View larger version (45K): [in this window] [in a new window]   FIG. 2.   Features of the puc upstream regulatory region. (A) The predicted 35 and 10 regions of the pucP1 promoter are highlighted by black boxes. The transcriptional start site is identified with a thick, filled arrow. Thin, open arrows show the locations of palindromes associated with CrtJ binding. Unshaded boxes show consensus R. capsulatus IHF binding sites. (B) The pucP1 promoter region and alignments with 70-type promoters. The sequence at the top of the panel represents the first 58 nucleotides upstream of the pucP1 transcriptional start site with the 10 and 35 regions boxed. Alignments with 70-type promoters from purple photosynthetic bacteria as well as the E. coli consensus sequence are shown below. Asterisks indicate bases identical to those of the lacZ UV5 promoter, a cyclic AMP receptor protein-independent promoter from E. coli. Rb. cap., R. capsulatus; Rb. sph., R. sphaeroides; Rs. pal., Rhodopseudomonas palustris.

Northern blot analysis. Northern blot analysis was performed to define regions that may be transcribed with ³²P-labeled oligonucleotide primers designed to hybridize to specific regions upstream of pucB. As shown in Fig. 3A, hybridization with primer Puc1, which is complementary to a region 46 to 67 nucleotides upstream of the transcribed pucB start codon, hybridized to a 2.4-kb mRNA segment as well as to discrete 1.4-, 1.0-, and 0.5-kb mRNA segments. Longer exposures also indicate the presence of two additional weak signals of approximately 3.0 and 3.4 kb in size that hybridize to the Puc1 probe (33). This supports a previous report that the puc operon contains transcripts of various sizes that are formed as a result of processing (23). Northern blot analysis with oligonucleotides Puc2, Puc3, and Puc4, which were designed to hybridize 288 to 310, 356 to 386, and 618 to 637 bp upstream of the pucB gene, respectively (Fig. 1), failed to give rise to a signal (33). This indicates that initiation of puc transcription most likely occurs in the region bracketed by the Puc1 and Puc2 probes (from 46 to 310 bp upstream of pucB).

View larger version (55K): [in this window] [in a new window]   FIG. 3.   Northern blot and primer extension analyses of the puc operon transcript using 5'-32P-labeled Puc1 oligonucleotide as a probe. (A) Lane 1 is a control loaded with E. coli RFS859 RNA, while lane 2 contains RNA from R. capsulatus SB1003. Estimated sizes (in kilobases) of puc mRNA fragments are shown on the right. (B) High-resolution end mapping of the 5' terminus of the puc operon by primer extension analysis using 5'-32P-labeled Puc1 oligonucleotide as a primer. Individual reactions from a DNA sequencing ladder were loaded in the first four lanes. Lane 1 contains the product of a primer extension reaction with total RNA isolated from high-light photosynthetic cells, lane 2 contains the product formed from low-light photosynthetic cells, and lane 3 contains the product from a dark-aerobic-grown culture. The arrow indicates the 116 base signal proposed as the pucP1 transcriptional start site.

Primer extension analysis. Detection of a stable mRNA 5' end in the region thought to initiate the puc transcript (based on the Northern blot results) was undertaken by performing primer extension analysis using a ³²P-labeled Puc1 primer as a probe. As shown in Fig. 3B, a predominant primer extension product located 116 bp upstream of the start of pucB was obtained with mRNA from photosynthetically grown cells under high-light conditions (Fig. 3B, lane 1). Primer extension with mRNA obtained from low-light-grown cells had higher amounts of the 116-bp product (lane 2), whereas mRNA obtained from aerobically grown cells had a significant reduction in the amount of the primer extension product (lane 3). This correlates well with previous studies which indicated that transcription of the puc operon is light regulated under anaerobic conditions and repressed under aerobic growth conditions (3, 4, 11, 22, 23, 32, 34, 35, 41). Primer extension with oligonucleotides that hybridize >288 bp upstream of pucB (Puc2, Puc3, and Puc4) exhibited no primer extension signal, which supports the negative Northern blot results obtained with these oligonucleotides (33). We concluded therefore that puc transcription is predominantly initiated from a cytosine located 116 bp upstream of pucB.

In vivo promoter probe analysis. To determine if the observed primer extension signal represents an actual transcription initiation start site rather than an artifact of mRNA processing, various regions of the pucURS were cloned into a promoter probe plasmid that uses lacZ as a reporter of transcription activity. As shown in Fig. 4A, -galactosidase activity was observed with plasmid pDN12S, which contains a 272-bp fragment 49 to 320 bp upstream of the start codon of pucB, and plasmid pDN13S, which contains a segment that overlaps that of pDN12S as well as an additional 760 bp of upstream DNA sequence (from 49 to 1,080 bp upstream of pucB) (Fig. 4). No activity could be detected with plasmids pDN11S, pDN20S, pDN21S, and pDN24S, which contained segments spanning sequences from 173 to 1,341 bp upstream of pucB. The in vivo activities observed with these vectors thus localizes the initiation of puc transcription to a region between 49 and 173 bp upstream of pucB, which corresponds well with the 116 bp location of the primer extension product.

View larger version (20K): [in this window] [in a new window]   FIG. 4.   In vivo promoter probe analysis of the puc upstream regulatory region. (A) The organization of the 5.75-kb EcoRI fragment used to clone the puc operon is shown at top (53). Below this is shown the PstI-ClaI fragment used for many subsequent subclones. The bottom of the panel shows DNA-restricted and PCR-generated fragments that were cloned into pZM400 as well as the results of transcriptional assays under photosynthetic growth conditions. +, levels of -galactosidase activity significantly above that observed with the control reporter plasmid pZM500, which has no insert DNA; , -galactosidase activity equal to or below that observed with the control plasmid pZM500. (B) Plasmids tested for expression are shown on the x axis, and -galactosidase units (micromoles of ONPG hydrolyzed per minute per milligram of protein) are shown on the y axis. Results are shown for extracts prepared from dark-aerobic-grown cells (O2), photosynthetic high-light-grown cells (HL), and photosynthetic low-light-grown cells (LL).

Regulation of promoter reporter constructs pDN12 and pDN13. Previous studies by Lee and Kaplan indicated that the puc promoter in R. sphaeroides contains regulatory elements that are located up to 600 bp upstream of the location of transcription initiation (25, 26). Since pDN12S exhibited -galactosidase activity with only 57 bp of DNA upstream of the site of transcription initiation, we decided to undertake an analysis of expression of this construct in comparison with that of plasmid pDN13S, which contains 964 bp of DNA upstream of the transcription start site. As shown in Fig. 4B, wild-type cells harboring either plasmid pDN12S or pDN13S exhibited similar patterns of expression under the tested conditions. However, pDN12S had a slightly lower level of expression than pDN13S when grown under both photosynthetic high-light and low-light conditions. There was also a slightly higher level of expression of pDN12S than of pDN13S when both constructs were grown under aerobic growth conditions. This indicates that pDN12S contains many, but not all, of the critical cis-acting sites needed for regulating puc expression.

View larger version (18K): [in this window] [in a new window]   FIG. 5.   Results of -galactosidase assays with plasmid pDN13S (pucP1) in wild-type (W.T.) and regulatory mutants of R. capsulatus. Strains tested are shown on the x axis, and -galactosidase activities (micromoles of ONPG hydrolyzed per minute per milligram of protein) are shown on the y axis. (A) Results with extracts prepared from high-light-grown cells. (B) Results with extracts prepared from low-light grown cells. (C) Results with extracts prepared from dark-aerobic-grown cells.

Gel mobility shift analysis. We also undertook in vitro gel mobility shift analysis with cell extracts to assay for the presence of transcription factors that may bind to the puc promoter region. As shown in Fig. 6, extracts derived from wild-type strain SB1003 retard probe I, which contains the distal upstream regulatory region from 618 to 860 bp upstream of pucB, to two different R_f positions of 0.81 and 0.84 (with the free probe having an R_f of 1.0) (Fig. 6B, lane 2). Probe II, which contains the proximal upstream regulatory region from 306 to 591 bp upstream of pucB, shifts to three different R_f positions of 0.87, 0.74, and 0.43 (Fig. 6B, lane 6). Probe III, which contains the pucP1 promoter region from 46 to 192 bp upstream of pucB, exhibits three shifts with R_fs of 0.80, 0.72, and 0.66 (Fig. 6B, lane 10).

View larger version (63K): [in this window] [in a new window]   FIG. 6.   Gel mobility shift analysis of the pucP1 promoter and upstream regulatory region. (A) Probes used for gel shift analysis. See text for details. (B) Gel mobility shifts are shown immediately below the respective probe that was used for each experiment with no protein (lanes 1, 5, 9) and with cell extracts derived from SB1003 (lanes 2, 6, and 10), IR4 rif10 (IHF) (lanes 3, 7, and 11), and TB1 (RegA) (lanes 4, 8, and 12) cells. Because each of the probes were run on different gels, the Rfs observed with individual probes are independent of each other.

View larger version (69K): [in this window] [in a new window]   FIG. 7.   Gel mobility shift analysis of the pucP1 promoter and upstream regulatory region with purified R. capsulatus IHF. (A) Probes used for gel shift analysis. (B) Gel shifts shown immediately below the respective probe used for each experiment with no protein (lanes 1, 4, and 7), 10.8 nM IHF (1X) (lanes 2, 5, and 8), and 21.6 nM IHF (2X) (lanes 3, 6, and 9).

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

Previous low-resolution S1 mapping studies indicated that the puc 5' mRNA terminus was located 110 to 125 bp upstream of the pucB translational start codon (23). This is in good agreement with the results of our high-resolution in vitro primer extension (Fig. 3B) and in vivo promoter probe analyses (Fig. 4), which demonstrate that a single puc operon transcription start site occurs 116 bp upstream of the pucB translational start codon. Analysis of the pucP1 promoter region (Fig. 2B) shows that it contains a ⁷⁰-type promoter consensus (TTGAtc-N₁₇-cATAgT) that has a high degree of sequence similarity with other proposed or identified promoters from R. capsulatus that are involved in carotenoid and bacteriochlorophyll biosynthesis (1, 30). Further inspection reveals strong similarities between the pucP1 35 and 10 regions and similar regions in the cyclic AMP receptor protein-independent lacZ UV5 promoter from E. coli, i.e., 16 of 20 nucleotides identical around the 35 region and 11 of 14 nucleotides identical around the 10 region (see bases highlighted with asterisks in Fig. 2B). A significant difference is that these regions are spaced 6 nucleotides closer in lacZ UV5 than is the case for pucP1 (37).

Previous studies using pucB::lacZ translational fusions gave results consistent with the degree of oxygen and light regulation reported here for the transcriptional vector pDN13S (11, 41). Specifically, we observed that transcription initiated from the pucP1 promoter increases two- to threefold when cultures are grown under low-light compared to high-light growth conditions, which correlates with the effect of light intensity on synthesis of the LH-II complex. These results contradict the results of Zucconi and Beatty (57), who concluded that puc mRNA levels vary inversely with LH-II protein levels with respect to changes in light intensity based on mRNA hybridization studies. However, our own primer extension (Fig. 3B) and Northern blot (33) experiments indicate that levels of puc mRNA are significantly greater under low-light compared to high-light growth conditions when equal amounts of total RNA were used. This would indicate that there is no substantial light-mediated posttranscriptional control of puc expression.

The results of our expression analysis with reporter plasmids in different regulatory mutant backgrounds, coupled with the in vitro gel mobility shift results, provide a low-resolution view of the complexity of the transcription factors that are responsible for controlling puc expression. Mutations in RegA and IHF had similar 2- to 2.5-fold reductions of puc expression under photosynthetic and aerobic-dark growth conditions (Fig. 5). Gel mobility shift patterns demonstrated that cell extracts derived from a regA-deleted strain affect a higher-order structure of proteins bound to a DNA segment (from 45 to 192 bp upstream of pucB) that codes for the pucP1 promoter region (Fig. 7). Indeed, this is confirmed by recent DNase I nuclease protection experiments with purified RegA, which demonstrate that RegA binds to a segment spanning from 52 to 80 bp upstream of the start of puc transcription initiation (16).

Cell extracts derived from the IHF mutant were also altered in mobility shifts in all three DNA probes tested when compared to that observed with extracts derived from wild-type cells (Fig. 7). This indicates that there may be multiple IHF binding sites located from the promoter region to up to several hundred bases upstream of the transcription start site. This conclusion is supported by our observation that purified R. capsulatus IHF shifted all three DNA probes encompassing the pucP1 promoter region. Inspection of the DNA sequence upstream of pucB indicates that there are multiple copies of a related sequence motif containing ATT (AAATTGC, AGATTCG, CAATTCG, AAATTCC, and AAATTCG) that are present in the DNA segments that exhibited a gel mobility shift with purified IHF (Fig. 2). Variants of this sequence are also located upstream of the R. capsulatus hip and himA genes, which code for subunits of IHF, as well as in the hupS promoter, which requires IHF for transcription initiation (46-48). Footprint analysis indicates that R. capsulatus IHF protects the variant sequence CCATTGA present in the hupS promoter (47) as well as in the pucURS (33), which contains three repeats of this sequence (AAATTCG-N₆-CAATTCG-N₆-AAATTCC, shown in Fig. 2 at position 489 to 521). Our current model is that IHF binds in the pucURS and stabilizes RegA binding which subsequently interacts with RNA polymerase bound at the pucP1 promoter.

Sequence analysis also indicated the presence of two putative CrtJ binding sites separated by 222 bp with one site overlapping with the pucP1 promoter (Fig. 2). Reporter plasmid expression from the crtJ-disrupted strain DB469 exhibited a 1.5-fold increase in transcriptional activity in the presence of oxygen, which is consistent with previous reports that CrtJ acts as an aerobic repressor of puc expression (36) (Fig. 6). Recent DNA binding studies indicate that CrtJ binds to these two palindromes in a cooperative manner, suggesting that the intervening DNA segment between the palindromes is "looped out" (17).

Conclusion. This study provides the first detailed analysis of the R. capsulatus puc operon promoter. In many respects, the results of our analyses indicate that the puc promoter region exhibits features that are typical of ⁷⁰ type promoters. Ongoing footprint analyses in our laboratory using purified transcription factors are beginning to define individual cis-acting sites that are involved in binding various activators and repressors that control puc expression in response to alterations in light intensity and oxygen tension (16, 36). The challenge will be in further defining the complex interactions that must be occurring among the transcription factors and RNA polymerase at the puc promoter.