Regulation of the Rhodobacter sphaeroides 2.4.1 hemA gene by PrrA and FnrL

Abstract

Part of the oxygen responsiveness of Rhodobacter sphaeroides 2.4.1 tetrapyrrole production involves changes in transcription of the hemA gene, which codes for one of two isoenzymes catalyzing 5-aminolevulinic acid synthesis. Regulation of hemA transcription from its two promoters is mediated by the DNA binding proteins FnrL and PrrA. The two PrrA binding sites, binding sites I and II, which are located upstream of the more-5' hemA promoter (P1), are equally important to transcription under aerobic conditions, while binding site II is more important under anaerobic conditions. By using phosphoprotein affinity chromatography and immunoblot analyses, we showed that the phosphorylated PrrA levels in the cell increase with decreasing oxygen tensions. Then, using both in vivo and in vitro methods, we demonstrated that the relative affinities of phosphorylated and unphosphorylated PrrA for the two binding sites differ and that phosphorylated PrrA has greater affinity for site II. We also showed that PrrA regulation is directed toward the P1 promoter. We propose that the PrrA component of anaerobic induction of P1 transcription is attributable to higher affinity of phosphorylated PrrA than of unphosphorylated PrrA for binding site II. Anaerobic activation of the more-3' hemA promoter (P2) is thought to involve FnrL binding to an FNR consensuslike sequence located upstream of the P2 promoter, but the contribution of FnrL to P1 induction may be indirect since the P1 transcription start is within the putative FnrL binding site. We present evidence suggesting that the indirect action of FnrL works through PrrA and discuss possible mechanisms.

FIG. 1.

Description of the relevant hemA wild-type and mutant upstream sequences used in this study. The +1 sites of transcription from hemA promoters P1 and P2 (11, 26) are labeled (P1) and (P2). The FNR consensuslike sequence is also labeled. DNA sequences protected from DNase I cleavage by PrrA, binding sites I and II (30), are indicated by shading, and the identical 9-bp motifs in these protected regions are highlighted. The P2 sequences that were deleted using the oligonucleotides described previously (11) are in parentheses. DNA oligonucleotides used to create mutations within PrrA binding sites I and II were described previously (30). Sequences corresponding to the primers used to generate DNA that was investigated by the EMSAs are indicated by arrows; labeled binding site I DNA was generated using primer “UP” and biotin-labeled primer “I,” labeled binding site II DNA was generated using primer “II” and biotin-labeled primer “DOWN,” and unlabeled competitor DNA was generated using primers “UP” and “DOWN.” For further details concerning the oligonucleotides and their use, see Materials and Methods.

FIG. 2.

Immunoblots of protein samples probed with anti-PrrA antisera. The samples examined were either crude lysates (Lysate) or protein eluted from a phosphoprotein affinity column (Eluate) prepared from cultures of R. sphaeroides wild-type strain 2.4.1 grown under dark and highly aerobic (30%), semiaerobic (2%), or anaerobic (0%) conditions. A 4.1-mg protein sample of each lysate was applied to the affinity columns. Immunoblots were prepared from the crude lysates (18.0 μg protein) or equal sample volumes of the peak fractions eluted from each column.

FIG. 3.

Immunoblots of protein samples probed with anti-PrrA or anti-HemA antisera. The samples examined were either crude lysates (Lysate) or protein eluted from phosphoaffinity columns (Eluate) prepared from cultures of R. sphaeroides wild-type strain 2.4.1 or cbb₃ oxidase⁻ mutant strain JZ722 grown under dark and highly aerobic (30% oxygen) conditions. The antisera used to probe the immunoblots are indicated at the top. A 2.1-mg protein sample of each lysate was applied to affinity columns. The immunoblot probed with anti-HemA antisera was prepared using samples of crude lysate (21 μg protein). The immunoblots probed with anti-PrrA antisera were prepared using samples of the crude lysates (2.0 μg protein) or equal sample volumes of the peak fractions eluted from each column. For further details see Materials and Methods.

FIG. 4.

Competition EMSA results for PrrA, target biotin-labeled DNA containing binding site I or binding site II sequences, and competitor unlabeled DNA containing both sites I and II. The DNA used in the assays was generated by PCR performed with the primers shown in Fig. 1, and either 1.33 μM PrrA or BeF₃⁻-PrrA was used. The amounts of competitor DNA used are indicated at the bottom. For further details see Materials and Methods.

FIG. 5.

β-Galactosidase activities in extracts of R. sphaeroides wild-type strain 2.4.1 with hemA(P1)::lacZ reporter plasmids having intact or altered PrrA binding sites that had been grown under highly aerobic (30% oxygen) or anaerobic (0% oxygen) conditions. The reporter plasmids used were pBRO75 (none), pBRO100 (I), pBRO109 (II), and pBRO61 (I & II). Alterations in the hemA sequences present on the plasmids are shown in Fig. 1, and additional information about the plasmids is shown in Table 1. The error bars indicate the standard deviations from the means. The values are the average values for duplicate assays of a minimum of three independent growth experiments. One unit of enzyme activity was defined as 1 μmol of o-nitrophenyl-β-d-galactopyranoside hydrolyzed per min. For further details concerning the growth conditions used, see Materials and Methods.

FIG. 6.

Immunoblots of protein samples probed with anti-PrrA antisera. The samples examined were either crude lysates (Lysate) or protein eluted from TALON PMAC magnetic beads (Eluate). The crude lysates were prepared from cultures of R. sphaeroides mutant strain JZ4141 or JZ4148 (Table 1) grown under anaerobic-dark conditions with DMSO, and the eluate was obtained by processing samples of crude lysates containing 5.6 mg of total protein. The immunoblots were prepared using samples of the crude lysates (24.0 μg protein) or equal sample volumes of the eluates. For further details, see Materials and Methods.

FIG. 7.

Growth of R. sphaeroides wild-type strain 2.4.1 with the vector or plasmids coding for the PrrA mutant proteins indicated following incubation under the conditions indicated at the top.

FIG. 8.

Immunoblots of protein samples probed with anti-PrrA antisera. The samples examined were either crude lysates (Lysate) or protein eluted from TALON PMAC magnetic beads (Eluate). The crude lysates were prepared from cultures of R. sphaeroides wild-type strain 2.4.1 with either the pRK415 vector or plasmid pD63K-PrrA from cultures grown under anaerobic-dark conditions with DMSO, and the eluate was obtained by processing samples of crude lysates containing 3.8 mg of total protein. The immunoblots were prepared using samples of the crude lysates (23.0 μg protein) and equal sample volumes of the eluates. For further details see Materials and Methods.

FIG. 9.

β-Galactosidase activities in extracts of R. sphaeroides wild-type strain 2.4.1 or FnrL⁻ mutant strain JZ1678 with hemA(P1)::lacZ reporter plasmids having intact or altered PrrA binding sites that had been grown under semiaerobic (2% oxygen) conditions. The reporter plasmids used were the plasmids described in the legend to Fig. 5. Additional information about the strains and plasmids is shown in Table 1. The error bars indicate the standard deviations from the means for duplicate assays of a minimum of three independent growth experiments. For an explanation of the units see the legend to Fig. 5. For further details concerning the growth conditions used, see Materials and Methods.