Iron-Dependent Cytochrome c₁ Expression Is Mediated by the Status of Heme in Bradyrhizobium japonicum

Bacterial strains and growth conditions. Escherichia coli strain DH5α was used for propagation of plasmids. Cells containing pBluescript II-based plasmids were grown in LB medium containing 200 μg/ml ampicillin. B. japonicum I110 and LO are parent strains used in the present work. Mutant strains GEM4, LODTM5, MLG1, and I110ek4 are defective in fur, irr, hemA, and hemH, respectively, and have been described previously (8, 10-12). Strain 2800 contains a Tn5 insertion in the cytochrome c₁-encoding portion of the fbcH gene (26). B. japonicum strains were routinely grown at 29°C in GSY (glycerol-salts-yeast extract) medium (9). The Tn5-containing mutants LODTM5 and 2800 were grown in media containing 50 μg/ml each of kanamycin and streptomycin. Strain GEM4 contains an Ω cassette and was grown in medium containing 50 μg/ml each of streptomycin and spectinomycin. Strain I110ek4 contains a kanamycin resistance-encoding cassette (8) and was grown on 50 μg/ml kanamycin. Cells containing pRK290-based plasmids were grown with 50 μg/ml tetracycline. All B. japonicum strains were grown in media containing 100 μg/ml cycloheximide to inhibit growth of eukaryotic microbes in the media. Strain I110ek4 was grown in media supplemented with 10 μM hemin (ferric heme hydrochloride) to satisfy its heme auxotrophy. For low-heme conditions, the medium was supplemented with 0.1 μM heme, which allows growth of strain I110ek4 comparable to the wild type (22). Strain MLG1 was grown in media supplemented with 150 μM δ-aminolevulinic acid (ALA) when needed. For experiments with cells grown on high- or low-iron media, a modified GSY medium was used in which 0.5 g/liter of yeast extract (Difco) was used rather than 1 g/liter. For low-iron medium, no exogenous iron source was added. The actual concentration of iron in this medium is 0.3 μM as determined by atomic absorption spectroscopy. A 6 μM concentration of FeCl₃ was added for the high-iron medium.

Analyses of cytochrome c₁ and GroEL proteins.Cytochrome c₁ and GroEL were measured in cells grown under various conditions of iron, ALA, or heme by immunoblot analysis. To prepare whole-cell extract, cells grown to mid-log phase were harvested by centrifugation. The cell pellets were washed once with phosphate-buffered saline (PBS) buffer (137 mM NaCl, 2.7 mM KCl, 4.3 mM Na₂HPO₄ · 7H₂O, 1.4 mM KH₂PO₄) and then resuspended in 200 μl of PBS buffer. The cell solutions were mixed with equal volumes of 2× sodium dodecyl sulfate (SDS) sample buffer (20 mM sodium phosphate, 20% glycerol, 4% SDS, 0.2 M dithiothreitol, 0.001% bromophenol blue) and boiled for 5 min, followed by centrifugation at 13,000 × g for 1 min. Aliquots of the supernatant were loaded onto 10% SDS-polyacrylamide gels and analyzed by immunoblotting using antibodies against cytochrome c₁ (a gift of H. Hennecke and L. Thöny-Meyer) or GroEL (StressGen, Vancouver, Canada). The anti-cytochrome c₁ antibodies were raised against a 16-amino-acid synthetic peptide representing positions 581 to 596 of FbcH (24).

Expression of cytochrome c₁ encoded by a plasmid-borne construct contains a His tag (see below) and was analyzed in membrane fractions. Membranes were prepared as described previously (8) with minor modifications. Briefly, a 500-ml culture was grown to an optical density at 540 nm of 0.6. Cells were harvested and washed with 50 mM sodium phosphate buffer (pH 6.8). The cells were then resuspended in 5 ml buffer and broken by passing twice through a French pressure cell press (18,000 lb/in²) in the presence of 1 mM phenylmethylsulfonyl fluoride (PMSF). DNase I and RNase A were added to the cell extract to a final concentration of 20 μg/ml. The cell extract was centrifuged twice at 10,000 × g for 10 min. The resulting supernatant was centrifuged at 140,000 × g for 90 min. The membrane pellet was suspended using a Pyrex homogenizer in PBS buffer in the presence of 1 mM PMSF. The membrane suspensions were mixed with equal volumes of 2× SDS sample buffer and boiled for 5 min, followed by centrifugation. The supernatant was used for SDS-polyacrylamide gel electrophoresis and immunoblot analysis using anti-tetra-His antibody (QIAGEN, Valencia, CA). Protein concentrations were determined using bicinchoninic acid assay (Sigma). Bovine serum albumin was used as a standard.

Isolation of B. japonicum RNA.Total RNA was prepared and analyzed as described previously (2). Briefly, cells grown to an optical density (at 540 nm) of 0.4 to 0.6 were harvested by centrifugation at 4°C, and cell pellets from 12.5 ml of the culture were resuspended in 550 μl of lysis buffer containing 10 mM NaCl, 10 mM Tris (pH 8), 5% SDS, and 200 μg of proteinase K per ml and incubated at 37°C for 5 min. Three hundred microliters of 5 M NaCl was added to the lysed cells, and the mixture was kept on ice for 10 min and then centrifuged at 13,000 × g at 4°C. RNA in the supernatant was precipitated by addition of 3 volumes of ethanol at −70°C for 1 h, followed by centrifugation. The crude RNA pellet was resuspended in 1 mM EDTA, acid phenol-chloroform extracted, and then precipitated with ethanol. This preparation was treated with DNase I in the presence of RNasin inhibitor (Promega) to eliminate any contaminating DNA.

Analysis of steady-state levels of cytochrome c₁, hemB, and groEL mRNAs.RNA levels of specific genes were determined by a quantitative RNase protection assay according to a protocol described previously (2). For mRNA encoding cytochrome c₁, a 367-bp fragment was obtained by PCR using the following primers: 5′-ATGGATCCACTACTTCCCCTCGCCATTTC-3′ and 5′-TCTAAGCTTCACATCAGGAACGTGGTGAC-3′. The PCR product was then cloned into the BamHI and HindIII sites of the pBluescript II SK. A 488-bp fragment within the hemB gene was obtained as described previously (2). For groEL, a 297-bp fragment was obtained by PCR using the following primers: 5′-TATGGATCCCCGTAGAAGCCATCGTCAAT-3′ and 5′-TCTAAGCTTTGAGGATGTAGGGGTCTTCG-3′.

The PCR product was then cloned into the HindIII/BamHI site of the pBluescript II SK. Each recombinant plasmid was linearized with StuI and used as a template for in vitro transcription from the vector-borne T7 promoter of non-template strand DNA for syntheses of antisense riboprobes. The transcription reaction was carried out by using a MAXIscript in vitro transcription kit (Ambion, Inc.) in accordance with the manufacturer's instructions. The [α-³²P]UTP-labeled riboprobes were purified by electrophoresis on a 5% denaturing polyacrylamide gel, followed by elution of excised fragments containing the riboprobes overnight at 37°C. RNase protection analysis to detect cytochrome c₁, hemB, and groEL mRNAs was performed by using the Hyberspeed RNase protection assay kit (Ambion Inc.) in accordance with the manufacturer's instructions. A total of 2 μg of total RNA was used for each reaction. The protected regions were analyzed by polyacrylamide gel electrophoresis, followed by autoradiography.

Construction of a plasmid encoding a His-tagged cytochrome c₁.In B. japonicum, the fbcH gene encodes a cytochrome b-c₁ fusion protein that is posttranslationally cleaved into two proteins. We made a construct encoding only the cytochrome c₁ portion of the protein, including a signal peptide and a C-terminal His tag, which was expressed under the control of the fbcFH promoter. The cytochrome c₁ coding region was amplified by PCR using genomic DNA as the template and the following primers: forward, 5′-TATGAATTCATATGCTCGCCAAGGGCGGCAAGGCG-3′; reverse,5′-TATGAATTCTCAGTGGTGGTGGTGGTGGTGGTGCGAGTCGGCCCA-3′. The forward primer includes an EcoRI site for cloning and an NdeI site which contains an ATG initiation codon. The reverse primer contains an EcoRI cloning site and six consecutive histidine codons. The PCR product was digested with EcoRI and ligated into the EcoRI site of pBluescript II SK(+). The resultant plasmid was linearized by digestion with SalI and NdeI, sites that originate with the vector and PCR primer, respectively. The linearized plasmid was ligated with a SalI/NdeI fragment that contains the fbcFH upstream region, including the promoter, ribosome binding site, and ATG start codon. This upstream region was obtained by PCR using the following primers: forward, 5′-GTTCGAAAACCAATCCTATCTGATGCCGAG-3′; reverse, 5′-TCACGTCGACATATGGATTCCAACCCTTTCTTCTTATGCTACCGG-3′. The PCR product was digested with SalI and NdeI, which recognize sites found in the genome and the PCR primer, respectively. This plasmid, pC1PSK, contains the coding region of cytochrome c₁ and the upstream region of fbcFH. The insert was ligated into the EcoRI site of the broad-host-range plasmid pRK290 to construct pC1-his6. This plasmid was introduced into B. japonicum strains for analysis.

Mutagenesis of the cytochrome c₁-encoding gene.Mutagenesis of cysteine codons 68 and 71 to serine codons from pC1PSK was carried out using QuikChange (Stratagene, La Jolla, CA) as described previously (27). The forward and reverse primers to mutate cysteine 68 to serine are 5′-GTCTACAAGGAAGTCAGCGCCAGCTGCCACG-3′ and 5′-CGTGGCAGCTGGCGCTGACTTCCTTGTAGAC-3′, respectively. The forward and reverse primers to mutate cysteine 71 to serine are 5′-GTCAGCGCCAGCAGCCACGGCCTGTC-3′ and 5′-GACAGGCCGTGGCTGCTGGCGCTGAC-3′, respectively. The plasmid containing the double mutant was then digested with EcoRI, ligated into the EcoRI site of pRK290, and introduced into B. japonicum strains by conjugation.

Cytochrome c₁ protein expression is affected by iron.Cytochromes are heme proteins, and heme is an iron-containing tetrapyrrole compound. Therefore, we wanted to address whether cytochrome levels are controlled by iron availability. B. japonicum cells were grown in media containing no additional iron (final concentration, 0.3 μM) or supplemented with 6 μM FeCl₃, and cytochrome c₁ was assessed by immunoblot analysis using antibodies directed against it (Fig. 1). The cytochrome level was much lower in cells grown under iron limitation than in those grown in iron-supplemented media for both wild-type B. japonicum strains I110 and LO. By contrast, accumulation of GroEL was relatively unaffected by iron. Thus, there is a positive correlation between iron and cytochrome c₁ protein levels, which makes physiological sense, since iron is a component of the holoprotein.

• Open in new tab
• Download powerpoint

FIG. 1.

Effect of iron on cytochrome c₁ apoprotein levels. Extracts of B. japonicum wild-type strain I110 or LO grown in low-iron (−) or high-iron (+) media were analyzed for cytochrome c₁ (Cyc1) or GroEL protein by immunoblotting using antibodies directed against the respective proteins. Thirty micrograms of protein was loaded per lane.

We examined steady-state levels of cytochrome c₁ transcripts to determine whether iron-dependent expression was at the RNA level (Fig. 2). Cytochrome c₁ mRNA levels were essentially the same in cells grown under low- or high-iron conditions. For a positive control, we examined hemB gene mRNA levels in the same cells, which has been shown previously to be regulated by iron (3). The findings indicate that iron-dependent accumulation of cytochrome c₁ protein is not due to changes in the synthesis or degradation of the mRNA that encodes it but rather is due to a posttranscriptional process.

• Open in new tab
• Download powerpoint

FIG. 2.

Effect of iron on cytochrome c₁ mRNA levels. RNA was isolated from cells grown in low-iron (−Fe) or high-iron (+Fe) media. The samples were analyzed for cytochrome c₁ (cyc1), hemB, or groEL mRNA by RNase protection analysis. Two micrograms of total RNA was loaded per lane.

Iron-dependent expression of cytochrome c₁ is not mediated by Irr or Fur.Although cytochrome c₁ mRNA expression was not affected by iron, it is possible that the system which regulates protein expression is under transcriptional control. Fur and Irr are the two known regulators of iron-dependent processes in B. japonicum. Irr functions under iron limitation to negatively control heme biosynthesis and positively affect ferric citrate transport (11, 22). Fur is also involved in iron transport, and it transcriptionally controls the irr gene as well (1, 7, 12, 13). Mutants defective in the irr or fur gene and their corresponding parent strains were grown in low-iron media, and cytochrome c₁ protein was measured by immunoblotting (Fig. 3). We found that expression in both mutants was iron responsive and similar to the parent strains. Thus, iron-dependent expression of cytochrome c₁ is independent of Fur or Irr.

• Open in new tab
• Download powerpoint

FIG. 3.

Effects of iron on cytochrome c₁ apoprotein levels in fur and irr mutants. Strain LO is the parent strain of the irr mutant LODTM5. Strain I110 is the parent strain for fur mutant GEM4. Cells grown in low-iron (−) or high-iron (+) media were analyzed for cytochrome c₁ (Cyc1) or GroEL protein by immunoblotting using antibodies directed against the respective proteins. Thirty micrograms of protein was loaded per lane. Wt, wild type.

Cytochrome c₁ protein expression is heme dependent.Heme is an iron porphyrin, and thus, its formation requires iron. Furthermore, heme biosynthesis is regulated by iron (3, 11). Thus, it seemed plausible that the effect of iron on cytochrome apoprotein formation is an indirect consequence of the heme status. To test this idea, we examined cytochrome c₁ protein accumulation in heme biosynthesis mutants. The first heme precursor, ALA, is synthesized from glycine and succinyl-coenzyme A by ALA synthase, encoded by the hemA gene (Fig. 4A). Cytochrome c₁ protein was very low in a hemA mutant strain grown in the absence of an exogenous heme precursor (Fig. 4B). However, wild-type levels of the protein were restored when cells were grown in media supplemented with ALA, the product of ALA synthase, or with heme. RNase protection analysis revealed that cytochrome c₁ mRNA was not diminished in the hemA strain, nor was it enhanced by addition of ALA or heme to the growth medium (Fig. 5). The cytochrome c₁ mRNA level was somewhat elevated in the unsupplemented hemA strain; we did not investigate this phenomenon further, because it does not explain the near absence of detectable protein. Thus, the deficiency in cytochrome c₁ protein and its response to ALA or heme were not due to differences in the quantity of the mRNA that encodes cytochrome c₁.

• Open in new tab
• Download powerpoint

FIG. 4.

Expression of cytochrome c₁ in heme biosynthesis mutants. (A) Abbreviated heme biosynthetic pathway showing ALA synthase (ALAS), encoded by hemA, and ferrochelatase (FC), encoded by hemH. The substrates of ALA synthase are glycine and succinyl coenzyme A (CoA). Proto, protoporphyrin IX. (B) Cells of wild-type (Wt) strain I110, hemA strain MLG1, or hemH strain I110ek4 were grown in media containing no supplement (−), 150 μM ALA, or 15 μM heme (H). Cell extracts were analyzed by immunoblotting using antibodies raised against cytochrome c₁ (Cyc1) or GroEL. Thirty micrograms of protein was loaded per lane.

• Open in new tab
• Download powerpoint

FIG. 5.

Effect of ALA or heme on expression of cytochrome c₁ mRNA. RNA was isolated from I110 (wild type [Wt]) or MGL1 (hemA) cells grown in medium containing no supplement, 150 μM ALA, or 15 μM hemin (H). The samples were analyzed for cytochrome c₁ (cyc1) or groEL mRNA by RNase protection analysis. Two micrograms of total RNA was loaded per lane.

Ferrochelatase catalyzes the insertion of iron into protoporphyrin in the final step of the heme pathway and is encoded by the hemH gene (Fig. 4A). A B. japonicum hemH mutant also expressed very low levels of cytochrome c₁ as observed in immunoblots (Fig. 4B). The protein was restored in the hemH mutant when the growth medium was supplemented with heme, as was observed for the hemA strain. However, addition of ALA did not rescue the cytochrome c₁-deficient phenotype in the hemH strain as it did for the hemA mutant. These observations indicate that rescue of the hemA strain by ALA was likely due to the conversion of ALA to heme, rather than a direct effect of ALA on cytochrome c₁ accumulation. Overall, these findings show that cytochrome c₁ accumulation is coordinated with heme availability and that expression is controlled at a step after mRNA synthesis and turnover.

To further address the possibility that control by iron is mediated by heme, we examined the effect of iron on cytochrome c₁ in heme biosynthesis mutants in the absence or presence of ALA or heme (Fig. 6). Whereas cytochrome expression was iron dependent in parent strain I110, protein levels remained high when cells were grown in the presence of heme, regardless of the iron status. In addition, the effects of iron and heme did not appear to be additive, suggesting that they do not act independently. Cytochrome c₁ was undetectable in the hemA strain under high- or low-iron levels. When ALA was added to the growth medium, iron responsiveness was restored to a level similar to what was observed in the wild type (Fig. 6). For the hemH strain, cytochrome c₁ levels were very low unless heme was present, and this expression was independent of iron. These data are consistent with the conclusion that iron exerts its regulatory effect through heme.

• Open in new tab
• Download powerpoint

FIG. 6.

Effects of ALA and heme on iron-dependent expression of cytochrome c₁ in heme biosynthesis mutants. Cells of the wild type (Wt) and hemA and hemH mutants were grown in low-iron (−) or high-iron (+) iron media that is not supplemented further (−) or is supplemented with ALA or heme. Cell extracts were analyzed by immunoblotting using antibodies raised against cytochrome c₁ (Cyc1) or GroEL. Thirty micrograms of protein was loaded per lane.

Control of cytochrome c₁ by iron or heme is independent of posttranslational processing of the cytochrome b-c₁ fusion protein.In B. japonicum cytochromes b and c₁ are synthesized as a fusion protein precursor encoded by the fbcH gene that is subsequently processed into two proteins by an endopeptidase activity (24, 26). To address whether control by iron occurred at the processing of the fusion protein, we constructed a plasmid-borne DNA, called pC1-his6 that encoded the cytochrome c₁ component alone, including the signal peptide for targeting to the periplasm, and a histidine tag at the C terminus (Fig. 7A). The gene was placed under the control of the fbcFH promoter. Plasmid-encoded cytochrome c₁ was detected in the membrane fraction using anti-His antibodies, which was expected of a membrane-anchored periplasmic protein (Fig. 7B). Furthermore, protein accumulation was strongly dependent on the iron status of the plasmid-encoded cytochrome c₁ (Fig. 7B), similar to what was observed for the endogenous protein (Fig. 1). Similar results were obtained in a wild-type background where endogenous cytochrome c₁ is also expressed or in mutant strain 2800, which does not express endogenous cytochrome c₁ due to a chromosomal transposon insertion (24).

• Open in new tab
• Download powerpoint

FIG. 7.

Expression of the portion of the fbcH gene encoding cytochrome c₁ in trans and the effects of iron, heme, or mutation of the heme-binding site. (A) Construction of pC1-His6. The region of fbcH encoding cytochrome c₁ (Cyt c₁), including the internal signal peptide (ISP), was fused to the fbcFH upstream region, which includes the promoter, the fbcF ribosome binding site, and initiation codon. This was ligated into the broad-host-range vector pRK290 and then mobilized into various B. japonicum strains. The cysteines involved in covalent linkage to heme and substitutions in the mutant are shown. (B) Iron-dependent expression from pC1-His6 in wild-type strain I110 and cytochrome c₁ mutant 2800. Membranes were isolated from cells grown in low-iron (−Fe) or high-iron (+Fe) media and analyzed by immunoblotting using anti-His₄ antibodies. Fifty micrograms of protein was loaded per lane. (C) Heme-dependent expression from pC1-His6 in I110 and hemA strain MLG1. Membranes were isolated from cells grown in media alone (−) or supplemented with ALA or heme and analyzed as described above for panel B. (D) Effect of mutation of the heme binding site on expression. Membranes were isolated from strain 2800 cells harboring pC1-His6 or pC1(C/S)-His6 expressing unmutated and mutated cytochrome c₁, respectively. Protein was analyzed as described above for panel B. Wt, wild type.

To address whether control of cytochrome c₁ expression by heme requires processing of the cytochrome b-c₁ fusion protein, we introduced pC1-his6 into a hemA strain and measured the protein in cells grown under various conditions (Fig. 7C). Little cytochrome c₁ was observed in the hemA strain grown in GSY medium. However, the protein was readily detected when cells were grown when the medium was supplemented with ALA or heme, as occurred with the endogenous cytochrome c₁. These findings suggest that regulation of cytochrome c₁ accumulation by iron or heme is not at the level of posttranslational cleavage of the fusion protein precursor.

Cytochrome c₁ accumulation requires heme binding. c-type cytochromes bind heme covalently via a thioether bond formed between the two vinyl groups of the porphyrin ring of heme and the thiol groups of two proximal cysteine residues within the conserved motif C-X₂-CH. The corresponding residues in B. japonicum cytochrome c₁, cysteines 68 and 71 of the protein synthesized from pC1-his6 (these cysteines are at positions 38 and 41 of the mature protein and 471 and 474 of the b-c₁ fusion protein). The cysteine residues were each mutated to serine in pC1-his6 and introduced into B. japonicum strain 2800 (Fig. 7D). The plasmid-borne unmutated protein was expressed in cells, but the C68S C71S mutant was not detectable in immunoblots. Thöny-Meyer et al. found essentially the same result using the entire fbcH gene encoding the b-c₁ fusion protein (24). Thus, mutations that abrogate heme biosynthesis or heme binding to cytochrome c₁ result in the failure of apoprotein to accumulate in cells.