Involvement of the C-Terminal Extension of the α Polypeptide and of the PucC Protein in LH2 Complex Biosynthesis in Rubrivivax gelatinosus

Bacterial strains, plasmids, and growth media. Escherichia coli strains were grown at 37°C in Luria Bertani medium (30). R. gelatinosus (39) was grown at 30°C in malate medium (1), either photosynthetically (light, 3,000 lx; anaerobiosis) or semiaerobically (cultures filled to 50% of the flask volume agitated at 100 rpm in the dark). Antibiotics were used at the following concentrations for E. coli and R. gelatinosus: kanamycin, 50 μg/ml; ampicillin, 25 μg/ml; chloramphenicol, 3 μg/ml; streptomycin, 25 μg/ml; and spectinomycin, 25 μg/ml. The bacterial strains and plasmids used in this work are listed in Table 1.

Gene transfer and strain selection.Plasmid DNA was introduced into R. gelatinosus by electroporation as previously described (24). Transformants were selected on malate plates supplemented with the appropriate antibiotics. For gene disruption, two different antibiotic resistance markers were used to distinguish between double- and single-crossover events, namely, the ampicillin resistance gene of pBluescript KS(+) and the antibiotic resistance carried by the cartridge inserted in the gene to be disrupted.

Molecular biology techniques.Unless otherwise indicated, the standard methods of Sambrook et al. (30) were used. Plasmid DNAs were purified using the Bio-Rad Quantum Prep plasmid kit. DNA was treated with restriction enzymes and other nucleic acid-modifying enzymes (Klenow fragment, T4 DNA polymerase, and T4 DNA ligase) according to the manufacturer's specifications. DNA fragments were analyzed on agarose gels, and different restriction fragments were purified using the Geneclean kit (BIO 101, Inc.). PCRs were carried out using TaqDNA polymerase from Appligene according to the manufacturer's instructions, except that dimethyl sulfoxide was added to a 5% final concentration. Twenty cycles of 30 s at 92°C, 40 s at 60°C, and 40 s at 72°C were performed in a Hybaid thermal cycler. We used the Quick-Change mutagenesis kit (Stratagene) to mutate the pucA gene in the pA100 plasmid. To synthesize the different truncated forms of PucA, we used the following oligonucleotides as primers: mutA1-13 (5′-GCCGCCCCGGCCCCGGTCTAATGACCGCAAGCCCCGGCGC-3′) and mutA2-13 (5′-GCGCCGGGGCTTGCGGTCATTAGACCGGGGCCGGGGCGGC-3′); mutA1-18 (5′-GCAAGAAGGCTGCTGCGTAATGAGCTGCTGCCGCCCCGGC-3′) and mutA2-18 (5′-GCCGGGGCGGCAGCAGCTCATTACGCAGCAGCCTTCTTGCCG-3′). DNA sequencing was performed on an ABI 310 automatic DNA sequencer (Applied Biosystems).

RNA isolation, Northern blot analysis, and primer extenion.Total RNA was isolated from R. gelatinosus in mid-logarithmic-phase culture by the procedure described by Heck et al. (11). Glass beads (0.11-mm diameter; Braun) were added to the breaking buffer to improve cell breakage. After extraction, the RNAs were treated with RNase-free DNase (Life Technologies), extracted with acidified phenol-chloroform, precipitated with ethanol, and resuspended in RNase-free water. Total RNA (10 μg per lane) and a 0.24- to 9.5-kb RNA ladder (3 μg per lane) (Bio-Rad) were denatured for 10 min at 70°C in MOPS (morpholinepropanesulfonic acid)-formaldehyde (37%)- formamide (50%) buffer (1:3 [vol/vol] RNA-buffer). RNA was run on a 1.6% (wt/vol) agarose-formaldehyde gel at 20 mV for 24 h. The gel was then placed for 30 min in 10× SSC (1× SSC is 0.15 M NaCl plus 0.015 M sodium citrate) buffer and blotted in 20× SSC with Hybond-N⁺ or Hybond-N membrane (Amersham). After being blotted, the membrane was exposed to a UV stratalinker (Autocrosslink; 120,000 J/cm²). The membrane was prehybridized in a buffer containing 5× SSC, 1× Denhardt's solution, 0.05 M phosphate buffer, 0.1% sodium dodecyl sulfate (SDS), 50% formamide, and denatured sheared salmon sperm DNA for 2 h at 45°C. The blots were then hybridized with the denatured labeled probes for 18 h at 45°C. Three probes (1, 2, and 3) were used as indicated in Fig. 2A. Probes 1 and 2 were DNA fragments excised from pA100 (the 0.2-kb probe 1 was isolated by PpuMI/MscI endonuclease restriction, and the 0.45-kb probe 2 was isolated by PpuMI/SphI endonuclease restriction). The 0.6-kb probe 3 was obtained by PCR from PA100 with the oligonucleotides C1 (5′-CCAGGTCGTCGGCCGGC-3′) and C2 (5′-GCGGCACCATCGCCGCG-3′) as primers. The probes were labeled with [α³²-P]dCTP using the Oligolabelling kit (Amersham Pharmacia). After hybridization, the membrane was washed twice for 5 min each time in 2× SSC- 0.1% SDS at 45°C and three times for 10 min each time in 1× SSC- 0.1% SDS at 65°C and exposed in a phosphorimager.

• Open in new tab
• Download powerpoint

FIG. 2.

(A) Physical map of the pucBAC genes of R. gelatinosus. The genes are represented by boxes, and putative transcripts are indicated by arrows. The positions of the putative transcription terminators (T, T1, T2, and Tc) are indicated by loops with stems. P1 corresponds to a σ⁷⁰-type promoter. The DNA probes (1, 2, and 3) used in Northern blot experiments are indicated. The inactivation sites of puc genes are the MscI site for pucA and BlpI for pucC. Km, kanamycin cartridge; Ω, streptomycin-spectinomycin cartridge. (B) DNA sequence of the pucBA promoter region. The putative σ⁷⁰-type promoters are boxed. The ribosome binding site (RBS) is represented by a dotted box. The PpsR binding sites (TGT-N₁₂-ACG and TGT-N₁₂-ACA) are shown in boldface uppercase letters. The coding sequence of pucB is in uppercase letters. The position of the oligonucleotide PE, used for the primer extension, is also shown.

Primer extension experiments.Total RNA was extracted from R. gelatinosus grown photosynthetically to an optical density of 0.5 to 0.6 at 680 nm, as described above. The 5′-end [γ³²-P]ATP-labeled PE (5′-TCAACCAGACCCTTTTGCAGTTCTTC-3′) (500,000 cpm) was precipitated with 20 μg of total RNA and resuspended in PXI buffer (20 mM Tris-HCl, pH 8.3, 200 mM KCl, 0.1 mM Na₂-EDTA). The sample was heated for 3 min at 100°C and then at 65°C for 90 min. Deoxynucleoside triphosphate (0.5 mM) and 25 U of avian myeloblastosis virus reverse transcriptase (Boehringer) in PXII buffer (73.3 mM Tris-HCl, pH 8.3, 133.3 mM KCl, 16.7 mM dithiothreitol, 16.7 mM MgCl₂) were added, and the mixture was incubated at 42°C for 2 h. After reverse transcription, the synthesized DNA was extracted and resuspended in Tris-EDTA buffer, 35 mM NaOH, and stop buffer. The DNA was electrophoresed in a 6% polyacrylamide gel containing 8 M urea. The sequencing gel was dried and exposed to film with an intensifying screen at −80°C. For the sequencing ladder, 20 ng of the pucPE probe was hybridized to 5 μg of pA100 DNA template. The sequencing reaction was performed with Sequenase using the dideoxy sequencing method of Sanger.

Half-life measurement of pucBA mRNA.Cells of R. gelatinosus were grown under semiaerobic and photosynthetic conditions until logarithmic growth phase, to an optical density of 0.5 to 0.6 at 680 nm. Then, rifampin at a final concentration of 200 μg/ml was added to the cultures, and cells were harvested at several time points. The mRNA pucBA half-lives were determined by measuring the radiolabeled bands on the Northern blots using the ImageQuant program. The 16S rRNA of R. gelatinosus was used as an internal-standard probe for the calibration of total RNA.

Membrane preparation.Membranes were prepared by differential ultracentrifugation after disruption of cells with a French press and resuspension in 0.1 M Na phosphate buffer, pH 7.4, containing 1 mM phenylmethylsulfonyl fluoride. The protein concentration was determined using the Pierce BCA Protein Assay Reagent method with bovine serum albumin as a reference standard, as prescribed by the manufacturer.

Spectroscopic analyses.The absorption spectra of whole cells were measured with a Cary 500 spectrophotometer at room temperature on 3 · 10⁹ cells resuspended in 1 ml of 60% sucrose. To discriminate between the LH2 and the LH1 absorption peaks, the absorption spectra of membranes were measured at 77 K with an Aminco DW2A spectrophotometer interfaced with a computer. Membranes were isolated from cells grown to an optical density of 0.8 to 1 at 680 nm; they were resuspended in 60% glycerol and frozen in liquid N₂ in 2-mm-thick cuvettes in the low-temperature accessory of the spectrophotometer.

Nucleotide sequence accession number.The pucBA operon, the pucC gene, and the deduced amino acid sequences were deposited in the GenBank Data Library under accession number AF312921 .

Characterization of the pucBA operon of R. gelatinosus. (i) Cloning and sequence analyses of the pucBA operon.Screening of a genomic library from R. gelatinosus with the 0.9-kb MscI/XhoI fragment of R. gelatinosus 151 pucBA genes led to the identification of pB3-24 containing a 6-kb hybridizing fragment. This fragment contains the pucBA genes and the 3′ end of the pucC gene. It was used to subclone a 2.3-kb MluI/Sau3AI fragment carrying the pucBA genes to give the plasmid pA100. To clone the entire pucC coding sequence, the genomic library of R. gelatinosus was screened again by PCR using the primers C1 and C2, leading to the identification of the plasmid pB19-143 containing a 4.5-kb fragment. This fragment contained the pucB, pucA, and pucC genes. It was used to subclone a 2.8-kb AatII fragment carrying the 5′ end of the pucC gene to give the plasmid pA108.

A physical map of the pucBAC genes is presented in Fig. 2A. Unlike Rhodobacter species (21, 38), in which the pucBAC genes form an operon, in R. gelatinosus the pucC gene is located 550 bp downstream of the pucA gene in an opposite transcriptional orientation, as reported for strain 151 (32). Putative elements involved in the transcription of pucBA genes are shown in Fig. 2B. Two potential promoters, P1 and P2, exhibiting an E. coli-like σ⁷⁰ sequence motif were identified at −126 and −180 bp from the pucB start codon. Two copies of a conserved repressor PpsR recognition palindrome (TGT-N₁₂-ACG/A) at −137 and −110 bp from the pucB initiation codon were also identified, involving PpsR in pucBA expression (34). Sequences that might form hairpin structures were found (Fig. 2A), one downstream of orf1 (T), two downstream of pucA (T1 and T2), and one downstream of pucC (Tc). These sequences are followed by a T-rich sequence, suggesting that they form terminator structures. A putative promoter sequence with a well-conserved −35 region sequence was identified for the pucC gene.

DNA sequence comparison between R. gelatinosus strain 1 and R. gelatinosus strain 151 shows that the percentages of identity are 94% for the intragenic sequences and 81% for the intergenic sequences. In addition, the resulting substitutions within the pucB and pucA coding sequences are silent, except for substitution A57V in PucA. Within the pucC gene, 16 amino acid substitutions have been identified.

(ii) Characterization of the puc gene transcripts of the wild-type strain.Transcription of the pucBA and pucC regions has been studied by Northern blot analyses with total RNA extracted from the wild-type strain grown either photosynthetically or semiaerobically and tested with several probes (Fig. 2A). Hybridization with probe 1 resulted in two transcripts of 0.8 and 0.65 kb for both growth conditions (Fig. 3). The amounts of the 0.65-kb transcript under both conditions are slightly smaller than those of the 0.8-kb transcript. This difference could be due to different stabilities. To investigate this possibility, we determined the half-life of each transcript (Fig. 4). For the 0.8-kb pucBA mRNA, the half-lives are 25 ± 3 and 14 ± 2 min under semiaerobic and photosynthetic conditions, whereas the half-lives of the 0.65-kb puc mRNA are 6 and 3 ± 0.5 min under semiaerobic and photosynthetic conditions, respectively. These results indicate that the transcripts are more stable under semiaerobic than under photosynthetic conditions and that the smaller transcript is much less stable. We have never detected any pucC transcript either by Northern blotting (probe 3) or by reverse transcription-PCR analyses irrespective of the growth conditions, suggesting a very low transcription level of the pucC gene and/or high instability of its transcript.

• Open in new tab
• Download powerpoint

FIG. 3.

Northern blot analysis of the pucBA operon. Total RNAs were prepared from R. gelatinosus grown under semiaerobic (lane 1) and photosynthetic (lane 2) conditions; 10 μg of RNA per lane and 3 μg of RNA ladder (lane 3) were loaded. The membrane was hybridized with probe 1, identified in Fig. 2A.

• Open in new tab
• Download powerpoint

FIG. 4.

puc mRNA decay kinetics. Total RNA was isolated several times after the addition of rifampin (200 μg/ml) to the culture grown under semiaerobic (A and B) or photosynthetic (C) conditions. (A) Northern blot analysis of puc mRNAs from the wild-type strain grown under semiaerobic conditions. The hybridization was performed with ³²P-labeled probe 1 and then rehybridized with a labeled probespecific to 16S rRNA as an internal standard for the RNA calibration. (B and C) Quantification of radiolabeled bands using the ImageQuant program. The squares represent the 0.8-kb transcript, and the diamonds represent the 0.65-kb transcript.

(iii) Identification of the functional promoter and terminator of the pucBA operon.The initiation site of pucBA transcription was determined by primer extension analyses with total RNA extracted from photosynthetically grown cultures of R. gelatinosus using the labeled PE primer located downstream of P1. As shown in Fig. 5A, the major and longest primer extension product obtained allowed us to localize the transcription initiation site at +10 bp from the σ⁷⁰-type promoter P1 immediately after the PpsR putative binding site and at ∼90 bp from the pucB start codon. This result is compatible with the size of the longer transcript and suggests that the transcription of pucBA starts from only one promoter. The other signals underneath the major primer extension are attributed to an early stop of the reverse transcriptase. Therefore, the two puc transcripts should have two different 3′ ends. To confirm this assumption, we used two different probes. Probe 1 hybridizes to the sequence upstream of the first terminator, T1, and probe 2 hybridizes to the sequence downstream of the T1 terminator (Fig. 2A). As shown in Fig. 5B, probe 2 revealed only the large transcript of 0.8 kb, whereas probe 1 revealed both transcripts of 0.65 and 0.8 kb (Fig. 3), indicating that the small transcript terminates at T1 and the larger one terminates at T2.

• Open in new tab
• Download powerpoint

FIG. 5.

(A) Mapping of 5′ ends of pucBA transcripts by primer extension using the PE primer shown in Fig. 2B. The G, A, T, and C lanes represent didesoxyribonucleotide sequencing ladders generated with the same primer. Lanes 1 and 2 contain the products of a primer extension reaction with total RNA isolated from photosynthetic cells. The −10 potentiel σ⁷⁰ promoter sequence (P1) is boxed. The boldface ACA represents the 3′ end of a putative PpsR binding site. The arrow indicates the pucBA transcriptional start site located at −90 nucleotides from the pucB initiation codon start. (B) Northern blot analyses of the pucBA operon. Total RNAs were prepared from R. gelatinosus grown under semiaerobic (lane 1) and photosynthetic (lane 2) conditions; 10 μg of RNA per lane and 3 μg of RNA ladder were loaded (lane 3). The membrane was hybridized with probe 2, shown in Fig. 2A.

(iv) Construction and characterization of R. gelatinosus mutants disrupted in pucA or pucC.To study the functions of the pucA and pucC gene products in LH2 biosynthesis, two mutant strains were constructed by insertional mutagenesis. The pucA and pucC genes were disrupted with the Ω (streptomycin-spectinomycin) and kanamycin cartridges, respectively (Fig. 2A). The suicide vectors pA102 and pA117 (Table 1) were introduced into wild-type R. gelatinosus by electroporation, and the recombinants resulting from double-crossover events were selected according to their antibiotic resistance profiles. The purified mutant strains were analyzed by Southern blot analysis (data not shown) and named PUCAΩ and PUCCK. While under semiaerobic conditions the wild type and the mutants grew with the same generation time (3 h), the mutants grew more slowly (generation time, 3.5 h) than the wild type (generation time, 2.5 h) under photosynthetic conditions.

The absorption spectra of the wild type and the two mutant strains, PUCAΩ and PUCCK, were recorded from cultures grown under either semiaerobiosis (Fig. 6A) or photosynthesis (Fig. 6B).

• Open in new tab
• Download powerpoint

FIG. 6.

Spectral analyses of ∼3 · 10⁹ cells, resuspended in 1 ml of 60% sucrose at room temperature, from the wild type (solid lines) and the mutants PUCAΩ (dashed lines) and PUCCK (dotted lines) grown semiaerobically (A) or photosynthetically (B).

In the wild-type strain, two peaks are observed at 804 and 860 nm, corresponding mainly to LH2 absorption. The amplitudes of these two peaks are greater in cells grown under light in the absence of oxygen than in cells grown under semiaerobiosis. The peak at 860 nm presents a shoulder toward the higher wavelength corresponding to LH1 absorption. In PUCAΩ, the two absorbance peaks at 875 and 802 nm correspond to the LH1 and RC major infrared bands, respectively. They are observed irrespective of the growth conditions. The absorbance spectrum of PUCCK grown in semiaerobiosis is very similar to that of PUCAΩ, with two peaks at 802 and 875 nm, indicating that PUCCK grown under these conditions has an LH2-deficient phenotype. In the spectrum of PUCCK grown under photosynthetic conditions, the peak amplitudes are increased and the 875-nm peak is less symmetrical, with a slight shoulder toward the lower wavelengths, indicating the possible presence of LH2 complexes.

Analyses of the mutant strains indicate that PUCAΩ is an LH2-deficient mutant and that PUCCK is a conditional LH2-deficient mutant that synthesizes these complexes only under photosynthetic conditions. To confirm these hypotheses and for a better resolution of the absorbance bands of the LH1, RC, and LH2 complexes, the low-temperature absorption spectra of the membranes have been measured for the wild type and for the two mutants grown under semiaerobiosis (Fig. 7A) or photosynthetic conditions (Fig. 7B). In the wild type, three infrared bands were resolved under both culture conditions. Two peaks at 812 and 874 nm are attributed to LH2, and one at 894 nm is attributed to LH1, as deduced from analyses of low-temperature spectra of isolated complexes (14). These bands overlap and hide the RC absorption and are red shifted compared with bands observed at room temperature. In PUCAΩ, two bands are observed at 802 and 894 nm. The former is attributed to the major RC absorption (which shifts only slightly with temperature), and the latter is attributed to LH1. Under semiaerobiosis, the spectrum collected from PUCCK membranes is nearly identical to that of PUCAΩ except for a slight broadening of the 894-nm band, which is at the limit of experimental accuracy. However, if this broadening originated from LH2 in small amounts, one would expect a concomitant shift of the 802-nm band toward 812 nm, which is not observed. We therefore conclude from these low-temperature absorption spectra that no detectable LH2 complex is present in the PUCCK mutant under semiaerobic conditions. Under photosynthesis, however, PUCCK presents the same infrared absorption as the wild type, but with different relative amplitudes. This demonstrates that PUCCK contains LH2 complexes, but in smaller amounts than the wild type. The semiaerobic photosynthetic-shift experiments confirm the conditional phenotype of PUCCK. Indeed, when the culture is shifted from photosynthesis to semiaerobiosis, a peak at 894 nm is observed, and when it is shifted back to photosynthesis, PUCCK again presents a shoulder at 874 nm. These results indicate that PucC is required for LH2 formation under semiaerobiosis. However, it is not essential for the formation of the LH2 complex under photosynthesis.

• Open in new tab
• Download powerpoint

FIG. 7.

Spectral analysis at low temperature (77 K) of the membrane fraction, resuspended in 60% glycerol, isolated from the wild type (solid lines) and the mutants PUCAΩ (dashed lines) and PUCCK (dotted lines) grown semiaerobically (A) or photosynthetically (B). All spectra are normalized with a minimal optical density of 0 and a maximal optical density of 1.

Role of PucA C-terminal extension in the assembly of LH2.Compared to LH2 from R. palustris, R. acidophila, and R. capsulatus, the α apoprotein (71 amino acids) of R. gelatinosus has a C-terminal hydrophobic extension (Fig. 1A). Characterization of LH2 complexes isolated from R. gelatinosus has shown an α apoprotein heterogeneity generated during purification by an endogenous protease (28). Mass spectrometry has demonstrated that the major cleavage site is located between alanine 64 (A64) and A65 and that proteolysis continues until A53. Cleavage until A51 could also be performed by limited thermolysin digestion. However, loss of as many as 21 amino acids from the C terminus did not affect the LH2 structure in vitro. To check its role in vivo, we decided to delete this extension.

Prior to mutagenesis, we analyzed the minimal sequence necessary and sufficient for the expression of pucBA. A set of replicative plasmids carrying sequences of different lengths upstream of pucB (776 to 495 bp) were constructed. The 596 bp upstream of the pucB initiation codon sequence are necessary and sufficient for pucBA gene expression (data not shown). This sequence should form the upstream regulatory sequence involved in the binding of various transcriptional factors, as observed in R. sphaeroides and R. capsulatus (20, 23). To study the function of the C-terminal extension of PucA, we subcloned a SacI/HindIII fragment from pA100 containing 776 bp upstream of the pucB initiation sequence and the total coding sequence of pucB and pucA into a replicative plasmid, giving the plasmid pbcA1.

Derivatives of the pucA gene in which a stop codon was introduced to generate a α apoproteins lacking 13 and 18 amino acids were constructed from the expression vector pbcA1, giving the plasmids pbcA1-13 and pbcA1-18, respectively. These deletions resulted in LH2 α subunits ending at A58 and A53, respectively. The LH2-deficient strain PUCAΩ was transformed either with pbcA1-13 or with pbcA1-18. Transformants were selected and grown under photosynthetic conditions. The infrared absorption spectrum analyses (Fig. 8A) show that strains transformed with the plasmid pbcA1-18 do not produce LH2 complexes, while the strain with pbcA1-13 recovered LH2 production. The puc gene transcripts have also been analyzed in PUCAΩ and in PUCAΩ complemented by pbcA1-13 or by pbcA1-18. As expected, no pucBA transcripts were detected in PucAΩ. In both complementation strains, the pucBA transcripts are still present (Fig. 8B). These results together indicate that the last 13 amino acids on the C-terminal end of PucA are not required for LH2 assembly in the membrane, while truncated PucA without the last 18 amino acids cannot assemble with PucB to form the LH2 complex.

• Open in new tab
• Download powerpoint

FIG. 8.

(A) Spectral analysis at low temperature (77 K) of the membrane fraction, resuspended in 60% glycerol, from the PUCAΩ mutant transformed by a plasmid carrying either the wild-type pucA gene (pbcA1) (solid line) or the mutated pucA gene leading to an α apoprotein lacking the last 13 (pbcA1-13) (dashed line) or 18 (pbcA-18) (dotted line) C-terminal amino acids. Each complemented strain was grown under photosynthetic conditions. All spectra are normalized with a minimal optical density of 0 and a maximal optical density of 1. (B) RNA blot analysis of the pucBA transcripts from the strains PUCAΩ (lane 1) and PUCAΩ transformed by pbcA1-18 (lane 2) or pbcA1-13 (lane 3) grown under photosynthetic conditions. Ten micrograms of total RNA was loaded per lane. The membrane was hybridized with probe 1, shown in Fig. 2A.