Inactivation of Mg Chelatase during Transition from Anaerobic to ...

Download PDF
4 downloads 0 Views 289KB Size Report
Comment
Sep 18, 2002 - BchH binds protoporphyrin IX, and BchH that has been expressed and .... dard Zn-protoporphyrin IX was dissolved in diethyl ether, and the ...
JOURNAL OF BACTERIOLOGY, June 2003, p. 3249–3258 0021-9193/03/$08.00⫹0 DOI: 10.1128/JB.185.11.3249–3258.2003 Copyright © 2003, American Society for Microbiology. All Rights Reserved.

Vol. 185, No. 11

Inactivation of Mg Chelatase during Transition from Anaerobic to Aerobic Growth in Rhodobacter capsulatus Robert D. Willows,1 Vanessa Lake,1 Thomas Hugh Roberts,1 and Samuel I. Beale2* Department of Biological Sciences, Macquarie University, North Ryde, 2109 Australia,1 and Division of Biology and Medicine, Brown University, Providence, Rhode Island 029122 Received 18 September 2002/Accepted 11 March 2003

The facultative photosynthetic bacterium Rhodobacter capsulatus can adapt from an anaerobic photosynthetic mode of growth to aerobic heterotrophic metabolism. As this adaptation occurs, the cells must rapidly halt bacteriochlorophyll synthesis to prevent phototoxic tetrapyrroles from accumulating, while still allowing heme synthesis to continue. A likely control point is Mg chelatase, the enzyme that diverts protoporphyrin IX from heme biosynthesis toward the bacteriochlorophyll biosynthetic pathway by inserting Mg2ⴙ to form Mgprotoporphyrin IX. Mg chelatase is composed of three subunits that are encoded by the bchI, bchD, and bchH genes in R. capsulatus. We report that BchH is the rate-limiting component of Mg chelatase activity in cell extracts. BchH binds protoporphyrin IX, and BchH that has been expressed and purified from Escherichia coli is red in color due to the bound protoporphyrin IX. Recombinant BchH is rapidly inactivated by light in the presence of O2, and the inactivation results in the formation of a covalent adduct between the protein and the bound protoporphyrin IX. When photosynthetically growing R. capsulatus cells are transferred to aerobic conditions, Mg chelatase is rapidly inactivated, and BchH is the component that is most rapidly inactivated in vivo when cells are exposed to aerobic conditions. The light- and O2-stimulated inactivation of BchH could account for the rapid inactivation of Mg chelatase in vivo and provide a mechanism for inhibiting the synthesis of bacteriochlorophyll during adaptation of photosynthetically grown cells to aerobic conditions while still allowing heme synthesis to occur for aerobic respiration. The purple photosynthetic bacteria are able to derive their cellular energy from light, inorganic compounds, or organic compounds depending on the chemical and physical environment. This remarkable versatility requires that each mode of energy metabolism be regulated so as to prevent unnecessary biosynthesis of alternative energy systems. Adaptation from one metabolic system to another is usually triggered by an environmental or chemical signal. During the anaerobic photosynthetic mode of growth in Rhodobacter capsulatus, genes that encode components needed for photosynthesis and the biosynthesis of pigments are coordinately regulated by aerobic repression circuits, anaerobic induction circuits, and a light control circuit (for a review, see reference 1). Although much of the research in this area has concentrated on adaptation from aerobic growth to photosynthetic growth, adaptation in the reverse direction also occurs (4, 9, 22). The coordinated transcriptional control circuits would be expected to function well for adaptation to photosynthetic growth. However, this mode of regulation may not be sufficient during adaptation from photosynthetic growth to aerobic growth, because many of the pigment biosynthetic enzymes that are present in anaerobically growing cells may continue to be active even after the synthesis of pigment-binding proteins has stopped. Continued pigment synthesis under these conditions is potentially lethal because bacteriochlorophyll and its precursors can be phototoxic when not bound in light-harvesting complexes or reaction centers. However, during adaptation

from photosynthetic growth to aerobic growth in the light, bacteriochlorophyll biosynthesis is almost immediately halted (4). This observation cannot be explained by the relatively slower transcriptional control of expression of bacteriochlorophyll biosynthetic genes and indicates that one or more enzymes involved in bacteriochlorophyll biosynthesis must be rapidly inactivated during adaptation to aerobic growth. There is a requirement for the inactivation of bacteriochlorophyll synthesis to be located specifically in the bacteriochlorophyll branch of the tetrapyrrole biosynthetic pathway, because heme biosynthesis must be maintained for aerobic metabolism. Mg chelatase is the first enzyme that is unique to bacteriochlorophyll biosynthesis, and it is therefore the likely target for this type of inactivation. During previous studies on the Mg chelatase of R. capsulatus, it was discovered that the BchH subunit of the Mg chelatase is susceptible to inactivation by light (23). We hypothesized that this inactivation may have a useful function in vivo by preventing accumulation of potentially toxic levels of bacteriochlorophyll and its precursors during adaptation. The results presented here indicate that the inactivation of BchH by light is enhanced by O2 and results in the formation of covalent adducts between BchH and its bound protoporphyrin IX substrate. BchH is the rate-limiting component of Mg chelatase in cell extracts, and its selective inactivation during adaptation to aerobic growth may account for the rapid inactivation of Mg chelatase in vivo when anaerobically growing cells are exposed to O2 in the light.

* Corresponding author. Mailing address: Division of Biology and Medicine, Brown University, Providence, RI 02912. Phone: (401) 8633129. Fax: (401) 863-1182. E-mail: [email protected].

Growth of R. capsulatus. R. capsulatus wild-type strain SG1001 and BchHdeficient strain ZY6 (3, 24) were gifts from C. E. Bauer (Indiana University, Bloomington, Ind.) and D. W. Bollivar (Illinois Wesleyan University, Blooming-

MATERIALS AND METHODS

3249

3250

WILLOWS ET AL.

ton, Ill.). Cells were grown in 200-ml Macartney bottles that were almost completely filled with RCV medium (3) and incubated in the light (250 ␮mol m⫺2 s⫺1 supplied by cool-white fluorescent tubes) at 28°C until they had reached the mid-exponential growth phase. For experiments to examine adaptation, the 200-ml cultures were diluted with 200 ml of fresh fully oxygenated RCV medium and aerated and shaken at 200 rpm on an orbital shaker in 2-l Erlenmeyer flasks in continued light or complete darkness. At the specified times, cultures were rapidly chilled in an ice-water bath, and the cells were harvested by centrifugation at 10,000 ⫻ g for 10 min. Preparation of cell-free R. capsulatus extracts. Sedimented cells from 400 ml of culture were resuspended to a maximum volume of 8 ml in a mixture of 50 mM Tricine-NaOH (pH 8.0), 10 mM MgCl2, and 4 mM dithiothreitol (DTT) at 0°C and lysed by passage through a French pressure cell at 1.38 ⫻ 108 Pa. ATP was immediately added to 4 mM, and the extract was centrifuged at 105,000 ⫻ g for 60 min. The supernatant was removed and used for assays of ␦-aminolevulinic acid (ALA) synthase (ALAS), porphobilinogen (PBG) synthase (PBGS), and Mg chelatase, as described below. The sediment was used for assay of ferrochelatase, as described below. Recombinant R. capsulatus Mg chelatase proteins. BchD and BchH proteins containing NH2-terminal His6 tags (HisBchD and HisBchH, respectively) and BchI were expressed in Escherichia coli from the respective cloned R. capsulatus genes and purified as described previously (23). For recombinant HisBchH, all expression and purification steps were done in the dark or under a dim green safe light (23). ALAS assay. ALAS activity was measured by the method described by Beale et al. (2). Cell extract (100 ␮l) was added to 1 ml (final volume) of solution containing 75 mM Tris (pH 7.8), 50 mM glycine, 25 mM succinic acid, 20 mM MgCl2, 10 mM levulinic acid, 5 mM EDTA, 0.1 mM coenzyme A, 0.1 mM pyridoxal phosphate, and 5 mM ATP. After 30 min of incubation at 30°C, the reaction was stopped by the addition of 50 ␮l of 100% (wt/vol) aqueous trichloroacetic acid. Precipitated proteins were removed by centrifugation for 2 min at 14,000 ⫻ g. Then, 500 ␮l of the clear supernatant was combined with 250 ␮l of 1 M sodium acetate (pH 6.0; final pH, 6.8) and 25 ␮l of acetylacetone. After being mixed, the sample was heated for 15 min at 95°C and then cooled to room temperature. An aliquot (500 ␮l) was mixed with an equal volume of Ehrlich Hg reagent (1 g of p-dimethylaminobenzaldehyde, 42 ml of glacial acetic acid, 8 ml of 70% [wt/wt] perchloric acid, 0.175 g of HgCl2), and the color was allowed to develop for 15 min. The product was detected spectrophotometrically by absorbance at 554 nm, and the original ALA concentration was calculated with the use of a molar absorption coefficient of 6.8 ⫻ 104 (15). PBGS assay. PBGS activity was measured by the method described by Senior et al. (20). Cell extract (25 ␮l) was preincubated for 5 min at 30°C in a solution containing 50 mM Tricine-NaOH (pH 8.0), 5 mM MgCl2, 100 ␮M ZnCl2, and 2 mM DTT. The reaction was started by the addition of ALA (final concentration, 5 mM) to give a final volume of 500 ␮l, and the mixture was incubated for 30 min at 30°C. The reaction was stopped by the addition of 500 ␮l of 0.1 M HgCl2 in 10% (wt/vol) trichloroacetic acid. Precipitated proteins were removed by centrifugation for 2 min at 14,000 ⫻ g. Then, 500 ␮l of the clear supernatant was mixed with an equal volume of Ehrlich reagent (1 g of p-dimethylaminobenzaldehyde, 42 ml of glacial acetic acid, 8 ml of 70% [wt/wt] perchloric acid), and the color was allowed to develop for 15 min. The product was detected spectrophotometrically by absorbance at 555 nm, and the original PBG concentration was calculated with the use of a molar absorption coefficient of 6.02 ⫻ 104. Ferrochelatase assay. Ferrochelatase activity was measured by the method described by Neuberger and Tait (17). Cell extract (25 ␮l) was incubated for 30 min at 30°C in a solution containing 50 mM Tricine-NaOH (pH 8.0), 5 mM MgCl2, 100 ␮M ZnSO4, and 10 ␮M protoporphyrin IX with a final volume of 120 ␮l. The reaction was stopped by the addition of 1 ml of acetone:water:14 M NH4OH (80:20:1, vol/vol/vol) and 200 ␮l of hexane. After being mixed, the precipitated proteins were removed and the phases were separated by centrifugation for 5 min at 14,000 ⫻ g. The reaction product, Zn-protoporphyrin IX, was detected fluorometrically in the lower acetone phase. The excitation wavelength used was 418 nm, and the fluorescence intensity was recorded at 592 nm. Standard Zn-protoporphyrin IX was dissolved in diethyl ether, and the concentration was determined from the absorbance at 415 nm, using the molar absorption coefficient of 2.40 ⫻ 105 (17). Fluorescence standards were prepared by dilution of this solution into acetone. Mg chelatase assay. A modification of the method of Gorchein (8) was used. Glycerol (final concentration 10%, vol/vol) was added to the supernatant from the ultracentrifugation step, and 1-ml portions were desalted on a NAP10 column (Amersham/Pharmacia Biotech., Piscataway, N.J.) and equilibrated with 10% (vol/vol) glycerol–50 mM Tricine-NaOH (pH 8.0)–10 mM MgCl2–4 mM DTT–4 mM ATP. Protein concentration was determined and adjusted to 2.5

J. BACTERIOL. mg/ml before assaying. Desalted cell extract (80 ␮l) was incubated at 30°C in a total volume of 100 ␮l of solution containing 25 mM Tricine-NaOH (pH 8.0), 10 mM MgCl2, 0.025% (wt/vol) NaN3, 4 mM ATP, 15 mM phosphocreatine, 1 unit of creatine kinase/ml, and 5 ␮M protoporphyrin IX. After incubation for 15 min, 50 ␮l of reaction mixture was removed and added to 1 ml of acetone:water:14 M NH4OH (80:20:1, vol/vol/vol). The remaining 50 ␮l was incubated for a further 45 min, and then the reaction was stopped by the addition of 1 ml of acetone: water:14 M NH4OH (80:20:1, vol/vol/vol). Hexane (200 ␮l) was added to all stopped reactions, and after mixing, the precipitated proteins were removed and the phases were separated by centrifugation for 5 min at 14,000 ⫻ g. The reaction product, Mg-protoporphyrin IX, was detected fluorometrically in the lower acetone phase. The excitation wavelength used was 418 nm, and the fluorescence intensity was recorded at 595 nm. Mg chelatase activity was calculated from the difference in product formed between the 15-min and 60-min samples to allow for the fact that an initial 10- to 15-min lag period occurs before the linear phase begins (23). Standard Mg-protoporphyrin IX was dissolved in diethyl ether, an aliquot was added to 5% (wt/vol) HCl to convert it to protoporphyrin IX, and the concentration was determined from the absorbance at 407.5 nm with a molar absorption coefficient of 2.78 ⫻ 105 (5). Fluorescence standards were prepared by dilution of the Mg-protoporphyrin IX solution into acetone. In vitro photoinactivation of BchH. HisBchH (2 mg/ml) was irradiated on ice with 240 ␮mol m⫺2 s⫺1 of light from cool-white fluorescent tubes in the presence of air for aerobic samples or in O2-free N2 after repeated degassing with O2-free N2 for lowered-O2 samples. Assays contained 50 ␮g of HisBchH, 5 ␮g of HisBchD, and 10 ␮g of BchI in 100 ␮l of 50 mM Tricine-NaOH (pH 8.5), 15 mM MgCl2, 1 mM DTT, 4 mM ATP, 10 mM phosphocreatine, and 2 U of creatine phosphokinase. Measurement of cellular protoporphyrin IX, Mg-protoporphyrin IX, and bacteriochlorophyll. Portions (100 ␮l) of the French pressure cell homogenates of R. capsulatus cells were added to 900 ␮l of acetone:water:14 M NH4OH (80:20:1, vol/vol/vol), and after being mixed, the mixtures were centrifuged for 5 min at 14,000 ⫻ g to remove cell debris and precipitated proteins. Hexane (200 ␮l) was added to the supernatant, and the two phases were allowed to separate. The concentrations of protoporphyrin IX and Mg-protoporphyrin IX were determined in the lower acetone phase spectrofluorometrically with excitation at 400 nm and emission at 630 nm for protoporphyrin IX and excitation at 418 nm and emission at 595 nm for Mg-protoporphyrin IX. Quenching was estimated by addition of a fixed amount of authentic protoporphyrin IX and Mg-protoporphyrin IX to each sample, and the fluorescence intensities were reread. No significant fluorescence quenching was observed in any samples. Fluorometric standard solutions of protoporphyrin IX and Mg-protoporphyrin IX were prepared as described above. The bacteriochlorophyll concentration was determined in upper hexane phase by measuring the absorbance at 770 nm and using the molar absorption coefficient of 9.11 ⫻ 104 reported for the pigment in diethyl ether (21). Affinity purification of BchH antibody. Purified His-BchH (1 mg) was coupled to a 1-ml HiTrap N-hydroxysuccinimide (NHS)-activated Sepharose column (Amersham/Pharmacia) in accordance with the manufacturer’s instructions. The coupling efficiency was 90%. Anti-BchH antiserum from rabbits (1 ml) was loaded onto this column, and the column was washed with 0.1 M Tris-HCl (pH 8.0) (10 ml) and then with 0.01 M Tris-HCl (pH 8.0) (10 ml). The bound immunoglobulin G was then eluted with 0.1 M glycine (pH 3.0), and 500-␮l fractions were collected into tubes containing 100 ␮l of 1 M Tris-HCl (pH 8.0). Immunoblots blots were performed with a 1:500 dilution of the affinity purified antibodies. Other methods. Protein concentration was determined by using Bio-Rad (Hercules, Calif.) protein estimation reagent according to the manufacturer’s instructions. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis was done according to the procedure of Fling and Gregerson (6). Gels were stained with colloidal Coomassie brilliant blue (18). Polyclonal antibodies were raised against the BchH, BchI, and BchD proteins in rabbits. Antibodies were partially purified by cross-absorbing the serum to an E. coli lysate column (Pierce, Rockford, Ill.) in accordance with the manufacturer’s instructions before use. Purified antibodies were used at 1:2,000 dilution in immunoblots by the method of Harlow and Lane (10), using alkaline phosphatase-conjugated goat antirabbit antibody for detection. Peptide mass fingerprinting by electrospray mass spectroscopy was performed by the Australian Proteome Analysis Facility, Macquarie University. HPLC separation of tryptic digestion products was done using the SMART system (Pharmacia/Amersham) and a Brownlee column (220 mm long with a 2.1-mm diameter; Perkin Elmer, Wellesley, Mass.) containing 5-␮m particles of PTH C18 with a mobile phase gradient of 5 to 95% (vol/vol) aqueous acetonitrile in 0.1% (wt/vol) trifluoroacetic acid and flowing at 200 ␮l/min. The effluent was monitored at 214, 280, and 405 nm.

VOL. 185, 2003

Mg CHELATASE INACTIVATION

FIG. 1. In vitro inactivation of HisBchH by light under atmospheric (F) and lowered (E) O2 concentration. The assays were done in duplicate and were monitored at 5-min intervals. Activities shown are the linear rates after the initial 5- to 10-min lag phase typically found with this reaction. Activities of control samples (containing unilluminated HisBchH) for atmospheric and low-O2 treatments were 899 and 476 pmol h⫺1 ␮g⫺1 of BchD, respectively. Duplicate samples for all experimental points agreed to within 2%.

Chemicals. Protoporphyrin IX, Mg-protoporphyrin IX, and Zn-protoporphyrin IX were purchased from Porphyrin Products, Inc. (Logan, Utah). Except where indicated otherwise, all other chemicals were from Sigma Chemical Co. (St. Louis, Mo.).

RESULTS In vitro photoinactivation of BchH. In vitro inactivation of HisBchH protein by light under aerobic and O2-depleted conditions was determined in assays containing reconstituted recombinant Mg chelatase components. White light (250 ␮mol m⫺2 s⫺1 from cool-white fluorescent tubes) rapidly and irreversibly inactivated the HisBchH protein, and the inactivation was more rapid and complete at atmospheric O2 concentration than at lowered O2 concentration (Fig. 1). Oxygen alone was not sufficient to inactivate HisBchH under these conditions. This is shown by comparing the unilluminated (zero time) samples for both treatments. The initial low-O2 sample had 60% of the activity of the initial atmospheric O2 sample, and the difference is attributed to the occurrence of some denaturation of the HisBchH when O2 was depleted from the sample by repeated degassing in a vacuum dessiccator followed by flushing with O2-free N2. It should be noted that some O2 probably remained in the O2-depleted sample, even after repeated degassing cycles, and the residual O2 might have been responsible for the slow photoinactivation of HisBchH in this sample. Spectral changes during in vitro photoinactivation of BchH. Recombinant HisBchH protein, after expression and purification in the dark or under dim green safe light, contained

3251

0.8 to 1.0 mol of protoporphyrin IX per mol of protein. Changes were observed in the absorption and fluorescence emission spectra of HisBchH after irradiation (Fig. 2). The absorbance in the 280 and 410 nm regions decreased. The decrease at 410 nm is due to photobleaching of the bound pigment, and this absorption progressively decreases with increasing light exposure time. In contrast, the decrease in absorbance of protein tryptophan at 280 nm is largely complete after 5 min of light exposure. This change indicates a rapid change in the environment of one or more tryptophan residues and/or covalent modification of one or more of them. There are 15 tryptophan residues in HisBchH, and the relatively small change observed at 280 nm may indicate that a single tryptophan residue is responsible for the change. The fluorescence emission spectra indicate that more than one product is formed, because there are no isosbestic points. The appearance of an emission peak around 670 nm suggests that one of the products may have a structure similar to that of photoprotoporphyrin IX (13), but free photoprotoporphyrin IX cannot be a major final product because this peak is seen only in the covalently bound pigment-protein complex (see below). Covalent attachment of protoporphyrin IX to HisBchH. HisBchH produces a straw-colored protein precipitate when it is denatured with 9 volumes of acetone:water:14 M NH4OH (80:20:1, vol/vol/vol), and the released protoporphyrin IX is found in the supernatant. In contrast, HisBchH that has been irradiated in the presence of O2 for 5 min produces a redbrown precipitate upon denaturation, indicating that some or all of the protoporphyrin IX remains bound to the protein after denaturation and suggesting that it is covalently bound to the protein. By determining the difference in the concentration of protoporphyrin IX that was released into the denaturing supernatants from irradiated and nonirradiated samples, it was calculated that 26 to 28% of the protoporphyrin IX that was bound to native HisBchH was retained by the denatured irradiated protein. Irradiated and nonirradiated HisBchH was digested with trypsin, and mass spectra were obtained for each peptide mixture. The two samples gave nearly identical mass spectra, and most of the major ions could be assigned to peptide fragments expected from either a trypsin digest of HisBchH or autodigestion of trypsin itself (Fig. 3). The peptide ions assigned to HisBchH comprise 57% of the protein. Six major differences were observed between the spectra of irradiated and nonirradiated HisBchH. Two new mass signals were observed in the irradiated sample at 1,044 and 1,060 m/z, and the other four differences were strong reductions in the intensities of four mass signals at 1,769, 1,860, 2,023, and 2,533 m/z. The two new mass signals were also found in some trypsin-treated control samples not containing HisBchH and are therefore unlikely to be due to HisBchH peptide fragments. Mass signals 1,860, 2,023, and 2,533 correspond to amino acids 122 to 138, 637 to 648, and 614 to 636, respectively, in the HisBchH sequence. It is possible that one or more of these regions form part of the protoporphyrin IX binding site. The mass signal at 1,769 could not be assigned. We attempted to identify tryptic digestion fragments from irradiated HisBchH that contain bound protoporphyrin IX derivatives by subjecting HPLC-separated fragments to mass

3252

WILLOWS ET AL.

J. BACTERIOL.

FIG. 2. Absorbance and fluorescence emission spectra of illuminated HisBchH. Purified recombinant His-tagged R. capsulatus BchH protein was illuminated for the times indicated below with light from cool-white fluorescent tubes at an intensity of 240 ␮mol photons m⫺2 s⫺1. (A) Absorption spectra at 0 min (——), 5 min (- - -), 15 min (– – –), and 30 min ( 䡠 䡠 䡠 䡠 䡠 ) of illumination. (B) Fluorescence emission spectra at 0 min (——), 0.5 min (- - -), 2 min (– – –), 5 min ( 䡠 䡠 䡠 䡠 䡠 ), and 10 min (– 䡠 – 䡠 –) of illumination. The fluorescence excitation wavelength was 408 nm.

VOL. 185, 2003

Mg CHELATASE INACTIVATION

3253

FIG. 3. Mass spectra of tryptic digests of HisBchH before illumination (A) and after irradiation for 5 min with light from cool-white fluorescent tubes at an intensity of 250 ␮mol photons m⫺2 s⫺1 (B) and the primary structure of HisBchH (C). Underlined residues in panel C indicate peptides that were identified in the mass spectra, and the residues shown in bold type indicate peptide fragments that are strongly reduced or absent in the spectrum of irradiated HisBchH. These signals are marked with asterisks in panel A.

spectral analysis. Several late-eluting peptide peaks that absorbed at 405 nm were present in the HPLC eluates of tryptic digests from irradiated HisBchH samples, indicating the presence of protoporphyrin IX derivatives in these digests. One of the fractions yielded a doubly charged mass signal of 748.5, which, after subtracting the mass of two protons and protoporphyrin IX, has a net peptide mass of 932.55. This mass corresponds to the mass of a predicted HisBchH tryptic fragment, (K)MLKFIPGK(A). However, further analysis of this parent ion by tandem mass spectroscopy did not yield any signals attributable to protoporphyrin IX, possibly because of poor ionization. Poor ionization may also account for the absence of signals from pigment-containing digestion fragments in the mass spectra of unseparated digests of irradiated HisBchH. No

405-nm absorbance peaks were detected in the HPLC eluates of digestion products from unirradiated HisBchH or from undigested irradiated HisBchH. Changes in cellular content of Mg chelatase components during adaptation of photosynthetically growing cells to aerobic metabolism. R. capsulatus cells that had been growing photosynthetically were transferred to aerobic conditions in either the light or the dark, and samples were taken over a 30-min period. Immunoblots of these fractions indicate that there is relatively little variation in the levels of the BchD, BchH, and BchI proteins during this treatment (Fig. 4). Tetrapyrrole concentrations during adaptation. Cells that had been growing photosynthetically under anaerobic conditions were transferred to aerobic conditions in either the light

3254

WILLOWS ET AL.

J. BACTERIOL.

FIG. 4. Immunoblots of sodium dodecyl sulfate-polyacrylamide gel electrophoresis-separated proteins extracted from R. capsulatus during adaptation to aerobic conditions in the light or the dark. Immunoblots were probed with anti-HisBchD (A), anti-BchI (B), or anti-His-BchH (C). For all immunoblots, lane are identified as follows: lane 1, anaerobic control; lane 2, 5 min light; lane 3, 5 min dark; lane 4, 15 min light; lane 5, 15 min dark; lane 6, 30 min light; lane 7, 30 min dark. For immunoblot A, the rightmost lane contains 5 ng of purified recombinant HisBchD; for immunoblot B, the rightmost lane contains 5 ng of purified BchI; for immunoblot C, the leftmost lane contains 100 ng of purified recombinant His-BchH. The top immunoreactive band corresponds with BchH, while the lower band may be the CobN subunit of cobaltochelatase, which is similar to, but smaller than, BchH.

or the dark. In the first 30 min following the transfer, the cellular levels of bacteriochlorophyll remained relatively constant, whether the cells were kept in the light or transferred to the dark (Table 1). In contrast, the levels of protoporphyrin IX

and Mg-protoporphyrin IX rapidly declined following transfer to aerobic conditions. The decreases in the levels of these pigments occurred more rapidly in the light than in the dark, especially that of Mg-protoporphyrin IX.

TABLE 1. Cellular pigment concentrations during adaptation to aerobic conditionsa Illumination

Time (min)

Protoporphyrin IX concn

Mg-protoporphyrin IX concn

Bacteriochlorophyll a concn

Light

0 5 15 30

0.033 ⫾ 0.002 (100) 0.011 ⫾ 0.001 (33) 0.009 ⫾ 0.002 (27) 0.010 ⫾ 0.001 (30)

0.069 ⫾ 0.003 (100) 0.011 ⫾ 0.000 (16) 0.006 ⫾ 0.004 (9) 0.009 ⫾ 0.000 (13)

20.7 ⫾ 1.6 (100) 19.8 ⫾ 0.9 (96) 23.6 ⫾ 0.5 (114) 24.6 ⫾ 0.7 (119)

Dark

5 25 30

0.023 ⫾ 0.002 (70) 0.021 ⫾ 0.001 (64) 0.018 ⫾ 0.001 (55)

0.040 ⫾ 0.003 (58) 0.039 ⫾ 0.002 (56) 0.038 ⫾ 0.002 (55)

21.8 ⫾ 0.5 (105) 22.0 ⫾ 0.0 (106) 22.6 ⫾ 0.9 (109)

a All pigment concentrations are expressed as nmol/mg of protein. Numbers in parentheses are percentages of initial values immediately before the cells were exposed to aerobic conditions. All values are the means of duplicate samples from the same cultures, and the tabulated errors are the standard deviations.

VOL. 185, 2003

Mg CHELATASE INACTIVATION

3255

TABLE 2. Enzyme activities in cell extracts after transfer of the cells to aerobic conditionsa Activity of: Illumination

Time (min)

ALAS (nmol/8)

PBGS (nmol/4)

Ferrochelatase (nmol)

Mg chelatase (nmol)

Mg chelatase (nmol)b

Light

0 5 15 30

19.4 ⫾ 0.8 (100) 19.7 ⫾ 0.0 (102) 19.1 ⫾ 0.3 (98) 21.5 ⫾ 0.3 (111)

112 ⫾ 13 (100) 132 ⫾ 18 (118) 144 ⫾ 17 (129) 109 ⫾ 2 (97)

0.966 ⫾ 0.089 (100) 0.831 ⫾ 0.095 (86) 0.990 ⫾ 0.076 (102) 1.083 ⫾ 0.020 (112)

0.0153 ⫾ 0.0011 (100) 0.0085 ⫾ 0.0012 (56) 0.0042 ⫾ 0.0003 (27) 0.0046 ⫾ 0.0006 (30)

0.533 ⫾ 0.079 (100) 0.393 ⫾ 0.024 (74) 0.362 ⫾ 0.031 (68) 0.356 ⫾ 0.010 (67)

Dark

5 15 30

20.2 ⫾ 0.5 (104) 20.4 ⫾ 0.3 (105) 23.3 ⫾ 0.6 (120)

114 ⫾ 8 (102) 116 ⫾ 13 (104) 157 ⫾ 24 (140)

1.078 ⫾ 0.134 (112) 1.063 ⫾ 0.064 (110) 0.913 ⫾ 0.070 (95)

0.0105 ⫾ 0.0021 (69) 0.0075 ⫾ 0.0019 (49) 0.0103 ⫾ 0.0026 (67)

0.464 ⫾ 0.040 (87) 0.445 ⫾ 0.039 (84) 0.358 ⫾ 0.029 (67)

a Chloramphenicol (20 ␮g/ml) was added to the cultures 20 min before the transfer to prevent protein synthesis from occurring after the transfer. All enzyme activities are expressed as nmol of product formed mg⫺1 protein h⫺1. To facilitate comparison of the enzyme activities, the tabulated amounts of each product formed were adjusted as indicated to show equivalent amounts of tetrapyrrole formation, taking into account that 8 mol of ALA or 4 mol of PBG are required to form 1 mol of tetrapyrrole. Numbers in parentheses are percentages of initial activities. All values are the means of results for duplicate samples from the same cultures, and the tabulated errors are the standard deviations. b Purified recombinant HisBchH (20 ␮g) was added to each sample before assay.

Activity of tetrapyrrole biosynthetic enzymes in vivo during adaptation. Cultures that had been growing photosynthetically were supplemented with chloramphenicol (20 ␮g/ml) to inhibit protein synthesis, and then 10 min later the cultures were transferred to aerobic conditions in either the light or the dark. Cells were collected at several time points after the transfer, and the activities of ALAS and PBGS, two enzymes that are common to the heme and bacteriochlorophyll biosynthetic pathways, and the activities of ferrochelatase and Mg-chelatase were determined. Ferrochelatase activity remained within 15% of its initial value throughout the course of the experiment (Table 2). The activities of ALAS and PBGS varied somewhat more widely, but none of these three enzymes underwent a significant decrease in activity. In contrast to the other three enzyme activities assayed, Mg chelatase activity decreased markedly over a 30-min period after the transfer. The rate of inactivation was significantly higher in cells that were kept in the light than in cells that were transferred to the dark. These results suggest that Mg chelatase is specifically inactivated after the cells are exposed to air and that the inactivation is stimulated by light. Addition of 20 ␮g of purified recombinant HisBchH protein to the cell extracts before assay greatly increased the in vitro Mg chelatase activity. Moreover, the addition of HisBchH to extracts of cells that were kept in the light increased the Mg chelatase activity to the level in the HisBchH-supplemented extracts of cells that were transferred to the dark. These results suggest that although all components of Mg chelatase decline in cells that are exposed to air, the BchH component is the one that is rate limiting and that the decline in BchH activity is stimulated by light. Although the in vitro Mg chelatase activity in extracts of cells undergoing the transition from anaerobic to aerobic growth decreased significantly, it nevertheless remained measurable for at least 30 min. This residual in vitro activity could be due to the presence in the cell extracts of BchH molecules that do not contain bound protoporphyrin IX. These BchH molecules would be active in the in vitro assay even though they would not contribute to the in vivo Mg chelatase activity (see Discussion). It is of interest that the activities of ALAS and PBGS, as

measured in vitro, were vastly greater than the activities of ferrochelatase and Mg chelatase, even when the activities are compared on an equivalent tetrapyrrole biosynthetic activity basis, which takes into account that 8 mol of ALA or 4 mol of PBG are required to form one mole of tetrapyrrole. However, it should be noted that unlike the other enzymes examined, ALAS is subject to feedback inhibition by heme, and the in vitro activity therefore may not accurately reflect the in vivo activity but instead may indicate the maximum potential in vivo activity. The activity of PBGS, which is not known to be allosterically regulated, appears not to be rate limiting and therefore does not affect the rate of synthesis of heme or bacteriochlorophyll. DISCUSSION It is known that porphyrins can be phototoxic to cells (16). In the presence of light and O2, they form singlet oxygen, which will then react indiscriminately with almost any organic molecule. Therefore, it is important that porphyrin concentrations be kept to minimum levels in cells that are exposed to light and O2. R. capsulatus cells that are undergoing the transition from anaerobic, photosynthetic growth to aerobic growth rapidly arrest the biosynthesis of photosynthetic components, including bacteriochlorophyll-binding antenna and reaction center proteins. However, this response by itself would be insufficient to prevent the continued synthesis of bacteriochlorophyll and its precursors by enzymes that are active in the cells at the time of the transfer to air. Our results indicate that Mg chelatase, the first step of the Mg branch of the tetrapyrrole biosynthetic pathway, is selectively inactivated in the cells and that this inactivation occurs more rapidly in cells that remain in the light than in cells that are transferred to the dark. In contrast to the instability of Mg chelatase, ALAS and PBGS, two enzymes that are involved in the synthesis of both heme and bacteriochlorophyll, as well as that of ferrochelatase, which catalyzes the terminal step of heme biosynthesis, appear to be relatively stable in photosynthetically growing cells that are exposed to aerobic conditions. The selective inactivation of Mg chelatase allows adapting cells to divert their tetrapyrrole biosynthetic capacity to the needs of aerobic energy metabo-

3256

J. BACTERIOL.

WILLOWS ET AL.

FIG. 5. Structures of protoporphyrin IX and two photoproducts collectively named photoprotoporphyrin.

lism, which includes allowing continued synthesis of heme for respiratory cytochromes. The purified BchH protein component of R. capsulatus Mg chelatase contains bound protoporphyrin IX. It is not surprising that the BchH protein with bound protoporphyrin IX is photosensitive in vitro. Our results suggest that R. capsulatus has taken advantage of this photosensitivity in vivo to increase the rate of Mg chelatase inactivation in aerobic cells in the light and thus to rapidly photoinhibit the formation of bacteriochlorophyll when cells are exposed to light and O2. This inactivation, and the subsequent bleaching of the protoporphyrin IX bound to the BchH, would help to prevent the accumulation of potentially lethal concentrations of bacteriochlorophyll and its precursors. Photodynamic inactivation of Mg chelatase probably occurs only in organisms that photosynthesize anaerobically and repress the formation of the photosynthetic apparatus in the presence of O2. The inactivation is likely to depend on the ability of the BchH to bind protoporphyrin IX with high affinity. This tight binding can additionally be used to advantage by R. capsulatus because it allows the cells to regulate how much protoporphyrin IX is diverted from heme biosynthesis into bacteriochlorophyll biosynthesis by simply regulating the amount of BchH that is synthesized. This effective sequestration of protoporphyrin IX by BchH is evident when the protein is overexpressed in E. coli, where the cells become reddish in color due to protoporphyrin IX bound to the BchH protein (23). In this context, it is relevant that Synechocystis sp. PCC 6803 contains an orthologue of BchH, ChlH, which did not remain red when purified (11). This suggests that in this oxygenic photosynthetic organism, the potential for photoinactivation of ChlH may be limited by a relatively low binding affinity for protoporphyrin IX. Protoporphyrin IX in solution is photodegraded to several products, including two isomers that are collectively named photoprotoporphyrin (Fig. 5) (13). In these pigments, the formal addition of two oxygen atoms to one or the other of the

two ␤-methyl, ␤⬘-vinyl pyrrole rings converts it to a ␤-methyl, ␤-hydroxy, ␤⬘-oxoethylidine dihydropyrrole ring. Photoprotoporphyrins are chlorins and have characteristic chlorin fluorescence emission spectra with a prominent peak at 670 nm. The emission spectrum of photoinactivated BchH contains a peak in this region, indicating that one or more compounds with electronic structures similar to those of photoprotoporphyrins may be among the photoproducts of BchH-bound protoporphyrin IX. Free photoprotoporphyrins cannot be major final products because the 670-nm emission peak is seen only in the covalently bound pigment-protein complex. However, photoprotoporphyrins are themselves photosensitizers (14), and it is possible that photoprotoporphyrins may be reaction intermediates in the conversion of noncovalently bound protoporphyrin IX to covalently bound pigments. The observation that signals corresponding to two nonadjoining BchH peptide fragments are strongly reduced or absent in the peptide mass spectrum of photoinactivated BchH, compared to their values in the spectrum of unirradiated BchH, suggests that the photoinactivation products contain pigment molecules that can be attached to the protein by two different covalent bonds. In view of the activation of the vinyl groups of protoporphyrin IX caused by illumination, as evidenced by the conversion to photoprotoporphyrins, it seems likely that the vinyl groups or their derivatives are the points of attachment of the photoactivated protoporphyrin IX to BchH. The very rapid, steep decrease in cellular Mg-protoporphyrin IX content that occurs within 5 min after photosynthetically growing cells are transferred to aerobic conditions in the light suggests that in vivo Mg-protoporphyrin IX formation abruptly halts upon the transfer. In contrast, although the in vitro Mg chelatase activity extracted from cells undergoing this transition decreases significantly, it nevertheless remains measurable for at least 30 min. It is likely that the residual in vitro activity is attributable to the fact that the extracted proteins may include some BchH molecules that do not contain bound protoporphyrin IX. These BchH molecules could include those that

VOL. 185, 2003

have recently been synthesized and have not yet bound protoporphyrin IX and those that have recently contributed protoporphyrin IX to form Mg-protoporphyrin IX in the Mg chelatase reaction and have not yet bound another protoporphyrin IX molecule. These pigment-free BchH molecules would not become photoinactivated and would therefore be active in the in vitro assay, where protoporphyrin IX is supplied as a substrate, but because they would transiently lack a substrate in vivo, they would not contribute to in vivo Mg-protoporphyrin IX formation. The cellular content of protoporphyrin IX, as well as that of Mg-protoporphyrin IX, decreases more rapidly in photosynthetically growing cells that are transferred to aerobic conditions in the light than in cells kept in the dark after transfer to aerobic conditions. Considering that some de novo biosynthesis of protoporphyrin IX probably occurs during this period, as is suggested by the relatively constant in vitro activities of ALAS and PBGS, two enzymes involved in the biosynthesis of protoporphyrin IX, as measured in extracts of adapting cells, the decrease in protoporphyrin IX content suggests that it is rapidly degraded during the transition. This degradation could be facilitated if a substantial fraction of the total cellular protoporphyrin IX is bound to BchH. To assess the reasonableness of this possibility, it can be calculated, given that the initial protoporphyrin IX content is 0.033 nmol/mg of protein (taken from Table 1) and that the relative molecular weight of BchH is approximately 129 (GenBank accession CAA77524), that even if all of the cellular protoporphyrin IX were bound to BchH in a 1:1 molar ratio, this protein would have to comprise only 0.4% of the total cell protein. If most of the cellular protoporphyrin IX were bound to BchH, this could provide an effective mechanism to regulate the proportional distribution of protoporphyrin IX to the synthesis of heme and bacteriochlorophyll in response to changes in the relative demand for these end products. In cells that are adapting to photosynthetic growth, transcriptionally regulated induction of BchH synthesis would cause protoporphyrin IX to be directed toward bacteriochlorophyll biosynthesis because of the high affinity of BchH for protoporphyrin IX. In contrast, in cells that are adapting to aerobic growth, the rapid photoinhibition of BchH by formation of protoporphyrin IX-BchH complexes would allow newly synthesized protoporphyrin IX to be used for heme biosynthesis without competition from binding to BchH. Added recombinant HisBchH greatly stimulates in vitro Mg chelatase activity in cell extracts, even in extracts of cells that have been transferred to aerobic conditions. This observation, together with the rapid decrease in the cellular Mg chelatase activity (Table 1) but no corresponding decrease of BchH, BchD, and BchI protein levels during adaptation (Fig. 4), suggests that it is inactivation of BchH rather than its degradation and/or turnover which is important. It is of interest that in angiosperm plants, the BchH orthologue, ChlH, undergoes diurnal variations in its cellular content and in the cellular content of its mRNA (12, 19). It appears that regulation of Mg chelatase activity via control over the expression and cellular content of the BchH/ChlH component of the enzyme, the component that interacts with the substrate protoporphyrin IX, has been conserved in photosynthetic organisms ranging from purple bacteria through angiosperm plants. Recent structural evidence suggests that in R. capsulatus, BchH may act as

Mg CHELATASE INACTIVATION

3257

a shuttle that brings protoporphyrin IX to a preformed multimeric BchD-BchI complex where catalysis occurs (7). As was discussed above, control over the concentration of BchH may provide a mechanism for selectively steering protoporphyrin IX, the last common precursor of heme and (bacterio)chlorophyll, into the two branches of the pathway. In conclusion, although transcriptional control of bacteriochlorophyll synthesis in R. capsulatus (1) is an adequate mechanism for increasing the synthesis of bacteriochlorophyll and its intermediates during adaptation to anaerobic photosynthetic growth, this mode of regulation is unlikely to be sufficiently rapid in arresting bacteriochlorophyll synthesis during adaptation to aerobic conditions, especially in the light, where the accumulation of tetrapyrroles would be deleterious. The results presented here indicate that down-regulation of bacteriochlorophyll synthesis in R. capsulatus is effected at the posttranslational level by specific aerobic inactivation of Mg chelatase, and that this process is aided by rapid photoinactivation of the BchH protein of the Mg chelatase. ACKNOWLEDGMENTS We thank C. E. Bauer and D. W. Bollivar for supplying R. capsulatus strains, K. M. Smith for information about photoprotoporphyrin, and A. Srivastava for helpful discussions. This work was supported by U.S. National Science Foundation Grant MCB-9506901 and U.S. Department of Energy Grant DEFG02-88ER13918 to S.I.B. and by a Macquarie University Research Grant and Australian Research Council Grant A09905713 to R.D.W. REFERENCES 1. Bauer, C. E., and T. H. Bird. 1996. Regulatory circuits controlling photosynthesis gene expression. Cell 85:5–8. 2. Beale, S. I., T. Foley, and V. Dzelzkalns. 1981. ␦-Aminolevulinic acid synthase from Euglena gracilis. Proc. Natl. Acad. Sci. USA 78:1666–1669. 3. Bollivar, D. W., J. Y. Suzuki, J. T. Beatty, J. M. Dobrowolski, and C. E. Bauer. 1994. Directed mutational analysis of bacteriochlorophyll a biosynthesis in Rhodobacter capsulatus. J. Mol. Biol. 237:622–640. 4. Cohen-Bazire, G., W. R. Sistrom, and R. Y. Stanier. 1957. Kinetic studies of pigment synthesis by non-sulfur purple bacteria. J. Cell. Comp. Physiol. 49:25–68. 5. Dalziel, K. 1969. Porphyrins and related compounds, p. 128–145. In R. M. C. Dawson (ed.), Data for biochemical research. Clarendon Press, Oxford, United Kingdom. 6. Fling, S. P., and D. S. Gregerson. 1986. Peptide and protein molecular weight determination by electrophoresis using a high-molarity Tris buffer system without urea. Anal. Biochem. 155:83–88. 7. Fodje, M. N., A. Hansson, M. Hansson, J. G. Olsen, S. Gough, R. D. Willows, and S. Al-Karadaghi. 2001. Interplay between an AAA module and an integrin I domain may regulate the function of magnesium chelatase. J. Mol. Biol. 311:111–122. 8. Gorchein, A. 1997. Cell-free activity of magnesium chelatase in Rhodobacter sphaeroides and Rhodobacter capsulatus. Biochem. Soc. Trans. 25:82S. 9. Gorchein, A., A. Neuberger, and G. H. Tait. 1968. Adaptation of Rhodopseudomonas spheroides. Proc. Roy. Soc. B 170:111–125. 10. Harlow, E., and D. Lane. 1988. Antibodies: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 11. Jensen, P. E., L. C. D. Gibson, and C. N. Hunter. 1998. Determinants of catalytic activity with the use of purified I, D, and H subunits of the magnesium protoporphyrin IX chelatase from Synechocystis PCC6803. Biochem. J. 334:335–344. 12. Jensen, P. E., R. D. Willows, B. L. Petersen, U. C. Vothknecht, B. M. Stummann, C. G. Kannangara, D. von Wettstein, and K. W. Henningsen. 1996. Structural genes for Mg-chelatase subunits in barley: Xantha-f, -g and -h. Mol. Gen. Genet. 250:383–394. 13. Konig, K., H. Schneckenburger, A. Ruck, and R. Steiner. 1993. In vivo photoproduct formation during PDT with ALA-induced endogenous porphyrins. J. Photochem. Photobiol. B 18:287–290. 14. Ma, L. W., S. Bagdonas, and J. Moan. 2001. The photosensitizing effect of the photoproduct of protoporphyrin IX. J. Photochem. Photobiol. B 60:108– 113. 15. Mauzerall, D., and S. Granick. 1956. The occurrence and determination of ␦-aminolevulinic acid and porphobilinogen in urine. J. Biol. Chem. 219:435– 446.

3258

WILLOWS ET AL.

16. Moan, J., and K. Berg. 1992. Photochemotherapy of cancer—experimental research. Photochem. Photobiol. 55:931–948. 17. Neuberger, A., and G. H. Tait. 1964. Studies on the biosynthesis of porphyrin and bacteriochlorophyll by Rhodopseudomonas spheroides. 5. Zinc-protoporphyrin chelatase. Biochem. J. 90:607–616. 18. Neuhoff, V., N. Arold, D. Taube, and W. Ehrhardt. 1988. Improved staining of proteins in polyacrylamide gels including isoelectric focusing gels with clear background at nanogram sensitivity using Coomassie Brilliant Blue G-250 and R-250. Electrophoresis 9:255–262. 19. Papenbrock, J., H.-P. Mock, E. Kruse, and B. Grimm. 1999. Expression studies in tetrapyrrole biosynthesis. Inverse maxima of magnesium chelatase and ferrochelatase activity during cyclic photoperiods. Planta 208:264–273. 20. Senior, N. M., K. Brocklehurst, J. B. Cooper, S. P. Wood, P. Erskine, P. M. Shoolingin-Jordan, P. G. Thomas, and M. J. Warren. 1996. Comparative

J. BACTERIOL.

21. 22. 23.

24.

studies on the 5-aminolaevulinic acid dehydratases from Pisum sativum, Escherichia coli and Saccharomyces cerevisiae. Biochem. J. 320:401–412. Smith, J. H. C., and A. Benitez. 1955. Chlorophylls: analysis in plant material, p. 142. In K. Paech and M. V. Tracey (ed.), Modern methods of plant analysis, vol. IV. Springer-Verlag, Berlin, Germany. Weaver, P. F., J. D. Wall, and H. Gest. 1975. Characterization of Rhodopseudomonas capsulatus. Arch. Microbiol. 105:207–216. Willows, R. D., and S. I. Beale. 1998. Heterologous expression of the Rhodobacter capsulatus bchI, -D, and -H genes that encode magnesium chelatase subunits and characterization of the reconstituted enzyme. J. Biol. Chem. 273:34206–34213. Yang, Z. M., and C. E. Bauer. 1990. Rhodobacter capsulatus genes involved in early steps of the bacteriochlorophyll biosynthetic pathway. J. Bacteriol. 172:5001–5010.