The enzyme BchM (S-adenosyl-L-methionine:magnesium-protoporphyrin IX O-methyltransferase) from Rhodobacter capsulatus catalyses an intermediate reaction in the bacteriochlorophyll biosynthetic pathway. Overexpression of His6-tagged protein in Escherichia coli resulted in the majority of polypeptide existing as inclusion bodies. Purification from inclusion bodies was performed using metal-affinity chromatography after an elaborate wash step involving surfactant polysorbate-20. Initial enzymatic assays involved an in situ generation of S-adenosyl-L-methionine substrate using a crude preparation of S-adenosyl-L-methionine synthetase and this resulted in higher enzymatic activity compared with commercial S-adenosyl-L-methionine. A heat-stable stimulatory component present in the S-adenosyl-L-methionine synthetase was found to be a phospholipid, which increased enzymatic activity 3–4-fold. Purified phospholipids also stabilized enzymatic activity and caused a disaggregation of the protein to lower molecular mass forms, which ranged from monomeric to multimeric species as determined by size-exclusion chromatography. There was no stimulatory effect observed with magnesium–chelatase subunits on methyltransferase activity using His–BchM that had been stabilized with phospholipids. Substrate specificity of the enzyme was limited to 5-co-ordinate square-pyramidal metalloporphyrins, with magnesium-protoporphyrin IX being the superior substrate followed by zinc-protoporphyrin IX and magnesium-deuteroporphyrin. Kinetic analysis indicated a random sequential reaction mechanism. Three non-substrate metalloporphyrins acted as inhibitors with different modes of inhibition exhibited with manganese III-protoporphyrin IX (non-competitive or uncompetitive) compared with cobalt II-protoporphyrin IX (competitive).

INTRODUCTION

The photosynthetic bacterium Rhodobacter capsulatus is an example of a facultative photoheterotroph that uses photosynthesis as an energy source under anaerobic conditions in the light [1]. The bacteriochlorophyll biosynthetic pathway is required for photosynthetic growth, and mutational analysis with R. capsulatus was previously carried out to determine the intermediates and genetic loci of the enzymes involved [26]. This allowed the photosynthetic gene cluster to be assembled [7] (EMBL accession number Z11165). The genetic locus bchM was found to encode the BchM (S-adenosyl-L-methionine:magnesium-protoporphyrin IX O-methyltransferase) [810]. Details of the enzymatic steps including the genes involved from aminolevulinic acid, derived from glutamate, to (bacterio)chlorophyll can be found in previously published reviews [1114].

The enzyme BchM (EC 2.1.1.11) catalyses the methylation of a carboxyl group of Mg-proto (magnesium-protoporphyrin IX) by the ubiquitous methylating agent SAM (S-adenosyl-L-methionine), yielding MgPE (magnesium-protoporphyrin IX monomethyl ester) and SAH (S-adenosylhomocysteine) [15]. The homologous enzyme in plants is called ChlM and the porphyrin substrate and product are important signalling molecules in plastid–nucleus communication in plants. A mutant of ChlM in Arabidopsis, that accumulated Mg-proto had a negative impact on photosystem gene expression [16] and the effect of Mg-proto and MgPE as a way of co-ordinating chlorophyll biosynthesis with photosystem assembly was reviewed recently [17]. Methyltransferase has been studied kinetically in some detail from plants, algae and photosynthetic bacteria with Km values in the range 20–230 μM for SAM and 10–48 μM for Mg-proto [1827], although Km for Mg-proto using Rhodobacter spheroides was not determined since it approached zero [22]. Most of the studies on the bacterial and plant enzyme were undertaken before nucleotide sequence information was available and therefore crude or partially purified protein was used. With the R. capsulatus gene sequences being identified, heterologous expression of BchM from R. spheroides and R. capsulatus in E. coli was possible with demonstration of enzymatic activity [9,10]. The ChlM protein from Synechocystis PCC 6803 complemented a BchM mutant of R. capsulatus [28]. Most recently Synechocystis PCC 6803 ChlM has been heterologously expressed in E. coli and purified to homogeneity [25].

The reaction mechanism of methyltransferase appears to vary between a plant, algal and photosynthetic bacterium. The wheat enzyme exhibited a ping-pong mechanism with SAM binding first [19,27], whereas a random sequential mechanism was elucidated for Euglena gracilis [23] and Synechocystis [25]. An ordered sequential mechanism was determined for R. spheroides with Mg-proto obligated to bind to the enzyme first [22].

Methyltransferase is mainly associated with membranes according to differentiation of the cell fractions and measurement of enzymatic activity. It was detergent solubilized from barley chloroplast membranes [26], located on membrane fragments from chromatophores in R. spheroides [20], membrane-bound and soluble in E. gracilis [29] and associated with chloroplast envelope and thylakoid membrane in spinach and Arabidopsis thaliana [30]. A soluble form of the enzyme was obtained using sucrose and crude wheat homogenates [18]. Amino acid analysis of A. thaliana and Oriza sativa revealed two putative transmembrane domains at the same location in each plant sequence with the N-terminal portion suspected of being membrane-associated. This N-terminal region is not present in bacterial sequences and there is an independent putative transmembrane region near the C-terminus [30].

There is indirect and direct evidence to suggest that there is an interaction between methyltransferase and magnesium–chelatase (E.C. 6.6.1.1). The latter is a multisubunit enzyme involved in the previous step of the bacteriochlorophyll biosynthetic pathway, magnesium-insertion into protoporphyrin IX [31,32]. It is ATP-dependent (AAA protein) [33] and is comprised of the subunits BchI (approx. 40 kDa), BchD (approx. 70 kDa), and BchH 9 (approx. 140 kDa) [34]. An interaction of BchM and magnesium–chelatase was first suggested 35 years ago by Gorchein [35], based on in vivo studies using R. spheroides. Exogenous protoporphyrin IX fed into cells showed the accumulation of MgPE. The addition of ethionine (converted into S-adenosylethionine in vivo), which is a competitive inhibitor of SAM-dependent methyltransferases [20] exhibited no accumulation of Mg-proto [35]. A build-up of Mg-proto is expected if the magnesium–chelatase and BchM enzymes were acting independently. More recently it was observed that the membrane fraction of heterologously expressed BchM from R. capsulatus supplemented with a soluble BchH fraction stimulated methyltransferase activity [36]. Although there was no observed stimulatory effect of purified Synechocystis ChlH on ChlM activity, there was increased activity using crude lysates of co-expressed magnesium–chelatase subunits ChlI, D, H and ChlM compared with ChlM alone [37]. Using purified protein, it was later found that the ChlH stimulatory effect is observed in a much shorter time period (milliseconds) using quenched flow techniques [38] compared with the previous 30 min stopped assay [37]. This contrasts with the study in which the heterologously expressed tobacco ChlM was significantly stimulated with expressed ChlH in similar minute-scale assays [39]. There is a correlation between the transcription and expression of tobacco methyltransferase and magnesium–chelatase with increased transcript and protein levels of methyltransferase together with magnesium–chelatase ChlH subunit in ChlM sense transgenic plants and a suppressed level of magnesium–chelatase ChlH-subunit together with methyltransferase in ChlM antisense tobacco plants [40].

The properties of the methyltransferase, using HisBchM (His6-tagged BchM), presented here show a previously unreported lipid requirement for optimal activity and the kinetic data reveal some new insights into the reaction mechanism and the nature of substrate binding. In addition we also re-examine the reported stimulation of methyltransferase activity by magnesium–chelatase using purified HisBchM with magnesium–chelatase subunits, BchI, HisBchD and HisBchH.

EXPERIMENTAL

Materials

Unless stated elsewhere all chemicals were obtained from Sigma–Aldrich, Ajax Finechem, Astral Scientific, APS Finechem, Calbiochem, Selby-Biolab, BDH chemicals and J.T. Baker. Growth medium was from Difco Laboratories. Surfactant polysorbate P-20 (P-20) was from Biacore AB (Uppsala, Sweden). Protoporphyrin IX, deuteroporphyrin, and hemin were from Porphyrin Products (Logan, UT, U.S.A.). Protein determinations were performed with a Bio-Rad protein assay reagent according to the manufacturer's instructions with BSA as a standard [41].

Cloning, expression and purification of proteins

Magnesium–chelatase subunits BchD and BchH were expressed and purified as their His6-tagged products (HisBchD and HisBchH), whereas BchI was purified as the native form as described previously [42]. SAM synthetase (E.C. 2.5.1.6) from the overproducing E. coli strain DM22pK8 [44] was partially purified using ammonium sulfate precipitation [45] and dialysed against 100 mM Tris/HCl (pH 8.0), 10% (v/v) glycerol, 1 mM EDTA and 0.1% 2-mercaptoethanol at 5 °C for 4 h, and then stored at −80 °C. BchM from R. capsulatus was heterologously expressed in E. coli. The plasmid pRPS404 [6], consisting of a 46 kb pair section of the R. capsulatus photosynthetic gene cluster [7], was used as a PCR template using the Eppendorf Triplemaster PCR system with the following primers: 5′-TCATGCCCGATACTCCAGGCATTC-3′ and 5′-CACCATGCCCTCCGATTACGCAGAGATC-3′. The expression plasmid pHisBchM was obtained by directly cloning this PCR product into a pET 100/D-TOPO vector (Invitrogen) and the fidelity of the construct was confirmed by DNA sequencing using the Dye termination method with an ABI Prism model 377 automated fluorescent DNA sequencer (Perkin–Elmer). E. coli (BL21DE3) Star strain containing this plasmid was used for expression of HisBchM in Luria–Bertani medium with 100 μg/ml ampicillin. The cells were grown at 37 °C to a D600 of 0.3–0.5 when the flask was cooled to 18 °C and induced for 13–18 h with 0.25 mM isopropyl β-D-thiogalactopyranoside. Cells were harvested by centrifugation at 8000 g for 15 min at 4 °C, and then resuspended in binding buffer (20 mM Tris/HCl, pH 7.9, 0.5 M NaCl and 5 mM imidazole). The cells were lysed using a Thermospectronic French Press at approximately 7×104 kPa and centrifuged for 30 min at 30000 g at 4 °C. The pellet was washed with distilled water and re-centrifuged and the supernatant was discarded. Enzymatically active HisBchM protein was extracted from the pellet with solubilization buffer [20 mM Tricine/NaOH, pH 8.2, 1 M KCl, 0.05% P-20 and 1 mM DTT (dithiothreitol)], and re-centrifuged. The supernatant was named the soluble fraction. The pellet was dissolved in solubilization buffer containing 6 M Guanidine/HCl, re-centrifuged and this supernatant was named the Guanidine-soluble fraction. Purification of HisBchM from the soluble fraction was performed in a single step as described in the pET system manual (Invitrogen) using a HiTrap Ni2+ FF chelating column (GE Healthcare BioSciences) with the following modifications: wash steps contained increasing imidazole concentrations, of 5, 60 and 100 mM in 20 mM Tris/HCl (pH 7.9), 0.5 M NaCl, 0.5 M KCl and 1 mM DTT, and used two column volumes for each. HisBchM was eluted by 500 mM imidazole and the above buffer components (elution buffer). Phosphatidylglycerol was added to a final concentration of 25–50 μg/ml and HisBchM was desalted into 50 mM Tricine/NaOH (pH 8.2), 10% (v/v) glycerol and 1 mM DTT with a NAP-10 column (GE Healthcare BioSciences). Purified protein was concentrated using a 10 kDa cut-off membrane (Millipore) by centrifugation at 4000 g for 15 min at 4 °C, and to give final concentrations of approx. 0.2–4.5 μg/μl with approximately similar phosphatidylglycerol concentrations. If the concentrated protein was turbid, the solution was clarified by centrifugation at 18000 g for 5 min at room temperature, and the pellet discarded. The supernatant was stored in assay-size aliquots at −80 °C with the activity retained for at least 3 months. The Guanidine-soluble fraction was purified as above with 6 M Guanidine/HCl present in the wash steps. Refolding was performed with stepwise reduction of Guanidine/HCl in binding buffer (two column volumes each) from 6 M, 3 M, 1 M, 0.5 M, to 0 M. The protein was eluted as described above with elution buffer and the desalting and concentration steps performed as for the soluble fraction. SDS/PAGE was performed according to the method of Schägger and von Jagow [45a] using 4–20% gradient gels (LifeGels) at 150 V, and the gel was stained with Coomassie Brilliant Blue.

Metalated magnesium, nickel, copper, zinc, cobalt and manganese protoporphyrin IX, and magnesium-deuteroporphyrin synthesis and purification

Synthesis of Mg-proto dipotassium salt was performed based on the method of Fuhrhop and Granick [46]. Other metal protoporphyrins were based on a modified method stated in Falk [47] using the metal acetates: nickel, copper, zinc, cobalt and magnesium with pyridine or glacial acetic acid as the solvent. The metalated porphyrins were precipitated with ice-cold water generally (one part pyridine or acetic acid to 5–10 parts water) and collected by centrifugation at 5000 g for 15 min at 4 °C. The precipitate was washed a further five times with water to remove most of the pyridine, and the final precipitate was dried under nitrogen flow, desiccated over silica beads for several days and stored at −20 °C. The synthesis of Mg-deutero (magnesium-deuteroporphyrin) was carried out in the same way as for Mg-proto, and the precipitation/drying stage was the same as for the other metal protoporphyrins. Metalloporphyrin stock solutions were newly prepared for each assay by dissolving a speck of solid in 1 μl of 1 M NaOH and adding 100–1000 μl water with concentrations determined using the appropriate solvent and molar extinction coefficients for Ni-, Cu-, Zn-, Co- and Mn-proto [48] or ϵ408 278000 M−1·cm−1 for Mg-proto, ϵ398 433000 M−1·cm−1 for Mg-deutero [49] and ϵ400 90000 M−1·cm−1 for hemin [50].

Mn(III), Co(III) and Fe(III) metalloporphyrins were reduced to Mn(II), Co(II) and Fe(II) forms under basic conditions [51] using 1 mM NaOH, 159 μM POPG [palmitoyl-oleoyl(C18:1,16:0)- phosphatidylglycerol] and 960 mM sodium dithionite. Spectral changes were observed in the visible absorbance region and the reduced metalloporphyrins did not re-oxidize when diluted in assay buffer (50 mM Tricine/NaOH, pH 8.0, and 1 mM DTT).

MgPE enzymatic synthesis

Synthesis of MgPE was performed on a 200 ml scale with the following components: 50 mM Tricine/NaOH (pH 8.0), 1 mM DTT, 1.8 mM MgCl2, 1.3 mM ATP, 3 mM KCl, 8.8 mM glycerol, 2 μM Mg-proto, 0.4 mM L-methionine, 0.18 mg crude SAM synthetase and 0.2–0.6 μM partially purified HisBchM (from cell lysate supernatant) at 30 °C with gentle agitation. The formation of MgPE was monitored periodically by HPLC until complete conversion was achieved (typically 3–4 h) and the sample was lyophilized. Extraction of MgPE was performed according to the method of Ellsworth and Murphy [52]. MgPE was applied as an aqueous extract to a Waters C18 Sep-pak Plus column which was washed in a step-wise fashion with 5 ml each of 10%, 20%, 30% and 35% (v/v) acetonitrile. MgPE was finally eluted by 45–80% (v/v) acetonitrile (typically 5–10 ml), and checked for purity by HPLC (this was typically ≥99% by Abs/fluorescence) and the solution lyophilized. MgPE stocks for assays were initially dissolved in 2% (v/v) ethanol, 1 mM NH3, and centrifuged repeatedly for 5 min at room temperature (20 °C) at 18000 g until no pellet was observed, and this supernatant was used for inhibition assays.

Phospholipid preparation for assays

Crude bacterial phospholipids were extracted from BL21 (DE3) E. coli according to the method described by Osborn and Rothfield [53] and stored in 3:1 chloroform/methanol under nitrogen or argon in the dark at −80 °C. The concentration of the prepared phospholipids was based on the mass of a dried sample. The dried phospholipids were not used in the assays. Both crude and commercial phospholipids (Sigma) used in assays were prepared in aqueous solution as follows: a fraction of the phospholipids dissolved in chloroform/methanol was evaporated under nitrogen or argon flow to a volume of approx. 10 μl, then 4 volumes of methanol were added, evaporated again until approx. 10 μl remained. At least a 10 volume excess of water was then added, and the methanol evaporated again under nitrogen or argon flow to leave the phospholipids as a transparent aqueous micellar suspension. This suspension was sonicated for 20 min at room temperature, followed by centrifugation for 5 min at room temperature and 18000 g with this supernatant used for assays.

Methyltransferase enzymatic assays

Assays were in a total volume of 110–650 μl with final concentrations of 50 mM Tricine/NaOH (pH 8.0 or 8.2) or Tris/HCl (pH 8.5), 1 mM DTT and varying phospholipid, HisBchM, SAM and metalloporphyrin concentrations for 10 s to 30 min at 30 °C. Assays involving Co(III)-proto as the substrate contained no DTT. For exact details of assay conditions refer to relevant Figure and Table legends. Coupled assays with SAM synthetase also contained 3 mM KCl, 1.8 mM MgCl2, 1.3 mM ATP and 2 μg crude SAM synthetase per 100 μl assay. Stock solutions of SAM p-toluenesulfonate salt were prepared by dissolving the solid in 10 mM HCl and measuring the concentration based on ϵ257 14700 M−1·cm−1 [54] and stored in aliquots at −80 °C. Purified HisBchM stock solutions were used once for each set of reactions and diluted into 50 mM Tricine/NaOH (pH 8.0/8.2) or Tris/HCl (pH 8.0/8.5) and 1 mM DTT (with phospholipid added if applicable), followed by SAM/porphyrin in 50 mM Tris/HCl or Tricine/NaOH and 1 mM DTT to start the assay. Assays were stopped by taking 125 μl aliquots from each assay mix and adding to 105 μl 80:20 (v/v) acetone/water, spiked with 20 μl deuteroporphyrin as an internal standard to a final concentration of 1.21 μM in each 250 μl stopped assay. The internal standard was used to check for any obvious discrepancies in the data. The assay tubes were centrifuged at 18000 g for 5 min at room temperature, and the supernatant was used for analysis by HPLC. Assays and extractions with acetone were performed in the dark or in dim light.

Assays involving HisBchH and HisBchM were performed essentially according to HisBchM enzymatic assays. Assays involving BchI and HisBchD involved refolding 1 μl of 61 μM HisBchD in 6 M urea, 50 mM Tricine/NaOH (pH 8.0) and 4 mM DTT by rapidly adding 577 μl (200 μl at a time) of 0.45 μM BchI in 50 mM Tricine/NaOH (pH 8.0) and 1 mM DTT, and either 0.54 mM MgCl2, 3.8 mM MgCl2 or 3.8 mM MgCl2 and 1.01 mM ATP on ice for 2 h [42] with any variations stated in the Figure and Table legends.

CD spectroscopy

CD spectroscopy measurements were carried out with a Jasco J-810 Spectropolarimeter. Wavelengths scans were from 190–300 nm at 100 nm/min using a 1 cm cuvette, with 1 nm bandwidth, 1 nm step size, 4 s response and 32 accumulations at 20 °C.

HPLC for analysing porphyrins

The HPLC method was modified from the method of Gibson and Hunter [10]. A Shimadzu HPLC system was used at 2 ml/min with an Alltech C8 column (150×4.6 mm), diode array detector and a Shimadzu RF-535 fluorescence detector. A gradient was used for separation of Mg-proto, Zn-proto and Mg-deutero with a 6 min linear gradient from 5% buffer A water or 20 mM phosphate (pH 7.8) to 67% buffer B (acetonitrile). The other metalloporphyrins used a 6 min linear gradient fom 5% buffer A1; 0.1 M ammonium acetate (pH 6.8) to 90% buffer B1; 48% (v/v) acetonitrile, 48% (v/v) methanol, 0.1 M ammonium acetate (pH 6.8). Then a 5 min gradient to 100% buffer B1.

RESULTS AND DISCUSSION

Purification and optimization of HisBchM activity

HisBchM 29 kDa was purified as described in the Experimental section, as shown in Figure 1, and is prone to aggregation with the majority of expressed HisBchM protein as inactive aggregates or inclusion bodies (Figure 1, lane 5). Although active HisBchM was recovered by refolding from inclusion bodies the yield was very low and this fraction was not used further. Purification of HisBchM from the cell lysate supernatant using Ni2+-affinity chromatography resulted in co-purification with other non-identified proteins (Figure 1, lane 8). This preparation was used in coupled HisBchM–SAM synthetase assays, characterizing the phospholipid effect, and synthesis of MgPE. The largest yields of active HisBchM, estimated at 98% purity by SDS/PAGE (Figure 1, lane 7), were obtained by washing inclusion bodies with P-20 (Figure 1, lane 6) before application to the Ni2+-affinity column, yielding approx. 0.5 mg of purified protein recovered from a 2 litre culture, which was used for experiments with magnesium–chelatase subunits, kinetics, substrate specificity and inhibition assays.

Expression and Ni2+-affinity chromatography purification of HisBchM

Figure 1
Expression and Ni2+-affinity chromatography purification of HisBchM

Lane 1, Fermentas PageRuler protein ladder; lane 2, uninduced cells; lane 3, induced cells; lane 4, cell lysate inclusion body pellet; lane 5, cell lysate supernatant; lane 6, P-20 solubilized pellet; lane 7, 10 μg HiTrap FF purified HisBchM from P-20 solubilized pellet; lane 8, 8 μg HiTrap FF purified HisBchM from cell lysate supernatant. The arrow shows the band corresponding to the correct molecular mass of HisBchM (29.2 kDa).

Figure 1
Expression and Ni2+-affinity chromatography purification of HisBchM

Lane 1, Fermentas PageRuler protein ladder; lane 2, uninduced cells; lane 3, induced cells; lane 4, cell lysate inclusion body pellet; lane 5, cell lysate supernatant; lane 6, P-20 solubilized pellet; lane 7, 10 μg HiTrap FF purified HisBchM from P-20 solubilized pellet; lane 8, 8 μg HiTrap FF purified HisBchM from cell lysate supernatant. The arrow shows the band corresponding to the correct molecular mass of HisBchM (29.2 kDa).

Some common detergents and simple alcohols were also tested to try and solubilise and further stabilise the HisBchM (Table 1). P-20 at 0.0004% (v/v) gave a 3.8-fold stimulatory effect on HisBchM activity, however it was unsuitable to help stabilize HisBchM as it aggregated when frozen. The other detergents tested, Tween-20 or dodecyl-β-D-maltoside, had no stimulatory effect at 0.0004% (v/v) and 0.004% (w/v) respectively. Several other detergents have been previously tested using R. spheroides chromatophores with mainly negative effects, albeit at relatively high detergent concentrations (0.5–1%) [22]. A detergent concentration of 0.05% (v/v) Tween-20 strongly inhibited purified HisBchM (in our experiments only 23% of activity was retained). The straight chain alcohols (C2, C3 and C4) had no stimulatory effect and were not inhibitory up to 1% (v/v).

Table 1
Effect of detergents and alcohols on HisBchM activity

Assays were for 10–20 min with the following concentrations of each component: 50 mM Tris/HCl (pH 8.5), 1 mM DTT, 2 μM Mg-proto and variable SAM and HisBchM concentrations as stated in the Table. The following concentration ranges were used: Tween 20 or P-20, 0.000003–0.05% (v/v); dodecyl-β-D-maltoside, 0.004–4% (w/v); alcohols, 0.06–10% (v/v). Assays were compared with a control (assigned a value of 100% activity) conducted in parallel.

Additive Name SAM (μM) HisBchM (μM) % Additive* % Control 
Detergents Tween20 250 0.015 0.0004 117 
 Dodecyl-β-D-maltoside 60 0.16 0.004 101 
Surfactant Polysorbate 20 (P-20) 250 0.015 0.0004 382 
Alcohols Methanol 62 0.2 90 
 Ethanol 62 0.2 109 
 Propanol 62 0.2 114 
 Butanol 62 0.2 118 
Additive Name SAM (μM) HisBchM (μM) % Additive* % Control 
Detergents Tween20 250 0.015 0.0004 117 
 Dodecyl-β-D-maltoside 60 0.16 0.004 101 
Surfactant Polysorbate 20 (P-20) 250 0.015 0.0004 382 
Alcohols Methanol 62 0.2 90 
 Ethanol 62 0.2 109 
 Propanol 62 0.2 114 
 Butanol 62 0.2 118 
*

Additives were used over a variety of concentrations with the optimal amounts stated here based on the greatest stimulatory effect.

Coupled HisBchM–SAM synthetase assay and a stimulatory component

Preliminary HisBchM assays involved an in situ generation of SAM with crude SAM synthetase coupled with HisBchM. This approach gave a 1.4-fold higher relative activity compared with assaying HisBchM using commercial SAM. Surprisingly the addition of crude SAM synthetase to the HisBchM assay together with SAM resulted in up to a 4-fold increase in activity compared with SAM alone (Table 2). Crude BL21 (DE3) E. coli cell lysate supernatant also resulted in heightened HisBchM activity, which was retained after boiling, and the stimulatory component was larger than 5 kDa. It was speculated that the stimulatory effect was due to the phospholipid micelles. A preparation of a crude phospholipid extract from E. coli membranes also gave a comparable stimulatory effect.

Table 2
Identification of stimulatory component of HisBchM activity

Crude phospholipids were prepared from E. coli as stated in the Experimental section. Assays were for 10 min using 50 mM Tris/HCl (pH 8.5), 1 mM DTT, 100 μM SAM and variable Mg-proto and HisBchM concentrations as stated in the Table. Assays were compared with a control (assigned a value of 100% activity) conducted in parallel.

Additive Mg-proto (μM) HisBchM (μM) Amount additive (μg)* % Control 
Crude SAM synthetase 0.1 0.0016 374±25 
E.coli BL21 supernatant 0.26 0.0017 4.2 321 
Boiled crude SAM synthetase supernatant 0.26 0.0017 370±17 
Boiled E.coli BL21 supernatant 0.26 0.0017 4.2 350 
<5 kDa 0.28 0.0017 † 146±19 
>5 kDa 0.28 0.0017 † 519±71 
BL21 E.coli crude phospholipids 0.27 0.001 0.6 355 
Additive Mg-proto (μM) HisBchM (μM) Amount additive (μg)* % Control 
Crude SAM synthetase 0.1 0.0016 374±25 
E.coli BL21 supernatant 0.26 0.0017 4.2 321 
Boiled crude SAM synthetase supernatant 0.26 0.0017 370±17 
Boiled E.coli BL21 supernatant 0.26 0.0017 4.2 350 
<5 kDa 0.28 0.0017 † 146±19 
>5 kDa 0.28 0.0017 † 519±71 
BL21 E.coli crude phospholipids 0.27 0.001 0.6 355 
*

Additives were used over a variety of concentrations with the optimal amounts stated here based on the greatest stimulatory effect. Crude SAM synthetase assays were performed in duplicate, whereas E. coli BL21 supernatant and crude phospholipid were single assays. A strain of BL21 (DE3) Star E. coli was grown to stationary phase, lysed and centrifuged (60 min at 4 °C and 100000 g) and the supernatant used. The crude SAM synthetase and BL21 supernatant were boiled for 10 min, and re-centrifuged.

0.5 ml of a 2.5 mg/ml fraction of boiled BL21 E. coli supernatant separated with a NAP-10 column. Extraction of crude phospholipids from BL21 E. coli cells was performed as described previously [53].

Phospholipid effect on HisBchM activity

Specific phospholipid requirements have been found for many membrane proteins [55]; however, the stimulatory effect of crude preparations of phospholipids in our assays was labile as the stimulatory effect was lost after storage for 1 week. Therefore various commercial phospholipids with different polar head groups and alkyl chains (stable for several months at −80 °C in 99:1 chloroform/methanol) were tested for their stimulatory effect. Phospholipids were titrated to determine the optimal concentration that gave the highest stimulation of activity (for an example see Supplementary Figure 1 at http://www.BiochemJ.org/bj/406/bj4060469add.htm) and the results are shown in Table 3. Pg (phosphatidylglycerol) is the main phospholipid present in R. capsulatus [56] and a mixed alkyl chain pg was initially tested for its effect on HisBchM activity and found to have a 10-fold stimulatory effect. Two other phospholipids were tested with variations in the polar head group, ps (phosphatidylserine) and pe (phosphatidylethanolamine), with an 8-fold and 1.3-fold stimulatory effect respectively. These results seem to correlate with the polarity and charge of the phospholipid, the greatest stimulatory effect seen with the most polar negatively charged phospholipid, pg. Ps, a negatively charged phospholipid of intermediate polarity not present in significant quantities in R. capsulatus [56], also had a strong stimulatory effect. A comparatively marginal effect was observed with the most non-polar neutral phospholipid tested, pe. This appears to coincide with the known phospholipid composition of R. capsulatus, which is mainly composed of pg, whether grown heterotrophically or photosynthetically. It has also been shown that there is a large increase in the amount of pg in photosynthetically grown cells (62.5%) compared with heterotrophic growth (39.3%) together with a large decrease in pe (33.8 to 18.7%) [56]. Variant pgs were tested with differing alkyl chains. There was little difference in activity between two unsaturated alkyl chains, dioleoyl (C18:1)2 phosphatidylglycerol (DOPG) or one, POPG with both exhibiting a 4–5-fold effect on HisBchM; however, no stimulatory effect was observed with a saturated alkyl chain phospholipid, dipalmitoyl (C16:0)2 phosphatidylglycerol (DPPG) highlighting the importance of the single double bond or the C18 chain. The oleoyl (C18:1) chain might be important since it is the major phospholipid fatty acid chain length of pg in R. capsulatus (83.5 to 85.7%), with the remainder being C16:0, C16:1 and C18:0 [56]. The greater stimulatory effect seen with mixed pg as opposed to DOPG or POPG may suggest the need for a variable alkyl chain length (possibly C16:1 or C18:0).

Table 3
Phospholipid effect on HisBchM activity

Additives of crude and commercial phospholipids were assayed over a variety of concentrations with the optimal amounts stated based on the greatest stimulatory effect in a 100 μl assay (see Supplementary Figure 1 at http://www.BiochemJ.org/bj/406/bj4060469add.htm for an example). Assays were as described in the Experimental section, with the concentrations and any variations stated in the Table. Assays were performed for 0.5 min (DPPG) or 10 min for the remaining phospholipids and compared with a control (assigned a value of 100% activity) conducted in parallel.

Additive pH Mg-proto (μM) SAM (μM) HisBchM (μM) Amount additive (μg)TF3-001 % Control 
Pg, mixed alkyl chain 8.5 250 0.018 1035 
Ps, mixed alkyl chain 8.5 250 0.018 865 
Pe, mixed alkyl chain 8.5 250 0.018 20 139 
Dioleoyl (C18:1)2 pg (DOPG) 8.5 77 0.041 442 
Palmitoyl-oleoyl (C16:0,18:1) pg (POPG) 8.5 77 0.041 494 
Dipalmitoyl (C16:0)2 pg (DPPG) 8.0 0.5 80 0.05 81 
Additive pH Mg-proto (μM) SAM (μM) HisBchM (μM) Amount additive (μg)TF3-001 % Control 
Pg, mixed alkyl chain 8.5 250 0.018 1035 
Ps, mixed alkyl chain 8.5 250 0.018 865 
Pe, mixed alkyl chain 8.5 250 0.018 20 139 
Dioleoyl (C18:1)2 pg (DOPG) 8.5 77 0.041 442 
Palmitoyl-oleoyl (C16:0,18:1) pg (POPG) 8.5 77 0.041 494 
Dipalmitoyl (C16:0)2 pg (DPPG) 8.0 0.5 80 0.05 81 
*

Additives were used over a variety of concentrations with the optimal amounts stated here based on the greatest stimulatory effect.

Dilution of HisBchM with a constant concentration of DOPG resulted in a protein-concentration-dependent production of MgPE (see Supplementary Figure 2 at http://www.BiochemJ.org/bj/406/bj4060469add.htm). It is not surprising that phospholipids are associated with HisBchM from R. capsulatus in the present study, since the enzyme activity has been localized in chromatophores and membrane fragments in R. spheroides [20,57] and a putative transmembrane domain has also been identified [30]. Since phospholipids are prevalent in membrane proteins and required for structural integrity and enzymatic activity [55], it is likely that the BchM protein from R. capsulatus is membrane-bound in vivo.

Physical properties of HisBchM

Phospholipids had a significant effect on the aggregation state of HisBchM in vitro using size-exclusion chromatography (see Supplementary Figure 3 at http://www.BiochemJ.org/bj/406/bj4060469add.htm). Without phospholipids the majority of purified 29 kDa HisBchM existed as a high molecular mass aggregate or multimer. This is in contrast with pure ChlM protein from Synechocystis, which exists solely as a monomer [25]. The addition of a crude phospholipid preparation from E. coli shifted the elution profile of the protein to lower molecular mass complexes, suggesting a disaggregation process. A more profound effect was observed on the aggregation state of HisBchM with the addition of pg (mixed alkyl chain) which resulted in a large proportion of the protein eluting as an apparent monomer. Analysis of the CD spectrum of this solubilised HisBchM (see Supplementary Figure 4 at http://www.BiochemJ.org/bj/406/bj4060469add.htm) using CDPro program [58], indicated that the majority of protein is α-helical 56%) with a significant proportion of disordered structure (20%) and some β-sheet (16%).

Effect of magnesium–chelatase on methyltransferase activity

It has been reported previously that the BchH subunit of the magnesium–chelatase stimulates the activity of the BchM methyltransferase [36]. In the light of the effect of phospholipids stimulating and stabilizing the HisBchM we decided to re-examine this effect using purified magnesium-chelatase subunits. Formation of the magnesium–chelatase complex requires magnesium ions and ATP and so the effect of magnesium ions on methyltransferase activity was also tested. Increasing magnesium ion concentrations caused a decrease in methyltransferase activity, possibly due to aggregation of the enzyme Figure 2). The HisBchH appears to slightly inhibit the methyltransferase activity using a limiting or excess molar amount of HisBchH compared with HisBchM (Table 4) and a low magnesium chloride concentration (0.54 mM). In addition the combination of other magnesium-chelatase subunits and BSA were also tested for stimulation of methyltransferase activity (see Supplementary Figure 5 at http://www.BiochemJ.org/bj/406/bj4060469add.htm). At a low magnesium chloride concentration (0.54 mM) there is little effect of these proteins on methyltransferase activity. Higher magnesium ion concentrations (3.8 mM) inhibited HisBchM activity in the presence or absence of magnesium–chelatase subunits or BSA. At 3.8 mM magnesium chloride and 1.01 mM ATP an active magnesium–chelatase complex was formed according to a positive magnesium–chelatase assay independent of methyltransferase (results not shown), and the effect on methyltransferase was the same as without ATP. The idea of a physical interaction was proposed by Gorchein [35], with an in vivo study using R. spheroides. Much later it has been shown that crude BchM from R. spheroides was stimulated by crude BchH expressed in E. coli [36]. It has been proposed that BchH (with Mg-proto bound) is the real substrate for BchM [38]. A stimulatory effect of Synechocystis ChlH on Synechocystis ChlM methyltransferase activity has been shown with a molar excess of ChlH on a millisecond time scale [38]; however, no increased activity was seen with purified protein from Synechocystis over 30 min [37], highlighting the speed of the interaction. This might be the reason we do not see a direct stimulatory effect of HisBchH on HisBchM.

Effect of MgCl2 on HisBchM activity

Figure 2
Effect of MgCl2 on HisBchM activity

Assays were performed in triplicate with the following concentrations: 50 mM Tricine/NaOH (pH 8.0), 1 mM DTT, 100 μM SAM, 0.4 μM Mg-proto, 0.1 μM HisBchM, 2.4 mM glycerol and variable MgCl2 (0, 0.625, 1.25, 2.5, 5, 10, 20 mM) for 0.5 min. The results are presented as means±S.D.

Figure 2
Effect of MgCl2 on HisBchM activity

Assays were performed in triplicate with the following concentrations: 50 mM Tricine/NaOH (pH 8.0), 1 mM DTT, 100 μM SAM, 0.4 μM Mg-proto, 0.1 μM HisBchM, 2.4 mM glycerol and variable MgCl2 (0, 0.625, 1.25, 2.5, 5, 10, 20 mM) for 0.5 min. The results are presented as means±S.D.

Table 4
HisBchH effect on methyltransferase activity

A 5 min pre-incubation of HisBchH with HisBchM was performed at room temperature prior to the assay. Assays were performed for 0.5 min with the following concentrations: 50 mM Tricine/NaOH (pH 8.0), 1 mM DTT, 100 μM SAM, 0.4 μM Mg-proto, 0.1 μM HisBchM, 51 mM glycerol and 0.54 mM MgCl2. The results are presented as means±S.D.

HisBchH (μM) H:M V (pmol/min per μg of protein) 
74.6±3.5 
0.0071 0.074 65.6±4.8 
0.021 0.22 58.6±6.7 
0.063 0.67 67.5±1.2 
0.19 2.0 60.5±6.6 
0.57 6.0 43.6±15 
HisBchH (μM) H:M V (pmol/min per μg of protein) 
74.6±3.5 
0.0071 0.074 65.6±4.8 
0.021 0.22 58.6±6.7 
0.063 0.67 67.5±1.2 
0.19 2.0 60.5±6.6 
0.57 6.0 43.6±15 

Enzymatic mechanism

Prior to performing detailed kinetic studies, preliminary experiments were performed to optimize the assay conditions including determining the optimal pH and linearity of the assay. The pH range for active HisBchM is reasonably broad and peaks at approx. 8.0 although pH 7.5–8.5 are also within the error range (Figure 3). There also seems to be a slight preference for Tris/HCl buffer over Tricine/NaOH. A previous study with chromatophores of R. spheroides showed an optimal pH value of 8.4 using Tris/HCl [20]. The HisBchM methyltransferase assay was optimized and was shown to give a constant rate of product formation up to approx. 70% conversion to MgPE, which occurred after ∼100 s (Figure 4). There is a rapid formation of MgPE product, even after 10 s, and previous recent quenched flow analysis using the BchM orthologue, ChlM, from Synechocystis indicated the presence of an intermediate [24]. We were unable to observe this intermediate in this system despite using a similar HPLC method for detection.

Optimal pH range for HisBchM activity

Figure 3
Optimal pH range for HisBchM activity

The following buffers were used: ○ NaHCO3, □ Tris/HCl, △ Tricine/NaOH, ● MOPS/NaOH, ■ MES/NaOH. The assay consisted of 40 mM buffer, 1 mM DTT, 2.4 mM glycerol, 0.4 μM Mg-proto, 100 μM SAM and 0.1 μM HisBchM for 0.5 min. The results are presented as means±S.D. for assays performed in triplicate

Figure 3
Optimal pH range for HisBchM activity

The following buffers were used: ○ NaHCO3, □ Tris/HCl, △ Tricine/NaOH, ● MOPS/NaOH, ■ MES/NaOH. The assay consisted of 40 mM buffer, 1 mM DTT, 2.4 mM glycerol, 0.4 μM Mg-proto, 100 μM SAM and 0.1 μM HisBchM for 0.5 min. The results are presented as means±S.D. for assays performed in triplicate

Linearity of the HisBchM assay

Figure 4
Linearity of the HisBchM assay

The following time points were used: 0, 10, 20, 30, 60, 120, 180 and 240 s, with the following assay components: 50 mM Tricine/NaOH (pH 8.2), 1 mM DTT, 0.25 μM Mg-proto, 75 μM SAM, 0.1 μM HisBchM, 3.1 μM DOPG and 2.2 mM glycerol.

Figure 4
Linearity of the HisBchM assay

The following time points were used: 0, 10, 20, 30, 60, 120, 180 and 240 s, with the following assay components: 50 mM Tricine/NaOH (pH 8.2), 1 mM DTT, 0.25 μM Mg-proto, 75 μM SAM, 0.1 μM HisBchM, 3.1 μM DOPG and 2.2 mM glycerol.

Kinetic analysis was performed by varying one of the substrates, while keeping the second substrate constant. Analysis of the kinetic data was performed by GraphPad Prism version 4.00, GraphPad Software (San Diego, CA, U.S.A.). Non-linear regression and the two models, sequential [Y=VmaxAX/(AU+KaX+KbA+KiaKb)], and ping-pong [Y=VmaxAX/(AX+KaX+KbA)] were tested [59]. The data clearly showed a sequential type of mechanism (P=0.0015), and the kinetic constants were determined (Table 5, and Supplementary Figures 6 and 7 at http://www.BiochemJ.org/bj/406/bj4060469add.htm) with a Km of 44±8 μM for SAM, 0.11±0.02 μM for Mg-proto and Vmax 93±5 pmol/min per μg of protein. These values were comparable with previous kinetic data for SAM (20–230 μM), whereas the Km for the porphyrin was approx. 100–400-fold lower than previous data with algal or plant BchM orthologues, 10–48 μM [19,21,25,26,29]. Although not quantitated, the Km value of Mg-proto from R. spheroides has been shown previously to be much lower than plant or algal methyltransferase [22].

Table 5
Summary of kinetic data for biologically active substrates

Kinetic data for metalloporphyrin (MP) substrates. Assays were performed in triplicate with graphical data shown in Supplementary Figures 6 and 7 (http://www.BiochemJ.org/bj/406/bj4060469add.htm). Apparent Km values for Zn-proto and Mg-deutero were determined using 250 μM SAM and previously determined constants, KiSAM 31 μM, KmSAM 45 μM and a global fit used for the Vmax and KmMP. The following metallopophyrin concentration ranges were used: 0.0625–4 μM for Mg-deutero and 0.07–0.87 μM for Zn-proto.

Substrate KmMP (μM) KiMg-proto (μM) KmSAM (μM) KiSAM (μM) Vmax (pmol/min per μg) Kcat (s−1Kcat/Km (M−1·s−1
Mg-proto 0.11±0.02 0.076±0.03 45±7 31±7 97±6 0.04 4×105 
Zn-proto 0.08±0.02 − − − 43±2 0.018 2.3×105 
Mg-deutero 0.06±0.03 − − − 7.3±0.3 0.003 5×104 
Substrate KmMP (μM) KiMg-proto (μM) KmSAM (μM) KiSAM (μM) Vmax (pmol/min per μg) Kcat (s−1Kcat/Km (M−1·s−1
Mg-proto 0.11±0.02 0.076±0.03 45±7 31±7 97±6 0.04 4×105 
Zn-proto 0.08±0.02 − − − 43±2 0.018 2.3×105 
Mg-deutero 0.06±0.03 − − − 7.3±0.3 0.003 5×104 

Product inhibition with both SAH and MgPE was performed to elucidate the reaction mechanism in terms of the order of substrate binding and product release. The following non-linear regression equations were used with GraphPad Prism: Non-competitive, {Y=[VmaxX/(1+I/Ki)]/Km+X}; and Competitive, Km(app)=Km[1+I/K], Y=VmaxX/[Km(app)+X]. Non-competitive inhibition was seen using SAH with respect to either Mg-proto or SAM substrate (Table 6, and see Supplementary Figure 8 at http://www.BiochemJ.org/bj/406/bj4060469add.htm). Product inhibition with MgPE was performed using Mg-deutero as the substrate instead of Mg-proto which enabled separation of the MgPE inhibitor from the Mg-deutero monomethyl ester product made. Inhibition of MgPE against Mg-deutero was assumed to be the same type of inhibition expected as if Mg-proto was used as a substrate. Non-competitive inhibition was found using MgPE against either substrate, Mg-deutero or SAM (Table 6, and see Supplementary Figure 9 at http://www.BiochemJ.org/bj/406/bj4060469add.htm). This product inhibition pattern is characteristic of a random sequential-type xmechanism (Figure 5), whereupon substrate binding is a random process with either Mg-proto or SAM able to bind first to the enzyme. The product release is also apparently non-ordered with either product SAH or MgPE released first from the enzyme. Also, both E–Mg-proto–SAH and E–SAM–MgPE (where E is the enzyme) form dead-end complexes and are unable to be out-competed with the substrates SAM and Mg-proto.

Table 6
Enzymatic mechanism of HisBchM

Product inhibition of SAH (see Supplementary Figure 8 at http://www.BiochemJ.org/bj/406/bj4060469add.htm), and product inhibition of MgPE (see Supplementary Figure 9 at http://www.BiochemJ.org/bj/406/bj4060469add.htm). The results are presented as means±S.D. N.D., not determined

Variable substrate SAH, inhibition type KiSAH (μM) MgPE, inhibition type KiMgPE (μM) 
Mg-proto Non-competitive 122±11 N.D. − 
Mg-deutero N.D. − Non-competitive 0.45±0.03 
SAM Non-competitive 141±16 Non-competitive 0.61±0.06 
Variable substrate SAH, inhibition type KiSAH (μM) MgPE, inhibition type KiMgPE (μM) 
Mg-proto Non-competitive 122±11 N.D. − 
Mg-deutero N.D. − Non-competitive 0.45±0.03 
SAM Non-competitive 141±16 Non-competitive 0.61±0.06 

Random-order reaction mechanism for methyltransferase from R. capsulatus

Figure 5
Random-order reaction mechanism for methyltransferase from R. capsulatus

Either substrate Mg-proto or SAM may bind to the enzyme E first and two dead-end complexes can exist, E–Mg-proto–SAH and E–SAM–MgPE.

Figure 5
Random-order reaction mechanism for methyltransferase from R. capsulatus

Either substrate Mg-proto or SAM may bind to the enzyme E first and two dead-end complexes can exist, E–Mg-proto–SAH and E–SAM–MgPE.

(Metallo)porphyrin substrate specificity of HisBchM

In addition to Mg-proto, only Zn-proto and Mg-deutero were found to be substrates for the enzyme, with no product formation detected using protoporphyrin IX, deuteroporphyrin, Cu-proto, Ni-proto, Fe(II)-proto, Fe(III)-proto, Co(II)-proto, Co(III)-proto, Mn(II)-proto and Mn(III)-proto as substrates. The Zn-proto and Mg-deutero were found to have similar Km values to Mg-proto; however, the Vmax values were 2-fold and 13-fold lower than Mg-proto respectively (Table 5). Previous studies showed that R. spheroides BchM was able to utilise the following porphyrins as substrates of decreasing efficiency with respect to Mg-proto (100%): Ca-proto (73%), Zn-proto (57%), Mg-deutero (23%) and Mg-meso (20%) [20]. This shows the importance of both the vinyl group and the metal for optimal catalytic activity. It has been shown previously that Zn-proto is a physiological substrate in Zea mays with approximately 26–44% efficiency of Mg-proto [60]. Several metalloporphyrins were not substrates with BchM from R. spheroides, which included Mn-proto, Fe-proto, Cu-proto, protoporphyrin IX and deuteroporphyrin [20]. There seems to be a correlation between the metalloporphyrin co-ordinate structure and HisBchM activity. Depending upon the metal ion inserted into the porphyrin, metalloporphyrins may have zero, one or two addition axial ligands (e.g. water molecules) bound to the metal atom [61]. The metalloporphyrins that exhibited activity (Mg-proto, Zn-proto and Mg-deutero) have one ligand bound to the metal in the tetrapyrrole nucleus forming a 5-co-ordinate square-pyramidal structure. Ni-proto and Cu-proto are stable with no ligands bound and have a square-planar 4-co-ordinate structure. Mn-proto, Co-proto and Fe-proto can exist as either octahedral 6-co-ordinate or 5-co-ordinate structures depending on the oxidation state, environment and ligand [61]. The 5-co-ordinate square-pyramidal structures are usually distorted with one ligand bound, which may be important for initiating catalysis. The 4- or 6-co-ordinate metalloporphyrins may not be able to fit into the active-site due to the lack of an exchange ligand or an extra ligand.

(Metallo)porphyrin inhibition

The (metallo)porphyrins that were not substrates for HisBchM were tested as inhibitors (Table 7). Mg-, Zn-, Cu- and Ni-porphyrin exist in the Metal(II) oxidation state, whereas Fe-, Co-, and Mn-proto exist primarily in the Metal(III) oxidation state, which can be reduced to the Metal(II) state with dithionite and both of these forms were tested. There were no inhibitory effects of the following porphyrins up to 1 μM: protoporphyrin IX, deuteroporphyrin, Cu-proto, Ni-proto, Fe(III)-proto and Co(III)-proto. Assays involving Mn(II)-proto or Fe(II)-proto gave variable results, as the inhibitors interfered with separation and detection in the HPLC chromatography, and Ki values could not be determined and in this case the unidentified values are stated as not determined. Fe(II)-proto was not inhibitory up to 3 μM using a substrate concentration of 0.4 μM Mg-proto, whereas Mn(II)-proto was approx. 30% inhibitory at 0.5 μM. The only strong inhibitors observed were Mn(III)-proto and Co(II)-proto, and the Ki values and inhibition type were determined. The following non-linear regression equations were used with GraphPad Prism: Non-competitive; Uncompetitive, Y=VmaxX/[Km+X(1+I/Kj)]; and Competitive. Mn(III)-proto and Co(II)-proto gave different inhibition patterns with Mn(III)-proto showing uncompetitive or non-competitive inhibition (no significant difference was found with either inhibition type) and Ki 1.6–2.7 μM, with Co(II)-proto seemingly being competitive (Ki 0.65 μM). Previous data with R. spheroides has shown that both Fe-proto and Mn-proto are strong inhibitors with Fe(III)-proto seemingly a non-competitive inhibitor. Cu-proto was a weak inhibitor, although much higher concentrations of the porphyrin inhibitors were used (110–240 μM and 125 μM Mg-proto substrate) [20] compared with our assays (0.5–3 μM inhibitor and 0.1–0.4 μM Mg-proto).

Table 7
Metalloporphyrin inhibitors of HisBchM

Assays were performed in duplicate for 0.5 min with 50 mM Tricine/NaOH, pH 8.0, 1 mM DTT, 4 mM glycerol, 100 μM SAM, 0.1, 0.2, 0.3 or 0.4 μM Mg-proto, 0.06 μM HisBchM and 6.5 μM POPG, with or without 25 mM dithionite. A global fit was used to determine the kinetic parameters with the results presented as means±S.D. N.D., not determined (refer to the text for more information).

Inhibitor used Inhibition Inhibition type Ki (μM) Km (μM) Vmax (pmol/min per μg) 
Protoporphyrin IX No − − − − 
Deuteroporphyrin No − − − − 
Cu-proto No − − − − 
Ni-proto No − − − − 
Fe(III)-proto No − − − − 
Fe(II)-proto No − − − − 
Co(III)-proto No − − − − 
Co(II)-proto Yes Competitive 0.65±0.12 0.22±0.05 234±26 
Mn(III)-proto Yes Non-competitive 2.7±0.4 0.12±0.03 193±15 
  Uncompetitive 1.6±0.3 0.15±0.04 213±21 
Mn(II)-proto Yes N.D. N.D. N.D. N.D. 
Inhibitor used Inhibition Inhibition type Ki (μM) Km (μM) Vmax (pmol/min per μg) 
Protoporphyrin IX No − − − − 
Deuteroporphyrin No − − − − 
Cu-proto No − − − − 
Ni-proto No − − − − 
Fe(III)-proto No − − − − 
Fe(II)-proto No − − − − 
Co(III)-proto No − − − − 
Co(II)-proto Yes Competitive 0.65±0.12 0.22±0.05 234±26 
Mn(III)-proto Yes Non-competitive 2.7±0.4 0.12±0.03 193±15 
  Uncompetitive 1.6±0.3 0.15±0.04 213±21 
Mn(II)-proto Yes N.D. N.D. N.D. N.D. 

One of the common features of the metalloporphyrins which act as either substrates or inhibitors is that these metalloporphyrins prefer to have five co-ordination bonds with the metal and adopt a geometry in which the metal ion forms a square pyramidal structure with the metal out of the average plane of the porphyrin macrocycle in the direction of the fifth ligand. Fe(II)-proto is an exception as it is not a substrate or inhibitor but has five co-ordination bonds with the metal, which can be in or out of plane with the porphyrin depending on the ligand and spin state of the metal. Of the remaining metalloporphyrins that were not substrates or inhibitors most generally prefer to have four or six co-ordination bonds with the metal ion remaining in the plane of the macrocycle. Zn(II) and Mg(II) derivatives of metalloporphyrins have the metal ion between 0.35 and 0.4 Å (1 Å=0.1 nm) out of plane of the porphyrin macrocycle (see [62] for review) and the fifth co-ordination bond to either an N or O group on the fifth ligand. This compares with known 5-co-ordinate Co(II) porphyrin structures in which the metal ion is 0.15 Å out of plane of the macrocycle, while similar 5-co-ordinate Mn(II) and Mn(III) metalloporphyrins have the metal ion between 0.5 and 0.74 Å out of plane of the macrocycle [62]. It appears that catalysis can only occur when the metal ion is out of plane from the macrocycle over a very narrow range, and thus the metal ion–ligand bond is important in the correct positioning of the porphyrin for methylation.

Conclusion

Phosphatidylglycerol plays an important role for BchM solubility, stability, and enzymatic activity. This seems to be synonymous with a requirement of this phospholipid for the bacteriochlorophyll biosynthetic pathway in R. capsulatus and it suggests a role for it in the regulation of bchM activity in vivo. No stimulation of activity was observed with magnesium–chelatase subunits but it is important to note that it is still possible that the BchM and magnesium–chelatase subunits could interact in vitro and in vivo even though there is no effect on the activity of BchM.

We would like to thank David Bollivar (Department of Biology, Illinois Wesleyan University, Bloomington, IL, U.S.A.) for supplying the clone of the R. capsulatus BchM and Professor G. D. Markham (Fox Chase Cancer Center, Philadelphia, PA, U.S.A.) for the SAM-synthetase expression clone. Part of this work was supported by a Macquarie University Research Development Grant.

Abbreviations

     
  • BchM

    S-adenosyl-L-methionine:magnesium-protoporphyrin IX O-methyltransferase

  •  
  • DOPG

    dioleoyl (C18:1)2 phosphatidylglycerol

  •  
  • DPPG

    dipalmitoyl (C16:0)2 phosphatidylglycerol

  •  
  • DTT

    dithiothreitol

  •  
  • HisBchM

    His16-tagged BchM

  •  
  • Mg-deutero

    magnesium-deuteroporphyrin

  •  
  • Mg-proto

    magnesium-protoporphyrin IX

  •  
  • MgPE

    magnesium-protoporphyrin IX monomethyl ester

  •  
  • pe

    phosphatidylethanolamine

  •  
  • pg

    phosphatidylglycerol

  •  
  • POPG

    palmitoyl-oleoyl (C18:1,16:0) phosphatidylglycerol

  •  
  • SAH

    S-adenosylhomocysteine

  •  
  • SAM

    S-adenosyl-L-methionine

References

References
1
Gregor
J.
Klug
G.
Regulation of bacterial photosynthesis genes by oxygen and light
FEMS Microbiol. Lett.
1999
, vol. 
179
 (pg. 
1
-
9
)
2
Biel
A. J.
Marrs
B. L.
Transcriptional regulation of several genes for bacteriochlorophyll biosynthesis in Rhodobacter capsulatus
J. Bacteriol.
1983
, vol. 
156
 (pg. 
686
-
694
)
3
Zsebo
K. M.
Hearst
J. E.
Genetic-physical mapping of a photosynthetic gene cluster from R. capsulata
Cell
1984
, vol. 
37
 (pg. 
937
-
947
)
4
Taylor
D. P.
Cohen
S. N.
Clark
W. G.
Marrs
B. L.
Alignment of genetic and restriction maps of the photosynthesis region of the Rhodopseudomonas capsulata chromosome by a conjugation-mediated marker rescue technique
J. Bacteriol.
1983
, vol. 
154
 (pg. 
580
-
590
)
5
Yen
H.-C.
Marrs
B. L.
Map of genes for carotenoid and bacteriochlorophyll biosynthesis in Rhodopseudomonas capsulata
J. Bacteriol.
1976
, vol. 
126
 (pg. 
619
-
629
)
6
Marrs
B.
Mobilization of the genes for photosynthesis from Rhodopseudomonas capsulata by a promiscuous plasmid
J. Bacteriol.
1981
, vol. 
146
 (pg. 
1003
-
1012
)
7
Alberti
M.
Burke
D. H.
Hearst
J. E.
Blankenship
R. E.
Madigan
M. T.
Bauer
C. E.
Structure and sequence of the photosynthetic gene cluster
Anoxygenic Photosynthetic Bacteria
1995
Amsterdam
Kluwer Academic Publishing
(pg. 
1083
-
1106
)
8
Yang
Z.
Bauer
C. E.
Rhodobacter capsulatus genes involved in early steps of the bacteriochlorophyll biosynthetic pathway
J. Bacteriol.
1990
, vol. 
172
 (pg. 
5001
-
5010
)
9
Bollivar
D. W.
Jiang
Z.-Y.
Bauer
C. E.
Beale
S. I.
Heterologous expression of the BchM gene product from Rhodobacter capsulatus and demonstration that it encodes S-adenosyl-L-methionine:Mg-protoporphyrin IX methyltransferase
J. Bacteriol.
1994
, vol. 
176
 (pg. 
5290
-
5296
)
10
Gibson
L. S. D.
Hunter
C. N.
The bacteriochlorophyll biosynthesis gene, bchM, of Rhodobacter sphaeroides encodes S-adenosyl:Mg protoporphyrin IX methyltransferase
FEBS Lett.
1994
, vol. 
352
 (pg. 
127
-
130
)
11
Beale
S. I.
Enzymes of chlorophyll biosynthesis
Photosynthesis Res.
1999
, vol. 
60
 (pg. 
43
-
73
)
12
Beale
S. I.
Weinstein
J. D.
Dailey
H. A.
Tetrapyrrole metabolism in photosynthetic organisms
Biosynthesis of Heme and Chlorophylls
1990
New York
McGraw-Hill Publishing
(pg. 
287
-
391
)
13
Willows
R. D.
Biosynthesis of chlorophylls from protoporphyrin IX
Natural Products Reports
2003
, vol. 
20
 (pg. 
1
-
16
)
14
Bollivar
D. W.
Recent advances in chlorophyll biosynthesis
Photosynthesis Res.
2006
, vol. 
89
 (pg. 
1
-
22
)
15
Bollivar
D. W.
Kadish
K. M.
Smith
K. M.
Guilard
R.
Intermediate steps in chlorophyll biosynthesis methylation and cyclization
The Porphyrin Handbook, Chlorophylls and Bilins: Biosynthesis, Synthesis, and Degradation
2003
Sydney
Academic Press
(pg. 
49
-
69
)
16
Pontier
D.
Albrieux
C.
Joyard
J.
Lagrange
T.
Block
M. A.
Knock-out of the magnesium protoporphyrin IX methyltransferase gene in Arabidopsis: effects on chloroplast development and on chloroplast-to-nucleus signaling
J. Biol. Chem.
2007
, vol. 
282
 (pg. 
2297
-
2304
)
17
Nott
A.
Jung
H.-S.
Koussevitzky
S.
Chory
J.
Plastid-to-nucleus retrograde signaling
Ann. Rev. Plant Biol.
2006
, vol. 
57
 (pg. 
739
-
759
)
18
Ellsworth
R. K.
Dullaghan
J. P.
Activity and properties of S-adenosyl-L-methionine:magnesium-protoporphyrin IX methyltransferase in crude homogenates from wheat seedlings
Biochim. Biophys. Acta
1972
, vol. 
268
 (pg. 
327
-
333
)
19
Ellsworth
R. K.
Dullaghan
J. P.
St Pierre
M. E.
The reaction mechanism of S-adenosyl-L-methionine:magnesium protoporphyrin IX methyltransferase of wheat
Photosynthetica
1974
, vol. 
8
 (pg. 
375
-
383
)
20
Gibson
K. D.
Neuberger
A.
Tait
G. H.
Studies on the biosynthesis of porphyrin and bacteriochlorophyll by Rhodopseudomonas spheroides
Biochem. J.
1963
, vol. 
88
 (pg. 
325
-
334
)
21
Hinchigeri
S. B.
Chan
J. C.-S.
Richards
W. R.
Purification of S-adenosyl-L-methionine:magnesium protoporphyrin IX methyltransferase by affinity chromatography
Photosynthetica
1981
, vol. 
15
 (pg. 
351
-
359
)
22
Hinchigeri
S. B.
Nelson
D. W.
Richards
W. R.
The purification and reaction mechanism of S-adenosyl-L-methionine:magnesium protoporphyrin IX methyltransferase from Rhodopseudomonas sphaeroides
Photosynthetica
1984
, vol. 
18
 (pg. 
168
-
178
)
23
Hinchigeri
S. B.
Richards
W. R.
The reaction mechanism of S-adenosyl-L-methionine:magnesium protoporphyrin IX methyltransferase from Euglena gracilis
Photosynthetica
1982
, vol. 
16
 (pg. 
554
-
560
)
24
Shepherd
M.
Hunter
C. N.
Transient kinetics of the reaction catalysed by magnesium protoporphyrin IX methyltransferase
Biochem. J.
2004
, vol. 
382
 (pg. 
1009
-
1013
)
25
Shepherd
M.
Reid
J. D.
Hunter
C. N.
Purification and kinetic characterization of the magnesium protoporphyrin IX methyltransferase from Synechocystis PCC6803
Biochem. J.
2003
, vol. 
371
 (pg. 
351
-
360
)
26
Shieh
J.
Miller
G. W.
Psenak
M.
Properties of S-adenosyl-L-methionine magnesium-protoporphyrin IX methyltransferase from barley
Plant Cell Physiol.
1978
, vol. 
19
 (pg. 
1051
-
1059
)
27
Yee
W. C.
Eglsaer
S. J.
Richards
W. R.
Confirmation of a ping-pong mechanism for S-adenosyl-L-methionine:magnesium protoporphyrin methyltransferase of etiolated wheat by an exchange reaction
Biochem. Biophys. Res. Commun.
1989
, vol. 
162
 (pg. 
483
-
490
)
28
Smith
C. A.
Suzuki
J. Y.
Bauer
C. E.
Cloning and characterization of the chlorophyll biosynthesis gene chlM from Synechocystis PCC6803 by complementation of a bacteriochlorophyll biosynthesis mutant of Rhodobacter capsulatus
Plant Mol. Biol.
1996
, vol. 
30
 (pg. 
1307
-
1314
)
29
Ebbon
J. G.
Tait
G. H.
Studies on S-adenosyl-magnesium protoporphyrin methyltransferase in Euglena gracilis strain Z
Biochem. J.
1969
, vol. 
111
 (pg. 
573
-
582
)
30
Block
M. A.
Tewari
A. K.
Albrieux
C.
Maréchal
E.
Joyard
J.
The plant S-adenosyl-L-methionine:Mg-protoporphyrin IX methyltransferase is located in both envelope and thylakoid chloroplast membranes
Eur. J. Biochem.
2002
, vol. 
269
 (pg. 
240
-
248
)
31
Walker
C. J.
Willows
R. D.
Mechanism and regulation of Mg-chelatase
Biochem. J.
1997
, vol. 
327
 (pg. 
321
-
333
)
32
Willows
R. D.
Hansson
M.
Kadish
K. M.
Smith
K. M.
Guilard
R.
Mechanism, structure, and regulation of magnesium chelatase
The Porphyrin Handbook
2003
Sydney
Academic Press
(pg. 
1
-
47
)
33
Vale
R. D.
AAA proteins: lords of the ring
J. Cell Biol.
2000
, vol. 
150
 (pg. 
F13
-
F19
)
34
Willows
R. D.
Gibson
L. C. D.
Kanangara
G. C.
Hunter
C. N.
von Wettstein
D.
Three separate proteins constitute the magnesium chelatase of Rhodobacter sphaeroides
Eur. J. Biochem.
1996
, vol. 
235
 (pg. 
438
-
443
)
35
Gorchein
A.
Magnesium protoporphyrin chelatase activity in Rhodopseudomonas sphaeroides: studies with whole cells
Biochem. J.
1972
, vol. 
127
 (pg. 
97
-
106
)
36
Hinchigeri
S. B.
Hundle
B.
Richards
W. R.
Demonstration that the BchH protein of Rhodobacter capsulatus activates S-adenosyl-methionine:magnesium protoporphyrin IX methyltransferase
FEBS Lett.
1997
, vol. 
407
 (pg. 
337
-
342
)
37
Jensen
P. E.
Gibson
L. C. D.
Shepherd
F.
Smith
V.
Hunter
C. N.
Introduction of a new branchpoint in tetrapyrrole biosynthesis in Escherichia coli by co-expression of genes encoding the chlorophyll-specific enzymes magnesium chelatase and magnesium protoporphyrin methyltransferase
FEBS Lett.
1999
, vol. 
455
 (pg. 
349
-
354
)
38
Shepherd
M.
McLean
S.
Hunter
C. N.
Kinetic basis for linking the first two enzymes of chlorophyll biosynthesis
FEBS J.
2005
, vol. 
272
 (pg. 
4532
-
4539
)
39
Alawady
A.
Reski
R.
Yaronskaya
E.
Grimm
B.
Cloning and expression of the tobacco CHLM sequence encoding Mg protoporphyrin IX methyltransferase and its interaction with Mg chelatase
Plant Mol. Biol.
2005
, vol. 
57
 (pg. 
679
-
691
)
40
Alawady
A. E.
Grimm
B.
Tobacco Mg protoporphyrin IX methyltransferase is involved in inverse activation of Mg porphyrin and protoheme synthesis
Plant J.
2005
, vol. 
41
 (pg. 
282
-
290
)
41
Bradford
M. M.
A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding
Anal. Biochem.
1976
, vol. 
72
 (pg. 
248
-
254
)
42
Willows
R. D.
Beale
S. I.
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.
1998
, vol. 
273
 (pg. 
34206
-
34213
)
43
Reference deleted
44
Boyle
S. M.
Markham
G. D.
Hafner
E. W.
Wright
J. M.
Tabor
H.
Tabor
C. W.
Expression of the cloned genes encoding the putrescine biosynthetic enzymes and methionine adenosylmethyltransferase of Escherichia coli speA, speB, speC and metK
Gene
1984
, vol. 
30
 (pg. 
129
-
136
)
45
Markham
G. D.
Hafner
E. W.
Tabor
C. W.
Tabor
H.
S-Adenosylmethionine synthetase from Escherichia coli
J. Biol. Chem.
1980
, vol. 
255
 (pg. 
9082
-
9092
)
45a
Schägger
H.
von Jagow
G.
Tricine-sodium dodecyl sulfate-polyacrylamide gel electrophoresis for the separation of proteins in the range from 1 to 100 kDa
Anal. Biochem.
1987
, vol. 
166
 (pg. 
368
-
379
)
46
Fuhrhop
J.
Granick
S.
Magnesium protoporphyrin dimethyl ester and the dipotassium salt of magnesium protoporphyrin
Biochemical Preparations
1971
, vol. 
13
 (pg. 
55
-
58
)
47
Falk
J. E.
Laboratory methods
Porphyrins and Metalloporphyrins: Their General, Physical And Coordination Chemistry, And Laboratory Methods
1964
New York
Elsevier Publishing Company
(pg. 
135
-
141
)
48
Brisbin
D. A.
Richards
G. D.
Kinetics of the reaction of some first-row transition metals with protoporphyrin IX dimethyl ester
Inorg. Chem.
1972
, vol. 
11
 (pg. 
2849
-
2851
)
49
Karger
G. A.
Reid
J. D.
Hunter
C. N.
Characterization of the binding of the deuteroporphyrin IX to the magnesium chelatase H subunit and spectroscopic properties of the complex
Biochemistry
2001
, vol. 
40
 (pg. 
9291
-
9299
)
50
Dalziel
K.
Dawson
R. M. C.
Elliott
D. C.
Elliott
W. H.
Jones
K. M.
Porphyrins and related compounds
Data for Biochemical Research
1959
London
Oxford University Press
pg. 
134
 
51
Loach
P. A.
Calvin
M.
Oxidation states of manganese hematoporphyrin IX in aqueous solution
Biochemistry
1963
, vol. 
2
 (pg. 
361
-
371
)
52
Ellsworth
R. K.
Murphy
S. J.
Enzymatic preparation of Mg-protoporphyrin IX monomethyl ester
Photosynthetica
1978
, vol. 
12
 (pg. 
81
-
82
)
53
Osborn
M. J.
Rothfield
L. I.
Formation of lipopolysaccharide in mutant strains of Salmonella typhimurium
Methods Enzymol.
1966
, vol. 
8
 (pg. 
456
-
466
)
54
Burton
K.
Dawson
R. M. C.
Elliott
D. C.
Elliott
W. H.
Jones
K. M.
Spectral data and pK values for purines, pyrimidines, nucleosides and nucleotides
Data for Biochemical Research
1959
London
Oxford University Press
pg. 
74
 
55
Opekarová
M.
Tanner
W.
Specific lipid requirements of membrane proteins: a putative bottleneck in heterologous expression
Biochim. Biophys. Acta
2003
, vol. 
1610
 (pg. 
11
-
22
)
56
Russell
N. J.
Harwood
J. L.
Changes in the acyl lipid composition of photosynthetic bacteria grown under photosynthetic and non-photosynthetic conditions
Biochem. J.
1979
, vol. 
181
 (pg. 
339
-
345
)
57
Tait
G. H.
Gibson
K. D.
The enzymic formation of magnesium protoporphyrin monomethyl ester
Biochim. Biophys. Acta
1961
, vol. 
52
 (pg. 
614
-
616
)
58
Sreerama
N.
Woody
R. W.
Estimation of protein secondary structure from circular dichroism spectra: comparison of CONTIN, SELCON, and CDSSTR methods with an expanded reference set
Anal. Biochem.
2000
, vol. 
287
 (pg. 
252
-
260
)
59
Cleland
W. W.
The kinetics of enzyme-catalyzed reactions with two or more substrates or products. I. Nomenclature and rate equations
Biochim. Biophys. Acta
1963
, vol. 
67
 (pg. 
104
-
137
)
60
Radmer
R. J.
Bogorad
L.
S-adenosyl-L-methionine:magnesium protoporphyrin methyltransferase, an enzyme in the biosynthetic pathway of chlorophyll in Zea mays
Plant Physiol.
1967
, vol. 
42
 (pg. 
463
-
465
)
61
Falk
J. E.
Further coordination by metalloporphyrins
Porphyrins and Metalloporphyrins: Their General, Physical And Coordination Chemistry, And Laboratory Methods
1964
New York
Elsevier Publishing Company
(pg. 
41
-
66
)
62
Scheidt
W. R.
Kadish
K. M.
Smith
K. M.
Guilard
R.
Systematics of the stereochemistry of porphyrins and metalloporphyrins
The Porphyrin Handbook
2000
Sydney
Academic Press
(pg. 
49
-
112
)

Supplementary data