Biosynthetic studies on chlorophylls: from protoporphyrin IX to protochlorophyllide.

The series of reactions leading from protoporphyrin IX to protochlorophyllide have been studied over the last 15 years in the authors' laboratories at Davis and Clemson. Here, two crucial steps are emphasized, the discovery of the ATP requirement for Mg2+ chelation, and the oxidative cyclization of Mg-protoporphyrin IX monomethyl ester to protochlorophyllide. The in vitro systems for the chelation of Mg2+ and for the oxidative cyclization of Mg-protoporphyrin IX monomethyl ester both require membrane-associated and soluble heat-labile components. We speculate about the enzymological mechanisms of these important reactions, their sub-plastidic localization and the relationship of these individual steps to the broader questions of chloroplast and cell development.

Further observations on the Mg-protoporphyrin IX monomethyl ester (oxidative) cyclase system.

The Mg-protoporphyrin IX monomethyl ester (oxidative) cyclase was strongly inhibited by CN- and N3- in a reconstituted system, but was inhibited slightly or not at all by the same reagents in intact developing chloroplasts. Known inhibitors of cytochrome P-450 processes showed no consistent effect. Benzoquinone and quinol, which can give rise to the same semiquinone by one-electron redox events, were strong inhibitors of the cyclase. It was previously shown that O2 and a source of electrons are required in the cyclization process. The substrates for the dehydrogenases of the pentose phosphate pathway (glucose 6-phosphate and 6-phosphogluconate) were effective reductants in the reconstituted system with supernatant that had been dialysed or passed through Sephadex G-50, in the absence of added NADP+. However, inhibitor studies suggested that the electrons from these sugar phosphates reached the cyclase system via NADPH. Therefore we infer the presence of protein-bound NADP+ that can be reduced by glucose 6-phosphate and 6-phosphogluconate and donate reducing equivalents to the cyclase system. This bound NADPH pool may be particularly effective in the cyclization process, owing to channeling. These findings are discussed in relation to the results of a companion paper [Whyte and Castelfranco (1993) Biochem. J. 290, 361-367] on the breakdown of chloroplast pigments in the same reconstituted system.

Breakdown of thylakoid pigments by soluble proteins of developing chloroplasts.

In the presence of Triton X-100 (TX-100) or imazalil, plastidic pigments were degraded by a soluble enzyme extracted from developing chloroplasts. This bleaching was not photochemical and required oxygen; it was not inhibited by superoxide dismutase or catalase, but was strongly inhibited by benzoquinone, quinol, phenazine methosulphate and, more weakly, by other reagents. Synthetic intermediates of chlorophyll biosynthesis, e.g. Mg(II)-protoporphyrin IX monomethyl ester, was also degraded. This reaction was compared with the bleaching catalysed by soybean (Glycine max) lipoxygenase. The plastidic system required TX-100 and was inhibited by unsaturated fatty acids, whereas lipoxygenase required a polyunsaturated fatty acid and was inhibited by TX-100. The bleaching capability of the stromal extract decreased with age if the seedlings were placed in the greenhouse to allow further development of the chloroplasts. A direct relationship was observed between the promotion of pigment bleaching by TX-100 and the inhibition of the in vitro synthesis of divinylprotochlorophyllide. This bleaching reaction is discussed on the basis of interference by TX-100 with the normal O2-requiring anabolic processes of developing chloroplasts.

Regulation of 5-Aminolevulinic Acid Synthesis in Developing Chloroplasts : IV. An Endogenous Inhibitor from the Thylakoid Membranes.

5-Aminolevulinic acid synthesis in isolated, intact, developing chloroplasts from greening cucumber (Cucumis sativus) cotyledons was inhibited by broken chloroplast fragments. It was shown that the inhibitory constituent was associated with the thylakoid membrane system. The inhibitor was resistant to boiling, was not a form of ribonuclease, and did not inhibit Mg-chelatase, indicating that massive organelle destruction was not involved. The inhibitor was also found in etioplast and mature chloroplasts; and it was found in barley as well as cucumber. 5-Aminolevulinate synthesis in the dark with exogenous ATP and NADPH, or in the light without added cofactors, were inhibited approximately equally. In the dark, 5-aminolevulinate synthesis and protochlorophyllide synthesis from glutamate were inhibited to about equal extent. The inhibition was decreased when the membranes were washed with aqueous acetone prior to incubation. The inhibition by the unknown factor was compared to the inhibition by gabaculine, 4-amino-5-hexynoic acid, protoheme, and glutathione. The unknown inhibitor appeared to have a number of similarities with protoheme.

Synthesis of divinyl protochlorophyllide. Enzymological properties of the Mg-protoporphyrin IX monomethyl ester oxidative cyclase system.

The resolution and reconstitution of the Mg-protoporphyrin IX monomethyl ester oxidative cyclase system into a supernatant and a pellet fraction was accomplished by a procedure involving salt treatment followed by osmotic shock. Recombination of pellet and supernatant fractions was required for cyclase activity. This restoration effect could be demonstrated using either Mg-protoporphyrin IX or Mg-protoporphyrin IX monomethyl ester as the cyclase substrate in the presence or absence of S-adenosylmethionine. Pretreatment of the pellet fraction with either 8-hydroxyquinoline or desferal mesylate inhibited cyclase activity, indicating that there is a heavy-metal-ion requirement in this fraction. The cyclase supernatant protein(s) was not internalized by Sephadex G-50 and did not bind to Blue Sepharose, suggesting that it has a molecular mass of over 30 kDa and that it does not bind the cofactor NADPH. The cyclase supernatant protein did bind to MgProtoMe2-bound Sepharose and could be eluted by raising the pH to 9.7 in the presence of 4 mM-n-octyl glucoside. The pH optimum of the cyclase was 9.0. About a 40-fold purification of the cyclase supernatant protein was achieved by a combination of (NH4)2SO4 fractionation and phenyl-Sepharose chromatography.

Regulation of 5-Aminolevulinic Acid (ALA) Synthesis in Developing Chloroplasts : III. Evidence for Functional Heterogeneity of the ALA Pool.

Gabaculine and 4-amino-5-hexynoic acid (AHA) up to 3.0 millimolar concentration strongly inhibited 5-aminolevulinic acid (ALA) synthesis in developing cucumber (Cucumis sativus L. var Beit Alpha) chloroplasts, while they hardly affected protochlorophyllide (Pchlide) synthesis. Exogenous protoheme up to 1.0 micromolar had a similar effect. Exogenous glutathione also exhibited a strong inhibitory effect on ALA synthesis in organello but hardly inhibited Pchlide synthesis. Pchlide synthesis in organello was highly sensitive to inhibition by levulinic acid, both in the presence and in the absence of gabaculine, indicating that the Pchlide was indeed formed from precursor(s) before the ALA dehydratase step. The synthesis of Pchlide in the presence of saturating concentrations of glutamate was stimulated by exogenous ALA, confirming that Pchlide synthesis was limited at the formation of ALA. The gabaculine inhibition of ALA accumulation occurred whether levulinic acid or 4,6-dioxohepatonic acid was used in the ALA assay system. ALA overproduction was also observed in the absence of added glutamate and was noticeable after 10-minute incubation. These observations suggest that although Pchlide synthesis in organello is limited by ALA formation, it does not utilize all the ALA that is made in the in organello assay system. Gabaculine, AHA, and probably also protoheme, inhibit preferentially the formation of that portion of ALA that is not destined for Pchlide. A model proposing a heterogenous ALA pool is described.

Regulation of 5-Aminolevulinic Acid (ALA) Synthesis in Developing Chloroplasts : II. Regulation of ALA-Synthesizing Capacity by Phytochrome.

When dark-grown cucumber (Cucumis sativus L.) seedlings previously exposed to white light for 20 hours were returned to darkness, the ability of isolated chloroplasts to synthesize 5-aminolevulinic acid dropped by approximately 70% within 1 hour. The seedlings were then exposed to light, and the synthetic ability of the isolated chloroplasts was determined. Restoration of the synthetic capacity was promoted by continuous white or red light of moderate intensity. Intermittent red light was also effective. Blue and far-red light did not restore the synthetic capability. Blue light given after a red pulse did not enhance the effect of the red light. Far-red light given immediately after each red pulse prevented the stimulation due to intermittent red light. Restoration of the biosynthetic activity by in vivo light treatments was inhibited by cycloheximide indicating the requirement for translation on 80 S ribosomes for the in vivo light response. These findings suggest that the majority of the plastidic 5-aminolevulinic acid synthesis is under phytochrome regulation.

Regulation of 5-aminolevulinic Acid synthesis in developing chloroplasts : I. Effect of light/dark treatments in vivo and in organello.

Intact chloroplasts isolated from greening cucumber (Cucumis sativus L. var Beit Alpha) cotyledons regenerated protochlorophyllide (Pchlide) in the dark with added cofactors from either exogenous glutamate or endogenous substrates. No other intermediates of the chlorophyll biosynthetic pathway accumulated. When inhibitors of 5-aminolevulinic acid (ALA) dehydratase were added, the Pchlide that failed to form was replaced by an excessive amount of ALA. When greening seedlings were returned to the dark, ALA-synthesizing activity in the isolated chloroplasts decreased dramatically and recovered if the dark-treated seedlings were again exposed to continuous white light prior to chloroplast isolation. Both the decline and the recovery of ALA-synthesizing activity were complete in approximately 50 minutes. Changes in chloroplast structure during in vivo light to dark and dark to light transitions (as evidenced by electron microscopy) were much slower. Exposing isolated chloroplasts from dark-treated seedlings to short white flashes before incubation transformed nearly all the endogenous Pchlide, but hardly stimulated ALA synthesis, suggesting that Pchlide does not act as a feed-back inhibitor on ALA synthesis. Chloroplasts isolated from dark-treated tissue did not form Pchlide from glutamate when incubated in the dark with added cofactors; moreover, the endogenous Pchlide did not turn over in organello. However, these chloroplasts did synthesize Pchlide from added ALA at the normal rate and synthesized ALA from glutamate at a reduced, but still significant, rate. Mg chelation was not affected by in vivo dark treatment.

Incorporation of atmospheric oxygen into the carbonyl functionality of the protochlorophyllide isocyclic ring.

Detached cucumber (Cucumis sativus L. var. Beit Alpha) cotyledons incubated in darkness with 5-aminolaevulinic acid and either 16O2 air (control) or 18O2 in N2 accumulated protochlorophyllide. This was converted into methyl phaeoporphyrin alpha 5 and analysed by mass spectrometry. The molecular ion of the methyl phaeoporphyrin alpha 5 derived from the 18O2 incubation was 2 mass units greater than that of the control, establishing that the oxo oxygen atom of the isocyclic ring is derived from atmospheric oxygen.

The magnesium-protoporphyrin IX (oxidative) cyclase system. Studies on the mechanism and specificity of the reaction sequence.

Mg-protoporphyrin IX monomethyl ester cyclase activity was assayed in isolated developing cucumber (Cucumis sativus L. var. Beit Alpha) chloroplasts [Chereskin, Wong & Castelfranco (1982) Plant Physiol. 70, 987-993]. The presence of both 6- and 7-methyl esterase activities was detected, which permitted the use of diester porphyrins in a substrate-specificity study. It was found that: (1) the 6-methyl acrylate derivative of Mg-protoporphyrin monomethyl ester was inactive as a substrate for cyclization; (2) only one of the two enantiomers of 6-beta-hydroxy-Mg-protoporphyrin dimethyl ester had detectable activity as a substrate for the cyclase; (3) the 2-vinyl-4-ethyl-6-beta-oxopropionate derivatives of Mg-protoporphyrin mono- or di-methyl ester were approx. 4 times more active as substrates for cyclization than the corresponding divinyl forms; (4) at the level of Mg-protoporphyrin there was no difference in cyclase activity between the 4-vinyl and 4-ethyl substrates; (5) reduction of the side chain of Mg-protoporphyrin in the 2-position from a vinyl to an ethyl resulted in a partial loss of cyclase activity. This work suggests that the original scheme for cyclization proposed by Granick [(1950) Harvey Lect. 44, 220-245] should now be modified by the omission of the 6-methyl acrylate derivative of Mg-protoporphyrin monomethyl ester and the introduction of stereo-specificity at the level of the hydroxylated intermediate.

Labeling of porphobilinogen deaminase by radioactive 5-aminolevulinic acid in isolated developing pea chloroplasts.

A protein had been previously described, which was labeled by radioactive 5-aminolevulinic acid in isolated developing chloroplasts. In the present study we have shown that this protein (Mr approximately equal to 43,000) probably exists as a monomer in the chloroplast stroma. The labeling is blocked if known inhibitors of 5-aminolevulinic acid dehydratase are added to the incubation mixture, and is markedly decreased in intensity if nonradioactive 5-aminolevulinate or porphobilinogen are added to the incubation mixture; other intermediates in the porphyrin biosynthetic pathway, uroporphyrinogen III, uroporphyrin III, and protoporphyrin IX, do not decrease the labeling of the 43-kDa protein appreciably. Nondenaturing gels of the proteins isolated from the incubation with radioactive 5-aminolevulinic acid were stained for porphobilinogen deaminase activity. A series of red fluorescent bands was obtained which coincided with the radioactive bands visualized by autoradiography. It is concluded that the soluble chloroplast protein that is labeled in organello by radioactive 5-aminolevulinic acid is porphobilinogen deaminase.

Regeneration of Magnesium-2,4-Divinylpheoporphyrin a(5) (Divinyl Protochlorophyllide) in Isolated Developing Chloroplasts.

A preparation of developing chloroplasts isolated from greening cucumber (Cucumis sativus L. var Beit Alpha) cotyledons was found capable of synthesizing divinyl protochlorophyllide (magnesium-2,4-divinylpheoporphyrin a(5)) in the presence of glutamate, adenosine triphosphate, reducing power, S-adenosyl-l-methionine, and molecular oxygen. Both adenosine triphosphate and molecular oxygen were absolutely required while each of the other three was strongly promotive. Organelle intactness was essential. The divinyl protochlorophyllide (Pchlide) formed in vitro could be completely phototransformed. Regeneration of Pchlide was not inhibited by 0.3 millimolar chloramphenicol. The initial in vitro rate of Pchlide regeneration was considerably higher than the rate of Pchlide synthesis observed when greened cucumber seedlings were returned to darkness. However, Pchlide synthesis in vitro fell off exponentially with a half-life of approximately 21 minutes, whereas Pchlide synthesis in vivo was linear for at least 100 minutes. It is likely that the leveling off of the in vitro rate is due to the loss of chloroplast integrity during the incubation, because neither adding more cofactors, nor phototransforming the accumulated Pchlide in the middle of the incubation period, restored the high initial rate of Pchlide synthesis.

Synthesis of a Putative c-Type Cytochrome by Intact, Isolated Pea Chloroplast.

In addition to chlorophyll-protein complexes, other proteins were labeled when isolated developing pea (Pisum sativum L.) chloroplasts were incubated with [(14)C]-5-aminolevulinic acid. The major labeled band (M(r) = 43 kilodaltons by lithium dodecyl sulfate-polyacrylamide gel electrophoresis) was labeled even in the presence of chloramphenicol. Heme-dependent peroxidase activity (as detected by the tetramethyl benzidine-H(2)O(2) stain) was not visibly associated with this band. The radioactive band was stable to heat, 5% HCl in acetone, and was absent if the incubation with [(14)C]-5-aminolevulinic acid was carried out in the presence of N-methyl protoporphyrin IX dimethyl ester (a specific inhibitor of ferrochelatase). Organic solvent extraction procedures for the enrichment of cytochrome f from chloroplast membranes also extracted this unknown labeled product. It was concluded that this labeled product was probably a c-type cytochrome; however, the possibility that it might be a protein containing a covalently linked linear tetrapyrrole was not ruled out.

Intermediates in the formation of the chlorophyll isocyclic ring.

Cell-free, organelle-free synthesis of Mg-2,4-divinylpheoporphyrin a(5) (MgDVP) from Mg-protoporphyrin IX monomethyl ester (Mg-Proto Me) has been described (Wong and Castelfranco 1984 Plant Physiol 75: 658-661). This system consists of plastid membrane and stromal fractions and requires O(2), NAD(P)H and S-adenosylmethionine (SAM). The synthetic 6-methyl-beta-ketopropionate analog of Mg-Proto Me was converted to MgDVP by the same catalytic system in the presence of O(2) and NADPH. SAM was not required. A compound (X) displaying the kinetic behavior of an intermediate was isolated from reaction mixtures with Mg-Proto Me as the substrate, but not with the 6-methyl-beta-ketopropionate analog as the substrate. X was identified as the 6-methyl-beta-hydroxypropionate analog of Mg-Proto Me by conversion to the dimethyl ester with CH(2)N(2) and comparison with authentic 6-beta-hydroxydimethyl ester. X was converted to MgDVP by the same catalytic system in the presence of O(2) and NADPH. We conclude that the conversion of Mg-Proto Me to MgDVP proceeds through the 6-beta-hydroxy and the 6-beta-ketopropionate esters in agreement with earlier suggestions.

Properties of the Mg-Protoporphyrin IX Monomethyl Ester (Oxidative) Cyclase System.

Mg-protoporphyrin IX monomethyl ester (oxidative) cyclase, the enzyme system responsible for the formation of the chlorophyll isocyclic ring, exhibits requirements for both essential sulfhydryls and essential disulfides. It is inhibited by N-ethylmaleimide, dithiothreitol, and beta-mercaptoethanol, but not by sodium arsenite. This enzyme system shows some substrate specificity: (a) the 6-side-chain of the macrocycle can either be a methyl propionate ester, or its beta-hydroxy or beta-keto derivatives; (b) the 7-side-chain can either be a propionic acid or a methyl propionate ester; (c) both the 4-vinyl and the 4-ethyl series can serve as substrates, at least at the beta-keto ester level; (d) the activity appears to be lost if the side-chain in the 2-position is reduced from a vinyl to an ethyl.

Chlorophyll biosynthesis and assembly into chlorophyll-protein complexes in isolated developing chloroplasts.

Isolated developing plastids from greening cucumber cotyledons or from photoperiodically grown pea seedlings incorporated (14)C-labeled 5-aminolevulinic acid (ALA) into chlorophyll (Chl). Incorporation was light dependent, enhanced by S-adenosylmethionine, and linear for 1 hr. The in vitro rate of Chl synthesis from ALA was comparable to the in vivo rate of Chl accumulation. Levulinic acid and dioxoheptanoic acid strongly inhibited Chl synthesis but not plastid protein synthesis. Neither chloramphenicol nor spectinomycin affected Chl synthesis, although protein synthesis was strongly inhibited. Components of thylakoid membranes from plastids incubated with [(14)C]ALA were resolved by electrophoresis and then subjected to autoradiography. This work showed that (i) newly synthesized Chl was assembled into Chl-protein complexes and (ii) the inhibition of protein synthesis during the incubation did not alter the labeling pattern. Thus, there was no observable short-term coregulation between Chl synthesis (from ALA) and the synthesis of membrane proteins in isolated plastids.

Resolution and Reconstitution of Mg-Protoporphyrin IX Monomethyl Ester (Oxidative) Cyclase, the Enzyme System Responsible for the Formation of the Chlorophyll Isocyclic Ring.

Mg-protoporphyrin IX monomethyl ester (oxidative) cyclase, the system responsible for the formation of the chlorophyll isocyclic ring in developing cucumber (Cucumis sativus L. cv Beit Alpha) chloroplasts, was resolved into two enzymic components: a high-speed supernatant and a membrane pellet. This reconstituted enzyme system required reduced pyridine nucleotide for activity.

Localization of Mg-Chelatase and Mg-Protoporphyrin IX Monomethyl Ester (Oxidative) Cyclase Activities within Isolated, Developing Cucumber Chloroplasts.

Magnesium chelatase and magnesium protoporphyrin IX monomethyl ester (oxidative) cyclase activities were both sensitive to inhibition by p-chloromercuribenzoate in intact, developing cucumber (Cucumis sativus L. var Beit Alpha) chloroplasts. Magnesium chelatase was also sensitive to the membrane-impermeable mercurial p-chloromercuribenzene sulfonate (PCMBS), while cyclase activity was only slightly sensitive. When the plastids were pretreated with PCMBS, triosephosphate dehydrogenase activity was inhibited very slightly, indicating that PCMBS does not readily penetrate through the chloroplast envelope. These results suggest that magnesium chelatase is located in the chloroplast envelope, while the cyclase is located deeper within the chloroplast.

Formation of Mg-Containing Chlorophyll Precursors from Protoporphyrin IX, delta-Aminolevulinic Acid, and Glutamate in Isolated, Photosynthetically Competent, Developing Chloroplasts.

Intact developing chloroplasts isolated from greening cucumber (Cucumis sativus L. var Beit Alpha) cotyledons were found to contain all the enzymes necessary for the synthesis of chlorophyllide. Glutamate was converted to Mg-protoporphyrin IX (monomethyl ester) and protoclorophyllide. delta-Aminolevulinic acid and protoporphyrin IX were converted to Mg-protoporphyrin IX, Mg-protoporphyrin IX monomethyl ester, protochlorophyllide and chlorophyllide a. The conversion of delta-aminolevulinic acid or protoporphyrin IX to Mg-protoporphyrin IX (monomethyl ester) was inhibited by AMP and p-chloromercuribenzene sulfonate. Light stimulated the formation of Mg-protoporphyrin IX from all three substrates. In the case of delta-aminolevulinic acid and protoporphyrin IX, light could be replaced by exogenous ATP. In the case of glutamate, both ATP and reducing power were necessary to replace light. With all three substrates, glutamate, delta-aminolevulinic acid, and protoporphyrin IX, the stimulation of Mg-protoporphyrin IX accumulation in the light was abolished by DCMU, and this DCMU block was overcome by added ATP and reducing power.

Mg-2,4-divinyl pheoporphyrin a5: the product of a reaction catalyzed in vitro by developing chloroplasts.

The major product of an aerobic reaction mixture containing developing chloroplasts, Mg-protoporphyrin IX, S-adenosylmethionine, and other cofactors was isolated and purified. Structural studies using nuclear magnetic resonance confirmed earlier reports, based on fluorescence and absorption spectra, that this compound is Mg-2,4-divinyl pheoporphyrin a5. The molecular weight determined by secondary-ion mass spectroscopy further confirmed the assigned structure. Absorption and fluorescence spectra indicate that this compound is identical to that reported previously by various workers in less-purified biological extracts. The nuclear magnetic resonance spectrum of the Mg-free base also supports the assigned structure.