Search

Insertion of magnesium into protoporphyrin IX is a complex ATP-dependent reaction catalysed by the enzyme Mg-chelatase. Three separate proteins (Mg-chelatase subunits), designated as D, H and I, are involved in the chelation reaction. The genes encoding the Mg-chelatase subunits of the green sulfur bacterium Chlorobium vibrioforme and of the cyanobacterium Synechocystis strain PCC6803 were expressed in Escherichia coli. The recombinant proteins were purified, tested for ATPase and phosphate exchange activities, and compared with the activities of the corresponding subunits of Rhodobacter sphaeroides. The Synechocystis strain PCC6803 I subunit and the C. vibrioforme H and I subunits hydrolysed ATP at the rates of 2.0, 1.8 and 0.16 nmol (mg protein)-1 min-1, respectively. The ATPase activity of the C. vibrioforme H subunit was similar to that reported for the R. sphaeroides H subunit. The Synechocystis strain PCC6803 H subunit failed to hydrolyse ATP. The I subunit of Synechocystis strain PCC6803 and C. vibrioforme catalysed a transfer of PO4 from ATP to ADP (exchange activity) at the rate of 1.75 +/- 0.15 nmol (mg protein)-1 min-1. This exchange rate was 300-fold lower than that reported for the R. sphaeroides I subunit. The PO4 exchange activities were correlated with the presence of the sequence GXRGTGKSTXVRALA in the primary structure of the three I subunits. Mg-chelatase activity was reconstituted by combining the three subunits of the same bacterium [rates of 41-89 pmol Mg-deuteroporphyrin (mg protein)-1 min-1]. Heterologous subunit combinations resulted in low or no Mg-chelatase activity.

Magnesium-protoporphyrin chelatase, the first enzyme unique to the (bacterio)chlorophyll-specific branch of the porphyrin biosynthetic pathway, catalyzes the insertion of Mg2+ into protoporphyrin IX. Three genes, designated bchI, -D, and -H, from the strictly anaerobic and obligately phototrophic green sulfur bacterium Chlorobium vibrioforme show a significant level of homology to the magnesium chelatase-encoding genes bchI, -D, and -H and chlI, -D, and -H of Rhodobacter sphaeroides and Synechocystis strain PCC6803, respectively. These three genes were expressed in Escherichia coli; the subsequent purification of overproduced BchI and -H proteins on an Ni2+-agarose affinity column and denaturation of insoluble BchD protein in 6 M urea were required for reconstitution of Mg-chelatase activity in vitro. This work therefore establishes that the magnesium chelatase of C. vibrioforme is similar to the magnesium chelatases of the distantly related bacteria R. sphaeroides and Synechocystis strain PCC6803 with respect to number of subunits and ATP requirement. In addition, reconstitution of an active heterologous magnesium chelatase enzyme complex was obtained by combining the C. vibrioforme BchI and -D proteins and the Synechocystis strain PCC6803 ChlH protein. Furthermore, two versions, with respect to the N-terminal start of the bchI gene product, were expressed in E. coli, yielding ca. 38- and ca. 42-kDa versions of the BchI protein, both of which proved to be active. Western blot analysis of these proteins indicated that two forms of BchI, corresponding to the 38- and the 42-kDa expressed proteins, are also present in C. vibrioforme.

A region comprising approximately 25 kbp of the genome of the strictly anaerobic and obligate photosynthetic green sulfur bacterium Chlorobium vibrioforme has been mapped, subcloned and partly sequenced. Approximately 15 kbp have been sequenced in it's entirety and three genes with significant homology and feature similarity to the bchI, -D and -H genes and the chlI, -D and -H genes of Rhodobacter and Synechocystis strain PCC6803, respectively, which encode magnesium chelatase subunits, have been identified. Magnesium chelatase catalyzes the insertion of Mg2+ into protoporphyrin IX, and is the first enzyme unique to the (bacterio)chlorophyll specific branch of the porphyrin biosynthetic pathway. The organization of the three Mg-chelatase encoding genes is unique to Chlorobium and suggests that the magnesium chelatase of C. vibrioforme is encoded by a single operon. The analyzed 25 kbp region contains five additional open reading frames, two of which display significant homology and feature similarity to genes encoding lipoamide dehydrogenase and genes with function in purine synthesis, and another three display significant homology to open reading frames with unknown function in distantly related bacteria. Putative E. coli sigma 70-like promoter sequences, ribosome binding sequences and rho-independent transcriptional stop signals within the sequenced 15 kbp region are related to the identified genes and orfs. Southern analysis, restriction mapping and partial sequencing of the remaining ca. 10 kbp of the analyzed 25 kbp region have shown that this part includes the hemA, -C, -D and -B genes (MOBERG and AVISSAR 1994), which encode enzymes with function in the early part of the biosynthetic pathway of porphyrins.

Magnesium-protoporphyrin chelatase catalyzes the first step unique to chlorophyll synthesis: the insertion of Mg2+ into protoporphyrin IX. Genes from Synechocystis sp. PCC6803 with homology to the bchI and bchD genes of Rhodobacter sp. were cloned using degenerate oligonucleotides. The function of these genes, putatively encoding subunits of magnesium chelatase, was established by overexpression in Escherichia coli, including the overexpression of Synechocystis chlH, previously cloned as a homolog of the Rhodobacter bchH gene. The combined cell-free extracts were able to catalyze the insertion of Mg2+ into protoporphyrin IX in an ATP-dependent manner and only when the products of all three genes were present. The ChlH, ChlI, and ChlD gene products are therefore assigned to the magnesium chelatase step in chlorophyll a biosynthesis in Synechocystis PCC6803. The primary structure of the Synechocystis ChlD protein reveals some interesting features; the N-terminal half of the protein shows 40-41% identity to Rhodobacter BchI and Synechocystis ChlI, whereas the C-terminal half displays 33% identity to Rhodobacter BchD. This suggests a functional as well as an evolutionary relationship between the "I" and "D" genes.

Barley mutants in the loci Xantha-f, Xantha-g and Xantha-h, when fed with 5-aminolevulinate in the dark, accumulate protoporphyrin IX. Mutant alleles at these loci that are completely blocked in protochlorophyllide synthesis are also blocked in development of prolamellar bodies in etioplasts. In contrast to wild type, the xan-f, -g and -h mutants had no detectable Mg-chelatase activity, whereas they all had methyltransferase activity for synthesis of Mg-protoporphyrin monomethyl ester. Antibodies recognising the CH42 protein of Arabidopsis thaliana and the OLIVE (OLI) protein of Antirrhinum majus immunoreacted in wild-type barley with 42 and 150 kDa proteins, respectively. The xan-h mutants lacked the protein reacting with antibodies raised against the CH42 protein. Two xan-f mutants lacked the 150 kDa protein recognised by the anti-OLI antibody. Barley genes homologous to the A. majus olive and the A. thaliana Ch-42 genes were cloned using PCR and screening of cDNA and genomic libraries. Probes for these genes were applied to Northern blots of RNA from the xantha mutants and confirmed the results of the Western analysis. The mutants xan-f27, -f40, -h56 and -h57 are defective in transcript accumulation while -h38 is defective in translation. Southern blot analysis established that h38 has a deletion of part of the gene. Mutants xan-f10 and -f41 produce both transcript and protein and it is suggested that these mutations are in the catalytic sites of the protein. It is concluded that X an-f -h genes encode two subunits of the barley Mg-chelatase and that X an-g is likely to encode a third subunit. The XAN-F protein displays 82% amino acid sequence identity to the OLI protein of Antirrhinum, 66% to the Synechocystis homologue and 34% identity to the Rhodobacter BchH subunit of Mg-chelatase. The XAN-H protein has 85% amino acid sequence identity to the Arabidopsis CH42 protein, 69% identity to the Euglena CCS protein, 70% identity to the Cryptomonas BchA and Olisthodiscus CssA proteins, as well as 49% identity to the Rhodobacter BchI subunit of Mg-chelatase. Identification of the barley X an-f and X an-h encoded proteins as subunits required for Mg-chelatase activity supports the notion that the Antirrhinum OLI protein and the Arabidopsis Ch42 protein are subunits of Mg-chelatase in these plants. The expression of both thet X an-f and -h genes in wild-type barley is light induced in leaves of greening seedlings, and in green tissue the genes are under the control of a circadian clock.

A structural gene encoding nitrite reductase (NiR) in bean (Phaseolus vulgaris) has been cloned and sequenced. The NiR gene is present as a single copy encoding a protein of 582 amino acids. The bean NiR protein is synthesized as a precursor with an amino-terminal transit peptide (TP) consisting of 18 amino acid residues. The bean NiR transit peptide shows similarity to the TPs of other known plant NiRs. The NiR gene is expressed in trifoliate leaves and in roots of 20-day old bean plants where transcript accumulation is nitrate-inducible. Gene expression occurs in a circadian rhythm and induced by light in leaves of dark-adapted plants. A particular 100 bp sequence is present in the promoter and in the first intron of the NiR gene. Several copies of this 100 bp sequence are present in the bean genome. Comparisons between the promoter of the bean NiR gene and of two bean nitrate reductase genes (NR1 and NR2) show a limited number of conserved motifs, although the genes are presumed to be co-regulated. Comparisons are also made between the bean NiR promoter and the spinach NiR promoter. Transformation of tobacco plants with the bean NiR promoter fused to the GUS reporter gene (beta-glucuronidase) shows that the bean NiR promoter is nitrate-regulated and that the presence of the 100 bp sequence influences the level of GUS activity. NiR-coding sequences are not required for nitrate regulation but have a quantitative effect on the measured GUS activity.

Eight chlorophyll b deficient nuclear mutants of pea (Pisum sativum L.) have been characterized by low temperature fluorescence emission spectra of their leaves and by the ultrastructure, photochemical activities and polypeptide compositions of the thylakoid membranes. The room temperature fluorescence induction kinetics of leaves and isolated thylakoids have also been recorded. In addition, the effects of Mg(2+) on the fluorescence kinetics of the membranes have been investigated. The mutants are all deficient in the major polypeptide of the light-harvesting chlorophyll a/b protein of photosystem II. The low temperature fluorescence emission spectra of aurea-5106, xantha-5371 and -5820 show little or no fluorescence around 730 nm (photosystem I fluorescence), but possess maxima at 685 and 695 nm (photosystem II fluorescence). These three mutants have low photosystem II activities, but significant photosystem I activities. The long-wavelength fluorescence maximum is reduced for three other mutants. The Mg(2+) effect on the variable component of the room temperature fluorescence (685 nm) induction kinetics is reduced in all mutants, and completely absent in aurea-5106 and xantha-5820. The thylakoid membranes of these 2 mutants are appressed pairwise in 2-disc grana of large diameter. Chlorotica-1-206A and-130A have significant long-wavelength maxima in the fluorescence spectra and show the largest Mg(2+) enhancement of the variable part of the fluorescence kinetics. These two mutants have rather normally structured chloroplast membranes, though the stroma regions are reduced. The four remaining mutants are in several respects of an intermediate type.

The gene for the 44 kD chlorophyll a-binding photosystem II polypeptide has been localized on the pea (Pisum sativum) chloroplast genome. The nucleotide sequence of the gene and its flanking regions has been analyzed. The gene codes for a polypeptide of 473 amino acid residues and is possibly cotranscribed with the gene for the D2 photosystem II polypeptide with which it has 50 bp in common. The amino acid sequences of the 44 kD polypeptides from pea, spinach and maize are approximately 95% homologous. Within the 1 kb fragment 3' to the 44 kD gene a 93 bp tRNA-Ser (UGA) gene and an open reading frame of 62 codons (ORF 62) were identified. Both show high homology to corresponding genes 3' to the 44 kD genes from spinach, maize and barley. The 44 kD gene and ORF 62 are encoded in the same strand, and have putative promoter sequences, ribosome binding sites and transcription termination signals.

The gene for the membrane polypeptide D2 has been mapped on the pea (Pisum sativum) chloroplast genome. The nucleotide sequence of the gene and its flanking regions is presented. The only large open reading frame in the sequence codes for a protein of MW 39.5 kD. A potential ribosome binding site is located 6 nucleotides upstream from the initiation codon and there are two sets of putative promotor sequences in the 5' flanking region. The polypeptide has a high content of hydrophobic amino acids, predominatly grouped in clusters of 20 or more residues. The 3' end of the D2 gene is overlapped by 50 nucleotides of a second open reading frame (UORF I) which is at least 369 nucleotides long. Based on current data we suggest the D2 polypeptide to be a constituent of photosystem II (PSII).

A simple, rapid, and inexpensive method for the preparation and purification of chloroplast DNA (cpDNA) from pea has been developed. The crucial step is the isolation of chloroplasts in a medium of high ionic strength (I congruent equal to 1.40 M). CpDNA from pea prepared according to this method has successfully been used for restriction enzyme mapping, Southern transfers, and cloning.

A cDNA clone encoding the barley photosystem I polypeptide which migrates with an apparent molecular mass of 16 kDa on SDS-polyacrylamide gels has been isolated. The 634 bp sequence of this clone has been determined and contains one large open reading frame coding for a 15,457 Da precursor polypeptide. The molecular mass of the mature polypeptide is 10,821 Da. The amino acid sequence of the transit peptide indicates that the polypeptide is routed towards the stroma side of the thylakoid membrane. The hydropathy plot of the polypeptide shows no membrane-spanning regions.

The nucleotide sequence of a 1082 bp fragment from the pea (Pisum sativum) chloroplast genome is presented. This fragment contains genes for tRNAGlu, tRNATyr and tRNAAsp as well as an open reading frame (ORF) of 91 codons on one strand and two ORFs of 52 and 59 codons on the complementary strand. The tRNAAsp gene is located entirely within the ORF of 91 codons. The first 366 bp of the fragment correspond to 376 bp at one end of a recently published (1) sequence from the broad bean (Vicia faba) chloroplast genome. These regions contain the tRNAGlu and tRNATyr genes, which are identical and separated by 60 bp in both species. These two genes are probably cotranscribed. The intergenic regions in the corresponding segments from the two species are, except for a 10 bp deletion in the pea sequence, 94% homologous.

The genes encoding the two P700 chlorophyll a-apoproteins of the photosystem I complex were localized on the pea (Pisum sativum) chloroplast genome. The nucleotide sequence of the genes and the flanking regions has been determined. The genes are separated by 25 bp and are probably cotranscribed. The 5' terminal gene (psaA1) codes for a 761-residue protein (MW 84.1 kD) and the 3' terminal gene (psaA2) for a 734-residue protein (MW 82.4 kD). Both proteins are highly hydrophobic and contain eleven putative membrane-spanning domains. The homology to the corresponding polypeptides from maize are 89% and 95% for psaA1 and psaA2, respectively. A putative promoter has been identified for the psaA1 gene, and potential ribosome binding sites are present before both genes.

Protochlorophyllide and chlorophyll(ide) holochromes (Pchl-H and Chl-H) were extracted from dark-grown and greening seedlings with saponin and partly purified by ammonium sulfate fractionation. Sephadex gel filtration in the presence of saponin showed that the photoactive saponin Pchl-H from dark-grown leaves of bean (Phaseolus vulgaris L. cv. Redlands Pioneer) or pea (Pisum sativum L. cv. Greenfeast) has an apparent molecular weight of about 170,000, compared with 51,000 to 75,000 for the saponin Pchl-H from barley (Hordeum vulgare L. cv. Svalöfs Bonus). Photoconversion of saponin Pchl-H from dark-grown barley seedlings yields Chl-H with an absorption maximum at 678 nm, and with no change in apparent molecular weight. Above 0 C, a spectral shift from 678 to 672 nm follows, and a change in apparent molecular weight from about 63,000 to 29,000 is observed.Saponin Chl-H extracted from barley leaves illuminated for 15 minutes has an absorption maximum at 670 nm and an apparent molecular weight greater than 100,000. This chlorophyll holochrome has photosystem I activity and it is eluted together with the cytochromes. Saponin holochrome extracted from barley leaves returned to darkness after a light period, contains chlorophyll(ide) and protochlorophyllide complexes. Gel chromatography yields a complete separation of Chl-H (apparent molecular weight > 100,000) and photoactive Pchl-H (63,000).It is proposed that Chl-H dissociates into a chlorophyll(ide) a carrier protein complex and a photoenzyme, before the incorporation of chlorophyll into the lamellar membrane.Spectrofluorimetry on partially photoconverted preparations of saponin holochrome from barley, bean, and pea gave no indication for resonance energy transfer from protochlorophyllide to chlorophyllide. The saponin holochromes gave high polarization values, in contrast with bean holochrome extracted without the aid of detergents and bean leaves.

Isolated chloroplasts from Pisum sativum were found to contain at least 32 tRNA species. Hybridization of in vitro labeled, identified, chloroplast tRNAs to Pisum chloroplast DNA fragments revealed the locations of the tRNA genes on the circular chloroplast genome. Comparison of this gene map to the maps of Vicia faba and Phaseolus vulgaris showed that the chloroplast genomes of Pisum and Phaseolus are otherwise more closely related than either genome is to the chloroplast genome of Vicia. Furthermore, the results suggest how possible recombination events could be involved in the evolution of these three closely related, but divergent, chloroplast genomes.

Chlorophyll-deficient barley (Hordeum vulgare) mutants were studied that had chlorophyll a/b ratios either higher or lower than the wild type. Mutants with high ratios (>5.2) had a reduced proportion of their photosynthetic lamellae appressed into grana ("grana-deficient" mutants) compared with wild type (chlorophyll a/b = 3.2), while the majority of lamellae in the chloroplasts with low chlorophyll a/b ratios (2.0-2.4) were organized into grana ("grana-rich" mutants).All mutants catalyzed photosystem I and photosystem II electron transport, were tightly coupled as evidenced by increased rates of electron transport in the presence of methylamine, and were able to generate a light-dependent transmembrane proton gradient. Differences were evident in rates of electron transport per mole of chlorophyll. The mutants having high chlorophyll a/b ratios catalyzed 15- to 50-fold higher rates of ferricyanide photoreduction than the mutants having low chlorophyll a/b ratios, and 5- to 7-fold higher than the wild type.Low temperature absorption spectra of chloroplast fragments showed that the grana-deficient mutant with a high a/b ratio had a chlorophyll spectrum characteristic of a PSI preparation while mutants with the low ratio had a spectrum typical of a PSII preparation.The temperature fluorescence emission spectra of thylakoid membrane fragments from the two types of mutants were also strikingly different from one another, as were the electrophoretic patterns of the thylakoid polypeptides.

The development of photochemical activity during the greening of dark-grown barley seedlings (Hordeum vulgare L. cv. Svalöfs Bonus) was studied in relation to the formation of the high potential form of cytochrome b-559 (cytochrome b-559(HP)). Photosynthetic oxygen evolution from leaves was detected at 30 minutes of illumination. The rate of oxygen evolution per gram fresh weight of leaf was as high at 2 to 2.5 hours of greening as at 24 hours or in fully greened leaves. On a chlorophyll basis, the photosynthetic rate at 90 minutes of greening was 80-fold greater than the rate at 45 hours. It is concluded that the majority of photosynthetic units are functional at an early stage of greening, and that chlorophyll synthesis during greening serves to increase the size of the units.Plastids showed substantial photochemical oxygen evolution after a seedling greening time of 1 hour. However, a comparison of the relative activity of leaves and plastids at 2 hours and 24 hours of greening suggests that there was some inactivation of greening plastids during isolation. Appreciable photosystem I activity was observed as early as 15 minutes of greening.The synthesis of cytochrome b-559(HP) during greening does not correlate with the onset of oxygen evolution. Cytochrome b-559(HP) was absent from etioplasts and in most preparations of 2-hour plastids. The average amount of cytochrome b-559(HP) at 2 hours of greening was well below the level needed to provide 1 molecule of the carrier for each functional photosynthetic chain. The results suggest that cytochrome b-559(HP) is not essential for oxygen evolution. Cytochrome f, cytochrome b(6), and the low potential form of cytochrome b-559 were present in the etioplast. There was little increase in the levels of these cytochromes during 24 hours of greening.

A stable, soluble, and photoactive protochlorophyll(ide) complex has been extracted from dark-grown barley (Hordeum vulgare L.) leaves with buffer containing saponin and glycerol. After ammonium sulfate precipitation, the redissolved pigment complex was partially purified by chromatography on Sephadex gels in the presence of saponin. With the assumptions that the pigment complex from barley has the same shape and density as the proteins used for calibration, its molecular weight is 63,000. Photoactive protochlorophyll(ide) complex isolated from bean (Phaseolus vulgaris L.) and chromatographed by the same procedures has an aparent molecular weight of 100,000 or greater. No chromatographic separation of photoactive and inactive protochlorophyll(ide) complexes was observed. Photoconversion of protochlorophyll(ide) to chlorophyll(ide) did not change the chromatographic behavior of the pigment complex.The protochlorophyll(ide) complex had an absorption maximum at 644 nanometers and a corresponding fluorescence emission maximum at 652 nanometers. Photoconversion yielded chlorophyll(ide) complex with an absorption maximum at 678 nanometers and a main fluorescence emission maximum at 685 nanometers. Spectrofluorometry on partially photoconverted preparations gave no indication of energy transfer from protochlorophyll(ide) to chlorophyll(ide). This result is consistent with the presence of a single protochlorophyll(ide) molecule per active unit of the pigment complex.It is concluded that the protochlorophyll(ide) complex obtained from barley represents an active subunit of protochlorophyll holochrome.

Sequential changes occurring in the etioplasts of the primary leaf of 7-day-old dark-grown barley seedlings upon continuous illumination with 20 lux have been investigated by electron microscopy, in vivo spectrophotometry, and thin-layer chromatography. Following photoconversion of the protochlorophyllide pigment to chlorophyllide and the structural transformation of the crystalline prolamellar bodies, the tubules of the prolamellar bodies are dispersed into the primary lamellar layers. As both chlorophyll a and b accumulate, extensive formation of grana takes place. After 4 hr of greening, protochlorophyllide starts to reaccumulate, and concomitantly both large and small crystalline prolamellar bodies are formed. This protochlorophyllide is rapidly photoconverted upon exposure of the leaves to high light intensity, which also effects a rapid reorganization of the recrystallized prolamellar bodies into primary lamellar layers.

The effect of seedling age and of the time of greening on the incorporation of 1-(14)C-acetate into lipids by isolated barley (Hordeum vulgare cultivar Svalöf's Bonus) plastids was examined. The fatty acid synthesizing capacity of plastids isolated from 5-day-old seedlings did not increase markedly from zero to 36 hours of greening nor was a light stimulation of fatty acid synthesis observed. However, an increasing capacity for fatty acid synthesis and an increasing light stimulation of this process with greening were attained by the plastids isolated from 7-, 9-, and 11-day-old seedlings.Plastids of 7-day-old dark-grown plants, which were illuminated at 2 foot-candles showed increasing capacity of (14)C-acetate incorporation with significant flow into phospholipids and sulfolipid, low flow into digalactosyl diglyceride, and considerable flow into 6-methyl salicylic acid. Exposure of these plants to high light intensity for an hour resulted in chloroplasts that after isolation had a 10-fold increased capacity to incorporate (14)C label into digalactosyl diglyceride, while the flow of (14)C label into phospho- and sulfolipids was unaltered, and that into 6-methyl salicylic acid was drastically curtailed.With plastids from 7-day-old dark-grown plants in early stages of greening, essentially all the (14)C label in the stroma fraction could be accounted for by 6-methyl salicylic acid, while the membrane lipids only contained small amounts of (14)C label. As greening proceeded, the flow of (14)C label into 6-methyl salicylic acid diminished sharply, and the lipids of the lamellar systems became increasingly labeled.Only palmitic and oleic acids were main sites of (14)C label in the membrane lipids.The activity of acetyl CoA carboxylase present in plastids of 5- and 7-day-old dark-grown plants fell sharply as the etioplasts differentiated into chloroplasts.

Protein bodies were isolated intact from dormant barley seeds, Hordeum vulgare, var. Kenia, by a combination of buffer extractions and centrifugations over a sucrose gradient. Examination of the protein bodies pellet in the electron microscope shows 2 types of protein bodies in a wide variation of sizes. The majority of them stain evenly with osmium, are contained within a single membrane, and have no other structural components. The other type, mostly the larger particles, has a fine structure of orderly dark and light-stained layers attached to the protein bodies. Two acid hydrolases are associated with these particles: acid phosphatase activity, specific for sodium phytate but inactive on beta-glycerol phosphate, glucose 1-phosphate, fructose 1,6-diphosphate and adenosine triphosphate; and acid protease activity.

The capacity of leaf slices from light-grown seedlings of wild type barley and 10 xantha mutants at six different gene loci to incorporate acetate-(14)C into various lipids has been investigated. The fatty acid composition of the leaf lipids in these lethal mutants was similar to that of the wild type, but the fatty acid labeling pattern in the individual lipid classes can be drastically altered by these mutations, which affect chloroplast differentiation. A genetic block in chlorophyll synthesis, caused by mutations in the xan-f locus, leads to a repression of the formation of chloroplast membranes and of acetate incorporation into phospho-, sulfo-, and galacto-lipids (the acetate being preferentially channeled into a lipid fraction containing steroids and free fatty acids). Two leucine "auxotrophs" at different loci, which in the absence of leucine in the growth medium produce giant grana and accumulate some chlorophyll, differed considerably in the amount of labeling of their polar lipids during incubation. Leaves of xan-a(11), containing plastids with little chlorophyll, highly disorganized membrane systems, and large bodies with osmiophilic deposits, were nonetheless equal to wild type in their capacity to incorporate acetate-(14)C into phospho-, sulfo-, and galacto-lipids. The mutants at the xan-m locus have plastids with undispersed prolamellar bodies and osmiophilic packages of grana-like membranes associations. Leaf slices of these mutants synthesized considerably more linolenic acid-(14)C, which was incorporated into monogalactosyl diglycerides, than did slices of the wild type. This led to a labeling pattern of the fatty acids in the monogalactolipids which was remarkably similar to their endogenous fatty acid composition.