Search

Mutations in ABCA1 uniformly decrease plasma HDL-cholesterol (HDL-C) and reduce cholesterol efflux, yet different mutations in ABCA1 result in different phenotypic effects in heterozygotes. For example, truncation mutations result in significantly lower HDL-C and apoliprotein A-I (apoA-I) levels in heterozygotes compared with nontruncation mutations, suggesting that truncation mutations may negatively affect the wild-type allele. To specifically test this hypothesis, we examined ABCA1 protein expression in response to 9-cis-retinoic acid (9-cis-RA) and 22-R-hydroxycholesterol (22-R-OH-Chol) in a collection of human fibroblasts representing eight different mutations and observed that truncation mutations blunted the response to oxysterol stimulation and dominantly suppressed induction of the remaining full-length allele to 5–10% of wild-type levels. mRNA levels between truncation and nontruncation mutations were comparable, suggesting that ABCA1 expression was suppressed at the protein level. Dominant negative activity of truncated ABCA1 was recapitulated in an in vitro model using transfected Cos-7 cells. Our results suggest that the severe reduction of HDL-C in patients with truncation mutations may be at least partly explained by dominant negative suppression of expression and activity of the remaining full-length ABCA1 allele. These data suggest that ABCA1 requires a physical association with itself or other molecules for normal function and has important pharmacogenetic implications for individuals with truncation mutations.

The segregating unit of mtDNA is a protein–DNA complex called the nucleoid. In an effort to understand how nucleoid proteins contribute to mtDNA organization and inheritance, we have developed an in organello formaldehyde crosslinking procedure to identify proteins associated with mtDNA. Using highly purified mitochondria, we observed a time-dependent crosslinking of protein to mtDNA as determined by sedimentation through isopycnic cesium chloride gradients. We detected ≈20 proteins crosslinked to mtDNA and identified 11, mostly by mass spectrometry. Among them is Abf2p, an abundant, high-mobility group protein that is known to function in nucleoid morphology, and in mtDNA transactions. In addition to several other proteins with known DNA binding properties or that function in mtDNA maintenance, we identified other mtDNA-associated proteins that were not anticipated, such as the molecular chaperone Hsp60p and a Krebs cycle protein, Kgd2p. Genetic experiments indicate that hsp60-ts mutants have a petite-inducing phenotype at the permissive temperature and that a kgd2Δ mutation increases the petite-inducing phenotype of an abf2Δ mutation. Crosslinking and DNA gel shift experiments show that Hsp60p binds to single-stranded DNA with high specificity for the template strand of a putative origin of mtDNA replication. These data identify bifunctional proteins that participate in the stability of ρ + mtDNA.

Intermolecular recombination of Chlamydomonas chloroplast genes has been analyzed in sexual crosses and following biolistic transformation. The pattern and position of specific exchange events within 15 kb of the 22-kb inverted repeat have been mapped with respect to known restriction fragment length polymorphism markers that distinguish the chloroplast genomes of the interfertile species Chlamydomonas reinhardtii and Chlamydomonas smithii. Recombinant progeny were selected from two- and three-factor crosses involving point mutations conferring herbicide (dr) and antibiotic resistance (er and spr) in the psbA, 23S and 16S ribosomal RNA genes, respectively. Exchange events were not randomly distributed over the 15-kb region, but were found to occur preferentially in a 0.7-kb sequence spanning the 3' end of the psbA gene and were much less common in an adjacent region of ca. 2.0 kb. These findings are corroborated by data showing that the dr mutation is unlinked genetically (3% recombination/kb) to the er and spr rRNA mutations, which are themselves linked and show ca. 1% recombination/kb. This discrepancy is significant since the dr-er and er-spr intervals are about the same length (ca. 7 kb). During chloroplast transformation, the 0.7-kb recombination hotspot also functions as a preferential site for exchange events leading to the integration of donor psbA gene sequences. The 0.7-kb hotspot region contains four classes of 18-37-bp direct repeats also found in other intergenic regions, but no open reading frame. Using deletion constructs in a chloroplast transformation assay, the hotspot was localized to a 500-bp region that lacks most of these repeats, which suggests that the repeats themselves are not responsible for the increased recombination frequency. Within this region, a 400-bp sequence is highly conserved between the chloroplast genomes of C. reinhardtii and C. smithii and includes several structural motifs characteristic of recombination hotspots in other systems.

We have developed an efficient procedure for the disruption of Chlamydomonas chloroplast genes. Wild-type C. reinhardtii cells were bombarded with microprojectiles coated with a mixture of two plasmids, one encoding selectable, antibiotic-resistance mutations in the 16S ribosomal RNA gene and the other containing either the atpB or rbcL photosynthetic gene inactivated by an insertion of 0.48 kb of yeast DNA in the coding sequence. Antibiotic-resistant transformants were selected under conditions permissive for growth of non-photosynthetic mutants. Approximately half of these transformants were initially heteroplasmic for copies of the disrupted atpB or rbcL genes integrated into the recipient chloroplast genome but still retained photosynthetic competence. A small fraction of the transformants (1.1% for atpB; 4.3% for rbcL) were nonphotosynthetic and homoplasmic for the disrupted gene at the time they were isolated. Single cell cloning of the initially heteroplasmic transformants also yielded nonphotosynthetic segregants that were homoplasmic for the disrupted gene. Polypeptide products of the disrupted atpB and rbcL genes could not be detected using immunoblotting techniques. We believe that any nonessential Chlamydomonas chloroplast gene, such as those involved in photosynthesis, should be amenable to gene disruption by cotransformation. The method should prove useful for the introduction of site-specific mutations into chloroplast genes and flanking regulatory sequences with a view to elucidating their function.

Summary Although the chloroplast genome of Chlamydomonas reinhardtii contains essentially the same complement of genes as chloroplast genomes of land plants, the arrangement of these genes within the circular DNA molecule is very different. Studies in several laboratories suggest that comparison of gene order, restriction fragment length polymorphisms, and sequences within chloroplast DNA may be valuable tools in phylogenetic analysis of Chlamydomonas species. We have found that six interfertile strains in the C. reinhardtii group differ by restriction fragment length polymorphisms throughout their chloroplast genomes. In two of these strains the chloroplast genomes are known to be colinear and the differences largely attributable to variations in the size and distribution of clusters of short dispersed repeat sequences occurring in intergenic regions. These repeats, which are marked by Aat II and Kpn I restriction sites, are present in the chloroplast genomes of all six interfertile strains of the C. reinhardtii group, but are absent from the chloroplast genomes of C. eugametos, C. moewusii, C. incerta and C. globosa. The latter two species, which are sensitive to the same vegetative cell autolysin as C. reinhardtii, do not appear to be interfertile with C. reinhardtii. Restriction fragment analysis of chloroplast DNA suggests that C. incerta and C. globosa are very similar to one another, but differ greatly from C. reinhardtii.

Studies demonstrating differences in chloroplast structure and biochemistry have been used to formulate hypotheses concerning the origin of algal plastids. Genetic and biochemical experiments indicate that significant variation occurs in ribulose-1,5-bisphosphate carboxylase (Rubisco) when supertaxa of eukaryotic algae are compared. These differences include variations in the organelle location of the genes and their arrangement, mechanism of Rubisco synthesis, polypeptide immunological reactivity and sequence, as well as efficacy of substrate (ribulose bisphosphate and CO2) binding and inhibitor (6-phosphogluconate) action. The structure-function relationships observed among chromophytic, rhodophytic, chlorophytic and prokaryotic Rubisco demonstrate that: (a) similarities among chromophytic and rhodophytic Rubisco exist in substrate/inhibitor binding and polypeptide sequence, (b) characteristic differences in enzyme kinetics and subunit polypeptide structure occur among chlorophytes, prokaryotes and chromophytes/rhodophytes, and (c) there is structural variability among chlorophytic plant small subunit polypeptides, in contrast to the conservation of this polypeptide in chromophytes and rhodophytes. Taxa-specific differences among algal Rubisco enzymes most likely reflect the evolutionary history of the plastid, the functional requirements of each polypeptide, and the consequences of encoding the large and small subunit genes in the same or different organelles.

Previous studies have established that the mitochondrial high mobility group (HMG) protein, Abf2p, of Saccharomyces cerevisiae influences the stability of wild-type (ρ+) mitochondrial DNA (mtDNA) and plays an important role in mtDNA organization. Here we report new functions for Abf2p in mtDNA transactions. We find that in homozygous Δabf2 crosses, the pattern of sorting of mtDNA and mitochondrial matrix protein is altered, and mtDNA recombination is suppressed relative to homozygous ABF2 crosses. Although Abf 2p is known to be required for the maintenance of mtDNA in ρ+ cells growing on rich dextrose medium, we find that it is not required for the maintenance of mtDNA in ρ− cells grown on the same medium. The content of both ρ+ and ρ− mtDNAs is increased in cells by 50–150% by moderate (two- to threefold) increases in the ABF2 copy number, suggesting that Abf2p plays a role in mtDNA copy control. Overproduction of Abf 2p by ≥10-fold from an ABF2 gene placed under control of the GAL1 promoter, however, leads to a rapid loss of ρ+ mtDNA and a quantitative conversion of ρ+ cells to petites within two to four generations after a shift of the culture from glucose to galactose medium. Overexpression of Abf2p in ρ− cells also leads to a loss of mtDNA, but at a slower rate than was observed for ρ+ cells. The mtDNA instability phenotype is related to the DNA-binding properties of Abf 2p because a mutant Abf 2p that contains mutations in residues of both HMG box domains known to affect DNA binding in vitro, and that binds poorly to mtDNA in vivo, complements Δabf2 cells only weakly and greatly lessens the effect of overproduction on mtDNA instability. In vivo binding was assessed by colocalization to mtDNA of fusions between mutant or wild-type Abf 2p and green fluorescent protein. These findings are discussed in the context of a model relating mtDNA copy number control and stability to mtDNA recombination.

DNA-protein complexes (nucleoids) are believed to be the segregating unit of mitochondrial DNA (mtDNA) in Saccharomyces cerevisiae. A mitochondrial HMG box protein, Abf2p, is needed for maintenance of mtDNA in cells grown on rich dextrose medium, but is dispensible in glycerol grown cells. As visualized by 4',6'-diamino-2-phenylindole staining, mtDNA nucleoids in mutant cells lacking Abf2p ( delta abf2) are diffuse compared with those in wild-type cells. We have isolated mtDNA nucleoids and characterized two mtDNA-protein complexes, termed NCLDp-2 and NCLDs-2, containing distinct but overlapping sets of polypeptides. This protocol yields similar nucleoid complexes from the delta abf2 mutant, although several proteins appear lacking from NCLDs-2. Segments of mtDNA detected with probes to COXII, VAR1 and ori5 sequences are equally sensitive to DNase I digestion in NCLDs-2 and NCLDp-2 from wild-type cells and from the delta abf2 mutant. However, COXII and VAR1 sequences are 4-to 5-fold more sensitive to DNase I digestion of mtDNA in toluene-permeabilized mitochondria from the delta abf2 mutant than from wild-type cells, but no difference in DNase I sensitivity was detected with the ori5 probe. These results provide a first indication that Abf2p influences differential organization of mtDNA sequences.

Of the more than 400 species of unicellular green algae in the genus Chlamydomonas (Ettl, 1976), only a few have been used for genetic investigations. The small cell size, rapid division time, and short sexual cycles of these green flagellates make them especially favorable organisms for genetic and molecular studies. The haploid heterothallic species C. reinhardtii has been the organism of choice for most molecular, cellular and developmental investigations (see Harris, 1989). Its nuclear genome of ca. 1-2 × 105 kb comprises 18 genetic linkage groups (Harris, 1989,1990), and physical maps are being developed for many of these (Ranum et al, 1988). Detailed molecular maps have also been established for the 196 kb chloroplast and 15.8 kb mitochondrial genomes of C. reinhardtii, and formal genetic studies are possible for both organelles (see Harris, 1989,1990). A fourth genome of 6–9 mb, localized in the basal body has recently been defined (Hall et al., 1989) and correlated with a unique quasi-Mendelian circular linkage group (Ramanis and Luck, 1986). Existence of closely related, interfertile isolates such as C. smithii mt + (CC-1373 = SAG 54.72 and UTEX 1062) and the Minnesota S-1 D-2 isolates (CC-2290 and CC-1952) that differ from C. reinhardtii by numerous restriction fragment length polymorphisms (RFLPs) in two or more genomes (Boynton et al, 1987; Gross et al., 1988; Hall et al., 1989; Ranum et al., 1988) makes this species group especially useful for molecular genetic studies.

The green alga Chlamydomonas reinhardtii is an ideal model organism in which to investigate the cooperation of nuclear and chloroplast genomes in chloroplast ribosome formation from both developmental and evolutionary perspectives. Mutations affecting a variety of ribosomal components have been isolated and mapped genetically (see Boynton et al., 1990a and Harris, 1989 for reviews), and both the chloroplast (Boynton et al., 1988, 1990b; Blowers et al., 1989) and nuclear genomes (Debuchy et al., 1989; Kindle et al., 1989; Mayfield and Kindle, 1990; Diener et al., 1990) can be manipulated using newly developed transformation methods. In this paper we review our recent work on 1) the genetics and molecular biology of mutations to antibiotic resistance affecting the rRNA genes; 2) identification of chloroplast and nuclear genes encoding chloroplast ribosomal proteins (r-proteins); 3) expression of chloroplast genes encoding r-proteins; and 4) the transformation of chloroplast rRNA genes. We also outline the direction our current research is taking to define more precisely the rRNA-r-protein interactions involved in chloroplast ribosome biogenesis and function, as well as the molecular mechanisms involved in regulating expression of chloroplast r-protein genes.

Meiotic progeny of Chlamydomonas reinhardtii normally receive chloroplast genomes only from the mt + parent. However, exceptional zygotes, which transmit the chloroplast genomes of both parents or, more rarely, only those of the mt - parent, arise at a low frequency. Mutations at the mt +-linked mat-3 locus were found previously to elevate the transmission of chloroplast genomes from the mt-parent, resulting in a much higher than normal frequency of exceptional zygotes. In this paper we demonstrate that an ultraviolet-sensitive nuclear mutation mapping at the uvsE1 locus, which is unlinked to mating type, also promotes chloroplast genome transmission from the mt - parent. This mutant, which was previously shown to reduce recombination of nuclear genes in meiosis, acts synergistically which the mat3-3 mutation to produce an extremely high frequency of exceptional zygotes. Through the use of restriction fragment length polymorphisms existing in the chloroplast genomes of C. reinhardtii and the interfertile strain C. smithii, we show that chloroplast DNA fragments from the mt - parent normally begin to disappear shortly after zygote formation. However, this process appears to be blocked totally in the absence of wild-type uvsE1 and mat-3 gene products. Our findings are consistent with the hypothesis that both gene products contribute to the mechanism responsible for uniparental inheritance of the chloroplast genome from the mt + parent.

Transformation of chloroplast ribosomal RNA (rRNA) genes in Chlamydomonas has been achieved by the biolistic process using cloned chloroplast DNA fragments carrying mutations that confer antibiotic resistance. The sites of exchange employed during the integration of the donor DNA into the recipient genome have been localized using a combination of antibiotic resistance mutations in the 16S and 23S rRNA genes and restriction fragment length polymorphisms that flank these genes. Complete or nearly complete replacement of a region of the chloroplast genome in the recipient cell by the corresponding sequence from the donor plasmid was the most common integration event. Exchange events between the homologous donor and recipient sequences occurred preferentially near the vector:insert junctions. Insertion of the donor rRNA genes and flanking sequences into one inverted repeat of the recipient genome was followed by intramolecular copy correction so that both copies of the inverted repeat acquired identical sequences. Increased frequencies of rRNA gene transformants were achieved by reducing the copy number of the chloroplast genome in the recipient cells and by decreasing the heterology between donor and recipient DNA sequences flanking the selectable markers. In addition to producing bona fide chloroplast rRNA transformants, the biolistic process induced mutants resistant to low levels of streptomycin, typical of nuclear mutations in Chlamydomonas.

In photosynthetic eukaryotes, genes encoding chloroplast proteins of multimeric complexes involved in photosynthesis and protein synthesis are partitioned between the cell’s nuclear and chloroplast genomes. The chloroplast genome’s relatively small size, high copy number and great evolutionary conservation have led to rapid progress in identifying and characterizing chloroplast genes. In fact, the entire chloroplast genomes of tobacco, liverwort and rice have been sequenced within the last three years (1–3). Somewhat less progress has been made in identifying the nuclear genes for the remainder of the chloroplast polypeptides, due to the larger number of genes and the much larger size and evolutionary diversity of the nuclear genomes themselves. However in many plant species, nuclear genes can be engineered in vitro, the constructs introduced by transformation, and their function studied in vivo.

The ribulose-1,5-bisphosphate carboxylase (Rubisco) large- and small-subunit genes are encoded on the chloroplast genome of the eukaryotic chromophytic alga Olisthodiscus luteus. Northern blot experiments indicate that both genes are co-transcribed into a single (>6 kb) mRNA molecule. Clones from the O. luteus rbc gene region were constructed with deleted 5′ non-coding regions and placed under control of the lac promoter, resulting in the expression of high levels of O. luteus Rubisco large and small subunits in Escherichia coli. Sucrose gradient centrifugation of soluble extracts fractionated a minute amount of carboxylase activity that cosedimented with native hexadecameric O. luteus Rubisco. Most of the large subunit synthesized in E. coli appeared insoluble or formed an aggregate with the small subunit possessing an altered charge: mass ratio compared to the native holoenzyme. The presence in O. luteus of a polypeptide that has an identical molecular mass and cross reacts with antiserum generated against pea large-subunit binding protein may indicate that a protein of similar function is required for Rubisco assembly in O. luteus.

Ribulose-1,5-bisphosphate carboxylase (RuBPCase) was purified from the marine chromophyte Olisthodiscus luteus. This study represents the first extensive analysis of RuBPCase from a chromophytic plant species as well as from an organism where both subunits of the enzyme are encoded on the chloroplast genome. The size of the purified holoenzyme (17.9 Svedberg units, 588 kilodaltons) was determined by sedimentation analysis and the size of the subunits (55 kilodaltons, 15 kilodaltons) ascertained by analytical sodium dodecyl sulfate gel electrophoresis. This data predicts either an 8:9 or 8:8 ratio of the large to small subunits in the holoenzyme. Amino acid analyses demonstrate that the O. luteus RuBPCase large subunit is highly conserved and the small subunit much less so when compared with the chlorophytic plant peptides. The catalytic optima of pH and Mg(2+) have been determined as well as the response of enzyme catalysis to temperature. The requirements of NaHCO(3) and Mg(2+) for enzyme activation have also been analyzed. The Michaelis constants for the substrates of the carboxylation reaction (CO(2) and ribulose bisphosphate) were shown to be 45 and 48 micromolar, respectively. Competitive inhibition by oxygen of RuBPCase-catalyzed CO(2) fixation was also demonstrated. These data demonstrate that a high degree of RuBPCase conservation occurs among widely divergent photoautotrophs regardless of small subunit coding site.

Ribulose-1,5-bisphosphate carboxylase (Rubisco) catalyzes the initial CO2-fixing reaction of photosynthesis in prokaryotes and eukaryotes. The active holoenzyme consists of 8 large subunits (LS) and 8 small subunits (SS), except in R. rubrum which contains only two large subunits (1). In chlorophytic plants (chlorophyll a,b), rbcL (large subunit gene) is chloroplast DNA encoded where as the small subunit genes (rbcS) are nuclear encoded and the small subunit gene products are post-translationally transported into the plastid. This organization of rbcL and rbcS is not representative of all plants. In the chromophytic (chl.a,c) alga O1isthodiscus luteus, both subunits are coded for by chloroplast DNA (ctDNA) (2). The proximity and arrangement (5′-LS-SS-3′) of these genes are identical to that seen in blue-green algae and in Cyanophora Chapaudoxa. Since O. luteus is a eukaryotic alga with a prokaryotic organization of rbcL and rbcS, the enzymology and mechanism of biosynthesis of the O. luteus Rubisco are expected to be unique when compared to the enzyme from other photoautotrophs. In this study, the biochemical analysis of Rubisco has been extended to include the enzyme from the rhodophytic (red) alga Griffithsia pacifica, in addition to using antibodies and proteases to determine the structural relatedness of this enzyme and its subunits among chromophytic, rhodophytic and chlorophytic plants.

Ribulose-1,5-bisphosphate carboxylase (Rubisco) from the algae Olisthodiscus luteus (chromophyte) and Griffithsia pacifica (rhodophyte) are remarkably similar to each other. However, both enzymes differ significantly in the structure and function when compared to Rubisco from green algae and land plants. Analysis of purified Rubisco from O. luteus and G. pacifica indicates that the size of the holoenzyme and stoichiometry of the 55 and 15 kilodalton subunit polypeptides are approximately 550 kilodaltons and eight:eight for both algae. Antigenic determinants are highly conserved between the O. luteus and G. pacifica enzymes and differ from those of the spinach subunit polypeptides. Sequence similarity between the two algal large subunits has been further confirmed by one-dimensional peptide mapping. Substrate ribulose bisphosphate has no effect on the rate of CO(2)/Mg(2+) activation of O. luteus and G. pacifica enzymes which contrasts to the extensive inhibition of spinach Rubisco activation at similar concentrations of this compound. In addition, the Michaelis constant for CO(2) and the inhibition constant for 6-phosphogluconate are similar for the O. luteus and G. pacifica catalyzed carboxylation reaction. Both values are intermediate to those observed for the tight binding spinach enzyme and weak binding prokaryotic (Rhodospirillum rubrum) enzyme. The biochemical similarities documented between O. luteus and G. pacifica may be due to a common evolutionary origin on the chromophytic and rhodophytic chloroplast but could also result from the fact that both subunit polypeptides are chloroplast DNA encoded in these algal taxa.