Plastid and Mitochondrial Genomes

Introduction

Three plant organelles contain functional, replicating, plant-derived DNA molecules: the nuclei, chloroplasts and mitochondria. We've spent the past six weeks on nuclear DNA, today we will cover the structure and the basics of function of plastid and mitochondrial genome organization.

Background

Plants contain three distinct classes of information containing organelles: nuclei, chloroplasts and mitochondria. Nuclei contain chromosomes with normal eukaryotic characteristics: centromeres, telomeres, nucleosomes replete with histones, synaptonemal complexes and meiosis, the whole nine yards. Chloroplast genomes are similar to bacterial genomes, but smaller and simpler. The chloroplast genome in any plant consists of a circular double stranded DNA molecule ranging in size from120kbp to 205kbp.

Each chloroplast contains from 2 to 200 genomes per plastid. Chloroplasts in young, expanding leaves tend to contain a higher copy number of genomes than those in older cells. In a few species, chloroplasts have been observed which lacked detectable genomes. If chloroplast replication does occur primarily in rapidly developing tissues, perhaps plastids may derive from fission which by accident lack genomes. Plastid genomes are mainly observed within the stroma in structures called nucleoids. Each nucleoid may contain several plastid genomes.

-The Structure

The chloroplast genome is a closed, simple circle. However, the ribosomal sequences and sometimes their adjacent regions are duplicated in inverted repeats, generating a structure which looks like a mace. The stem varies in length with the variation in the amount of sequence adjacent to ribosomal sequences which is also duplicated.

- The Numbers

An average plastid genome is about 150 kbp. If you cut it with a 6-cutter, you wind up with fewer than 40 fragments. This is resolvable on a high quality agarose gel. Consequently, if you have also cloned the plastid genome it is a relatively simple thing to generate a restriction site map of the chloroplast genome. Further, since plastids are relatively simple to isolate, cloning and sequencing the plastid genome is a relatively simple project. The first (liverwort, Marchantia polymorpha was sequenced by Ohyama et al. in 1986 (Nature 332:572-574). Using the ABI377, 16kbp of useful sequence data can be gathered per gel. Sequencing a plastid genome demands about ten gels, or about five days with good planning and execution.

-Inheritance

Plastids are generally maternally inherited. Not always. Our lab demonstrated that alfalfa chloroplasts are generally paternally inherited. Similar observations have been made in pine species, rye, Oenothera and several other species.

PlastidGenomes

Most of the gene products contained within a plastid are derived from the nuclear genome, but many important genes reside within the plastid genome and are transcribed and translated within the plastid. These include the large subunit of RuBisco, ribosomal RNAs, transfer RNAs, and about 100 structural protein genes.

The chloroplast genome of most higher plants contains a single large inverted repeat which contains the 16S and 23S ribosomal genes. These structures flank a small unique sequence and a large unique sequence. In a few legumes there are three copies of this inverted repeat structure There are a few primarily monomer repeats (e.g. TTTTTT) which may act over time as points at which intramolecular recombination may occur. Such an event occurred during the evolution of wheat which defines the cytoplasmic differences between T. aestivum, Ae. crassa and Ae. squarossa.(see Ogihara et al., PNAS 85:8573).

Plastids are predominantly maternally inherited. There are several exceptions, one of which we (Lee et al., 1985, 1986) characterized. Ted Bingham (U. Wisc) had identified alfalfa plants which exhibited the characteristics of plastid mutants, but their inheritance was not maternal. We identified restriction fragment length polymorphisms which differentiated the three different plastid phenotypes we observed, and followed these plastid markers through generations. We determined that these chloroplast mutations were predominantly paternally inherited. Similar observations have been made with rye and several pine species.

Steve Smith (Ted's postdoc) observed two alfalfa plants which exhibited chlorophyll deficiencies in sectors, one with yellow and the other with albino sectors of leaves, and occasionally entire branches would show the defective phenotype. Alfalfa is easy to propagate from cuttings, and therefore it's easy to grow plants from heavily sectored plant parts. The white-sectored portions of the albino sectored plants are obviously non-photosynthetic and survive parasitically. Both yellow-green and albino sectors can produce flowers and seed.
Female parent Male Parent normal progeny chlorophyll deficient progeny
Yellow green sector normal 77 58 (43%)
normal YGS sector 96 149 (61%)
Normal sector on YGS plant Normal plant 129 0
Normal Normal sector on YGS plant 158 0
Albino Normal 92 18 (16%)
Normal Albino 34 164 (83%)
Normal Normal 94 0
Normal normal sector on AS plant 123 0


We took these plants and their progeny and did the chloroplast marker analyses needed to show that plastid restriction fragment length polymorphisms tracked the differences between progeny, and that more than half the time the plastid genotype of a progeny was determined by the paternal plastid genotype. In albino sectored plants, it turned out to be more than 80% of the time.

Chloroplast genomes are small and relatively simple. While a great way for my lab to learn how to do electron microscopy and Southern blots, the project turned out to be little more than a single PhD project. On the positive side, Don Lee turned this into a professorship at U. Nebraska, where he's been tenured and lives still.

Interesting Stuff

Day and Ellis (1984) noted that several workers had developed anther culture doubled haploid programs in many species, including barley and wheat. They noted that in both barley and wheat plastids are known to be maternally inherited, and asked the question: how can you get green plants produced from microspores? It had long been noted that when doing microspore haploid plant production that many of the plants were albino.

Day and colleagues focussed their attention on the albinos produced in their doubled haploid development process and asked "what happed to the plastids?". They found that massive deletions had occurred.

Chloroplasts are obviously important. However, with the exception of plastid-encoded genes conferring resistance to herbicides, no 'positive' useful variation has been observed. Lots of plastid mutations have been identified, but are generally deleterious. However, plastid genomes accumulate variation relatively slowly, and have turned out to be excellent cladistics tools. Restriction fragment length polymorphisms have been the most widely utilized markers for cladistics analyses, and if you take Matt Lavin's evolution course you will undoubtedly utilize restriction sites for evaluation of relationships among plant genera.

Historical note: Jeff Palmer was the big guy in the field. His manuscript (Comparative Organization of Chloroplast Genomes, J.D. Palmer, Annual Review of Genetics 1985: 325-254) provides a fine review of much of his work.

Mitochondrial Genome Organization

Plant mitochondrial genomes are much larger and more interesting than are chloroplast genomes. They gained the nation's attention when Southern Corn Leaf Blight damaged the US corn crop in 1970. This disease was a 'breeder's disease', and the significant race (race 'T') of Helminthosporium (or Dreschleria) maydis was virulent only on plants which carried cytoplasm which conferred cytoplasmic male sterility. These cytoplasms also conferred sensitivity to the insecticide 'Methomyl.

Cytoplasmic male sterility (cms) was thought to be useful because in the presence of nuclear 'restorer genes' plants carrying cms were fully fertile. Four independent cytoplasms conferring cms were identified, the first (since lost) by Marcus Rhoades in 1931 (Rhoades MM. 1931. Cytoplasmic inheritance of male sterility in Zea mays. Science 73:340-341.) The most commonly utilized cms was the T cms which is defined by the genes which result in fertility restoration, Rf1 and Rf2. These are dominant, one copy of each being sufficient for full fertility restoration. This is a sporophytic sort of cms, a plant heterozygous for both dominant genes will produce all healthy pollen, even though only 1/4 of the pollen grains have the Rf1,Rf2 genotype.

CMS (especially t-cms) provided seed corn companies with two desirable characteristics. Plants which were cms, rf1rf1 rf2 rf2 were fully sterile and made excellent female parents for hybrid seed production. No detasseling was necessary which reduced labor costs. These plants, when pollinated with pollen from Rf1Rf1,Rf2Rf2 pollen wound up heterozygous at both loci and produced 100% functional pollen. However, 1/4 of their progeny would wind up completely male sterile. Cytoplasmic male sterility spread rapidly through the corn industry. In 1969 SCLB affected the Puerto Rico corn crop and in 1970 it hit the southern US. In 1971, seed corn companies went back to manual detasseling and normal cytoplasms.

Interesting phenomena

Revertants to fertility and resistance to SCLB occurred relatively frequently. These occurred in production fields, and in tissue culture labs. Berle Gegenbach ( U. Minn) took H. maydis toxin and incorporated it into media on which he grew cms-t callus. A few cells grew which he subcultured and eventually regenerated into plants. They were toxin tolerant, resistant to H. maydis race T, and fully fertile. Many labs were devoted to finding a resistant male sterile cytoplasm. None were found. None of the straightforward technological fixes seemed to work, so this problem became the premier research problem of applied plant science of the 1970s. It wasn't solved until 1986, and seedcorn companies still use detasselers.

Plant Mitochondrial Genomes

Plant mitochondrial genomes are (relative to chloroplast genomes) large and complex. They range in size from around 200kbp in Brassicas (Palmer and Shields 1984. Nature 307:437-440) to over 2500kbp. When evaluated by microscope, each mitochondrion contains circular DNA molecules of several sizes and often linear DNA molecules. Mapping the mitochondrial genomes was more challenging than would have been expected, because the various circular molecules derived via recombination from what should have been (but apparently rarely or never actually is) a single large circular genome. Internal recombination via homologous recombination between direct repeat sequences within the genome result in the production of variously sized minor circular molecules. The Brassica mitochondrial genome contains two of these, the maize mitochondrial genome contains many more.

These recombination events provide a fertile testing ground for recombinationally-generated mutations.

The T cytoplasm : Dewey's experiments

Ralph Dewey (Sam Leving's grad student and son of Doug Dewey, USDA wheat cytogeneticist) did the work that demonstrated how cms-T works. See Dewey. Levinga and Timothy. 1986. Cell 44:439-449.

I. a BamH1 library was constructed from isolated mitochondrial DNA.

II. End-labeled cms-T mitochondrial RNA was hybridized to these BamH1 fragments, which were also hybridized to end-labeled normal mitochondrial RNA. A 9kb fragment provided a far stronger signal with cms-T RNA than normal RNA. This 9kb fragment was subcloned using HindIII, and three relatively small HindIII fragments were the site of this differential hybridization. They were consecutively oriented, and comprised a 3,147 bp DNA fragment which was completely sequenced. The sequence was analyzed and two significant open reading frames were observed. When compared with current databases, these open reading frames shared homology with a tRNA, atp 6, and a portion of 26s RNA. When used as a probe against BamH1 digested normal and cmsT mitochondrial DNA, the normal genome lacked the 9kbband, and only a 6.5kb fragment was observed. The 9kb fragment appears to be the product of an unlucky fusion of gene fragments, resulting in production of a product which destabilizes mitochondria. Transcript analysis confirmed this view.

III. The final data: 1) transformation of E. coli with Turf-13 renders E. coli sensitive to both methomyl and to H. maydis toxin. Transfer of the Turf-13 gene product into tobacco mitochondria likewise renders them sensitive to methomyl. Turf-13 gene products look like the causal agent for cytoplasmic male sterility. Nice story.