Genetic Control O F Meiosis in Rice , Oryza Sativa L . 111 . Effect

The recombination frequency as influenced by five independent recessive ds genes was measured on three segments of different chromosomes of rice, Oryza sativa L. Each ds gene in the homozygous condition resulted in an almost equally reduced recombination frequency in the three segments. When the mean reduction in recombination frequency was related to the reduction of chiasma frequency, the five ds genes were divided into two types: in one type the reduction of chiasma frequency almost corresponded to the mean reduction of recombination frequency, and in the other the chiasma frequency was greatly reduced in comparison with the mean reduction of recombination frequency. Three of the five ds genes were found to belong to the former group. In both types, normal synaptonemal complexes were observed in pachytene cells homozygous for ds genes. This finding suggests that ds genes do not affect the formation of synaptonemal complexes which are regarded as the prerequisite structure for crossing over. large number of desynaptic mutants have been reported in various plants A in which various numbers of univalents occur at metaphase I following normal pairing of homologous chromosomes at pachytene (BAKER et al. 1976; GOLUBOVSKAYA 1979; GOTTSCHALK and KAUL 1980). The precocious dissociation of bivalents could be due to the lack or precocious dissolution of chiasmata between homologous chromosomes. Reduction of chiasma frequency has been recognized in several desynaptic mutants (ENNS and LARTER 1960; THOMAS and RAJHATHY 1966; KLEIN 1969; SINGH et al. 1977). Two kinds of defects conducive to a reduction of chiasma frequency at metaphase 1 are considered. One is concerned with the formation of chiasmata in which two types of defects can occur: first, the chiasma frequency is reduced but chiasmata are distributed as in wild type, and, second, the distribution of chiasmata is altered relative to the wild type (BAKER et al. 1976). Since chiasmata are thought to be the cytological equivalent of genetic crossing overs, it is assumed that the frequency of crossing over would either be equally reduced in all sites of exchanges or be altered differentially among the chromosomal sites of exchanges. ' Present address: Laboratory of Molecular Genetics, Mitsuhishi-Kasei Institute of Life Sciences, Minamiooya, Machida, Tokyo 194, Japan. Genetics 108: 697-706 November, 1984. 698 K. KITADA AND T. OMURA The second possible defect that would result in a reduction of chiasma frequency is the failure of chiasmata to maintain themselves once initiated. In this case, the frequency of crossing over is expected to be normal at all sites of exchanges since the reduction of chiasma frequency is not due to a defect in crossing over. In only a very few desynaptic mutants has genetic recombination been examined. A reduction of recombination frequency has been reported in a desynaptic mutant of barley (ENNS and LARTER 1962), whereas an increase of recombination frequency has been reported for three ds genes in tomato (MOENS 1969). It was previously reported that several desynaptic mutants in rice induced by N-methyl-N-nitrosourea were controlled by different recessive genes and that the frequency of univalent per cell was parallel to the reduction of chiasma frequency per bivalent as well as per cell (KITADA and OMURA 1983). In the present paper the effect of different ds genes on the frequency of genetic recombination is examined and its effect is discussed in relation to the reduction of chiasma frequency. MATERIALS AND METHODS Five ds genes, dsa (MM-9), dssr (MM-11), dsg, (MM-13), dslol (MM-14) and dsll, (MM-18), were examined. All were induced by the treatment of fertilized egg cells of rice variety "Kinmaze" with N-methyl-N-nitrosourea. The original variety Kinmaze was used as the wild type. It was previously reported that the mean chiasma frequencies per cell were 17.38, 13.48, 11.68, 14.86, 16.08 and 18.91 for dsel, ds8,, dsgr, dslor, d s , , , mutants and the wild type, respectively (KITADA and OMURA 1983). Cytological obseruations: Meiosis in pollen mother cells (PMCs) and embryo sac mother cells (EMCs) usually occurred synchronously within the same spikelet. When the meiotic stage in EMCs was metaphase 1, that in PMCs was from late pachytene to diakinesis. For observation of both EMCs and PMCs at metaphase I using paraffin sectioning, ovaries and anthers were fixed for 24 hr in 3:l ethanol-acetic acid solution. The tissues were embedded in paraffin, sectioned at a thickness of 8-10 pm and stained with gentian violet. For observation of pachytene chromosomes, young panicles with the leaf sheath intact were cut off and kept for 4 hr in water at room temperature. Anthers teased from the spikelets were squashed with a needle in drops of 3:l methanol-acetic acid solution and flame dried. The prep arations were stained with Giemsa solution. For the electron microscopic observations of synaptonemal complexes, meiotic stages of anthers were determined by aceto-carmine squashes, and adjacent anthers were used. Initial fixation was for 3 hr with 3% glutaraldehyde in 0.1 M phosphate buffer at pH 7.0. After three buffer washes, anthers were postfixed for 2 hr with 1% Os04 in 0.1 M phosphate buffer, dehydrated through an alcohol series and embedded in Epon resin. Ultrathin sections were stained with uranyl acetate and lead citrate. Determination of genetic recombination: Recombination frequencies were measured for wild-type and homozygous ds genes at three interchromosomal segments, viz., the ch:, (ch1orina)-dl (drooping leaf) region of chromosome 5, the wx (waxy endosperm)-vg (virescent seedling) region of chromosome 6 and the la (lazy growth habit)-v4 region of chromosome 9. Desynaptic plants doubly heterozygous for marker genes were selected in FZ plants of the cross (dsJds., ++/++) X (ds:/ds:, chzdl/ch:,dl), and the segregation of marker genes was measured in the seedling of their progenies. On the assumption that the effect of ds genes on genetic recombination was the same in both sexes, we calculated recombination frequency from the segregating populations on the basis of the maximum likelihood method. Similarly, the recombination of wild type was measured from the cross Kinmaze: (ds+/ds+, ++/++) X (ds+/ds+, ch:,dl/chpdl). EFFECT OF DS GENES ON RECOMBINATION 699 FIGURE I.-Photomicrographs of chromosomes at pachytene in PMCs. a. Normal plant; b, desynaptic mutant. Magnification bars represent 15 pm.