Institut Francois de Recherche Scientifique pour le Developpement en Cooperation (ORSTOM), Centre de Montpellier, BP 5045, 34032 Montpellier Cedex, France. Present address: 1) Department of Plant Breeding and Biometry, 252 Emerson Hall, Cornell University, Ithaca, N.Y. 14853-1902, USA. 2) Institut fur Entwicklungs- und Molekularbiologie der Pflanzen, Universitdtsstr 1, D-4000 Dilsseldorf, FRG. 3) Department of Biology, Zhongshan University, Guangzhou 510275, China
As earlier reported (Dally 1988; Dally and Second 1988), some cytoplasniic male sterile lines carrying the so-called' 'wild abortive" (WA) cytoplasm are heterogeneous for their chloroplast DNA restriction patterns when digested by the enzymes Ava1 (recognition sequences 5' CpyCGpuG) and Sma1 (recognition sequence 5' CCCGGG, also recognized by Ava1). Most of the lines studied had the plastotype a2 which is frequent in O. rufipogon originating in South and South-East Asia. Other lines had the plastotype a1 which is most frequent in the indica type of O. sativa, including the "maintainer" cultivars used to propagate the male sterile lines. One plant (accession IRRI 103919, line V41A, plant No. 84 in Dally 1988) was found to have a mixed a1+a2 plastotype. The two patterns were thought to differ by a single restriction site gain recognized by both Ava1 and Sma1 enzymes.
The few wide-cross hybrids we studied in rice showed maternal inheritance of plastotype. The observation of the above presumed "cybrid" plant suggested, however, occasional biparental inheritance of chloroplast DNA in rice. We now report a few more observations on this line:
1) Seven other plants issued of the same lot of seeds as plant No. 84 were analyzed by S. Zhang, according to the method previously used (Dally and Second 1989), for the Ava1 restriction pattern. Three plants had a mixed restriction pattern showing different proportions of the bands No. 37 and No. 41 (in Dally 1988) which differentiate the patterns a1 and a2: approximatively 20% and 80%, 50% and 50%, 80% and 20% respectively for the three plants. The four other plants had the plastotype a1. It may be concluded that the observed mixed patterns do not correspond to accidental mixtures but are a characteristic of the line V41A.
2) The restriction site responsible for the differences between the patterns a1 and a2 was mapped in the following way. Based on the total sequence of rice chloroplast DNA (Hiratsuka et al. 1989), restriction maps were drawn. The sequence most probably corresponds to the plastotype el, because a japonica type cultivar was used by Hiratsuka et al. (1989) to extract the DNA. However, the plastotypes a1 and e1 appear identical in terms of electrophoretic patterns for the restriction fragments presently considered. The fragments observed by Dally (1988) in pattern a1 could thus unambiguously be positioned on the restriction site gain clearly mapped in the inverted repeats, near the junction with the short single copy region, as shown on Fig. 1. There are thus actually two sites, one on each inverted repeat, included respectively in the Pst1-5 and Pst1-9 cloned fragments (Hirai et al. 1985). In al pattern, the Sma1 generated fragment 2 (20.9 kbp) is replaced by three fragments (13.3 and 2 X 3.8 kbp) in two bands. In the Ava1 restriction patterns, the two sites are in two different fragments. One is located near one end of fragment 1. Its effect is not detected by the technique used by Dally (1988). The other cuts the generated fragment 16 (2.2 kbp) in two, with the approximate molecular sizes 1.8 + 0.4 kbp. The size of the 0.4 kbp fragment generated by Ava1 was used to position the mutation as accurately as possible in Fig. 1.
Fig. 1. Restriction maps of the short single copy region (between, around 101
and 114 kbp), with the adjacent portions of the inverted repeat regions of
the rice chloroplast DNA, for the enzymes Ava1 and Sma1 and the plastotype
a1, as deduced from the total sequence (Hiratsuka et al. 1989). The
approximate deduced positions of the site mutations distinguishing the
plastotypes al and a2 are shown by the arrows.
A molecular size scale is shown on the upper line in kbp. The numbers on the fragments indicate, in the order of decreasing molecular weights, the number of the bands to which they belong in the restriction patterns (plastotype el in Dally 1988). The upper bold line shows the extent of the inverted repeat. The lower bold line shows the position of the cloned Pst1-5 and Pst1-9 fragments used in the Southern blot analysis.
3) Following gel electrophoresis, the DNA of which the restriction patterns were analyzed in 1988 with the enzyme Sma1 had been saved on filter after Southern blotting. lt was hybridized with the chloroplast DNA clones Pst1-5 and Pst1-9 (Hirai et al. 1985). As expected, both hybridizations gave the same pattern of bands which correspond to the map in Fig. 1. The result of the hybridization with the fragment Pst1-5 is shown on Fig. 2. It appears clearly that the extra bands seen by Dally (1988) in plant No. 84 (second lane from left) correspond to a mixed pattern a1 and a2 and not to DNA of different nature.
As many of the WA lines studied has the a2 plastotype, but the maintainer lines used to pollinate them have the a1 plastotype, it seems likely that there had been a biparental transmission of chloroplasts. Because the chloroplasts multiply clonally, the "cybrid" state may be expected to disappear rapidly due to drift in the course of generations. That explains why we found most lines are fixed for the plastotype a1 or a2 in the WA group. Because several lines of the WA cytoplasm were found to have the a1 plastotype, it is suggested that the event of
Fig 2. Southern blot analysis of chloroplast DNA with the cloned probe Pst1-5
(Hirai 1985). Each sample (1 ug chloroplast DNA) was digested with Sma1
enzyme, fractionated on a 0.9% agarose gel in separate lanes, observed for
its restriction pattern under fluorescence in ethidium bromide, transferred
to a filter and then hybridized to a 32P labelled cloned Pst1-5 fragment.
The restriction patterns (Dally 1988) as read on the gel under fluorescence
are indicated. The bands shown here were also seen in fluorescence, however,
hybridization shows their homology to the cloned fragment. The second lane
from left is plant No. 84 in the text. The molecular size were measured on
the gel from size markers and further precised according to the map in Fig.
1.
biparental transmission occurred more than once. There may be a correlation between this phenomenon and the instability of some of the male sterile lines (sometimes with an heterogeneity within a given plant) but that remains to be investigated.
These results strengthen our earlier statement that biparental inheritance of chloroplast DNA occurs occasionally in rice. Hiratsuka et al. (1989) showed that intermolecular recombination of chloroplast DNA occurred during the evolution of the cereals. There appears a possibility that intermolecular recombination of chloroplast DNA might have occurred in the various wild and cultivated species of rice which have been subjected to introgressive hybridization and allotetraploidization. Although we have some evidence on that line, it remains to be investigated further.
We deeply thank Dr S. S. Virmani for providing the seeds and Dr. M. Sugiura for providing the rice chloroplast DNA clones and a personal computer readable copy of the total rice chloroplast DNA sequence.
References
Dally, A. M., 1988. Analyse cladistique de mutations de l'ADN chloroplastique et phylogenie des riz (section Eu-Oryza du genre Oryza). Coll. Etudes et Theses. ORSTOM, Paris. 153p.
____ and G. Second, 1988. Analysis of total chloroplast DNA RFLPs in cultivated and wild species of rice. RGN 5: 77-81.
____ and _____, 1989. Chloroplast DNA isolation from higher plants: an improved nonaqueous method. Plant Mol. Biol. Rep. 7, 2: 135-143.
Hirai, A., H. Ichikawa, N. lwatsuki, M. Sugiura, 1985. Rice chloroplast DNA: a physical map and the location of the genes for the large subunit of ribulose 1, 5-bisphosphate carboxylase and the 32 kd photosystem II reaction center protein. Theor. Appl. Genet. 70: 117-122.
Hiratsuka, J., H. Shimada, S. Whittier, T. Ishibashi, M. Sakamoto, M. Mori, C. Kondo, Y. Honji, C. R. Sun, B. Y. Meng, Y. Q. Li, A. Kanno, Y. Nishizawa, A. Hirai, K. Shinozaki and M. Sugiura, 1989. The complete sequence of the rice (Oryza sativa) chloroplast genome: Intermolecular recombination between distinct tRNA genes accounts for a major plastid DNA inversion during the evolution of the cereals. Mol. Gen. Genet. 217: 185-194.