23. A nuclear gene inducing fertility restoration in cytoplasmic male-sterile rice

Yoshio SANO and Mitsugu EIGUCHI

National Institute of Genetics, Mishima, 411 Japan

In S type male-sterile cytoplasm of maize, spontaneous reversions to male fertility including cytoplasmic and nuclear changes occur not infrequently (Laughnan and Gabay-Laughnan 1983). The nuclear genotype governs the frequency of reversions as well as the relative frequency of cytoplasmic and nuclear reversions. Recent molecular studies revealed that cytoplasmic reversions are highly associated with reorganization of mitochondrial genomes. However, the genetic mechanism inducing spontaneous reversions has not been investigated. The present study gives evidence on the presence of a single nuclear gene which induces nuclear reversion in CMS rice.

The (ms-bo) cytoplasm of Chinsurah boro II exhibits male sterility when combined with the nucleus of Taichung 65 (T65R) as reported by Shinjyo (1975). Fertility restoration of (ms-bo) cytoplasm is gametophytically controlled by Rf1. Although male sterility is stably expressed in the genetic background of T65R, Taichung 65 (T65T) preserved in our laboratory showed about 8% seed fertility when pollinated to male-sterile plants (ms-bo)rf1rf1. Unexpectedly, a strange phenomenon was detected in later generations of the hybrid, namely, seed fertility gradually increased with regeated selfings and finally almost fully fertile plants were obtained in the F6 generation. In this experiment, seed fertility was examined by strictly bagging to eliminate cross pollination. The occurrence of fertile segregants might be simply explained by the assumption that T65T carries recessive restorers and their accumulation gives rise to fertile plants in later generations of the hybrid. The assumption was tested by establishing plants with (ms-bo) cytoplasm and a T65T nucleus by successive backcrossings. The assumption was ruled out since resultant plants still showed about 6% seed fertility. Accordingly, the fertile plants observed in F6 might have resulted from mutational events such as found in the S-type of maize cytoplasm.

The cytoplasm of the fertile revertant (FR37-4) was examined in relation to cytoplasmic reversion. FR37-4 was repeatedly backcrossed with T65R possessing rf1rf1. Results indicated that no cytoplasmic change was detected in FR37-4 since the cytoplasm was completely male-sterile in the T65R genetic background. Therefore, mutational events seem to have occurred in the nuclear genes. A gradual increase of fertility strongly suggested that multiple genes are involved in the fertile revertant. FR37-4 X T65T F2 gave more fertile plants than (ms-bo)rf1rf1 X T65T F2 supporting the assumption that FR37-4 carries mutant nuclear genes. Conventional genic analysis for them, however, seems to be impossible because of instability.

The segregation patterns observed in later generations of (ms-bo)rf1rf1 X T65T suggested that genes controlling partial fertility in the hybrid are associated with the induction of fertility reversion since F2 plants showing different degrees of fertility similarly exhibited a gradual increase in fertility with selfings. In other words, partially fertile plants seemed to produce fertility restoration with repeated selfings. Therefore, the genetic control for partial fertility was examined. When (ms-bo)rf1rf1 was pollinated with the pollen of T65RXT65T F1, completely sterile and partially fertile plants segregated into a 1:1 ratio, suggesting that partial fertility is controlled by a single dominant gene (Table 1). From the assumption, when T65R X T65T F2 individuals were crossed to (ms-bo)rf1rf1, three different types of F2 individuals which produce only sterile plants, only partially fertile ones or both would arise. Results supported the above assumption showing the expected ratio of 1:1:2 (Table 2). In addition, an allelism test revealed that the gene seemed to be independent of Rf1(Table 1). The gene detected was designated Ifr(t). At present, whether Ifr(t) gives rise to mutational events only in the presence of (ms-bo) cytoplasm is under investigation.

Table 1. Segregations for seed fertility in test crosses
Cross                  ====================       Ratio      X2
                        CS   PF   F   Total
MSrfx(T65RxT65T)        81   77        158         1:1       0.10ns
MSrfx(T65TxMSRf)        35   30  76    141         1:1:2     1.21ns
Note: CS, PF and F show completely sterile, partially ferile and fully fertile, respectively. MSrf and MSRf are (ms-bo)rf1rf1 and (ms-bo)Rf1Rf1.


Table 2. Segregations for seed fertility in F2 individuals of T65R X T65T as
determined by crosses with MSrf
      No. of F2 plants                 Ratio              X2
   CS  Seg.  PF   Total
    8   17    5      30                 1:2:1             1.40ns
Note: T65R-A line of Taichung 65 from University of Ryukyus,

T65T-Another line from Taichung Agric. Improvement Station.

We called them T65A and T65B, respectively (cf. Tsai 1986).


Laughnan, J. R. and Gabay-Laughnan S. 1983. Cytoplasmic male sterility in maize. Annu. Rev. Genet. 17: 27-48.

Shinjyo, C. 1975. Genetical studies of cytoplasmic male sterility and fertility restoration in rice, Oryza sativa L. Bull. Col. Agric., Univ. Ryukyus 22: 1-57.

Tsai, K. H., 1986. Possible genic differences between two Tichung 65 strains, one preserved at Taichung and the other from Ryukus. RGN 3: 75-76.