Genetic and Cytological Investigation of Genic Male Sterility in Brassica rapa ssp. rapa ‘Kida-aokabu’ and ‘Tennhoji-kabu’

Abstract

A male sterile (ms) plant was identified in each of the two turnip (Brassica rapa) cultivars ‘Kida-aokabu’ and ‘Tennhoji-kabu,’ which are two of the main domestic varieties in Japan. Morphological analysis of organs constituting the flowers in ‘Kida-aokabu’ and ‘Tennhoji-kabu’ showed differences in the features of ms plants between both cultivars, in addition to the differences in size between ms and male fertile (mf) plants in each cultivar. The results of a cross between an ms plant and two cultivars of B. rapa indicated that male sterility is controlled by a single genic recessive gene without any cytoplasmic effect in both cultivars. Light microscopic observations showed that at an early microspore stage, pollen and tapetum development of the ms plant was different from that of the mf plant in ‘Kida-aokabu’ and ‘Tennhoji-kabu,’ respectively. In ms anthers of both cultivars, microspores and tapetum have degenerated, and then completely disappeared before or during the developmental stage to pollen grains. Although the collapse of microsporogenesis in both cultivars occurred at a similar stage, the modes of collapse differed. This is because the collapse of microspores and the tapetum in the locule did not occur synchronously in the ms anther during early microspore development in ‘Tennhoji-kabu,’ which was different from that in ‘Kida-aokabu.’ Our results suggest that the ms plant of ‘Kida-aokabu’ and ‘Tennhoji-kabu’ may represent a new type of Brassica genic male sterility.

The use of male sterile (ms) plants has become an important technique in heterosis breeding to simplify and reduce the cost of seed production. Heterosis is a common strategy in the breeding of Brassica crops, with great potential for increasing yield (Grant and Beverdorf 1985, Kaul 1988, Brandle and Mcvetty 1990). There are two types of male sterility in Brassica crops: genic male sterility (GMS) and cytoplasmic male sterility (CMS) (Van der Merr 1987). CMS is more widely used since ms plants can be obtained at a frequency of 100%, whereas utilization of a GMS system is limited for the production of hybrid seeds because removal of 50% of segregating fertile plants in the sterile female line is required to produce seeds when a trait controlled by a single recessive gene. However, GMS also has specific advantages for hybrid seed production, including complete and stable male sterility, no negative cytoplasmic effects on yield, and ease of transfer of the male sterile gene(s) to diverse genetic backgrounds (Perez-Prat and van Lookeren Campagne 2002, Ke et al. 2005, Wang et al. 2012, Zhou et al. 2017).

Brassica rapa is an allogamous species owing to its self-incompatibility, comprising some of the most important arable vegetable subspecies such as Chinese cabbage (ssp. pekinensis), turnip (ssp. rapa), Mizuna (ssp. japonica), Taina (ssp. chinensis), and Yukina (ssp. narinosa). Several GMS varieties have been reported in two subspecies of B. rapa, including dominant GMS (Van der Merr 1987), recessive GMS (Takahata et al. 1996, Feng et al. 2009, Liu et al. 2016), and multiple-allele GMS (Zhou et al. 2017) plants in ssp. pekinensis, and a recessive GMS (Cao and Li 1981, Ying et al. 2003, Xie et al. 2005) in ssp. chinensis. However, only one GMS in B. rapa spp. rapa plant, derived from a European turnip variety, has been reported to date (Wakui et al. 2013).

We discovered GMS lines of ssp. rapa from populations of ‘Kida-aokabu’ and ‘Tennhoji-kabu,’ domestic varieties originating from Hukui and Osaka prefecture in Japan, respectively. To better characterize the mechanism of GMS of this cultivar, we examined the mode of inheritance and pollen and tapetum development in these newly identified ms lines.

Materials and methods

The ms plant was identified in each population of B. rapa ssp. rapa, ‘Kida-aokabu’ and ‘Tennhoji-kabu,’ grown in an uncontrolled-environment greenhouse, respectively. The ms plant was crossed with a male fertile (mf) plant of ‘Kida-aokabu’ and ‘Tennhoji-kabu,’ respectively, and then their F₁ plants were selfed and backcrossed to the ms plants to obtain F₂ and BC₁ progenies, respectively. Six flowers were arbitrarily selected from one ms and one mf individual of the segregating population, respectively, and the sizes of the stamen, petal, pistil, and sepal were measured. Measurements were repeated three times using three different individuals in the ms and mf populations, respectively, and size differences were statistically evaluated by paired t-tests.

In addition, the ms plant was crossed with B. rapa ssp. rapa ‘Tokyo-nagakabu’ and B. rapa ssp. pekinensis ‘Ho Mei,’ respectively. Their F₁ plants were selfed and backcrossed to ms plants of each cultivar to obtain F₂ and BC₁ progenies, respectively. ‘Ho Mei,’ as the female plant, was crossed with F₁ plants produced by crossing ‘Kida-aokabu’ and ‘Tennhoji-kabu’ ms and mf plants, respectively. Moreover, the F₁ progenies of 12 and 14 randomly selected plants were selfed to obtain F₂ seeds from the resulting plants of ‘Kida-aokabu’ and ‘Tennhoji-kabu,’ respectively.

Floral buds more than 1.0 mm in length were collected every 0.5 or 1.0 mm from the ms and mf plants, respectively, and fixed in FAA (50% ethanol, 5% glacial acetic acid, 3.7% formaldehyde) for 18–24 h. Samples were rinsed several times in 50% ethanol, dehydrated through a graded ethanol series, and stained in 100% ethanol containing 1% safranin for easy orientation of the material. The buds were embedded in paraffin, cut into 2–3-µm sections, and stained with hematoxylin. Photomicrographs of the section were obtained using a Nikon OPTIPHOT-2 X2 microscope.

Results and discussion

In both ‘Kida-aokabu’ and ‘Tennhoji-kabu,’ the ms plants were morphologically similar to the mf plants except for the flowers, though the flowers of the ms plants were characterized by shriveled anthers without fertile pollen. Figure 1 shows the open flowers of ms and mf plants in both cultivars, demonstrating a clear size difference in the constitutive organs (Table 1). In ‘Kida-aokabu,’ the ms flowers had significantly (p＜0.01) shorter stamens and longer pistils than the mf flowers. A longer pistil in an ms flower was also reported in the previously identified ms in B. rapa ssp. rapa by Takahata et al. (1996) and Wakui et al. (2013). The sizes of the petal and sepal of ms flowers were similar to or slightly smaller than those of the fertile flowers. In ‘Tennhoji-kabu,’ on the other hand, the ms flowers had significantly (p＜0.01) shorter stamens and shorter sepals than the mf flowers, but the petal size and the length of the pistil in ms flowers were similar to those of the fertile flowers. These results show the differences in the size of the organs constituting each ms flower between ‘Kida-aokabu’ and ‘Tennhoji-kabu,’ suggesting that the ms differed between the two cultivars.

Fig. 1. Open flower morphology of male fertile (mf, left) and male sterile (ms, right) plants of B. rapa ssp. rapa ‘Kida-aokabu’ (upper) and ‘Tennhoji-kabu’ (lower).

Table 1. Comparison of flower characteristics between male sterile (MS) and male fertile (MF) lines of B. rapa ssp. rapa ‘Kida-aokabu’ and ‘Tennhoji-kabu.’

Phenotype	Characteristics (mm)
	Length of long stamen	Length of short stamen	Length of petal	Width of petal	Length of pistil	Length of sepal
‘Kida-aokabu’
MF	6.8±0.14	4.5±0.19	11.0±0.22	5.5±0.16	6.8±0.20	5.8±0.14
MS	5.2±0.11	3.1±0.16	10.5±0.15	5.4±0.16	8.8±0.24	5.8±0.15
t value	8.76**	5.59**	1.43	0.27	6.46**	0.07
‘Tennhoji-kabu’
MF	6.3±0.14	4.0±0.15	9.7±0.22	3.7±0.18	7.4±0.22	6.9±0.19
MS	5.2±0.15	3.2±0.15	9.4±0.15	4.0±0.12	7.2±0.19	6.2±0.10
t value	5.00**	3.96**	0.83	1.26	0.50	3.35**

** Significantly different at the 1% level

All of the F₁ plants obtained from the cross between the ms and two cultivars (‘Ho Mei’ and ‘Tokyo-nagakabu’) were mf (Table 2). However, ms plants also segregated in the F₁ population of the cross between the ms plant and each of the two parental cultivars (‘Kida-aokabu’ and ‘Tennhoji-kabu’). This indicates that each population used as the male parent included both heterozygous and homozygous mf plants in ‘Kida-aokabu’ and ‘Tennhoji-kabu,’ respectively. The segregation patterns in the F₂ and BC₁ generations are shown in Table 2. Segregation of the F₂ population of msבHo Mei’ and msבTokyo-nagakabu’ showed a 3 : 1 mf : ms ratio in ‘Kida-aokabu’ and ‘Tennhoji-kabu,’ respectively. In BC₁ plants obtained from the crosses ms×(msבHo Mei’) and ms×(msבTokyo-nagakabu’), a 1 : 1 mf : ms ratio was obtained in ‘Kida-aokabu’ and ‘Tennhoji-kabu,’ respectively. This suggests that male sterility in both cultivars is controlled by a single recessive gene.

Table 2. Segregation of male sterility obtained in F₁ populations of male sterile (ms) B. rapa ‘Kida-aokabu’ and ‘Tennhoji-kabu’ crossed with three cultivars and their progenies, respectively.

Cross combination		Phenotype		χ2	p
		MF1)	MS2)
‘Kida-aokabu’
msבHo Mei’	Obs.	16	0
msבTokyo-nagakabu’	Obs.	18	0
msבKida-aokabu’	Obs.	23	27	0.32	0.5＜p＜0.6
	Exp. (1 : 1)	25	25
(msבKida-aokabu’) F2	Obs.	30	8	0.32	0.5＜p＜0.6
	Exp. (3 : 1)	28.5	9.5
(msבHo Mei’) F2	Obs.	29	9	0.04	0.8＜p＜0.9
	Exp. (3 : 1)	28.5	9.5
(msבTokyo-nagakabu’) F2	Obs.	26	9	0.02	0.8＜p＜0.9
	Exp. (3 : 1)	26.25	8.75
ms×(msבKida-aokabu’) F1	Obs.	24	28	0.30	0.5＜p＜0.6
	Exp. (1 : 1)	26	26
ms×(msבHo Mei’) F1	Obs.	22	18	0.42	0.5＜p＜0.6
	Exp. (1 : 1)	20	20
ms×(msבTokyo-nagakabu’) F1	Obs.	17	22	0.64	0.4＜p＜0.5
	Exp. (1 : 1)	19.5	19.5
‘Tennhoji-kabu’
msבHo Mei’	Obs.	19	0
msבTokyo-nagakabu’	Obs.	18	0
msבTennhoji-kabu’	Obs.	25	24	0.02	0.8＜p＜0.9
	Exp. (1 : 1)	24.5	24.5
(msבTennhoji-kabu’) F2	Obs.	46	12	0.57	0.4＜p＜0.5
	Exp. (3 : 1)	43.5	14.5
(msבHo Mei’) F2	Obs.	37	11	0.11	0.7＜p＜0.8
	Exp. (3 : 1)	36	12
(msבTokyo-nagakabu’) F2	Obs.	28	4	2.67	0.1＜p＜0.2
	Exp. (3 : 1)	24	8
ms×(msבTennhoji-kabu’) F1	Obs.	16	19	0.26	0.6＜p＜0.7
	Exp. (1 : 1)	17.5	17.5
ms×(msבHo Mei’) F1	Obs.	16	14	0.13	0.7＜p＜0.8
	Exp. (1 : 1)	15	15
ms×(msבTokyo-nagakabu’) F1	Obs.	21	14	1.40	0.2＜p＜0.3
	Exp. (1 : 1)	17.5	17.5

¹⁾ Male fertile, ²⁾ Male sterile.

To examine the effect of ms on the cytoplasm, ‘Ho Mei’ as the female parent was crossed with the mf progeny of the msבKida-aokabu’ cross and msבTennhoji-kabu’ cross, respectively, as the male parent. All of the F₁ plants were mf. In F₂ families derived from 12 randomly selected F₁ plants, that is ‘Ho Mei’×(msבKida-aokabu’), four lines were composed of only mf plants, and eight lines had segregated ms plants, in agreement with the expected 3 : 1 mf : ms ratio (Table 3). The presence of four homozygous and eight heterozygous mf plants in the F₁ population agreed with the expected 1 : 1 ratio following the chi-square test (χ²=1.33, 0.2＜p＜0.3). Similarly, five lines were composed of only mf plants and nine lines were segregated ms plants in F₂ derived from 14 randomly selected F₁ plants, that is ‘Ho Mei’×(msבTennhoji-kabu’) (Table 4). The segregation data of the nine lines were in agreement with the expected 3 : 1 mf : ms ratio. The presence of five homozygous and nine heterozygous mf plants in F₁ was also in agreement with the expected 1 : 1 ratio according to the chi-square test (χ²=1.14, 0.2＜p＜0.3). These results confirm that the newly identified ms lines in ‘Kida-aokabu’ and ‘Tennhoji-kabu’ are controlled by a monogenic recessive gene.

Table 3. Segregation of male sterile plants obtained in the F₂ generation of ‘Ho Mei’×(msבKida-aokabu’).

Plant no. of F1 between ‘Ho Mei’ and (msבKida-aokabu’) F1			No. of plants in F2		χ2 (3 : 1)	p
			MF1)	MS2)
‘Ho Mei’×(msבKida-aokabu’)	F1-1	Obs.	32	13	0.36	0.6＜p＜0.7
		Exp.	33.75	11.25
	F1-2	Obs.	44	0
	F1-3	Obs.	32	12	0.12	0.7＜p＜0.8
		Exp.	33	11
	F1-4	Obs.	33	13	0.26	0.6＜p＜0.7
		Exp.	34.5	11.5
	F1-5	Obs.	45	0
	F1-6	Obs.	37	9	0.72	0.3＜p＜0.4
		Exp.	34.5	11.5
	F1-7	Obs.	35	8	0.94	0.2＜p＜0.3
		Exp.	32.25	10.75
	F1-8	Obs.	20	8	0.19	0.6＜p＜0.7
		Exp.	21	7
	F1-9	Obs.	29	0
	F1-10	Obs.	30	0
	F1-11	Obs.	24	10	0.35	0.5＜p＜0.6
		Exp.	25.5	8.5
	F1-12	Obs.	19	9	0.76	0.3＜p＜0.4
		Exp.	21	7
Total of F1-1, 3, 4, 6–8, 11, 12		Obs.	232	82	0.21	0.6＜p＜0.7
		Exp.	235.5	78.5

¹⁾ Male fertile, ²⁾ Male sterile.

Table 4. Segregation of male sterile plants obtained in the F₂ generation of ‘Ho Mei’×(msבTennhoji-kabu’).

Plant no. of F1 between ‘Ho Mei’ and (msבTennhoji-kabu’) F1			No. of plants in F2		χ2 (3 : 1)	p
			MF1)	MS2)
‘Ho Mei’×(msבTennhoji-kabu’)	F1-1	Obs.	25	13	1.72	0.1＜p＜0.2
		Exp.	28.5	9.5
	F1-2	Obs.	28	11	0.21	0.6＜p＜0.7
		Exp.	29.25	9.75
	F1-3	Obs.	40	0
	F1-4	Obs.	32	9	0.20	0.6＜p＜0.7
		Exp.	30.75	10.25
	F1-5	Obs.	32	10	0.03	0.8＜p＜0.9
		Exp.	31.5	10.5
	F1-6	Obs.	28	9	0.01	0.9＜p＜1.0
		Exp.	27.75	9.25
	F1-7	Obs.	31	6	1.52	0.2＜p＜0.3
		Exp.	27.75	9.25
	F1-8	Obs.	28	12	0.53	0.4＜p＜0.5
		Exp.	30	10
	F1-9	Obs.	34	9	0.38	0.5＜p＜0.6
		Exp.	32.25	10.75
	F1-10	Obs.	18	0
	F1-11	Obs.	29	0
	F1-12	Obs.	23	9	0.17	0.6＜p＜0.7
		Exp.	24	8
	F1-13	Obs.	36	0
	F1-14	Obs.	31	0
Total of F1-1, 2, 4–9,12		Obs.	261	88	0.01	0.9＜p＜1.0
		Exp.	261.75	87.25

¹⁾ Male fertile, ²⁾ Male sterile.

Transverse sections of mf and ms anthers in ‘Kida-aokabu’ and ‘Tennhoji-kabu’ were prepared and stained with hematoxylin to compare the process of anther development. Sanders et al. (1999) divided anther development into 14 stages, at which distinctive cellular events can be visualized with a light microscope. Ma (2005) described the characteristic four-lobed morphology of the anther, which undergoes morphogenesis by stage 5, followed by meiosis of pollen mother cells, which form the tetrad and dissociate from the tapetum during the transition from stage 6 to 7. In ‘Kida-aokabu,’ tetrads in the anther derived from buds 1.0–2.0 mm in length were observed in all locules of mf and ms plants, which were dissociated from the tapetum (Fig. 2A, G). This indicates that the anther in both the mf and ms at this size corresponds to stage 7 (Sanders et al. 1999, Ma 2005). At stage 7, although a similar phenotype was observed in both the ms and mf plants, as previously described, some key differences were also noted. In the ms anther, the tetrads were partially separated compared with those of the mf anthers, and the tapetum was swollen (Fig. 2A, G). Moreover, in the mf anther, the microspores released from the tetrads contained densely packed cytoplasmic cells with a nucleus and cell wall, which ultimately developed into pollen grains (Fig. 2B–D). The tapetum gradually degenerated, and then completely disappeared at the late microspore stage (Fig. 2E, F). By contrast, after the tetrads stage in ms anthers, there was a clear phenotypic difference in the anther in comparison to that of mf plants. In addition to tapetum swelling, the size of the microspores released from the tetrads was smaller than that of mf microspores (Fig. 2H), and these small microspores collapsed and degenerated in ms anthers (Fig. 2I). The tapetum also degenerated, and then both the microspores and the tapetum completely disappeared in the locule before developing into pollen grain (Fig. 2J–L). The anther locules sequentially shrunk and shrived, resulting in male sterility.

Fig. 2. Cytological characterization of microsporogenesis in male sterile (ms) plants of B. rapa ssp. rapa ‘Kida-aokabu.’ Bud sizes are as follows: (A and G) 1.0–2.0 mm, (B and H) 2.0–2.5 mm, (C and I) 2.5–3.0 mm, (D and J) 3.0–4.0 mm, (E and K) 4.0–5.0 mm, and (F and L) 5.0–6.0 mm. T: Tapetum, Tds: tetrads, Msp: microspore, PG: pollen grain. Scale bar=50 µm.

On the other hand, in ‘Tennhoji-kabu,’ the microspores in the anthers derived from buds 1.0–2.0 mm in length were observed immediately following release from each locule of both the mf and ms plants (Fig. 3A, G). This indicated that the anthers of both mf and ms plants at this size correspond to stage 8 (Sanders et al. 1999, Ma 2005). In the mf anther, the microspores released from the tetrads contained densely packed cytoplasmic cells with a nucleus and cell wall (Fig. 3A–C), which ultimately developed into pollen grains (Fig. 3D–F). The tapetum degenerated in association with the development of the microspore, and then completely disappeared at the late microspore stage (Fig. 3D–F).

Fig. 3. Cytological characterization of microsporogenesis in male sterile (ms) plants of B. rapa ssp. rapa ‘Tennhoji-kabu.’ Bud sizes are as follows: (A and G) 1.0–2.0 mm, (B, H–J) 2.0–3.0 mm, (C and K) 3.0–4.0 mm, (D and L) 4.0–5.0 mm, (E and M) 5.0–6.0 mm, and (F, N and O) 6.0–7.0 mm. T: Tapetum, Tds: tetrads, Msp: microspore, PG: pollen grain, R: Residue by adhesion of the substance in the locule. Scale bars=50 µm in A–E, G–I, K–N and scale bars=500 µm in F, J, O.

At stage 8, no phenotypic difference between mf and ms anthers was observed (Fig. 3A, G). However, at the subsequent stage, after temporary swelling of the tapetum (Fig. 3H), the microspores and the tapetum both gradually degenerated in the ms anther (Fig. 3I), whereas some locules appeared to proceed through normal microsporogenesis at this point (Fig. 3J). Thus, the collapse of the microspores and tapetum in the locule did not occur synchronously in the ms anther, and the same phenomenon was observed among four locules comprising a single anther in an ms plant at the same stage (Fig. 3J). Subsequently, almost all of the ms anthers showed microsporogenesis collapse, and the microspores and tapetum were degenerated, although a very small number of microspores remained to develop into pollen grain (Fig. 3K, L). However, the microspores and tapetum both completely disappeared at the late stage of pollen development, and no pollen grains and adhered residue of the substance formed by the degeneration of the components in locule were observed (Fig. 3M–O).

In most cases of cytological male sterility, microsporogenesis breaks down at the tetrad stage or more rarely at the uninucleate vacuolate microspore stage (Delourme and Budar 1999). Breakdown of microsporogenesis at the tetrad stage in male sterility was reported in B. rapa var. brown sarson and yellow sarson (Das and Pandey 1961, Chowdhury and Das 1966, 1968). On the other hand, Takahata et al. (1996) and Xie et al. (2005) reported the swelling and vacuolation of the tapetum and microspore abortion after the uninucleate microspore stage. These results are similar to the stage-observed microsporogenesis breakdown in the ms anthers of ‘Kida-aokabu’ and ‘Tennhoji-kabu,’ whereas the tapetum, which was swollen but not vacuolated, disappeared at almost the same point as the degradation of the microspore, which differs from that reported by Takahata et al. (1996) and Xie et al. (2005). By comparison between ‘Kida-aokabu’ and ‘Tennhoji-kabu,’ it was observed that the mode of microsporogenesis breakdown differed between the two cultivars. That is, we found a difference in the time at which the microspores and the tapetum begin to collapse among locules in ‘Tennhoji-kabu,’ though the microspore and the tapetum disappeared in all locules at the same time in ‘Kida-aokabu.’ Furthermore, cytological observation of the ms of ‘Kida-aokabu’ further suggested that abnormal microsporogenesis occurs when the pollen mother cells undergo meiosis, which is the stage after anther morphogenesis. Indeed, the outward appearance of the anther was maintained in the stamens of ms of ‘Kida-aokabu’ as well as those of the ms of ‘Tennhoji-kabu,’ whereas the anthers of ms plants shriveled compared to those of mf plants.

We identified a GMS plant in each of the turnip populations ‘Kida-aokabu’ and ‘Tennhoji-kabu,’ which are local Japanese varieties. Our discovery suggests that the ms gene was introduced into a part of the Japanese turnip population. Since turnip differentiated into several hundreds of local varieties in each district after its introduction in Japan (Shinohara 1984, Yoshikawa and Yui 1991), the discovery of another type of male sterility could be expected from the broad gene pool of Japanese turnip varieties. Indeed, we have found some other Japanese varieties showing ms traits, and their characterization is in progress. These studies will generate a useful resource for analyzing the molecular mechanism(s) causing male sterility and may provide useful information to use GMS for plant breeding in B. rapa in the future.

References

• Brandle, J. L. and Mcvetty, P. B. E. 1990. Geographical diversity, parental selection and heterosis in oilseed rape. Can. J. Plant Sci. 70: 935–940.
• Cao, S. C. and Li, S. J. 1981. Breeding of the genic male sterile line of Brassica campestris L. Chinensis. Aijiaohuang. Acta Hort. Sinica 8: 35–41.
• Chowdhury, J. B. and Das, K. 1966. Male sterility in Brassica campestris var. yellow sarson. Indian J. Genet. Plant Breed. 26: 374–380.
• Chowdhury, J. B. and Das, K. 1968. Cytomorphological studies on male sterility in Brassica campestris L. Cytologia 38: 195–199.
• Das, K. and Pandey, B. D. 1961. Male sterility in brown sarson. Indian J. Genet. Plant Breed. 21: 185–190.
• Delourme, R. and Budar, F. 1999. Male Sterility. In: Gómez-Campo, C. (ed.). Biology of Brassica Coenospecies. Elsevier, Amsterdam. pp. 185–216.
• Feng, H., Wei, P., Piao, Z.-Y., Liu, Z.-Y., Li, C.-Y., Wang, Y.-G., Ji, R.-Q., Ji, S.-J., Zou, T., Choi, S.-R. and Lim, Y.-P. 2009. SSR and SCAR mapping of a multiple-allele male-sterile gene in Chinese cabbage (Brassica rapa L.). Theor. Appl. Genet. 119: 333–339.
• Grant, I. and Beverdorf, W. D. 1985. Heterosis and combining ability estimates in spring oilseed rape (Brassica napus L.). Can. J. Genet. Cytol. 27: 472–478.
• Kaul, M. L. H. 1988. Male Sterility in Higher Plants. Springer-Verlag, Berlin.
• Ke, L. P., Sun, Y. Q., Hong, D. F., Liu, P. W. and Yang, G. S. 2005. Identification of AFLP markers linked to one recessive genic male sterility gene in oilseed rape, Brassica napus. Plant Breed. 124: 367–370.
• Liu, C., Liu, Z., Li, C., Zhang, Y. and Feng, H. 2016. Comparative transcriptome analysis of fertile and sterile buds from a genetically male sterile line of Chinese cabbage. In Vitro Cell. Dev. Biol. Plant 52: 130–139.
• Ma, H. 2005. Molecular genetic analyses of microsporogenesis and microgametogenesis in flowering plants. Annu. Rev. Plant Biol. 56: 393–434.
• Perez-Prat, E. and van Lookeren Campagne, M. M. 2002. Hybrid seed production and the challenge of propagating male sterility plants. Trends Plant Sci. 7: 199–203.
• Sanders, P. M., Bui, A. Q., Weterings, K., McIntire, K. N., Hsu, Y.-C., Lee, P. Y., Truong, M. T., Beals, T. P. and Goldberg, R. B. 1999. Anther developmental defects in Arabidopsis thaliana male-sterile mutants. Sex. Plant Reprod. 11: 297–322.
• Shinohara, S. 1984. Vegetable Seed Production Technology of Japan, Particulars Elucidated with Respective Variety Development History, Particulars 1. Shinohara’s Authorized Agricultural Consulting Engineer Office, Tokyo.
• Takahata, Y., Nagasaka, M., Kondo, H. and Kaizuma, N. 1996. Genic male sterility in Brassica campestris. Acta Hortic. 147–150.
• Van der Merr, Q. P. 1987. Chromosomal monogenic dominant male sterility in Chinese cabbage (Brassica rapa subsp. pekinensis (Lour.) hanelt). Euphytica 36: 927–931.
• Wakui, K., Shinohara, T., Hujihara, E., Igarashi, D. and Fujigaki, J. 2013. Genic male sterility in Brassica rapa ssp. rapa. J. Agri. Sci. Tokyo Univ. Agric. 57: 287–292.
• Wang, Z.-W., Gao, L., Liu, H. Z., Mei, S. Y., Zhou, Y., Xiang, C. P. and Wang, T. 2012. Genetic and cytological analysis of a new spontaneous male sterility in radish (Raphanus sativus L.). Euphytica 186: 313–320.
• Xie, C. T., Yang, Y. H., Qiu, Y. L., Zhu, X. Y. and Tian, H. Q. 2005. Cytochemical investigation of genic male-sterility in Chinese cabbage. Sex. Plant Reprod. 18: 75–80.
• Ying, M., Dreyer, F., Cai, D. and Jung, C. 2003. Molecular marker for genic male sterility in Chinese cabbage. Euphytica 132: 227–234.
• Yoshikawa, H. and Yui, S. 1991. Classification and identification of crucifer vegetables in Japan. JICA Ref. Series 4: 97–114.
• Zhou, X., Li, Z., Ji, R. and Feng, H. 2017. Comparative transcript profiling of fertile and sterile flower buds from multiple-allele-inherited male sterility in Chinese cabbage (Brassica campestris L. ssp. pekinensis). Mol. Genet. Genomics 292: 967–990.