GS9 acts as a transcriptional activator to regulate rice grain shape and appearance quality

Identification of grain shape determining genes can facilitate breeding of rice cultivars with optimal grain shape and appearance quality. Here, we identify GS9 (Grain Shape Gene on Chromosome 9) gene by map-based cloning. The gs9 null mutant has slender grains, while overexpression GS9 results in round grains. GS9 encodes a protein without known conserved functional domain. It regulates grain shape by altering cell division. The interaction of GS9 and ovate family proteins OsOFP14 and OsOFP8 is modulated by OsGSK2 kinase, a key regulator of the brassinosteroids signaling pathway. Genetic interaction analysis reveals that GS9 functions independently from other previously identified grain size genes. Introducing the gs9 allele into elite rice cultivars significantly improves grain shape and appearance quality. It suggests potential application of gs9, alone or in combination with other grain size determining genes, in breeding of rice varieties with optimized grain shape.

1. The signal indicating the interactions in BiFC analysis looks weak and the method is prone to bring false positive result, especially in tobacco, a heterogeneous system. Additional verification should be required.
2. It should be detailed whether OFP8 or OFP14 also regulates grain shape. I noted that a recent report showed OFP1 is also involved in regulating grain shape (Front. Plant Sci. 2017, 8: 1698, which may be referred for the authors' consideration or discussion.
3. When the authors tested the interactions among GS9, GSK2, OFP1 and OFP14 and claimed that GS9 or OFP14 cannot interact with GSK2, a positive control confirming the interaction between GSK2 and OFP8 should be included to make sure the vector expressing GSK2 is alright.
4. The authors speculated that GS9 majorly regulates cell division on the longitudual direction. If this is true, it should be clarified whether OFPs, or GSK2, or BR also regulate grain length by majorly regulating cell division.
5. I suggested the authors to perform the subcellular localization analysis of both GS9 and OFP14 in rice cells, but not only in tobacco leaves, as the two systems sometimes led to different results.

Reviewer #3 (Remarks to the Author):
In this study, the authors identified novel grain shape regulating gene, GS9 by map-based cloning. The authors found that the NIL-gs9 shows increased cell number in longitudinal direction. The authors showed that the GS9 protein functions to a novel transcriptional activator and it interact with OsOFP14 and OsOFP8 in Yeast. The authors also analyzed the epistasis between GS9 and other grain size genes. However, GS9 had an additive effect with GW5 and GS3 in determining grain size and shape. Although the cloning of GS9 gene is solid and reliable, molecular mechanism of grain shape regulation is insufficient.
Major comments: 1. Although the authors showed that the GS9 protein interacts with OsOFP14 and OsOFP8 in yeast, no evidence of this interaction in vivo. The authors should confirm this.
2. Although the authors guessed that the GS9, OsOFP14 and OsOFP8 are involved in transcription co-regulation and this activity seems to be modulated by OsGSK2 kinase, there are no evidences in this manuscript. A transcriptome analysis between WT and gs9 lines may provide some candidate target genes.

Response to reviewers' comments
Response: Thanks for your comment and suggestion. Following your suggestion, a transcriptome analysis was performed to explore the functional mechanism of GS9 in the revised manuscript. The new data were shown in Supplementary Tables 3 and 4 Response: Thanks for your comment and suggestion. As you mentioned, the plant architecture is a key parameter to determine rice yield. In the dense planting conditions, rice plants with erect leaf phenotype, for example, mild BR-deficient or BR-insensitive mutant, will have enhanced per-unit area grain yield (Sakamoto et al., 2005). Rice with a certain increment of BR biosynthesis or signaling will led to the changes of rice architecture, including increased leaf angle. In conventional growth conditions, the per-plant grain yield of the BR-enhanced rice will increase (Wu et al., 2008). The data from our yield experiment exhibit no difference between the gs9 line and its wild-type control ( Figure 2g). There might be two possible reasons to this result. Firstly, the increment of the leaf angle is mild in gs9 mutant. Second, under our conventional planting conditions, the increased leaf angle in gs9 mutant line might not be serious enough to affect grain yield. The field space arrangement and the population to capture light and other resources still maintain a suitable level in our experiment. Thus, we have revised the discussion in the revised text as followings (Lines 537-541): "In present study, although slightly increased leaf angle was observed in NIL-gs9 plants (Fig. 2a), which may affect the performance of population photosynthesis efficiency, it did not altered population yield under conventional management in the field (Fig. 2g). It is possible that the NIL-gs9 population maintains a suitable level of the field spaced arrangement among individuals to capture sunlight and other resources.".

Comment: Another question is that except for the null-mutant allele occurred during the development of the SSSL N138, why function-defective gs9 alleles are not present in natural rice cultivars? Is this gene important for fitness?
Response: Thank you for your critical comments. In our study, a total of 114 germplasms, including 83 rice cultivars and 31 wild rice samples, were used for genotyping of GS9 locus. No function-defective gs9 allele was identified in these rice germplasms. However, this cannot completely eliminate the possibility that the function-defective gs9 exists in the untested rice germplasms. Another possibility is that due to a yield preference of early human selection, function-defective gs9 with slender grain shape but lack of yield contribution may have escaped human selection.
Due to the lack of selection pressure, the naturally reserved sequence variation of GS9 gene showed no consistent phenotype alterations. We have revised the description in the revised Discussion section (Lines 561-566) as followings: "…Therefore, it implied that sequence variation in the GS9 locus had been accrued during rice domestication, and they were naturally reserved due to no or limited effects on plant growth as well as morphology. Besides, it is also possible that the loss-of-function allele of GS9 may have been escaped during artificial selection, as the gs9 mutation lacks an effect on grain weight as well as final yields, and high yielding is highly preferred during modern rice breeding program." Comment: Please correct "grain weigth" to "grain weight" in Figure 2f.

Reviewer #2:
The present study by Zhao et al. reported the identification of a novel gene, named GS9, controlling grain shape in rice. The study highlighted the function of GS9 in regulating the length/width ratio of grains without affecting other agronomic traits including grain yield. Although the loss-of-function allele was actually derived from a natural mutation but not existed among the natural variations, the functional specificity of the gene render it practical value as a target for improvement of the grain appearance quality, an important trait for marketing consideration. Apparently, the major flaw of the study is about mechanism work regarding the functional relevance between GS9 and its interacting proteins OFPs, which lacks the basic genetic evidence as well as additional supporting information as detailed below.
Response: Thank you for the comments and suggestions. Based on your suggestions, we have carried out several additional experiments and got more solid data to support the interaction between GS9 and its interacting proteins. (1) An additional BiFC analysis in rice and Arabidopsis protoplast system was applied to further confirm the GS9-OsOFP8 and GS9-OsOFP14 interactions in vivo (Figure 5b, Supplementary   Figure 14). Moreover, the interaction between OsGSK2 and OsOFP8 was further confirmed by Y2H analysis (Figure 5f).
(2) We further tested the interplays among GS9, OsOFP14, OsOFP8 and OsGSK2 in co-regulating GS9 transcriptional activity by using dual-luciferase assays system. The results confirmed the repression effects of OFPs on GS9 transcriptional activity, and this repression effect could be partially recovered by OsGSK2 (Figure 5c believe that all these evidences support the interaction between GS9 with OsOFP14 in vivo.

Comment 2. It should be detailed whether OFP8 or OFP14 also regulates grain
shape. I noted that a recent report showed OFP1 is also involved in regulating grain shape (Front. Plant Sci. 2017, 8: 1698, which may be referred for the authors' consideration or discussion.
Response: Thank you very much. As you suggested, we got the mature seeds of OFP8-overexpression lines from Jianxiong Li's lab as reported (Reference#34), and our data showed that the mature grains also exhibited slender grain shape similar to that of gs9 mutant (Supplementary Figure 16), which consistent with its role in negatively regulating GS9 transcription activity. We have added related descriptions in the Results section (Lines 283-296) of revised text as followings: "With the end to clarify whether OsOFP8 and OsGSK2 also regulate grain size, we obtained their transgenic rice lines as previous reported 34,35 . Interestingly, the OsOFP8-overexpression line (OsOFP8-OE7) 34 exhibited a more slender grain phenotype than that of wild-type Zhonghua11 (ZH11), with a 7% increase in grain length ( Supplementary Fig. 16a-c). … The results of scanning electron microscopy showed no significant differences in longitudinal cell density on the outer surface of the glume between OsOFP8-OE7 or … and their corresponding wild-types ( Supplementary Fig. 16d, 17d). Thus, it suggested that the longer grain length phenotype of OsOFP8-overexpression or OsGSK2-RNAi lines was the result from an increase in longitudinal cell numbers of their spikelet hull. Overall, these results further indicated that GS9-OsOFP8-OsGSK2 complex participate in regulation of grain shape, at least in part, by modulating cell division." Moreover, we thank for your introduction of the recent report on OsOFP1 (Front. Plant Sci. 2017, 8: 1698, and we also learned another recent report on OsOFP19 (Plant J. 2017(Plant J. , 10.1111. Both OFPs were reported to be involved in regulating grain shape in rice, though with different extent of effects. In higher plants, OFPs were firstly reported to determine the transition from round to pear-shaped fruit in tomato. All these data suggest that rice OFPs might involve in regulating grain shape in different participatory approaches. We have discussed these in the revised Response: Thanks so much for your suggestion. According to your suggestion, we have performed an additional experiment and added the data on the interaction between OsGSK2 and OsOFP8 as a positive control, which was added in Figure 5f  whether OsOFP8 and OsGSK2 also regulate grain size, we obtained their transgenic rice lines as previous reported 34,35 . Interestingly, the OsOFP8-overexpression line (OsOFP8-OE7) 34 exhibited a more slender grain phenotype than that of wild-type Zhonghua11 (ZH11), with a 7% increase in grain length ( Supplementary Fig. 16a-c).
In OsGSK2-Gi-2 transgenic line 35 , suppression of OsGSK2 expression by RNA interference (RNAi) resulted in a 22% increase in grain length compared with that of the wild-type ZH11 (Supplementary Fig. 17a-c). The results of scanning electron microscopy showed no significant differences in longitudinal cell density on the outer surface of the glume between OsOFP8-OE7 or OsGSK2-Gi-2 and their corresponding wild-types ( Supplementary Fig. 16d,17d). Thus, it suggested that the longer grain length phenotype of OsOFP8-overexpression or OsGSK2-RNAi lines was the result from an increase in longitudinal cell numbers of their spikelet hull.
Overall, these results further indicated that GS9-OsOFP8-OsGSK2 complex participate in regulation of grain shape, at least in part, by modulating cell division." Comment 5. I suggested the authors to perform the subcellular localization analysis of both GS9 and OFP14 in rice cells, but not only in tobacco leaves, as the two systems sometimes led to different results.
Response: Thanks so much. According to your suggestion, we have performed the subcellular localization analyses of both GS9 and OsOFP14 in rice protoplast cells.

Reviewer #3:
In this study, the authors identified novel grain shape regulating gene, GS9 by map-based cloning. The authors found that the NIL-gs9 shows increased cell number in longitudinal direction. The authors showed that the GS9 protein functions to a novel transcriptional activator and it interact with OsOFP14 and OsOFP8 in Yeast.
The authors also analyzed the epistasis between GS9 and other grain size genes.
However, GS9 had an additive effect with GW5 and GS3 in determining grain size and shape. Although the cloning of GS9 gene is solid and reliable, molecular mechanism of grain shape regulation is insufficient.
Response: Thank you for your comments. By referring to all editorial and reviewers' suggestions, we have designed and performed additional experiments to facilitate the further exploration of the molecular mechanism of grain shape regulation. A number of new data were integrated into the revised manuscript. For example, we further confirmed the protein-protein interaction by performing BiFC in the rice and Arabidopsis protoplast system. Also, OsOFP14, OsOFP8 and OsGSK2 involved co-regulation of the transcriptional activity of GS9 was also dissected. Furthermore, we demonstrated that GS9, OsOFP8 and OsGSK2 are all involved in the regulation of grain length by modulating cell division in the longitudinal direction of the rice glume.
Finally, RNA-sequencing data provided some candidate downstream target genes that might mediate GS9-regulated grain development.

Major comment 1. Although the authors showed that the GS9 protein interacts with
OsOFP14 and OsOFP8 in yeast, no evidence of this interaction in vivo. The authors should confirm this.
Response: Thanks for the suggestion. In our previous version of manuscript, we confirmed the interaction between GS9 and OFP14 in both yeast and tobacco epidermal cells. In the revised manuscript, we first repeated the interaction between GS9 and OsOFP14 in tobacco leaf cells, and then we further verified the GS9-OFP14 and GS9-OFP8 interactions in rice and Arabidopsis protoplast by using BiFC analysis.
The result is consistent with that observed in the tobacco system, hence confirming that GS9 could interact with both OsOFP14 and OsOFP8 in the nucleus of plant cells.
The new results (Figure 5b and Supplementary Figure 14) and the interpretation of the results (Lines 256-259) were added in the revised manuscript.

Major comment 2. Although the authors guessed that the GS9, OsOFP14 and
OsOFP8 are involved in transcription co-regulation and this activity seems to be modulated by OsGSK2 kinase, there are no evidences in this manuscript.
Response: Thank you for your comments and suggestions. According to your suggestions, we have carried out more experiments and tested again the hypothesis by using the pMN6 transient transcriptional activity assay system. Our data showed that both OsOFP14 and OsOFP8 can suppress GS9 transcriptional activity, while OsGSK2 alleviates the suppression effects likely by directly interacting with OFPs ( Figure 5d).
And we have revised the text (Lines 275-279) as followings: "The dual-luciferase assays also revealed the repression effect of OsOFP8 on the transcriptional activity of GS9 (Fig. 5c,d), and this repression could be partly recovered by OsGSK2 (Fig. 5d and Supplementary Fig. 15). As expected, co-expression of OsOFP14 and OsOFP8 had a more serious repression effect on GS9 transcriptional activity, and it could also be attenuated by OsGSK2 (Fig. 5d).". Overall, these findings suggest that OsGSK2