Tillering dwarf mutants [dwarf3 (d3), d10, d14, d17 and d27] in rice, which reduced plant stature and increased tiller number were previously characterized. Among them, the D3 gene and the D10 gene were identified by map-based cloning and were shown to encode proteins orthologous to Arabidopsis MAX2/ORE9 and MAX4, respectively. The Arabidopsis max2/ore9 mutant was originally isolated by screening for mutants that show a delay of senescence. Thus, I examined whether the rice d3 mutant also delays leaf senescence or cell death as in the case of the Arabidopsis max2/ore9 mutant. On the other hand, while investigating characteristics of the rice d3 mutant, I noticed that the lengths of coleoptile and mesocotyl were significantly longer in d3 than in its reference line, cv. Shiokari during germination under darkness. However, the coleoptile and mesocotyl growth under light conditions was comparable between d3 and Shiokari. It was interesting for us to clarify the reason for difference of the habits of coleoptile and mesocotyl grown under darkness or under light conditions between d3 and Shiokari. The phenotype of longer coleoptile and mesocotyl in d3 under darkness was similar to the phenotype observed in rice jasmonic acid (JA)-deficient mutant cpm1 (coleoptile photomorphogenesis1). Thus, I investigated whether the plant hormones such as JA were involved in the elongation of coleoptile and mesocotyl in d3 under dark conditions.

1. The d3 mutant has increased leaf longevity during senescence or cell death induced by darkness, H2O2 or jasmonic acid.

Senescence or cell death in plant leaves is known to be inducible by darkness, H2O2 and JA. When the Arabidopsis gene MAX2/ORE9 is disrupted, leaf senescence or cell death in response to the above stimuli is delayed. Because the rice gene D3 is orthologous to MAX2/ORE9, I wished to know whether disruption of D3 also results in more longevity in leaves. I found that senescence or cell death induced by darkness, JA and H2O2 in the third leaf (as measured by chlorophyll degradation and membrane ion leakage) in d3 rice mutant was delayed by 1-3 d compared to that in its reference line Shiokari. To confirm the delay of leaf senescence in d3, I examined the expression of three Senescence Associated Genes (SAGs; Osl20, Osl85 and Osl295) that are known to be induced during leaf senescence. The mRNA levels of the three SAGs started to increase dramatically in Shiokari at 1 d after transfer to dark conditions, peaked at 3 d, and then decreased. However, in d3, induction of the SAGs was delayed for 1-3 d compared with the induction observed in Shiokari. Moreover, the mRNA levels of D3, HTD1 and D10, which are orthologs of Arabidopsis MAX2/ORE9, MAX3 and MAX4, respectively, increased during senescence or cell death. These results suggest that the induction of gene expression of the D3, HTD1 and D10 genes is associated with the occurrence of senescence or cell death in leaves. Together, it is suggested that D3 protein in rice, like MAX2/ORE9 in Arabidopsis, is involved in leaf longevity during leaf senescence or cell death induced by darkness, JA and H2O2.

2. Coleoptile and mesocotyl in the d3 mutant are highly elongated under darkness, but not under light conditions.

When the d3 mutant and Shiokari were germinated under darkness, I found that the lengths of coleoptile and mesocotyl were significantly longer in d3 than in Shiokari. However, such differences of the coleoptile and mesocotyl lengths were not observed between the d3 mutant and Shiokari seedlings grown under light conditions. In contrast, the lengths of coleoptiles grown under complete submergence (i.e., anaerobic conditions) were comparable between d3 and Shiokari even if the seedlings were germinated under darkness. Moreover, the timing of the stop of the coleoptile elongation and timings of start of the coleoptile splitting and the induction of Senescence-Associated Genes (SAGs), both of which are senescence associated markers, were delayed in d3 grown under darkness, but not under light conditions. These results indicate that the phenotype of elongated coleoptile in d3 is not due to enhancement of the coleoptile elongation, but is due to delay of the coleoptile senescence under darkness.

The phenotypes of the mesocotyl and coleoptile elongation under darkness were similar to the phenotypes observed in rice JA-deficient mutant cpm1. Thus, I examined whether JA is involved in the coleoptile and mesocotyl elongation in d3 under darkness. Treatment of JA (500 nM MeJA) under darkness reduced the lengths of mesocotyl and coleoptile of d3, and their lengths were almost the same between d3 and Shiokari, suggesting that the JA treatment can suppress the enhanced elongation of coleoptile and mesocotyl in d3. The LC-MS analyses showed that the JA amounts in the coleoptiles were comparable between d3 and Shiokari. These results suggested that the phenotype of the enhanced mesocotyl and coleoptile elongation was not due to the difference of JA contents between d3 and Shiokari, but rather due to reduced sensitivity of coleoptile and mesocotyl to JA in d3 comparing to Shiokari. Thus, higher concentration of JA may be required for reduction of the lengths of coleoptile and mesocotyl in the d3 mutant.

The LC-MS analyses showed that the content of ABA in the d3 coleoptile was nearly two fold higher than that in the Shiokari coleoptile. It was likely to be due to the activation of the ABA biosynthesis and the repression of the ABA catabolism in d3 mutant coleoptiles compared to that in Shiokari. On the other hand, the GA content in the d3 coleoptile was about half of the GA content in the Shiokari coleoptile.

I also investigated sensitivity of the d3 coleoptile to red light of far-red light by checking the rate of inhibition of coleoptile growth. The coleoptile of the d3 mutant initially had only a slight responsiveness to red light comparing to the case of Shiokari, and then the d3 coleoptile gained greater responsiveness during elongation of coleoptile.

In conclusion, I found that the d3 mutant increased leaf longevity during senescence or cell death induced by darkness, JA and H2O2. The d3 coleoptile and mesocotyl were highly elongated and the senescence of coleoptile in the d3 mutant was also delayed under darkness, but not under light conditions. The d3 coleoptile initially had only a slighter responsiveness to red light compared to wild-type, and gained greater responsiveness during elongation of coleoptile. This study suggested that the deficiency of D3 protein affected to seedling development and senescence of leaves and coleoptile as well as tillering and stature of plant in rice. The target protein of rice D3 is not known. Thus, future characterization of D3-target protein appears to be necessary to understand the D3-dependent control of senescence and seedling development and to understand the relationship among tillering and stature of plant, senescence and seedling development in rice.

本研究は、イネの腋芽の成長に関わるDWARF3 (D3) 遺伝子が、葉の老化時における老化速度の調節および発芽時における子葉鞘・メソコチルの伸長の調節にも関与することを証明したものであり、次の2つの章から構成されている。

1. 暗黒、ジャスモン酸(JA)、過酸化水素(H2O2)処理によって誘導される老化の遅延に関与するイネdwarf3 (d3) 変異体の解析

イネdwarf3 (d3)は、多分げつ矮性の形質を示す変異体として単離され、原因遺伝子はシロイヌナズナの腋芽成長抑制因子MAX/SMSの受容・シグナル伝達を調節するMAX2/ORE9のオルソログをコードする。シロイヌナズナのmax2/ore9変異体は、分枝数の増加のほかに葉の老化遅延を示すことが報告されていたため、イネd3においても老化の遅延を示す可能性が考えられた。そこで、d3と対照品種であるしおかりの幼植物体に対して、それぞれ暗黒処理、JA処理、H2O2処理をすることによって、老化を誘導させ、d3としおかりの間で老化速度に違いがあるかどうかを調査した。そのために、老化の指標としてよく用いられるクロロフィル分解の割合、細胞膜からのイオン漏出量、Senescence Associate Genes (SAGs)の遺伝子発現パターンを、d3としおかりの間で比較した。その結果、どの処理に対しても、両系統ともクロロフィルの分解、イオン漏出、SAGsの発現が誘導されたが、それらの開始がしおかりと比べd3の方で1-3日遅延していた。これに対し、止葉を用いた自然老化(natural senescence)の実験では、クロロフィルの分解、SAGsの発現ともに、d3としおかりの間で有為な差は観察されなかった。

一方、暗黒、JA、H2O2処理によって誘導される老化の過程において、MAX/SMS経路に関わるD3, HTD1, D10遺伝子の発現を調査したところ、3遺伝子とも老化の過程で発現が誘導されるが、SAGsの場合と異なり、d3において発現誘導の遅延は観察されなかった。

以上の結果より、イネd3変異体ではシロイヌナズナmax2/ore9変異体と同様に、分枝の増加のほかに老化の遅延が観察された。しかし、max2/ore9変異体の場合とは異なり、自然老化の遅延には影響がなかった。このことより、イネD3とシロイヌナズナMAX2/ORE9の役割は、植物種間で必ずしも完全に一致していないことが明らかになった。また、MAX/SMS経路は老化のプロセスの上流に位置し、老化速度を調節することが示唆された。

2. 暗所発芽時の子葉鞘、メソコチルの伸長促進に関与するイネd3変異体の解析

暗所で発芽させたd3の芽生えでは、子葉鞘、メソコチルの長さが対照品種しおかりと比べ有為に長いことが明らかになった。子葉鞘は特殊な器官で、発芽後2, 3日間は伸長するが、その後老化が誘導され、伸長が停止することが知られている。d3としおかりを用いて、子葉鞘の伸長を経時的に測定したところ、しおかりでは発芽後4日目で子葉鞘の伸長が停止するのに対し、d3の子葉鞘は7, 8日目まで伸長を続けることが明らかになった。さらに、老化の指標遺伝子であるSAGsの発現を調査したところ、これらの遺伝子の発現誘導がしおかりの場合と比較して1-3日遅延していた。これらのことから、d3では子葉鞘の老化が遅延するために、しおかりと比べ子葉鞘が長いことが明らかになった。

d3の子葉鞘、メソコチルが伸長するという形質は、イネのJA欠損変異体であるcpm1やhebibaと酷似していた。そこでd3の形質にもJAが関与するかどうかを確かめるために、発芽時にJA処理を行ったところ、d3の子葉鞘、メソコチルの長さがしおかりと同程度になることが明らかになった。この結果より、d3の芽生えでは、JAの蓄積量が少ないか、もしくはJAに対する感受性が低い可能性が考えられた。しかし、吸水後4日目および5日目の子葉鞘内のJA量を測定したところ、d3としおかりの間で有為な差は観察されなかった。一方、d3の子葉鞘におけるJA応答性遺伝子の発現を調査したところ、低濃度のJAではあまり誘導されないが、高濃度のJAで処理した場合はしおかり内の遺伝子発現と同程度まで回復することが明らかになった。これらの結果より、d3の芽生えではJA感受性が低下しているために、しおかりと比べ、子葉鞘、メソコチルがより伸長することが示唆された。

以上、本研究では、MAX/SMS経路のシグナル受容・伝達に関わるイネD3遺伝子が、腋芽の休眠だけでなく、老化の調節にも関与することを明らかにできた。さらに、D3は芽生えにおけるJA感受性も調節するということを世界に先駆けて発見することができ、学術上価値が高いといえる。したがって、審査員一同は、本論文が博士(農学)の学位論文として価値があるものと認めた。