研究概要 (Research Summary in Japanese)
共生の仕組みと進化の解明
マメ科植物は根粒菌と相互作用することによって、感染糸の形成や皮層細胞分裂を誘導し、「根粒」と呼ばれる窒素固定器官を形成する。一方、陸上植物の根にはアーバスキュラー菌根菌（AM菌）が共生し、成長に必要なリンや水分を効率よく吸収している。近年、マメ科植物の根粒共生は、 4 億年よりも前に起源を持つAM共生を基盤として、茎頂メリステム（SAM）や側根形成に必要とされる遺伝子を流用して進化してきたことが見えてきた。一方で窒素固定共生はバラ科やウリ科植物などで失われたことも分かってきた。
本研究部門では、日本に自生するマメ科モデル植物ミヤコグサとAM菌を主に用いて、共生の分子メカニズムと進化の解明を目指している。さらには進化過程で失われた根粒共生の復元や発生過程に連動した代謝システムの解明にも取り組んでいる。
根粒形成過程と共生遺伝子群
根粒の形成過程では、根粒菌の感染を契機に宿主植物のこれまで分化した組織であった根の皮層細胞が脱分化し、根粒原基形成に向けた新たな発生プログラムが実行される（図1）。
私たちはマメ科のモデル植物ミヤコグサを用いて網羅的な共生変異体の単離を行い、根粒菌の感染や窒素固定、さらには根粒形成の全身制御に関わる遺伝子を特定してきた。興味深いことに、根粒形成のごく初期に関わる遺伝子の多くは、植物にリンを与えるAM菌との共生にも必須であった（赤字で示した遺伝子）。共生の分子メカニズムと進化の解明を目指している。
図 1. 根粒形成過程の概要と根粒共生と菌根共生に必要な宿主遺伝子群
根と葉の遠距離コミュニケーションを介した根粒形成の全身制御
マメ科植物は根粒菌との共生により大気中の窒素を利用することができるが、窒素固定には多く生体エネルギーが消費されるため、植物は根粒の数を適正にコントロールしている。私たちは、ミヤコグサの根粒過剰着生変異体を用いて、根粒数が根と葉の間の遠距離コミュニケーションにより制御される分子メカニズムを解明してきた。根から葉へ遠距離移動する糖修飾CLEペプチド、そのレセプターであるHAR1、さらにはシュート由来因子を根で受けるTML等の解析を行っており、全身的なフィードバック制御の全容解明を目指している（図2）。また、植物が根の感染や窒素情報をあえて葉に伝達する理由は不明である。その謎の解明に取り組んでいる。
図 2. 「根」と「葉」の遠距離シグナル伝達を介した根粒形成の全身制御モデル
根粒形成シグナリングと共生系の進化
根粒共生とAM共生で初期応答に関わる遺伝子が共通するように、マメ科植物はどの植物にも保存される遺伝子をうまく利用しながら、根粒の形成を制御、調節していると考えられる。通常、植物は側根を発達させることで土壌中の限られた栄養を効率的に吸収できるように工夫している。私たちの研究成果から、根粒共生に特異的なNIN転写因子の下流で、側根の発達に関わる遺伝子が根粒の形成に流用されていることが分かってきた。根粒共生のために進化した因子が側根の発達経路とどのように相互作用しているのかを探ることで、根粒形成の進化やその仕組みの理解を目指している。
多様な AM菌の純粋培養及び形質転換技術開発
AM菌は宿主との共生なくして増殖できない絶対共生菌であり、形質転換系が確立されてないために、その分子機構はほとんど不明である。私たちはオミクス解析の情報を元に、多様なAM菌の純粋培養技術開発を試みている。また多核であるAM菌の形質転換法の開発にも挑戦している。
DIVISION OF SYMBIOTIC SYSTEMS
Rhizobium–legume symbiosis is one of the most successful mutually beneficial interactions on Earth. In this type of symbiosis, soil bacteria called rhizobia supplies the host legumes with ammonia produced through bacterial nitrogen fixation. In return, host plants provide the rhizobia with their photosynthetic products. To accomplish this biotic interaction, leguminous plants develop nodules on their roots. However, more than 80% of land plant families have symbiotic relationships with arbuscular mycorrhizal (AM) fungi. Despite marked differences between the fungal and bacterial symbioses, common genes are required for both interactions. Using a model legume, Lotus japonicus, we are trying to elucidate the molecular mechanisms of both symbiotic systems.
Visual overview of this lab’s work.
I. Root nodule symbiosis
Legumes develop de novo organs known as root nodules to accommodate symbiotic bacteria called rhizobia. Nodule formation involves two distinct processes: rhizobial infection in epidermis and nodule development accompanied by cell division in cortex (Figure 1).
Figure 1. Processes for nodule formation.
The cortical cell division provides an indispensable scaffold for rhizobia progression from epidermis to cortex (Figure 1; pink line), leading to successful nodule formation. Cortical cell division occurs just below the site of rhizobial infection in epidermis. Therefore, there appears to be a spatiotemporal coordination across epidermis and cortex in this symbiotic organogenesis. Epidermal expression of genes required for early symbiotic genes is sufficient for nodule development in cortex, suggesting that some kinds of signals generated in epidermis trigger cortical cell division (Figure 1; blue dotted arrow). However, little is known about the mechanism that coordinates these two events.
In this study, we conducted a time-course transcriptome analysis using a L. japonicus non-nodulation mutant “daphne”, which has uncoupled symbiotic events in epidermis and cortex, in that it promotes excessive rhizobial infection in epidermis but does not produce nodule primordia in cortex. Among genes that showed different expression patterns in daphne and wild type, we found IAA CARBOXYL METHYLTRANSFERASE 1 (IAMT1), which encodes the enzyme that specifically converts auxin (indole-3-acetic acid; IAA) into its methyl ester (MeIAA). A significant MeIAA increase after rhizobial infection was detected by using daphne roots, in which excess rhizobia are infected (Figure 2).
Figure 2. Auxin methylation in nodule symbiosis. IAMT1 converts IAA into MeIAA (left). Relative MeIAA amount of wild type (WT) and daphne at 0 (non-inoculation) or 2 d after inoculation (DAI) (right).
In Arabidopsis, IAMT1 reportedly serves development and differential growth of shoot. On the other hand, we found that IAMT1 is duplicated in the legume lineage, and one of the duplicates (named IAMT1a) was expressed by rhizobial infection in root of L. japonicus. LjIAMT1a-knockdown inhibited nodule formation (Figure 3). LjIAMT1a overexpression promoted MeIAA accumulation and nodule formation. In contrast, overexpression of LjMES17, a MeIAA esterase, decreased MeIAA levels and nodule number. These showed that auxin methylation is essential for nodule formation.
Figure 3. LjIAMT1a-knockdown inhibited nodule formation. Hairy roots harboring empty vector (for control; left) or LjIAMT1a-RNAi vector (for knock-down; right). Nodules are indicated by arrowheads. (Scale bars, 1 cm.)
Notably, application of MeIAA significantly induced expression of a symbiotic gene, NIN, without rhizobia (Figure 4). NIN is a key regulator for cortical cell division; therefore, auxin methylation could act on the process of cortical cell division. Whereas IAMT1a was expressed mainly in root epidermis, it would be interesting to see how MeIAA affects cortex. The result also suggests that auxin methylation is not simply due to alteration of auxin homeostasis, contrary to what many studies assumed, MeIAA, as well as other auxin secondary metabolites, is also an inactive molecule of IAA. Understanding the function of auxin methylation and MeIAA in nodule development should open a new avenue for auxin metabolisms in the plant biology.
Figure 4. Exogenous MeIAA significantly induced NIN expression without rhizobia. Relative expression levels of NIN after treatment with DMSO as mock (white bar), IAA (gray bar), or MeIAA (blue bar) for 24 h.
I. Arbuscular mycorrhizal symbiosis
Arbuscular mycorrhizal (AM) symbiosis is the oldest beneficial mutualistic relationship between the majority of terrestrial plants and AM fungi. AM fungal spores germinate and elongate their hyphae toward the root of host plants then enter the root cells and form finely branched structures, arbuscules, for exchanging nutrients with plants(Figure 5). AM fungi deliver mineral nutrients especially phosphorus in the soil to host plants. In return, they obtain organic carbon sources such as sugars which are produced through photosynthesis. AM fungi are obligate symbionts depending on their host plants for essential nutrients, thus they are incapable of propagation by themselves.
Figure 5. AM fungus R. clarus HR1 growing with carrot hairy roots. (A) Expanded extraradical hyphae and formed symbiotically generated spores between the roots. Bar, 500 µm. (B) Fungal structures stained by green fluorescent (Wheat germ agglutinin Oregon green 488) in the root. Arrowheads, arbuscules inside root cortical cells. Bar, 100 µm.
In the history of AM fungal culture research, it was shown that an AM fungus Rhizophagus irregularis could complete its life cycle in co-culture with mycorrhiza-helper bacterium Paenibacillus validus in the absence of host plants. Over the last several years, it was reported that AM fungi receive not only sugars but also fatty acids. Subsequently, from these experiments, it was demonstrated that some fatty acids such as palmitoleate and myristate induce R. irregularis spore formation under an asymbiotic (completely host-free) culture media.
We found that a fungus isolated from Aichi prefecture, Rhizophagus clarus HR1, whose genome has been sequenced, shows better growth on the asymbiotic culture medium, thus is more suitable for this culture method. A fatty acid, myristate promotes R. clarus hyphal growth, but the addition of only myristate could not ensure induction of asymbiotically-generated spores (AS) formation. Furthermore, the number of AS in this culture was much smaller than those of symbiotically-generated spores (SS) which are produced in the co-culture system with hairy roots.
Some phytohormones have been reported that they have important roles as signals during AM symbiosis. Especially, strigolactone is well known as a major signal which induces hyphal elongation and branching and stimulates their mitochondrial activity in the pre-symbiotic stage. By supplementing synthetic strigolactone, GR24 with myristic acid, the time of germination and AS formation in R. clarus was accelerated. Although not all R. clarus spores formed AS in the medium supplemented only with myristic acid, the addition of GR24 resulted in a quite high efficiency of AS formation to nearly 100 %. Besides strigolactone, it was reported that there was an increase of jasmonate levels in AM roots and the upregulation of genes involved in jasmonic acid biosynthesis in the cells containing arbuscules. Application of jasmonates in the medium with myristic acid and GR24 led R. clarus to produce a higher number of AS with larger sizes (Figure 6).
Figure 6. Produced spores on asymbiotic culture medium, supplemented with only myristate (A), myristate and GR24 (B), myristate, GR24 and methyl jasmonate (C). Arrowheads, parent spores. Bars, 500 µm.
Although the number of AS produced in the medium with myristate, GR24 and jasmonate has been close to that of SS, the size of AS is still smaller than SS. However, AS can be subcultured with the same efficiency as the first asymbiotic culture. Moreover, AS was capable of infecting hairy roots and plants. The growth of Welsh onions was significantly promoted by inoculation with AS as well as with SS (Figure 7). Our findings would lead us to more extensive studies of AM symbiosis and to the application of AS as an inoculum for basic research and for sustainable agriculture.
Figure 7. Inoculation of AM fungal spores to Welsh onions. Numbers following AS or SS are the number of spores for inoculation. (A) An appearance of plants inoculated with AM spores (upper). Bar, 5 cm. Ink-stained fungal structures in the roots (bottom). Bars, 50 µm. (B) Dry weight of Welsh onion shoots. Different letters indicate significant differences among treatments in each trial using Wilcoxon rank-sum test with Bonferroni correction, p < 0.05.