Phytoplankton bloom is the most important event to stimulate the productivity of ocean ecosystems. This event largely contributes to a process known as biological pump, a major natural process of carbon dioxide sequestration in the ocean. About a half of the photosynthetically produced organic matters are consumed by heterotrophic microorganisms in marine surface layers. Active growth of these heterotrophs facilitates organic matters flux to higher trophic levels via microbial loop and controls sinking flux of organic matters as well as their remineralization for recycling inorganic nutrients. Therefore, the interactions between phytoplankton and heterotrophic bacteria lie at the heart of the carbon cycle in the oceans.

During the last three decades, microbial communities in the ocean have been described in regard to phylogenetically diverse microbial taxa, due to the developments of culture-independent molecular ecological approaches such as nucleic acid sequencing of phylogenetic marker genes directly retrieved from environmental DNA samples. In addition, total bacterial production has been routinely measured by the incorporation of radiolabeled substrates. The fundamental questions are "Which phylogenetic groups account for total bacterial productivity?" and "What is the relative contribution of each?" These are substantially important to our understanding of the food web dynamics and biogeochemical cycles in the ocean but remain unknown very well.

To answer these questions, I have firstly developed the novel single-cell based method, named bromodeoxyuridine immunocytochemistry-fluorescence in situ hybridization (BIC-FISH). This is a combined method of two different techniques that can measure single-cell activity or growth rates identifying their phylotypes. Bromodeoxyuridine (BrdU) immunocytochemistry is a technique to detect BrdU-incorporating (presumably growing) cells with the use of fluorescently labeled antibody. Fluorescence in situ hybridization (FISH) is a technique to detect specific phylogenetic groups of microorganisms with the use of fluorescently labeled oligonucleotide probes targeting to 16S ribosomal RNA.

In Chapter 2, the newly developed method has successfully applied to seawater bacterial assemblages in order to compare phylotype-specific growth rates. Also the method worked well with eight bacterial isolates tested in the study. Significant positive correlation between average cellular fluorescence intensity and cell-specific BrdU contents indicated the potential of the method to calculate the growth rate at single-cell levels identifying the phylotype of each cell. When the method was applied to a eutrophic seawater, analysis of single-cell immunofluorescence intensity revealed the differences of cellular growth rates and percentage in actively growing cells among various phylotypes, in which SAR86 and Vibrio group bacteria were growing more rapidly than others.

In Chapter 3, I applied the BIC-FISH method to study year-round changes of phylotype-specific bacterial production in a eutrophic seawater and explored the environmental factors controlling these changes. The monitoring by BIC-FISH throughout the year revealed the importance of the Roseobacter/Rhodobacter group bacteria as a constantly proliferating basic population (27% of BrdU-positive cells), although their contribution was not significantly correlated with water temperature or with chlorophyll a or organic matter concentrations. Also, the Bacteroidetes group bacteria was another important group, as they greatly increased in abundance after the end of phytoplankton blooms. Two other phylotypes tested in this study, the SAR86 and Vibrio group bacteria, changed their activities corresponding with water temperature.

In Chapter 4, the BIC-FISH method was used in a mesotrophic seawater to explore whether I can find a similar pattern of phylotype-specific growth response as described in the Chapter 3 in other environmental settings. The method was improved to increase the sensitivity of FISH and applied to less productive oceanic region than that studied in the Chapter 3. The improved protocol was successfully applied to the samples collected during the spring phytoplankton bloom in the western North Pacific. The growth pattern of Roseobacter/Rhodobacter and Bacteroidetes group bacteria was in agreement with the results as described in the Chapter 3. Roseobacter/Rhodobacter group bacteria were highly active regardless of the presence and absence of phytoplankton bloom, and the activity and production of Bacteroidetes group bacteria were especially high at the aged phytoplankton bloom stations. The dominance of these subgroups in total bacterial production indicated their importance as key bacterial groups in this season and area. The SAR86 group bacteria were more abundant in active cells at the bloom stations than other stations, although their activity was not correlated with several environmental factors including water temperature as indicated by the Chapter 3. The activities of SAR86 and SAR11 group bacteria (percentage of BrdU-positive cells) were high in the bloom stations. However they were not significantly correlated with chlorophyll a and organic matter concentrations. Phytoplankton bloom conditions, e.g. fresh and aged, might be affected their activity. The study also showed that Alteromonas group bacteria known as numerically minor group in in situ seawater were abundant in actively proliferating cells in the bloom stations. Their abundance and activity were strongly correlated with chlorophyll a and organic carbon concentrations.

In Chapter 5, I described succession patterns in biomass and productivity of major phylogenetic groups of bacteria during an artificial phytoplankton bloom of mesocosm experiment using the BIC-FISH method, and revealed phylotype-specific contribution to total bacterial production and decomposition of the phytoplankton bloom. The biomass and activity of Roseobacter/Rhodobacter and Bacteroidetes group bacteria were maintained at high levels of contribution to total bacterial biomass and production through the experiment and further increased following the peak of the phytoplankton bloom. Especially, the activity of Bacteroidetes group bacteria increased earlier than other bacteria just after the peak of the phytoplankton bloom, indicating that they triggered to decompose the phytoplankton bloom. Alteromonas group bacteria were only appeared as a major group in bacterial biomass and production at the peak of the phytoplankton bloom, showing their rapid response and high growth rate. The biomass and activity of SAR11 group bacteria increased with increase of phytoplankton. However, SAR86 group bacteria didn't respond to the phytoplankton bloom in this experiment, implying their limited response to some specific organic matters derived from phytoplankton under high temperature conditions.

This study revealed growth performance of each phylogenetic group of marine bacteria in response to phytoplankton blooms, their patterns of response and relative contribution to total bacterial biomass and production by the use of newly developed single-cell method. Roseobacter/Rhodobacter, Bacteroidetes, Alteromonas, SAR11 group bacteria were suggested to be important subgroups as prominent consumers of the photosynthetically produced organic matters. Based on my spatiotemporal and experimental study of phytoplankton blooms, their relative contributions to total bacterial production were estimated to be 20-40% for Roseobacter/Rhodobacter group bacteria, 20-40% for Bacteroidetes group bacteria, 0-30% for Alteromonasgroup bacteria, and 5-15% for SAR11 group bacteria. The sequential growth response of these subgroups to a phytoplankton bloom also illustrated a degradation mechanism of phytoplankton derived organic matters. I have hypothesized that Bacteroidetes group bacteria are firstly proliferated and trigger the collapse of phytoplankton bloom, then Roseobacter/Rhodobacter and SARI group bacteria actively proliferate and facilitate degradation. Finally, the activity of Alteromonas group bacteria contributes the degradation. I found that some phylogenetic groups of bacteria (Roseobacter/Rhodobacter, Bacteroidetes, and Alteromonas) enlarged their cell sizes during the phytoplankton bloom. In addition to the change of bacterial community structure, up-shift of averaged cell volumes due to the increase of large phylotype abundance may drastically change the organic matter flux from bacteria to their grazers flagellates. The exponential increase of bacterial biomass should stimulate a predator-prey interaction between protozoa and bacteria, which may lead to the change of ecosystem structures through a microbial "trophic cascade".

The BrdU approach described in this study would be one of the powerful tools for better understanding of the linkage between organic matters and bacterial communities. The diversity and function of microorganisms in the ocean ecosystems will be revealed combining this approach with some other culture-independent molecular approaches and also culture-dependent approaches.

審査委員会の各委員に本論文を提出し、研究内容に関する査読審査期間を経た後に、口頭発表と質疑応答による最終試験を行った。最終試験は、平成22年1月19日15時~17時、東京大学海洋研究所大講義室において行った。最終試験の席上において、各委員から以下のような論文審査の結果が示された。

研究の意義について

海洋における有機物生産場としての植物プランクトンブルームは、生物活動が最もダイナミックに変動する場である。ブルームによって生産された大量の有機物の質的最的変化、特に細菌群集による利用・分解過程を明らかにすることは、海洋生態系における物質循環過程を理解する上で不可欠の研究課題である。これまで、植物プランクトンブルーム時における細菌群集の動態については、16SrRNA遺伝子をターゲットにしたクローンライブラリ法やフィンガープリンティング法による群集構造解析を中心とした定性的な解析が行われ、その利用・分解過程に関わる主要細菌系統群が明らかとなってきた。しかしながら、こうした主要系統群の生産量の把握や全体の細菌生産への寄与率といった定量的な解析については、その解析の困難さから、未だほとんど行われていない。本研究では、新しい手法の条件検討から始めて、最終的に現場環境に適用できるレベルまで洗練することに成功した。審査委員からは、さらに定1定量性を高める余地が残されており検討すべき点が指摘されたが、現状でもこれまで困難であった解析を可能とする手法を確立し、世界的にみても他では真似のできない解析データを取得したことが、新規性や独自性の面から非常に高く評価された。

全体的な研究内容について

本研究は、植物プランクトンブルームに由来する有機物の利用・分解過程に関わる主要細菌系統群について、系統群別の細胞数、細胞サイズ、増殖(DNA合成活性)を、独自に開発した細胞レベルの手法を用いて定星的に解析することで、各細菌系統群の増殖応答や細菌生産への寄与率の時空間変動パターンを明らかにすることを目的としたものである。研究の結果、植物プランクトンブルームの時空間変動に応じて、細菌系統群ごとに特徴的な異なる増殖のパターンがあることが明らかとなった。ブルームの生成消滅に伴って全体の細菌生産量が変動するが、同時に細菌生産を担う系統群も時空間的に遷移してゆく。自然海水中の細菌系統群を識別しながら、その増殖をモニターするという新しい独自の手法を用いることによって、現場環境中における細菌群集のダイナミックな遷移過程を初めて定量的に明らかにしたものとして、その結果の新規性が評価された。一方で、従来の知見に対する研究成果の位置づけにおいて、論文中で曖昧な部分があることが指摘された。本研究で得られた結果の価値や新規性を明確に位置づけることができるように、引用文献の追加、論文の加筆、修正が要望された。また、植物プランクトンブルームや対象とした海域の環境に関する記載、主要細菌群との関係についての考察に不足する点があることが指摘され、これらの点についても適宜本論文の加筆、修正が要望された。

具体的な研究内容について

論文中では以下に示したような具体的な成果が丁寧にまとめられ、注意深く考察されていると評価された。富栄養海域における年間を通した連続モニタリングでは、Roseobacter系統群は安定的に細菌生産に対して高い寄与率を維持すること、Bacteroidetes系統群はブルームの崩壊過程において活発に増殖する系統群であることが示された。中栄養海域に形成された複数の春季植物プランクトンブルームを調査した結果では、それぞれ各細菌系統群ごとに特徴的な遷移パターンがあることがわかった。特に、これまで数的に少ないとされ物質循環的視点からは研究されてこなかったAlteromonas系統群が、ブルーム内で劇的に増加することを初めて見出し、群集動態解析の新たなターゲットとして注目された。人為的に植物プランクトンブルームを発生させるメソコスム実験では、複数の細菌系統群の増殖活性の遷移から、これらの系統群の連動的な増殖によって植物プランクトン起源の有機物の分解が進むことが明らかとなった。

以上、海洋における有機物の利用・分解に関わる主要細菌系統群の遷移過程を定量的に明らかにした本研究の成果は、喫緊の研究課題として近年注目されている環境変動に伴う海洋生態系の応答や有機物質循環の変動機構の解明など、海洋生態系における物質循環過程の理解に新たな視点を加えるものである。審査委員からの要望をふまえつつ、研究内容の独自性、新規性、完成度などを総合的に審査した結果、本論文は博士(農学)の学位論文としてふさわしいものであると判断された。