[M-ZZ45] 地球化学の最前線：その魅力や将来の展望を語り合う

[MZZ45-12] 本邦成層型湖沼における微生物群集組成のマルチ空間スケール比較解析と地球化学的示唆

*板井 啓明¹、平山 耕太郎¹、石水 浩喜¹、佐藤 佑磨¹、高橋 嘉夫¹、砂村 倫成¹ (1.東京大学大学院理学系研究科　地球惑星科学専攻)

キーワード：バクテリア、環境DNA、湖沼、群集組成解析

Comparing the composition of bacterioplankton communities at various spatial scales is crucial for characterizing aquatic environments. However, achieving comparability across different studies poses a challenge due to variability in processing methods. In this study, we present a multiscale comparison of bacterial communities utilizing 16S rRNA gene metabarcoding data collected from multiple oligotrophic lakes in Japan, alongside geochemical properties analysis. Our study delved into variations within individual lakes, encompassing horizontal, vertical, and seasonal differences in two lakes: Lake Nojiri (NJR) and Lake Kizaki (KZK). Data from nine thermally stratified Japanese lakes were also employed to compare bacterial communities among adjacent lakes and across catchments. Principal component analysis revealed distinguishable geochemical properties among three catchments: Fuji Five Lakes (FF), Nishina Three Lakes (NT), and Urabandai Three Lakes (UT), with differences attributed to bedrock geologies. Cell densities tended to be higher in hypolimnetic waters compared to epilimnetic waters. Estimated phosphorus (P) levels derived from bacterial cell counts relative to total particulate P were higher in the epilimnion (47-96%) than in the hypolimnion (38-77%). Alpha diversities, as assessed using Shannon indices, ranged from 3.01 to 4.52 for FF, 3.34 to 4.40 for NT, and 3.78 to 4.84 for UT. Significant positive correlations with Shannon indices were observed relative to latitude, total and dissolved Rb, while a negative correlation was noted with pH levels. Beta-diversity indices revealed distinct differences between epilimnetic and hypolimnetic bacterial communities, with dispersion prevalence confirmed across horizontal scales. The Mantel test indicated temperature as the most influential factor in shaping phylogenetically distinct populations, followed by dissolved oxygen (DO) levels and pH. Certain trace elements (Co, Mn, Fe, and Si) exhibited high correlation coefficients with Weighted-Unifrac distance. The abundance of metabolic pathways, as assessed using PICRUSt pipeline, demonstrated the strongest negative correlation with the reverse tricarboxylic acid (TCA) cycle and DO levels. In the presentation, we will also show our attempt to compare the sequence data to the other studies in the world to complement spatial scale dependence of microbial diversity. The data provided in this study, obtained using identical data processing protocol across 9 lakes in Japan, can provide a useful baseline for aquatic microbiology in the freshwater system.