Genome analysis of bacteria inhabiting the mucosal surface of intestine

我们研究的最终目标是阐明肠道微生物区系的组成和基因组特征以及它们的位置差异。然而，通过粪便样本的元基因组分析不可能知道它们的位置差异。我们需要对直接从肠道获得的样本进行分析，但在人体内进行这种采样伴随着许多技术和伦理问题。因此，本研究以小鼠肠道为模型系统。在目前的研究项目中，我们重点研究了位于粘膜表面的微生物区系，它们显然在宿主与驻留微生物的相互作用中发挥着最重要的作用，特别是在节段性丝状细菌(SFB)的基因组序列测定方面。SFB是一种不能培养的细菌，在断奶过程中以优势菌的形式出现在黏膜表面。我们首先用显微操作仪从断奶小鼠的肠道中分离出约5000个SFB细胞，用滚环扩增的方法扩增出它们的基因组DNA，并尝试用随机鸟枪测序的方法进行测序。然而，我们发现，质粒DNA是优先扩增的，因此，用这种方法很难确定整个基因组序列。因此，我们在诺生菌小鼠体内培养了SFB细胞，并制备了它们的DNA用于基因组测序。然后我们构建了一个随机的SFB散弹枪文库，并产生了大约50,000个散弹枪阅读。来自其他细菌物种的DNA污染严重干扰了我们的分析，但通过基于PCR的缺口闭合，我们获得了50个超重叠群。在最后的整理过程中，我们重新分离了SFB细胞，并制备了BAC文库。目前，我们正在确定1200个BAC克隆的末端序列，以进行最终的整理过程。我们在这项研究中采用的方法是一种新的策略，用于肠道微生物区系中不可培养细菌的基因组分析。

The final goal of our research is to clarify the compositional and genomic features of gut microbiota and their positional differences. However, it is impossible to know the positional differences by metagenomic analysis of fecal samples. We need to analyze the samples that are obtained directly from intestine, but such sampling in human is accompanied by many technical and ethical problems. Therefore, the mouse intestine was used as a model system in our study. In the present research project, we focused on the microbiota residing on the mucosal surface, which are obviously playing most important roles in term of host-resident microbe interaction, particularly on genome sequence determination of segmented filamentous bacterium (SFB). SFB is an unculturable bacterium which appears on the mucosal surface as a dominant bacterial species during the period of weaning.We first isolated about 5,000 SFB cells from the intestine of weanling mice by a micromanipulator, amplified their genomic DNA by rolling circle amplification, and tried to determine the sequence by the random shotgun sequencing method. However, we found that plasmid DNA was preferentially amplified, and thus, it is very hard to determine the whole genome sequence by this strategy. Therefore, we cultivated SFB cells in gnotobiotic mice, and prepare their DNA for genome sequencing. We then constructed a random shotgun library of SFB, and generated about 50,000 shotgun reads. Contamination of DNA derived from other bacterial species severely disturbed our analysis, but by PCR-based gap closing, we have obtained 50 super-contigs. For the final finishing process, we have re-isolated SFB cells and prepared a BAC library. Currently, we are determining the end-sequences of 1,200 BAC clones for the final finishing process. The approach we employed in this study is a new strategy for the genome analysis of unculturable bacteria in gut microbiota.