| 摘要: | 本研究選用嗜高溫古生菌Aeropyrum pernix K1，鑑定高溫環境中具有熱穩定性之外泌蛋白質並探討其功能。此微生物為1993年於日本西南方Kodakara-Jima島沿岸的硫磺噴口處篩選而得，是一株絕對好氧並嗜生長在超高溫環境的古生菌，其最適生長條件為絕對好氧、90~95oC、pH 7.0、鹽度3.5%；目前已知其他與A. pernix K1同科(Desulfurococcaceae)之古生菌皆為絕對厭氧，並以醱酵或無氧呼吸獲得能量，因之，A. pernix K1的能量運用、生理代謝及其酵素特性與應用潛力，因其獨特性而具有研究價值。 日本研究團隊Yamazaki等人已於2006年發表A. pernix K1胞內蛋白體之研究，本研究則探討其分泌蛋白體。將A. pernix K1培養於最適生長溫度(90oC)之JXTm培養液，並於對數期收集菌液，去除菌體後，經充氮氣正壓式過濾濃縮後可得到分泌到細胞外蛋白質體(分泌體)，經由三種蛋白質鑑定方法：(1)一維或(2)二維電泳分離蛋白質並經膠體內胰蛋白酶消化和(3)直接將分泌體消化水解，以電噴霧四極桿飛行時間質譜儀分析胺基酸序列，可鑑定出在高溫具穩定性之分泌性蛋白質。 本研究共鑑定到在對數期生長時分泌體中的136個熱穩定蛋白質，依生物資訊方法進行次細胞位置分類與預測，確認17個為分泌性蛋白質：如耐90oC之水解酵素subtilisin-like protease (pernisine) (APE0263.1)、surface layer-associated protease precursor (APE1213.1)和putative homocysteine methyltransferase (APE2050.1)，及多種能夠與amino acid (APE0917, APE2521.1)、oligopeptide (APE2257, APE1583.1, APE0945.1)、phosphate (APE0045)或其他基質(APE1630.1, APE1893.1, APE2254.1, APE1390.1)結合的binding proteins。這些分泌蛋白質中，水解酵素扮演分解、利用胞外(培養基)的營養物質的角色，而binding proteins扮演結合和運輸的重要功能，將營養物質送至細胞內，在細胞生長、代謝的過程具有重要的生理意義。 本研究鑑定到的136個蛋白質中，有131個屬於A. pernix K1，5個屬於其他古生菌：Caldivirga maquilingensis IC-167之膜蛋白質alpha-glucosidase (Cmaq1692)與胞內蛋白質hypothetical protein Cmaq0545 (Cmaq0545)、Methanococcus maripaludis C7之DNA-directed RNA polymerase (MmarC7_0607)、Methanococcus maripaludis S2之intermediate filament protein (MMP0676)、Aciduliprofundum boonei T469之NAD+ synthetase (ABOONEI1219)，上述物種中僅C. maquilingensis IC-167與A. pernix K1親源關係較接近。此5個蛋白質被鑑定屬於他種古生菌，其中2個基因(MmarC7_0607和MMP0676)，其起始密碼分別為TTG和TTA，皆轉譯為Leucine而非傳統的Methionine；而另外3個基因雖以ATG為起始密碼，但在先前四篇A. pernix K1以生物資訊方法的gene annotation文獻中，並不認定其為ORFs，原因尚有待分析，但本研究顯示，濕式研究能直接探索到特定條件下表達的基因產物，可補足乾式實驗的不足。因為在136個蛋白質中有5個為之前未發現的基因產物，推測在A. pernix K1基因體中應還有為數尚多的基因及其產物待進一步探索。 在次細胞位置的預測方面，我們分析本研究所鑑定之蛋白質，與前人所發表之A. pernix K1胞內蛋白質，評估四種次細胞預測工具(CELLO、PSORTb、Proteome Analyst和PRED-SIGNAL)之準確性與效能，其中，以古生菌蛋白質為training data的PRED-SIGNAL具有較佳的效能(Precision 95.9%, Recall 0.78)，其次為以細菌蛋白質為training data的CELLO (Precision 94.4%, Recall 0.78)；若針對胞外蛋白質而言，則僅PRED-SIGNAL之效能較好。未來期望能以更多的胞外蛋白質實驗數據，再對此四種預測工具做更客觀的評比，提出修正胞外蛋白質之預測工具。 In this study, Aeropyrum pernix K1 is chose to study its thermostable secretomes. Aeropyrum pernix K1 was found in the coastal solfataric thermal vent of Kodakara-Jima Island, south west of Japan in 1993. This archaeon is aerobic, living in a hyperthermophilic condition. It grows in the range of 90~95 oC, pH 7.0, and salinity level at 3.5%. Unlike A. pernix K1, other archaea are anaerobic, and uses fermentation and anaerobic respiration for energy; thus, the energy utilization, physiological metabolism, and enzyme characteristics of A. pernix K1 has a great value in microbiology research. Yamazaki et al. published the research achievements about the cytosolic proteome from A. pernix K1in 2006. In our study, we investigate the thermostable proteins secreted by A. pernix K1. It is grown at 90 oC in JXTm medium. The archaeon is harvested at log phase, after removing of the cell, the remaining the supernatant is filtered using ultrafiltration, thus secretome is obtained. Three proteomic methods are performed: (1) 1D SDS-PAGE, or (2) 2D SDS-PAGE followed by in-gel digestion, and (3) in-solution digestion. Finally, ESI Q-TOF MS/MS is used to analyze the amino acid sequence and its high-throughput is able to identify thermostable hyperthermophilic proteins. According to Pfam functional search and subcellular localization prediction, we identified a total of 136 thermostable proteins that harvested at log phase and confirmed 17 of secreted proteins are found to be subtilisin-like protease (pernisine) (APE0263.1) that can endure 90oC, surface layer-associated protease precursor (APE1213.1), putative homocysteine methyltransferase (APE2050.1) and binding proteins that can bind amino acid (APE0917, APE2521.1), oligopeptide (APE2257, APE1583.1, APE0945.1), phosphate (APE0045) or other substrates (APE1630.1, APE1893.1, APE2254.1, APE1390.1). Among these secreted proteins, hydrolase is able to hydrolyze and utilize the medium for nutrients, and the binding proteins play an important role in binding and transport, sending nutrients to the cytosol and physiological functions such as cell growth and metabolic processes. In our research, 136 were identified as thermostable proteins with 131 of them belongs to A. pernix K1 and five belongs to other archaea, such as the membrane protein alpha-glucosidase (Cmaq1692) and cytosolic protein hypothetical protein Cmaq0545 of Caldivirga maquilingensis IC-167, the DNA-directed RNA polymerase (MmarC7_0607) of Methanococcus maripaludis C7, the intermediate filament protein (MMP0676) of Methanococcus maripaludis S2 and NAD+ synthetase (ABOONEI1219) of Aciduliprofundum boonei T469. From the above mentioned organisms, C. maquilingensis IC-167 is closest related to A. pernix K1. There are five proteins belong to other arachaea. Among the 2 genes (MmarC7_0607 and MMP0676), the start codons are TTG and TTA, respectively. Their translated proteins are Leucine instead of the traditional Methionine; although the other 3 genes start with ATG, but the previous A. pernix K1 literatures using bioinformatics as a tool were shown that the gene annotation was not the ORFs. The questions are still waiting to be answered, but it has shown that the wet lab can directly explore the expression of genes under certain conditions, thus fulfills the insufficiency of the dry lab. Since 5 of the 136 genes were not found in the gene product, suggesting that there are a few more genes and products that needed to be explored further. In predicting the subcellular localization, we compared the identified proteins coupled with the published literature, we exam the precision and efficiency of the four cellular prediction tools (CELLO, PSORTb, Proteome Analyst and PRED-SIGNAL) and found that using arachaea as a training data, PRED-SIGNAL has the highest efficiency (Precision 95.9%, Recall 0.78), follow by using bacteria proteins as a training data in CELLO (Precision 94.4%, Recall 0.78); however, when testing only the extracellular proteins, only PRED-SIGNAL can give a better result. Hopefully we will be able to obtain more extracellular proteins in the lab, then compare the four prediction tools in order to correct any misperceptions about extracellular proteins. |
