牛樟芝萜類化合物生合成之系統生物學研究

Systems Biology Research on Terpenoids Biosynthesis in Antrodia cinnamomea

林彥良(Yan-Liang Lin)

指導教授 ： 曲芳華

國立臺灣大學/生物資源暨農學院/森林環境暨資源學研究所/博士(2015年)

https://doi.org/10.6342/NTU.2015.00966

摘要

牛樟芝是臺灣重要的藥用真菌，所具有之多樣且特殊的代謝產物提供牛樟芝在醫藥發展上莫大的潛力，其中萜類化合物為牛樟芝代謝產物中重要生物活性的來源之一，人工培養之菌絲體難以誘導子實體生成且缺乏多種子實體所特有之代謝產物，為克服牛樟芝在人工培養上所面臨的難處，解答菌絲體與子實體之差異，本論文結合不同體學的研究方法針對牛樟芝萜類生合成進行系統生物學的研究。
利用蛋白質二維膠體電泳方法比較牛樟芝菌絲體品系B506與野生型子實體蛋白質體學上的差異，分別從菌絲體與子實體中定位出606與514個蛋白質點，並挑選60個只存在於子實體之蛋白質點進行質譜定序，經由序列比對後雖然沒有找到和萜類生合成相關之蛋白質，但也從中發現牛樟芝子實體與生理代謝、逆境反應、細胞結構等重要的蛋白質。
為了建立牛樟芝重要的遺傳訊息，利用次世代定序的方式針對牛樟芝菌絲體單核品系S28進行染色體核酸的定序，推測牛樟芝的染色體大小為45.58 Mb，定序覆蓋率約為32×，經由組裝後共獲得1,242條scaffolds總長度為27,717,145 bp，包含6,522個預測的基因。針對牛樟芝的液態培養菌絲體品系B479與野生型子實體利用次世代定序進行轉錄體的定序，分別獲得19,875,544及25,296,502條clean reads，混合組裝後共得到13,109條unigenes，平均長度1,615 bp，N50為2,770 bp，總長度為21,174,312 bp，其中7,851條具有功能註解，經由差異性分析可得到共2,282個基因在牛樟芝菌絲體與子實體有不同的表現程度。針對牛樟芝的液態培養菌絲體品系WSY-01與野生型子實體利用次世代定序進行微核糖核酸的定序，分別獲得10,879,926及18,291,833條clean reads，經由後續分析與篩選總共獲得4條具有保守性與63條牛樟芝全新未註解的微核糖核酸序列，微核糖核酸長度分布最多為23 nt與偏好5′ uracil，經由微核糖核酸目標基因的預測，可能參與在調控三萜類生合成途徑與其他細胞分化重要的相關基因。
在三萜類生合成途徑的研究中，本論文鑑定了一個參與在生合成途經中重要的關鍵環化酵素AcOSC，利用功能缺失之酵母菌進行異源蛋白質表現並配合GC/MS驗證其功能為lanosterol synthase，隨後利用農桿菌感染法針對牛樟芝菌絲體進行轉基因實驗，經過篩選後獲得多個牛樟芝AcOSC轉基因品系並命名為ACT，並利用LC-MS/MS、HPLC與NMR等代謝體學的分析方式分析牛樟芝菌絲體轉殖株，藉由轉基因的方式提昇AcOSC基因而得到提升下游三萜類產物(dehydrosulphurenic acid與dehydroeburicoic acid)最高達3倍之多。
藉由次世代定序所建立之轉錄體基因資料庫，經由保守性序列的比對搜尋找到12個預測的萜類生合成酵素，利用大腸桿菌進行異源重組蛋白質表現並搭配GC/MS進行活性分析，最終有7個重組蛋白質具有萜類生合成的活性並可以產生一個或是多種的單、倍半萜類產物，並加以命名為AcTPS，同時結合了SPME吸附GC/MS來觀察菌絲體與子實體之萜類產物與萜類基因的相互關係，發現仍有多種難以鑑定的萜類產物，據此也將AcTPS序列與牛樟芝單核品系S28之基因體序列進行比對，尋找相關的基因叢集，推測尚有多種未知的合成酵素或是後修飾酵素參與在萜類的生合成途徑中。
最後本論文利用次世代定序所獲得之轉錄體基因資料庫相關資料設計了牛樟芝的客製化微矩陣列晶片，並進行牛樟芝四個特殊發育時期的基因表現分析(AL：液態培養菌絲體、EAP：固態培養轉化為子實體前之菌絲體、AF：固態培養之子實體、AT：野生型子實體)，將所得數據正規化後進行ANOVA分析、表現倍率之篩選與相互關係之計算，進而建立牛樟芝的基因相互關係網絡圖，從中找到許多參與在萜類生合成途徑中的重要基因如：acetyl-CoA acetyltransferase、HMG-CoA reductase、diphosphomevalonate decarboxylase、isopentenyl pyrophosphate isomerase、AcTPS3、AcTPS5、cytochrome p450與AcCYP51，而調控因子如：transcription factor PacC、transcription factor IIIC、TFIIH basal transcription factor、transcription initiation factor IIF、C2H2 transcription factor、transcription factor Gf. BMR1與jumonji superfamily，雖然本論文經由基因網絡圖的建立，找到許多與萜類生合成相關的調控因子與相關基因，但是仍然有許多的基因尚未加以討論與驗證，而本論文所獲得之所有資料實為未來牛樟芝在進一步生產特定二次代謝產物與型態轉化之重要的參考。

關鍵字

牛樟芝 ； 萜類 ； 生合成 ； 次世代定序 ； 子實體 ； 菌絲體

並列摘要

Antrodia cinnamomea is a precious, host-specific brown-rot fungus that has been used as a folk medicine in Taiwan for centuries is known to have diverse bioactive compounds with potent pharmacological activity. Terpenoids one of the most important secondary metabolites contribute the bioactivity of A. cinnamomea. Artificial culture mycelium is hard to transform into fruiting body and deficient in certain bioactive compounds. Systems biology approach was use to study the terpenoids biosynthesis in A. cinnamomea in order to improve the artificial culture codition and answer the different between mycelium and fruiting body.
Different fermentation states of A. cinnamomea (liquid cultured mycelium B506 and wild-type fruiting bodies) were analyzed with 2 dimentional polyacrylamide gel electrophoresis and found 606 and 514 protein spots in mycelium and fruiting body respectively. Although no protein spots were associated with terpenoid synthesis, several proteins involved in metabolism, cell rescue and ROS scavenging, cell structure and regulator specify expressed in fruiting body.
To establish the basic genetic information of A. cinnamomea, monokaryotic mycelium S28 was sequenced with next-generation sequencing technique (NGS). A predicted 45.58 Mb genome size with 32× sequencing coverage and 1,242 scaffolds with total length 27,717,145 bp contain 6,522 predicted genes was completed. Liquid cultured mycelium B479 and wild-type fruiting body’s transcriptome were also sequenced with NGS. 19,875,544 and 25,296,502 clean reads were obtained from mycelium and fruiting body respectively. De novo assembly generate 13,109 unigenes with average length 1,615 bp, N50 = 2,770 bp and total length is 21,174,312 bp. 7,851 out of 13,109 unigenes with gene annotation against to NCBI nr database. After differential expressed genes (DEGs) analysis, 2,282 genes are assigned to have differential expressed among mycelium and fruiting body. Liquid cultured mycelium WSY-01 and wild-type fruiting body were used to construct the small RNA library with NGS. 10,879,926 and 18,291,833 clean reads were obtained from mycelium and fruiting body respectively. Using a previously constructed pipeline to search for microRNAs (miRNAs), then identified 4 predicted conserved miRNA and 63 novel predicted miRNA-like small RNA (milRNAs) candidates with most abundant in 23 nt and prefer 5′ uracil. Target prediction revealed several proteins involved in triterpenoid synthesis, mating type recognition, chemical or physical sensory protein and transporters predicted to be regulated by the miRNAs and milRNAs.
An oxidosqualene cyclase (AcOSC) involved in triterpenoid compound synthesis pathway was identified. The lanosterol synthase function was characterized with heterologous protein expression system in a functional deficient yeast and GC/MS analysis. Agrobacteium-mediated transformation was used to generate the overexpression of AcOSC in A. cinnamomea artificial culture system. LC-MS/MS, HPLC and NMR were also used to identified the triterpenoid compound dehydrosulphurenic acid (DSA) and dehydroeburicoic acid (DEA) in transformed lines (called ACT) and found generate more DSA and DEA than un-transformed mycelium.
12 predicted terpene synthase genes were screened out from A. cinnamomea transcriptome library. After recombinant protein expression with Escherichia coli BL21 (DE3) and functional characterization with GC/MS. 7 recombinant proteins named with AcTPS had terpene synthase activity and generate one to multi products. With SPME absorption GC/MS analysis, several terpenoid compounds were identified in mycelium and fruiting body and correlated to AcTPS’s products and also found several un-identified terpenoid compounds. According to this, the AcTPS sequences are mapping back to the genomic sequence of A. cinnamomea isolate S28 to identify the gene clusters. Several unknown terpene synthases or modification enzymes may be involved in terpene synthesis pathway.
At last, the transcriptome data was used to design a custom micro-array chip to analyze gene expression at different developmental stages of A. cinnamomea, including liquid cultured mycelium (AL), mycelium before transform to fruiting body on agar plate (EAP), fruiting body on agar plate (AF), wild-type fruiting body (AT). The microarray data was analyzed with normalization, ANOVA analysis, fold-change filtering and correlation calculation to create a putative genetic relation network. Several important genes such as: acetyl-CoA acetyltransferase, HMG-CoA reductase, diphosphomevalonate decarboxylase, isopentenyl pyrophosphate isomerase, AcTPS3, AcTPS5, cytochrome p450 and AcCYP51 are involved in the genetic network. There also several regulatory factors such as: transcription factor PacC, transcription factor IIIC, TFIIH basal transcription factor, transcription initiation factor IIF, C2H2 transcription factor, transcription factor Gf. BMR1 and jumonji superfamily are involved. Although several terpene related genes and transcription factors are predicted to involved in the genetic network, further experiments are needed. This thesis provides several genetic information of A. cinnamomea for the basis of further secondary metabolites and fungal differentiation research.

並列關鍵字

Antrodia cinnamomea ； terpenoids ； biosynthesis ； next-generation sequencing (NGS) ； fruiting body ； mycelium

參考文獻

Chang, T.T., Chou, W.N., 2004. Antrodia cinnamomea reconsidered and A. salmonea sp. nov. on Cunninghamia konishii in Taiwan. Bot. Bull. Acad. Sin. 45, 347–352.

Abe, I., Prestwich, G.D., 1995. Molecular cloning, characterization, and functional expression of rat oxidosqualene cyclase cDNA. Proc. Natl. Acad. Sci. U. S. A. 92, 9274–9278.

Altschul, S.F., Gish, W., Miller, W., Myers, E.W., Lipman, D.J., 1990. Basic Local Alignment Search Tool. J Mol Biol. 215, 403–410.

Arnold, K., Bordoli, L., Kopp, J., Schwede, T., 2006. The SWISS-MODEL workspace: a web-based environment for protein structure homology modelling. Bioinformatics 22, 195–201.

Balciunas, D., Ronne, H., 2000. Evidence of domain swapping within the jumonji family of transcription factors. Trends Biochem. Sci. 25, 274–276.