基于Methylovorus sp. J1-1基因组尺度代谢网络优化吡咯喹啉醌合成

doi:10.13560/j.cnki.biotech.bull.1985.2021-0345

摘要/Abstract

摘要：

通过构建Methylovorus sp. J1-1基因组尺度代谢网络模型（genome scale of metabolic network model,GSMM）,发掘能够提升吡咯喹啉醌（pyrroloquinoline quinone,PQQ）产量的发酵策略和相关靶基因。基于其注释的基因组和生化信息,构建J1-1的GSMM。再通过COBRApy预测可能使PQQ产量提高的氨基酸和潜在的靶基因并进行验证。构建了Methylovorus sp. J1-1的GSMM模型iKH584,共有584个基因,779个生化反应,121个交换反应与765个代谢产物,且可以用于后续模拟。根据iKH584模拟,外源添加谷氨酸、谷氨酰胺和脯氨酸、过表达glyA与hps1以及敲除hps2基因,均具有促进PQQ合成效果。结果表明：添加谷氨酸与脯氨酸时,PQQ的产量分别提高了10.4%与22.9%;过表达基因glyA与hps1,PQQ产量分别提高20.6%与14.6%;敲除基因hps2,PQQ产量最高可达140.84 mg/L,提高了8.0%,这与模拟结果基本一致,表明构建的模型基本正确。最后,在5 L发酵罐进行J1-1△hps2的分批补料发酵,PQQ的产量为812.64 mg/L,比亲本菌株提高了11.1%。建立的模型可以用来指导J1-1的发酵和菌株改造,提高PQQ产量。

关键词: Methylovorus sp. J1-1, 基因组尺度代谢网络模型, 吡咯喹啉醌, 氨基酸, 靶基因预测

Abstract:

The aim of this study is to explore the fermentation strategies and related target genes that can increase PQQ production by constructing genome-scale metabolic model of Methylovorus sp. J1-1. A genome-scale metabolic model（GSMM）of Methylovorus sp. J1-1 was constructed based on its annotated genome and biochemical information. Subsequently,amino acids and several potential genetic targets that may increase PQQ yield were predicted by COBRApy and validated. The GSMM iKH584 of J1-1 was constructed,containing 584 genes,779 biochemical reactions,121 exchange reactions and 765 metabolites,and it may be used in follow-up simulation. According to the iKH584 simulation,addition of glutamic acid,glutamine and proline,overexpression of the glyA and hps1,and knockout of hps2 all promoted PQQ synthesis. The results showed that the addition of glutamic acid and proline increased PQQ production by 10.4% and 22.9% respectively,overexpression of the glyA and hps1 increased the extracellular concentration of PQQ by 20.6% and 14.6% respectively,and hps2 knockout enhanced PQQ concentration up to 140.84 mg/L and increased by 8.0%,which were consistent with the simulated results,indicating that the metabolic model iKH584 of J1-1 was basically correct. Finally,fed-batch fermentations of J1-1△hps2 in 5 L fermentor were carried out,and the PQQ product reached to 812.64 mg/L,improved 11.1% compared with the parent strain J1-1. Conclusively,the established metabolic model iKH584 of J1-1 can be used to guide the fermentation and strain modification of J1-1 for improving the PQQ yield.

Key words: Methylovorus sp. J1-1, genome-scale metabolic model, pyrroloquinoline quinone, amino acids, target prediction

寇航, 王艳梅, 李彤, 薄明井, 张惟材, 熊向华, 黎明. 基于Methylovorus sp. J1-1基因组尺度代谢网络优化吡咯喹啉醌合成[J]. 生物技术通报, 2022, 38(2): 173-183.

KOU Hang, WANG Yan-mei, LI Tong, BO Ming-jing, ZHANG Wei-cai, XIONG Xiang-hua, LI Ming. Fermentation Optimization for PQQ Synthesis Based on the Genome-scale Metabolic Model of Methylovorus sp. J1-1[J]. Biotechnology Bulletin, 2022, 38(2): 173-183.

图/表 15

表1 本实验所用引物

Table 1 Primers used in this study

引物名称Primer name	引物序列Primer sequence（5'-3'）	引物功能Primer function
glyA-F	GCCTGCAGGTCGACTCTAGATATAGCGCTCAACAAGGACCTC	glyA基因扩增 Gene amplification of glyA
glyA-R	GTGAATTCGAGCTCGGTACCTTAAGCGCCGTATACCGGG	glyA基因扩增 Gene amplification of glyA
hps1-F	GCCTGCAGGTCGACTCTAGATATTGCTGACTTGAATAGCGCTAAT	hps1基因扩增 Gene amplification of hps1
hps1-R	GTGAATTCGAGCTCGGTACCTTAGTGAGCCAGCGAAGTGA	hps1基因扩增 Gene amplification of hps1
hps2-up-F	AGAAAAGATCAAAGGATCTTCGGATCCTGCGGAATTCCTTTTGTGG	hps2基因敲除质粒构建 Construction of hps2 knockout plasmid
hps2-up-R	CACTATAGGGCGAATTGCTCGAGTGAAGTGGCACAGAACCAG
hps2-kan-F	CCCACAAAAGGAATTCCGCAGGATCCGAAGATCCTTTGATCTTTTC
hps2-kan-R	TAAAGCTCAGCGGGGCTGATACTAGTCAGGTGGCACTTTTCGG
hps2-down-F	TTCCCCGAAAAGTGCCACCTGACTAGTATCAGCCCCGCTGAGC
hps2-down-R	GAGCTCCACCGCGGTGGCGGCCGCATTGCCTTGCTGGGGG
hps2-CF	GCACCATGGGCCAGTCTGA	hps2敲除菌株鉴定 Identifying the strain with hps2 knockout
hps2-CR	GGAGCACTTTCACCAGAAGCT	hps2敲除菌株鉴定 Identifying the strain with hps2 knockout
glyA-Y-F	AGTGGATGCTGGTGCCAACAT	实时荧光定量PCR qRT-PCR
glyA-Y-R	GCCGCTGGAAGTCGAACCTT
hps1-Y-F	ATCGGTTCGCCAGCCATCAC
hps1-Y-R	GCCTTGGTCGCAGCAATCAC
fdh-Y-F	AGCTGCACGTACCAAGGAAGTT
fdh-Y-R	TGTTGGCACCAGCATCCACTAC
fae-Y-F	GCCGCTGGAAGTCGAACCTT
fae-Y-R	GCCACGCAGGTTGATCCACTT

图1 基因组代谢网络的反应分布 AM（氨基酸代谢）;BOSM（其他次级代谢）;CM（碳水化合物代谢）;EM（能量代谢）;GBM（糖合成代谢）;LM（脂质代谢）;CV（辅因子和维生素代谢）;OAM（其他氨基酸代谢）;NM（核苷酸代谢）;TP（萜类和多酮类代谢）;TS（运输系统）;UM（未分类反应）

Fig. 1 Distributions of reactions in GSMM AM（Amino Acids Metabolism）,BOSM（Biosynthesis of Other Secondary Metabolism）,CM（Carbohydrates Metabolism）,EM（Energy Metabolism）,GBM（Glycan Synthesis and Metabolism）,LM（Lipids Metabolism）,CV（Cofactors and Vitamins Metabolism）,OAM（Other Amino Acid Metabolism）,NM（Nucl-eotides Metabolism）,TP（Terpenoids and Polyketides）,TS（Transport system）,and UM（Unclassified Reactions）

表2 J1-1在不同碳源和氮源上的生长模拟

Table 2 Simulation of growing in carbon and nitrogen sources

底物Substrate	实验值In vivo	模拟值In silico
碳源 Carbon source
葡萄糖 Glucose	+	+
甲醇 Methanol	+	+
乙醇 Ethanol	+	+
果糖 Fructose	+	+
氮源 Nitrogen source
尿素 Urea	+	+
硝酸钠 Sodium nitrate	+	+
硝酸铵 Ammonium nitrate	+	+
氯化铵 Ammonium chloride	+	+
硫酸铵 Ammonium sulfate	+	+

图2 模拟不同氨基酸添加对菌体生长和PQQ合成的影响

Fig. 2 Influence of simulated different amino acid additions on growth rate and PQQ synthesis

表3 氨基酸添加对菌体生长和PQQ合成的影响

Table 3 Effects of amino acid addition on cell growth and PQQ synthesis

氨基酸Amino acids	OD₆₀₀	PQQ/（mg·L^-1）
对照 Control	2.57±0.21	106.18±2.35
谷氨酸 Glutamic acid	2.94±0.08	117.15±0.56
谷氨酰胺 Glutamine	2.69±0.18	106.26±1.25
脯氨酸 Proline	2.91±0.06	130.46±0.56

图3 单基因敲除和过表达靶基因预测

Fig. 3 Prediction of single gene knockout and overexpres-sed target genes

图4 pCM66-glyA与pCM66-hps1菌液PCR鉴定 M：Trans 2k plus II DNA marker;1-2：pCM66- hps1转化子;5：pCM66-gl- yA转化子;3,6：阳性对照;4,7：阴性对照

Fig. 4 Identification of pCM66-glyA and pCM66-hps1 strain culture liquid by PCR M：Trans 2k plus II DNA marker.1-2：pCM66- hps1 transformants. 5：pCM66- glyA transformants. 3 and 6：Positive control. 4 and 7：Negative control

图5 各个菌株对数期基因表达比较 *P<0.05;**P<0.01;***P<0.001. The same below

Fig. 5 Comparison of gene expression in each strain during exponential phase

表4 过表达菌株表型分析

Table 4 Phenotype analysis of each overexpression strain

菌株Strain	OD₆₀₀	PQQ/（mg·L^-1）
J1-1/pCM66	2.63±0.15	82.51±1.56
J1-1/pCM66-glyA	2.86±0.18	99.48±2.15
J1-1/pCM66- hps1	2.91±0.09	94.58±3.14

图6 J1-1△hps2菌株的PCR验证 M：Trans 2K plus II DNA marker;1：hps2敲除菌-Up-kan-片段PCR验证;2：J1-1菌株Up-kan-片段PCR验证;3：hps2敲除菌-kan-Down-片段PCR验证;4：J1-1菌株kan-Down-片段PCR验证;5：阴性对照

Fig.6 PCR verification of J1-1△hps2 strain M：Trans 2K plus II DNA marker. 1：pCR verification of hps2 with Up-kan-fragment knockout. 2：Up-kan-fragment PCR verification of J1-1 strain. 3：PCR verification of hps2 strain with kan-Down-fragment knowckout. 4：kan-Down-fragment PCR validation of J1-1 strain. 5：Negative control

表5 hps2基因敲除菌株表型分析

Table 5 Phenotype analysis of hps2 gene knockout strain

菌株Strain	OD₆₀₀	PQQ/（mg·L^-1）
J1-1	3.15±0.21	130.45±2.35
J1-1∆hps2	3.46±0.08	140.84±3.41

图7 hps2敲除菌5 L发酵的实验结果

Fig.7 Experimental results of hps2-knockout strain in 5 L fermenter

图8 hps2敲除菌关键基因转录水平的变化

Fig.8 Transcription level of key genes in hps2 - knockout strain

图9 J1-1菌株甲醇核心代谢途径

Fig. 9 Methanol core metabolic pathways of J1-1 strain

图10 导入外源质粒对J1-1生长及PQQ合成的影响

Fig. 10 Influence of foreign plasmid on the cell growth of J1-1 strain and PQQ synthesis

参考文献 37

[1]	Hauge JG. Glucose dehydrogenase of bacterium anitratum:an enzyme with a novel prosthetic group[J]. J Biol Chem, 1964, 239(11):3630-3639. doi: 10.1016/S0021-9258(18)91183-X URL
[2]	Nakano M, Kamimura A, Watanabe F, et al. Effects of orally administered pyrroloquinoline quinone disodium salt on dry skin conditions in mice and healthy female subjects[J]. J Nutr Sci Vitaminol:Tokyo, 2015, 61(3):241-246. doi: 10.3177/jnsv.61.241 URL
[3]	Shankar BS, Pandey R, Amin P, et al. Role of glutathione in augmenting the anticancer activity of pyrroloquinoline quinone(PQQ)[J]. Redox Rep, 2010, 15(4):146-154. doi: 10.1179/174329210X12650506623762 URL
[4]	Rucker R, Storms D, Sheets A, et al. Is pyrroloquinoline quinone a vitamin?[J]. Nature, 2005, 433(7025):10-11. doi: 10.1038/433010a URL
[5]	Klinman JP, Mu D. Quinoenzymes in biology[J]. Annu Rev Biochem, 1994, 63(1):299-344. doi: 10.1146/biochem.1994.63.issue-1 URL
[6]	柯崇榕. 适应性驯化选育高产吡咯喹啉醌的生丝微菌突变株[J]. 生物工程学报, 2020, 36(1):152-161.
	Ke CR. Breeding of Hyphomicrobium denitrificans for high production of pyrroloquinoline quinone by adaptive directed domestication[J]. Chin J Biotechnol, 2020, 36(1):152-161.
[7]	Urakami T, Yashima K, Kobayashi H, et al. Production of pyrroloquinoline quinone by using methanol-utilizing bacteria[J]. Appl Environ Microbiol, 1992, 58(12):3970-3976. doi: 10.1128/aem.58.12.3970-3976.1992 URL
[8]	郑玲辉, 杜敏娜一种生丝微菌和吡咯喹啉醌的制备方法：中国, 106282044A[P], 2015-05-20.
	Zheng LH, Du MN. A kind of Hyphomicrobium sp. and method for the preparation of pyrroloquinoline quinone：CN, 106282044A[P], 2015-05-20.
[9]	Shi W, Sun Q, Fan G, et al. gcType:a high-quality type strain genome database for microbial phylogenetic and functional research[J]. Nucleic Acids Res, 2021, 49(1):694-705.
[10]	Thiele I, et al. A protocol for generating a high-quality genome-scalemetabolic reconstruction[J]. Nat Protoc, 2010(1):93-121.
[11]	Monk JM, et al. iML1515, a knowledgebase that computes Esche-richia coli traits[J]. Nat Biotechnol, 2017(10):904-908.
[12]	Brochado AR, Patil KR. Overexpression of O-methyltransferase leads to improved vanillin production in baker’s yeast only when complemented with model-guided network engineering[J]. Biotechnol Bioeng, 2013, 110(2):656-659. doi: 10.1002/bit.24731 pmid: 23007522
[13]	Huang D, Li S, Xia M, et al. Genome-scale metabolic network guided engineering of Streptomyces tsukubaensis for FK506 production improvement[J]. Microb Cell Fact, 2013, 12:52. doi: 10.1186/1475-2859-12-52 pmid: 23705993
[14]	王歆, 汪建华, 刘党生, 等. 吡咯喹啉醌产生菌筛选方法建立及菌种筛选[J]. 微生物学报, 2007, 47(6):982-986.
	Wang X, Wang JH, Liu DS, et al. Establishment of the screening method and isolation of PQQ producing strains[J]. Acta Microbiol Sin, 2007, 47(6):982-986.
[15]	Ebrahim A, Lerman JA, Palsson BO, et al. COBRApy:COnstraints-based reconstruction and analysis for Python[J]. BMC Syst Biol, 2013, 7:74. doi: 10.1186/1752-0509-7-74 pmid: 23927696
[16]	Orth JD, Thiele I, Palsson BØ. What is flux balance analysis?[J]. Nat Biotechnol, 2010, 28(3):245-248. doi: 10.1038/nbt.1614 URL
[17]	Choon YW, Mohamad MS, Deris S, et al. Differential Bees Flux Balance Analysis with OptKnock for in silico microbial strains optimization[J]. PLoS One, 2014, 9(7):e102744. doi: 10.1371/journal.pone.0102744 URL
[18]	Razmilic V, Castro JF, Andrews B, et al. Analysis of metabolic networks of Streptomyces leeuwenhoekii C34 by means of a genome scale model:Prediction of modifications that enhance the production of specialized metabolites[J]. Biotechnol Bioeng, 2018, 115(7):1815-1828. doi: 10.1002/bit.26598 pmid: 29578590
[19]	李大攀, 等. 甲基营养菌MP688甲醇脱氢酶基因mpq1818的敲除及功能研究[J]. 生物技术通讯, 2014(5):632-635.
	Li DP, et al. Knockout and Characterization of mpq1818 Gene of Methylovorus sp.[J]. Lett Biotechnol, 2014(5):632-635.
[20]	Xiong XH, et al. Complete genome sequence of the bacterium Methylovorus sp. strain MP688, a high-level producer of pyrroloquinolone quinone[J]. J Bacteriol, 2011, 193(4):1012-1013. doi: 10.1128/JB.01431-10 URL
[21]	Moriya Y, Itoh M, Okuda S, et al. KAAS:an automatic genome annotation and pathway reconstruction server[J]. Nucleic Acids Res, 2007, 35(web server issue):W182-W185. doi: 10.1093/nar/gkm321 URL
[22]	Kanehisa M, et al. KEGG:integrating viruses and cellular organisms[J]. Nucleic Acids Res, 2021, 49(d1):D545-D551. doi: 10.1093/nar/gkaa970 URL
[23]	King ZA, Lu J, Dräger A, et al. BiGG Models:a platform for integrating, standardizing and sharing genome-scale models[J]. Nucleic Acids Res, 2016, 44(d1):D515-D522. doi: 10.1093/nar/gkv1049 URL
[24]	Bairoch A. The ENZYME database in 2000[J]. Nucleic Acids Res, 2000, 28(1):304-305. pmid: 10592255
[25]	Consortium U. UniProt:a worldwide hub of protein knowledge[J]. Nucleic Acids Res, 2019, 47(D1):D506-D515. doi: 10.1093/nar/gky1049 URL
[26]	Yang J, Zhang CT, Yuan XJ, et al. Metabolic engineering of Methylobacterium extorquens AM1 for the production of butadiene precursor[J]. Microb Cell Fact, 2018, 17(1):194. doi: 10.1186/s12934-018-1042-4 URL
[27]	Lapidus A, Clum A, Labutti K, et al. Genomes of three methylotrophs from a single niche reveal the genetic and metabolic divergence of the Methylophilaceae[J]. J Bacteriol, 2011, 193(15):3757-3764. doi: 10.1128/JB.00404-11 pmid: 21622745
[28]	Kim B, Kim WJ, Kim DI, et al. Applications of genome-scale metabolic network model in metabolic engineering[J]. J Ind Microbiol Biotechnol, 2015, 42(3):339-348. doi: 10.1007/s10295-014-1554-9 URL
[29]	Saier MH Jr, Reddy VS, et al. The transporter classification database[J]. Nucl Acids Res, 2014, 42(D1):D251-D258. doi: 10.1093/nar/gkt1097 URL
[30]	Feist AM, Palsson BØ. The growing scope of applications of genome-scale metabolic reconstructions using Escherichia coli[J]. Nat Biotechnol, 2008, 26(6):659-667. doi: 10.1038/nbt1401 pmid: 18536691
[31]	Latendresse M. Efficiently gap-filling reaction networks[J]. BMC Bioinformatics, 2014, 15:225. doi: 10.1186/1471-2105-15-225 pmid: 24972703
[32]	Ge X, Wang W, Du B, et al. Multiple pqqA genes respond differently to environment and one contributes dominantly to pyrroloquinoline quinone synjournal[J]. J Basic Microbiol, 2015, 55(3):312-323. doi: 10.1002/jobm.v55.3 URL
[33]	Kim J, Coradetti ST, Kim YM, et al. Multi-omics driven metabolic network reconstruction and analysis of lignocellulosic carbon utilization in Rhodosporidium toruloides[J]. Front Bioeng Biotechnol, 2020, 8:612832. doi: 10.3389/fbioe.2020.612832 URL
[34]	Müller JE, Litsanov B, Bortfeld-Miller M, et al. Proteomic analysis of the thermophilic methylotroph Bacillus methanolicus MGA3[J]. Proteomics, 2014, 14(6):725-737. doi: 10.1002/pmic.201300515 pmid: 24452867
[35]	Ochsner AM, Sonntag F, Buchhaupt M, et al. Methylobacterium extorquens:methylotrophy and biotechnological applications[J]. Appl Microbiol Biotechnol, 2015, 99(2):517-534. doi: 10.1007/s00253-014-6240-3 pmid: 25432674
[36]	van Dien SJ, et al. Stoichiometric model for evaluating the metabolic capabilities of the facultative methylotroph Methylobac-terium extorquens AM1, with application to reconstruction of C3 and C4 metabolism[J]. Biotechnol Bioeng, 2002(3):296-312.
[37]	Peyraud R, Schneider K, et al. Genome-scale reconstruction and system level investigation of the metabolic network of Methylobacterium extorquens AM1[J]. BMC Syst Biol, 2011, 5:189. doi: 10.1186/1752-0509-5-189 pmid: 22074569