三种鲑科鱼类比较基因组学分析鉴定出参与细胞内寄生菌鲑立克次体抗性的功能候选基因
Comparative Genomic Analysis of Three Salmonid Species Identifies Functional Candidate Genes Involved in Resistance to the Intracellular Bacterium Piscirickettsia salmonis
- 发表期刊: Frontiers in Genetics (2019) 10:665
- 出版机构: Frontiers Media S.A.
- DOI 链接: https://doi.org/10.3389/fgene.2019.00665
- 论文类型: Original Research (原始研究论著)
- 作者团队: José M. Yáñez1,2,, Grazyella M. Yoshida1, Ángel Parra1,3,4,5, Katharina Correa6, Agustín Barría1,7, Liane N. Bassini8, Kris A. Christensen9, Maria E. López1,10, Roberto Carvalheiro11,12, Jean P. Lhorente6, Rodrigo Pulgar3,
- 通讯作者: José M. Yáñez (jmayanez@uchile.cl), Rodrigo Pulgar (rpulgar@inta.uchile.cl)
- 作者单位:
- 智利大学 兽医与畜牧科学学院,智利圣地亚哥 (Facultad de Ciencias Veterinarias y Pecuarias, Universidad de Chile, Santiago, Chile)
- 智利千禧侵略生物学科学中心 (INVASAL),智利康塞普西翁 (Núcleo Milenio INVASAL, Concepción, Chile)
- 智利大学 营养与食品技术研究所 (INTA),智利圣地亚哥 (Instituto de Nutrición y Tecnología de los Alimentos, Universidad de Chile, Santiago, Chile)
- 水产养殖合作博士培养计划(智利大学、北部天主教大学、瓦尔帕莱索天主教大学),智利瓦尔帕莱索
- 北部天主教大学 海洋科学学院,智利科金博 (Facultad de Ciencias del Mar, Universidad Católica del Norte, Coquimbo, Chile)
- Benchmark Genetics Chile,智利蒙特港 (Benchmark Genetics Chile, Puerto Montt, Chile)
- 爱丁堡大学 罗斯林研究所与皇家(迪克)兽医研究学院,英国米德洛锡安 (The Roslin Institute, University of Edinburgh, Midlothian, UK)
- 安德烈斯·贝洛大学 生命科学学院兽医学院,智利圣地亚哥 (Escuela de Medicina Veterinaria, Universidad Andres Bello, Santiago, Chile)
- 加拿大渔业及海洋部,加拿大西温哥华 (Fisheries and Oceans Canada, West Vancouver, BC, Canada)
- 瑞典农业科学大学 动物育种与遗传系,瑞典乌普萨拉 (Department of Animal Breeding and Genetics, Swedish University of Agricultural Sciences, Uppsala, Sweden)
- 圣保罗州立大学 农业与兽医科学学院,巴西雅博蒂卡巴尔 (School of Agricultural and Veterinarian Sciences, São Paulo State University, Jaboticabal, Brazil)
- 巴西国家科学技术发展委员会 (CNPq),巴西巴西利亚 (CNPq, Brasília, Brazil)
- 学术编辑: Christos Palaiokostas (瑞典农业科学大学)
- 审稿专家: Alastair Hamilton (荷兰 Hendrix Genetics BV), Marieke Verleih (德国莱布尼茨家畜生物学研究所)
- 收稿日期: 2019-03-24 | 接受日期: 2019-06-25 | 在线发表: 2019-08-05
- 专栏专刊: 畜禽基因组学 (Livestock Genomics)
目录与快速导航 / Quick Navigation
- [术语对照总表 / Terminology Ledger](#术语对照总表 / Terminology Ledger)
- 公式索引
- [图表索引 / Figure & Table Index](#图表索引 / Figure & Table Index)
- [摘要与关键词 / Abstract & Keywords](#摘要与关键词 / Abstract & Keywords)
- [引言 / Introduction](#引言 / Introduction)
- [材料与方法 / Materials and Methods](#材料与方法 / Materials and Methods)
- [人工攻毒感染试验 / Challenge Tests](#人工攻毒感染试验 / Challenge Tests)
- [基因分型与质量控制 / Genotyping](#基因分型与质量控制 / Genotyping)
- [贝叶斯全基因组关联分析 / GWAS Analysis](#贝叶斯全基因组关联分析 / GWAS Analysis)
- [比较基因组学与候选基因优先级分类 / Comparative Genomic Analysis](#比较基因组学与候选基因优先级分类 / Comparative Genomic Analysis)
- [研究结果 / Results](#研究结果 / Results)
- [攻毒试验表型变异与遗传参数估计 / Challenge Test and Genetic Parameters](#攻毒试验表型变异与遗传参数估计 / Challenge Test and Genetic Parameters)
- [贝叶斯 GWAS 剖析遗传架构 / GWAS Analysis](#贝叶斯 GWAS 剖析遗传架构 / GWAS Analysis)
- [跨物种比较基因组学与核心候选基因挖掘 / Comparative Genomic Analysis](#跨物种比较基因组学与核心候选基因挖掘 / Comparative Genomic Analysis)
- [讨论 / Discussion](#讨论 / Discussion)
- [鲑立克次体抗性的遗传架构特征 / Genetic Architecture of Resistance](#鲑立克次体抗性的遗传架构特征 / Genetic Architecture of Resistance)
- [抗性相关候选蛋白质与分子功能机制 / Candidate Proteins Associated With Resistance](#抗性相关候选蛋白质与分子功能机制 / Candidate Proteins Associated With Resistance)
- [结论 / Conclusions](#结论 / Conclusions)
- [数据可用性、伦理声明与补充信息 / Back Matter](#数据可用性、伦理声明与补充信息 / Back Matter)
- [参考文献 / References (完整64篇题录)](#参考文献 / References (完整64篇题录))
- [专家精读提示与科研实战研判 / Critical Reading Notes](#专家精读提示与科研实战研判 / Critical Reading Notes)
- [闭环质检报告 / Closed-Loop Verification Checklist](#闭环质检报告 / Closed-Loop Verification Checklist)
术语对照表
| 规范英文术语 (Canonical Term) | 规范中文译名 | 学术内涵、计算模型与本研究应用语境 |
|---|---|---|
| Piscirickettsia salmonis (P. salmonis) | 鲑立克次体 / 鲑立克次氏体 | 革兰氏阴性胞内兼性/专性寄生菌,引发鲑科鱼类严重败血性传染病(SRS),水产养殖头号杀手 |
| Salmon rickettsial syndrome (SRS) | 鲑立克次体综合征 / 鱼类立克次体败血症 | 又称 piscirickettsiosis,引起三文鱼多器官点状出血、肝肾肿大溃疡并造成高达 90% 以上死亡率 |
| Coho salmon (CS; Oncorhynchus kisutch) | 银鲑 / 共轭大马哈鱼 | 太平洋鲑属经济鱼类,在智利网箱养殖中受 SRS 危害严重,具有独特的寡基因抗性遗传架构 |
| Rainbow trout (RT; Oncorhynchus mykiss) | 虹鳟 | 太平洋鲑属重要冷水养殖品种,对 SRS 极度易感,养殖期死亡率中 SRS 占比超 92% |
| Atlantic salmon (AS; Salmo salar) | 大西洋鲑 | 大西洋鲑属全球产量最大的水产鱼类,对 SRS 抗性呈现典型微效多基因连续变异遗传控制 |
| Days to death (DD) | 死亡天数 / 染毒至死亡时间 | 连续型抗病表型,记录个体攻毒后存活的具体天数,反映宿主对病原感染进程的耐受与潜伏延缓能力 |
| Binary survival (BS) | 二元存活(终极存活状态) | 二分类抗病表型(0 = 试验终点存活,1 = 死亡),直接衡量个体的终极生/死抗病结局与群体抗性水平 |
| Double-digest RAD-seq (ddRAD) | 双酶切限制性位点关联 DNA 测序 | 一种高性价比的简化基因组测序策略,利用两种限制性内切酶富集基因组特定片段并检出高密度 SNP |
| Axiom SNP array | Affymetrix Axiom 高密度 SNP 芯片 | 商业化与定制化固相基因分型芯片(虹鳟 57K 芯片、大西洋鲑 50K 芯片),提供全基因组覆盖标记 |
| Bayes C | 贝叶斯 C 混合回归模型 | 经典全基因组关联与基因组选择模型,假设绝大多数 SNP 效应为零(概率 π = 0.99 \pi=0.99 π=0.99),仅少量位点服从正态先验 |
| Proportion of genetic variance explained (GEV) | 解释遗传变异比例 | 单个 SNP 标记所解释的加性遗传方差占总多基因方差的百分比,计算公式为 V g i = 2 p i q i a i 2 / σ u 2 Vg_i = 2p_i q_i a_i^2 / \sigma_u^2 Vgi=2piqiai2/σu2 |
| Oligogenic control | 寡基因控制 / 寡基因遗传架构 | 性状遗传变异主要由少数几个效应中等或较显著的主效数量性状基因座 (QTL) 所主导的遗传模式 |
| Polygenic control | 多基因控制 / 多基因微效架构 | 性状由全基因组成百上千个微效等位基因共同决定的经典数量性状遗传模式(如 AS 对 SRS 的抗性) |
| Synteny / Syntenic regions | 染色体同源共线性 / 同源共线区段 | 不同物种在演化过程中来自共同祖先的同源染色体片段中基因排列顺序与相对位置的高度保守性 |
| Leader SNP | 领头 SNP / 先导主效标记 | 在特定染色体或候选区间内解释遗传变异比例(GEV)最高、统计关联信号最显著的代表性核心标记 |
| P-loop_NTPase superfamily | 含 P-环的三磷酸核苷水解酶超家族 | 参与 ATP/GTP 结合与水解的大型蛋白超家族,在三种鲑鱼抗病候选基因中显著富集,主导细胞动力学与免疫信号 |
| Non-receptor tyrosine kinase (nr-TPK) | 非受体型酪氨酸蛋白激酶 | 胞质内催化酪氨酸残基磷酸化的关键信号激酶(如 Src 家族 YES1, LYN 及 SYK),常被胞内菌毒力效应蛋白劫持 |
| ADP-ribosylation factor / EspJ | ADP-核糖基化因子 / 细菌效应蛋白 EspJ | 细菌通过分泌系统注入宿主细胞的毒力修饰酶,可通过 ADP-核糖基化抑制宿主 nr-TPK 激酶活性以逃避清除 |
| RAB GTPases (RAB1, RAB18) | Rab 小 GTP 酶亚家族 | 胞内膜泡运输与细胞器衍生的主控分子;鲑立克次体侵染后劫持宿主膜系统构建含菌空泡 (PCV) 的关键靶点 |
| Motor proteins (Kinesins / Myosin) | 细胞骨架马达蛋白 (驱动蛋白/肌球蛋白) | 驱动蛋白 KIF15/KIF2C 与肌球蛋白 MYO3B,参与肌动蛋白与微管重塑;病原菌借此在宿主吞噬细胞内运动与扩散 |
| NLR Family (NLRC3 / NLRP12) | NOD 样受体模式识别家族 | 胞质先天免疫传感器,作为负向调控因子抑制 NF- κ \kappa κB 激活,对于控制致死性过度炎症风暴至关重要 |
| Liver fatty acid-binding protein (L-FABP) | 肝型脂肪酸结合蛋白 | 肝细胞脂质代谢核心载体,在抗病大西洋鲑家系中显著上调而在易感家系中下调,被确认为潜在表达生物标志物 |
| Linkage Disequilibrium (LD) | 连锁不平衡 ( r 2 r^2 r2) | 染色体上不同基因座等位基因非随机共同遗传的统计关联程度,决定了 GWAS 分辨率及所需标记密度 |
| Genomic Selection (GS) | 基因组选择 | 现代水产动物分子育种核心体系,利用覆盖全基因组的高密度 SNP 预测候选个体的基因组育种值 (GEBV) |
| Marker-Assisted Selection (MAS) | 标记辅助选择 | 当性状存在主效 QTL(如传染性胰腺坏死 IPN 抗性 QTL)时,直接利用紧密连锁标记实施精准基因型淘汰与选育 |
摘要与关键词 / Abstract & Keywords
Source: p.1 S001
Original:
Piscirickettsia salmonis is the etiologic agent of salmon rickettsial syndrome (SRS) and is responsible for considerable economic losses in salmon aquaculture. The bacterium affects coho salmon (CS; Oncorhynchus kisutch ), Atlantic salmon (AS; Salmo salar ), and rainbow trout (RT; Oncorhynchus mykiss ) in several countries, including Norway, Canada, Scotland, Ireland, and Chile. We used Bayesian genome-wide association study analyses to investigate the genetic architecture of resistance to P. salmonis in farmed populations of these species. Resistance to SRS was defined as the number of days to death and as binary survival (BS). A total of 828 CS, 2130 RT, and 2601 AS individuals were phenotyped and then genotyped using double-digest restriction site-associated DNA sequencing and 57K and 50K Affymetrix® Axiom® single nucleotide polymorphism (SNP) panels, respectively. Both traits of SRS resistance in CS and RT appeared to be under oligogenic control. In AS, there was evidence of polygenic control of SRS resistance. To identify candidate genes associated with resistance, we applied a comparative genomics approach in which we systematically explored the complete set of genes adjacent to SNPs, which explained more than 1% of the genetic variance of resistance in each salmonid species (533 genes in total). Thus, genes were classified based on the following criteria: i) shared function of their protein domains among species, ii) shared orthology among species, iii) proximity to the SNP explaining the highest proportion of the genetic variance, and iv) presence in more than one genomic region explaining more than 1% of the genetic variance within species. Our results allowed us to identify 120 candidate genes belonging to at least one of the four criteria described above.
中文:
鲑立克次体(Piscirickettsia salmonis )是鲑立克次体综合征(salmon rickettsial syndrome, SRS)的致病原,导致了全球三文鱼养殖业惨重的经济损失。该细菌侵染包括挪威、加拿大、苏格兰、爱尔兰和智利等多个国家的银鲑(coho salmon, CS; Oncorhynchus kisutch )、大西洋鲑(Atlantic salmon, AS; Salmo salar )以及虹鳟(rainbow trout, RT; Oncorhynchus mykiss)。在本研究中,我们采用贝叶斯全基因组关联分析(Bayesian genome-wide association study, GWAS)方法,深入探究了这些物种人工养殖群体对鲑立克次体抗性的遗传架构。对 SRS 的抗性表型分别定义为感染后的死亡天数(days to death, DD)以及二元存活状态(binary survival, BS)。研究总计对 828 尾银鲑、2130 尾虹鳟和 2601 尾大西洋鲑进行了严格的表型测定,并分别利用双酶切限制性位点关联 DNA 测序(ddRAD-seq)技术以及 57K 和 50K Affymetrix® Axiom® 单核苷酸多态性(SNP)芯片进行了高密度基因分型。研究表明,在银鲑和虹鳟中,两种抗病性状均呈现出寡基因(oligogenic)控制特征;而在大西洋鲑中,则呈现出经典多基因(polygenic)控制的遗传证据。为了挖掘与抗病性关联的核心候选基因,我们构建了一套跨物种比较基因组学分析策略,系统检索了各鲑科物种中解释遗传变异比例(GEV)超过 1% 的所有 SNP 侧翼物理区间内的完整基因集(共计 533 个独特基因)。在此基础上,按照以下四个标准对基因进行了分级归类:i)在物种间具有共享的蛋白结构域功能;ii)在物种间具有共享的直系同源性(orthology);iii)邻近解释遗传变异比例最高的领头 SNP(leader SNP);iv)在单一物种内部同时位于多个解释遗传变异大于 1% 的重叠基因组区域中。基于上述标准,我们最终筛选出 120 个至少符合上述四项标准之一的抗性候选基因。
Source: p.2 S002
Original:
Of these, 21 of them were part of at least two of the criteria defined above and are suggested to be strong functional candidates influencing P. salmonis resistance. These genes are related to diverse biological processes, such as kinase activity, GTP hydrolysis, helicase activity, lipid metabolism, cytoskeletal dynamics, inflammation, and innate immune response, which seem essential in the host response against P. salmonis infection. These results provide fundamental knowledge on the potential functional genes underpinning resistance against P. salmonis in three salmonid species.
中文:
在上述候选基因中,有 21 个基因同时满足至少两项评价标准,被鉴定为调控鲑立克次体抗性的高置信度功能候选基因。这些关键基因富集于多样化的生物学通路与细胞分子过程,包括蛋白激酶活性、GTP 水解、解旋酶活性、脂质代谢、细胞骨架动态重塑、炎症调控以及先天免疫应答等,这些过程在宿主抵御鲑立克次体胞内侵染的防御机制中发挥着核心功能。本研究结果为阐明三大鲑科鱼类抗鲑立克次体感染的潜在功能基因网络提供了关键的理论依据与理论支撑。
Source: p.2 S003
Original:
Keywords: coho salmon, rainbow trout, Atlantic salmon, Piscirickettsia salmonis, genome-wide association study, comparative genomics, piscirickettsiosis
中文:
关键词: 银鲑;虹鳟;大西洋鲑;鲑立克次体;全基因组关联分析;比较基因组学;鲑立克次体病 (piscirickettsiosis)
引言 / Introduction
Source: p.2 S004
Original:
Infectious diseases are responsible for large economic losses in salmon farming. Piscirickettsia salmonis , the causal agent of salmon rickettsial syndrome (SRS), affects several salmon species and is considered one of the major pathogens affecting the salmon farming industry (Rozas and Enríquez, 2014). P. salmonis was identified in 1989 from farmed coho salmon (CS; Oncorhynchus kisutch ) sampled in Chile (Cvitanich et al., 1991). Since then, P. salmonis has been confirmed as the causative agent for clinical and chronic SRS in CS, Atlantic salmon (AS; Salmo salar ), and rainbow trout (RT; Oncorhynchus mykiss) in several countries, including Norway, Canada, Scotland, Ireland, and Chile (Fryer and Hedrick, 2003). Current control protocols and treatments are based on antibiotics and vaccines. The effectiveness of both strategies in field conditions is not optimal (Rozas and Enríquez, 2014). From the total mortalities ascribed to infectious diseases in Chile, SRS is responsible for 18.3%, 92.6%, and 67.9% in CS, RT, and AS, respectively (Sernapesca, 2018). These mortality rates, together with other factors such as antibiotic treatments and vaccinations, have generated economic losses up to USD $450 million per year (Camusetti et al., 2015).
中文:
传染性疾病是造成全球三文鱼养殖业巨额经济损失的最主要诱因。鲑立克次体(Piscirickettsia salmonis )作为鲑立克次体综合征(salmon rickettsial syndrome, SRS)的病原体,侵染多种鲑科鱼类,被公认为对全球三文鱼养殖产业构成最严重威胁的病原菌之一(Rozas and Enríquez, 2014)。鲑立克次体于 1989 年在智利网箱养殖的银鲑(coho salmon, CS; Oncorhynchus kisutch )病鱼中首次被分离并鉴定(Cvitanich et al., 1991)。自此以后,多项研究证实该细菌是导致银鲑、大西洋鲑(Atlantic salmon, AS; Salmo salar )以及虹鳟(rainbow trout, RT; Oncorhynchus mykiss)发生急性临床型和慢性型 SRS 的直接病因,受害国家广泛分布于挪威、加拿大、苏格兰、爱尔兰和智利等主要产区(Fryer and Hedrick, 2003)。目前针对该病的日常防控与治疗方案主要依赖于抗生素化学药物与疫苗接种。然而在实际海水网箱养殖工况下,这两种应对策略的防护效果均极不理想(Rozas and Enríquez, 2014)。在智利海水养殖中因传染性疫病引起的总死亡率统计中,SRS 在银鲑、虹鳟和大西洋鲑中的死亡致因占比分别高达 18.3%、92.6% 和 67.9%(Sernapesca, 2018)。如此高企的死亡率,叠加抗生素药物治疗、全群疫苗接种及死鱼处理等衍生运营成本,每年给产业造成的综合经济损失高达 4.5 亿美元(Camusetti et al., 2015)。
Source: p.2 S005
Original:
A feasible and sustainable alternative to prevent disease outbreaks is genetic selection for disease resistance (Bishop and Woolliams, 2014). The estimated levels of heritability for resistance to P. salmonis in CS, AS, and RT range from 0.11 to 0.41 (Correa et al., 2015; Yáñez et al., 2016; Bangera et al., 2017; Barría et al., 2018a; Yoshida et al., 2018a; Bassini et al., 2019), demonstrating the feasibility of improving P. salmonis resistance through artificial selection in farmed salmon species.
中文:
从遗传育种维度进行抗病性人工选育,是阻断疫病暴发、实现水产养殖健康可持续发展的一条兼具可行性与生态效益的根本途径(Bishop and Woolliams, 2014)。已有文献报道显示,银鲑、大西洋鲑和虹鳟对鲑立克次体的抗性遗传力估计值介于 0.11 至 0.41 之间(Correa et al., 2015; Yáñez et al., 2016; Bangera et al., 2017; Barría et al., 2018a; Yoshida et al., 2018a; Bassini et al., 2019),处于中等偏高水平,这充分证实了通过人工选择育种手段有效提升人工养殖鲑科鱼类对鲑立克次体抗病能力的可行性与巨大潜力。
Source: p.2 S006
Original:
Currently, the advancement of molecular technologies has allowed the generation of dense marker panels for salmonid species (Houston et al., 2014; Palti et al., 2015; Yañez et al., 2016; Macqueen et al., 2017). The use of genotypes from dense panels of single nucleotide polymorphism (SNP) markers, together with phenotypes for the traits of interest, assessed in a large number of individuals could provide opportunities to discover the genetic architecture of complex traits. When genetic markers are linked to a major effect of quantitative trait loci (QTL), marker-assisted selection (MAS) could then be implemented into breeding programs. For instance, a QTL explaining ~80% of the genetic variance for resistance to infectious pancreatic necrosis virus (IPNV) has been identified in Scottish and Norwegian AS farmed populations (Houston et al., 2008; Moen et al., 2009). To date, the number of IPNV outbreaks has been significantly reduced in Norwegian AS populations because of MAS for IPNV resistance (Hjeltnes et al., 2018). Interestingly, Moen et al. (2015) mapped the QTL to a region containing an epithelial cadherin (cdh1) gene encoding a protein that binds to IPNV, indicating that the protein is part of the machinery used by the virus for host internalization.
中文:
当前,高通量分子生物学技术的飞速演进使得针对主要鲑科鱼类开发超高密度基因组标记芯片成为现实(Houston et al., 2014; Palti et al., 2015; Yañez et al., 2016; Macqueen et al., 2017)。在海量个体规模下,将高密度单核苷酸多态性(SNP)标记基因型与严密测定的目标经济性状表型数据深度整合,为全面揭示复杂性状的精细遗传架构创造了前所未有的契机。一旦检测到与主效数量性状基因座(QTL)紧密连锁的特征标记,便可直接将标记辅助选择(marker-assisted selection, MAS)落地整合至商业育种方案中。一个著名的成功范例是:在苏格兰和挪威的大西洋鲑商业养殖群体中,科研人员成功定位出了一个能够解释抗传染性胰腺坏死病毒(IPNV)约 80% 遗传变异的主效 QTL(Houston et al., 2008; Moen et al., 2009)。截至目前,由于全面实施了针对抗 IPNV 关键位点的 MAS 选育,挪威大西洋鲑养殖群体中 IPNV 的疫病暴发频次已显著骤降(Hjeltnes et al., 2018)。尤为引人注目的是,Moen 等人(2015)进一步将该主效 QTL 精细定位到了包含上皮钙粘蛋白(cdh1)基因的基因组区域,该基因编码的跨膜蛋白能够直接与 IPNV 结合,表明该蛋白正是病毒入侵宿主细胞并实现胞吞内化所必需的关键受体组件。
Source: p.2 S007
Original:
P. salmonis resistance has been suggested to be polygenic, with many loci explaining a small amount of the total genetic variance (Correa et al., 2015; Barría et al., 2018a), suggesting that the implementation of genomic selection (GS) is the most appropriate strategy to accelerate the genetic progress for this trait. Methods that can model all available SNPs simultaneously, including Bayesian regression methods (Fernando and Garrick, 2013), appear to be better for estimating marker effects than conventional methods of modeling each SNP individually and therefore are becoming increasingly more popular for genome-wide association study (GWAS; Goddard et al., 2009).
中文:
与 IPN 抗性的单一主效模式不同,以往研究表明鲑立克次体抗性呈现出多基因控制模式,即由散布在全基因组上的大量微效基因座共同解释遗传方差(Correa et al., 2015; Barría et al., 2018a)。这种遗传特性意味着,实施基因组选择(genomic selection, GS)是加快该性状育种遗传进展的最理想策略。相比于逐个检验单个 SNP 的传统单标记回归方法,能够对全基因组所有可用 SNP 进行同步联合建模的算法(尤其是贝叶斯回归方法体系,如 Fernando and Garrick, 2013 所述),在准确估计标记效应值方面展现出了明显优势,因此在全基因组关联分析(GWAS)领域正受到越来越广泛的应用与推崇(Goddard et al., 2009)。
Source: p.2 S008
Original:
Due to the fact that P. salmonis affects farmed populations of three phylogenetically related salmonid species, including CS, AS, and RT, generating mortalities in a similar manner and that genetic variation for P. salmonis resistance has been already reported, we believe that exploring the genetic architecture of this trait simultaneously in the three species can provide further insights into the biology of the differential response against this intracellular bacteria among individuals. Thus, a comparative genomics approach aiming at evaluating and comparing genomic regions involved in P. salmonis resistance in CS, AS, and RT would help in narrowing down the list of potential candidate genes associated with the trait for further functional validation in salmonid species.
中文:
鉴于鲑立克次体能够同时侵染系统发育关系密切的三种鲑科鱼类(银鲑、大西洋鲑和虹鳟),并诱发高度相似的组织病理学变化与群体致死过程,加之这三种鱼在以往研究中均已证实存在丰富的抗病遗传变异,我们认为:在三种鲑科鱼类中同步剖析该抗病性状的遗传架构,能够为深入解析宿主个体抵御该胞内寄生菌差异化应答背后的深层生物学机制提供全新视阈。因此,构建一套系统评估并横向比对银鲑、大西洋鲑和虹鳟中抗鲑立克次体相关基因组区域的比较基因组学策略,将有助于大幅缩小与抗病性状关联的候选基因范围,从而为后续在鲑科鱼类中开展更深入的功能验证与分子机理研究奠定坚实基础。
Source: p.2 S009
Original:
The aims of this study were i) to dissect the genetic architecture of resistance to P. salmonis in CS, AS, and RT using SNP and phenotype data modeled together using Bayesian GWAS approach, ii) to identify genomic regions involved in P. salmonis resistance among the three salmonid species, and iii) to identify candidate genes associated with P. salmonis resistance through a comparative genomics analysis.
中文:
本研究的核心科学目标包括:i)基于高密度 SNP 分型与精细表型测定数据,通过贝叶斯 GWAS 联合建模,深入剖析银鲑、大西洋鲑和虹鳟对鲑立克次体抗性的全基因组遗传架构;ii)横向系统挖掘并比对三种鲑科鱼类中参与抗鲑立克次体的同源共线与共有基因组区域;iii)依托跨物种比较基因组学分析流程,精准鉴定并按功能优先级排序与鲑立克次体抗性密切相关的核心候选功能基因。
材料与方法 / Materials and Methods
人工攻毒感染试验 / Challenge Tests
Source: p.2 S010
Original:
A total of 2,606, 2,601, and 2,416 fish belonging to 107, 118, and 105 full-sib families from CS, AS, and RT, respectively, were independently challenged with an isolate of P. salmonis (strain LF-89; Mandakovic et al., 2016) as described in Barría et al. (2018a), Bassini et al. (2019), and Yáñez et al. (2013), Yáñez et al. (2014), Yáñez et al. (2016).
中文:
本试验分别选取来自 107 个银鲑全同胞家系(共 2,606 尾)、118 个大西洋鲑全同胞家系(共 2,601 尾)以及 105 个虹鳟全同胞家系(共 2,416 尾)的试验鱼群,严格参照 Barría 等人(2018a)、Bassini 等人(2019)以及 Yáñez 等人(2013, 2014, 2016)所建立的成熟攻毒感染规程,独立接种鲑立克次体标准参考分离株(LF-89 株系;Mandakovic et al., 2016)开展受控人工感染试验。
Source: p.3 S011
Original:
Before the beginning of each experimental challenge, quantitative polymerase chain reaction (qPCR) was performed in a sub-sample of each population to confirm the absence of Flavobacterium spp., infectious salmon anemia virus, and IPNV. Subsequently, fish were intraperitoneally (IP) injected with 0.2 ml of an LD50 inoculum of P. salmonis. Although an IP challenge is not a natural form of infection, it is an effective method for presenting a naïve animal with a known and controlled amount of bacteria, making sure that the bacterial load and the time of infection are the same in every fish (Pulgar et al., 2015).
中文:
在各批次人工攻毒试验正式启动前,均从各个供试群体中随机抽取具有统计代表性的健康子样本,通过荧光定量聚合酶链反应(qPCR)进行常见病原学筛查,严格排除感染黄杆菌属(Flavobacterium spp.)、传染性鲑鱼贫血病毒(ISAV)以及传染性胰腺坏死病毒(IPNV)的潜在风险。随后,对所有供试个体实施腹腔注射(intraperitoneally, IP),注射剂量为 0.2 ml 处于半数致死量(LD50)浓度的鲑立克次体悬液。尽管腹腔注射并非自然环境中的天然感染途径,但该给药方法能确保向处于初始未暴露免疫状态(naïve)的试验个体精准呈递剂量已知且均一受控的活菌载量,从而保证试验群中每一尾鱼所承受的初始菌负荷与初始感染时间节点高度均一恒定(Pulgar et al., 2015)。
Source: p.3 S012
Original:
After IP injection, infected fish were equally distributed by family into three different test tanks. Each challenge was maintained until mortalities returned to baseline levels. At the end of the challenges, all surviving fish were anesthetized and euthanized. A sample of caudal fin was taken from each survivor and dead fish from each of the experimental challenges for DNA extraction. Body weight was measured at the beginning of the challenge and at the time of death for each individual. The presence of P. salmonis was confirmed in a random sample of dead fish through qPCR and necropsy. Each experimental challenge was performed at Aquainnovo's Research Station, Xth Region, Chile.
中文:
完成腹腔注射后,属于不同家系的感染试验鱼被随机且均等地分流饲养于三个平行的独立试验水槽中。整个攻毒监测周期持续推进,直至群体每日新增死亡率回落至初始基线平稳水平为止。试验终点时,对所有存活个体实施深度麻醉并作人道化安乐死处理。从每场攻毒试验中的所有死亡个体以及终末存活个体尾鳍部位采集组织样本,用于后续基因组 DNA 提取。在攻毒启动时以及每尾个体死亡时分别精确测定记录体重。通过系统的病理解剖学观察与特异性 qPCR 检测,在随机抽检的死亡鱼样本中 100% 确认了鲑立克次体的致死性感染与组织病变。所有人工攻毒试验均于智利第十区的 Aquainnovo 水产遗传育种试验基地标准化完成。
基因分型与质量控制 / Genotyping
Source: p.3 S013
Original:
A total of 828 CS, 2130 RT, and 2601 AS were genotyped using double-digest restriction site-associated DNA (ddRAD) and 57K and 50K Affymetrix® Axiom® SNP panels, respectively. Total DNA was extracted using commercial kits following the manufacturer's protocols. For CS, we used the Wizard SV Genomic DNA purification System (Promega), whereas DNeasy Blood & Tissue (Qiagen) was used for RT and AS.
中文:
本研究总计选取 828 尾银鲑、2130 尾虹鳟和 2601 尾大西洋鲑进行全基因组高密度分子标记分型,其中银鲑采用双酶切限制性位点关联 DNA(ddRAD)简化基因组测序技术,虹鳟和大西洋鲑则分别采用 57K 和 50K Affymetrix® Axiom® SNP 固相芯片。基因组总 DNA 依据商业化试剂盒的标准说明书严格提取:银鲑样本采用 Wizard SV 基因组 DNA 纯化系统(Promega 公司),而虹鳟和大西洋鲑样本则采用 DNeasy 血液与组织 DNA 提取试剂盒(Qiagen 公司)。
Source: p.3 S014
Original:
For CS, 10 ddRAD libraries were prepared following the protocol proposed by Peterson et al. (2012) and sequenced on an Illumina Hiseq2500 (150 bp single-end). Raw sequences were analyzed using STACKS version 1.41 (Catchen et al., 2011; Catchen et al., 2013). rad-tags that passed the process_radtags quality control (QC) were aligned to the CS reference genome (GCF_002021735.1). Loci were built with pstacks setting a minimum depth coverage of three. After catalog construction, rad-tags were matched using sstacks followed by populations using default parameters. QC included the removal of SNPs below the following thresholds: Hardy-Weinberg equilibrium (HWE) P < 1 × 10 − 6 P < 1 \times 10^{-6} P<1×10−6, minor allele frequency (MAF) < 0.05 < 0.05 <0.05, and genotyping call rate < 0.80 < 0.80 <0.80. Individuals with a call rate below 0.70 were removed from the subsequent analysis. For a detailed protocol of library construction and SNP identification, see Barría et al. (2018a).
中文:
对于银鲑,严格按照 Peterson 等人(2012)提出的标准化建库方案构建了 10 个 ddRAD 文库,并置于 Illumina HiSeq 2500 测序平台进行高通量测序(150 bp 单端测序)。下机原始测序数据采用 STACKS 软件包 1.41 版进行生物信息学深度分析(Catchen et al., 2011; Catchen et al., 2013)。通过 process_radtags 质控清洗后的高质量 rad-tags 标签比对至银鲑 NCBI 参考基因组(版本号:GCF_002021735.1)。利用 pstacks 构建基因座,并将最低测序深度阈值设定为 3。在完成基因座目录编目(catalog)后,运行 sstacks 进行匹配,进而采用默认参数执行 populations 模块完成群体 SNP 分型。随后的质量控制(QC)阶段剔除了不满足下列严苛阈值的 SNP 位点:哈代-温伯格平衡(HWE)检验检验显著性 P < 1 × 10 − 6 P < 1 \times 10^{-6} P<1×10−6、微效等位基因频率(MAF) < 0.05 < 0.05 <0.05 以及位点检出率(call rate) < 0.80 < 0.80 <0.80。此外,样本整体检出率低于 0.70 的低质量个体亦从后续分析中予以剔除。关于文库构建与 SNP 检出的详尽生信流程,详见 Barría 等人(2018a)的研究。
Source: p.3 S015
Original:
RT individuals were genotyped using the commercial 57K Affymetrix® Axiom® SNP array developed by the National Center of Cool and Cold Water Aquaculture at the U.S. Department of Agriculture (Palti et al., 2015). SNPs were filtered with the following QC parameters: HWE P < 1 × 10 − 6 P < 1 \times 10^{-6} P<1×10−6, MAF < 0.05 < 0.05 <0.05, and SNP call rate < 0.95 < 0.95 <0.95. Individuals with call rates lower than 0.95 were also removed.
中文:
虹鳟个体采用美国农业部国家冷水水产养殖中心联合研发的商业化 57K Affymetrix® Axiom® 虹鳟高密度 SNP 芯片进行基因分型(Palti et al., 2015)。分型后对所有 SNP 进行严格的质控参数过滤:偏离哈代-温伯格平衡 P < 1 × 10 − 6 P < 1 \times 10^{-6} P<1×10−6、MAF < 0.05 < 0.05 <0.05 以及标记检出率 < 0.95 < 0.95 <0.95 的位点均被过滤;同时,整体基因型检出率低于 0.95 的个体也被彻底剔除。
Source: p.3 S016
Original:
The 50K Affymetrix® Axiom® SNP array used to genotype AS was developed by Universidad de Chile and Aquainnovo (Correa et al., 2015; Yañez et al., 2016). These markers were selected from a 200K array, as described in detail by Correa et al. (2015). Genotypes were subjected to QC using the following criteria: HWE P < 1 × 10 − 6 P < 1 \times 10^{-6} P<1×10−6, MAF < 0.05 < 0.05 <0.05, SNP, and samples were discarded when the genotype rate was < 0.95 < 0.95 <0.95.
中文:
用于大西洋鲑基因分型的 50K Affymetrix® Axiom® SNP 芯片由智利大学与 Aquainnovo 公司联合研制(Correa et al., 2015; Yañez et al., 2016)。芯片中的 5 万个高多态性标记是从更高密度的 200K 探针文库中精挑细选而来,详细开发细节见 Correa 等人(2015)。所获基因型数据接受了同等严格的质量控制:剔除 HWE P < 1 × 10 − 6 P < 1 \times 10^{-6} P<1×10−6、MAF < 0.05 < 0.05 <0.05 的 SNP,且当标记或样本个体的分型检出率 < 0.95 < 0.95 <0.95 时直接废弃。
贝叶斯全基因组关联分析 / GWAS Analysis
Source: p.3 S017
Original:
Resistance to SRS was defined as both the number of days to death (DD) after experimental challenge and the binary survival (BS; 0 for surviving individuals at the end of the experimental challenge and 1 for deceased fish). GWAS analyses were performed using the Bayes C method that assumes distributed mixture distribution for marker effects. All model parameters are defined in the following equation:
中文:
在本研究中,对鲑立克次体综合征(SRS)的宿主抗病表型被精细划分为两个互补的性状指标:其一是攻毒后的死亡天数(days to death, DD),作为表征感染病程耐受潜伏期的连续型数量性状;其二是二元存活状态(binary survival, BS;试验终点健康存活个体赋值为 0,因病死亡个体赋值为 1),作为表征终极抗病结局的二分类性状。全基因组关联分析采用经典的贝叶斯混合模型------Bayes C 方法开展,该方法显式假定标记效应服从两点混合先验分布。统计模型的全部参数由以下数学方程式统一形式化定义:
Equation (1):
y = X b + Z u + ∑ i = 1 n g i a i δ i + e (EQ1) y = Xb + Zu + \sum_{i=1}^{n} g_i a_i \delta_i + e \quad \text{(EQ1)} y=Xb+Zu+i=1∑ngiaiδi+e(EQ1)
Source: p.3 E001
Variable Definitions / 变量释义:
- y y y: 表型观测值向量(连续变量 DD 或二分类变量 BS)。
- X , Z X, Z X,Z: 分别为固定效应和加性多基因效应的模型关联设计阵(incidence matrices)。
- b b b: 已知固定效应向量(本研究包括试验水槽环境效应 tank 以及个体起始体重 body weight 协变量)。
- u u u: 基于系谱亲缘关系的个体随机多基因遗传效应向量(random polygenic effects)。
- g i g_i gi: 包含所有个体在第 i i i 个 SNP 位点上的基因型编码向量(通常编码为 0, 1, 2)。
- a i a_i ai: 第 i i i 个 SNP 位点的随机等位基因替换效应值(random allele substitution effect)。
- δ i \delta_i δi: 标记服从二项分布的指示变量(取值为 0 或 1),其混合比例参数 π \pi π 设定为 0.99,即在任一 MCMC 抽样迭代中,只有 1 − π = 1 % 1-\pi = 1\% 1−π=1% 的 SNP 具有非零遗传效应。
- e e e: 随机残差效应向量(vector of residual effects)。
Source: p.3 S018
Original:
where y y y is the vector of phenotypic records (DD or BS); X X X and Z Z Z are the incidence matrix of fixed effects and polygenic effect, respectively; b b b is the vector of fixed effects (tank and body weight); u u u is the random vector of polygenic effects of all individuals in the pedigree; g i g_i gi is the vector of the genotypes for the i i ith SNP for each animal; a i a_i ai is the random allele substitution effect of the i i ith SNP; δ i \delta_i δi is an indicator variable ( 0 , 1 ) (0, 1) (0,1) sampled from a binomial distribution with parameters determined such that π \pi π value of 0.99; and e e e is a vector of residual effects.
中文:
式中: y y y 代表表型记录向量(死亡天数 DD 或二元存活 BS); X X X 和 Z Z Z 分别代表固定效应与多基因效应的设计关联矩阵; b b b 代表固定效应向量(包含试验水槽效应和个体体重协变量); u u u 代表系谱中所有个体的随机多基因加性遗传效应向量; g i g_i gi 代表每个个体在第 i i i 个 SNP 上的基因型构成的向量; a i a_i ai 代表第 i i i 个 SNP 的随机等位基因替换效应; δ i \delta_i δi 为从二项分布中抽样得到的指示变量(取值 0 或 1),其分布参数满足先验概率 π = 0.99 \pi = 0.99 π=0.99; e e e 代表随机残差效应向量。
Source: p.3 S019
Original:
The prior assumption is that SNP effects have independent and identical mixture distributions, where each SNP has a point mass at zero (with probability π \pi π) and a univariate Gaussian distribution (with probability 1 − π 1 - \pi 1−π) with a mean equal to zero and variance equal to σ a 2 \sigma_a^2 σa2 having in turn a scaled inverse χ 2 \chi^2 χ2 prior, with v a = 4 v_a = 4 va=4 and v e = 10 v_e = 10 ve=10 degrees of freedom and scale parameter, respectively (Fernando and Garrick, 2013). These hyperparameter values were chosen based on previous studies (Peters et al., 2012; Santana et al., 2016; Wolc et al., 2016; Yoshida et al., 2017; Yoshida et al., 2018a).
中文:
Bayes C 模型的核心先验假设是:全基因组 SNP 效应服从独立同分布的两阶段混合分布,即每一个 SNP 标记以先验概率 π \pi π 取值为零(点质量集中于零),并以概率 1 − π 1 - \pi 1−π 服从均值为零、方差为 σ a 2 \sigma_a^2 σa2 的单变量正态分布;而该方差分量 σ a 2 \sigma_a^2 σa2 本身又服从逆向标度卡方分布(scaled inverse χ 2 \chi^2 χ2 prior),其中自由度和尺度超参数分别指定为 v a = 4 v_a = 4 va=4 与 v e = 10 v_e = 10 ve=10(Fernando and Garrick, 2013)。这些先验超参数的经验取值严格基于团队前期多项经过同行评议的大量动植物统计基因组学权威研究(Peters et al., 2012; Santana et al., 2016; Wolc et al., 2016; Yoshida et al., 2017; Yoshida et al., 2018a)。
Source: p.3 S020
Original:
The analyses were performed using the GS3 software (Legarra et al., 2013). A total of 200,000 iterations in Gibbs sampling were used, with a burn-in period of 20,000 cycles, and the results were saved every 50 cycles. Convergence was assessed by visual inspection of trace plots of the posterior density of genetic and residual variances.
中文:
全基因组贝叶斯关联运算通过开源遗传评估套件 GS3 软件执行(Legarra et al., 2013)。马尔可夫链蒙特卡洛(MCMC)吉布斯抽样总迭代轮数设定为 200,000 次,其中前 20,000 次循环作为老化预烧期(burn-in)予以舍弃,此后每隔 50 次循环进行一次抽样记录(thinning interval = 50)以消除样本自相关性。马尔可夫链的抽样收敛性通过对遗传方差与残差方差后验概率密度的链轨迹图(trace plots)进行严格的可视化检验与平稳性评估。
Source: p.3 S021
Original:
The proportion of the genetic variance explained (GEV) by each significant SNP was calculated as:
中文:
每个达到显著性水平的 SNP 标记所解释的遗传方差比例(proportion of the genetic variance explained, GEV)通过以下理论公式严密求算:
Equation (2):
V g i = ( 2 p i q i a i 2 σ u 2 ) (EQ2) Vg_i = \left( \frac{2 p_i q_i a_i^2}{\sigma_u^2} \right) \quad \text{(EQ2)} Vgi=(σu22piqiai2)(EQ2)
Source: p.3 E002
Variable Definitions / 变量释义:
- V g i Vg_i Vgi: 第 i i i 个 SNP 标记所解释的遗传变异比例(GEV)。
- p i , q i p_i, q_i pi,qi: 第 i i i 个 SNP 标记的主等位基因与微效等位基因频率( q i = 1 − p i q_i = 1 - p_i qi=1−pi)。
- a i a_i ai: 第 i i i 个 SNP 标记对目标表型(DD 或 BS)后验估计的加性等位基因替换效应值。
- σ u 2 \sigma_u^2 σu2: 全基因组多基因加性遗传方差估计值(Lee et al., 2013)。
Source: p.3-4 S022
Original:
where p i p_i pi and q i q_i qi are the allele frequencies for the i i ith SNP, a i a_i ai is the estimated additive effect of the i i ith SNP on the phenotype, and σ u 2 \sigma_u^2 σu2 is the estimate of the polygenic variance (Lee et al., 2013). The association between the SNPs and the phenotypes was assessed using the proportion of the GEV by each marker. To be inclusive regarding the genomic regions to be compared across the three species, we selected each of the regions explaining at least 1% of the genetic variance for the trait in each species.
中文:
式中: p i p_i pi 与 q i q_i qi 代表第 i i i 个 SNP 位点的两个等位基因频率; a i a_i ai 为该 SNP 对表型性状的后验加性效应估计值; σ u 2 \sigma_u^2 σu2 代表总多基因方差后验估计值(Lee et al., 2013)。SNP 与抗病表型间的统计关联强度完全基于各标记所估算的 GEV 比例进行排序与界定。为了在跨物种横向比对中最大程度保持基因组候选区间的包容性与代表性,我们将筛选门槛设定为:在各物种内部,凡解释该性状遗传方差至少达到 1%(GEV ≥ 1 % \ge 1\% ≥1%)的全部基因组染色体区域均被纳入后续深入挖掘。
Source: p.4 S023
Original:
The heritability values were calculated as:
中文:
性状的广义基因组遗传力数值依据下式综合计算:
Equation (3):
h 2 = V A ′ V A ′ + σ e 2 (EQ3) h^2 = \frac{V_A'}{V_A' + \sigma_e^2} \quad \text{(EQ3)} h2=VA′+σe2VA′(EQ3)
Source: p.4 E003
Variable Definitions / 变量释义:
- h 2 h^2 h2: 基因组窄义遗传力(genomic heritability)。
- V A ′ V_A' VA′: 总加性遗传方差,由全基因组 SNP 加性标记方差与基于系谱亲缘关系的多基因遗传方差加和构成:
V A ′ = 2 σ a 2 π ∑ i = 1 n p i q i + σ u 2 V_A' = 2 \sigma_a^2 \pi \sum_{i=1}^n p_i q_i + \sigma_u^2 VA′=2σa2πi=1∑npiqi+σu2 - σ e 2 \sigma_e^2 σe2: 环境残差方差(residual variance)。
Source: p.4 S024
Original:
where V A ′ V_A' VA′ is the total additive genetic variance estimated as the sum of additive marker ( 2 σ a 2 π ∑ p i q i ) \left( 2 \sigma_a^2 \pi \sum p_i q_i \right) (2σa2π∑piqi) and the polygenic pedigree based ( σ u 2 ) (\sigma_u^2) (σu2) additive genetic variance.
中文:
式中: V A ′ V_A' VA′ 代表总加性遗传方差,定义为所有有效加性标记方差之和 ( 2 σ a 2 π ∑ p i q i ) \left( 2 \sigma_a^2 \pi \sum p_i q_i \right) (2σa2π∑piqi) 与基于系谱亲缘矩阵的多基因加性遗传方差 ( σ u 2 ) (\sigma_u^2) (σu2) 的总和;分母中的 σ e 2 \sigma_e^2 σe2 则为环境残差方差。
比较基因组学与候选基因优先级分类 / Comparative Genomic Analysis
Source: p.4 S025
Original:
Initially, sequence homologies between chromosomes containing regions with SNPs explaining more than 1% of the genetic variance were compared. Synteny among these chromosomes was identified using Symap (Soderlund et al., 2011). The relationship between the chromosomes from CS, RT, and AS and the association between SNPs and resistance to P. salmonis (Manhattan plot) was plotted using Circos (Krzywinski et al., 2009).
中文:
在比较基因组学分析的第一阶段,我们首先横向比对了在三种鲑鱼中包含解释遗传方差超过 1% 的显著 SNP 所落座的对应染色体序列同源性。利用 Symap 软件系统识别并建立了这些同源染色体之间的全基因组同源共线性区块(synteny blocks;Soderlund et al., 2011)。随后,借助 Circos 基因组多组学校验可视化平台(Krzywinski et al., 2009),将银鲑、虹鳟和大西洋鲑染色体间的同源共线拓扑关联与各物种抗鲑立克次体 GWAS 关联信号曼哈顿图(Manhattan plot)统合绘制在同心环状图中呈现。
Source: p.4 S026
Original:
To identify candidate genes associated with P. salmonis resistance, we used a comparative genomic analysis among CS, RT, and AS. For this, we mapped the location of each SNP that explained 1% or more of the genetic variance for the trait on the reference genome (NCBI_RefSeq) of each species: CS (GCF_002021735.1), RT (GCF_002163495.1; Pearse et al., 2018), and AS (GCF_000233375.1; Lien et al., 2016). Subsequently, we retrieved the sequences of all the genes (and their protein products) adjacent to each SNP within a window of 1 Mb (500 kb downstream and 500 kb upstream to the associated SNP). We then used this information to apply the following criteria to classify and prioritize functional candidate genes by comparing the genomic regions involved in P. salmonis defined as DD and BS within and among the three species:
中文:
为了高精度锁定与鲑立克次体抗性紧密连锁的核心功能候选基因,我们在银鲑、虹鳟和大西洋鲑三者之间实施了深入的比较基因组学联合挖掘。首先,将各物种中解释遗传变异比例不低于 1%(GEV ≥ 1 % \ge 1\% ≥1%)的所有关键 SNP 探针的物理染色体坐标,精准锚定映射到各大物种最新的 NCBI_RefSeq 官方参考基因组中:银鲑选用 GCF_002021735.1,虹鳟选用 GCF_002163495.1(Pearse et al., 2018),大西洋鲑选用 GCF_000233375.1(Lien et al., 2016)。随后,以每个关联 SNP 位点为物理几何中心,向上下游各拓展 500 kb 构建总跨度为 1 Mb 的候选物理窗口(即关联 SNP 上游 500 kb 至下游 500 kb),完整抓取并提取该窗口内落座的所有基因全长序列及其编码的蛋白质多肽产物。基于抓取获取的完整基因信息,我们在物种内部及物种之间对比由 DD 和 BS 两种表型所界定的抗病关联区段,并确立执行了以下四项逐级优先级归类标准:
Source: p.4 S027
Original:
i) The complete set of genes was identified and classified into homologous superfamilies based on InterPro (Mitchell et al., 2019) protein domain signatures using Blast2GO software version 5.2.5 (Götz et al., 2008; referred to as Group A);
中文:
i)A 组(Group A - 结构域同源超家族): 基于 InterPro 数据库中权威收录的保守蛋白特征结构域签名(protein domain signatures;Mitchell et al., 2019),利用 Blast2GO 5.2.5 版专业生物学功能注释平台(Götz et al., 2008),对窗口内所有编码蛋白进行全景扫描与结构域分类,聚类归纳出各个同源超家族;
Source: p.4 S028
Original:
ii) Orthologous and paralogous genes among species were identified using the ProteinOrtho tool (Lechner et al., 2011). Multidirectional alignments were performed using the full-length sequences among complete sets of proteins encoded in each of the three species to obtain orthologous groups, with a 35% threshold for identity and similarity (Group B);
中文:
ii)B 组(Group B - 跨物种直系/旁系同源基因): 借助高性能大规模直系同源检测工具 ProteinOrtho(Lechner et al., 2011),以三种鲑科鱼类候选窗口编码的全长蛋白质氨基酸序列为输入,开展高维全对全多向序列比对;以同一性(identity)与相似性(similarity)均不低于 35% 作为严苛界值,鉴定出跨物种间保守共有的直系同源群与旁系同源簇;
Source: p.4 S029
Original:
iii) The complete set of genes within 1 Mb windows adjacent to SNPs explaining the highest proportion of the genetic variation for each trait (leader SNP) was recovered and classified as high priority genes (Group C); and
中文:
iii)C 组(Group C - 先导主效标记紧邻基因): 在每个物种针对每种表型(DD 或 BS)解释遗传方差比例最高的核心主效应 SNP(即 leader SNP,先导标记)周围 1 Mb 物理窗口内的全部落座基因,全部提取并赋予高优先级候选基因评级;
Source: p.4 S030
Original:
iv) The complete set of genes located at the intersection of more than 1 Mb windows within a species was also identified and considered as high priority genes (Group D).
中文:
iv)D 组(Group D - 多标记窗口重叠交集基因): 在某一单一物种内部,凡同时位于两个或两个以上 1 Mb 显著 SNP 侧翼窗口相重叠交集物理区域中的全部候选基因,同样被遴选确立为重点高优先级功能候选基因。
研究结果 / Results
攻毒试验表型变异与遗传参数估计 / Challenge Test and Genetic Parameters
Source: p.4 S031
Original:
There was considerable phenotypic variation for P. salmonis resistance across fish species (Figure 1). The average cumulative mortality for different families ranged from 5% to 81%, 8% to 100%, and 8.3% to 73.7% for CS, RT, and AS, respectively. This result suggests that the phenotypic variation for this trait could be related to the genetic background on each species.
中文:
人工接种攻毒试验显示,三种鲑科鱼类对鲑立克次体的抗病力均表现出极其显著且广泛的家系间表型变异(图 1)。在各个物种不同全同胞家系间,试验期末的平均累积死亡率波动范围巨大:银鲑家系为 5% 至 81%,虹鳟家系为 8% 至 100%,而大西洋鲑家系则介于 8.3% 至 73.7% 之间。如此巨大的表型离散分布表明,宿主对该病原菌的抗感染存活差异与其各自独特的群体遗传背景及个体遗传分化紧密相关。
Fig. 1. 银鲑、虹鳟与大西洋鲑攻毒试验后全同胞家系累积死亡率分布图
Placed near: p.4 S031
Source: p.4 F001

Original caption:
FIGURE 1 | Cumulative mortality by family after P. salmonis experimental infection of CS, RT, and AS. For CS, RT, and AS, a total of 107, 105, and 118 full-sib families were experimentally challenged.
中文图注:
图 1 | 银鲑(CS)、虹鳟(RT)和大西洋鲑(AS)人工感染鲑立克次体后各家系的累积死亡率分布曲线。对于 CS、RT 和 AS,分别对 107、105 和 118 个全同胞家系实施了严格的受控人工攻毒试验。图中各鱼种家系按累积死亡率升序排列,直观展现了三大物种内部巨大的抗病遗传分化阶梯。
Reading note:
该图直观揭示了家系间悬殊的存活表现差。虹鳟对鲑立克次体最为敏感脆弱,部分极度易感家系最终死亡率达到 100%,最抗病家系死亡率亦有 8%;银鲑与大西洋鲑则展现出从高抗(<10% 死亡率)到极度易感(>70-80% 死亡率)的宽幅连续梯度,证明三个物种群体内部均蕴藏着极其丰富的抗病选育优良等位基因变异。
Source: p.5 S032
Original:
Estimated heritabilities for P. salmonis resistance were significant for the three species, indicating the feasibility to improve the trait by means of artificial selection (Table 1). The genomic heritability values for DD were 0.32 for CS, 0.48 for RT, and 0.24 for AS. When resistance was defined as BS, genomic heritability estimates increased to 0.88, 0.64, and 0.32 for CS, RT, and AS, respectively, representing moderate to high levels of genetic variation for P. salmonis resistance.
中文:
统计模型求算表明,三种鲑鱼对鲑立克次体的抗性遗传力估计值均达到了统计学极显著水平,强有力地证实了通过人工选择育种大幅改良鱼群抗病表型的理论与实践可行性(表 1)。以死亡天数(DD)度量时,银鲑、虹鳟和大西洋鲑的基因组遗传力估计值分别为 0.32、0.48 和 0.24。而当将抗病性状定义为试验终结时是否存活的二元存活(BS)指标时,三种鱼的基因组遗传力估计值均呈现显著跃升,分别高达 0.88(银鲑)、0.64(虹鳟)和 0.32(大西洋鲑),表明三大鲑科鱼类群体内部均保留着对鲑立克次体抗性从中等至极高水平的丰富加性遗传变异。
Table 1. 三种鲑科鱼类抗鲑立克次体性状(死亡天数 DD 与二元存活 BS)的加性遗传方差、残差方差及遗传力估计表
Placed near: p.5 S032
Source: p.5 T001
| 物种代码 (Species) | 英文名称 (Common Name) | 拉丁学名 (Scientific Name) | 性状指标 (Trait) | 加性遗传方差 ( V a ′ V_a' Va′) | 残差方差 ( σ e 2 \sigma_e^2 σe2) | 基因组遗传力 ( h 2 h^2 h2) | 估计标准误 (SD) |
|---|---|---|---|---|---|---|---|
| CS | Coho salmon (银鲑) | Oncorhynchus kisutch | DD (死亡天数) | 28.91 | 60.70 | 0.32 | 0.07 |
| CS | Coho salmon (银鲑) | Oncorhynchus kisutch | BS (二元存活) | 7.53 | 1.00 | 0.88 | 0.03 |
| RT | Rainbow trout (虹鳟) | Oncorhynchus mykiss | DD (死亡天数) | 30.42 | 32.71 | 0.48 | 0.04 |
| RT | Rainbow trout (虹鳟) | Oncorhynchus mykiss | BS (二元存活) | 1.87 | 1.00 | 0.64 | 0.05 |
| AS | Atlantic salmon (大西洋鲑) | Salmo salar | DD (死亡天数) | 16.52 | 53.17 | 0.24 | 0.04 |
| AS | Atlantic salmon (大西洋鲑) | Salmo salar | BS (二元存活) | 0.47 | 1.00 | 0.32 | 0.05 |
Original Table Caption:
TABLE 1 | Estimates of total additive genetic variance ( V a ′ V_a' Va′), residual variance ( σ e 2 \sigma_e^2 σe2), heritability ( h 2 h^2 h2), and standard deviation (SD) for resistance against P. salmonis in three salmonid species.
中文表注:
表 1 | 三种鲑科鱼类对鲑立克次体抗性的总加性遗传方差 ( V a ′ V_a' Va′)、残差方差 ( σ e 2 \sigma_e^2 σe2)、遗传力 ( h 2 h^2 h2) 及其后验标准差 (SD) 估计值。DD:攻毒后存活死亡天数(连续变量);BS:攻毒终期二元存活状态(阈值模型,残差方差固定为 1.00)。所有物种在 BS 定义下的遗传力均高于 DD,银鲑 BS 遗传力更达到惊人的 0.88。
贝叶斯 GWAS 剖析遗传架构 / GWAS Analysis
Source: p.5 S033
Original:
A total of 580 CS (9,389 SNPs), 1,929 RT (24,916 SNPs), and 2,383 AS (42,624 SNPs) were retained after QC. For CS and RT, we found relatively few SNPs explaining a moderate to high percentage of genetic variance for P. salmonis resistance. In contrast, for AS, a large number of SNPs with small effect were found and the percentage of GEV by a single marker was not higher than 5% (Figure 2; Supplementary Figure 1).
中文:
在执行严格的样本与标记质量控制(QC)过滤后,最终用于全基因组贝叶斯关联分析的有效数据集包含:银鲑 580 尾个体(9,389 个合格 SNP)、虹鳟 1,929 尾个体(24,916 个合格 SNP)以及大西洋鲑 2,383 尾个体(42,624 个合格 SNP)。关联分析显示:在银鲑和虹鳟中,仅有相对极少数的几个关键 SNP 标记解释了鲑立克次体抗性中等至极高比例的遗传变异;与此形成鲜明对照的是,在大西洋鲑中,抗病性状由遍布全基因组的海量微效 SNP 共同调控,其中没有任何单一标记所解释的遗传方差比例(GEV)能够超过 5%(图 2;附图 1)。
Source: p.5 S034
Original:
Although there were multiple shared syntenic regions with associated SNPs (4 for DD and 5 for BS) in two species, there were no shared syntenic regions where all three species had common associated SNPs (Figure 2).
中文:
从跨物种同源共线性(synteny)角度审视,尽管在任意两个鲑科鱼类之间能够检测到多处包含抗病关联显著 SNP 的共享同源共线染色体区段(其中死亡天数 DD 检测到 4 个共享区段,二元存活 BS 检测到 5 个共享区段),但在全基因组范围内,并未发现任何一个共线染色体区段能够被三种鲑鱼全部同时共享并均定位到显著关联 SNP(图 2)。
Fig. 2. 死亡天数 (A) 与二元存活 (B) 全基因组曼哈顿图与跨物种染色体同源共线性 Circos 环状图
Placed near: p.5 S034
Source: p.5 F002

Original caption:
FIGURE 2 | Circos plot for P. salmonis resistance as DD (A) and as BS (B). The inner ribbons mark syntenic regions among CS (green), RT (red), and AS (blue). Manhattan plots are shown on the outer ring, with significant associations plotted in red (values ≥ 1 \ge 1 ≥1).
中文图注:
图 2 | 鲑立克次体抗性在以死亡天数 DD (A) 与二元存活 BS (B) 评估时的 Circos 全景关联环状图。最内圈彩色缎带标注了银鲑(CS,绿色)、虹鳟(RT,红色)与大西洋鲑(AS,蓝色)之间的染色体同源共线对应区段。外层环圈展示了各物种的全基因组 GWAS 曼哈顿图,其中解释遗传变异比例 ≥ 1 % \ge 1\% ≥1% 的显著关联位点以高亮红点标出。
Reading note:
该图直观反映了三文鱼祖先经历第四轮全基因组加倍(SS4R)后分化出的三大物种染色体演化与抗病信号重排。虽然银鲑与虹鳟、或虹鳟与大西洋鲑之间存在局部抗病共线对应块,但三者并未在同一染色体共线锚点上发生趋同收敛,提示抗病机制在三者中可能存在不同的主效基因位点或上位性重组。
Source: p.5 S035
Original:
Figure 3 (and Supplementary Figure 2) highlights the different genetic architecture for resistance to P. salmonis among the three salmonid species studied. For CS, the top 200 SNPs explained about 70% and 90% of genetic variance for DD and BS, respectively, and just a marker located in chromosome 29 represented more than 50% of total genetic variance for BS. For RT, the top 200 SNPs explained 90% and 80% for DD and BS, respectively, whereas, in AS, they explained slightly more than 30% for both traits.
中文:
图 3(以及附图 2)清晰突显了所研究的三种鲑科鱼类在鲑立克次体抗性遗传架构上存在的根本性分歧。对于银鲑,全基因组关联排名前 200 个的核心 SNP 分别解释了死亡天数 DD 约 70% 以及二元存活 BS 约 90% 的绝大部分遗传方差;尤为震撼的是,仅位于银鲑第 29 号染色体上的单个先导主效 SNP,就独立解释了二元存活 BS 超过 50% 的全部遗传变异!对于虹鳟,前 200 个核心 SNP 对 DD 和 BS 的累积解释变异比例也分别高达 90% 和 80%。然而,在大西洋鲑中,排名前 200 个 SNP 对这两个抗病性状的累积解释比例仅仅略超 30%。
Fig. 3. 贝叶斯 GWAS 前 200 个核心标记对死亡天数 (A) 与二元存活 (B) 遗传变异的累积解释百分比曲线
Placed near: p.5 S035
Source: p.6 F003

Original caption:
FIGURE 3 | Cumulative percentage of the GEV by the top 200 markers from Bayesian GWAS for resistance to P. salmonis measured as DD (A) and BS (B) in CS, RT, and AS.
中文图注:
图 3 | 银鲑(CS,绿色实线)、虹鳟(RT,红色实线)与大西洋鲑(AS,蓝色实线)在贝叶斯 GWAS 分析中排名前 200 个 SNP 标记对鲑立克次体抗性遗传方差(GEV)的累积解释百分比曲线。图 (A) 为死亡天数 DD;图 (B) 为二元存活 BS。
Reading note:
该曲线极其生动地界定了"寡基因控制"与"多基因微效控制"的数学分野:银鲑与虹鳟曲线在极少标记处便呈现陡峭攀升并迅速趋近 80%-90% 平台期,证实少数 QTL 掌控全局;而大西洋鲑曲线呈现低斜率匀速上升,前 200 个标记仅解释约 30%,表明其抗病性状是由全基因组海量微效基因座共同驱动的典型复杂性状。
Source: p.5 S036
Original:
These results suggested that CS and RT both appear to have oligogenic control with few markers having large effect loci, whereas the small effect of loci suggested the polygenic nature for resistance to P. salmonis in AS.
中文:
上述数学关联与累积方差证据明确表明:银鲑和虹鳟两者对鲑立克次体的抗性均呈现出由极少数大效应或中等效应基因座主导的寡基因控制(oligogenic control)模式;而大西洋鲑中微弱分散的基因座效应值则无可辩驳地揭示了其抗病性状遵循典型的微效多基因控制(polygenic control)遗传本质。
跨物种比较基因组学与核心候选基因挖掘 / Comparative Genomic Analysis
Source: p.5 S037
Original:
We mapped the location of each SNP that explained 1% or more of the genetic variance for both DD and BS to the reference genome of CS, RT, and AS and searched for genes within 1 Mb windows flanking each SNP. This search allowed us to identify 533 unique genes that encoded 957 proteins. The complete list of genes and proteins can be found in Supplementary Table S1: Sheets 1 to 6.
中文:
我们将针对 DD 和 BS 两种性状解释遗传方差不低于 1%(GEV ≥ 1 % \ge 1\% ≥1%)的所有关键 SNP 探针精准映射到银鲑、虹鳟和大西洋鲑的参考基因组上,并在每个 SNP 两侧拓展出 1 Mb 的候选物理窗口进行基因遍历检索。这一全基因组比对检索使我们成功捕获并鉴定了 533 个非冗余独特基因,共编码 957 种剪接体蛋白质多肽。这 533 个候选基因及蛋白质的完整详尽清单收录于本论文补充材料表 S1(Supplementary Table S1: Sheets 1 至 6)。
Source: p.6 S038
Original:
To prioritize functional candidate genes, we annotated and classified the complete set of encoded proteins in homologous superfamilies for each trait and species based on InterPro protein domain signatures. We identified 194 and 129 homologous superfamilies for DD and BS, respectively, 103 of which were shared between traits (Supplementary Table S1: homologous superfamilies). The homologous superfamilies and the number of proteins present in at least two salmonid species are shown in Figure 4. Remarkably, about 30% of the proteins from genes present in regions associated with DD belong to five homologous superfamilies P-loop containing nucleoside triphosphate hydrolase (also known as P-loop_NTPase), immunoglobulin-like fold, zinc finger C2H2 superfamily, zinc finger RING/FYVE/PHD-type, and protein kinase-like domain superfamily. A total of 30% of proteins from genes present in regions associated with BS belong to only three homologous superfamilies (P-loop_NTPase, immunoglobulin-like fold, and immunoglobulin-like domain superfamily). Interestingly, the P-loop_NTPase superfamily contained the largest group of proteins for both traits, and at least one representative protein from each salmonid species belonged to this superfamily. Thirty-one of the proteins identified in this study are part of this superfamily, including some GTPases, kinesin and myosin proteins, and ATP-dependent RNA helicases Supplementary Table S1, sheet: P-loop NTPases (Group_A).
中文:
为对候选基因实施基于生物学功能的科学优先级排序,我们依托 InterPro 保守蛋白质特征结构域签名,针对不同表型和物种对提取的全部编码蛋白实施了同源超家族(homologous superfamilies)系统注释与聚类。在 DD 和 BS 候选区间内分别鉴定出 194 个和 129 个同源超家族,其中两个性状共有重叠的超家族达 103 个(补充材料表 S1: homologous superfamilies)。图 4 详细展示了在至少两种鲑科鱼类中保守共有的同源超家族及其包含的蛋白质数量分布。尤为瞩目的是,在与死亡天数(DD)显著关联的基因组区域所编码的全部蛋白质中,约有 30% 集中归属于五个核心同源超家族:即含 P-环的三磷酸核苷水解酶超家族(P-loop containing nucleoside triphosphate hydrolase,简称为 P-loop_NTPase)、免疫球蛋白样折叠超家族(immunoglobulin-like fold)、C2H2 型锌指超家族(zinc finger C2H2 superfamily)、RING/FYVE/PHD 型锌指超家族(zinc finger RING/FYVE/PHD-type)以及蛋白激酶样结构域超家族(protein kinase-like domain superfamily)。而在与二元存活(BS)显著关联的区域中,同样有高达 30% 的蛋白质仅仅汇聚于三个超家族(P-loop_NTPase、免疫球蛋白样折叠以及免疫球蛋白样结构域超家族)。极其引人深思的是,无论对于 DD 还是 BS,P-loop_NTPase 超家族均是富集蛋白规模最大的第一大超家族,且三种鲑鱼均至少有一类具有代表性的核心抗病蛋白归属该家族。本研究锁定的候选蛋白中有 31 个隶属于该超家族,包括多种关键 GTP 酶(GTPases)、驱动蛋白(kinesin)、肌球蛋白(myosin)以及 ATP 依赖性 RNA 解旋酶等详见补充材料表 S1 工作表:P-loop NTPases (Group_A)。
Fig. 4. 位于解释遗传变异 >1% 的 SNP 侧翼窗口中至少在两种鲑鱼中共享的 InterPro 同源超家族蛋白丰度分布柱状图
Placed near: p.6 S038
Source: p.6 F004

Original caption:
FIGURE 4 | Homologous superfamilies (InterPro) adjacent to the complete set of SNPs that explain more than 1% of the genetic variance of resistance to SRS measured as DD (A) and BS (B). Bars represent the abundance of genes in each homologous superfamily present in at least two salmonid species. CS, RT, and AS.
中文图注:
图 4 | 邻近解释 SRS 抗性遗传方差 >1% 的完整 SNP 集合侧翼窗口内、在至少两种鲑科鱼类中共有检出的 InterPro 同源超家族分布柱状图。(A) 为死亡天数 DD;(B) 为二元存活 BS。柱状条长度代表各同源超家族中包含的蛋白质丰度总数,颜色区分银鲑(CS,绿色)、虹鳟(RT,红色)与大西洋鲑(AS,蓝色)。
Reading note:
该图揭示了抗病分子演化中的功能收敛现象。虽然三种鱼的染色体共线定位不完全重叠,但其受选择的候选基因高度汇聚在 P-loop_NTPase、免疫球蛋白折叠(Ig-like)、锌指转录因子和激酶结构域上,证实了宿主抵御胞内细菌感染所依赖的底层生化机制与信号级联在鲑科演化历史上高度保守。
Source: p.7 S039
Original:
To complement these analyses, we looked for orthologous proteins through multi-directional alignments using full-length sequences of the complete set of proteins for each species (Group B). Only five groups of orthologous genes were identified in at least two species, highlighting three non-receptor tyrosine-protein kinases (nr-TPK) with representative genes in the three species for DD and two species for BS. In addition, for DD, two ATP-dependent RNA helicases (DDX) and two Ras-related proteins (RAB) were identified in CS and RT, whereas two FYVE, RhoGEF/PH domain-containing proteins (FGD) were identified in RT and AS. For BS, two fatty acid-binding proteins (L-FABP) and two ankyrin repeat domain-containing proteins were identified in CS and RT Supplementary Table S1, sheet: Orthologous genes (Group_B). The proteins nr-TPK, DDX, and L-FABP are also encoded by genes adjacent to SNPs that explained the highest proportion for the genetic variance (leader SNP) for both trait definitions (Group C).
中文:
为与上述结构域分析形成互补互证,我们进一步采用三大鲑鱼完整候选蛋白组的全长氨基酸序列实施了高维多向比对,旨在鉴定真正的跨物种直系同源蛋白(Group B)。在全基因组范围内,仅在至少两个物种中保守鉴定出五组直系同源基因群,其中最瞩目的是非受体型酪氨酸蛋白激酶(non-receptor tyrosine-protein kinase, nr-TPK)基因群------其代表性直系同源基因在 DD 性状中被三种鱼类共同检出,在 BS 性状中被两种鱼类共同检出。此外,针对死亡天数(DD),在银鲑和虹鳟中共同鉴定出两个 ATP 依赖性 RNA 解旋酶(DDX)基因和两个 Ras 相关蛋白(RAB)小 GTP 酶基因;而在虹鳟和大西洋鲑中共同检测到两个含 FYVE、RhoGEF 与 PH 结构域的蛋白(FGD)。针对二元存活(BS),在银鲑和虹鳟中共同检测出两个肝型脂肪酸结合蛋白(L-FABP)和两个含锚蛋白重复结构域的蛋白详见补充材料表 S1 工作表:Orthologous genes (Group_B)。特别关键的是,nr-TPK、DDX 和 L-FABP 这三类蛋白所对应的编码基因,在针对两种抗病性状定义下均同时精准落座于解释遗传方差比例最高的主效先导标记(leader SNP)紧邻物理区间内(即同时归属 Group C)。
Source: p.7 S040
Original:
Group C contained other genes ( n = 42 n=42 n=42) that encoded proteins such as myosin-IIIb (MYO3B), ATP-dependent RNA helicase (TDRD9), kinesin protein (KIF15), and kinesin protein (KIF2C) that are also included into the P-loop_NTPase superfamily as well as members of the orthologous groups such as FABP. Other genes encoding proteins classically associated with immune response such as tripartite motif-containing protein 35 (TRIM35) and lysozyme C II (LYZ2) are also part of this group. A complete list of these genes and proteins is in Supplementary Table S1, sheet: Adjacent to leader SNP (Group_C).
中文:
先导标记窗口核心基因组 C(Group C,共计 n = 42 n=42 n=42 个基因)还包含了诸如肌球蛋白 IIIb(myosin-IIIb, MYO3B)、ATP 依赖性 RNA 解旋酶(TDRD9)、驱动蛋白家族成员 KIF15 以及驱动蛋白 KIF2C 等高价值编码基因。这些蛋白质分子在结构上无一例外均归入 P-loop_NTPase 超家族,且包含直系同源群成员 FABP。此外,经典免疫防御级联相关的关键分子,如含三联基序蛋白 35(tripartite motif-containing protein 35, TRIM35)以及 C-II 型溶菌酶(lysozyme C II, LYZ2)等,亦被牢牢锚定在该高优先级组别中。该类核心基因及多肽产物的完整收录清单见补充材料表 S1 工作表:Adjacent to leader SNP (Group_C)。
Source: p.7 S041
Original:
Group D was composed of genes ( n = 58 n=58 n=58) located adjacent to more than one SNP simultaneously (within overlapped windows). Among them, we identified GTPase IMAP family member 4 (GIMAP4), GTPase IMAP family member 8 (GIMAP8), NLR family CARD domain-containing protein 3 (NLRC3), ADP-ribosylation factor protein 5B (ARL5B), voltage-dependent L-type calcium channel subunit beta-2 (CACNB2), and heparan sulfate glucosamine 3-O-sulfotransferase 3A1 (HS3ST3A1), all of which are also P-loop_NTPases. In addition, we identified histidine triad nucleotide-binding protein 1 (HINT1), which is also adjacent to the leader SNP for DD in AS, and other genes associated with immune response such as collectin-12 (COL12), macrophage mannose receptor 1 (MRC1), and tapasin-related protein (TAPBPR). A complete list of these genes and proteins can be found in Supplementary Table S1, sheet: Genes overlapped windows (Group_D). Additionally, the gene that codes for NACHT, LRR, and PYD domains-containing protein 12 (NLRP12) was found in Groups A, C, and D.
中文:
多标记交集基因组 D(Group D,共计 n = 58 n=58 n=58 个基因)由同时落座于多个独立显著关联 SNP 侧翼重叠候选窗口交集区域内的基因所构成。在该组别中,我们深度锁定了免疫相关 GTP 酶 IMAP 家族成员 4(GIMAP4)、IMAP 家族成员 8(GIMAP8)、NLR 家族含 CARD 结构域蛋白 3(NLRC3)、ADP-核糖基化因子蛋白 5B(ARL5B)、电压门控 L 型钙通道 β − 2 \beta-2 β−2 亚基(CACNB2)以及硫酸乙酰肝素氨基葡萄糖 3-O-磺基转移酶 3A1(HS3ST3A1),所有这些基因同样属于广义 P-loop_NTPases 家族。此外,我们鉴定了组氨酸三联体核苷酸结合蛋白 1(HINT1,该基因同时紧邻大西洋鲑 DD 表型的先导主效 SNP),以及一系列与宿主免疫清除直接相关的经典分子,如凝集素-12(collectin-12, COL12)、巨噬细胞甘露糖受体 1(macrophage mannose receptor 1, MRC1)以及抗原呈递伴侣 tapasin 相关蛋白(tapasin-related protein, TAPBPR)。此类基因全景列表见补充材料表 S1 工作表:Genes overlapped windows (Group_D)。特别值得强调的是,编码含 NACHT、LRR 与 PYD 结构域蛋白 12(NLRP12)的基因被同时检出且横跨 Group A、Group C 和 Group D 三大评判组别!
Source: p.7 S042
Original:
We identified several candidate genes associated with P. salmonis resistance ( n = 120 n=120 n=120), which were present in at least one of the groups described previously. These genes are associated with the following biological processes: dependence on kinase activity, GTP hydrolysis, helicase activity, lipid metabolism, cytoskeletal dynamics, and inflammation. To rank the genes, we scored them based on the counting of each of them across following categories: i) species (CS, RT, and AS), ii) trait definitions (DD and BS), and iii) groups (A--D); thus, the maximum score for one particular gene was equal to 9. The prioritized functional candidate genes based on the score described above are shown in Table 2 and the complete list of unique candidate genes ( n = 120 n=120 n=120) can be found in Supplementary Table S1. sheet: Candidate genes.
中文:
经过上述系统筛选,我们共锁定并汇总了 120 个至少符合上述分类标准之一的抗鲑立克次体核心候选基因( n = 120 n=120 n=120)。生物学功能注释显示,这些候选基因主要调控以下关键细胞生命进程:激酶活性依赖性级联、GTP 水解调控、解旋酶催化活性、脂质代谢运转、细胞骨架超微动态变化以及机体炎症应答。为对全部候选基因建立客观严谨的优先级排序矩阵,我们构建了一套多维综合计分模型,根据每个基因在以下三个维度的检出频次进行累加计分:i)涵盖物种跨度(CS、RT、AS,最高 3 分);ii)涵盖表型性状定义跨度(DD 与 BS,最高 2 分);iii)涵盖功能优先级评级分组(Group A--D,最高 4 分)。因此,任一单一基因在理论上所能获得的综合得分最高上限为 9 分。表 2 详细列出了依据上述评分模型严格排序的前 21 个最高优先级功能候选基因,120 个候选基因的完整全景数据收录于补充材料表 S1 工作表:Candidate genes。
Table 2. 综合物种、性状与分组标准打分排序的 21 个抗鲑立克次体核心高优先级功能候选基因汇总表
Placed near: p.7 S042
Source: p.7 T002
| 基因符号 (Gene Symbol) | 编码蛋白质功能描述 (Protein Description) | 检出物种 (Species) | 关联抗病性状 (Trait) | 归属分类组别 (Group) | 综合优先级评分 (Score)a |
|---|---|---|---|---|---|
| NRTPK | 非受体型胞质酪氨酸蛋白激酶 (nr-TPK, cytosolic) | CS, RT, 和 AS (全物种检出) | DD 和 BS (双性状关联) | B--D | 8 |
| DDX | ATP 依赖性 RNA 解旋酶 DDX 家族 (DDX24 / DDX47) | CS 和 RT | DD | A--C | 6 |
| ARL5B | ADP-核糖基化因子蛋白 5B (ARL5B) | CS | DD 和 BS | A 和 D | 5 |
| LFABP | 肝型脂肪酸结合蛋白 (Fatty acid-binding protein, liver) | CS 和 RT | BS | B 和 C | 5 |
| GIMAP4 | GTP 酶 IMAP 家族成员 4 (GTPase IMAP family member 4) | RT | DD 和 BS | A 和 D | 5 |
| HS3ST3A1 | 硫酸乙酰肝素氨基葡萄糖 3-O-磺基转移酶 3A1 | AS | DD 和 BS | A 和 D | 5 |
| KIF2C | 驱动蛋白家族成员 KIF2C (Kinesin protein KIF2C) | RT | DD 和 BS | A 和 C | 5 |
| MYO3B | 肌球蛋白 IIIb (Myosin-IIIb) | CS | DD 和 BS | A 和 C | 5 |
| NLRP12 | 含 NACHT, LRR 与 PYD 结构域蛋白 12 | AS | DD | A, C, 和 D | 5 |
| RAB | Ras 相关小 GTP 酶 Rab 蛋白 (RAB1 / RAB18) | CS 和 RT | DD | B 和 C | 5 |
| CACNB2 | 电压门控 L 型钙通道亚基 β − 2 \beta-2 β−2 (L-type calcium channel β − 2 \beta-2 β−2) | CS | DD 和 BS | A 和 D | 5 |
| TDRD9 | ATP 依赖性 RNA 解旋酶 TDRD9 (Tudor domain-containing 9) | CS | DD | A 和 C | 4 |
| FGD | 含 FYVE, RhoGEF 与 PH 结构域蛋白 (FGD family) | RT 和 AS | DD | B | 4 |
| GIMAP8 | GTP 酶 IMAP 家族成员 8 (GTPase IMAP family member 8) | RT | DD | A 和 D | 4 |
| HINT1 | 组氨酸三联体核苷酸结合蛋白 1 (Histidine triad protein 1) | AS | DD | C 和 D | 4 |
| KIF15 | 驱动蛋白家族成员 KIF15 (Kinesin protein KIF15) | CS | DD | A 和 C | 4 |
| NLRC3 | 含 NACHT, LRR 与 CARD 结构域蛋白 3 | RT | DD | A 和 D | 4 |
| COL12 | 凝集素-12 (Collectin-12 / CL-12) | CS | BS | D | 3 |
| LYZ2 | 溶菌酶 C II (Lysozyme C II) | AS | DD | C | 3 |
| MRC1 | 巨噬细胞甘露糖受体 1 (Macrophage mannose receptor 1) | CS | BS | D | 3 |
| TAPBPR | Tapasin 相关蛋白 / MHC-I 专职抗原装载伴侣 | RT | DD | D | 3 |
Original Table Caption:
TABLE 2 | Summary of candidate genes associated with P. salmonis resistance for CS, RT, and AS ranked by score, which is simply based on the number of appearance of each gene across the following categories: i) species (CS, RT, and AS), ii) trait definitions (DD and BS), and iii) groups (A--D).
中文表注:
表 2 | 银鲑(CS)、虹鳟(RT)和大西洋鲑(AS)抗鲑立克次体核心候选基因汇总表。表中基因按综合评分从高到低排列。评分规则基于各基因在以下三项分类中的出现频次直接累加:i)检出物种跨度(CS、RT 和 AS);ii)表型定义跨度(DD 和 BS);iii)功能组别跨度(A--D)。
a 单个基因所能获得的理论综合评分上限为 9 分。排名前 21 位的核心候选基因涵盖激酶、解旋酶、GTP 酶、细胞骨架、脂质代谢及先天免疫等核心分子通路。
讨论 / Discussion
Source: p.8 S044
Original:
The comparative genomic strategy used in this study allowed us to identify groups of homologous superfamilies and orthologous genes common to more than one species of salmonids among genes adjacent to SNPs that explain more than 1% of the genetic variance for P. salmonis resistance. To our knowledge, this is the first study that aims at identifying and prioritizing functional candidate genes involved in the differential response against bacterial infection by means of comparing results from GWAS mapping across different phylogenetically related salmonid species.
中文:
本研究所构建的跨物种比较基因组学策略,使我们得以在解释抗鲑立克次体遗传方差超过 1% 的关键 SNP 标记侧翼基因中,精准挖掘出在多种鲑科鱼类之间保守共有的同源超家族群与直系同源基因簇。据我们所知,这是国际上首次通过横向比对不同但在系统发育上高度近缘的鲑科物种的全基因组 GWAS 关联图谱,来系统鉴定并分级排序参与宿主对细菌感染产生差异化免疫应答的核心功能候选基因的开创性研究。
鲑立克次体抗性的遗传架构特征 / Genetic Architecture of Resistance to P. salmonis
Source: p.8 S045
Original:
Heritability estimates are in agreement with previous studies aimed to estimate levels of genetic variation for resistance to bacterial diseases in salmonid species. For instance, Vallejo et al. (2016), Vallejo et al. (2017) presented heritabilities ranging from 0.26 to 0.54 and from 0.31 to 0.48 for resistance to bacterial cold water disease in a farmed RT population. The levels of genetic variation observed in the current study are consistent or somewhat higher than previous estimates of heritabilities for resistance to P. salmonis depending on the species and the trait definition. For instance, previous heritability values for P. salmonis resistance estimated based on pedigree information reached a maximum of 0.16, 0.44, and 0.41 for CS, RT, and AS, respectively (Yáñez et al., 2013; Yáñez et al., 2014; Yáñez et al., 2016; Bassini et al., 2019). When heritability for P. salmonis resistance was estimated based on genomic information, the maximum values reported previously were 0.39 and 0.62 for AS and RT, respectively (Bangera et al., 2017; Yoshida et al., 2018a).
中文:
本研究估算获得的遗传力参数与先前针对鲑科鱼类细菌性疾病抗性遗传变异的大量学术报道高度吻合。例如,Vallejo 等人(2016, 2017)报道指出,虹鳟人工养殖群体对细菌性冷水病(BCWD)的抗性遗传力分别介于 0.26 至 0.54 以及 0.31 至 0.48 之间。本研究所测得的抗病遗传变异丰度水平,依据不同鱼种及不同表型定义,与先前针对鲑立克次体抗性的文献报道基本一致或略偏高。作为参照,以往基于传统家系系谱信息估算的抗鲑立克次体遗传力,在银鲑、虹鳟和大西洋鲑中的最高报道值分别为 0.16、0.44 和 0.41(Yáñez et al., 2013; Yáñez et al., 2014; Yáñez et al., 2016; Bassini et al., 2019)。而当依托高密度全基因组分子标记信息估算基因组遗传力时,先前文献在大西洋鲑和虹鳟中的最高记录值分别为 0.39 和 0.62(Bangera et al., 2017; Yoshida et al., 2018a)。
Source: p.8 S046
Original:
Our results show evidence of alleles of medium to large effect involved in resistance to P. salmonis in CS and RT. In contrast, for AS, our results suggest that if alleles of large effect do exist, they are at such low frequency that they individually explain a small proportion of the variance for resistance to P. salmonis. The identification of genomic regions harboring associated SNPs was based on GWAS using the Bayes C approach, which is more suitable for oligogenic traits (Habier et al., 2011). In a few cases, the same SNP was significantly associated with both trait definitions (DD and BS). This could be the result of pleiotropy, closely linked genes local linkage disequilibrium (LD), or by a strong correlation between both traits. For example, we observed the same SNP associated with DD and BS in CS (58185_41 and 24601_47) and RT (AX-89926208 and AX-89966072) among the top 10 SNPs explaining most of the genetic variance for the trait.
中文:
我们的研究明确证实:在银鲑和虹鳟中存在调控鲑立克次体抗性的中等乃至大效应等位基因;相反,在大西洋鲑中,分析结果表明即便存在大效应等位基因,其在群体中的发生频率也极低,以至于单个位点仅能解释极小份额的抗病遗传方差。本研究中定位富集关联 SNP 的基因组候选区间完全依托 Bayes C 贝叶斯统计建模,该方法在统计特性上被公认对寡基因控制性状(oligogenic traits)具有极高的探测效力与拟合优度(Habier et al., 2011)。在部分关键位点上,同一个 SNP 标记同时与两种表型性状定义(死亡天数 DD 与二元存活 BS)呈现出高度一致的极显著关联。这种现象可能是基因一因多效性(pleiotropy)、紧密物理连锁基因簇之间的局部连锁不平衡(local linkage disequilibrium, LD)所致,或者是由于两种表型性状本身所蕴含的高度内在生物学强相关所驱动。例如,在解释性状遗传方差最高的前 10 个核心 SNP 中,我们在银鲑(标记 58185_41 与 24601_47)以及虹鳟(标记 AX-89926208 与 AX-89966072)中均敏锐观察到了完全相同的 SNP 位点对 DD 和 BS 双重性状的同步主导控制。
Source: p.8 S047
Original:
Based on the LD of the AS population (measured as r 2 r^2 r2), the number of SNPs used for AS (~43K) should be enough to cover the entire genome (Barría et al., 2018b). There is a lack of studies aimed at evaluating the LD and population structure of the current farmed RT population. Based on results from a different RT farmed population, at least 20K SNPs are necessary to cover the whole genome (Vallejo et al., 2018). If the LD levels of the present RT population are similar to those reported by Vallejo et al. (2018), the 23K SNPs used here will most likely cover the whole genome. However, this is not the case for CS. Using a high-density SNP array, Rondeau et al. (In preparation) and Barría et al. (2019) suggested that at least 74K SNPs are necessary for whole-genome studies of the current CS population. The small number of SNPs assayed in this study for CS (9389) most likely affected the identification of markers with a moderate to high effect on resistance to P. salmonis in this species.
中文:
从群体连锁不平衡(LD衰减,以 r 2 r^2 r2 度量)的统计学规律来看,大西洋鲑养殖群体所使用的约 4.3 万个高质量 SNP 标记已完全足以无死角覆盖全基因组(Barría et al., 2018b)。对于现阶段人工养殖的虹鳟群体,系统评估其全基因组 LD 衰减与群体精细结构的专项研究目前仍显匮乏;然而参照另一处独立虹鳟商业选育群体的基准研究,至少需要 2 万个高质量 SNP 才能实现对虹鳟全基因组的有效覆盖(Vallejo et al., 2018)。若本研究中虹鳟群体的 LD 衰减速率与 Vallejo 等人(2018)的报道相仿,则本研究所采用的 2.3 万个合格标记将极大概率已充分覆盖其整个基因组。然而,对于银鲑而言情况却大不相同:利用超高密度 SNP 芯片的最新深度评估(Rondeau et al., 准备中;Barría et al., 2019)明确指出,针对当前的银鲑商业育种群体开展全基因组精细定位,至少需要 7.4 万个高密度 SNP 探针。因此,本研究中银鲑所分型的标记通量相对偏少(仅检出 9,389 个合格 SNP),这在客观上很可能会漏检部分对银鲑抗鲑立克次体具有中等至高等效应的潜在关键位点。
抗性相关候选蛋白质与分子功能机制 / Candidate Proteins Associated With the Resistance to P. salmonis
Source: p.8 S048
Original:
Whereas the complete set of proteins predicted from reference genomes of CS, RT, and AS consisted of 57,592, 58,925, and 97,738, respectively, the proteins neighboring SNPs associated with resistance (range of 1 Mb) represent less than 1% of the different proteomes. The characterization of the complete set of proteins among species established that the most prevalent homologous superfamily was the P-loop_NTPase. However, as this superfamily contains proteins with at least 21 functions (Shalaeva et al., 2018), it is possible that the high frequency of proteins identified from this group was due to the overall high representation in salmonid genomes. For this reason, we retrieved the sequences of 100 randomly selected proteins from the genomes of CS, RT, and AS and classified them into subfamilies (Supplementary Figure S3). The results indicate that P-loop_NTPase is not the most prevalent in any of the salmonid species, which suggests that this homologous superfamily is actually enriched in the regions analyzed and is not a consequence of their high representation in CS, RT, and AS genomes.
中文:
尽管从银鲑、虹鳟和大西洋鲑官方参考基因组中预测注释的完整全蛋白质组分别多达 57,592、58,925 和 97,738 个蛋白分子,但在抗病关联显著 SNP 侧翼 1 Mb 物理窗口内的落座蛋白质总量仅占各自全蛋白质组的不足 1%。跨物种全景功能特征分析确立了富集丰度最高的同源超家族当属含 P-环的三磷酸核苷水解酶(P-loop_NTPase)。然而,由于该超家族成员在真核生物中行使着至少 21 种极为广泛的生化功能(Shalaeva et al., 2018),一种潜在质疑在于:本研究所捕获的高频检出是否仅仅是由该家族在鲑鱼基因组中的天然高拷贝基底效应所致?为科学证伪这一假设,我们从银鲑、虹鳟和大西洋鲑全基因组中各自随机抽取了 100 个蛋白序列作为阴性背景对照集,并将其同等聚类划分为亚家族(补充材料图 S3)。对照实验确凿表明,在任何一个随机选取的鲑科鱼类背景蛋白集中,P-loop_NTPase 均绝非优势超家族。这一严谨的对照分析坚实证明:P-loop_NTPase 超家族在抗病关联区域中的高频现身,是真正在抗病基因座物理区间中受到显著统计学生物学富集的真实结果,而绝非基因组固有高丰度丰采所造成的虚假假象。
Source: p.8 S049
Original:
When traits are polygenic in nature, the identification of genes underlying them is a challenging task and often depends on previous knowledge of the function of genes adjacent to the associated SNPs (Jiang et al., 2014; Bouwman et al., 2018; Robledo et al., 2019). Our strategy was based on identifying orthologous proteins between the salmonid species and families of homologous proteins in the complete set of proteins adjacent to all the SNPs that explained more than 1% of the genetic variance, without searching for a specific function. The identification of genes directly associated with the innate immune response, after applying all the classification criteria, such as LYZ2, MRC1, COL12, and TAPBPR, suggests that our strategy was successful in finding strong functional candidate genes involved in resistance to P. salmonis. Interestingly, about 100 genes not classically associated with the immune system were also identified; among which, 17 were part of at least two of the groups described previously and hence are considered strong candidates for being responsible on trait variation (Table 2).
中文:
当目标表型具有多基因微效本质时,解析其底层的因果功能基因向来是一项极具挑战性的前沿难题,传统研究往往高度依赖于对关联 SNP 邻近基因先验已知功能的经验性推断(Jiang et al., 2014; Bouwman et al., 2018; Robledo et al., 2019)。与传统假说驱动法不同,我们的策略完全立足于无偏的数据驱动:在所有解释遗传方差大于 1% 的 SNP 邻近蛋白质集合中,全面搜寻在鲑科物种间共享的直系同源蛋白簇以及同源蛋白质超家族,而不事先人为锚定或偏倚于任何特定生化功能。在严格执行全部分类筛选标准后,所成功鉴定的基因集中直接涌现出一批与宿主先天免疫应答直接相关的权威经典分子,如 LYZ2、MRC1、COL12 以及 TAPBPR 等,这一结果强有力地印证了我们的跨物种比较挖掘策略在搜寻抗鲑立克次体核心功能候选基因上的极高准确性与可靠性。更为引人入胜的是,本研究还挖掘出约 100 个在传统免疫学文献中并未被归为经典免疫分子的"非典型"功能基因;其中有 17 个基因同时跨越满足至少两个前述优先级组别(表 2),因而被高度推荐作为驱动抗病表型变异的强效候选因果基因。
Source: p.8-9 S050
Original:
Previously, lysozymes have primarily been described as having a bacteriolytic activity against Gram-positive bacteria; however, the expression of LYZ2 has been shown to be induced in a resistant RT line in response to Flavobacterium psychrophilum infection (Langevin et al., 2012) and in AS families in response to P. salmonis infection (Pulgar et al., 2015), indicating that the transcriptional regulation of this enzyme in salmonids responds to Gram-negative bacterial infection. MRC1 and COL12 are membrane receptors that display several functions associated with innate immunologic defense, particularly in the recognition of carbohydrate structures of pathogens and as phagocytic receptors of bacteria, yeasts, and other pathogenic microorganisms (Harris et al., 1992; Ma et al., 2015). It has been reported that enhanced infection in human phagocytes with Francisella tularensis , a bacterium phylogenetically related to P. salmonis, is mediated by MRC1 (Schulert and Allen, 2006), whereas COL12 led to the activation of the alternative pathway of complement via association with properdin, a key positive regulator of the pathway by increment of the half-life of the C3 and C5 convertases (Ma et al., 2015). TAPBPR has been described as a second major histocompatibility complex class I-dedicated chaperone essential to providing specificity for T-cell responses against viruses and bacteria (Hermann et al., 2015) and the related protein tapasin has been shown to be induced in monocyte/macrophage in RT by chum salmon reovirus infection (Sever et al., 2014).
中文:
在传统生物化学认知中,溶菌酶主要被定义为特异性水解革兰氏阳性菌细胞壁肽聚糖的溶菌酶;然而近年来前沿免疫学证据揭示,在虹鳟抗病品系抵抗嗜冷黄杆菌(Flavobacterium psychrophilum )侵染过程中(Langevin et al., 2012)以及在大西洋鲑抗病家系抵抗鲑立克次体侵染应答中(Pulgar et al., 2015),LYZ2 基因的表达水平均呈现出显著的特异性上调诱导。这明确表明,鲑科鱼类该酶的转录调控网络在应对革兰氏阴性病原菌侵染时同样发挥着活跃的免疫应答功能。巨噬细胞甘露糖受体 1(MRC1)与凝集素-12(COL12)均为关键跨膜模式识别受体,在先天免疫防御中承担多重防御职责,特别是对病原体表面特定糖类碳水化合物结构的特异性识别,并作为介导细菌、酵母菌及其他致病微生物吞噬内吞的核心吞噬受体(Harris et al., 1992; Ma et al., 2015)。文献记载,在与鲑立克次体在系统发育上高度同源的土拉弗朗西斯菌(Francisella tularensis)侵染人类吞噬细胞的过程中,MRC1 直接充当了介导细菌高效入侵吞噬的关键介质(Schulert and Allen, 2006);而 COL12 跨膜受体则能通过直接结合备解素(properdin)------一种通过延长 C3 与 C5 转化酶半衰期来正向激活补体替代途径的核心关键因子------从而强力启动补体旁路途径级联激活(Ma et al., 2015)。此外,TAPBPR 已被权威确认为专司主要组织相容性复合体 I 类(MHC-I)抗原肽装载与质量监控的专职分子伴侣,对于确保针对病毒和胞内细菌的特异性 T 细胞细胞毒应答具有不可替代的关键作用(Hermann et al., 2015);而其同工分子 tapasin 在虹鳟单核/巨噬细胞系中已被证实可被大马哈鱼呼肠孤病毒感染强效激活表达(Sever et al., 2014)。
Source: p.9 S051
Original:
Another set of candidate proteins for SRS resistance in the three salmonid species studied are a cluster of cytosolic nr-TPKs. These proteins are a subgroup of the tyrosine kinase family, enzymes that phosphorylate tyrosine residues, and regulate many cellular functions, such as cell growth and survival, apoptosis, cell adhesion, cytoskeleton remodeling, and differentiation (Neet and Hunter, 1996). Although these proteins are not classically related to the response to pathogens, it has been described that the interaction of T- and B-cell antigen receptors with some nr-TPKs is critical to the activation of lymphocytes by an antigen (Sefton and Taddie, 1994). Moreover, some cellular signaling pathways are hijacked by intracellular pathogens, which can subvert protein phosphorylation to control host immune responses and facilitate invasion and dissemination (Haenssler and Isberg, 2011). It has been described that some bacterial effectors are injected into host cells through their secretion systems where they inhibit the Src kinase. In particular, the effector EspJ, an ADP-ribosyltransferase of the bacteria Escherichia coli and Citrobacter rodentium , regulates multiple host nr-TPKs in vivo by ADP-ribosylation, demonstrating that part of its target protein repertoire involves Src kinases such as YES1 and LYN as well as the adapter SYK (Young et al., 2014; Pollard et al., 2018), all of which were identified in this study in CS, RT, and AS. Remarkably, among the candidate genes, we also identified the small ARL5B, suggesting that an adequate regulation of the activity of nr-RTKs by ADP-ribosylation could be critical to combat P. salmonis infection.
中文:
在三大鲑科鱼类中被一致检出的另一组高优先级核心抗病候选蛋白,是一组定位于细胞质中的非受体型酪氨酸蛋白激酶(nr-TPKs)簇。这类蛋白质分子隶属于酪氨酸激酶超家族,专门催化靶蛋白酪氨酸残基发生特异性磷酸化,精细调控诸如细胞增殖与存活、细胞凋亡、细胞黏附、微丝细胞骨架重构以及细胞定向分化等多重基础生命活动(Neet and Hunter, 1996)。尽管在传统分类中它们并未被划为病原应答专职分子,但免疫学机理研究早已表明,T 细胞和 B 细胞抗原受体与特定 nr-TPKs 的物理相互作用,是外来抗原刺激触发淋巴细胞活化与克隆扩增的分子开关(Sefton and Taddie, 1994)。更为关键的是,许多致命性胞内寄生病原菌进化出了极其隐蔽的"宿主信号劫持"机制:它们通过篡改和颠覆宿主细胞原有的蛋白质磷酸化级联反应,以抑制宿主免疫杀伤并加速病原体自身的胞内侵入与全身播散(Haenssler and Isberg, 2011)。大量前沿机理研究证实,某些细菌毒力效应蛋白通过其特化分泌系统被直接注入宿主细胞浆内部,在胞内特异性抑制 Src 激酶的活性。特别是细菌毒力因子 EspJ------一种在大肠杆菌和啮齿枸橼酸杆菌中广泛存在的 ADP-核糖基转移酶------能够在活体内通过共价催化 ADP-核糖基化修饰,直接失活宿主的多株 nr-TPKs 激酶;其靶向攻击的蛋白储备库精确包含了 Src 家族核心激酶如 YES1、LYN 以及关键接头蛋白 SYK(Young et al., 2014; Pollard et al., 2018),而令人震撼的是,所有这些关键靶标激酶分子均在本研究中被三种鲑鱼跨物种无偏捕获!极为巧合而又耐人寻味的是,在我们的核心候选基因库中还精准检出了编码小 GTP 酶 ADP-核糖基化因子 5B(ARL5B)的基因,这高度提示:通过严密调控 ADP-核糖基化水平来维持宿主 nr-TPKs 激酶网络的活性稳态,是决定宿主能否抵御并清除鲑立克次体侵染的核心胜负手。
Source: p.9 S052
Original:
Other orthologous candidate genes identified in this study encode for proteins RAB1 and RAB18, both members of the GTPase superfamily. GTPases are a large family of hydrolase enzymes that bind and hydrolyze GTP and play an important role in signal transduction, protein translation, control and cellular differentiation, intracellular transport of vesicles, and cytoskeletal reorganization, among other cellular processes (Bourne et al., 1991). Specifically, RAB GTPases constitute a subfamily of small GTPases known as master regulators of intracellular membrane traffic (Stenmark, 2009). As P. salmonis drives the formation of host membrane-derived organelles, the development of these P. salmonis -containing vacuoles is dependent on the bacterium's ability to usurp the intracellular membrane system of the fish. Furthermore, two orthologous of FGD were identified in RT and AS. These proteins activate CDC42, a GTPase involved in the organization of the actin cytoskeleton and with a role in early contractile events in phagocytes (Ching et al., 2007). As it has been described that the infective process of P. salmonis depends on the exploitation of the actin monomers (Ramírez et al., 2015), the identification in this study of candidate genes that encode for cytoskeletal motor proteins (two kinesins and a myosin) highlights their relevance not only for the reorganization of the cytoskeleton but also for its motility and involvement in the development of the infection (Hoyt et al., 1997). Remarkably, two other candidate proteins associated with SRS resistance are also members of the GTPase superfamily, GIMAP4 and GIMAP8. This is a family of proteins abundantly expressed in lymphocytes and whose function is to contribute in the regulation of apoptosis and the maintenance of T-cell numbers in the organism (Yano et al., 2014).
中文:
本研究鉴定的另一组高价值跨物种直系同源候选基因编码了 RAB1 和 RAB18 蛋白,二者均属于广阔的 GTP 酶超家族成员。GTP 酶是一大类能够特异性结合并水解 GTP 的水解酶分子群,在细胞跨膜信号传导、蛋白质核糖体翻译、细胞命运控制与分化、胞内囊泡物质定向运输以及微丝细胞骨架空间重构等众多核心生命过程中扮演着中枢枢纽角色(Bourne et al., 1991)。特别是在亚家族层面,RAB GTP 酶小分子蛋白被公认为统领真核生物细胞膜泡靶向运输的"主控调节器"(Stenmark, 2009)。鉴于鲑立克次体在入侵鱼类巨噬细胞后,高度依赖于驱动宿主内膜系统重塑以构建由宿主膜衍生而来的特化细胞器------即"含鲑立克次体空泡"(P. salmonis-containing vacuoles, PCVs),该病原性空泡的成熟与扩增完全取决于细菌强行劫持宿主鱼类胞内膜泡运输系统的分子能力。此外,我们在虹鳟和大西洋鲑中共同锁定了两个 FGD 家族直系同源基因;FGD 蛋白能够催化激活 CDC42------一种专门调控肌动蛋白细胞骨架空间聚合并在吞噬细胞早期吞噬收缩事件中起关键作用的 Rho 家族 GTP 酶(Ching et al., 2007)。正如已有细胞生物学证据所明确揭示,鲑立克次体的侵染病理进程高度依赖于强行盗用宿主肌动蛋白单体(actin monomers)进行微丝组装(Ramírez et al., 2015);本研究在此类位点精准挖掘出编码细胞骨架马达蛋白的候选基因(包括两个驱动蛋白 KIF15/KIF2C 和一个肌球蛋白 MYO3B),进一步从遗传层面深刻凸显了细胞骨架重组以及微丝/微管介导的胞内动力学运输在支持病原菌胞内移动、复制扩增及病理发展过程中的绝对核心地位(Hoyt et al., 1997)。同样令人拍案叫绝的是,与 SRS 抗性紧密连锁的另外两个核心候选蛋白------GIMAP4 和 GIMAP8------亦隶属于 GTP 酶超家族。GIMAP 家族蛋白在淋巴细胞中呈现出特异性高丰度表达,其核心生理功能在于协同维持机体内成熟 T 细胞群体稳态并严密调控活化诱导的细胞凋亡进程(Yano et al., 2014)。
Source: p.9 S053
Original:
Another group of orthologous genes code for ATP-dependent RNA helicases DDX24 in CS and DDX47 in RT for DD. The ATP-dependent RNA helicase DDX family, also known as DEAD-box helicases, is required for different cellular processes such as transcription, pre-mRNA processing, ribosome biogenesis, nuclear mRNA export, translation initiation, RNA turnover, and organelle function. The protein structure is very similar to viral RNA helicases and to DNA helicases, which suggests that the fundamental activities of these enzymes are similar (Rocak and Linder, 2004). Viruses also use RNA helicases at various stages of their life cycle. Many viruses carry their own helicases to assist with the synthesis of their genome, but others synthesize their genome within the cell nucleus, which tends to exploit cellular helicases and thus do not encode their own. We also identified the ATP-dependent RNA helicase TDRD9, which has not been directly implicated in infection but was differentially expressed in channel catfish in response to Aeromonas hydrophila infection (Li et al., 2013). Mechanistic studies of RNA helicases will allow the determination of the precise role of these helicases in the host-pathogen interaction.
中文:
另一组针对死亡天数(DD)被跨物种一致捕获的直系同源基因,分别编码了银鲑中的 DDX24 以及虹鳟中的 DDX47 两种 ATP 依赖性 RNA 解旋酶。ATP 依赖性 RNA 解旋酶 DDX 家族(亦被称为 DEAD-box 家族解旋酶)在真核生物基因表达的几乎所有生命环节中均是不可或缺的:包括染色质转录、前体 pre-mRNA 剪接加工、核糖体生物发生组装、成熟 mRNA 穿核孔转运出核、核糖体翻译起始、RNA 代谢降解更新以及细胞器稳态维持等。其空间三维蛋白质拓扑结构与病毒编码的 RNA 解旋酶以及真核生物自身的 DNA 解旋酶具有极高同源相似性,表明这些酶在能量偶联核酸解旋等基础生物物理活性上遵循高度统一的保守催化机制(Rocak and Linder, 2004)。许多病毒在侵染生命周期的各个阶段都需要借力 RNA 解旋酶:部分病毒自身基因组便编码专职解旋酶以辅助其病毒核酸的复制合成,但更多病毒在宿主细胞核或胞质内合成基因组时,往往直接盗用并强行霸占宿主细胞内原有的内源性 RNA 解旋酶。与此同时,我们还鉴定了同属于该酶系的 ATP 依赖性 RNA 解旋酶 TDRD9;尽管该分子此前鲜有在细菌侵染中被直接报道,但在斑点叉尾鮰(channel catfish)抵御嗜水气单胞菌(Aeromonas hydrophila)感染的免疫转录组学研究中,TDRD9 呈现出极其强烈的差异化表达应答(Li et al., 2013)。未来针对此类 RNA 解旋酶分子机制的深入功能剖析,必将精细厘清它们在宿主-病原互作界面上究竟承担着怎样的抗侵染或被病原菌所盗用的因果角色。
Source: p.9 S054
Original:
The last group of orthologous genes identified code for two L-FABPs in CS and RT for BS. L-FABPs are abundant in hepatocytes and are known to be associated with lipid metabolism. In addition, these proteins are up-regulated in several types of cancer, but their role in infection remains unclear (Ku et al., 2016). Nevertheless, it has been recently reported that serum and urine L-FABP may be a new diagnostic marker for liver damage in patients with both acute and chronic hepatitis C infection (Cakir et al., 2017). Interestingly, in AS challenged with P. salmonis , L-FABP was up-regulated in resistant families and simultaneously down-regulated in susceptible families (Pulgar et al., 2015), suggesting a transcriptional regulation in response to P. salmonis infection and a putative expression marker of resistance to SRS.
中文:
最后一组针对二元存活(BS)在银鲑和虹鳟中共同鉴定的直系同源基因,分别编码了两个肝型脂肪酸结合蛋白(liver fatty acid-binding proteins, L-FABPs)。L-FABP 在鱼类及哺乳动物肝实质细胞中含量极高,是公认主导游离脂肪酸转运与脂质稳态代谢的核心载体分子。此外,该类蛋白在多种人类恶性肿瘤中已被证实显著高表达,但在感染性疾病免疫应答中的明确角色以往尚不明确(Ku et al., 2016)。然而,临床流行病学最新报道揭示,在急性与慢性丙型肝炎病毒感染患者中,血清及尿液中的 L-FABP 水平可作为早期精准诊断严重肝实质损伤与组织炎症的新型临床标志物(Cakir et al., 2017)。极其耐人寻味的是,在我们前期对大西洋鲑开展的人工攻毒转录组追踪研究中(Pulgar et al., 2015),L-FABP 在抵御鲑立克次体成功存活的高抗病家系中被特异性强效上调诱导,但在极度易感家系中却出现剧烈下调抑制!这一显著的转录调控分歧表明,L-FABP 的转录活化稳态不仅是宿主肝脏应对鲑立克次体侵袭的关键代谢应激表征,更有望作为预测鲑鱼抗 SRS 能力的极具潜力的特异性表达生物标志物(biomarker)。
Source: p.9-10 S055
Original:
Genes coding NLRP12, CACNB2, HS3ST3A1, and HINT1 were also selected as candidate genes for SRS resistance. NLRP12 and NLRC3 are two cytosolic proteins that share two functional domains (NACHT and LRR). NLRP12 was one of the best ranked genes, adjacent to the leader SNP and adjacent to more than one SNP simultaneously for DD in AS. This protein functions as an attenuating factor of inflammation in monocytes by negative regulation of the nuclear factor-κB (NF-κB) activation (Fata et al., 2013). In murine macrophages, a significant expression increase has been shown in cells infected with the intracellular parasite Leishmania major compared to non-infected macrophages (Fata et al., 2013). NLRC3 is also a negative regulator of the innate immune response mediated by the inhibition of Toll-like receptor-dependent activation of the transcription factor NF-κB (Schneider et al., 2012). The presence of these genes suggests that the control of the inflammatory reaction in response to P. salmonis infection could be essential to combat SRS.
中文:
编码 NLRP12、CACNB2、HS3ST3A1 和 HINT1 的基因同样被严格选定为抗 SRS 综合征的核心候选基因。NLRP12 与 NLRC3 是两种共享 NACHT 和 LRR 两大特征功能结构域的细胞质模式识别受体蛋白。其中,NLRP12 是全篇综合评分排名前列的核心明星基因之一------在大西洋鲑死亡天数 DD 分析中,它不仅紧邻解释方差最高的先导主效 SNP,还同时被多个显著 SNP 的重叠物理窗口交叉捕获。细胞免疫学机理证实,NLRP12 蛋白在单核/巨噬细胞系中行使着极为关键的"炎症阻尼器"功能,它能够特异性阻断核转录因子- κ \kappa κB(NF- κ \kappa κB)的过度激活,从而强力负向衰减致死性失控炎症反应(Fata et al., 2013)。在小鼠巨噬细胞模型中,当受到胞内专性寄生原虫巨型利什曼原虫(Leishmania major )侵染时,NLRP12 较未感染细胞出现极具统计学意义的剧烈表达上调(Fata et al., 2013)。无独有偶,同家族的 NLRC3 蛋白同样被公认为宿主先天免疫应答的关键负调节因子,它通过直接干预并抑制依赖 Toll 样受体(TLR)信号通路的转录因子 NF- κ \kappa κB 激活级联来实现对免疫风暴的精准下调抑制(Schneider et al., 2012)。上述炎症抑制基因的集中涌现深刻启示我们:在应对鲑立克次体高度破坏性的侵袭过程中,机体能否有效驾驭并严密调控过度白细胞浸润与过度炎症反应,是决定宿主鱼类最终避免组织严重坏死溃疡并逃脱死亡宿命的决定性生命屏障。
Source: p.10 S056
Original:
To the best of our knowledge, this is the first time that functional candidate genes underpinning resistance to P. salmonis are proposed based on a comparative genomics approach comparing GWAS results for the same trait in different fish genus/species. We hypothesize that variations in the sequences of these genes could play important roles in the host response to P. salmonis infection, which could be tested through new genetic approaches such as gene editing using CRISPR-Cas9 and used through GS or more traditional selection practices. All this information together can be used to generate better control and treatment measures for one of the most important bacterial diseases affecting salmon aquaculture.
中文:
据我们所知,这是国际上首次基于跨不同鱼类属/种对相同抗病表型的 GWAS 关联结果进行横向比较基因组学挖掘,从而系统提出支撑鲑立克次体抗性的关键功能候选基因集。我们提出明确的科学假说:这些核心候选基因编码区或调控区内的天然等位基因序列变异,在决定宿主对鲑立克次体侵染的防御应答能力上发挥着决定性作用。这一科学假说在未来可通过尖端逆向遗传学工具(如利用 CRISPR-Cas9 系统对靶基因实施定点编辑改造)在鱼类模型上进行直接的功能敲除与机理求证,并可作为先验生物学功能权重矩阵直接融入全基因组选择(GS)算法模型,或直接赋能更为精准的现代分子标记辅助商业育种。全篇所汇聚的这套高质量功能基因图谱与遗传参数基线,必将协同推动针对这一严重威胁全球三文鱼产业的头号细菌性病害研发出更具针对性、更高效的遗传改良选育方案与新型精准药物免疫综合防治措施。
结论 / Conclusions
Source: p.10 S057
Original:
Although P. salmonis resistance has previously been described as a polygenic trait, our comparative genomics approach based on GWAS results for the same trait in different salmonid species allowed us to identify about 100 candidate genes that may explain resistance to P. salmonis. Of these, 21 are suggested to be strong functional candidates influencing the trait. These genes are associated with multiple biological processes, including dependence on kinase activity, GTP hydrolysis, helicase activity, lipid metabolism, cytoskeletal dynamics, inflammation, and the innate immune response.
中文:
尽管在传统的单物种遗传学评估中,鲑立克次体抗性常被普遍归类为由微效多基因决定的复杂性状,但本研究通过构建跨越三大不同鲑科鱼类针对同一抗病表型的全基因组 GWAS 联合比较基因组学挖掘体系,成功突破了单一物种的视野局限,锁定了约 100 个在物种间具有高度保守生物学特征、且能够科学解释鲑立克次体抗性遗传变异的候选基因。在这些基因中,有 21 个核心基因因同时符合多项严苛的跨物种直系同源及主效应窗口重叠标准,被确立为影响抗病表型的极高置信度功能候选基因。这些关键基因在分子层面深度参与多种核心生命代谢与免疫信号通路,包括依赖激酶活性的信号转导、GTP 水解催化级联、RNA 解旋催化、脂质转运代谢、微丝/微管细胞骨架动态重排、过度炎症消退负调控以及宿主先天免疫防御等。
Source: p.10 S058
Original:
We hypothesize that variations in the sequences of these genes could play an important role in the expression and/or activity of their encoded proteins and consequently in the resistance to P. salmonis. This information could be used to generate better control and treatment measures, based on selective breeding or new drug development, for one of the most important bacterial diseases affecting salmon aquaculture.
中文:
我们由此构建核心学术假说:这些基因内部存在的天然因果核苷酸序列多态性变异,直接主导了其编码蛋白质分子的转录表达丰度及/或空间催化活性的个体间分化,进而最终在宏观表型上决定了个体对鲑立克次体侵染的高度抗性或严重易感表型。本研究所揭示的这一整套全新候选靶标基因网络及精细遗传参数,为今后依托高精度分子选择育种(GS / MAS)选育高抗病品系,以及面向分子靶标开发新型抗感染特异性防治药物创造了坚实的理论基石,对有效遏制和根除这一困扰全球三文鱼养殖业数十年的最重要细菌性疫病具有深远的产业价值与科学意义。
数据可用性、伦理声明与补充信息 / Back Matter
数据可用性声明 / Data Availability
Source: p.10 S059
Original:
Genotype and phenotype data generated for this study are available as supplementary material. CS, RT, and AS data are found on Supplementary files 2 to 4, respectively.
中文:
本研究所产生的所有全基因组 SNP 基因型与抗病表型原始数据均已作为开放获取补充材料随刊公开发布。其中,银鲑(CS)、虹鳟(RT)以及大西洋鲑(AS)的完整分型与试验表型数据集分别收录并可在补充文件 2 至 4(Supplementary files 2 to 4)中直接下载与验证。
致谢 / Acknowledgments
Source: p.10 S063
Original:
Aguas Claras, Pesquera Antares, and Salmones Chaicas provided the CS, RT, and AS datasets, respectively. GY and RC acknowledge the FAPESP (2014/20626-4; 2015/25232-7) and CNPq (308636/2014-7) for the financial support. AB and KCo acknowledge the National Commission of Scientific and Technologic Research (CONICYT) for the funding through the national Ph.D. funding program. RP acknowledges the CONICYT for the funding through the Fondecyt program (11161083). AB acknowledges the Government of Canada for the funding through the Canada--Chile Leadership Exchange Scholarship. We acknowledges the Centro de Investigación en Alimentos para el Bienestar en el Ciclo Vital (ABCvital) for funding this publication. We also want to thank the World Congress on Genetics Applied to Livestock Production as this work has been partially presented on this conference (Yoshida et al., 2018b).
中文:
衷心感谢 Aguas Claras、Pesquera Antares 以及 Salmones Chaicas 养殖企业分别为本研究慷慨提供银鲑(CS)、虹鳟(RT)和大西洋鲑(AS)的珍贵试验群体与基础数据。GY 与 RC 衷心感谢巴西圣保罗州研究基金会 FAPESP(项目号:2014/20626-4;2015/25232-7)以及巴西国家科学技术发展委员会 CNPq(项目号:308636/2014-7)提供的科研资助支持。AB 与 KCo 诚挚感谢智利国家科学与技术研究委员会(CONICYT)国家博士研究生专项基金的资助培养。RP 感谢 CONICYT 通过 Fondecyt 青年科学家项目(项目号:11161083)所给予的研究资助。AB 感谢加拿大政府通过"加拿大-智利领导力交流学者计划"所提供的高级访学奖学金。作者衷心感谢生命周期福祉食品研究中心(ABCvital)资助本篇论文的出版版面费。作者还要向世界家畜遗传应用大会(WCGALP)组委会致以谢忱,本论文的部分前期关联分析结果曾在此次国际学术盛会上做过阶段性展示(Yoshida et al., 2018b)。
精读提示与科研实战研判 / Critical Reading Notes
针对课题组当前核心科研方向------《鱼类跨物种个体表型预测与全基因组选择》(特别结合《20261105预测思路.md》中制定的三大创新方向),对 Yáñez 等人(2019)这篇奠基性论著进行的深层技术拆解、机理批判与实战转化研判。
一、 遗传架构的"物种特异性分歧"对跨物种表型预测模型的深层挑战
-
寡基因架构 (CS, RT) vs 典型多基因架构 (AS) 的本质碰撞:
- 本文最重磅的遗传学发现之一,是揭示了同为鲑科的三种鱼在面对同一种致病菌(P. salmonis LF-89)时,其遗传抗性架构呈现出截然不同的全基因组拓扑结构:银鲑(CS)和虹鳟(RT)表现为典型的寡基因(oligogenic)控制 ------前 200 个 SNP 解释了 70%~90% 的遗传方差,银鲑第 29 号染色体单个主效位点甚至独立解释了超过 50% 的二元存活(BS)变异;相反,大西洋鲑(AS)则表现为极度分散的微效多基因(polygenic)控制,没有单个位点能解释超过 5% 的遗传方差。
- 对跨物种预测的重大启示: 传统动植物育种中的 GBLUP 或 Bayes 模型高度依赖目标物种内部的固定连锁不平衡(LD)架构。如果直接用大西洋鲑(AS)的微效分散模式训练传统线性加性模型,去预测由少数大效应位点主导的银鲑(CS)或虹鳟(RT),预测准确率必然遭遇断崖式下跌;反之,若用包含超大效应位点的银鲑训练模型,模型很可能将极大的回归权重分配给特定的几个非保守位点,从而在留出的测试鱼种(如 AS)上完全失效。这为本实验室开发基于深度学习特征嵌入(如 DNABERT-2 或卷积序列编码器)的非线性泛化预测模型提供了极其坚实的立论依据。
-
表型指标定义的分水岭:死亡天数 (DD) vs 二元存活 (BS):
- 论文表 1 显示,三种鱼在二元存活(BS)下的基因组遗传力(CS: 0.88, RT: 0.64, AS: 0.32)显著高于死亡天数(DD)下的遗传力(CS: 0.32, RT: 0.48, AS: 0.24)。
- 实战建模建议: 死亡天数(DD)本质上是"耐受性/感染潜伏期延长",而二元存活(BS)是"终极存活生/死结局"。在跨物种表型预测任务中,由于不同鱼种攻毒试验的试验周期长短、水温调控及终点判定存在微小差异,优先以二元存活(BS,存活为 0,死亡为 1,或转化为存活概率)作为跨物种模型的主预测目标,其信噪比更高、遗传可塑性更强、AUROC 评价标准更加刚性。
二、 为什么染色体共线性 (Synteny) 未发生位点重合,但功能结构域却高度收敛?
-
第四轮全基因组加倍 (SS4R) 遗留的演化痕迹:
- 鲑科鱼类共同祖先在距今约 8000 万至 1 亿年前经历了第四轮全基因组加倍事件(salmonid-specific whole-genome duplication, SS4R),目前仍处于尚未完全结束的"再二倍体化"(rediploidization)进程中(Lien et al., 2016)。
- 这一独特的基因组演化历史导致了大量同源染色体的大规模重组、易位、倒位以及旁系同源基因的非对称丢失。因此,我们在 Circos 图(图 2)中看到了看似矛盾的现象:三种鱼虽然来自同一祖先,但在染色体物理共线性水平上,并没有任何一个关联 SNP 窗口能够被三种鱼共同锁定。
-
从"位点共线"到"分子通路功能共线":
- 尽管物理坐标无法跨物种对齐,但比较基因组学分析(图 4、表 2)表明,533 个候选基因高度收敛于 P-loop_NTPase(占 30%)、非受体型酪氨酸激酶(nr-TPKs)、小 GTP 酶(RAB1/RAB18)、细胞骨架马达蛋白(KIF/MYO3B)以及炎症抑制因子(NLRP12/NLRC3)。
- 对课题组《方向三(加入 SNP 相对于同源基因的位置)》的理论支撑: 这一现象深刻证明:跨物种表型预测绝对不能依赖绝对染色体坐标的简单平移,而必须建立在直系同源基因(Orthologs)的功能对齐之上。通过将 SNP 锚定到保守同源基因的相对位置(如距离转录起始位点 TSS 的上/下游距离、相对剪接位点的外显子/内含子坐标),模型才能在不同物种间捕获到一致的生物学调控上下文。
三、 21 个核心高优先级候选基因的分子互作网络与致病机理全景
根据本研究表 2 锁定的核心候选基因,结合现代细胞分子生物学与宿主-病原互作前沿进展,我们梳理出鲑立克次体在鲑鱼体内的致病与宿主抗性因果作用链网络:
【鲑立克次体 Piscirickettsia salmonis 侵染】
│
┌────────────────────────────────┴────────────────────────────────┐
▼ ▼
【膜表面模式识别与吞噬】 【细菌分泌系统效应蛋白注入】
- MRC1 (甘露糖受体介导内吞) - 毒力因子 EspJ (ADP-核糖基转移酶)
- COL12 (激活补体替代途径) │
- TAPBPR (MHC-I 专职抗原装载) ▼
│ 【劫持宿主胞内激酶网络】
▼ - 抑制 nr-TPKs (Src 家族: YES1, LYN, SYK)
【宿主细胞内膜泡系统被盗用】 - ARL5B (小 GTP 酶逆向调控核糖基化)
- RAB1 / RAB18 (调控囊泡运输) │
- 构建特化"含菌空泡" (PCV) │
│ │
▼ ▼
【微丝与微管细胞骨架动力学剥削】 【过度炎症反应与免疫风暴调控】
- FGD (催化激活 CDC42) - NLRC3 (抑制 TLR-TRAF6-NF-κB 级联)
- 单体肌动蛋白 (Actin) 聚合 - NLRP12 (强力负调控 NF-κB 激活,阻尼炎症)
- 马达蛋白驱动: KIF15, KIF2C, MYO3B - 决定宿主是"组织耐受修复"还是"败血溃疡致死"
│ │
▼ ▼
【核酸代谢与能量运转】 【肝脏脂质代谢与全身应激】
- DEAD-box RNA 解旋酶 (DDX24, DDX47) - L-FABP (肝型脂肪酸结合蛋白)
- TDRD9 (能量偶联 RNA 催化代谢) - 抗病家系特异性上调 / 易感家系剧烈下调
-
nr-TPKs 与 ARL5B 的攻防平衡(评分 8 分与 5 分):
- 细菌通过分泌系统向宿主巨噬细胞注入效应蛋白(如类似大肠杆菌/枸橼酸杆菌的 EspJ),通过共价催化 ADP-核糖基化修饰,直接"废掉"宿主胞质内的 Src 家族酪氨酸激酶(YES1、LYN、SYK),导致 T/B 细胞受体信号中断,阻断淋巴细胞活化。
- 宿主抗性个体中检出的先导主效基因 ARL5B 和 nr-TPK 序列多态性,极可能是赋予了激酶对细菌毒力修饰的抗性突变,或者通过强化内源性去核糖基化维持了激酶级联的正常传导。
-
RAB GTP 酶与细胞骨架马达蛋白对"含菌空泡(PCV)"的制约:
- 鲑立克次体是胞内寄生菌,它不会在胞浆内自由裸露游动,而是迫使巨噬细胞内膜系统为其构建一层专属保护膜------PCV 空泡。RAB1 和 RAB18 是高尔基体与内质网囊泡运输的中枢;而病原体在细胞内的穿梭与胞吐播散,则完全借助宿主肌动蛋白(Actin)聚合及微管驱动蛋白(KIF15、KIF2C)与肌球蛋白(MYO3B)。抗性鱼类在这些马达蛋白和 GTP 酶位点上的变异,可能破坏了细菌构建空泡或利用骨架轨道移动的能力。
-
NLRP12 与 NLRC3:为什么"抑制炎症"反而能提高存活率?
- SRS 的典型临床解剖症状是患鱼严重贫血、脾脏和肾脏极度充血肿大、全身肌肉广泛点状出血溃疡。大量鱼类致死并非细菌本身的直接毒害,而是细菌触发了失控的全身性"细胞因子风暴"(cytokine storm)和重度过度炎症。
- NLRP12 和 NLRC3 作为胞质内下调 NF- κ \kappa κB 激活的关键"阻尼刹车",能够有效平抑过度巨噬细胞活化,使宿主在清除病原的同时维持机体器官稳态,这是抗病鱼存活的关键保障。
-
L-FABP(肝型脂肪酸结合蛋白):抗病表达生物标志物:
- 攻毒后,抗病家系 L-FABP 表达显著上调,易感家系剧烈下调。该基因编码的蛋白控制肝脏长链脂肪酸代谢与能量稳态,直接关系到鱼体在抵御败血症急性消耗阶段的能量代谢供给与肝细胞自我修复。
四、 面向课题组三大研究方向的"实操转化落地路线图"
结合本实验室在《20261105预测思路.md》中制定的科研计划,本文及其配套数据集(Supplementary files 2--4)是开展跨物种个体表型预测的最理想"天然试验场":
-
关于数据源的精准调用(Supplementary files 2--4):
- 本文已在公开补充材料中完整提供了三大鱼种的个体编号、全同胞家系 ID、起始体重、死亡天数(DD)、二元存活状态(BS)以及完整的合格 SNP 基因型矩阵。
- 关键提示: 银鲑(CS)是 ddRAD 测序(9,389 个 SNP,参考基因组 GCF_002021735.1);虹鳟(RT)是 57K 芯片(24,916 个 SNP,参考基因组 GCF_002163495.1);大西洋鲑(AS)是 50K 芯片(42,624 个 SNP,参考基因组 GCF_000233375.1)。
- 由于三个物种分型平台和标记密度不同,传统基于同一芯片坐标的对齐方法彻底失效,但完全契合课题组所提出的**"按 SNP 截取两侧 DNA 序列片段"**的核心范式!
-
方向一(截取 SNP 中心片段,设置 A/B/C 三组对照)的落地验证:
- 片段截取方案: 针对每个物种通过 QC 的 SNP,从各自参考基因组中提取以 SNP 为中心、两侧各拓展 L L L 个碱基(如 L = 250 L=250 L=250 bp 或 L = 500 L=500 L=500 bp,形成 501 bp 或 1001 bp 片段)的双等位 DNA 片段。
- A 组(遮住两侧): 仅保留中心等位基因(
TC[A]GT→ \to →NN[A]NN)。 - B 组(真实两侧): 包含真实参考基因组上下游上下文。
- C 组(交换两侧): 同一鱼种内部打乱互换上下游序列。
- 科学预测与检验假说:
- 如果利用例如大西洋鲑(AS)训练模型,在留出的虹鳟(RT)或银鲑(CS)个体上测试:若 B 组的 AUROC 显著且稳定高于 A 组与 C 组,则决定性证明:即使 SNP 物理坐标无法跨物种对齐,SNP 侧翼包含的转录因子结合基序(TF binding motifs)、启动子/增强子顺式调控元件序列,能够在跨物种表型预测中提供关键的通用表征信息!
-
方向二(显式引入个体基因型剂量 0/1/2)的协同增益:
- 银鲑和虹鳟存在大效应主效位点(如银鲑 Chr 29 位点)。纯粹基于无权重的 DNA 片段嵌入平均可能稀释大效应纯合子(
1/1)与杂合子(0/1)的剂量分化。 - 方向二通过在序列嵌入向量后拼接个体等位基因拷贝数(0, 1, 2)或进行双线性交互,能够显著增强模型对显性效应与加性主效应的感知灵敏度。
- 银鲑和虹鳟存在大效应主效位点(如银鲑 Chr 29 位点)。纯粹基于无权重的 DNA 片段嵌入平均可能稀释大效应纯合子(
-
方向三(融合直系同源基因与共线区段相对坐标)的终极攻关:
- 重点提取本文 Table 2 中鉴定的 21 个核心高优先级功能基因(尤其是 nr-TPK、DDX、RAB、NLRP12、ARL5B、L-FABP 等)!
- 以这些保守直系同源基因的转录起始位点(TSS)为参考锚点,计算各鱼种内部 SNP 相对该基因的相对物理距离与链方向。
- 将这种跨物种含义高度一致的结构化先验位置特征输入模型,预期将在留出物种测试中进一步突破跨物种预测精度的天花板!