【作者】 万裴琦；

【导师】 吴继周；

【作者基本信息】 广西医科大学 ， 内科传染病学， 2014， 博士

【摘要】 背景：肝癌是全世界最常见的恶性肿瘤之一，近几十年来发病率及死亡率均居高不降。全球每年原发性肝癌新发例数超过74万，居恶性肿瘤的第五位，死亡59.8万例，死亡人数仅次于胃、食道癌而居第三位，已经严重威胁到人类的健康及寿命[1]。全球约有一半的肝癌发生在中国，其分布具有明显的地区差异性。广西是全国肝癌高发区之一，多次的流行病学调查结果显示广西肝癌发病率及死亡率均明显高于全国平均水平，并且广西肝癌的发生呈现家族聚集倾向现象，很大一部分的肝癌病例有肝癌家族史，其一级亲属、二级亲属罹患肝癌的风险性增高。大多数肝癌起病隐匿、进展快、病程短，迄今为止，没有特效的治疗手段来阻止肿瘤的进展，故有必要从内在遗传因素及外在环境暴露因素着手对原发性肝癌及其家族聚集性的发病机理进行深入的研究。国内外专家经过多年来不懈的努力，对于原发性肝癌的病因研究取得了一定成果。通过对原发性肝癌的高发地区进行调查，专家普遍认为肝癌是由于以下多种因素共同引起的：肝炎病毒慢性感染、黄曲霉毒素及化学毒物（亚硝胺、重金属等）摄入、饮用水污染、放射线接触、酒精、嗜肝寄生虫、遗传因素等，但其中生物、环境、遗传因素间错综复杂的交互作用的具体机制尚未完全清楚。导致肝癌家族聚集的原因更复杂，为数不少的研究提示乙肝病毒感染是乙肝相关性肝癌的致病因子，我们既往的研究HBV感染是我区家族聚集性肝癌高发的主要致病因子，另环境因素包括饮用水污染、黄曲霉毒素及丙肝病毒感染外，尚有遗传因素。不管生物、环境、遗传因素的具体致癌机制如何，机体都会出现免疫功能的改变，多种免疫细胞的功能和（或）细胞因子表达或表达水平异常[2-5]。体内炎症抑制因子和炎症促进因子网络平衡被打破，机体体液免疫和细胞免疫功能异常，免疫应答功能异常形成恶性循环。譬如，原发性肝细胞癌患者体内IL-1、2、12、18、TNF-a、IFN-r等Th2类促炎因子与IL-4、5、8、10等Th1类炎症抑制因子网络平衡失调，CD4+、CD8+、NK细胞、细胞毒性T细胞、巨噬细胞等专职免疫细胞功能下降或异常，致使肿瘤逃避免疫监视及免疫清除，肿瘤得以发生、进展及转移，同时也说明肝癌是很典型的炎症相关性肿瘤[6-14]。已知多种细胞因子如IL-1、IL-10、IL-12、IL-18、TGF-β1、TNF-α、NQo1、XPD、COX2等细胞因子基因及其多态性与肝癌有关。其中转化生长因子β1（TGF-β1）及其多态性在原发性肝癌发生中起到了十分重要的作用。TGF-β1是一类调节细胞增殖、分化、细胞功能作用的激素样活性多肽，对淋巴细胞、上皮细胞增殖、分化具有抑制作用，对间充质细胞增殖分裂具有促进作用，同时调节细胞外基质蛋白质的合成，在组织修复、细胞恶性突变中具有重要的作用。正常水平的TGF-β1通过抑制T淋巴细胞、B淋巴细胞活化，抑制NK细胞、细胞毒性T细胞杀伤功能而发挥适度的免疫抑制功能[13,14]。病理情况下，TGF-β1水平明显升高，破坏机体免疫监督功能，使机体失去对外来病原或肿瘤细胞的正常免疫应答，导致疾病或肿瘤的发生、发展[15]。近年来，以上各种细胞因子与肝癌家族聚集的相关性也陆陆续续有报道，TGF-β1及其基因多态性在肝癌家族聚集的发生中起到的作用迄今为止尚未见有相关报道，TGF-β1蛋白质表达水平与肝癌家族聚集性的关系也尚不十分清楚。为更好地了解广西肝癌家族聚集发生的相关因素，我们从转化生长因子β1基因多态性及其蛋白表达的角度探讨肝癌家族聚集发生的原因。本研究分为两大部分。第一部分对广西11个肝癌高发区的35个肝癌高发家族的214名成员（其中壮族105人、瑶族49人、汉族60人）、37个无癌对照家族中214名成员（其中壮族105人、瑶族49人、汉族60人）进行转化生长因子6个位点基因多态性的分析，获得研究组与对照组6个位点基因型分布频率的遗传学数据，并重点探讨在吸烟、饮酒、饮用水及HBV-DNA阳性等外暴露因素共同作用下，TGF-β1基因多态性与原发性肝癌家族聚集性的关系。第二部分研究分两小部分：第一小部分是用免疫酶联吸附法检测全部428例样本的血清TGF-β1水平，比较肝癌高发家族与无癌对照家族血清TGF-β1水平差异、比较肝癌风险位点各基因型对应的血清TGF-β1水平差异，按血缘关系分层分析血清TGF-β1水平差异。第二小部分是用免疫蛋白印迹法检测82例（包括11名先证者及其配对、10名随机抽取的肝癌高发家族Ⅰ级亲属及其配对、10名随机抽取的肝癌高发家族Ⅱ级亲属及其配对、10名随机抽取的肝癌高发家族Ⅲ级亲属及其配对）研究对象全血TGF-β1蛋白表达情况，比较肝癌高发家族与无癌对照家族全血TGF-β1蛋白灰度值差异，按血缘关系分层分析全血TGF-β1蛋白灰度值差异。从细胞因子TGF-β1表达水平角度探讨其与原发性肝癌家族聚集性的关系。第一部分TGF-β1基因多态性等因素与中国广西人群肝癌家族聚集的相关性研究目的：探讨TGF-β1（转化生长因子β1）基因多态性及外暴露因素与广西壮、汉、瑶族肝癌家族聚集发生的关系。方法：在广西11个肝癌高发区收集到符合条件的肝癌高发家族35个，共214名成员，无癌对照家族37个，以相同年龄（±5岁）、乙肝表面抗原（HBsAg）、民族、居住地、性别为配对条件，选取无癌家族成员214名作为无癌对照家族组。研究对象的基本信息及肝癌相关风险因素采用流行病学调查表采集。采集所有研究对象的空腹外周静脉血抗凝标本5毫升，用于提取全血基因组DNA，采集非抗凝血5毫升（乙二胺四乙酸抗凝）分离出血清用于检测HBV、HCV病毒标志物、TGF-β1水平。以上所提取的DNA样品，经引物特异性PCR方法扩增TGF-β1rs1800469、rs2241715、rs2241716、rs11466345、rs8105161、rs747857共6个位点，采用Snapshop方法进行基因分型检测。应用Haploview软件进行哈-温（Hardy-Weinberg）遗传平衡检测，SHEsis软件进行连锁不平衡检验及单倍型频率分析统计。经SPSS17.0统计软件进行数据录入及统计分析，采用卡方检验和条件Logistic回归模型分析TGF-β1SNP位点与肝癌家族聚集性的关系，选取有统计学意义的位点及环境因素，采用叉生分析法进行基因-环境交互作用分析，应用SAS9.1编程假设验证。结果：（1）肝癌可能的风险因素分布：饮酒、饮用塘水、HBV-DNA阳性在肝癌高发家族与无癌对照家族中分布差异具有统计学意义（P<0.05）。（2）各基因型分布的哈-温平衡检验：经Haploview软件检验，6个位点基因型频率在无癌对照家族与肝癌高发家族中分布符合遗传平衡定律，观察值与理论值符合（P＞0.05），选取样本具有群体代表性。（3）TGF-β1各位点基因型及等位基因在无癌对照家族与肝癌高发家族组的分布：1. TGF-β1rs1800469T/C位点基因型TT、CT、CC在肝癌高发家族组与无癌对照家族组中分布频率分别为47.2%、41.1%及11.7%，33.7%、45.3%及21．0%，两组间分布具有统计学差异（P=0.004）；等位基因T、C在肝癌高发家族组与无癌对照家族组中分布频率分别为：67.8%和32.2%，56.3%和43.7%，等位基因在两组间分布具有统计学差异（P=0.001）。rs1800469TT基因型及T等位基因在肝癌高发家族分布明显高于无癌对照家族（P=0.004； P=0.001）。2.TGF-β1rs2241715G/T位点基因型TT、GT、GG在肝癌高发家族组与无癌对照家族组中分布频率分别为46.7%、41.6%、11.7%与33.2%、44.9%、21．9%，两组间分布具有统计学差异（P=0.003）；等位基因T、G在肝癌高发家族组与无癌对照家族组中分布频率分别为：67.5%和32.5%，55.6%和44.4%，等位基因在两组间分布具有统计学差异（P=0.000）。rs2241715TT基因型及T等位基因在肝癌高发家族分布明显高于无癌对照家族（P=0.004； P=0.000）。3. TGF-β1rs8105161C/T位点基因型TT、CT、CC在肝癌高发家族组与无癌对照家族组中分布频率分别为29.4%、51.9%、18.7%与42.5%、45.8%、11．7%，两组间分布具有统计学差异（P=0.009）；等位基因T、C在肝癌高发家族组与无癌对照家族组中分布频率分别为：55.4%和44.6%，65.4%和34.6%，等位基因在两组间分布具有统计学差异（P=0.003）。rs8105161TT基因型及T等位基因在无癌对照家族分布明显高于肝癌高发家族（P=0.005；P=0.003）。4.TGF-β1rs747857A/G位点基因型AA、AG、GG在肝癌高发家族组与无癌对照家族组中分布频率分别为82.7%、15.0%及2.3%，90.7%、9.3%及0%，两组间分布具有统计学差异（P=0.005）；等位基因G、A在肝癌高发家族组与无癌对照家族组中分布频率分别为：90.2%和9.8%，95.3%和4.7%，等位基因在两组间分布具有统计学差异（P=0.004）。rs747857GG基因型及G等位基因在无癌对照家族分布明显高于肝癌高发家族（P=0.005；P=0.003）5.TGF-β1rs2241716C/T位点基因型及等位基因在肝癌高发家族组与无癌对照家族组中分布频率无统计学差异（P=0.105；P=0.160）。6.TGF-β1rs11466345T/C位点基因型在肝癌高发家族组与无癌对照家族组中分布频率无统计学差异（P=0.161）；等位基因T、C在两组间分布具有统计学差异（P=0.049）。（4）TGF-β1多态性位点连锁不平衡分析：rs2241715、 rs2241716、rs1800469三个位点间存在强连锁不平衡。 TGF-β1rs1800469与TGF-β1rs2241715间D‘=1，r2=0.984,与TGF-β1rs2241716间D‘=1，r2=0.547。TGF-β1rs2241715与TGF-β1rs2241716也存在强LD，D‘=1，r2=0.538。（5）环境与遗传的多因素条件Logistic回归分析：将rs1800469、rs2241715、rs8105161、rs747857连同肝癌可能的相关危险因素吸烟、饮酒、饮用水、HBV-DNA带入Logistic回归模型分析（因所有研究对象均以大米为主食、抗HCV均阴性，故该两项因素未列入模型），最终进入模型的危险因素由高到低为：饮用塘水＞饮酒＞HBV-DNA阳性＞rs1800469TT基因型＞rs2241715TT基因型。（6）环境与遗传的多因素条件Logistic回归分析： rs1800469、rs2241715、rs8105161、rs747857、rs2241716、rs11466345连同肝癌可能的相关危险因素吸烟、饮酒、饮用水、HBV-DNA带入Logistic回归模型分析（因所有研究对象均以大米为主食、抗HCV均阴性，故该两项因素未列入模型），最终进入模型的危险因素由高到低为：饮用塘水＞饮酒＞HBV-DNA阳性＞rs1800469TT基因型＞rs2241715TT基因型。（7）风险基因型的分层分析：rs1800469TT基因型及T等位基、rs2241715TT基因型及T等位基因在各级血缘亲属中总体分布无统计学差异（P=0.416,P=0.185）；(P=0.348,P=0.133），但二者分布频率随着与先证者血缘级别降低而降低。（8）基因-环境交互作用分析：基于相加模型的TGF-β1rs1800469TT与饮用塘水存在正交互作用（OR=6.44，S=1.50，AP=0.28,P＜0.05）,经相加模型的叉生分析假设检验U=0．394，P＜0．05，具有统计学意义；TGF-β1rs1800469TT与饮酒交互作用也有统计学意义（OR=3.45，S=1.29，AP=0.16,P＜0.05），假设检验U=0．432， P＜0.05具有统计学意义；TGF-β1rs1800469TT与HBV-DNA相加交互作用同样也有统计学意义（OR=5.38，S=4.13，AP=0.62,P＜0.05），假设检验U=1．105，P＜0.05）。结论：（1）TGF-β1rs1800469T/C基因多态性可能是广西肝癌家族聚集发生的遗传因素。（2）TGF-β1rs2241715G/T基因多态性可能是广西肝癌家族聚集发生的遗传因素。（3）饮用塘水、饮酒、HBV-DNA阳性可能是广西肝癌家族聚集发生的主要环境因素。（4）广西肝癌家族聚集现象的发生是综合因素所致，多种遗传因素之间、多种环境因素之间或遗传与环境因素之间有相互协同作用。第二部分TGF-β1基因表达与中国广西人群肝癌家族聚集的相关性目的：探讨转化生长因子β1血清中表达水平、全血中蛋白表达情况与广西人群肝癌家族聚集的相关性。方法：在广西11个肝癌高发区收集到符合条件的肝癌高发家族35个，共214名成员，无癌对照家族37个，以相同年龄（±5岁）、乙肝表面抗原（HBsAg）、民族、居住地、性别为配对条件，选取无癌家族成员214名作为无癌对照家族组。本研究分两小部分：第一小部分是用免疫酶联吸附法检测全部428例样本的血清TGF-β1水平，比较肝癌高发家族组（FHCC）与无癌对照家族组（FNC）血清TGF-β1水平差异、比较肝癌家族聚集性风险位点各基因型间的血清TGF-β1表达水平差异、按血缘关系分层分析血清TGF-β1水平差异。第二小部分是用免疫蛋白印迹法（Western blot）检测82例（包括11名先证者及其配对、10名随机抽取的肝癌高发家族Ⅰ级亲属及其配对、10名随机抽取的肝癌高发家族Ⅱ级亲属及其配对、10名随机抽取的肝癌高发家族Ⅲ级亲属及其配对）研究对象全血TGF-β1蛋白表达情况（包括蛋白表达的量、蛋白结构），比较肝癌高发家族与无癌对照家族全血TGF-β1蛋白灰度值差异，按血缘关系分层分析全血TGF-β1蛋白灰度值差异。探讨其与肝癌家族聚集性是否相关。采用SPSS17.0软件进行数据分析，指标以均数±标准差(x±S)表示。两组间比较采用配对样本t检验，多组间比较采用多独立样本Kruskal-Wallis检验，以P＜0.05为有统计学意义，检验水准α=0.05。结果：1血清TGF-β1水平结果（1）血清TGF-β1水平在肝癌高发家族组TGF-β1水平明显高于无癌对照家族组（28.495±17.495ng/mlVS20.24±7.56ng/ml，P=0.000）。（2）按基因型分析TGF-β1水平：Rs2241715TT基因型对应的血TGF-β1水平明显高于GT型（30.33±15.66ng/mlVS23.55±11.45ng/ml，P=0.000）及GG型（30.33±15.66ng/mlVS20±10ng/ml，P=0.000）。而GT与GG基因型对应的血TGF-β1水平无差异（23.55±11.45ng/mlVS20±10ng/ml，P=0.061）。Rs1800469TT基因型对应的血TGF-β1水平明显高于CT基因型（39.445±7.445ng/mlVS26.33±13.65ng/ml，P=0.000）及CC基因型（39.445±7.445ng/mlVS26.245±13.555ng/ml，P=0.000）。CT及CC基因型对应的血清TGF-β1水平无差异（26.33±13.65ng/mlVS26.245±13.555，P=0.347）。（3）风险位点基因型对应的血清TGF-β1在肝癌高发家族组无癌对照家族组的比较：肝癌高发家族组TT基因型对应的血清TGF-β1水平明显高于无癌对照家族组rs1800469TT基因型（43.395±3.495ng/mlVS35.95±3.95ng/ml，P=0.000）。肝癌高发组rs2241715TT基因型对应的血TGF-β1水平明显高于无癌对照家族组（18.815±4.145ng/mlVS38.995±6.995ng/ml，P=0.000）。2. Western blot检测结果（1）肝癌高发家族组与无癌对照家族组均表达TGF-β1蛋白且未发现异常条带，但表达程度有明显不同。（2）肝癌高发家族组血浆TGF-β1蛋白含量比无癌对照家族组多，差异有统计学意义（p＜0.05）。（3）分层分析：血浆TGF-β1蛋白含量在肝癌高发家族先证者、Ⅰ级、Ⅱ级、Ⅲ级亲属中无差异（p＞0.05）。先证者血浆TGF-β1蛋白灰度值很高，但随着与先证者血缘关系的变远，TGF-β1蛋白灰度值减少。结论：（1）血清高TGF-β1水平可能与肝癌家族聚集性相关。（2）全血高TGF-β1蛋白水平可能与肝癌家族聚集性相关。（3）血清高TGF-β1水平或全血TGF-β1蛋白高表达可能与机体免疫抑制状态有关，尚待进一步验证T细胞、B细胞功能。更多还原

【Abstract】 Background: Hepatocellular carcinoma is one of the most commonmalignant tumors in the world, in recent years the rate of incidence andmortality keep high level. The number of new cases worldwide is600000,ranking the fifth malignant tumor,and deaths to598000cases each year. Thedeath rate of liver cancer ranked third,just following the stomach, esophaguscancer. Hepatocellular carcinoma has been a serious threat to human health andlife. Statistic data shows that half of new hepatocellular carcinoma cases anddeaths reported worldwide occur in China. Its distribution has obvious regionaldifferences. Accorrding to the epidemiological investigation, the incidence andmortality rate of liver cancer in Guangxi were significantly higher than that ofthe national average,and the liver cancer in Guangxi showed tendency offamilial aggregation. Most patients with hepatocellular carcinoma had familyhistory of liver cancer. The risk of suffering from HCC increased greatly withintheir Ⅰ and Ⅱdegree relatives. The majority of hepatocellular carcinoma presented insidious onset, rapid progress, short course, so far,there is no specifictreatments to prevent tumor, so it is necessary to conduct in-depth research onthe pathogenesis of liver cancer and familial aggregation focusing on theinternal genetic factors and environmental factors.After many years of hard work, domestic and foreign experts make someachievement in finding out the pathogenesis of liver cancer. Accorrding tothe epidemiological investigation in some areas of Guangxi with highincidence of hepatocellular carcinoma, experts agree that HCC is due to thefollowing factors: infection with chronic hepatitis B virus, intake of aflatoxinand chemical toxicant(including nitrosamines, heavy metals elements), intakeof water with algal toxins, exposure to radiation, alcohol consumption,infection by hepatotropic parasites, genetic factors and so on. But the specificmechanism of interaction among biological, perplexing environment, geneticfactors remain unclear. The cause of familial clustering of hepatocellularcarcinoma was more complex, many studies suggest that infection with hepatitisB virus was a main pathogenic factor of HBV related HCC. Our previous studyfound that HBV infection is the main reason causing familial clustering ofhepatocellular carcinoma in Guangxi population. Other factors included geneticfactors, HCV infection and environmental factors, such as water pollution oraflatoxin. No matter how the carcinogenic mechanism of biological factor,environment and genetic factors, the body will appear immune function change.A variety of immune cell function and (or) the expression of cytokines wasabnormal. The balance between anti-inflammatory factors and inflammatoryfactors was broken, the humoral immunity and cellular immunity weredysfunction, causing vicious spiral of abnormal immune response function. Forexample, The body with hepatocellular carcinoma lost balance of cytokine between Th1cytokines IL-1, IL-2, IL-12, IL-18, TNF-a, IFN-r and Th2cytokines IL-4, IL-5, IL-8, IL-10,which were belongs to inflammationpromoting factor and inflammation inhibition factor respectively.That will leadCD4+cells,CD8+cells, NK cells, cytotoxic T cells, macrophages and otherfull-time work immune cells to lower ability or abnormal function,resulting inthat tumor escaped from immune surveillance and immune clearance,and thatmake tumor produce, progress and metastasis. It means that liver cance is atypical tumor associated with inflammation at the same time. It was knownthat cytokines such as IL-18, IL-10, IL-1, IL-12, TGF-β1, TNF-α, NQo1, XPDand COX2gene polymorphism was associated with HCC. Among thesecytokines, the transforming growth factor beta1(TGF-β1) gene polymorphismplays an important role in production of primary hepatocellular carcinoma.TGF-β1is a kind of hormone like peptides to regulate cell proliferation,differentiation and function, inhibit lymphocyte, epithelial cell proliferation anddifferentiation, can promote proliferation of mesenchymal cell, regulatesynthesis of extracellular matrix proteins, plays an important role in the processof tissue repairation and cells mutation. Normal levels of TGF-β1suppressimmune function properly through inhibiting activation of T lymphocyte and Blymphocyte, inhibiting of killing function of NK cells and cytotoxic T cell.Under pathological conditions, TGF-β1level increases obviously, whichdestroys the immune surveillance function, makes the body out of the normalimmune response to foreign antigen or tumor cells, which gives chance todevelop diseases or tumor. Relationships between cytokines above and familialclustering of hepatocellular carcinoma continued to be reported, but there wasno report about the effect of genetic polymorphisms of TGF-β1on familialclustering of liver cancer recent years. The relationship between the expression of TGF-β1and familial aggregation of hepatocellular carcinoma is unclear. Inorder to learn about factors associated with familial clustering of hepatocellularcarcinoma occurred in Guangxi better, we explore the causes of familialclustering of hepatocellular carcinoma in the perspective of transforming growthfactor β1gene polymorphism and its protein expression.There are two parts ofthe study. Firstly, to obtain case group and control group genetics data of sixsites of transforming growth factor β1,we analysis this sites gene polymorphismof transforming growth factor β1,which associated with HCC, with214familymembers(including105cases of zhuang、60cases of Han and49cases of Yao)in34hepatocellular carcinoma clustering families (FHCC) and214familymembers (including105cases of zhuang、60cases of Han and49cases of Yao)from the families without any cancer (FNC), those of all that from11highincidence area of liver cancer in Guangxi population, probe the effect oftransforming growth factor β1gene polymorphism on hepatocellular carcinomaclustering families exposure to smoking, drinking, HBV infection together. Thesecond part of the study is divided into two small parts: Firstly, compare thetransforming growth factor β1level in serum between the hepatocellularcarcinoma clustering families (FHCC) group and the families without anycancer (FNC), compare serum TGF-β1levels among different groups of riskloci genotypes, analysis stratification to serum TGF-β1levels according to theblood relationship, with ELISA assay to detect the serum TGF-β1level of428cases. Second, use western blot assay to detect blood protein expression of82cases including11probands and their matches,10cases of first-degree relativesselected randomly in hepatocellular carcinoma clustering families and theirmatches,10cases of second degree relatives selected randomly inhepatocellular carcinoma clustering families and their matches,10cases of third degree relatives selected randomly in hepatocellular carcinoma clusteringfamilies and their matches, compare the difference of blood TGF-β1proteingray value between case group and control group, analysis blood TGF-β1protein gray value according to the blood relationship with stratification,explore relation of TGF-β1and liver cancer familial aggregation in perspectiveof protein expression.Part Ⅰ Study on the relationship between TGF-β1genepolymorphisms in GuangXi China population and familialclustering of hepatocellular carcinomaObjectives: To probe the susceptibility of cytokine TGF-β1genepolymorphism and exposure factors with primary hepatocellular carcinoma inGuangxi Zhuang, Han, Yao population.Methods: confirm214family members from34hepatocellular carcinomaclustering families (FHCC) as the cases and37families without any cancer(FNC),select214members from37families without any cancer (FNC) as thecontrol matched214member of the cases in terms of ages(±5), HBsAg,nationality, habitation and gender. All subjects were from11areas of highincidence rate of hepatocellular carcinoma in Guangxi population. Acquire basicinformation and liver cancer risk factors of the subjects by epidemiologicalquestionaire. Collect5ml non-anticoagulant blood to separate of serum fordetection of markers of HBV, anti-HCV, TGF-β1levels.5ml fasting venousblood of428subjects were collected in anticoagulant tube in the morning, andgenome DNA were extracted. The specific primer PCR amplified6SNPloci of TGF-β1rs1800469, rs2241715, rs2241716, rs11466345, rs8105161, rs747857, which were genotyped through the method of Snapshop. Haploviewsoftware was applied to test Hardy-Weinberg genetic balance balance, andSHEsis software was applied to analysis linkage disequilibrium and haplotypefrequency. All data were input and analyzed statistically in SPSS17.0. Chisquare test and conditional Logistic regression model were used to analysisrelationship between TGF-β1SNP and liver cancer familial aggregation.Crossover analysis for SAS9.1performed to gene-environmental interactionstudies for statistically significant Loci and environmental factors.Result:(1) Distributions of suspected risks of HCC in the subjects: thefrequency of drinking、smoking、Drinking pond water and HBV-DNA(+) weresignificant different in the members of FHCC and those of FNC.(2) Hardy Weinberg equilibrium test: the result show that all sis sitesgenotypes of the observations fit well with the theoretical value(P>0.05) byHaploview software test in hepatocellular carcinoma clustering families (FHCC)group and the families without any cancer (FNC) group, show that samples ofthis experiment with a group representative.(3) TGF-β1gene polymorphism genotypes and allele frequency in FHCCgroup and FNC group:1.the frequency of TGF-β11800469sites of genotypesTT,CT,CC were47.2%、41.1%、11.7%in FHCC group respectively,33.7%、45.3%、21．0%in FNC group respectively. There was a statistical differencebetween the FHCC group and FNC group（P=0.004）.The distribution ofTGF-β11800469of alleles T/C were67.8%and32.2%in FHCC grouprespectively,56.3%、43.7%in FNC group respectively. There was astatistical difference between the FHCC group and FNC group（P=0.001）.Thefrequency of TGF-β11800469TT and TGF-β11800469T was more higher inFHCC group than that of FNC group（P=0.004；P=0.001）.2. the frequency of TGF-β2241715sites of genotypes TT、GT、GG were46.7%、41.6%and11.7%in FHCC group respectively,33.2%、44.9%and21．9%in FNC grouprespectively. There was a statistical difference between the FHCC group andFNC group（P=0.003）.The distribution of TGF-β12241715of alleles T/Gwere67.5%and32.5%in FHCC group respectively,55.6%and44.4%inFNC group respectively. There was a statistical difference between the FHCCgroup and FNC group（P=0.000）.The frequency of TGF-β12241715TT andTGF-β12241715T was more higher in FHCC group than that of FNC group（P=0.004；P=0.000）.3. the frequency of TGF-β18105161sites of genotypesTT,CT,CC were29.4%、51.9%and18.7%in FHCC group respectively,42.5%、45.8%and11．7%in FNC group respectively. There was a statisticaldifference between the FHCC group and FNC group（P=0.009）.The distributionof TGF-β18105161of alleles T/C were55.4%and44.6%in FHCC grouprespectively,65.4%and34.6%in FNC group respectively, that was astatistical difference between the FHCC group and FNC group（P=0.003）.Thefrequency of TGF-β18105161TT genotype and TGF-β18105161T allele wasmore higher in FHCC group than that of FNC group（P=0.005；P=0.003）.4. thefrequency of TGF-β1747857site of genotypes AA,AG and GG were82.7%、15.0%and2.3%in FHCC group respectively,90.7%、9.3%and0%in FNCgroup respectively. There was a statistical difference between the FHCC groupand FNC group（P=0.005）.The distribution of TGF-β1747857of alleles G/Awere90.2%and9.8%in FHCC group respectively,95.3%and4.7%inFNC group respectively, that was a statistical difference between the FHCCgroup and FNC group（P=0.004）.The frequency of TGF-β1747857GG genotypeand TGF-β1747857G allele was more higher in FHCC group than that of FNCgroup（P=0.016；P=0.004）.5.There were no statistical differences between theFHCC group and FNC group in the frequencies of the genotypes CC/TC/TT and alleles T/C at the rs2241716locus（P=0.105；P=0.160）.6. The distributionof the genotypes GG/GA/AA at the rs11466345site was no significantdifference between the cases group and the control group（P=0.161）,while thefrequency of rs11466345T alleles in the FHCC group were high than that ofFNC group（P=0.049）.(4) TGF-β1polymorphisms in linkage disequilibrium test: three sites ofrs2241715、 rs2241716and rs1800469exist in strong linkage disequilibrium.The strength of linkage showed D‘=1，r2=0.984between rs1800469andrs2241715. The two sites of rs1800469and rs2241716exist in linkagedisequilibrium(D‘=1， r2=0.547).Also, the sites of TGF-β1rs2241715andTGF-β1rs2241716showed linkage disequilibrium(D‘=1，r2=0.538).(5) TGF-β1haplotypes analysis:SHSsis software was applied to performhaplotyping analysis. CGC、CGT and TTC were the most common haplotypesin rs1800469、rs2241715、rs2241716among the two groups. The frequency ofthem was more than30%. The distribution of haplotype CGC、CGT and TTChad no significant difference among the two groups（all P>0.05）. The frequencyof FHCC group was11.2%、23.3%and64.8%，as compared with12.6%、19.6%and67.9%in the FNC group. TGC only distributed in the FNC group, thefrequency was0.7%.(6) Conditional logistic regression analysis was applied to perform multiplefactors analysis of genetic and environmental factors: list rs1800469、rs2241715、rs8105161、rs747857、rs2241716、rs11466345and the exposurefactors of smoking、drinking、drinking pond water、HBV infection intoConditional logistic regression model, lastly, the risk factors entered the modelthat were sorted in ascending order were rs2241715TT, rs1800469TT,HBV-DNA positive, alcohol consumption and drinking pond water.(7)Stratification analysis of risk genotypes: the result showed that the frequency of rs1800469TT genotype and rs1800469T allele decreased with thealienation of consanguinity, although with no statistical difference in the ratesbetween FHCC group and FNC group.(8) Application of crossover analysis on interaction of gene andenvironment: In additive model, OR value of interactive effect ofTGF-β1rs1800469TT and drinking of pool water was6.44，S=1.50，AP=0.28;OR value of interactive effect of TGF-β1rs1800469TT and drinking was3.45，S=1.29，AP=0.16;OR value of interactive effect of TGF-β1rs1800469TT andHBV-DNA was5.38，S=4.13，AP=0.62.All the interactions mentioned abovewere statistically significant(P<0.05). The assumption test U=0.394,0.432,1.105respectively, and all p<0.05.Conclusion:(1)TGF-β1rs1800469and TGF-β1rs2241715SNPs may begenetic factors of HCC familial clustering in Guangxi population.(2) TGF-β1rs8105161TT and TGF-β1rs747857GG may be protectivefactors for HCC familial clustering in Guangxi population.(3)Drinking of pond water、alcohol consumption and HBV-DNA positivemay be main exposure factors of HCC familial clustering in Guangxipopulation.(4) Composite factors caused HCC familial clustering in Guangxipopulation, multiple genetic factors, environmental factors or gene-environmentfactors exist a interaction effect. Part Ⅱ Study on the relationship between TGF-β1expressionin GuangXi China population and familial clustering ofhepatocellular carcinomaObjectives: To approach the effect of expression of TGF-β1level in serumand plasma on HCC familial clustering in Guangxi population qualitatively andquantitatively.Methods: confirm214family members from34hepatocellularcarcinoma clustering families (FHCC) as the cases and37families without anycancer (FNC),select214members from37families without any cancer (FNC)as the control matched214member of the cases in terms of ages(±5), HBsAg,nationality, habitation and gender. All subjects were from11areas in Guangxiwhere the incidence of hepatocellular carcinoma keep high level. The study isdivided into two small parts:Firstly, compare the transforming growth factor β1level in serum between the hepatocellular carcinoma clustering families (FHCC)group and the families without any cancer (FNC), compare serum TGF-β1levels among different groups of risk loci genotypes, analysis stratification toserum TGF-β1levels according to the blood relationship, with ELISA assay todetect the serum TGF-β1level of428cases. Second, use western blot assay todetect blood protein expression of82cases including11probands and theirmatches,10cases of first-degree relatives selected randomly in hepatocellularcarcinoma clustering families and their matches,10cases of second degreerelatives selected randomly in hepatocellular carcinoma clustering families andtheir matches,10cases of third degree relatives selected randomly inhepatocellular carcinoma clustering families and their matches, compare thedifference of plasma TGF-β1protein gray value between case group and control group, analysis blood TGF-β1protein gray value according to the bloodrelationship with stratification, explore the relationship of TGF-β1and livercancer familial aggregation in perspective of protein expression. Datas wereinput and analyzed statistically in SPSS17.0.Variable values can be summarisedas a mean value and standard deviation. Paired-sample t-test was used tocompare mean value of the sample within two groups. Use Kruskal-Wallis t-testto compare population distribution among multiple groups, withα=0.05P＜0.05as statistically significant.Result：1．The effect of serum TGF-β1level on HCC familial clustering(1)The serum TGF-β1level in HCC familial clustering group was muchhigher than that in FNC group.（28.495±17.495ng/mlVS20.24±7.56ng/ml，P=0.000）。(2)Stratification analysis on serum TGF-β1level according to genotype:Serum concentrations of TGF-b1were significantly higher in all rs1800469TTsubjects than that in all rs1800469CT subjects （39.445±7.445ng/mlVS26.33±13.65ng/ml， P=0.000） and all rs1800469CC subjects（39.445±7.445ng/mlVS26.245±13.555ng/ml，P=0.000）.no significantdifferences in serum TGF-b1levels were found in CT subjects with CC subjects（26.33±13.65ng/mlVS26.245±13.555，P=0.347）. Serum concentrations ofTGF-b1were significantly higher in all rs2241715TT subjects than that in allrs2241715GT subjects（30.33±15.66ng/mlVS23.55±11.45ng/ml，P=0.000）and all rs2241715GG subjects （30.33±15.66ng/mlVS20±10ng/ml， P=0.000）.no significant differences in serum TGF-b1levels were found in GTsubjects with GG subjects（23.55±11.45ng/mlVS20±10ng/ml，P=0.061）.(3) Compare of TGF-b1serum concentrations in subjects of FHCC groupand FNC group with risk genotypes: Serum concentrations of TGF-b1were significantly higher in rs1800469TT subjects in FHCC group than that inrs1800469TT subjects in FNC group（43.395±3.495ng/mlVS35.95±3.95ng/ml，P=0.000）. Serum concentrations of TGF-b1were significantly higher inrs2241715TT subjects in FHCC group than that in rs2241715TT subjects inFNC group（18.815±4.145ng/mlVS38.995±6.995ng/ml，P=0.000）.2. Results of Western blot test(1) All subjects in the case and the control group express TGF-β1proteinand no abnormal protein bands were found, but the expression of protein levelswere obviously different.(2) Compared to the FNC group, TGF-β1protein content of FHCC groupwas higher.The difference had statistically significant（p＜0.01）.(3) Stratification analysis showed that TGF-β1protein content had nostatistically difference among probands、first degree、second degree、threedegree relatives(P>0.05).TGF-β1protein gray value very high in probands, butit decreased with blood relationship.Conclusion:(1)High serum TGF-β1levels may be associated withfamilial clustering of hepatocellular carcinoma.(2)High levels of plasma TGF-β1may be associated with familial clustering of hepatocellular carcinoma.(3)High serum TGF-β1levels or high levels of plasma TGF-β1may be associatedwith immune status, But it need to be confirmed through T cell or B cellfunction.更多还原