遺傳基因於肥胖兒童非酒精性脂肪肝疾病的影響

Genetic Influences in Non-alcoholic Fatty Liver Disease in Obese Children

林裕誠(Yu-Cheng Lin)

指導教授 ： 張美惠

共同指導教授 ： 倪衍玄(Yen-Hsuan Ni)

國立臺灣大學/醫學院/臨床醫學研究所/博士(2011年)

https://doi.org/10.6342/NTU.2011.01231

摘要

非酒精性脂肪肝疾病的範疇包含從單純性肝臟脂肪堆積到潛在有致死性風險的非酒精性脂肪肝炎，甚至是已經到達肝硬化的程度。隨著兒童肥胖逐漸盛行，非酒精性脂肪肝疾病是愈來愈惡化的全球性兒童青少年健康問題。雖然非酒精性脂肪肝疾病是全世界造成慢性肝病的最常見原因，但是其致病機轉仍然不明。肥胖是非酒精性脂肪肝疾病發生的最重要危險因子，會引起胰島素抗拒。然而不是所有肥胖兒童都會發生非酒精性脂肪肝疾病，這表示其他環境與基因遺傳因子對於非酒精性脂肪肝疾病的易感受性與是否會惡化具有決定性的關鍵影響。
有關非酒精性脂肪肝疾病的致病機轉，Day等人於1998年提出雙擊假說，第一擊是肝臟內脂肪的過度堆積，與胰島素抗拒有關。經過第一擊後的肝臟細胞會容易受第二擊的影響，使得肝臟細胞受損、發炎與纖維化。第二擊是由於脂肪過度氧化產生的氧化壓力與細胞激素造成傷害。我們選擇候選基因的原則也是基於雙擊理論中對第一擊與第二擊具有重要影響力的基因。
對於第一擊部分， Patalin-like phospholipase (PNPLA3)與脂肪細胞內三酸甘油酯的水解有關，對於非酒精性脂肪肝疾病第一擊致病過程有重要影響。Romeo等人首先報告PNPLA3 rs738409單一核甘酸多型性在成人與肝臟脂肪堆積的嚴重度有很強的相關性。我們假說PNPLA3 rs738409 G對偶基因會增加在肥胖兒童中發生非酒精性脂肪性肝疾病的危險性。我們收案520名6到18歲的肥胖兒童青少年，以肝臟超音波評估是否有非酒精性脂肪性肝疾病，同時測量他們的肝功能指數血清ALT值，並鑑定PNPLA3 rs738409單一核甘酸多型性為CC、CG或是GG基因型。以多變量邏輯迴歸分析與線性迴歸分析分別檢定PNPLA3 rs738409單一核甘酸多型性對於兒童非酒精性脂肪肝疾病與血清ALT值的影響。我們發現19.6%的肥胖兒童青少年有非酒精性脂肪肝疾病。相較於PNPLA3 rs738409 CC野生基因型，CG基因型增加了2.96倍發生非酒精性脂肪肝疾病的危險性(95%信賴區間：157－5.59；P = 0.0008)；GG基因型甚至增加了5.84倍發生非酒精性脂肪肝疾病的危險性(95%信賴區間：2.59－13.16；P < 0.0001)。另外，相較於PNPLA3 rs738409 CC野生基因型，CG基因型組的平均血清ALT值增加了4.77 IU/L (P = 0.0435)；GG基因型組的平均血清ALT值增加了10.86 IU/L (P < 0.0001)。總結來說，對於台灣地區肥胖兒童，PNPLA3 rs738409單一核甘酸多型性基因變異增加了發生非酒精性脂肪肝疾病的危險性。由於PNPLA3 rs738409 同型合子GG基因型增加非酒精性脂肪肝疾病的危險性為異型合子CG基因型所增加的2倍，表示PNPLA3 rs738409 G對偶基因對非酒精性脂肪肝疾病的影響效應為顯性遺傳模式。
對於第二擊部分，氧化壓力扮演非酒精性脂肪肝疾病第二擊致病過程中很重要的角色。膽紅素具有抗氧化的功能，而UGT1A1基因的變異是影響膽紅素濃度的主要因素。在第一部分研究，我們假說UGT1A1基因變異會減少非酒精性脂肪肝疾病的發生。我們收案了234名6到13歲的肥胖兒童，以肝臟超音波評估是否有非酒精性脂肪肝疾病。UGT1A1基因型，包括UGT1A1*6與UGT1A1*28，以聚合酶連鎖反應－限制酶片段長度多型性方法來鑑定。以多變量邏輯迴歸分析檢定UGT1A1基因變異對於兒童非酒精性脂肪性肝疾病發生的影響。我們發現12%肥胖兒童受試者有非酒精性脂肪肝疾病。相較於沒有非酒精性脂肪肝疾病組，非酒精性脂肪肝疾病組的血清總膽紅素值組顯著為低(0.36 ± 0.38 mg/dL vs. 0.25 ± 0.30 mg/dL；P = 0.021)。於控制了身體質量指數、腰臀圍比與adiponectin值的效應後，UGT1A1*6基因變異為兒童非酒精性脂肪肝疾病的一項獨立顯著性保護因子，勝算比為0.31 (95%信賴區間：0.11－0.91；P = 0.033)。至於UGT1A1*28基因變異則與我們受試者的非酒精性脂肪肝疾病沒有顯著相關性。總結來說，對於台灣地區肥胖兒童，UGT1A1*6基因變異減少了發生非酒精性脂肪肝疾病的危險性。UGT1A1基因型是對於兒童非酒精性脂肪肝疾病發生的一個新影響因子。
根據文獻上家族集簇性研究顯示兒童非酒精性脂肪肝疾病的遺傳性很高。因此，我們分析與非酒精性脂肪肝疾病致病過程第一擊與第二擊有重要相關的11個候選基因中的24個單一核甘酸多型性，在控制了PNPLA3 rs738409基因型的干擾因素後，以多變量邏輯迴歸分析檢定這些單一核甘酸多型性基因變異對於兒童非酒精性脂肪肝疾病的影響。挑選的候選基因，包括了自體吞噬途徑上的ATG16L1、 PIK3C3及 IRGM；類鐸受體途徑上的TOLLIP及CD14；發炎反應途徑上的TNF-α；脂肪酸代謝途徑上的PPARG及PPARGC1A；與adiponectin訊息傳導途徑上的ADIPOQ、ADIPOR1及ADIPOR2。這些單一核甘酸多型性在華人漢族中的少數對偶基因頻率至少要大於5%。我們總共挑選出95名有非酒精性脂肪肝疾病的個案組與91名沒有非酒精性脂肪肝疾病的控制組。在控制了腰圍、三酸甘油酯、adiponectin值與PNPLA3 rs738409 GG基因型等干擾因子後，PPARGC1A rs8192678基因變異顯著地增加了發生兒童非酒精性脂肪肝疾病的危險性(勝算比：2.21；95%信賴區間：1.04－4.69；P = 0.0389)；TNF-α rs1799964基因變異則顯著地減少了發生兒童非酒精性脂肪肝疾病的危險性(勝算比：0.49；95%信賴區間：0.24－0.99；P = 0.0487)。我們進一步分析所有520名受試者的PPARGC1A rs8192678單一核甘酸多型性基因型，於控制PNPLA3 rs738409基因變異型後，相較於野生基因型，PPARGC1A rs8192678基因變異型增加了1.90倍發生非酒精性脂肪肝疾病的危險性(95%信賴區間1.02－3.57；P = 0.045)。總結來說，對於台灣地區肥胖兒童，PPARGC1A rs8192678單一核甘酸多型性基因變異與TNF-α rs1799964單一核甘酸多型性基因變異會獨立顯著性地影響發生非酒精性脂肪肝疾病發生的危險性。
另外，非酒精性脂肪肝疾病病患的肝臟組織檢體會有較高的膽固醇堆積。相對於單純性肝臟脂肪堆積病患，非酒精性脂肪肝炎病患會有更高的肝臟組織內膽固醇堆積。這結果顯示肝臟膽固醇過度堆積對於非酒精性脂肪肝疾病的第二擊致病過程有重要影響。我們以油酸引發的Huh7脂肪肝細胞模式來探討肝臟細胞脂肪堆積影響膽固醇代謝異常的致病機轉。我們發現膽固醇新生合成關鍵酵素HMG-CoA reductase的訊息核醣核酸表現量，與控制組相較，在0.5 mM油酸組增加19%基因表現量(P = 0.024)。HMG-CoA reductase是膽固醇新生合成的速率決定酵素，增加HMG-CoA reductase基因表現量會導致肝臟細胞增加膽固醇合成。我們的實驗結果顯示脂肪肝細胞內過多的膽固醇堆積可能是連結非酒精性脂肪肝疾病第一擊與第二擊致病過程的重要分子因素。
我們的研究結果將有助於進一步了解影響台灣地區肥胖兒童發生非酒精性脂肪肝疾病的重要基因遺傳因子。由於文獻報告上極少有亞洲裔兒童的相關基因資料，因此我們的研究成果對於全球性分析不同種族間發生非酒精性脂肪肝疾病的基因遺傳因子，像是比較西班牙裔、非洲裔、北美或是歐洲裔白人等，具有重要的學術價值。

關鍵字

兒童 ； 非酒精性脂肪肝疾病 ； 單一核甘酸多型性 ； UGT1A1 ； PNPLA3 ； PPARGC1A ； TNF-α ； 膽固醇 ； 脂肪肝細胞模式 ； HMG-CoA reductase

並列摘要

Nonalcoholic fatty liver disease (NAFLD) represents a spectrum of conditions ranging from simple hepatic steatosis to potentially fatal nonalcoholic steatohepatitis and cirrhosis. Concurrent with the epidemic of childhood obesity, pediatric NAFLD is a growing global problem. The pathogenesis of NAFLD is multifactorial; it is strongly associated with obesity and insulin resistance. Not all children with obesity will develop NAFLD, suggesting that genetic factors may contribute to the NAFLD susceptibility.
NAFLD is a complex metabolic disease that is strongly associated with obesity and insulin resistance. Day and James proposed a two-hit hypothesis. Fat accumulation in the liver is the first hit. Additional hits, including oxidative stress, lipotoxicity, adipocytokines, etc., are potential mediators in inducing persistent liver injury. Candidate genes studied in our studies were selected based on the two-hit hypothesis.
With regard to the genetic susceptibility to the first hit, PNPLA3 rs738409 single nucleotide polymorphism (SNP), causing a nonsynonymous sequence variation I148M, had a strong association with increased liver fat content in adults. In our study, we hypothesized that the presence of the PNPLA3 rs738409 G allele would increase the susceptibility of NAFLD in obese Taiwanese children. We recruited a total of 520 obese children aged 6-18 years. Their PNPLA3 rs738409 genotypes-CC, CG, or GG-were detected by the 5'-nuclease assay. The effects of the PNPLA3 rs738409 G allele on pediatric NAFLD were evaluated based on liver ultrasonography findings and mean serum alanine aminotransferase levels in these children. We found that NAFLD was present in 19.6% of the obese children. In comparison to the subjects with CC alleles, the risk of NAFLD was increased by 2.96-fold (95% confidence interval, 1.57-5.59, P = 0.0008) in the subjects with CG alleles and by 5.84-fold (95% confidence interval, 2.59-13.16; P < 0.0001) in those with GG alleles. Variant PNPLA3 rs738409 genotypes were associated with increases in mean serum alanine aminotransferase level of 4.77 IU/L (P = 0.0435) in subjects with CG alleles and of 10.86 IU/L (P < 0.0001) in those with GG alleles compared with subjects with CC alleles. In conclusion, the variant PNPLA3 rs738409 genotypes increased the risk of NAFLD in our population of obese Taiwanese children. The effect of the G allele on pediatric NAFLD followed a dominant genetic model.
With regard to the genetic susceptibility to the second hit, oxidative stress plays an important role in the pathogenesis of NAFLD. Variants in the UGT1A1 gene contribute to increased bilirubin levels, and bilirubin can act as an antioxidant. In our study, we hypothesized that variant UGT1A1 genotypes would reduce the risk for NAFLD development. We recruited 234 obese children 6 to 13 years of age. NAFLD was determined through liver ultrasonography. The UGT1A1 genotypes UGT1A1*6 and UGT1A1*28 were detected. We assessed the effects of UGT1A1 genotypes on pediatric NAFLD. We found that in total, 12% of the obese children had NAFLD. The subjects with NAFLD had lower serum total bilirubin levels (0.25 ± 0.30 mg/dL) than did those without NAFLD (0.36 ± 0.38 mg/dL; P = 0.021). With conditioning on the effects of age- and gender-adjusted BMI, waist/hip ratio, and adiponectin levels, variant UGT1A1*6 genotypes were a protecting factor for NAFLD, with an estimated adjusted odds ratio (OR) of 0.31 [95% confidence interval (CI): 0.11-0.91; P = 0.033], but variant UGT1A1*28 genotypes were not significantly associated with the occurrence of NAFLD. In conclusion, variant UGT1A1*6 genotypes are associated with a lower risk of NAFLD in obese Taiwanese children. The UGT1A1 genotype is a new risk factor for pediatric NAFLD.
The heritability of pediatric NAFLD, as estimated by family aggregation study, is high. We genotyped 24 selected SNPs in 11 NAFLD-related candidate genes which may be involved in the first hit or second hit on the pathogenesis of NAFLD . We examined the associations between SNPs and the risk of pediatric NAFLD in obese children with conditioning on the effect of PNPLA3 rs738409 polymorphism. We selected these 24 SNPs by a pathway-driven approach, including autophagy (ATG16L1, PIK3C3, IRGM), toll-like receptor (CD14, TOLLIP), inflammatory (TNF-α), fatty acid metabolism (PPARG, PPARGC1A), and adiponectin signaling (ADIPOQ, ADIPOR1, ADIPOR2) pathways. SNPs were chosen based on minor allele frequency higher than 5% among Han Chinese. NAFLD was determined by liver ultrasonography. Associations between SNPs and pediatric NAFLD were examined using multiple logistic regression models. We found that a total of 95 cases and 91 controls were studied. The two groups matched each other in terms of age, gender and body mass index. With conditioning on the effects of waist circumference, triglyceride, adiponectin and PNPLA3 rs738409 polymorphism, one PPARGC1A SNP (rs8192678) was significantly associated with an increased risk for pediatric NAFLD OR, 2.21; 95% CI, 1.04–4.69; P=0.0389) and one TNF-α SNP (rs1799964) was significantly associated with a decreased risk for pediatric NAFLD (OR, 0.49; 95% CI, 0.24–0.99; P=0.0487). We further genotyped all 520 participants and analyzed the effect of PPARGC1A rs8192678 SNP variants on pediatric NAFLD. With conditioning on the effects of gender, adujsted BMI, waist circumference, adiponectin and PNPLA3 rs738409 polymorphism, variant PPARGC1A rs819267 genotypes increased the risk for pediatric NAFLD by 90% (OR, 1.9; 95% CI, 1.02–3.57; P=0.045). In conclusion, the variant PPARGC1A rs8192678 and TNF-α rs1799964 genotypes significantly modified the risk of NAFLD independent of the effect of PNPLA3 rs738409 polymorphism in our population of obese Taiwanese children.
Excessive cholesterol accumulation in the liver is involved in the deterioration of NAFLD. The underlying mechanism was unknown. We used an oleic acid-induced steatotic Huh7 cell model to investigate the changes in the expression levels of key cholesterol metabolism genes. We found that the mRNA expression of HMG-CoA reductase was increased by 19% in 0.5 mM oleic acid treated steatotic Huh7 cells (P = 0.024). HMG-CoA reductase is the rate-limiting key enzyme in the regulation of cholesterol biosynthesis in liver. We proved that the excessive cholesterol accumulation in the NAFLD liver is induced by increased expression of HMG-CoA reductase gene. Our results suggest that cholesterol is a possible lipid mediator linking the first hit and the second hit on the pahtogenesis of NAFLD.
In summary, we focused on finding associations between pediatric NAFLD and variants in candidate genes encoding proteins involved in NAFLD pathogenesis. Our findings may contribute to the understanding of the genetic susceptibility to pediatric NAFLD. Considering the lack of data in Asian children, the results of our genetic association studies in obese Taiwanese children could be valuable for global comparisons with data from other ethnic groups, including Hispanic, African, and North American or European origin.

並列關鍵字

children ； non-alcoholic fatty liver disease ； UGT1A1 ； PNPLA3 ； PPARGC1A ； TNF-α ； cholesterol ； steatotic cell model ； HMG-CoA reductase

參考文獻

Adams LA, Lymp JF, St Sauver J, Sanderson SO, Lindor KD, Feldstein A, Angulo P. The natural history of nonalcoholic fatty liver disease: a population-based cohort study. Gastroenterology. 2005 Jul;129(1):113-21.

Aithal GP, Thomas JA, Kaye PV, Lawson A, Ryder SD, Spendlove I, Austin AS, Freeman JG, Morgan L, Webber J. Randomized, placebo-controlled trial of pioglitazone in nondiabetic subjects with nonalcoholic steatohepatitis. Gastroenterology. 2008 Oct;135(4):1176-84.

Ambros V. The functions of animal microRNAs. Nature. 2004 Sep 16;431(7006):350-5.

Angulo P, Alba LM, Petrovic LM, Adams LA, Lindor KD, Jensen MD. Leptin, insulin resistance, and liver fibrosis in human nonalcoholic fatty liver disease. J Hepatol. 2004 Dec;41(6):943-9.

Angulo P, Lindor KD. Non-alcoholic fatty liver disease. J Gastroenterol Hepatol. 2002 Feb;17 Suppl:S186-90.