應用外顯子定序來找出年輕人之成熟期發病型糖尿病的基因

Application of whole-exome sequencing to identify genes for Maturity-Onset Diabetes of the Young (MODY)

陳遊(Yu Chen)

指導教授 ： 楊偉勛

共同指導教授 ： 陳沛隆(Pei-Lung Chen)

國立臺灣大學/醫學院/分子醫學研究所/碩士(2018年)

https://doi.org/10.6342/NTU201802065

摘要

年輕人之成熟期發病型糖尿病屬於家族性體染色體顯性遺傳糖尿病，特徵為單基因失活性變異，所造成的胰島素分泌與作用機制發生缺陷。發病年齡較早，通常小於25歲，且體型外觀較瘦，因此容易與第1型糖尿病混淆。不同於第1型糖尿病患者須終身施打胰島素，特定基因型的年輕人之成熟期發病型糖尿病患者，如：GCK-MODY，青少年時期可藉由運動及飲食控制血糖。其他如：HNF4A-MODY、HNF1A-MODY、ABCC8-MODY、KCNJ11-MODY首選的降血糖藥物建議為磺胺尿素類，而非第2型糖尿病首選用藥二甲雙胍類。基因診斷對此類遺傳性糖尿病，扮演著調整治療方針與評估疾病進展的重要角色。目前已知14個基因（HNF4A、GCK、HNF1A、PDX1、HNF1B、NEUROD1、KLF11、CEL、PAX4、INS、BLK、ABCC8、KCNJ11、APPL1）會導致年輕人之成熟期發病型糖尿病，然而亞洲族群中，超過80%的患者，其基因型並非以上的14個，屬於基因未明型。在考量樣本數較少以及花費成本後，我們應用全外顯子定序的方式，試圖尋找台灣此類型遺傳性糖尿病的致病基因。
本研究經臨床診斷為年輕人之成熟期發病型糖尿病表現型患者，可納入本研究。納入條件為個案25歲前患有糖尿病或糖尿病前期的症狀，不具有第1型和第2型糖尿病典型的臨床表徵，且至少有兩代家族性遺傳的糖尿病史，以及父母其中一方是糖尿病患者，可加入研究與接受遺傳諮詢。在2017年，我們收案了2個符合年輕人之成熟期發病型糖尿病表現的家族，包含5名患者與1名未罹病者，男女比1：1，並以NA12878作為對照組標準品，共計7個樣本。委外使用Agilent SureSelect V6 Exon kit擷取人類基因體中的全外顯子序列後，以Illumina Hiseq進行次世代定序，平均7個樣本生成的序列425926569±6273250.0條，高達99.72%的序列是成對且對齊在參考序列hg19上，在10倍深度下標的區域覆蓋率為93.77%，平均定序深度是51.46倍，其中89.30%鹼基的定序品質（Q值）至少等於30。經由標準化的生物資訊處理流程（BWA-MEM、Picard、GATK、GATK/HaplotypeCaller、ANNOVAR），平均註解的SNVs與Indels數為390775.6±44822.0個，藉由親本三元體與共分離的分析方式，分別選出患者家族內共同的變異點位，其最小等位基因頻率在基因資料庫中（Esp6500、1000 Genomes Project、ExAC）均小於0.01，而在多項軟體預測的條件中，排除多型性表現型預測為良性的變異點位（Polyphen2_HVAR_score ≦0.446）或排除胺基酸置換不影響蛋白質功能的點位（SIFT > 0.05）。利用罕見單基因糖尿病之105個基因清單中，篩選出致病點位，再利用桑格定序做確認。最後，成功地在兩個家族分別找出致病點位（INS, c.125T>C;p.V42A）和（GCK,c.1318G>T;p.E440X）。
我們在台灣族群第一次發現這2個點位，皆位於國外文獻記載的致病基因中，且藉由基因型的確診，使我們將更加邁向精準醫療的時代。應用全外顯子次世代定序的基因診斷流程，同樣具有確診性，並提供機會尋找新的致病基因，對於小樣本類型的研究，帶來更有效率及便利的診斷方式。

關鍵字

年輕人之成熟期發病型糖尿病 ； 全外顯子定序 ； 分子學基因診斷 ； 胰島素 ； 葡萄糖激酶

並列摘要

Maturity-onset diabetes of the young (MODY) is a monogenic diabetes mellitus characterized by an autosomal dominant mode of inheritance. The genetic defects result in insulin secretion or functional impairment. Because of their younger onset age usually less than 25 and their lean physical appearance, MODY patients are easily misdiagnosed as type 1 diabetes (T1D). Compared to T1D patients who need the insulin supplement for whole life, some patients of MODY can be managed with oral hypoglycemic agents or diet. For instance, the first line recommended drug in HNF4A-MODY, HNF1A-MODY, ABCC8-MODY, KCNJ11-MODY is sulfonylurea, not metformin, which is the first line recommended drug for most type 2 diabetes (T2D). Therefore, a correct genetic diagnosis is important to guide the treatment strategies, also to predict disease progress. Although there are at least 14 MODY genes (HNF4A, GCK, HNF1A, PDX1, HNF1B, NEUROD1, KLF11, CEL, PAX4, INS, BLK, ABCC8, KCNJ11, APPL1) more than 80% of Asian MODY cases cannot be assigned to a specific genetic diagnosis in previous reports. Our aim is to find disease causative genes of MODY families in Taiwan. Considering the cost-effect in a small-sample-size study, we used whole-exome sequencing (WES) for genotyping.
In this study, the inclusion criteria were probands who had atypical manifestations of T1D and T2D, whose age at onset of diabetes or pre-diabetes was less than 25, who had at least 2 generations of familial diabetes history, and one of their parents had diabetes. In 2017, six subjects from two Taiwanese MODY families were enrolled in this study. Genomic DNA of six subjects and one international reference DNA sample (NA12878) were captured by Agilent SureSelect V6 Exon kit and sequenced by Illumina Hiseq. The total generated reads were 425926569±6273250, and 99.72% of reads were properly paired and mapped to the reference sequence (GRCh37/hg19). Eighty-nine point thirty percent of readed bases were at least large than 30 in Phred quality scores (Q scores). The average of targeted coverage of seven samples was 93.77% at 10 folds read depth. The average read depth of seven samples was 51.46 folds. Variants calling were performed by bioinformatic pipelines (BWA-MEM, Picard, GATK, GATK/HaplotypeCaller, ANNOVAR). The average variants of 7 samples were 390775.6±44822.0. After a trio and a co-segregation analysis, disease-causing variants were selected sequentially by minor allele frequency (MAF) less than 0.01 in population databases (Esp6500, 1000 Genomes Project, ExAC), by in silicon functional predictions (by excluded Polyphen2_HVAR_score ≦0.446 as benign or by excluded SIFT > 0.05 as amino acid substitution tolerated), and by a list of 105 DM-related genes (MODY, NDM and other rare form of monogenic diabetes). The final results were confirmed by Sanger sequencing.
Two variants (INS, c.125T>C;p.V42A of insulin) and (GCK, c.1318G>T;p.E440X of glucokinase) were successfully identified respectively in 2 MODY families. In Taiwanese, these two variants were found in the MODY disease-causing genes for the first time. We are going further toward the era of precision medicine by diagnosing diseases with the correct genotyping. This study shows us that WES is a practical and effective method for diagnosing unexplained monogenic disease in small-sample-size studies.

並列關鍵字

Maturity-onset diabetes of the young (MODY) ； whole-exome sequencing ； molecular genotyping ； insulin ； glucokinase

參考文獻

1. Ogurtsova, K., et al., IDF Diabetes Atlas: Global estimates for the prevalence of diabetes for 2015 and 2040. Diabetes Res Clin Pract, 2017. 128: p. 40-50.

2. 國民健康署, 2017國民健康署年報中文版. https://health99.hpa.gov.tw/educZone/edu_detail.aspx?CatId=21980&kw=, 2018.

3. American Diabetes, A., 2. Classification and Diagnosis of Diabetes: Standards of Medical Care in Diabetes-2018. Diabetes Care, 2018. 41(Suppl 1): p. S13-S27.

4. DAROC, Clinical Practice Guidelines for Diabetes Care- 2018, Taiwan, Diabetes, Association of the R.O.C. 2018.

5. Barbour, L.A., et al., Cellular mechanisms for insulin resistance in normal pregnancy and gestational diabetes. Diabetes Care, 2007. 30 Suppl 2: p. S112-9.