乙醯基轉移酶Naa10p 與去氧核醣核酸 體外結合能力之探討

Abstract

人類N端乙醯轉移酶10蛋白質(hNaa10p)的突變與不正常表現，在嚴重發育遲緩，如致死性奧格登症候或智能障礙，以及癌症生成中扮演重要的角色。我們實驗室發現hNaa10p促使去氧核醣核酸甲基轉移酶1(DNMT1)降低抑癌基因的表現，且在人類肺癌細胞中hNaa10p可結合至特定抑癌基因啟動子。我們實驗室也發現在老鼠胚胎及幹細胞中，Naa10p可與許多特定的印記基因調控區域結合。然而，hNaa10p如何結合至去氧核醣核酸依舊未知，且hNaa10p去氧核醣核酸結合能力與病理角色的關聯也尚未清楚。本論文中，我們利用人類上皮鈣離子依賴性之黏合蛋白(E-cadherin)啟動子以及老鼠H19印記基因調控區域(H19-ICR)作為體外研究模式，並藉由電泳膠遷移遲緩分析(EMSA)與hNaa10p大規模的點突變實驗，以瞭解去氧核醣核酸序列如何被hNaa10p辨識，而hNaa10p上哪些胺基酸對於去氧核醣核酸結合是重要的。第一部份的結果中，我們發現hNaa10p特定結合至E-cadherin啟動子 -53到 -33鹼基對的位置。K165A或奧格登綜合症相關的S37P則降低了hNaa10p結合去氧核醣核酸的能力。K165A突變依然保有N端乙醯轉移酶活性，也說明此突變不太可能影響hNaa10p的整體結構。第二部份的結果證實(1) hNaa10p結合至H19-ICR的GCXGXG序列; (2)老鼠Naa10p變異體一(Variant 1)結合至H19-ICR片段，而老鼠Naa10p變異體二(Variant 2)則失去去氧核醣核酸結合能力; (3)老鼠幹細胞中內源性的Naa10p可結合至H19-ICR片段; (4) hNaa10p可結合至核小體(nucleosomal templates)，而人類疾病相關之點突變蛋白則降低此功能; (5) hNaa10p臨床點突變S37P、Y43S、V107F或F128L造成活體外(in vitro)蛋白質穩定性下降，而R83C或R116W則不影響; (6)相較於半甲基與全甲基化的去氧核醣核酸，hNaa10p對於無甲基化的去氧核醣核酸具有較高親和力; (7) hNaa10p促使Dnmt1結合至H19-ICR，而S37P或K165A突變則失去此能力; (8) 在老鼠幹細胞的核萃取蛋白中，Naa10p維持Dnmt1的活性而不影響Dnmt3a/b。總結來說，此體外研究找出Naa10p中對於去氧核醣核酸結合的重要胺基酸，以及去氧核醣核酸序列。更重要是，本研究指出降低去氧核醣核酸結合能力，可能是造成hNaa10p點突變所產生的人類疾病原因之一。

Mutation or abnormal expression of human N-α-acetyltransferase 10 protein (hNaa10p) leads to severe developmental delays or cancer formation, respectively. Our lab previously reported that hNaa10p recruited DNA methyltransferase 1 (DNMT1) to silence tumor suppressor genes (TSG) (Lee et al., 2010). In that study, we demonstrated that hNaa10p associated with specific TSG promoters in human lung cancer cells. Our lab also found that Naa10p bound to multiple imprinting control regions in mouse embryos and ESCs. However, how hNaa10p binds to DNA remains elusive. Nor do we understand whether DNA binding of hNaa10p correlates with its pathological role. In this study, we used the human E-cadherin promoter and the imprinting control region of mouse H19 (H19-ICR) gene as models to characterize the DNA element recognized by Naa10p and the amino acids required for Naa10p DNA binding by the in vitro electromobility shift assay (EMSA) combined with extensive site-directed mutagenesis. Part I of the Result Section indicates that hNaa10p specifically bound to -53 to -33 (relatively to the transcription start site, TSS) of the E-cadherin promoter. Moreover, the K165A or the Ogden syndrome-associated S37P mutation impaired the DNA binding of hNaa10p. Importantly, the K165A mutant still maintains the N-acetyltransferase activity, suggesting that the mutation does not disrupt the overall hNaa10p structure. Part II of the Result Section demonstrates that (1) hNaa10p bound to the GCXGXG motif of H19-ICR; (2) mouse Naa10p (mNaa10p) Variant 1, but not Variant 2, bound to H19-ICR oligo; (3) the endogenous mNaa10p from mouse ESCs bound to H19-ICR oligo; (4) WT hNaaa10p but not human disease-associated mutants bound to H19-ICR oligo and nucleosomal templates; (5) hNaa10p with clinical mutation S37P, Y43S, V107F or F128L but not R83C or R116W, showed impaired protein stability in vitro; (6) hNaa10p preferentially bound to non-methylated mouse H19-ICR, compared to hemi- or fully-methylated DNA; (7) WT hNaa10p but not S37P or K165A mutant enhanced Dnmt1 binding to H19-ICR; and (8) Naa10p maintained the activity of Dnmt1 but not Dnmt3a/b in nuclear extracts of mouse ESCs. In summary, the study reveals the essential residues and DNA element for Naa10p binding to DNA in vitro and suggests that the DNA binding activity might contribute to hNaa10p mutation-associated disease formation.