建構第五型核酸內切酶修復試管中及生物體試驗之含亞黃嘌呤受質

Abstract

在正常的生理情況下，腺嘌呤(Adenine, A)可能會因為自發性脫胺作用產亞黃嘌呤(hypoxanthine, Hx)，在DNA上則稱為脫氧肌苷(deoxyinosine, dI)。如果dI無法及時被DNA修復系統移除，細胞持續利用dI作為模板股進行DNA複製則傾向於誤置C，最終會導致A: T to G: C transition mutation的結果。 研究發現在大腸桿菌主要是由第五型核酸內切酶系統進行dI的修復。於修復系統中，第五型核酸內切酶(endonuclease V, Endo V)會辨識DNA上的dI並在5’端第二個磷酸酯鍵進行切割，由第一型DNA聚合酶(polymerase, PolI)將dI移除，隨後合成正確鹼基配對的核苷酸，最後再藉由接合酶(ligase)把缺口黏合。為了更深入研究第五型核酸內切酶修復系統，本研究製備G-I, T-I, A-I配對並於5’端帶有C-C標誌的DNA受質，透過in vitro和in vivo的方式進行修復分析。 含有dI的DNA受質是使用來自phagemid的lesion containing strand DNA、template strand DNA和含有dI之人工合成寡核酸共同組成。我們將dI的位置設計在NheI限制酶識別序列中，分別產生G-I, T-I, A-I損傷形式。當dI存在時，NheI對此序列的水解會受到影響；若dI成功被修復，NheI可以將DNA受質完全水解。在in vitro repair assay中，測定修復反應後NheI水解的程度即可進行修復的定量。在in vivo repair assay中，抽取dI受質轉型後單一菌落質體進行限制酶分析，如被NheI完全水解為修復，如NheI僅水解半數DNA則為不修復。我們也利用細菌對於C-C配對錯誤修復的不敏感，將C-C marker置於重疊的AflII和XhoI限制酵素識別位置，則可以用來扣除in vivo assay常發生的template strand loss與lesion strand loss等副反應結果。 我們利用Endo V nicking assay來檢測dI受質的品質，結果顯示100ng G-I、T-I和A-I超螺旋受質經過1U Endo V作用後，可完全轉成nicked form。接著進行Endo V滴定試驗以觀察在不同unit數的Endo V作用下，不同dI配對的受質劑量依賴性的現象(dose dependency)，結果顯示G-I、T-I受質有劑量依賴性，而A-I受質要使用0.5 U Endo V進行反應才觀察到A-I的受質被Endo V作用變成nicked form。三種dI受質以含有第五型核酸內切酶，第一型核酸聚合酶及DNA連接酶的純化蛋白修復系統進行測試，dI也可以成功被修復且G-I受質的修復效率最佳。將G-I, T-I, A-I受質藉由轉型作用送入大腸桿菌BW25113和JW5547中進行細菌體內修復試驗。發現在BW25113野生株in vivo修復比率分別為：G-I為34%，T-I為26％，A-I為31.42%，相對應JW5547突變株in vivo背景雜訊為：G-I為5%，T-I為3%，A-I為4%。

Under physiological condition, adenine could spontaneously deaminates to generate hypoxanthine. In DNA, hypoxanthine is called deoxyinosine (dI). An A: T to G: C transition mutation would occur if dI was not repaired before DNA replication. In previous study, it was found that dI lesion was mainly repaired by endonuclease V (Endo V) repair pathway in E. coli Endo V recognized dI lesion and incised on the second phosphodiester bond 5’ to the lesion. Polymerase I (PolI) removed dI by 5’-3’ exonuclease activity and synthesized correct nucleotides, then ligase sealed the nick. To further study Endo V repair pathway, G-I, T-I and A-I substrate with 5’ C-C marker were designed and used in the in vitro and in vivo assay system. The dI substrate was composed of phagemid-based lesion containing strand DNA, template strand DNA and dI containing oligo. The dI lesion was designed in the NheI recognition sequence and the existence of dI would interrupt NheI digestion. The degree of substrate digested by NheI was used to evaluate repair efficiency in the in vitro repair assay. In the in vivo repair assay, dI substate was after transformed into E. coli and the phagemid was then extracted by alkaline lysis and analyzed by AflII, XhoI and NheI digestion. If dI substrate was repaired, DNA could be completely digested by NheI. If dI substrate was not repaired, DNA was incompletely digested by NheI. Bacteria was unsensitive to C-C mismatch, therefore C-C marker was inserted into overlapped AflII and XhoI recognition sequences and used to evaluate template strand loss and lesion strand loss. The quality of dI substrate was determined by Endo V nicking assay, and the result showed that 100ng G-I, T-I and A-I substrate were completely nicked by 1U Endo V. In Endo V titration tests, G-I and T-I substrate repair was dose-dependent to Endo V. But it was not observed in A-I substrate repair. In the in vitro repair assay using purified protein systems, different kinds of dI substrates were successfully repaired by Endo V, PolI and ligase. In the in vivo repair condition, the repair efficiency was 34% to G-I, 6% to T-I and 32% to A-I in E. coli BW25113. The repair efficiency in the nfi- strain JW5547 was 5% to G-I, 3% to T-I and 4% to A-I.