【作者】 史强；

【导师】 楼宜嘉；

【作者基本信息】 浙江大学 ， 药理学， 2006， 博士

【摘要】 微粒体GST1(microsomal glutathione S-transferase 1,MGST1)是一个Ⅱ相药物代谢酶,在体内分别催化两类反应,(1)催化还原型谷胱甘肽(glutathione,GSH)与亲电子物质的结合反应,使其毒性更低更易排出体外;(2)催化氧化应激产物和GSH的结合,而执行抗脂质过氧化功能。 MGST1最显著的特性体现在活性调控方面,其基因和蛋白表达水平都相对恒定;但多种翻译后修饰可成倍提高其催化活性,修饰位点一般位于酶活性中心第49位半胱氨酸残基(Cys49)上的巯基,例如巯基试剂N-乙基马来酰亚胺(N-ethylmaleimide,NEM)和5,5’-二硫代双(2-硝基苯甲酸)[5,5’-dithionbis(2-nitrobenzonic acid),DTNB],及烷化剂等都可对Cys49特异性修饰而使酶催化区域得以激活。Cys49上的修饰方式还包括S-亚硝酰化修饰以及氧化修饰。另外,MGST1的酪氨酸(tyrosine,Tyr)也可发生硝基化修饰和分子间Tyr交联反应,后者导致二聚体形成,均可引起酶活性升高。 上述MGST1各种修饰激活方式具有组织专属性,仅发生在哺乳动物肝脏,而其余组织,如小肠、肾上腺和睾丸所含的MGST1并不能被激活。可见MGST1激活现象是哺乳动物肝组织特有的一种保护机制,推测其在肝脏损伤过程中具有重要意义。对乙酰氨基酚(acetaminophen,AP)导致的肝脏损伤模型中,MGST1呈现短暂激活现象,但激活机制迄今未明。 MGST1激活的诸多修饰方式基本来自体外实验,如利用纯化的酶或微粒体,但其体内生物学意义并不明确。微粒体和纯酶的实验结果往往只代表一种体外的化学反应,在体内复杂的调控体系中并不一定具有实际意义;且文献报道大多缺乏直接的实验证据,结论一般来自间接推测,导致某些修饰方式仍存在较大争议。但在体内情况下,迄今尚无任何一种修饰方式被确切地验证过,所以MGST1体内激活机制是目前亟需解决的一个焦点问题。另一方面,某些新近报道的体外修更多还原

【Abstract】 Microsomal glutathione S-transferase 1 (MGST1) is a phase two drug metabolism enzyme which catalyzing both the conjugation reaction between glutathione (GSH) and endogenous/ extraneous electrophiles, and the reaction of various oxidant with GSH. The consequence of the former reaction is that the electrophiles are transformed into less toxic substances and become more easily to be excreted, thus serving as a detoxification process. The latter reaction is apparently an anti-lipidperoxidation mechanism.The most striking feature of MGST1 lies in the regulation of its catalytic activities. The expression of MGST1, both at the gene and protein level, was relative stable and constant. However, a number of post-translational modifications would greatly enhance the catalytic activity. These modifications usually occur to its single cysteine (Cys49), which locates in the catalytic centre facing the cytosolic side. The most well established modifications include the binding of N-ethylmaleimide (NEM), etc. to the Cys49, all of which lead to activation of enzyme itself. Other cysteine related modifications include S-nitrosylation and oxidation. Both S-nitrosylation and dimer formation could apparently promote the catalytic activity. In addition to Cys49, recently reports also point to modifications on tyrosine, such as tyrosine nitration and tyrosine cross-linking;these tyrosine-related modifications also greatly enhance the MGST1 activity.All the above-mentioned modifications are found in the mammalian livers. Extra-hepatic tissues, such as the intestine, the adrenal, and the testis, also contain certain amount of MGST1, but their activities can not be increased by pos t-translational modifications. MGST1 activation usually leads to enhanced capabilityin excreting reactive metabolite and other oxidants. It is therefore clearly that activation of MGSTl is a liver-specific protective mechanism in mammalians, which might play a role during liver injury.For the mechanism studies in MGSTl activation, nearly all data, with few exceptions, are from in vitro experiments, i.e., experiments using purified MGSTl or microsomal fractions. This renders the in vivo significance of these modifications in question, as many studies show that result from only purified MGSTl or microsomal fractions represents nothing more than a chemical reaction under in vitro conditions. At the same time, previous reports, in most cases if not all, failed to present direct experimental evidence. It is therefore not surprisingly that many controversies exist for certain modifications. Three decades of efforts have witnessed an increasing number of modifications on MGSTl, rendering the discovery of novel modifications almost near impossible, however, till now, none of the known modifications has been univocally and definitively proved by experimental evidence under in vivo conditions. It is thus clear that in vivo mechanism for MGSTl activation is an urgent issue awaiting clarification.In view of the potential protective effects of MGSTl activation in liver damage, this study was carried out to observe whether MGSTl activation indeed occur in different models of liver injury. Taking into account that the real activation mechanism in vivo has not been revealed, animal models are chosen in the present study, which is complemented by in vitro studies using the purified MGSTl or the microsomal fractions. This study orientates to bridging of two aspects, first, the observation that whether MGSTl is indeed activated in certain animal models of liver injuries, and second, the examination whether a specific modification indeed occur in certain animal models of liver damage.Part 1 Purification of rat liver MGSTl and preparation of rabbit anti-MGSTl polyclonal antibodyAim: To establish the methods for both the purification of MGSTl and the generation of its antibody, thus solve the problem of their unavailability incommercial resource;therefore facilitate the studies of various modifications, especially those aimed at the comparison of in vivo and in vitro results. Methods: Rat liver microsomes were prepared by ultra-centrifuge. Microsomal proteins were solubilized by detergent Triton X-100, followed by a two-step chromatography on hydroxyapatite (HA) and CM-Sepharose CL-6B. Enzymatic activity toward glutathione and l-chloro-2, 4-dinitrobenzene was measured by spectrometry. The parameters of enzyme kinetics are calculated from the Lineweaver-Burk curve. NEM was used to test the activation properties of MGSTl. SDS-PAGE was performed to examine the purity of MGSTl. Purified MGSTl was blended with Freund’s adjuvant;and the resulting water-in-oil emulsion was injected at the back of rabbit to trigger production of polyclonal antibody. Anti-MGSTl serum was test by dot blotting and Western blot. The antibody was further utilized to establish the immuno-precipitation method to purify MGSTl antigen from liver microsomal fractions.Results: As expected, a combination of HA and CM-Sepharose CL-6B chromatography produced a single protein band at 17 kDa on SDS-PAGE. The parameters of enzyme kinetics and its capability to be activated by NEM agreed well with the reported data. Both these properties indicate the successful purification of MGSTl from rat liver microsomes. By using the conventional immunological methods, rabbit anti-MGSTl polyclonal antibody was successfully generated. This antibody can be adopted in the dot blotting and Western blot detection, as well as the immuno-precipitation of MGSTl from microsomal proteins, which represents a rapid isolation method of MGSTl and would facilities the detection of post-translational modifications in vivo.Part 2 Studies on the S-ntirosylation of MGSTl and related microsomal sulfhydryl proteins in respond to NO donor treatment or iNOs induction Aim: To observe the S-nitrosylation of MGSTl in microsomal preparations incubated with NO donor in vitro, or in rats that were subjected to lipopolysaccharide (LPS) challenge, which is characterized by in vivo NO overproduction via iNOs induction.Other microsomal proteins susceptible to S-nitrosylation in vivo were also studied. Methods: In vitro S-nitrosylation was triggered by incubation of microsomal proteins with NO donor S-nitrosoglutathione (GSNO) at a series of concentrations ranging from 10 uM to 10 mM, which covers nearly the whole range of reported GSNO concentrations used under in vitro experiments. In vivo S-nitrosylation was induced by an i.p. injection of LPS, which triggers NO overproduction via iNOs induction and thus enhances liver protein S-nitrosylation. Biotin Switch method was introduced to transform the S-nitrosylated proteins into biotin labeled ones, which were then affinity purified by Streptavidin-agarose. SDS-PAGE was performed to separate the enriched S-nitrosylated proteins, followed by Western blot detection of the MGSTl. Other in vivo S-nitrosylated proteins were subjected to proteomic identification by matrix-assisted laser dissociation/ionization time-of-flight mass spectrometry (MALDI-TOF MS).Results: Biotin Switch method confirmed the speculation that purified MGSTl is indeed S-nitrosylable in vitro. However, after incubation of microsomes with NO donor, no activation of MGSTl was observed, neither did MGSTl S-nitrosylation detected. Similar results were obtained from rats subjected to LPS challenge, which indeed significantly enhanced microsomal protein S-nitrosylation, indicating that MGSTl is possibly not an S-nitrosylable protein in the microsomes and under in vivo conditions. Except for MGSTl, many other microsomal proteins were found to be S-nitrosylated in respond to LPS induced liver injury. Four of them were identified by the proteomic methods, including retinol dehydrogenase type I (RODH I), aldolase B, cytochrome P450 2C11, and peroxiredoxin 1, among which the last one has been proven to be a well-established target for S-nitrosylation, but this study represents the first evidence of its in vivo involvement, and the remaining three are novel.Part 3 Mechanisms for MGSTl activation in response to acetaminophenoverdoseAim: To examine whether MGSTl activation in acetaminophen (AP) overdose isassociated with its metabolite [N-acetyl-p-benzoquinone imine, (NAPQI)] adduction9on Cys49. Other possible mechanisms including tyrosine nitration and protein dimer formation were also examined.Methods: Anti-NAPQI antibody was prepared by the established method. Immuno-precipitation was adopted to isolate total NAPQI-binding proteins from liver microsomes of mice treated with AP. After SDS-PAGE, the isolated proteins were probed with anti-MGSTl antibody to test whether MGSTl formed adducts with NAPQI. In parallel, monoclonal anti-nitratyrosine antibody was used to enrich the nitrated proteins, which was then probed with anti-MGSTl antibody on Western Blot. Also, Western Blot was performed to test whether MGSTl forms a dimer after AP overdose. DTT was incubated with microsomes to observe whether the dimers are reversible. GSH pretreatment was adopted to determine whether dimer is related to peroxynitrite overproduction.Results: Anti-NAPQI antibody was able to recognize the BSA-NAPQI adducts, but not the BSA itself. Also, it cross-reacts with several microsomal proteins in mice treated with AP but not un-treated ones. These results show that antibody production was successful. After AP overdose, activation of MGSTl was indeed observed. But at no time course were MGSTl-NAPQI adducts detected. Neither did nitrated MGSTl observed in the damaged livers. However, MGSTl formed an apparent dimer on Western blot, which is resistant to DTT treatment in vitro, but is completely abolished by GSH pre-treatment, indicating that dimerization is mediated by peroxynitrite over-production. In addition, dimers of MGSTl were released into serum as with cGSTs.Conclusions1. Rat liver MGSTl and its polyclonal antibody were successfully prepared;the antibody is useful in both Western blot and immuno-precipitation experiments.2. MGSTl is not a S-nitrosylable protein in both the microsomes and the animal model of LPS induced liver injury. However, other proteins, including retinol dehydrogenase type I (RODH I), aldolase B, cytochrome P450 2C11, and peroxiredoxin 1 were identified as in vivo protein targets for S-nitrosylation, among which the former three are novel protein targets for in vivo S-nitrosylation, while thelast one has not been revealed in animal models before.3. The catalytic activity of aldolase is subjected to regulation by S-nitrosylation, which might be a regulatory mechanism in response to nitrosative stress and play a role in controlling glycolysis.4. Anti-NAPQI antibody was successfully raised, and by using this antibody, it was proved that MGSTl did not form an NAPQI adduct in vivo after AP overdose. Neither did MGSTl undergo nitration in this model. The only observable modification on MGSTl was protein dimerization, which is mediated by peroxynitrite overproduction.5. MGSTl dimers were released out from hepatocytes in the later stage of AP damage, accompanied by the disappearance of dimers in the liver and the appearence of dimers in the serum. This is a novel discovery in the study.更多还原