# 在健康與疾病中的氧化壓力：Nrf2啟動的治療潛能 Oxidative stress in health and disease: The therapeutic potential of Nrf2 activation
## Abstract 摘要
For the past 40 years or so, oxidative stress has been increasingly recognized as a contributing factor in aging and in various forms of pathophysiology generally associated with aging. Our view of oxidative stress has been largely “superoxide-centric”, as we focused on the pathological sources of this oxygen-derived free radical and the types of molecular havoc it can wreak, as well as on the protection provided by the antioxidant enzymes, especially the superoxide dismutases, catalases, and glutathione peroxidases. In the last decade our view of oxidative stress has broadened considerably, and it is now often seen as an imbalance that has its origins in our genes, and the ways in which gene expression is regulated. At the center of this new focus is the transcription factor called nuclear factor (erythroid-derived 2)-like 2, or Nrf2. Nrf2 is referred to as the “master regulator” of the antioxidant response, modulating the expression of hundreds of genes, including not only the familiar antioxidant enzymes, but large numbers of genes that control seemingly disparate processes such as immune and inflammatory responses, tissue remodeling and fibrosis, carcinogenesis and metastasis, and even cognitive dysfunction and addictive behavior. Thus, the dysregulation of Nrf2-regulated genes provides a logical explanation for the connections, both direct and indirect, between observable oxidative stress and perhaps 200 human diseases involving these various physiological processes, each reflecting a network involving many gene products. The evolutionary self-association of these many genes under the common control of Nrf2 suggests that the immune and inflammatory systems may present the largest demand for increased antioxidant protection, apart from constitutive oxidative stress resulting from mitochondrial oxygen consumption for metabolic purposes. Gene expression microarray data on human primary vascular endothelial cells and on the SK-N-MC human neuroblastoma-derived cell line have been obtained in response to the dietary supplement Protandim, a potent composition of highly synergistic phytochemical Nrf2 activators. Pathway analysis of results shows significant modulation by Protandim of pathways involving not only antioxidant enzymes, but of those related to colon cancer, cardiovascular disease, and Alzheimer disease.
> **在过去大约40年左右的时间里，氧化应激越来越被认为是与衰老和各种与衰老有关的病理生理现象有关的因素。**
我们对氧化应激的看法主要集中在「超氧负荷中心」，我们关注这种氧导出的自由基的病理来源以及它可能造成的分子破坏，以及由**抗氧化酶（特别是超氧化物歧化酶 SOD、过氧化氢酶 CAT 和谷胱甘肽过氧化酶 GSH-Px ）提供的保护。**
在过去的十年里，我们对氧化应激的看法有了很大的拓展，现在通常将其视为一种失衡，其起源在于我们的基因以及基因表达的调节方式。
在这个新的焦点中，核因子（赤红素源性2）类似物2，或简称 NRF2，位于中心地位。NRF2 被称为抗氧化应激反应的「主调节因子」，调节数百个基因的表达，其中不仅包括我们熟悉的抗氧化酶，还包括许多控制看似不相关的过程的基因，如免疫和炎症反应、组织重塑和纤维化、癌症发生和转移，甚至认知功能障碍和成瘾行为。
因此，NRF2 调控基因的失调提供了对可观察的氧化应激和大约 200 种涉及这些不同生理过程的人类疾病之间的直接和间接联系的合理解释，每一种疾病都反映了涉及许多基因产物的网络。这些多个基因在 NRF2 的共同控制下的进化自组装表明，免疫和炎症系统可能是对增加抗氧化保护需求最大的部分，除了由于线粒体氧气消耗导致的结构性氧化应激。
根据对人类主要血管内皮细胞和 SK-N-MC 人类神经母细胞瘤细胞株进行的基因表达微阵列数据，通过膳食补充剂 Protandim（一种含有高度协同作用的植物化学物质 NRF2 激活剂的强效配方），得到了响应的结果。
> **结果的通路分析显示 Protandim 显著调节不仅包括抗氧化酶，还包括与结直肠癌、心血管疾病和阿尔茨海默病相关的通路。**
---
## 1\. Introduction 1. 介绍
### 1\.1. The concept of “oxidative stress” 1.1 “氧化应激”的概念
The term “oxidative stress” began to be used frequently in the 1970s, but its conceptual origins can be traced back to the 1950s to researchers pondering the toxic effects of ionizing radiation, free radicals, and the similar toxic effects of molecular oxygen (Gerschman et al., 1954), and the potential contribution of such processes to the phenomenon of aging (Harman, 1956). The acceptance of free radical biology was remarkably slow, probably due to the largely theoretical and hypothetical nature of its beginnings, the evanescent nature of free radicals, and the lack of experimental tools to study them. The recognition in 1968 that biological systems could produce substantial quantities of the superoxide free radical, O2radical dot−, through normal metabolic pathways (McCord and Fridovich, 1968) and that enzymes, the superoxide dismutases (SOD), had evolved with the apparent sole purpose of protecting aerobic organisms from the presumed toxicity of this free radical (McCord et al., 1969, McCord et al., 1971) spurred much interest. These enzymatic tools to both produce (via xanthine oxidase) and eliminate superoxide (via SOD) facilitated additional research in a number of areas of physiology and pathology.
「氧化应激」这个术语在 20 世纪 70 年代开始被频繁使用，但其概念起源可以追溯到 20 世纪 50 年代，当时研究人员思考电离辐射、自由基以及分子氧类似的毒性效应，以及这些过程对衰老现象的可能贡献（Gerschman等，1954 年）。
> 「自由基生物学」的认可非常缓慢，可能是因为其起步阶段大部分是理论和假设性的，自由基的瞬息即逝的特性，以及缺乏研究它们的实验工具。
1968 年的一个重要的认识是，生物系统可以通过正常代谢途径产生大量超氧自由基，即 O2 补充，而酶类——超氧化物歧化酶（SOD），似乎演化出来的目的就是保护有氧生物免受这种自由基的潜在毒性（McCord 和 Fridovich，1968年）。
这些用于产生（通过黄嘌呤氧化酶）和消除超氧（通过 SOD）的酶类工具，促进了生理学和病理学等领域的进一步研究。
For several decades free radical biology has been “superoxide-centric”, owing largely, perhaps, to the fact that superoxide is quantitatively the predominant free radical produced by biological systems. An example of a biologically-important free radical process that does not necessarily involve superoxide is lipid peroxidation, propagated by the characteristic “free radical chain reaction”. Oxidative stress, however, is a broader term than free radical biology, as few oxidants are actually free radicals. The superoxide radical, in fact, is a fairly good reducing agent in addition to being a mild oxidizing agent. In the dismutation reaction one superoxide radical acts as an oxidant, the other acts as a reductant. As the term “oxidative stress” came into broad usage in the 1970s it frequently described imbalances in redox couples such as reduced to oxidized glutathione (GSH/GSSG) or NADPH/NADP+ ratios. Such metabolic disturbances need not involve the overproduction of reactive free radicals at all. Thus, the terms “oxidative stress” and “free radical damage” are not synonymous and may not always be interchangeable. Similarly, the terms “free radicals” and “reactive oxygen species” (ROS) are also not synonymous, as many reactive oxygen species (singlet oxygen, hydrogen peroxide, peroxynitrite) are not free radicals.
几十年来，自由基生物学一直以「超氧根为中心」，这可能主要是因为超氧根在生物系统中是定量优势的自由基。一个在生物学上重要的自由基过程的例子，并不一定涉及超氧根，而是被特征性的「自由基链反应」所推动的脂质过氧化反应。
然而，氧化应激这个术语比自由基生物学更广泛，因为**实际上很少有氧化剂是自由基**。事实上，超氧根除了是一种温和的氧化剂外，它也是一种相当好的还原剂。在去螯合反应中，一个超氧根作为氧化剂，另一个作为还原剂。
在 20 世纪 70 年代，「氧化应激」这个术语开始广泛流行，它经常用来描述氧化还原对偶体之间的不平衡，例如还原型到氧化型谷胱甘肽（GSH/GSSG）或 NADPH/NADP+ 的比率。
这种代谢紊乱并不一定完全涉及过量产生反应性自由基。因此，「氧化应激」和「自由基损伤」这两个术语并不是同义词，也不一定总能互换使用。
同样，术语「自由基」和「活性氧物质（ROS）」也不是同义词，因为许多活性氧物质（单线态氧、过氧化氢、过氮酸根）并不是自由基。
---
### 1\.2. Oxidative stress, inflammation, reperfusion injury, fibrosis, and cancer 1.2. 氧化应激、炎症、再灌注损伤、纤维化和癌症
Even before the discovery of SOD’s enzymatic ability to scavenge the superoxide radical (McCord et al., 1969), it was recognized that the protein (also known then as Orgotein, Ontosein, and Palosein) had substantial anti-inflammatory activity (Cushing et al., 1973, Huber et al., 1968, Marberger et al., 1975, McGinness et al., 1977). A search for “orgotein” in PubMed returns well over 100 publications from the past 40+ years, describing many veterinary and human clinical trials. The biochemical connection between superoxide and the inflammatory process followed soon after the discovery of SOD activity. Bernard Babior reported in 1973 that phagocytosing polymorphonuclear leukocytes produced significant amounts of superoxide radical (Babior et al., 1973). It was quickly shown that the depolymerization of hyaluronic acid, as an example of molecular damage resulting from the inflammatory process, was indeed due to this ability of activated leukocytes to produce superoxide radical and to cause oxidative stress (McCord, 1974). The ability of SOD to prevent various sorts of oxidative stress-associated damage resulting from the inflammatory process supported the earlier empirical evidence that SOD appeared to be useful as an anti-inflammatory therapy, but also began to reveal that superoxide’s role in the inflammatory process is rather complex, serving constructive as well as destructive roles (McCord et al., 1980, Petrone et al., 1980, Salin and McCord, 1975). By the 1980s it became apparent that superoxide was involved in pathophysiological conditions beyond the inflammatory process, such as post-ischemic reperfusion injury (Granger et al., 1981, McCord, 1985), even though in vivo reperfusion injury ultimately involves inflammation as well. Overexpression of SOD2 was even found to suppress the malignant phenotype of human melanoma cells (Church et al., 1993). Furthermore, the clinical work with SOD as “orgotein” suggested that the protein may be anti-fibrotic in some applications (Ludwig, 1991, Sanchiz et al., 1996), in addition to being anti-inflammatory. Thus, for more than four decades research has suggested that superoxide-dependent oxidative stress may be involved in the pathophysiology of inflammation, fibrosis, cancer, and reperfusion injury.
即使在发现 SOD 酶能清除超氧自由基的能力之前（McCord 等，1969年），人们就已经认识到这种蛋白质（当时也称为 Orgotein、Ontosein 和 Palosein）具有显著的抗炎活性（Cushing 等，1973年；Huber 等，1968年；Marberger 等，1975年；McGinness 等，1977年）。
「orgotein」在 PubMed上 的搜索结果返回了过去 40 多年里 100 多篇发表的文章，描述了许多兽医和人类临床试验。
在 SOD 活性的发现之后不久，超氧和炎症过程之间的生化联系也随之揭示出来。Bernard Babior 在 1973 年报告说，吞噬性多核白细胞产生了大量超氧自由基（Babior 等，1973年）。很快就证明，透明质酸的降解，作为炎症过程导致的分子损伤的一个例子，确实是由激活的白细胞产生超氧自由基和引起氧化应激的能力所致（McCord，1974年）。
SOD 防止与炎症过程相关的各种氧化应激损伤的能力，支持了早期的经验证据，即 SOD 似乎有用作为抗炎疗法，但也开始揭示超氧在炎症过程中的作用相当复杂，既起到建设性的作用，又起到破坏性的作用（McCord 等，1980年；Petrone 等，1980年；Salin 和 McCord，1975年）。到了 20 世纪 80 年代，人们发现超氧在炎症过程以外的病理生理条件中也起着作用，比如术后缺血再灌注损伤（Granger 等，1981年；McCord，1985年），尽管体内再灌注损伤最终也涉及炎症。
> **SOD2 的过度表达甚至被发现抑制人类黑色素瘤细胞的恶性表型（Church 等，1993年）。**
此外，关于 SOD 作为「orgotein」的临床工作还表明，该蛋白质在某些应用中可能具有抗纤维化作用（Ludwig，1991年；Sanchiz 等，1996年），除了具有抗炎作用。
> **因此，四十多年的研究表明，超氧依赖性氧化应激可能参与了炎症、纤维化、癌症和再灌注损伤的病理生理过程。**
---
### 1\.3. SOD as a drug? 1.3. SOD作为一种药物？
Much effort has been expended in the past several decades in attempts to turn SOD into a drug. Proteins and enzymes generally make very poor drugs for a variety of reasons: possible immunogenicity, high cost of production, problems associated with purification and stability, non-availability by oral administration, and poor pharmacokinetic properties. The SODs studied all suffered from these limitations—some more than others. In an effort to create an SOD with a better set of properties than any of the three human gene products, we created by genetic engineering techniques a chimeric SOD that combined the structure of the mature human mitochondrial SOD (SOD2) with the “sticky” polycationic C-terminal “tail” of the human extracellular SOD (SOD3), naming the chimeric recombinant product “SOD2/3” (Gao et al., 2003). This chimeric form of SOD did indeed possess pharmacological properties that were superior to any of the three naturally occurring forms of the human enzyme. Being smaller than SOD3 and nearly neutral in charge, it extravasated more easily into the tissue spaces. By having the ability to bind to cell surfaces, it displayed greater efficacy at lower doses than either SOD1 or SOD2, as well as much slower renal clearance than either. This ability to bind to cell surfaces and components of the extracellular matrix also seemed to buffer the effective concentration in vivo. While the superiority of SOD2/3 was subsequently demonstrated in a variety of models (for a review, see Hernandez-Saavedra et al., 2005), many of the problems enumerated above remained. Was there a better solution for reducing levels of oxidative stress? More importantly, if the effective treatment of complex pathological conditions is the goal, perhaps a broader approach is called for than the administration of a single gene product such as SOD.
过去几十年中，人们在将 SOD 转化为药物方面付出了很多努力。由于可能产生免疫原性、生产成本高、纯化和稳定性问题、无法通过口服获得以及药代动力学性能差等种种原因，蛋白质和酶通常作为药物表现不佳。研究中发现，所有的 SOD 都受到了这些限制，其中一些限制更为严重。
为了创造出一种比人类三个基因产物都具有更好性能的 SOD，我们采用基因工程技术创建了一个嵌合 SOD，它结合了成熟人类线粒体 SOD（SOD2）的结构和人类细胞外 SOD（SOD3）的「粘性」多阳离子 C-末端「尾巴」，并将这种嵌合重组产物命名为「SOD2/3（Gao et al., 2003）」。
这种嵌合形式的 SOD 确实具有比人体内任何一种天然 SOD 都优越的药理特性。它比 SOD3 更小，几乎不带电荷，更容易从血管外渗透进入组织空间。通过能够与细胞表面结合，它在较低剂量下显示出比 SOD1 和 SOD2 更高的疗效，并且排泄速度也比它们慢得多。这种与细胞表面和细胞外基质成分结合的能力似乎还可以在体内缓冲有效浓度。
虽然 SOD2/3 的优越性在多种模型中得到了证明（参见 Hernandez-Saavedra et al., 2005 的综述），但上述许多问题仍然存在。减少氧化应激水平的方法是否有更好的解决方案？更重要的是，如果目标是有效治疗复杂病理条件，也许需要采用比单一基因产物（如 SOD）更宽泛的方法。
---
### 1\.4. The Keap1/Nrf2 pathway 1.4 Keap1/Nrf2 通路
In 1994 a transcription factor was identified as a regulator of expression of the beta-globin genes, and was named Nrf2 (Moi et al., 1994). Soon it was discovered that Nrf2 is a positive regulator of the human Antioxidant Response Element (ARE) that drives expression of antioxidant enzymes such as NAD(P)H:quinone oxidoreductase 1 (NQO1) (Venugopal and Jaiswal, 1996). The mechanism of Nrf2 activation was described by Itoh et al., to involve a protein they named Keap1, a suppressor protein anchored in the cytoplasm that physically binds Nrf2, preventing its translocation to the nucleus and its access to ARE-containing promoters (Itoh et al., 1999b). What followed was a flurry of discoveries of additional Nrf2-regulated genes, including antioxidant-related genes such as those involved in glutathione synthesis (Wild et al., 1999), Phase II detoxification or “stress-response” genes (Itoh et al., 1999a), genes involved in limiting the inflammatory process (Itoh et al., 2004), genes involved in limiting pulmonary fibrosis (Kikuchi et al., 2010), and genes conferring protection against ischemia/reperfusion injury (Cao et al., 2006). Thus, the same spectrum of pathophysiological processes that had been found to be favorably modulated by attempts to use SOD as a drug was now also found to be favorably modulated by Nrf2 activation. Is this purely coincidental? Probably it is not. Rather, it seems likely that the process of evolution has assigned large numbers of genes, the products of which are required for survival in stressful conditions, to a common control mechanism—the Nrf2 pathway. It is reassuring to think that our cells have evolved the resources necessary to extricate themselves from many dire circumstances, in effect by making their own “medicines”. Perhaps all we need to do is assist with the signaling process—to help the cells perfect the timing and degree of Nrf2 activation. The idea becomes especially attractive in view of the fact that Nrf2 expression appears to decline with aging, leading to dysregulation of oxidative stress responses (Tomobe et al., 2011, Ungvari et al., 2011). Why our antioxidant defense system appears to abandon us as we age is not clear, but one possibility is that a programmed decrease in Nrf2 expression is Nature’s way of eliminating the drain on resources imposed on the species by old, post-reproductive individuals.
1994 年，一种转录因子被确认为 β-珠蛋白基因表达的调节因子，并被命名为 NRF2（Moi 等，1994 年）。很快发现，NRF2 是人类抗氧化应答元件（ARE）的正调节因子，推动抗氧化酶（如NAD（P）H：酮醌氧还蛋白1）表达（Venugopal 和 Jaiswal，1996年）。
Itoh 等人描述了 NRF2 激活的机制，涉及一种被他们命名为 Keap1 的蛋白质，它是一个位于细胞质固定的抑制蛋白，与 NRF2 物理结合，阻止其转位到细胞核并进入含有 ARE 的启动子（Itoh 等人，1999b 年）。接下来，一系列 NRF2 调控的基因的发现如雨后春笋般出现，包括与抗氧化相关的基因，如参与谷胱甘肽合成的基因（Wild 等，1999年）、第二期解毒或「应激反应」基因（Itoh 等，1999a 年）、限制炎症过程的基因（Itoh 等，2004年）、限制肺纤维化的基因（Kikuchi 等，2010年）以及提供对缺血/再灌注损伤保护的基因（Cao 等，2006年）。
> **因此，尝试将 SOD 用作药物时发现的优化的病理生理过程谱系，现在也发现其可以通过激活NRF2 来进行优化。**
这纯属巧合吗？很可能不是。相反，似乎进化过程已将大量基因分配给一个共同的控制机制—— NRF2 途径，它们的产物在应对压力条件下是生存所必需的。想到我们的细胞已经进化出了解救自身于困境的必要资源，实际上是在制造它们自己的「药物」，这是令人放心的。也许我们需要做的就是协助信号传导过程，帮助细胞完善 NRF2 激活的时机和程度。考虑到 NRF2 表达似乎随着年龄增长而下降，导致氧化应激反应的失调（Tomobe 等，2011年；Ungvari 等，2011年），这个想法尤为有吸引力。
为什么我们的抗氧化防御系统在衰老时似乎放弃了我们还不清楚，但一个可能性是，NRF2 表达的程序性降低是大自然为了减少旧的、已过生殖期的个体对种群资源的消耗而采取的措施。
---
### 1\.5. Eliminating oxidative stress by Nrf2 activation 1.5. 通过激活Nrf2来消除氧化应激
It seems entirely plausible that cells possess all the genetic resources required to maintain proper oxidative balance, as young healthy individuals seem not to be oxidatively stressed. It seems unlikely that the condition we describe as “oxidative stress” brings forth new types of oxidizing molecules, heretofore unseen, against which our cells have evolved no specific antioxidant defenses. Rather, it seems more likely that oxidative stress merely reflects an imbalance between the quantities of oxidants our cells are producing and the quantities of antioxidant gene products (SOD, catalase, GSH peroxidases, etc.) required to restore balance. Instead of attempting to restore oxidative balance by the administration of relatively tiny amounts of one antioxidant enzyme or another (e.g. SOD), perhaps our attention should be directed at Nrf2 activation, which can modulate the expression levels of hundreds of gene products that can affect oxidative stress and the related pathophysiological states. In a number of clinical trials in osteoarthritis, the intra-articular injection of about 50,000 U of SOD has been seen to be efficacious (McIlwain et al., 1989); in a recent clinical trial of Protandim (a composition of multiple synergistic phytochemical Nrf2 activators) the average individual showed an increase of erythrocyte SOD of 34%. As the entire human body contains roughly 7 g of SOD, this 34% increase, if seen in all organs, would result in a steady-state increase of more than 6,000,000 U of SOD activity distributed throughout the body (Nelson et al., 2006). Thus, the Nrf2-induced increase produced more than 100 times the amount of SOD activity provided by a 15 mg injection of the purified enzyme. This, coupled with the fact that hundreds of other so-called “survival” genes are modulated by Nrf2 (in addition to SOD1), makes Nrf2 activation appear to be a very attractive alternative to the use of antioxidant enzymes, or of synthetic mimetics of antioxidant enzymes, or of natural or synthetic molecules touted to be “antioxidants” by virtue of their abilities to react stoichiometrically with oxidants or free radicals.
似乎完全可以認為，細胞擁有所有必要的遺傳資源，以保持正確的氧化平衡，因為年輕健康的個體似乎沒有受到氧化壓力。我們所描述的「氧化壓力」條件不太可能產生我們之前未見過的新型氧化分子，而我們的細胞對此並未進化出特定的抗氧化防禦。
相反，**氧化壓力更有可能僅僅反映出我們的細胞產生的氧化劑數量與恢復平衡所需的抗氧化基因產物（如 SOD、過氧化氫酶、GSH 過氧化物酶等）之間的不平衡**。而不是試圖通過給予相對微小量的某一或另一種抗氧化酶（例如 SOD）來恢復氧化平衡，也許我們應該專注於啟動 NRF2，它可以調節影響氧化壓力和相關的病理生理狀態的數百種基因產物的表達水平。
在多次骨關節炎的臨床試驗中，約 50,000 U 的 SOD 的關節腔內注射已被證明是有效的（McIlwain 等人，1989年）；在最近的 Protandim（多種協同的植物化學 NRF2 活化劑組合）的臨床試驗中，平均每個人的紅細胞 SOD 增加了 34%。
由於整個人體大約含有 7 克的 SOD，如果所有器官都出現這 34% 的增加，那麼會有超過6,000,000 U 的 SOD 活性穩定分佈在整個身體中（Nelson 等人，2006 年）。因此，NRF2 誘導的增加比純化酶 15 mg 注射提供的 SOD 活性多出 100 倍以上。
再加上除了 SOD1 外，還有數百個所謂的「生存」基因受到 NRF2 的調節，這使得 NRF2 啟動似乎成為非常吸引人的替代選擇，取代使用抗氧化酶，或合成的抗氧化酶模擬劑，或被宣稱具有「抗氧化」特性的天然或合成分子，因為它們能與氧化劑或自由基進行化學反應。
---
### 1\.6. How is Nrf2 activated? 1.6. Nrf2 如何被激活？
The discovery of Keap1, a Nrf2-binding protein anchored to the cytoskeleton, revealed how the Keap1/Nrf2 complex functions as the cell’s “oxidative stress sensor” (Itoh et al., 1999b). Four particularly reactive cysteine residues were identified in Keap1 as the most likely candidates for being the direct sensors of oxidative stress (Dinkova-Kostova et al., 2002). The formation of adducts with electrophiles or their subsequent rearrangement to form protein disulfide linkages was suggested as the molecular basis for the cellular chemostat capable of regulating oxidative stress levels by modulating the production of ARE-regulated antioxidant enzymes. Soon, however, alternative mechanisms for Nrf2 activation were found, and they are dependent upon kinase pathways, including those of mitogen-activated protein kinases (MAPK) (Yu et al., 1999), phosphatidylinositol-3 kinase (Kang et al., 2002, Zheng et al., 2009), and atypical protein kinase(s) C (Numazawa et al., 2003), among others. Recent refinements to our understanding of Nrf2 activation suggest that the oxidant sensor function of Keap1 may be primarily to slow the ubiquitination and subsequent degradation of Nrf2 at higher levels of oxidative stress, such that more Nrf2 accumulates in the cell under these conditions. Nrf2 itself may contain an oxidant sensor that facilitates nuclear translocation, but that function remains poorly defined (Hu et al., 2010). The phosphorylation of Nrf2 at serine 40 appears to be an important event in the release of Nrf2 from Keap1 and the translocation of Nrf2 to the nucleus (Huang et al., 2002). Many early studies interpreted the action of Nrf2 activators to be mediated solely via adduct formation with, or by oxidation of, the reactive cysteine residues of Keap1, but it seems more likely that kinase signaling pathways are nearly always involved as well, with phosphorylation of Nrf2 ultimately responsible for most of its migration to the nucleus. The actions of sulforaphane and phenethyl isothiocyanate have recently been reviewed in this light (Cheung and Kong, 2010).
Keap1 的发现揭示了 Keap1/Nrf2 复合物作为细胞的「氧化应激传感器」的功能（Itoh 等，1999b）。在 Keap1 中确定了四个特别活泼的半胱氨酸残基，被认为是氧化应激的直接传感器的最有可能的候选者（Dinkova-Kostova 等，2002）。
通过与电泳亲合物的形成或其后续重排形成蛋白质二硫键连接，被认为是通过调节 ARE 调节型抗氧化酶的产生来调节氧化应激水平的细胞化学统计互联作用的分子基础。然而，很快就发现了 NRF2 激活的替代机制，它们依赖于激酶途径，包括丝裂原活化蛋白激酶（MAPK）的途径（Yu 等，1999），磷脂酰肌醇-3激酶（Kang 等，2002，Zheng 等，2009），和非典型的蛋白激酶 C（Numazawa 等，2003），以及其他途径。对 NRF2 激活理解的最新改进表明，Keap1 的氧化应激传感器功能可能主要是在较高水平的氧化应激下减缓 NRF2 的泛素化和随后的降解，以便更多的 NRF2 在这些条件下积累于细胞中。
NRF2 本身可能含有一个氧化应激传感器，有助于核内转运，但该功能尚未明确定义（Hu 等，2010）。NRF2 在丝氨酸 40 位点的磷酸化似乎是 NRF2 从 Keap1 释放和转运至细胞核的重要事件（Huang 等，2002）。许多早期研究将 NRF2 活化剂的作用解释为仅通过与 Keap1 的活泼半胱氨酸残基形成互联物，或氧化这些残基，但更可能的是激酶信号通路几乎始终参与其中，其中NRF2 的磷酸化最终负责其大部分迁移到细胞核。最近以此角度对芥末硫素和苯乙烯异硫氰酸酯的作用进行了回顾（Cheung 和 Kong，2010）。
Literally dozens of compounds have been reported to have some ability to activate Nrf2, at least in cell culture experiments. Quantitative comparison of these compounds is nearly impossible, as there is no “standard” system in which such evaluations are made. Observed fold induction of an ARE-driven gene depends on a long list of variables, including the structure and origin of the ARE-containing promoter, the type of cell expressing the reporter gene, the concentration of the inducer, the composition of the culture medium used, the basal level of Nrf2 activation, and many other parameters. Often it is implied that an observed induction in vitro means that this Nrf2 activator may be useful in vivo, when the concentration tested in vitro may be impossible to achieve pharmacologically due to poor absorption, lack of bioavailability, rapid metabolism and clearance, etc.
据报道，至少在细胞培养实验中有几十种化合物被发现具有激活 NRF2 的能力。由于缺乏「标准」的评估系统，这些化合物的定量比较几乎是不可能的。
观察到的 ARE 驱动基因的倍增取决于许多变量，包括含有 ARE 的启动子的结构和来源、表达报告基因的细胞类型、诱导剂浓度、使用的培养基组成、NRF2 活化的基础水平以及许多其他参数。经常会暗示在体外观察到的诱导作用意味着这种 NRF2 激活剂在体内可能有用，但是由于吸收差、生物利用度低、快速代谢和清除等原因，体外测试的浓度在药理学上无法实现。
---
### 1\.7. Nrf2 activators as potential therapies for oxidative stress, inflammation, and chemoprevention 1.7. Nrf2激活剂作为氧化应激、炎症和化学预防的潜在治疗方法
Many Nrf2 activators are naturally-occurring and plant-derived, but many others are synthetic compounds not found in Nature. Several Nrf2 activators have progressed to animal experiments and even to human clinical trials. Among the more interesting is bardoxolone methyl \[or methyl 2-cyano-3,12-dioxooleana-1,9(11)dien-28-oate\] (Reata Pharmaceuticals), currently in Phase 2/3 clinical trials. It was recently reported that a clinical trial of bardoxolone methyl in patients with moderate chronic kidney disease found significant and sustained improvements in estimated glomerular filtration rate with parallel improvements in other measures of kidney function, in a 52-week study (Pergola et al., 2011). BG-12 (dimethyl fumarate) (Biogen Idec) is in clinical trials for the treatment or relapsing-remitting multiple sclerosis (Kappos et al., 2008). Protandim® (LifeVantage Corp.) is a patented dietary supplement consisting of five low-dose natural Nrf2 activators that achieves its effect through a 9-fold synergy obtained when all five components are present together (Velmurugan et al., 2009). The mechanism of Nrf2 activation was concluded to be through multiple kinase pathways, including PI-3 kinase, p38 MAPK, and PKCdelta. A study in humans with oral administration showed significant elevations in SOD1 and catalase, with a decrease in plasma markers of lipid peroxidation (Nelson et al., 2006). Protandim induced Nrf2 and HO-1 in a rat model of SU5416/hypoxia-induced pulmonary hypertension, reducing oxidative stress and cardiac fibrosis, preserving right ventricular microcirculation, maintaining right heart function, and reducing expression of osteopontin-1 (Bogaard et al., 2009). Osteopontin-1, a marker of fibrosis, was also decreased by oral Protandim supplementation in mdx mice, a model of Duchenne muscular dystrophy (Qureshi et al., 2010), a disease where fatal heart and diaphragm fibrosis are thought to be regulated by osetopontin (Vetrone et al., 2009).
许多 NRF2 活化剂是自然存在且植物来源的，但还有许多人工合成的化合物在自然界中没有发现。
几种 NRF2 活化剂已经进行了动物实验甚至人体临床试验。其中更有趣的是巴多索隆甲酯（或甲基2-氰基-3,12-二氧代熊果酸酯）（Reata 制药），目前正进行第 2/3 阶段的临床试验。
最近有报道称，在患有中度慢性肾脏疾病的患者中，巴多索隆甲酯的临床试验发现估计的肾小球滤过率明显和持续改善，并且肾脏功能的其他指标也得到改善，在一项为期 52 周的研究中（Pergola 等，2011）。BG-12（二甲基富马酸酯）（Biogen Idec）正在进行治疗复发缓解型多发性硬化症的临床试验（Kappos 等，2008）。
Protandim®（LifeVantage Corp.）是一种专利的膳食补充剂，由五种低剂量的天然 NRF2 活化剂组成，当这五种成分同时存在时，通过 9 倍的协同作用达到其效果（Velmurugan 等，2009）。研究得出结论，NRF2 活化的机制包括多个激酶途径，包括 PI-3 激酶、p38 MAPK 和PKCdelta。
> 给人体口服后的研究显示，SOD1 和过氧化氢酶显著升高，而血浆脂质过氧化的标志物减少（Nelson 等，2006）。
在 SU5416/缺氧诱导的肺动脉高压大鼠模型中，Protandim 诱导 Nrf2 和 HO-1，减少氧化应激和心脏纤维化，保护右心室微循环，维持右心功能，并减少骨牌状蛋白-1 的表达（Bogaard 等，2009）。在杜氏肌营养不良（mdx）小鼠模型中，口服 Protandim 补充剂也降低了成纤维标志物骨牌状蛋白-1，这是一种致命性心脏和膈肌纤维化的模型（Qureshi 等，2010），这种疾病被认为由骨牌状蛋白调节（Vetrone 等，2009）。
A substantial literature documents the chemopreventive effect of Nrf2 activators, particularly those that are naturally occurring (such as sulforaphane and curcumin) and found in foods (Giudice and Montella, 2006, Surh et al., 2008). In a two-stage mouse skin carcinogenesis model, a Protandim-supplemented diet was found to reduce skin tumor incidence and multiplicity by 33% and 57%, respectively, compared to mice on basal diet (Liu et al., 2009). Suppression of p53 and induction of mitochondrial SOD are thought to play an important role in the tumor suppressive activity of Protandim (Robbins et al., 2010).
> 大量文献记载了 NRF2 活化剂的抗肿瘤预防效果，尤其是一些天然存在于食物中的活性物质（如芥子硫酸酯和姜黄素）（Giudice 和 Montella，2006 年；Surh 等，2008 年）。
在一个两阶段小鼠皮肤致癌模型中，与基础饮食的小鼠相比，摄取 Protandim 补充饮食被发现可以分别将皮肤肿瘤的发病率和多样性降低 33% 和 57%（Liu 等，2009年）。
抑制 p53 并诱导线粒体 SOD 被认为在 Protandim 的肿瘤抑制活性中起着重要作用（Robbins 等，2010年）。
---
## 2\. Materials and methods 2. 材料和方法
### 2\.1. Reagents 2.1. 试剂
Protandim was provided by LifeVantage Corp. (Salt Lake City, UT). d,l-Sulforaphane was purchased from Axxora LLC (San Diego, CA). d-Luciferin was from Gold Biotechnology (St. Louis, MO). Bardoxolone methyl (NSC 713200, also known as “RTA 402” and “CDDO-methyl ester”) was obtained from the NCI/DTP Open Chemical Repository (http://dtp.cancer.gov). Unless specified, all other chemicals were from Sigma–Aldrich (St. Louis, MO).
Protandim 由 LifeVantage Corp 提供（位于盐湖城，犹他州）。d,l-Sulforaphane 来自 Axxora LLC（位于圣地亚哥，加利福尼亚州）。d-Luciferin 来自 Gold Biotechnology（位于圣路易斯，密苏里州）。Bardoxolone methyl（NSC 713200，也被称为「RTA 402」和「CDDO-methyl ester」）来源于 NCI/DTP 开放化学品存储库（http://dtp.cancer.gov）。
除非另有指定，其他化学品均来自 Sigma-Aldrich（位于圣路易斯，密苏里州）。
---
### 2\.2. Bioassay for Nrf2-activation 2.2. Nrf2激活的生物测定
The assay is based on the AREc32 cell line, developed and generously provided by Dr. C.R. Wolf and colleagues of the University of Dundee, Scotland (Wang et al., 2006). The AREc32 cell line is a stable transfectant derived from the MCF7 human breast cancer cell line. It contains a promoter with eight copies of the rat glutathione-S-transferase-A2 Antioxidant Response Element (ARE) and the SV40 promoter sequence upstream of a firefly luciferase reporter gene. In these cells, luciferase activity is increased up to 50-fold following treatment with 50 μmol/L tert-butyl-hydroquinone. Luciferase activity is increased up to 100-fold by Protandim at 30 μg/ml, the most potent Nrf2 activator that we have observed.
该分析基于 AREc32 细胞系，该细胞系由苏格兰邓迪大学的 C.R. Wolf 博士和他的同事们开发并慷慨提供（Wang 等，2006）。AREc32 细胞系是从人类乳腺癌细胞系 MCF7 中稳定转染而来。它含有一个启动子，其中包含 8 个大鼠谷胱甘肽S-转移酶 A2 抗氧化应激响应元件（ARE）的拷贝和上游的 SV40 启动子序列，以及一个荧光素酶报告基因。
在这些细胞中，经过 50 μmol/L叔丁基对苯二酚处理后，荧光素酶活性增加了多达 50 倍。我们所观察到的最强 NRF2 激活剂Protandim在30 μg/ml 浓度下，可将荧光素酶活性提高多达 100 倍。
The AREc32 cells were grown in Opti-MEM (GIBCO, Carlsbad, CA) supplemented with 4% fetal bovine serum (FBS, GIBCO) and 1% Antibiotic–Antimycotic (GIBCO) at 37 °C and in a 10% CO2-supplemented air atmosphere. The cells were seeded at 1% to 5% of confluent cell density in T75 tissue culture flasks and cultivated until they approached confluence. The medium was aspirated off and the adherent cells trypsinized with 1 ml of 1× Trypsin–EDTA solution (GIBCO) for 10 min. Ten milliliters of medium was added to the flask and the cells transferred to a 50 ml centrifuge tube and centrifuged at 1000 rpm for 5 min at room temperature. Cells were washed once with 10 ml of medium, then resuspended in 10 ml of medium. Cells were counted using a hemocytometer and diluted to a concentration of 50,000 cells/ml. Four hundred microliters of this cell suspension were seeded into each well of a 24-well plate (i.e. 20,000 cells/well). Cells were then returned to the incubator for 24 h. After 18–24 h the cells were reattached and growing, and ready for treatment with putative Nrf2 activating agents. The agents were added to the wells in an appropriate concentration range, in volumes ranging from 1 to 10 μl/well. Vehicles used were aqueous or organic solvents such as ethanol or DMSO, and appropriate vehicle controls were included. The cells were then returned to the incubator for 18 h. All operations up to this point must be conducted under sterile conditions.
AREc32 细胞在 37℃ 和 10% CO2 辅助的空气气氛中，培养于 Opti-MEM 培养基（GIBCO, Carlsbad, CA）中，补充 4% 胎牛血清（FBS, GIBCO）和 1% 抗生素-抗菌素（GIBCO）。细胞密度为初始接触面的 1% 至 5%，在 T75 组织培养瓶中培养至细胞接近充满状态。移除培养基并用 1 ml 的1×胰酶-EDTA溶液（GIBCO）酶解吸附的细胞，处理时间为 10 分钟。向瓶中添加 10 ml 的培养基，将细胞转移至 50 ml 离心管中，在室温下以 1000 rpm 离心，离心 5 分钟。用 10 ml 培养基洗涤一次细胞，然后再悬浮于 10 ml 培养基中。
使用血细胞计数板计数细胞，并稀释至浓度为 50,000 个细胞/毫升。取 400 微升的这种细胞悬液分装入 24 孔板中的每个孔（即每孔 20,000 个细胞）。然后将细胞返回孵育箱中培养 24 小时。在 18 至 24 小时后，细胞重新附着并生长，可以开始使用潜在的 NRF2 激活剂进行处理。
激活剂被加入到孔中，浓度范围适当，并以 1 至 10 纳升/孔的体积加入。使用的载体是水性或有机溶剂，如乙醇或二甲基亚砜，同时包含适当的载体对照组。然后将细胞返回孵育箱中培养 18 小时。以上所有操作必须在无菌条件下进行。
After 24 h the cells were checked under the microscope for any abnormalities or detachment. The medium was aspirated and the cells were washed with phosphate buffered saline, pH 7.4 (100 μl/well). Following aspiration of the wash solution, the cells were lysed by application of 0.1 M potassium phosphate buffer, pH 7.8, containing 1% Triton X-100, 2 mM dithiothreitol, 2 mM EDTA, 10% glycerol and 3.5 mM sodium pyrophosphate (100 μl/well). The plate was incubated at 4 °C for 20 min. Lysate (20 μl from each well in a new 12 × 75 mm glass test tube) was assayed for luciferase activity using a Monolight 3010 autoinject luminometer (Analytical Luminescence Laboratory, Ann Arbor, MI), automatically injecting 50 μl of Luciferase Assay Buffer after background measurement. Luciferase Assay Buffer was prepared by mixing 9 ml of Solution A (15 mM Tricine, pH 7.8, containing 1.5 mM ATP, 7.5 mM MgSO4, and 5 mM dithiothreitol) with 1 ml of Solution B (10 mM d-luciferin). After a 4 s delay following injection, luminescence was measured for 10 s. Relative Light Units (RLU) were recorded for the contents of each well. Fold Induction of luciferase activity was calculated by dividing the RLU obtained for the test well by the average RLU obtained for control wells (which received no putative Nrf2 activator). Each assay or control was performed in duplicate. Parameters noted were the concentration of test substance providing maximal fold induction of luciferase (Cmax), and the maximal fold induction observed (FImax).
24 小时后，细胞在显微镜下检查是否有异常或脱离情况。将培养基吸出后，用磷酸盐缓冲盐水（pH 7.4，100 μl/孔）洗涤细胞。洗涤溶液吸出后，使用含有 1% Triton X-100、2 mM 二硫苏糖醇、2 mM EDTA、10% 甘油和 3.5 mM 焦磷酸钠的 0.1 M 磷酸钾缓冲液（pH 7.8，100 μl/孔）裂解细胞。板子在 4 °C 下孵育 20 分钟。取新的 12×75 毫米玻璃试管中每个孔的裂解液（20 μl），使用 Monolight 3010 自动进样发光仪（Analytical Luminescence Laboratory, Ann Arbor, MI）测定荧光素酶活性，注入 50 μl 的荧光素酶检测缓冲液进行背景测量后。荧光素酶检测缓冲液由 9 ml A溶液（含有15 mM Tricine，pH 7.8，1.5 mM ATP，7.5 mM MgSO4，5 mM二硫苏糖醇）和 1 ml B 溶液（含有 10 mM d-荧光素）混合制成。注射后延迟 4 秒，测量 10 秒的荧光发光。
记录每个孔的相对光单位（RLU）。通过将测试孔得到的 RLU 除以对照孔（未添加预计的 NRF2 激活剂）得到的平均 RLU，计算荧光素酶活性的折叠诱导。每个实验或对照均进行两次。记录的参数是提供最大荧光素酶折叠诱导的试验物质浓度（Cmax）和观察到的最大折叠诱导（FImax）。
---
### 2\.3. Gene expression experiments 2.3. 基因表达实验
Primary human umbilical vein endothelial cells (HUVEC) were obtained from Dr. Sonia Flores (University of Colorado Denver), and cultured to near confluency as two groups, Control and Protandim-treated. Protandim (as an extract of 200 mg/ml in 95% ethanol) was added to the growth medium of the treated group to produce a final concentration equivalent to 40 μg of Protandim per ml. Both groups were incubated for an additional 18 h.
人类脐静脉内皮细胞 (HUVEC) 从索尼娅·弗洛雷斯博士(科罗拉多丹佛大学)处获得，并被培养至接近充分密集的状态，分为两组，对照组和 Protandim 处理组。Protandim（以 95％ 乙醇中 200 毫克/毫升作为提取物）添加到处理组的生长培养基中，以使最终浓度相当于每毫升 40 微克的Protandim。两组都再培养 18 小时。
---
#### 2\.3.1. RNA preparation 2.3.1. RNA制备
For our experiments, the cell culture treatment groups were performed in triplicate, and each sample was used for an individual GeneChip array assay, resulting in three sets of gene expression data per treatment group. Total RNA was extracted from the cultured HUVEC cells (RNeasy Total RNA Isolation Kit, Qiagen, Valencia, CA), treated with DNase I, then the DNase was inactivated (DNA-free, Ambion, Austin, TX) and the sample purified further using RNeasy (Qiagen, Valencia, CA). The concentration of each sample was determined based on the absorbance at 260 nm (A260). The purity of each sample was determined based on the ratio of A260 to A280, and a range of 1.9–2.1 was considered adequately pure. The integrity of Total RNA samples was verified by Agilent 2100 Bioanalyzer.
对于我们的实验来说，细胞培养处理组在三个复位，每个样本都用于个别的 GeneChip 阵列检测，从而得到每个处理组三组基因表达数据。总 RNA 是从培养的 HUVEC 细胞中提取的（RNeasy 总 RNA 提取试剂盒，Qiagen，加利福尼亚州瓦伦西亚），经过 DNase I 处理，然后 DNase 被灭活（DNA-free，Ambion，德州奥斯汀），样本进一步经过 RNeasy纯化（Qiagen，加利福尼亚州瓦伦西亚）。每个样本的浓度是基于 260 nm（A260）的吸光度来确定的。
每个样本的纯度是基于 A260 和 A280 的比值来确定的，范围在 1.9-2.1 被认为足够纯净。通过 Agilent 2100 Bioanalyzer 验证了总 RNA 样本的完整性。
---
#### 2\.3.2. GeneChip analysis of gene expression 2.3.2. 基因芯片对基因表达的分析
Briefly, RNA samples were converted into double-stranded cDNA (ds-cDNA) using an oligodeoxythymidylic acid 24 primer with a T7 RNA polymerase promoter site added to the 3′ end (Superscript cDNA Synthesis System; Life Technologies, Inc., Rockville, MD). Double-stranded cDNA was purified using a GeneChip sample cleanup module (Affymetrix, Santa Clara, CA) and then used for in vitro transcription with an ENZO BioArray RNA transcript labeling kit (Enzo, Farmingdale, NY), transcribing the ds-cDNA template in the presence of a mixture of biotin-labeled ribonucleotides. Biotin-labeled cRNA was purified by affinity column (RNeasy, Qiagen, Valencia, CA) and randomly fragmented into 50–200 base cRNA fragments by incubation at 94 °C for 35 min in fragmentation buffer before hybridization to Affymetrix GeneChips Human 133 plus 2.0 arrays (45 °C, 16 h) using a GeneChip Hybridization Oven 640 (Affymetrix, Santa Clara, CA). The hybridized GeneChip microarray was stained with streptavidin–phycoerythrin using a GeneChip® Fluidics Station 450 (Affymetrix, Santa Clara, CA), and scanned at 2.5–3 μm resolution by GeneChip Scanner 3000 (Affymetrix, Santa Clara, CA).
简要来说，使用含有 T7 RNA 聚合酶启动子位点的寡聚脱氧胸腺苷酸 24 引物（Superscript cDNA合成系统；Life Technologies, Inc., Rockville, MD）将 RNA 样本转化为双链 cDNA(ds-cDNA)。采用 GeneChip 样品纯化模块 (Affymetrix, Santa Clara, CA) 纯化双链 cDNA, 然后使用 ENZO BioArray RNA 转录标记试剂盒 (Enzo, Farmingdale, NY) 进行体外转录，转录时在存在生物素标记核糖核苷酸混合物的情况下，将双链 cDNA 模板转录成生物素标记的 cRNA。
经过纯化（RNeasy, Qiagen, Valencia, CA）后，生物素标记的 cRNA 通过亲和柱提取，再通过在 94 °C 下的 35 分钟的反应光解制备成 50-200 个碱基的 cRNA 片段，然后在一个 GeneChip Hybridization Oven 640（Affymetrix, Santa Clara, CA）中与 Affymetrix GeneChips Human 133 plus 2.0 arrays 进行杂交（温度为 45 °C，时间为 16 小时）。
杂交后的 GeneChip 微阵列通过 GeneChip® Fluidics Station 450 (Affymetrix, Santa Clara, CA) 使用链霉亲和素-藻红蛋白染色，并通过 GeneChip Scanner 3000(Affymetrix, Santa Clara, CA)以2.5-3 μm 分辨率进行扫描。
---
#### 2\.3.3. Gene expression data analysis 2.3.3. 基因表达数据分析
Hybridization intensities were quantified and normalized across all arrays using the Robust Multichip Average (RMA) algorithm with an adjustment for guanine/cytosine content of probesets, available as an array processing tool on Partek Genomics Suite software 6.5 (St. Louis, MO) (Wu et al., 2004a). Data were filtered to remove all transcripts considered ‘absent’ or below “Detection Above BackGround (DABG)” in all samples, as determined by the Affymetrix GeneChip Operating Software (GCOS). Remaining transcripts (22,737 of 54,675) were used for all subsequent statistical and visual analysis. Partek Genomics Suite software was used to identify differentially expressed transcripts using a one-way ANOVA model with a stringent false discovery rate of less than 2.5% (corresponding to a p < 0.0033) to control for multiple testing. An arbitrary expression change cutoff of more than 1.5 was applied to generate a set of transcripts with differential expression between the experimental groups. These cutoff criteria resulted in discovery of 3000 gene transcripts that were significantly modulated by Protandim treatment in cultured HUVEC cells. For pathway analysis, we entered the gene probe identification numbers of the transcripts that met our cutoff criteria and corresponding fold change values into Ingenuity Pathway Analysis (IPA) software (Ingenuity Systems, Redwood City, CA), which facilitated the evaluation of our Protandim-modulated gene transcripts in the context of known, published biological pathways, functions, and networks.
杂交强度使用 Robust Multichip Average (RMA) 算法进行了量化和归一化处理，通过针对探针集的鸟嘌呤/胞嘧啶含量进行调整。该算法可在 Partek Genomics Suite 软件 6.5 版（位于 MO 州圣路易斯）中作为一种阵列处理工具使用（Wu 等，2004a）。
根据 Affymetrix GeneChip Operating Software (GCOS) 确定，数据在所有样本中被过滤以删除所有被认为是「缺失」或在「Detection Above BackGround (DABG)」以下的转录本。剩余的转录本（22,737/54,675）用于所有后续的统计和可视化分析。
使用 Partek Genomics Suite 软件使用一种一元方差分析模型和严格的虚假发现率小于 2.5%（对应p < 0.0033）进行差异表达转录本的识别，以控制多重检验。我们应用了 1.5 以上的任意表达变化值作为截断标准，以生成一组在实验组之间具有差异表达的转录本。
这些截断标准导致在培养的 HUVEC 细胞中，发现了 3000 个基因转录本受 Protandim 治疗显著调控。对于通路分析，我们将满足截断标准的基因探针鉴定号和相应的折叠变化值输入 Ingenuity Pathway Analysis (IPA) 软件（位于加利福尼亚州Redwood City），该软件有助于评估我们在已知的发表生物通路、功能和网络上调控的 Protandim 基因转录本。
---
## 3\. Results and discussion 3. 结果与讨论
### 3\.1. Quantitative comparison of Nrf2 activators using the AREc32 bioassay 3.1. 使用AREc32生物测定法对Nrf2活化剂进行定量比较
Fig. 1 provides a comparison using the AREc32-based bioassay for Nrf2 activation among Protandim, sulforaphane, bardoxolone methyl, and dimethyl fumarate. Sulforaphane is often considered a “gold standard” among naturally-occurring Nrf2 activators (Agyeman et al., 2011). As seen here, the two important parameters Cmax and FImax are easily observed. The greatest FImax was observed with Protandim at 135-fold, followed by bardoxolone methyl at 67-fold, dimethyl fumarate at 55-fold, and sulforaphane at 21-fold. Of the three pure compounds tested, bardoxolone methyl showed the lowest Cmax at 0.3 μM, with sulforaphane at 6 μM, and dimethyl fumarate 60 μM. Protandim, a mixture of five active ingredients, showed a Cmax of 48 μg/ml. This concentration of Protandim would contain approximately 26 μM silybinin, 13.6 μM curcumin, 5 μM EGCG, 0.07 μM withaferin A, and 5 μM bacopasides. Bardoxolone methyl appeared to produce a biphasic induction, producing near maximal FI over a range of concentrations from less than 40 nM to 0.4 μM.
图1使用基于 AREc32 的生物活性试验对 Protandim、芥子硫酸甲酯、巴多沙隆甲酯和富马酸二甲酯的Nrf2激活进行了比较。芥子硫酸甲酯通常被认为是天然产生的 NRF2 激活剂的“黄金标准”（Agyeman 等人，2011）。如图所示，两个重要参数 Cmax 和 FImax 可以轻松观察到。Protandim 呈现最大的 FImax 为 135 倍，其次是巴多沙隆甲酯为 67 倍，富马酸二甲酯为 55 倍，芥子硫酸甲酯为 21 倍。经过测试的三种纯化合物中，巴多沙隆甲酯的 Cmax 最低，为 0.3μM，芥子硫酸甲酯为 6μM，富马酸二甲酯为 60μM。Protandim 是由五种活性成分混合而成，其 Cmax 为 48μg/ml。这种浓度的 Protandim 大约含有 26μM 水飞蒲素、13.6μM 姜黄素、5μM 表绿素儿茶素、0.07μM 伴随视黄甙A和 5μM 叶子酸。巴多沙隆甲酯似乎产生了双相诱导，在浓度从低于 40nM 到 0.4μM 范围内产生接近最大FI的效果。
Fig. 1. Induction of luciferase in AREc32 cells by Protandim, bardoxolone methyl, dimethyl fumarate, and sulforaphane. Cells (20,000/well) were treated with the indicated concentrations of Nrf2 activator and incubated for 18 h. Maximum fold inductions (FImax) and the concentrations producing those maxima (Cmax) were as follows: Protandim, FImax = 135 at Cmax = 48 μg/ml; bardoxolone methyl, FImax = 67 at Cmax = 0.3 μM; dimethyl fumarate, FImax = 55 at Cmax = 60 μM; sulforaphane, FImax = 21 at Cmax = 6 μM.
图1. Protandim、bardoxolone methyl、dimethyl fumarate 和 sulforaphane在 AREc32 细胞中诱导荧光酶的作用。细胞（每孔2万个）被处理以所示的 NRF2 活化剂浓度，并在孵育 18 小时后。最大倍数诱导（FImax）和产生这些最大值的浓度（Cmax）如下：Protandim, FImax = 135，Cmax = 48 μg/ml；bardoxolone methyl, FImax = 67，Cmax = 0.3 μM；dimethyl fumarate, FImax = 55，Cmax = 60 μM；sulforaphane, FImax = 21，Cmax = 6 μM。
---
#### 3\.1.1. Problems associated with quantifying Nrf2 activators 3.1.1. 与定量Nrf2活化剂相关的问题
There is no universally accepted method for quantifying Nrf2 activation elicited by any given agent or for comparing potencies of agents that share this property. Sometimes the claim is based on microscopic evidence of nuclear translocation of Nrf2, detected by immunofluorescence. While important to demonstrate, this technique is qualitative, and does not actually reflect gene expression. More often, Western blot analysis is used to demonstrate an increase of a particular gene product, but this is rarely used to determine FImax and Cmax with any degree of accuracy. Moreover, the use of these techniques is reported in many different cell types, and not all cells respond equally to any given agent due to myriad biological variables. A more quantitative technique has been transient transfection of a cell line with a reporter gene controlled by an ARE-containing promoter. While this approach works well for a given study, similar experiments performed in different laboratories generally involve use of various expression vectors, relying on different promoters and different reporter genes, transfected with different efficiencies into various cell lines. The creation of stably transfected cell lines such as the AREc32 cell line used here (Wang et al., 2006) or a similar recently described cell line derived from the human keratinocyte HaCaT cell line (Natsch and Emter, 2008) provides opportunities for defining “standard assays” that may be performed under standardized conditions in any laboratory. These standard, economical, high-throughput assays will greatly facilitate comparisons among the growing number of putative Nrf2 activators, whether phytochemicals or synthetic pharmaceuticals.
目前还没有公认的方法来量化由任何给定物质引发的 NRF2 激活，也没有比较具有相同特性的物质的效能的一种普遍接受的方法。有时，基于通过免疫荧光检测到的 NRF2 细胞核转位的显微镜证据来进行声明。
尽管这是一个重要的证明手段，但这种技术是定性的，实际上并不能反映基因表达。更常见的是使用 Western 印迹分析来证明特定基因产物的增加，但这很少用于准确确定 FImax 和 Cmax。此外，这些技术的应用报道在许多不同的细胞类型中，并且并非所有细胞对任何给定物质的反应都相同，这是由于众多的生物变量。
一种更定量的技术是将一个包含 ARE 启动子的报告基因转染到细胞系中进行过渡性转染。尽管这种方法在特定研究中效果很好，但在不同实验室进行的类似实验通常涉及使用不同表达载体、依赖不同启动子和不同报告基因以及转染到不同细胞系的不同效率。在这里使用的 AREc32 细胞系（Wang 等，2006）或类似最近从人类角质形成细胞 HaCaT 细胞系中衍生出的细胞系（Natsch 和 Emter，2008）的建立为定义在任何实验室中可以在标准条件下执行的「标准检测」提供了机会。这些标准的、经济的、高通量的检测方法将极大地促进在越来越多的潜在 NRF2 激活剂（无论是植物化学物质还是合成药物）之间进行比较。
Protandim, sulforaphane, bardoxolone methyl, and dimethyl fumarate have all been tested in vivo in humans and are therefore of potential therapeutic interest. When compared contemporaneously in the AREc32-based assay, FImax observed was in the order Protandim > bardoxolone methyl > dimethyl fumarate > sulforaphane. A notable difference among these four agents is that Protandim consists of five active ingredients which interact with substantial synergy, whereas the other three are single compounds. The nature of the synergistic action between any two of the five active components is to increase FImax well beyond the sum of the two individual values and to substantially decrease Cmax for each. At the Cmax of Protandim, each of its five components is therefore well below that component’s individual Cmax and FImax, such that the induction caused by the composition is up to nine times the sum of the five component contributions (Velmurugan et al., 2009).
Protandim, 苦参素，巴多沙龙甲酯和富马酸二甲酯均在人体内进行活体实验，并因此具有潜在的治疗兴趣。当在 AREc32 基础测定中进行比较时，观察到的 FImax 顺序为 Protandim > 巴多沙龙甲酯 > 富马酸二甲酯 > 苦参素。
这四种药物中的一个显著差异在于 Protandim 由五种活性成分组成，这些成分之间存在着相当大的协同作用，而其他三种则是单一化合物。这五个活性成分之间的协同作用性质是使FImax远远超过两个个体数值之和，并且显著降低每个成分的 Cmax。在 Protandim 的 Cmax 下，因此其五个组分中的每一个都远低于该组分的个体 Cmax 和 FImax，导致由该组合引发的感应较五个组分贡献之和高出九倍（Velmurugan et al.，2009）。
The problems of variability in the bioassay of Nrf2 activators are not completely eliminated, even by the use of a stably transfected cell line such as AREc32. One reason is that fold induction is calculated by dividing the relative light units (RLU, representing luciferase concentration) in the presence of the inducer by the relative light units observed in the absence of the inducer, with the latter value representing the “basal” level of gene expression. Basal level is affected by growth medium composition—especially by concentration and source of the fetal bovine serum it contains. It is also affected by degree of confluency of the cells, and certainly by the type of cell. This basal level, however, may be a much smaller number than the induced number (as small as <1%) such that minor fluctuations in it have a great effect on the calculated fold induction. The variability can be largely eliminated if FImax is calculated relative to the contemporaneous standard. Using the data of Fig. 1, if sulforaphane is considered the standard, then the fold inductions relative to sulforaphane would be: Protandim, 6.4, bardoxolone methyl, 3.2, and dimethyl fumarate, 2.6. When these ratios are calculated, the basal levels of induction cancel out. Thus, the use of a stably transfected cell line such as AREc32, coupled with contemporaneous assessment of a well-characterized “standard” such as sulforaphane or tert-butylhydroquinone would seem to be a great improvement over currently used and often poorly controlled methods to assess Nrf2 activation.
NRF2 活化剂的生物检测中的变异性问题并没有完全消除，即使使用了一个稳定转染的细胞系，比如 AREc32。一个原因是通过将诱导剂存在下计算相对光单位（RLU，代表荧光素酶浓度）除以未加诱导剂时观察到的相对光单位来计算折叠诱导。后者表示基因表达的「基础」水平。
基础水平受到培养基组成的影响，尤其是胎牛血清的浓度和来源。它还受到细胞的密度以及细胞类型的影响。然而，这个基础水平可能比诱导水平小得多（小到 < 1%），因此其中的微小波动对计算的折叠诱导有很大影响。如果相对于同时期的标准计算FImax，变异性可以大部分消除。使用图1的数据，如果考虑辣根苗作为标准，那么相对于辣根苗的折叠诱导为：Protandim 为 6.4，bardoxolone 甲基为 3.2，dimethyl fumarate 为 2.6。当计算这些比率时，基础诱导水平被消除。因此，使用一个稳定转染的细胞系，如 AREc32，结合与已知良好的“标准”（如辣根苗或叔丁基对苯醌）同时评估，似乎是比当前使用的、常常控制不良的方法更大的改进，用来评估 NRF2 活化。
A recent laboratory study of dimethyl fumarate found that the compound activates Nrf2 in primary astrocytes, but not in the C6 glioma-derived cell line (Wilms et al., 2010), demonstrating that different cells may respond quite differently to Nrf2 activators. These authors suggest that the increased metabolic demands of transformed cells may play a role, but data showing a strong response of the transformed MCF7-derived AREc32 cell line argues against that rationale. Rather, it seems more likely that Nrf2 activators will not be found to be “one-size-fits-all” but instead may have to be selected based on the primary mechanism of activation (i.e., via thiol alkylation versus kinase pathway activation, and even which kinase pathway predominates) or on the particular biochemical and physiological idiosyncrasies of the cell type or organ being targeted.
最近的一项关于二甲基富马酸酯的实验室研究发现，该化合物会激活原代星形胶质细胞中的 NRF2，但不会激活 C6 胶质瘤诱导的细胞系（Wilms et al., 2010），这表明不同细胞对 NRF2 激活剂的反应可能会有相当大的差异。
这些作者认为，转化细胞的增强代谢需求可能起到一定作用，但显示转化的 MCF7 诱导的 AREc32 细胞系强烈反应的数据反驳了这种推理。
相反，似乎更有可能的是，NRF2 激活剂不会是「一刀切」的，而是可能需要根据激活的主要机制（即通过硫醇烷基化还是激酶信号通路的激活，甚至是哪种激酶信号通路占主导地位）或是基于所针对的细胞类型或器官的特定生化和生理特点来选择。
---
#### 3\.1.2. Why do Nrf2 activators display bell-shaped dose curves? 3.1.2. 为什么Nrf2激活剂呈现钟形剂量曲线？
A feature common to all Nrf2 activators examined here is bell-shaped dose–response curves. The reason for this behavior is not understood, but may reflect self-limitation imposed by the induction, at higher levels of Nrf2 activation, of enzymes that reverse the activation process, such as deacetylases. An additional layer of complexity has been added to the Nrf2 activation story with the demonstration that acetylation–deacetylation of Nrf2 determines its nuclear translocation, its ability to promote transcription, and its egress from the nucleus to terminate its transcriptional activity (Kawai et al., 2011). Sirtuin 1 (SIRT1) was shown to decrease acetylation of Nrf2, as well as Nrf2-dependent transcription. \[In our gene expression data, SIRT1 was induced 1.75-fold by Protandim (p = 0.015).\] The study also found that resveratrol, a putative activator of SIRT1 (Howitz et al., 2003), inhibited Nrf2-dependent transcription, apparently contradicting earlier reports that resveratrol activates Nrf2 (Chen et al., 2005, Ungvari et al., 2010). It may, however, do both, depending on concentration.
所有 NRF2 活化剂的一个共同特征是呈现钟形剂量-反应曲线。这种行为的原因尚不清楚，但可能反映了诱导作用的自我限制，在 NRF2 活化的较高水平上诱导反转活化过程的酶（如去乙酰酶）。
关于 NRF2 活化，还增加了一层复杂性，即乙酰化-脱乙酰化调控 NRF2 的核移位、促进转录和终止转录活性的核外迁 (Kawai et al., 2011) 研究还发现，Sirtuin 1 (SIRT1) 减少了 NRF2 的乙酰化，以及依赖于 NRF2 的转录。\[在我们的基因表达数据中，Protandim 诱导了 SIRT1 的 1.75 倍表达 (p = 0.015)。\]
这项研究还发现，白藜芦醇，作为 SIRT1 的潜在活化剂 (Howitz et al., 2003)，抑制了依赖于 NRF2 的转录，这显然与早期报道白藜芦醇激活 NRF2 的结果相矛盾 (Chen et al., 2005, Ungvari et al., 2010)。然而，它可能在不同浓度下既有激活 NRF2 的作用，也有抑制 NRF2 的作用。
---
### 3\.2. Gene expression data 3.2 基因表达数据
Among the 10 genes most highly upregulated by Protandim are a number of notables that encode antioxidant and anti-inflammatory proteins. SLC7A11, induced 76-fold, encodes a cystine/glutamate antiporter responsible for maintaining extracellular glutamate in the brain, and for importing cystine necessary for glutathione synthesis (Albrecht et al., 2010). This antiporter was recently found to be decreased by repeated cocaine exposure, and restoration of the activity prevented cocaine-primed drug seeking behavior in rats (Baker et al., 2003). AKR1B10, induced 72-fold, encodes aldo–keto reductase family 1 member B10 which efficiently detoxifies mutagenic and carcinogenic alpha, beta-unsaturated carbonyls such as 4-hydroxynonenal (Zhong et al., 2009). PTGR1 (aka LTB4DH), induced 68-fold, encodes leukotriene B4-12-hydroxydehydrogenase, which is considered to be a key enzyme responsible for biological inactivation of prostaglandins and related eicosanoids (Tai et al., 2002). It was recently found to suppress the oncogenic transformation of HepG2 cells (Wei et al., 2011). HMOX1, induced 56-fold, encodes heme oxygenase-1, an antioxidant enzyme considered a hallmark of Nrf2 activation. The induction of heme oxygenase-1 is now seen as a novel and alternative therapeutic target in the management of cardiovascular disease (Chan et al., 2011). AIFM2 (aka AMID), induced 29-fold, is implicated in caspase-independent apoptosis and was found to be downregulated in a majority of human tumors (Wu et al., 2004b). OSGIN1 (aka OKL38), induced 29-fold, is an oxidative stress response gene and a tumor suppressor gene (Yao et al., 2008). GPX3, induced 20-fold, encodes an important antioxidant enzyme, glutathione peroxidase-3, found normally in plasma and kidney but underexpressed in head and neck cancers (Chen et al., 2011). SQSTM1, induced 20-fold, encodes sequestosome-1, a participant in the autophagy pathway recently shown to be necessary to avoid premature senescence in human fibroblasts (Kang et al., 2011). HSPB8, induced 19-fold, is a heat shock protein that forms a complex with BAG3 (also induced 1.43-fold). Overexpression of the HSPB8-BAG3 complex also stimulates autophagy and facilitates the clearance of mutated aggregation-prone proteins, the accumulation of which characterizes many neurodegenerative disorders such as Alzheimer disease, Parkinson disease, and amyotrophic lateral sclerosis (Seidel et al., 2011). TNFSF9 (aka CD137L), induced 19-fold, can induce maturation of human immature monocyte-derived dendritic cells leading to an enhanced capacity of the dendritic cells to stimulate protective T cell responses, as compared to classical dendritic cells (Tang et al., 2011). It is interesting to note that even in this small sampling of the genes most highly upregulated by Protandim are genes that have been shown to be preventive or protective against cancer, cardiovascular disease, and neurodegenerative disease. These associations were further supported by pathway analysis.
在被 Protandim 高度上调的 10 个基因中，有一些值得注意的编码抗氧化和抗炎蛋白的基因。 SLC7A11 的表达增加了 76 倍，编码了一种负责维持脑部细胞外谷氨酸和进口半胱氨酸（谷胱甘肽合成所必需）的半胱氨酸/谷氨酸交换转运蛋白（Albrecht et al., 2010）。最近发现该抗转运蛋白在反复可卡因暴露后减少，而恢复其活性可以阻止大鼠的可卡因诱导的药物寻求行为（Baker et al., 2003）。
AKR1B10 的表达增加了 72 倍，编码了能有效解毒致突变和致癌的 α, β-不饱和羰基物质如4-羟基壬醛的醛酮还原酶家族 1 成员 B10（Zhong et al., 2009）。PTGR1（又称 LTB4DH）的表达增加了 68 倍，编码了白三烯 B4-12-羟基脱氢酶，被认为是生物去活化前列腺素和相关芳香族酸的关键酶（Tai et al., 2002）。最近发现该酶可以抑制 HepG2 细胞的致癌转化（Wei et al., 2011）。HMOX1 的表达增加了 56 倍，编码了血红素加氧酶-1,这是一种抗氧化酶,被认为是 NRF2 活化的标志。血红素加氧酶-1的诱导现在被视为管理心血管疾病的一种新颖和可替代的治疗靶点（Chan et al., 2011）。
AIFM2（又称 AMID）的表达增加了 29 倍，与 caspase 非依赖性的细胞凋亡有关，同时在大多数人类肿瘤中表达下调（Wu et al., 2004b）。OSGIN1（又称OKL38）的表达增加了 29 倍，它是一个氧化应激反应基因和肿瘤抑制基因（Yao et al., 2008）。GPX3 的表达增加了 20 倍，编码了一种重要的抗氧化酶—谷胱甘肽过氧化物酶-3，正常情况下在血浆和肾脏中存在，但在头颈部癌症中表达减少（Chen et al., 2011）。
SQSTM1 的表达增加了 20 倍，编码了隔离蛋白 1，是自噬途径中的参与者，最近研究表明它对于避免人类成纤维细胞的过早衰老是必需的（Kang et al., 2011）。HSPB8 的表达增加了 19 倍，它是一种热休克蛋白，与 BAG3 形成复合物（也增加了1.43倍）。HSPB8-BAG3 复合物的过度表达也会刺激自噬并促进聚集易聚集的突变蛋白的清除，这些蛋白的累积特征是许多神经退行性疾病（如阿尔茨海默病、帕金森病和肌萎缩侧索硬化症）的特点（Seidel et al., 2011）。TNFSF9（又称 CD137L）的表达增加了 19 倍，它能够诱导人体未成熟单核细胞源性树突状细胞成熟，从而增强树突状细胞刺激保护性T细胞反应的能力，与传统的树突状细胞相比（Tang et al., 2011）。有趣的是，即使在被 Protandim 高度上调的这些基因中,也有一些被证明对于癌症、心血管疾病和神经退行性疾病具有预防或保护作用，这些关联结果得到了通路分析的进一步支持。
#### 3\.2.1. Genes associated with specific disease states by pathway analysis 3.2.1. 通过路径分析确定与特定疾病状态相关的基因
Ingenuity Pathway Analysis (IPA) was used to examine gene transcripts that were increased or decreased by Protandim in HUVEC cells. The analysis revealed that atherosclerosis, colon carcinoma, and Alzheimer disease are each characterized by a number of genes significantly modulated by Protandim (see Table 1). For example, 19 genes products have been associated with atherosclerosis and are up or down-regulated by Protandim. Of these 19 genes, the first 16 listed (84%) were regulated by Protandim in the opposing direction to that taken by the atherosclerosis disease process. The probable benefit of this effect of Protandim is further supported by the fact that of the 11 gene products currently being targeted by drug interventions (Table 1, in bold type), nine of them (Table 1, marked by asterisks) are modulated by Protandim in the same direction that is proposed to be beneficial and caused by the therapeutic intervention.
Ingenuity Pathway Analysis (IPA) 被用于检查 HUVEC 细胞中 Protandim 增加或减少的基因转录本。分析结果显示，动脉粥样硬化、结肠癌和阿尔茨海默病的特征基因数量均受到 Protandim 的显著调节（见表1）。例如，19 个基因产物与动脉粥样硬化有关，并受到 Protandim 的调节上下调。
在这 19 个基因中，前 16个（84%）被 Protandim 以与动脉粥样硬化疾病进程相反的方向调节。Protandim 引起的这种效应的可能益处进一步得到支持，因为 11 个目前正受到药物干预靶向的基因产物（表 1，用粗体表示）中，有 9 个（表 1，用星号标记）被 Protandim 以有益的方式调节，并由治疗干预引起。
Table 1. Gene symbols in bold indicate gene products targeted by specific drugs used in the clinical treatment of the disease (determined by Ingenuity IPA analysis as described under Section 2.3.3). Bolded gene symbols with an asterisk indicate genes for which Protandim treatment modulates gene expression in the same direction as achieved with the drug therapy. Within a disease section, for genes above the horizontal line Protandim opposes the change caused by the disease process, and for genes below the line Protandim changes the gene in the same direction as the disease process.
表1. 加粗的基因符号表示临床治疗该疾病所使用的特定药物所靶向的基因产物（根据 2.3.3 节中所述的 Ingenuity IPA 分析确定）。带有星号的加粗基因符号表示 Protandim 治疗与药物疗法在同一方向上调节基因表达的基因。在疾病部分内，对于位于水平线上方的基因，Protandim 与疾病过程引起的变化相反，而对于位于线下的基因，Protandim 与疾病过程的基因变化方向相同。
In colon carcinoma, IPA analysis revealed 28 genes associated with the disease that were also modulated by Protandim treatment. Of these, the first 25 listed (89%) were regulated by Protandim in the opposing direction to that taken by the colon carcinoma disease process. In addition, Protandim downregulated the one gene targeted by a chemotherapeutic drug, an antimetabolite inhibitor for that gene’s product, thymidylate synthetase.
> 在结直肠癌中，IPA 分析显示与该疾病相关的 28 个基因也受 Protandim 治疗的调节。
其中，列出的前 25 个基因（89%）的调节方向与结直肠癌疾病进程相反。此外，Protandim 还下调了一种化疗药物的靶向基因，即抗代谢物酶抑制剂，用于抑制该基因的产物嘧啶核苷酸合成酶。
In Alzheimer disease, 66 genes were identified that are also modulated by Protandim at the gene expression level. Of these 66 genes, the first 43 of them (65%) were regulated by Protandim in the opposing direction to that taken by the Alzheimer disease process. The beneficial effect of Protandim is further supported by the fact that of the 10 gene products currently targeted by drug therapies, eight of them are modulated by Protandim in the same direction that is proposed to be beneficial and caused by the drug.
> 在阿尔茨海默病中，发现了 66 个基因，这些基因在基因表达水平上也受 Protandim 的调控。
在这 66 个基因中，前 43 个基因（65%）的调控方向与阿尔茨海默病过程相反，受 Protandim 调控的基因呈现有益效果的事实得到了进一步支持。此外，目前有药物治疗靶向的 10 个基因产物中，有 8 个在调控方向上也受 Protandim 调控，这种调控据推测亦有益并由药物引起。
Notably, among the relatively small number of genes for which Protandim regulates in the same direction as caused by the disease processes, several are antioxidant genes that are upregulated by Protandim and reported to be upregulated in colon carcinoma and Alzheimer disease. A likely explanation for the increased expression of GLRX2 (glutaredoxin 2) and NQO1 (NAD(P)H dehydrogenase, quinone 1) in colon carcinoma and of GLRX (glutaredoxin), HMOX1 (heme oxygenase-1), NQO1, and SOD1 (superoxide dismutase 1) in Alzheimer is that it represents an adaptive attempt to partially compensate for the increased level of oxidative stress associated with these diseases. These antioxidant genes are also upregulated by Protandim, which would provide additional antioxidant protection beyond that achieved by the ROS-dependent induction of these enzymes in the diseased tissues.
值得注意的是，相对较少数量的基因中，Protandim 调控的方向与疾病过程引起的方向相同，其中有几个是抗氧化基因，被 Protandim 上调，并且据报道在结肠癌和阿尔茨海默病中也有上调。 GLRX2（谷胱氧还蛋白 2）和 NQO1（NAD（P）H脱氢酶，喹啉1）在结肠癌中以及 GLRX（谷胱氧还蛋白）、HMOX1（血红素氧合酶-1）、NQO1和SOD1（超氧化物歧化酶1）在阿尔茨海默病中的表达增加，可能是为了部分补偿与这些疾病相关的氧化应激水平的增加而进行的适应性尝试。
这些抗氧化基因也被 Protandim 上调，这将提供超过在患病组织中这些酶的依赖于 ROS 诱导的抗氧化保护。
#### 3\.2.2. Do different Nrf2 activators produce identical gene expression patterns? 3.2.2 不同的Nrf2激活剂是否会产生相同的基因表达模式？
While Protandim, bardoxolone methyl, BG-12, and sulforaphane all have been demonstrated to modify gene expression profiles by activation of Nrf2, they have not been compared side by side, in the same cell line, under identical conditions. It is nearly certain that none of them is exclusively an Nrf2 activator, so significant differences may exist among their gene expression profiles. These differences would reflect differences in activation of transcription factors other than Nrf2, and could produce additional positive effects or could be responsible for unwanted or adverse effects. A published report exists providing a comparison between gene expression profiles for Keap-1-null mice (which have constitutive and presumably pure Nrf2 activation) and wild-type mice treated with CDDO-Imidazole, a derivative similar to bardoxolone methyl (Yates et al., 2009). Indeed, significant differences in gene expression patterns were seen in livers of mice from these two groups, particularly with regard to genes involved in detoxification and lipid metabolism. A similar study has recently been published comparing sulforaphane modulated gene expression to Keap-1 knockdown in the non-malignant human breast epithelial cell line MCF10 (Agyeman et al., 2011). Similar patterns were observed by both microarray and proteomic analysis. Using the microarray data, only 14% of the genes modulated by sulforaphane were similarly modulated by Keap-1 knockdown, indicating that the majority of sulforaphane-regulated transcripts appear not to be regulated through the KEAP1/NRF2 pathway.
虽然 Protandim，bardoxolone methyl，BG-12 和 sulforaphane 都被证明通过激活 NRF2 来改变基因表达谱，但它们尚未在相同细胞系和相同条件下进行副比较。
几乎可以确定它们中没有一种是独特的 NRF2 活化剂，因此它们的基因表达谱可能存在显着差异。这些差异将反映出除 NRF2 以外的转录因子激活的差异，并可能产生其他积极效应，或者可能导致不需要或不良影响。
已经存在一份发表的报告，对 Keap-1 缺失小鼠（其具有构成型和可能是纯净的 NRF2 活化）和用类似于 bardoxolone methyl 的 CDDO-Imidazole 治疗的野生型小鼠的基因表达谱进行了比较（Yates 等，2009）。的确，在这两组小鼠的肝脏中观察到了基因表达模式的显着差异，特别是涉及解毒和脂质代谢的基因。最近发表了一项类似研究，比较了 sulforaphane 调节的基因表达与Keap-1基因沉默在非恶性人类乳腺上皮细胞系 MCF10 中的情况（Agyeman 等，2011）。使用微阵列数据，只有 14% 的由 sulforaphane 调节的基因也被 Keap-1 基因沉默调节，表明大多数 sulforaphane 调节的转录本似乎不是通过 KEAP1/NRF2 途径调节的。
### 3\.3. Prospects for human therapeutic applications of Nrf2 activators 3.3. Nrf2活化剂在人类治疗应用的前景
Results of bardoxolone methyl therapy in a Phase II human clinical trial for chronic kidney disease in type II diabetics were recently reported (Pergola et al., 2011). After 52 weeks, the estimated glomerular filtration rate in the 75 mg/day treatment group had increased by 10.5 ± 1.8 ml/min/1.73 m2 (p < 0.001), representing an increase of about 32% when compared to entry values. The study suggests that Nrf2 activation represents a viable new therapeutic approach for renal disease, as similar results are not achievable with currently available therapies.
巴多沙酮甲基治疗II型糖尿病慢性肾病的II期人体临床试验结果最近被报道（Pergola et al., 2011）。经过 52 周，每天75毫克的治疗组估计肾小球滤过率增加了 10.5 ± 1.8 毫升/分钟/1.73平方米（p<0.001），相比入组值增加了约 32 %。
> 该研究表明，NRF2 的激活代表了一种可行的新的肾脏疾病治疗方法，因为目前可用的治疗方法无法达到类似的效果。
Patients with relapsing-remitting multiple sclerosis treated with BG-12 for 24 weeks showed significantly fewer new gadolinium-enhancing lesions, with significantly reduced probability of their evolution to T1-hypointense lesions than patients treated with placebo (Macmanus et al., 2011). BG-12 treatment reduced the annualized relapse rate by 32% (Kappos et al., 2008). These studies suggest that Nrf2 activation may represent a promising new therapeutic approach for multiple sclerosis.
接受 BG-12 治疗 24 周的复发缓解型多发性硬化症患者显示出明显减少的新钆增强病变，与接受安慰剂治疗的患者相比，这些病变的发展为 T1 低信号病变的概率也显著降低（Macmanus 等，2011）。
BG-12 治疗使年发作速率降低了 32%（Kappos 等，2008）。这些研究表明，NRF2 激活可能是多发性硬化症的一种有希望的新治疗方法。
The early successes of these two experimental Nrf2-activating drugs in diseases where currently used therapies have largely failed inspire hope for the future. These Nrf2 activators may well spawn a new class of drugs to target the so-called “diseases of aging”, including cancer, cardiovascular diseases, inflammatory and autoimmune diseases, and neurodegenerative diseases.
这两种实验性的 NRF2 激活药物早期在传统治疗失败的疾病中取得的成功为未来带来了希望。
这些 NRF2 激活剂有可能衍生出一类针对所谓的「老年疾病」的新药，包括癌症、心血管疾病、炎症和自身免疫性疾病，以及神经退行性疾病。
---
## 4\. Concluding remarks 4. 总结观点
The fact that as many as 200 human diseases have been associated with increased levels of oxidative stress has always been puzzling. Oxidative stress, because it is tied to mitochondrial oxidation of foodstuff and the generation of the energy necessary to sustain life, occupies a place of central importance. Even though reactive oxygen species are capable of disrupting nearly any metabolic pathway through their attack on proteins, lipids, and nucleic acids, is it reasonable that exposure to reactive oxygen species alone could cause such a diversity of disease processes? That still may be the case, and such arguments are supported by the early experiments to use SOD as a drug. It actually appeared to work single-handedly in many clinical and laboratory applications. A more useful paradigm, however, may be to focus on Nrf2 as the regulator of several thousand genes, including, not coincidentally, the family of antioxidant enzymes. Thus, if the real initiator of a disease process is dysfunctional activation of Nrf2, oxidative stress would inevitably be a symptom associated with whatever else may result. That is to say, oxidative stress may indeed be associated with 200 diseases, and even contributory to all of them, but not necessarily causative in every case. The data in Table 1 seem to support this view. Of the 66 Protandim-regulated genes that are associated with Alzheimer disease, only five (SOD1, NQO1, HMOX1, GLRX, and TXN) appear to be in the antioxidant family. Protandim upregulated all five of them, but clearly there is more to the story than genes associated with oxidative stress. The focus on Nrf2 will not only broaden our view, it will provide practical solutions.
> 据有关研究表示，多达 200 种人类疾病与氧化应激水平的增加有关，这一事实一直令人困惑。
氧化应激由于与线粒体对食物的氧化以及产生维持生命所需能量密切相关，因此具有重要的地位。尽管反应性氧自由基通过对蛋白质、脂质和核酸的攻击，能够破坏几乎任何代谢通路，但仅仅接触反应性氧自由基是否足以引起如此多样的疾病过程？在许多临床和实验室应用中，它似乎单独有效。
> 这可能仍然是可能的，早期使用超氧化物歧化酶（SOD）作为药物的实验证据支持了这种观点。
然而，一个更有用的观点可能是将 NRF2 作为数千个基因的调节因子来关注，包括与抗氧化酶家族不谋而合的基因。
因此，如果一种疾病过程的真正启动因子是 NRF2 的功能异常激活，那么氧化应激必然是与任何可能的结果相关的症状。也就是说，氧化应激确实与 200 种疾病相关，甚至在其中所有疾病中起到了一定的贡献，但并非在每种情况下都是原因。
表 1 中的数据似乎支持这种观点。在与阿尔茨海默病相关的 66 个由 Protandim 调节的基因中，只有五个（SOD1、NQO1、HMOX1、GLRX和TXN）属于抗氧化酶家族。
Protandim 上调了其中的五个基因，但显然与氧化应激相关的基因并非完全是所有问题的答案。关注 NRF2 不仅会扩大我们的视野，还会提供实际的解决方案。
---
## Disclosure statement 披露声明
Dr. McCord is Chief Science Officer for LifeVantage Corp. (the manufacturer of Protandim, used in this study, and primary sponsor of the project). He holds equity in the Company and serves on its Board of Directors. Dr. Hybertson serves as a paid consultant to the Company, and holds equity. Dr. Gao holds equity in the Company.
McCord 博士是 LifeVantage Corp. 的首席科学官（该公司是本研究中使用的 Protandim 的制造商，也是该项目的主要赞助商）。他在公司拥有股权并在董事会任职。Hybertson 博士是该公司的有偿顾问，并持有股权。Gao 博士持有该公司的股权。
---
## Acknowledgements 致谢
The authors wish to thank Dr. Michael Edwards for his assistance in analyzing the gene array data. This work was supported in part by LifeVantage Corp.
作者希望感谢 Michael Edwards 博士在分析基因数组数据方面的帮助。本研究部分由LifeVantage Corp. 资助。
---
資料來源：<https://www.sciencedirect.com/science/article/pii/S0098299711000501?via%3Dihub>
Sign in with Wallet
Connect another wallet