Non‐alcoholic fatty liver disease (NAFLD) is one of the most common liver disorders worldwide. NAFLD is considered to be the hepatic component of metabolic syndrome, because its features are very similar to those of metabolic disorders, such as obesity, inflammation, insulin resistance and type 2 diabetes. It is clear that NAFLD and type 2 diabetes have a close relationship. However, the exact mechanisms underlying the pathogenesis and progression of NAFLD are still incompletely understood1-3. It is well known that the regulation of glucose and lipid metabolism in the liver is properly carried out by insulin through either direct or indirect mechanisms. Therefore, two possible hypotheses are proposed to explain their pathogenesis. As one possibility, insulin resistance and excessive fatty acids in the bloodstream lead to simple hepatic steatosis. The second hypothesis implicates oxidative stress, lipid peroxidation and mitochondrial dysfunction. Hepatic rho‐associated coiled‐coil‐containing kinase 1 has been introduced as a new possible pathogenesis of NAFLD, playing an important role in high‐fat diet‐induced hepatic lipogenesis through the inhibition of adenosine 5´‐monophosphate‐activated protein kinase3. This hypothesis occurs under the condition of a high‐fat diet. This approach and the data are highly useful and meaningful in view of the current diet therapy for patients with diabetes. Recently, a very attractive and interesting hypothesis of the pathogenesis of NAFLD was proposed from the viewpoint of excessive intake of fructose (sugar or corn syrup)2. This completely new theory has a close connection with the polyol pathway hyperactivity activated by uric acid.

It is well known that hyperglycemia‐induced polyol pathway hyperactivity can lead to the development of diabetic complications, such as microangiopathies, macroangiopathies and others4-7. Furthermore, it has been thoroughly confirmed by many studies that the inhibition of aldose reductase (AR), a key enzyme in this pathway, is useful to prevent these complications. The polyol pathway consists of just two steps: glucose is first reduced to sorbitol by AR, and the resulting sorbitol is then changed to fructose by sorbitol dehydrogenase (Figure 1). During normoglycemia, the use of glucose through the polyol pathway accounts for <3% of glucose consumption in cells. However, during hyperglycemia, the utilization of glucose through this pathway represents up to 30%, resulting in the progress of diabetic complications in target tissues7. The mechanisms of diabetic complications induced by hyperglycemia‐activated polyol pathway hyperactivity are not as simple as we expect. Generally, metabolic factors, such as protein kinase C, glycation, oxidative stress and others, are involved in the lower reaches of this pathway and contribute to the progress of diabetic complications with complex issues6, 7. Furthermore, the following factors in relation to polyol pathway hyperactivity might also be partially involved, depending on the types of diabetic complications: inflammation, endothelial nitric oxide synthase, thromboxane, matrix metalloproteinases, nicotineamide phosphoriboxyl transferase, nitric oxide, tissue factor, vascular cell adhesion molecule and the expression of multiple genes of the transforming growth factor‐β pathway6, 7.

Recently, Sanchez‐Lozada et al .2 proposed a very interesting and attractive hypothesis of the pathogenesis of the development of NAFLD through uric acid‐induced polyol pathway hyperactivity. The concept of that study is based on the fact that the two primary sweeteners, sugar (sucrose) and high fructose corn syrup, induce fatty liver in animals. Furthermore, previous studies by this group have shown that the pathogenesis of inducing fatty liver using fructose is due to the generation of uric acid in the course of fructose metabolism, resulting in mitochondrial oxidative stress and an impairment of adenosine triphosphate production. They also confirm that hyperuricemia itself is not only strongly related with hypertriglyceridemia and NAFLD, but also predicts the progression of NAFLD. Furthermore, they found that uric acid upregulated fructokinase/ketohexokinase (KHK) and fructose metabolism through the activation of the transcription factor, carbohydrate response element‐binding protein. Their recent study based on these facts evaluates whether uric acid regulates AR expression both in cultured hepatocytes (HepG2 cells) and in the liver of hyperuricemic rats, and also whether this stimulation is associated with endogenous fructose production and fat (triglyceride) accumulation.

Their latest results are summarized as follows2. In human HepG2 cells exposed to uric acid of 4 mg/dL (normouricemia), 8 mg/dL and 12 mg/dL (hyperuricemia) for 72 h, AR expression was upregulated by uric acid in a dose‐dependent manner, and significant upregulation of sorbitol dehydrogenase and fructokinase/KHK was also observed. Interestingly, coexistence of uric acid and probenecid, a uric acid transporter inhibitor, prevented AR upregulation, signifying the regulation of AR expression by intracellular uric acid. However, as sorbitol and fructose did not increase in AR‐deficient cells, it is clear that these products in the polyol pathway were mediated through the upregulation of AR. In the next step, they observed that uric acid‐dependent AR expression was mediated by increased transcriptional activity. Namely, they found a significant enrichment of the transcription factor, nuclear factor of activated T cells 5 (NFAT5) in pure nuclear fractions of HepG2 cells exposed to uric acid. Furthermore, AR upregulation by uric acid decreased remarkably in NFAT5‐deficient cells. It is interesting to note that the luciferase signal system (obtained by cloning the human AR promoter upstream) activated by uric acid was strongly prevented by the anti‐oxidant molecule, apocynin, suggesting that NFAT5‐dependent activation of AR is induced by oxidative stress. Finally, they attempted to confirm whether the activation of AR can shift glucose into the polyol pathway for endogenous fructose production and metabolism, and then fat accumulation in human hepatocytes and in rats. They observed that AR upregulation induced from high glucose (25 mmol/L) led to a marked increase in both sorbitol and fructose in hepatocytes, resulting in the elevation of intracellular triglycerides. However, none of these increments was found in AR‐deficient cells. Intracellular oxidative stress, as well as sorbitol, fructose and triglycerides, were markedly greater in HepG2 cells exposed to both high glucose (12.5 and 25 mmol/L) and high uric acid (12 mg/dL), as compared with the control group. They also found that elevated hepatic uric acid induced upregulation of AR and KHK in rats, as well as an increase in intrahepatic sorbitol and fructose levels, and an increase of NFAT5 expression appeared in the nucleus of hyperuricemic rats as compared with the control or allopurinol, xanthine oxidase inhibitor‐treated animals. In Figure 1, their new results are summarized, along with their previous data that uric acid also upregulated KHK and fructose metabolism through the activation of the transcription factor, carbohydrate response element‐binding protein.

This observation by Sanchez‐Lozada et al. 2 is very important, because it means that hyperglycemia‐induced polyol pathway hyperactivity in diabetes might cause further activation with uric acid, contributing not only to the development of NAFLD, but also the progress of diabetic complications in related tissues. Incidentally, the relationship between uric acid and diabetes is still inconclusive. However, a recent study of non‐diabetic individuals showed that uric acid levels in plasma increase with 2‐h plasma glucose, but not with fasting plasma glucose, and uric acid levels are inversely associated with both fasting plasma glucose and 2‐h plasma glucose in the diabetic population8. Furthermore, uric acid levels correlate well with clinical and electrophysiological severity of diabetic sensorimotor polyneuropathy9. Thus, the management of uric acid in patients with diabetes is important and necessary for the prevention of diabetic complications together with NAFLD. Therefore, we cannot ignore the management of uric acid while maintaining good long‐term glucose control and carrying out proper diet therapy to avoid foods and drinks containing high fat and high fructose. Due to the fact that both NAFLD and diabetic complications are markedly related with polyol pathway hyperactivity, it would not be surprising that the suppression of AR, a key enzyme in the polyol pathway, by AR inhibitors might be useful as a tool to manage these disorders. In addition, it is supposed that sodium–glucose cotransporter 2 inhibitors, a type of oral hypoglycemic agent, might be helpful to prevent the development of either NAFLD or diabetic complications caused by hyperuricemia, because it is well known that sodium–glucose cotransporter 2 inhibitors can reduce serum uric acid through the urinary excretion of uric acid10. Actually, sodium–glucose cotransporter 2 inhibitor in a clinical trial was shown to reduce hepatic fat content in type 2 diabetes patients11.

非酒精性脂肪肝病（NAFLD）是世界范围内最常见的肝脏疾病之一。 NAFLD被认为是代谢综合征的肝脏成分，因为其特征与肥胖、炎症、胰岛素抵抗和2型糖尿病等代谢性疾病非常相似。很明显，NAFLD与2型糖尿病有着密切的关系。然而，NAFLD 发病机制和进展的确切机制仍不完全清楚1-3 。众所周知，肝脏中葡萄糖和脂质代谢的调节是由胰岛素通过直接或间接机制正确进行的。因此，提出两种可能的假设来解释其发病机制。一种可能性是，胰岛素抵抗和血液中过多的脂肪酸导致单纯性肝脂肪变性。第二个假设涉及氧化应激、脂质过氧化和线粒体功能障碍。肝脏 rho 相关卷曲螺旋激酶 1 被认为是 NAFLD 的一种新的可能发病机制，通过抑制 5'-单磷酸腺苷激活蛋白激酶3 ，在高脂饮食诱导的肝脏脂肪生成中发挥重要作用。这个假设是在高脂肪饮食的情况下发生的。鉴于目前糖尿病患者的饮食治疗，这种方法和数据非常有用和有意义。最近，从过量摄入果糖（糖或玉米糖浆）的角度提出了一个非常有吸引力且有趣的 NAFLD 发病机制假说2 。这一全新理论与尿酸激活的多元醇途径亢进有着密切的联系。

众所周知，高血糖引起的多元醇途径过度活跃可导致糖尿病并发症的发生，例如微血管病、大血管病等4-7 。此外，许多研究已彻底证实，抑制醛糖还原酶（AR）（该途径中的关键酶）有助于预防这些并发症。多元醇途径仅包含两个步骤：首先通过 AR 将葡萄糖还原为山梨醇，然后通过山梨醇脱氢酶将所得山梨醇转变为果糖（图 1）。在血糖正常期间，通过多元醇途径使用葡萄糖占<3 id=17>7 。高血糖激活多元醇途径过度活跃诱发糖尿病并发症的机制并不像我们想象的那么简单。一般来说，蛋白激酶C、糖化、氧化应激等代谢因素参与该通路的下游，并导致复杂问题的糖尿病并发症的进展6, 7 。此外，根据糖尿病并发症的类型，与多元醇途径过度活跃相关的以下因素也可能部分参与：炎症、内皮一氧化氮合酶、血栓素、基质金属蛋白酶、烟酰胺磷酸羧基转移酶、一氧化氮、组织因子、血管细胞粘附分子和转化生长因子-β 途径多个基因的表达6, 7 。

最近，Sanchez‐Lozada等人。 2提出了一个非常有趣且有吸引力的假设，即通过尿酸诱导的多元醇途径过度活跃来发展 NAFLD 的发病机制。该研究的概念基于以下事实：两种主要甜味剂，糖（蔗糖）和高果糖玉米糖浆，会诱发动物脂肪肝。此外，该课题组前期研究表明，果糖诱发脂肪肝的发病机制是由于果糖代谢过程中产生尿酸，导致线粒体氧化应激，三磷酸腺苷生成受损。他们还证实，高尿酸血症本身不仅与高甘油三酯血症和NAFLD密切相关，而且还可以预测NAFLD的进展。此外，他们发现尿酸通过激活转录因子、碳水化合物反应元件结合蛋白来上调果糖激酶/酮己糖激酶（KHK）和果糖代谢。他们最近基于这些事实的研究评估了尿酸是否调节培养的肝细胞（HepG2细胞）和高尿酸血症大鼠肝脏中的AR表达，以及这种刺激是否与内源性果糖的产生和脂肪（甘油三酯）的积累有关。

他们的最新成果总结如下2 。在人 HepG2 细胞暴露于 4 mg/dL（正常尿酸血症）、8 mg/dL 和 12 mg/dL（高尿酸血症）尿酸 72 小时时，AR 表达以剂量依赖性方式被尿酸上调，并且显着上调还观察到山梨醇脱氢酶和果糖激酶/KHK 的存在。有趣的是，尿酸和丙磺舒（一种尿酸转运蛋白抑制剂）的共存可以阻止 AR 上调，这表明细胞内尿酸对 AR 表达的调节。然而，由于山梨醇和果糖在 AR 缺陷细胞中没有增加，很明显多元醇途径中的这些产物是通过 AR 上调介导的。在下一步中，他们观察到尿酸依赖性 AR 表达是由转录活性增加介导的。也就是说，他们发现暴露于尿酸的 HepG2 细胞的纯核部分中转录因子、活化 T 细胞核因子 5 (NFAT5) 显着富集。此外，在 NFAT5 缺陷细胞中，尿酸对 AR 的上调显着降低。有趣的是，由尿酸激活的荧光素酶信号系统（通过克隆人类 AR 启动子上游获得）被抗氧化剂分子夹竹桃麻素强烈阻止，这表明 NFAT5 依赖性 AR 激活是由氧化应激诱导的。最后，他们试图确认AR的激活是否可以将葡萄糖转移到多元醇途径中，用于内源性果糖的产生和代谢，然后在人肝细胞和大鼠中积累脂肪。 他们观察到，高葡萄糖（25 mmol/L）引起的AR上调导致肝细胞中山梨醇和果糖显着增加，导致细胞内甘油三酯升高。然而，在 AR 缺陷细胞中没有发现这些增量。与对照组相比，暴露于高葡萄糖（12.5 和 25 mmol/L）和高尿酸（12 mg/dL）的 HepG2 细胞中的细胞内氧化应激以及山梨醇、果糖和甘油三酯明显更高。他们还发现，与对照组或别嘌呤醇相比，肝尿酸升高会引起大鼠AR和KHK的上调，以及肝内山梨醇和果糖水平的增加，并且高尿酸血症大鼠的细胞核中NFAT5表达增加。黄嘌呤氧化酶抑制剂治疗的动物。图 1 总结了他们的新结果以及之前的数据，即尿酸还通过激活转录因子、碳水化合物反应元件结合蛋白来上调 KHK 和果糖代谢。

Sanchez‐Lozada等人的观察结果。 2非常重要，因为这意味着糖尿病中高血糖引起的多元醇途径过度活跃可能会导致尿酸进一步激活，不仅促进NAFLD的发展，而且还会促进相关组织糖尿病并发症的进展。顺便说一下，尿酸与糖尿病之间的关系目前还没有定论。然而，最近一项针对非糖尿病个体的研究表明，血浆中的尿酸水平随 2 小时血糖升高而升高，但不随空腹血糖升高，并且尿酸水平与空腹血糖和 2 小时血糖呈负相关。在糖尿病人群中8 .此外，尿酸水平与糖尿病感觉运动性多发性神经病的临床和电生理严重程度密切相关9 。因此，糖尿病患者的尿酸管理对于预防糖尿病并发症和 NAFLD 非常重要且必要。因此，我们在长期保持良好的血糖控制、进行适当的饮食治疗、避免高脂肪、高果糖的食物和饮料的同时，不能忽视尿酸的管理。由于 NAFLD 和糖尿病并发症均与多元醇途径过度活跃显着相关，因此 AR 抑制剂对 AR（多元醇途径中的关键酶）的抑制可能可作为治疗这些疾病的工具，这一点并不奇怪。 。 此外，据推测钠-葡萄糖协同转运蛋白 2 抑制剂（一种口服降糖药）可能有助于预防 NAFLD 或高尿酸血症引起的糖尿病并发症的发生，因为众所周知，钠-葡萄糖协同转运蛋白 2 抑制剂可降低血尿酸，通过尿排出尿酸10 ．事实上，临床试验表明钠-葡萄糖协同转运蛋白 2 抑制剂可以降低 2 型糖尿病患者的肝脏脂肪含量11 。