研究目的

5’-肌苷酸(inosine 5’-monphosphate，简称IMP)调控核酸代谢、能量转移等重要生理功能。动物体内除了自身合成外，也通过饮食摄取以维持代谢平衡。IMP的二钠盐是食品行业中的鲜味剂，尚未对单日摄入容许量做出规定。前期我们发现以50 mg/(kg m_b·d)的IMP对db/db小鼠灌胃给药8周，小鼠出现了非酒精性脂肪性肝硬化（NAFLD-LC）症状，首次揭示了IMP具有潜在的肝脏毒性。  

鉴于db/db小鼠和正常小鼠的遗传学差异，必须了解IMP对这两种体质小鼠的作用差别，才能全面地理解IMP的食品安全性及损害肝脏的详细作用机制。

实验方法

参照IMP对db/db小鼠危害剂量，开展了以10，50，100 mg/(kg m_b·d)的IMP（以下简称低、中、高剂量）对4月龄C57BL/6J小鼠(正常小鼠)的肝损伤研究，连续给药4个月后，对小鼠体重、体脂率进行检测，测定小鼠血清内血脂、炎症因子、肝损伤因子和乙酰辅酶A等含量以及肝脏组织油红O染色观测脂肪蓄积，肝脂及乙酰辅酶A的含量分析。

为探究IMP是否具有致癌效果，对正常小鼠、IMP给药各组正常小鼠、db/db小鼠以及IMP给药db/db小鼠的肝脏开展了DNA彗星电泳实验，彗星电泳实验用于检测细胞DNA损伤情况。DNA损伤越严重，DNA超螺旋结构越松散，产生的断裂点越多，电泳结果显示出彗星形状，根据彗尾的长度、面积和荧光强度判断细胞是否癌变。

在前期研究中，我们发现IMP可以与AMPKγ1亚基形成稳定的复合体，通过促进AMPK的磷酸化而活化ACC-脂肪酸β氧化，在体外实验中IMP显示强烈的降脂活性，但在db/db小鼠肝脏内显著促进了以乙酰辅酶A合成甘油三酯（TG）而导致NAFLD恶化，出现了NAFLD-LC。为证明AMPK是IMP调控脂肪代谢的作用靶点，利用油酸诱导脂肪蓄积的HepG2细胞模型研究AMPK正常和被抑制功能期间 IMP的降脂活性，根据IMP的功效差异验证AMPK是IMP的作用靶点。利用蛋白质免疫印迹（Western blot）实验探究IMP给药对正常小鼠肝脏内AMPK下游的脂质分解代谢相关蛋白表达的影响情况。IMP诱导正常小鼠产生早期NAFLD症状，db/db小鼠自发产生NAFLD症状和IMP诱导db/db小鼠产生的肝硬化症状反映了高嘌呤饮食环境中的人类因瘦素抵抗逐渐恶化，出现不同阶段的NAFLD的原因。发现每个阶段的差异表达的蛋白质种类可解析NAFLD演化的分子机制。基于TMT定量蛋白质组学发现不同诱因导致小鼠NAFLD发病的肝细胞内蛋白种类和利用生物信息学方法研究IMP通过诱发这些蛋白质表达变化导致NAFLD发生和演化的作用机制。

实验结果

IMP对正常小鼠体重无显著影响，100 mg/(kg m_b·d) IMP引起小鼠体脂率显著升高，表明高剂量IMP促进正常小鼠体内脂质积累。IMP给药后各组小鼠的血脂较正常组升高，乙酰辅酶A含量随IMP给药剂量的升高而升高，炎症因子随IMP给药剂量的升高而降低，肝损伤因子无显著变化。肝组织切片经油红O染色观察发现低剂量组小鼠肝脏无明显的脂肪蓄积，中、高剂量组小鼠肝脏内出现脂质蓄积现象，脂肪蓄积量和IMP剂量为正相关，表明高剂量的IMP给药依然促进正常小鼠NAFLD，但即使是高剂量IMP，诱导的正常小鼠NAFLD的症状可能逆转，因此不能武断地判定IMP有食品危害性。但对严重瘦素抵抗体质的小鼠或人，不宜在饮食中额外添加IMP，同时要避免高嘌呤饮食。

DNA彗星电泳实验显示仅IMP给药的db/db小鼠肝脏DNA被损伤，表明IMP对db/db这样的极端瘦素抵抗体质的小鼠有促进肝细胞癌变的活性，验证了IMP诱导db/db小鼠出现的症状为癌变前期的NAFLD-LC。

利用AMPK抑制剂进一步验证AMPK是否为IMP在肝细胞内的作用靶点，证明IMP和AMPK形成复合体导致AMPK激活，引发了脂肪酸β氧化反应加速，促进了乙酰辅酶A蓄积、TG合成和肝脏氧化应激等诱发db/db小鼠的NAFLD-LC。

蛋白质免疫印迹（Western blot）实验发现IMP活化AMPK促进ACC1和ACC2表达和磷酸化，加速脂肪酸β氧化生成乙酰辅酶A，促进脂肪合成；同时诱导ATGL表达促进脂肪分解。表明IMP作用下正常小鼠肝脏内出现了促进脂肪蓄积和脂肪分解的互相拮抗。低剂量IMP作用时脂肪分解能力大于合成能力，中、高剂量IMP作用时脂肪分解能力小于蓄积能力，因而观察到IMP促进脂肪蓄积和剂量之间存在正相关。

TMT定量蛋白质组学和生物信息学方法发现IMP导致正常小鼠肝脏内Hepcidin等4种蛋白含量显著增加，补体成分 C9(C9)等23种蛋白表达显著减少；IMP导致db/db小鼠肝脏内Phospho1等65种蛋白显著增加，Camlg等110种蛋白表达显著减少。db/db小鼠相对于正常小鼠，肝脏内Get4等260种蛋白表达显著增加，Jagn1等148种蛋白表达显著减少。IMP诱导正常小鼠、IMP诱导db/db小鼠以及db/db小鼠产生各自表型的NAFLD进程中C9表达量都降低，表明IMP通过下调C9表达引起肝细胞炎症是促进NAFL发生和进展为NAFLD-LC的原因，这种炎症发生和肝细胞内脂质变性无因果关系。此外，db/db小鼠相对于正常小鼠，肝细胞胞间质（ECM）沉积相关的蛋白质如ITB3、LAMB2等的表达量显著提升，是导致NAFLD-LC发生的必要条件。

结论

对照IMP诱导正常饮食的C57/BL6J小鼠和db/db小鼠出现的NAFLD的表型差异和作用机制证明NAFLD-LC发病的必要条件是高度瘦素抵抗体质动物体内出现IMP过剩。此外，IMP导致肝脏内补体蛋白C9的表达下调是引发非酒精性脂肪肝炎，加速NAFLD恶化，为研究NAFLD-LC的发病机制提供新思路。这些结果证明了常规饮食条件下，由于代谢异常或过量摄入IMP引发体内嘌呤核苷酸过剩是NAFLD发生和演化的原因。因此，正常饲养条件下用IMP诱导正常小鼠可建立NAFLD的疾病新模型，IMP诱导db/db小鼠可建立NAFLD-LC疾病新模型，这两种模型分别反映了高嘌呤饮食环境中的人类NAFLD或NAFLD-LC的病理特征，为NAFLD发病和演化为肝硬化分子机制研究指明了新方向。

Aim

Inosine 5’-monphosphate (IMP) regulates nucleic acid metabolism, energy transfer and other important physiological functions. IMP can be produced in different organs of higher animals through de novo or remedial synthesis pathways. IMP disodium salt is the food industry as an additive to increase the fresh place, has not yet made provisions on a single day intake tolerance.

Given the genetic differences between db/db mice and normal mice, it is essential to understand the effects of IMP on these two types of mice in order to understand the food safety and liver damage mechanism of IMP in detail.

Methods

According to the harmful dose of IMP to db/db mice, the liver injury of 4-month-old C57BL/6J mice (normal mice) was studied by using 10, 50 and 100 mg/(kg·m_b·d) IMP (hereinafter referred to as low, medium, and high dose). After 4 months of continuous administration, the body weight and body fat percentage of mice were detected. Serum lipid, inflammatory factor, liver injury factor and acetyl-coA were analyzed. The contents of liver fat and acetyl-coA were determined, and the fat accumulation in liver tissue was observed by oil red O staining.

To investigate whether IMP has carcinogenic activity, DNA comet electrophoresis was performed using the liver tissues of normal mice, normal mice of IMP administration groups, db/db mice and the liver of IMP administration db/db miceas the samples. Comet electrophoresis was used to detect the DNA damage of liver cells. The more serious the DNA damage, the more loose the DNA superhelix structure, the more broken points, the smaller the DNA fragments, the more DNA fragments appeared in the tail of the comet, the greater the length, area, and fluorescence intensity of the comet tail.

In previous studies, we found that IMP can form a stable complex with AMPK-γ1 subunit promoting the fatty acid β-oxidation. In vitro experiments, IMP showed remarkable high lipid-lowering activity, but significantly promoted the synthesis of acetyl-CoA-TG in liver of db/db mice, leading to the NAFLD-LC emerged. In order to make sure that AMPK is the target of IMP regulating fat metabolism, the lipid-lowering activity of IMP during normal and inhibited AMPK function was studied in HepG2 cells with oleic acid induced fat accumulation, and AMPK was the target of IMP based on the efficacy difference of IMP. Western blot assay was used to investigate the effect of IMP administration on the expression of lipid-catabolism related proteins downstream of AMPK in liver of normal mice. IMP induced early NAFLD symptoms in normal mice, spontaneous NAFLD symptoms in db/db mice and liver cirrhosis symptoms in IMP induced db/db mice reflect the causes of different stages of NAFLD in humans in a high-purine diet due to increasing leptin resistance. The discovery of differentially expressed protein species at each stage can elucidate the molecular mechanism of NAFLD evolution. Based on TMT quantitative proteomics, the types of proteins in liver cells that are caused by different inductions of NAFLD in mice were found, and bioinformatics methods were used to study the mechanism of IMP inducing changes in the expression of these proteins to cause NAFLD occurrence and evolution.

Results

IMP had no significant increase of the body weight of normal mice, but 100 mg/(kg·m_b·d) IMP induced that significantly increase of the body fat percentage of the normal mice, suggesting that high dose IMP promoted the accumulation of lipid content in normal mice. After IMP administration, the serum lipid of mice in each group was higher than that in the normal group, acetyl-CoA content was increased with the increase of IMP dose, inflammatory factors were decreased with the increase of IMP dose, liver injury factor had no significant change. Lipid accumulation was found in liver of mice in medium and high dose groups. There was a positive correlation between the amount of fat accumulation and IMP dose, indicating that high-dose IMP still promoted NAFLD in normal mice. The symptoms of NAFLD induced in normal mice may be reversed, and the harm to normal mice is less, therefore it can not arbitrarily determine IMP food hazard. For mice or people with severe leptin resistance, IMP should not be added in the diet, and the high purine diet should be avoided.

DNA comet electrophoresis showed that only the db/db mice treated by IMP was damaged, indicating that IMP has the activity of promoting hepatocyte carcinogenesis in db/db mice with extreme leptin resistance, confirming that the symptoms induced by IMP in db/db mice were NAFLD-LC in pre-cancerous stage.

AMPK inhibitors were used to further verify whether AMPK is the target of IMP in liver cells, and it was proved that the combination of IMP and AMPK leads to the activation of AMPK, which leads to the accelerated oxidation reaction of fatty acid β, and promotes the accumulation of acetyl-CoA, triglyceride synthesis and liver oxidative stress to induce NAFLD-LC in db/db mice.

Western blot assay showed that IMP activated AMPK promoted the expression and phosphorylation of ACC1 and ACC2, accelerated the fatty acid β-oxidation to produce acetyl-CoA, promoted fat synthesis and inhibited lipid decomposition. At the same time, ATGL expression can be increased to promote fat decomposition. The results showed that there was antagonism to promote fat accumulation and fat decomposition in liver of normal mice treated with IMP. The fat decomposition capacity was greater than the synthesis capacity at low dose of IMP, while the fat decomposition capacity was less than the accumulation capacity at medium and high dose of IMP. Therefore, a positive correlation between the dosage and the promotion of fat accumulation was observed.

TMT quantitative proteomics and bioinformatics methods showed that IMP significantly increased the contents of 4 proteins, including Hepcidin, and significantly decreased the expression of 23 proteins, including complement C9 (C9) in liver of normal mice. IMP significantly increased 65 proteins, such as Phospho1, and decreased 110 proteins, such as Camlg, in liver of db/db mice. Compared with normal mice, the expressions of 260 proteins, including Get4, were significantly increased in db/db mice, while the expressions of 148 proteins, including Jagn1, were significantly decreased. The expression levels of C9 in NAFLD process induced by IMP in normal mice and db/db mice  were all reduced, suggesting that IMP induces hepatocyte inflammation by down-regulating C9 expression, which is responsible for promoting the occurrence and progression of NAFLD-LC. There is no causal relationship between this inflammation and intrahepatic lipid degeneration. In addition, compared with normal mice, the expression levels of proteins related to ECM deposition in hepatocytes, such as ITB3 and LAMB2, were significantly increased in db/db mice, which is a necessary condition for the occurrence of NAFLD-LC.

Conclusion

The phenotypic differences and pharmacological mechanism of NAFLD in C57/BL6J mice and db/db mice induced by IMP respectively suggested that high leptin resistance caused by IMP excess was the necessary condition for the occurrence of NAFLD-LC. IMP down-regulates the expression of C9 in liver, which is an independent factor causing liver inflammation, providing new ideas for understanding the pathogenesis and evolution of NAFLD and NAFLD-LC. Even under normal diet, liver inflammation and fat accumulation induced by excess purine nucleotides caused by abnormal metabolism or excessive intake are the causes of the occurrence and evolution of NAFLD. Therefore, under normal feeding conditions, normal mice induced by IMP can establish a new model of NAFLD, and db/db mice induced by IMP can establish a new model of NAFLD-LC disease. These two models, respectively, reflect the pathological features of human NAFLD or NAFLD-LC in a high-purine diet setting a new direction for the study of the molecular mechanisms underlying the pathogenesis and evolution of NAFLD into cirrhosis.