第一部分

母体孕期和哺乳期高脂饮食对子鼠在幼年期糖脂代谢的影响和机制研究

【研究目的】大量的临床研究和动物实验表明母代营养过剩可增加子代在成年期甚至老年期患肥胖、胰岛素抵抗、糖尿病等代谢性疾病的易感性，但是其作用机制一直不是十分清楚。另外，既往的研究多关注母代长期的高脂饮食对子代的远期影响，而关于对子代的早期影响并不清楚。因此，本研究运用跨代小鼠模型主要探讨母代高脂饮食对子代在生命早期（断乳时）糖脂代谢的影响及其潜在的机制。

【研究方法】建立跨代高脂饮食小鼠模型，将C57BL/6J雌性小鼠随机分为两组，分别在孕期和哺乳期给予正常饲料（11.4%脂肪、62.8%碳水化合物、25.8%蛋白质，总热量：3.9 千卡/克）和高脂饲料（58%脂肪、25.6%碳水化合物、16.4%蛋白质，总热量：5.56 千卡/克）。检测子鼠出生体重，在3周龄即断乳时行腹腔糖耐量试验，随之每窝取1只雄性子鼠、1只雌性子鼠处死采血，进行相关生化指标的检测，包括血清胰岛素、血脂（甘油三脂、胆固醇）、炎症因子（肿瘤坏死因子、白介素-6）等。另立即分离肝脏组织，采用全基因组表达谱芯片和microRNA芯片寻找差异表达的基因和microRNA分子，并利用生物信息学技术进行相关分析，随后进一步开展相关的分子生物学实验进行验证及深入分析，检测基因的表达情况。

【研究结果】

1、相比于母体正常饮食，母体短期（仅妊娠期和哺乳期）高脂饮食能够导致雄性子代在生命早期（3周龄时）出现肥胖、糖耐量异常、胰岛素抵抗、血脂异常升高及肝脏脂肪变性。而雌性子代在糖脂代谢指标方面未见明显差异。

2、通过基因芯片技术检测子鼠肝脏组织基因的表达情况，结果提示高脂饮食组和正常饮食组之间存在大量差异表达的基因，而这些基因可以富集在8条信号通路上。过氧化物酶体增殖物激活受体（PPAR）信号通路是其中一条十分重要的信号通路，实时荧光定量PCR发现PPAR信号通路上的4个基因表达均明显增加。提示PPAR信号通路可能在母代高脂饮食导致子代在生命早期出现糖脂代谢异常的过程中起着重要的作用。

3、另外研究表明母鼠高脂饮食能够调控子代在断乳时肝脏组织miRNAs的差异表达。结合生物信息学分析提示所有差异表达miRNAs的靶基因可以富集到7条与炎症密切相关的通路上。进一步发现高脂饮食组血清炎症因子明显升高，肝脏组织炎症调控因子蛋白表达水平明显上升，提示这些异常表达的miRNAs可能与子代机体炎症及糖脂代谢紊乱相关。

【研究结论】母代孕期和哺乳期高脂饮食摄入能够导致子代在生命早期（3周龄时）出现体重增加、糖耐量异常、胰岛素抵抗、脂代谢紊乱及肝脏脂肪变性。另外在高脂饮食组子鼠肝脏组织中发现部分差异表达基因和miRNA分子。而差异表达的基因富集在PPAR信号通路上，提示PPAR信号通路可能在母代高脂饮食导致子代在生命早期出现糖脂代谢异常的过程中起着重要的作用。而差异表达的miRNA可能参与调控机体的炎症状态。因此这些差异表达基因和miRNA可能是母体不良的发育营养导致子代在生命早期出现糖脂代谢异常的潜在作用机制。

第二部分

母体孕期和哺乳期高脂饮食对子鼠成年期-中老年期糖脂代谢的动态影响及miRNA的动态调控作用

【研究目的】大量的临床研究和动物实验均表明母代营养过剩可增加子代患肥胖、胰岛素抵抗、糖尿病等代谢性疾病的易感性。我们前期的研究提示母代高脂饮食能够导致子代在生命早期出现糖脂代谢的紊乱及miRNA的差异表达。那么这种作用是否会持续存在，差异表达的miRNA是否会持续起作用，因此，本研究主要探讨母代高脂饮食对子代在生命不同阶段（从幼年至中老年）糖脂代谢的动态影响及miRNA的动态调控作用。

【研究方法】建立跨代高脂饮食小鼠模型，将C57BL/6J雌性小鼠随机分为两组，分别在孕期和哺乳期给予正常饲料（Normal chow，NC）和高脂饲料（High-fat，HF）。子鼠于3周龄时断乳，在断乳后，每窝取1-2只雄性子鼠继续喂养，正常饲料组和高脂饲料组子鼠再随机给予正常饲料或高脂饲料，故将子鼠共分为四组：NC-NC组、NC-HF组、HF-NC和HF-HF组。研究对象为这四组子鼠，一直饲养至32周龄，期间监测体重的变化，每1-2月行糖耐量试验。在子鼠32周龄时处死取血进行生化指标检测，包括血清胰岛素、血脂（甘油三脂、胆固醇）、瘦素等。另立即分离肝脏组织，采用miRNA芯片寻找差异表达的miRNA分子，并与3周龄的结果进行对比及综合分析，动态观察miRNA分子表达的情况。

【研究结果】

1、母体和断乳后饮食的交互作用对子代糖脂代谢有显著影响，其中断乳后高脂饮食能导致子代出现糖脂代谢紊乱，包括肥胖、糖耐量异常、胰岛素抵抗、血脂水平升高及脂肪肝。而母体高脂饮食、断乳后给予正常饮食的子鼠与正常对照组在糖脂代谢上无明显差异，提示断乳后正常饮食可以逆转母体高脂饮食所带来的不良影响。

2、miRNA芯片结果提示：与正常对照组相比，母体高脂饮食、断乳后予正常饮食的子鼠在32周龄时肝脏组织miRNA的表达存在明显差异。其中8个miRNA表达明显下调，分别为mmu-miR-1948-3p、mmu-miR-409-3p、mmu-miR-592-5p、mmu-miR-127-3p、mmu-miR-541-5p、mmu-miR-149-5p、mmu-miR-379-5p和mmu-miR-344g-5p。 而mmu-miR-1247-5p和mmu-miR-150-5p表达显著上调。与3周龄子鼠miRNA表达谱相比，差异表达的miRNA明显不同，表明miRNA的表达呈动态变化。

3、通过实时荧光定量PCR验证及结合生物信息学分析提示所有差异表达的miRNAs的靶基因可以富集到AMPK等4条信号通路上。其中AMPK信号通路是最重要的一条信号通路，AMPK信号通路上的12个靶基因表达明显上调，AMPK信号通路的激活可抑制糖异生、激活线粒体生物合成以及抑制脂肪酸合成。

【研究结论】母体和断乳后饮食的交互作用对子代糖代谢有显著影响，其中断乳后高脂饮食能导致子代出现糖脂代谢紊乱，而断乳后正常饮食可以逆转母体高脂饮食所带来的不良影响。母鼠高脂饮食能够调控子代在中老年时（32周龄）肝脏组织miRNAs的差异表达，并且与3周龄子鼠肝脏差异表达的miRNA明显不同，呈现动态变化。在32周龄时，差异表达的miRNA的靶基因可能通过激活AMPK信号通路起调控作用，包括抑制糖异生、激活线粒体生物合成以及抑制脂肪酸合成。

第三部分

运动干预对子代在生命不同阶段糖代谢的影响及β细胞功能的研究

【研究目的】我们前期的研究结果提示母鼠妊娠期和哺乳期高脂饮食能够导致子代在生命早期出现肥胖、糖耐量异常、胰岛素抵抗和脂肪肝，一直持续至成年期甚至中老年期。近年来越来越多的研究表明父代不良饮食状况对子代糖脂代谢也有深远的影响。那么这些跨代的不良影响是否可以通过早期干预得到有效预防甚至逆转。因此，本研究的主要目的是探讨母代和父代运动对子代在生命不同阶段糖代谢的影响及其潜在的机制。

【研究方法】通过建立C57BL/6J小鼠跨代自主跑轮运动模型，母鼠在交配前2周随机分为运动组和非运动组，运动干预采用小鼠自主跑轮，而非运动小鼠则饲养于普通鼠笼。父鼠在交配前3周随机分为运动组和非运动组，随后与母鼠进行随机交配。交配结束后母鼠继续交配前的运动与非运动模式，直至分娩。期间母鼠和父鼠均喂养高脂饲料，母鼠喂养至哺乳结束。根据母代和父代的运动情况，将子鼠分为4组：非运动组、母代运动组、父代运动组、共同运动组。所有子鼠在出生后3周全部断乳，子鼠断乳后均给予正常饮食，饲养于普通鼠笼，一直饲养至52周龄。期间定期监测摄食量、体重、评估糖耐量和胰岛素抵抗情况，在3周龄和52周龄分别处死雄性子鼠取血进行生化指标的检测。另分离胰腺组织对胰腺β细胞各类指标进行评估，包括β细胞质量、β细胞增殖、β细胞凋亡和β细胞大小。

【研究结果】

1、母鼠在交配前经过2周的运动之后体重的变化无明显差异，但在孕期第2周运动组母鼠空腹血糖和胰岛素水平均低于非运动组。父鼠在交配前经过3周的运动可以显著减轻体重、降低空腹血糖和胰岛素水平、改善糖耐量异常、增强胰岛素敏感性。提示母代和父代运动可以显著改善自身糖代谢的状况。

2、母鼠及父鼠高脂饮食能导致子鼠在52周龄出现肥胖、糖耐量异常和胰岛素抵抗。但对于同样高脂饮食喂养的母鼠和父鼠，对母鼠和/或父鼠进行运动干预可以显著改善子代在52周龄的糖代谢异常，包括体重下降、减轻糖耐量异常、增加胰岛素的敏感性。同时还可以降低子代血清瘦素和抑胃肽的水平。

3、母鼠及父鼠高脂饮食能够导致子代在52周龄出现较大的β细胞质量，而仅母鼠或者仅父鼠运动的子代的β细胞质量较之变小，而父母鼠共同运动子代的β细胞质量最小。我们进一步检测影响β细胞质量的因素，包括β细胞增殖、β细胞凋亡以及β细胞大小。与父母代非运动的子鼠和父母代共同运动的子鼠相比，仅母代运动和仅父代运动的子鼠β细胞增殖的比例较高，同时各组子鼠之间β细胞的凋亡未见明显差异。但是仅母代运动组、仅父代运动组及父母代共同运动组子鼠在52周龄时β细胞的大小均明显小于父母代均未运动组。由此，结合上述结果可知各组子鼠之间β细胞质量的差异主要影响因素是β细胞的增殖和大小。

4、母鼠和/或父鼠运动干预无法改善子鼠在生命早期（3周龄）、12周龄及20周龄糖脂代谢的异常，另外对3周龄子鼠β细胞质量无明显影响。

【研究结论】母代和父代高脂饮食可以导致子鼠在52周龄出现肥胖、糖耐量异常和胰岛素抵抗，而父代和母代运动可以对子代糖代谢有明显的改善作用，然而这种对子代的保护效应可能会出现得较晚。其中β细胞的质量和β细胞的大小可能在其中起着重要的调节作用。

Part 1

The mechanisms between maternal high-fat diet and metabolic health in the early life of offspring

Objectives: Maternal high-fat diet feeding in mice predisposes offspring for impaired glucose homeostasis and obesity. However, the mechanisms underlying these detrimental effects of maternal nutrition, especially during early life of offspring, are incompletely understood. We aimed to determine the underlying mechanisms that influence these phenotype, especially the epigenetic modifications.

Materials and Methods: Using C57BL/6J mice, we examined the effects on the offspring from dams were fed a normal chow diet (NCD; 11.4% kcal fat, 62.8% from carbohydrate, 25.8% from protein) or high-fat diet (HFD; 58% kcal from fat; 25.6% from carbohydrate, 16.4% from protein) during gestation and lactation. Metabolic health was assessed in offspring at weaning. Gene array, hepatic levels of miRNAs and target genes were investigated in the liver tissues of the offspring mice from NCD- or HFD-fed dams at weaning.

Results: The offspring of the dams fed the high-fat diet had a heavier body weight, impaired glucose tolerance, decreased insulin sensitivity, increased serum cholesterol and hepatic steatosis at weaning. Bioinformatic analyses of gene array indicated that all differentially expressed genes of the offspring between the two groups were mapped to nine pathways. Genes in the PPAR signalling pathway were verified by quantitative real-time PCR and these genes were significantly upregulated in the high-fat diet offspring.

Microarray analysis indicated that expression of miR-615-5p, miR-3079-5p, miR-124* and miR-101b* was downregulated, whereas miR-143* was upregulated, in livers from offspring from HFD-fed dams. Our functional enrichment analysis indicated that the target genes of these differentially expressed miRNAs, including tumor necrosis factor-α (TNF-α) and mitogen activated protein kinase 1 (MAPK1), were mapped to inflammatory pathways. Finally we verified that both mRNA and protein levels of the pro-inflammatory modulators TNF-α and MAPK1 were significantly increased in livers of offspring from HFD-fed dams at weaning.

Conclusions: Maternal high-fat diet during pregnancy and lactation predisposes the offspring to the development of obesity, glucose intolerance, dyslipidemia and liver steatosis in the early stages of life, as early as at weaning age. Gene array analyses and gene validation indicated the PPAR signaling pathway may be the underlying mechanism that accounts for the phenomenon. Our study further showed that maternal high-fat diet consumption during pregnancy and lactation may regulate the expression of several miRNAs. After validation, it indicates that differentially expressed miRNAs may be associated with chronic inflammation status and glucose intolerance in offspring as early as at weaning age.

Part 2

The dynamic effects of maternal high-fat diet on metabolic health and miRNAs expression in offspring mice with the aging

Objective: Substantial studies demonstrated that maternal nutrition can significantly determine the susceptibility to developing some metabolic diseases in offspring. However, the dynamic effects of maternal high-fat diet on metabolic health and miRNAs expression in offspring mice, such as from weaning to later life, has not been well documented. Our objective was to explore the long-term effects of maternal and post-weaning diet interaction on offspring’s metabolic health with the aging.

Materials and Methods: Using C57BL/6J mice, we examined the effects on the offspring from dams were fed a normal chow (NC) diet or high-fat (HF) diet during gestation and lactation. At weaning, the male offspring of dams fed on either a HF or NC and then weaned to either a HF or NC diet, generating four groups post-weaning: NC–NC, NC–HF, HF–NC and HF–HF. Metabolic health was assessed in offspring from weaning to 32 weeks old. Hepatic levels of miRNAs and target genes were investigated in the liver tissues of the offspring mice at 32 weeks old.

Results: The NC-NC offspring had lower body weight than NC-HF group at 16 weeks of age (p<0.01) and both C-HF and HF-HF offspring had higher body weight than NC-NC group at 24 and 32 weeks of age (p<0.001, respectively). The blood glucose levels of the male offspring from the NC and HF dams weaned HF diet were significantly higher at 30 min, 60 min and 120 min (P<0.001) after intraperitoneal glucose administration compared with those of the NC-NC group. The NC-HF group had higher blood glucose at 30 min than HF-HF group (P<0.01). Furthermore, AUC in NC-HF and HF-HF groups was also significantly larger than NC-NC group (P<0.001). Fasting blood glucose and HOMA-IR of the offspring were significantly higher in NC-HF and HF-HF groups than NC-NC group at 32 weeks of age (P<0.05). It showed that post-weaning high-fat diet could contribute to abnormal glucose metabolism in the later-life of offspring, which is independent of dams fed with high-fat diet or normal chow diet.

Microarray analysis indicated that expression of mmu-miR-1948-3p, mmu-miR-409-3p, mmu-miR-592-5p, mmu-miR-127-3p, mmu-miR-541-5p, mmu-miR-149-5p, mmu-miR-379-5p, mmu-miR-344g-5p were downregulated, whereas mmu-miR-1247-5p and mmu-miR-150-5p was upregulated, in livers from offspring from HFD-fed dams. Our functional enrichment analysis indicated that the target genes of these differentially expressed miRNAs were mapped to AMPK pathways. Activations of AMPK pathway can inhibite expression of glucogenic genes, ttenuate gluconeogenic programme, promote mitochondrial biogenesis, repress fatty acid synthesis and increased free fatty acid oxidation.

Conclusions: Our result showed that offspring fed with high-fat diet from weaning to adulthood had heavier body weight, severer impaired glucose tolerance and lower insulin sensitivity level at 32 weeks of age. However, there is no difference of glucose and lipid metabolism between the offspring of dams fed with high-fat and normal chow diets. Furthermore, hepatic miRNAs expression is dynamically changed along with the aging of offspring. Thus, dynamic hepatic miRNAs expression may be the underlying mechanism between early life nutrition and metabolic health in offspring.

Part 3

The effects of maternal and paternal exercise on metabolic health and β cell function in offspring mice

Objective: Maternal and paternal environment play a critical role in determining risks of obesity and diabetes in offspring. Exercise has multiple health benefits, targeted to different organs and diseases. However, the effects of parental exercise on the metabolic phenotype of adult offspring are unknown and the mechanisms underlying these effects of parental exercise are incompletely understood. Our objective was to determine the effect of parental exercise on the metabolic health and β cell function of offspring mice.

Materials and Methods: Male and female mice were divided into sedentary and voluntary wheel cage exercise for 3 weeks and 2 weeks preconception, respectively and were fed a high-fat diet (60% kcal from fat). Male and female mice were mated together for 4 days. The exercised female mice continued to exercise for the 3 more weeks during gestation. This breeding scheme generated four groups of offspring: Sedentary (Sed), Maternal exercise (Mat Ex), Paternal exercise (Pat Ex) and both Maternal and Paternal exercise (Mat+Pat Ex). All male offspring were fed a chow diet and were sedentary from weaning thru 52 weeks.

Results: There was no difference in body weight change between exercised and sedentary dams. The fasting blood glucose and insulin concentration were significantly lower in exercised dams during the second week of pregnancy. However, there was no difference in glucose tolerance between sedentary and exercised dams. The exercised sires had a lower body weight increase when compared with sedentary sires. After breeding, the exercised sires had significantly lower fasting blood glucose and insulin concentrations as well as improved glucose and insulin tolerance. These data demonstrate that the metabolic health was improved in exercised dams and sires. 

At 52 weeks of age, a lower body weight was observed in male offspring of Mat+Pat Ex group, compared with the three other groups. Fasting blood glucose was significantly lower in Mat Ex, Pat Ex and Mat+Pat Ex offspring. However, there was no difference in insulin concentration in the offspring. The leptin concentration was significantly decreased in Mat+Pat Ex offspring at 52 weeks of age. And the gastric inhibitory polypeptide (GIP) concentration was strikingly decreased in Mat Ex, Pat Ex and Mat+Pat Ex offspring. Glucose tolerance and insulin sensitivity were significantly improved in male offspring at 52 weeks of age from high-fat fed, exercised parents.

Histological analysis of the pancreas revealed smaller islet size, lower total islet area and decreased islet density in Mat+Pat Ex offspring at 52 weeks of age. The immunofluorescence staining also showed that the insulin-positive area and beta cell mass were significantly decreased in Mat+Pat Ex offspring compared to Sed and had a tendency to be lower than Mat Ex and Pat Ex offspring. Both Ki67 and Phosphohistone H3 (PHH3) staining showed lower β cell proliferation in Mat+Pat Ex offspring, compared with Mat Ex and Pat Ex offspring. There was no difference of β cell apoptosis in the 52-week old male offspring. We further examined whether maternal and paternal can affect β cell size in male offspring at 52 weeks of age, then regulated the β cell mass. The co-staining of insulin and GLUT2 showed that β cell size was significantly decreased in male offspring from high-fat fed, exercised parents at 52 weeks of age. 

Another cohort of mice was used to determine the effect of maternal and paternal exercise on glucose metabolism and β cell function in offspring at weaning. There were no differences in the metabolic health of the offspring at weaning, including body weight, blood glucose, glucose tolerance and insulin sensitivity. However, these studies showed lower islet density and β cell proliferation (co-staining of Ki67 and insulin) in Pat Ex weaned offspring and lower β cell proliferation (co-staining of BrdU and insulin) in Mat+Pat Ex offspring. When studied at 12 and 20 weeks of age there were no differences in glucose and insulin tolerance in the offspring.

Conclusions: Taken together, our present study suggested that both maternal and paternal exercise has a profound effect on the offspring β cell mass due to, at least in part, β cell size, accompanied by the improved glycemic control at an older age. Thus, further studies to clarify the underlying mechanisms are urgently warranted