UC Davis

Vitamin A deficiency (VAD) has shown to be present in 334 million children, with high rates occurring in young children and pregnant mothers. VAD in early life can have severe negative effects on development, growth, and health. Additionally, VAD is often concurrently present in areas with high rates of obesity, a form of double-burden malnutrition which is characterized by coexistence of undernutrition (e.g. micronutrient deficiency) and overnutrition. Given the rise of double-burden malnutrition the need for further investigation into how vitamin A status may regulate the development of chronic metabolic conditions is warranted. In our first study, using a mouse model of vitamin A deficiency, we aimed to examine how maternal parity and extended dietary treatment post-weaning shapes the risk of developing vitamin A deficiency, the role of vitamin A in regulating iron homeostasis, and the impacts on cognition. In the second study, we explored how vitamin A deficiency affects lipid metabolism in the liver, brain, and plasma. We further investigated how these changes influence brain and liver health, as well as the underlying molecular mechanisms driving the altered lipid profile observed in these tissues. Finally, in the third study, we investigated how maternal and post-weaning vitamin A deficiency interact with the effects of excess energy intake from a high-fat diet. We assessed the combined impact of these dietary challenges on growth, lipid metabolism, and the metabolome. In the first study, C57/BL6J dams were randomly assigned to a vitamin A-replete (VR) or vitamin A-free (VD) diet throughout the first two consecutive gestation and lactation periods. Mouse pups were weaned at 3 weeks of age and fed the same diet as their dams until 6 or 9 weeks of age. Offspring were subjected to a battery of behavioral assays for cognition, motor function, and anxiety at 6 or 9 weeks of age. Dietary deprivation of vitamin A depleted hepatic and plasma retinol in offspring born from both gestations. Growth retardation was observed in VD offspring born from the first gestation, exhibiting lower body weight, shorter tail and intestinal length (P < 0.05). Unexpectedly, at 9 weeks of age, motor dysfunction shown as abnormal posture, gait, and impaired balance and coordination on Rotarod test (P < 0.05) was the leading clinical manifestation of early-life VAD, exceeding ocular anomalies. In contrast, grip strength and spatial cognition were not significantly affected compared to the VR group. Previous studies reported interaction between vitamin A and trace mineral metabolism (e.g. zinc and iron). However, the current study did not find consistent changes in hepatic mineral levels in offspring of either gestation or age groups.Polyunsaturated fatty acids (PUFA) mobilized from membrane phospholipids are primary precursors for synthesis of intracellular oxylipins, which play a critical role in modulating tissue inflammation and progression of chronic metabolic diseases. Considering the evidence that VAD interrupt membrane phospholipid metabolism and PUFA composition, we further explored oxylipin profile in liver, brain and plasma and hepatic inflammation from 9-week-old mice born from the first gestation (Chapter III). VAD significantly increased 21 oxylipins in liver (P < 0.05) including 12 eicosanoids from arachidonic acid (AA), most of which promote inflammation. Consistently, the gene expression of pro-inflammation cytokine (TNF) and acute phase protein (CRP) were increased in liver of VD mice (P < 0.05). This was corroborated by histological observation of increased immune cell infiltration observed in some liver tissues from VD mice. The mRNA expression of liver fibrosis biomarkers (TGFB1 and COL1A1) was unaffected by VAD. Enzymes that participate in oxylipin synthesis and metabolism were analyzed for their mRNA expression. However, the transcriptional changes of these enzymes were not always aligned with the changes in oxylipin profile suggesting presence of additional regulatory mechanisms. VAD downregulated the mRNA expression of the retinoid transporter (STRA6) in the cerebrum, indicating a reduction in cerebral retinol levels. In contrast to the coordinated increase in hepatic oxylipins, the majority of altered oxylipins in the cerebrum (27 out of 34) were decreased by VAD. Notably, more of these affected oxylipins were derived from ω-3 PUFAs (e.g., ALA, DHA, and EPA) than from ω-6 PUFAs (e.g., LA and AA). VAD did not affect mRNA expression of inflammatory cytokines or oxylipin-synthesizing enzymes tested in the cerebrum. Despite the significant changes in tissue oxylipins, plasma oxylipin profile was unaffected by VAD, suggesting a localized, tissue-specific regulation of oxylipin synthesis. To further investigate the role of vitamin A in regulating lipid metabolism and the development of hepatic metabolic diseases under conditions of double-burden malnutrition (DBM), we conducted a second mouse study using a 2 × 2 factorial design with dietary treatments varying by fat content (normal fat, 10% kcal vs. high fat, 45% kcal) and vitamin A level (VD vs. VR). Like the first study, C57/BL6J dams were fed the experimental diets throughout gestation and lactation. Mouse pups were either euthanized at weaning (3 weeks of age) or continued on the same diet as their dams until 6 weeks of age. Regardless of fat level, VD significantly decreased hepatic retinol stores at 6 weeks. Interestingly, we observed significant interaction effects (P < 0.05) on body weight (BW), liver weight, and tail length at 3 weeks: VAD increased these developmental parameters in pups exposed to the high-fat diet (HF) but decreased them in those fed the normal-fat diet (CN). This interaction effect was not observed at 6 weeks. Histological analysis of hepatic tissue revealed clear signs of hepatic steatosis in HF-fed pups at 6 weeks, while VAD appeared to have minimal effect on vacuole area % covered. However, based on qualitative gross morphology, we observe smaller vacuoles in the VAD HF group compared to the CN HF group, suggesting differential lipid pathogenesis. A key finding from the gene expression analysis showed significant interaction effect on ACOX1, indicating diminished fatty acid β-oxidation associated with VAD in those fed the HF diet (P < 0.05). While genes involved in inflammation was not altered by dietary treatment, the HF intake significantly upregulated mRNA expression of fibrosis biomarkers in the liver irrespective of vitamin A status. Untargeted metabolome analysis revealed no significant main effect of VAD on liver metabolism at 3 weeks of age. Instead, 16 and 7 metabolites were significantly altered by the main effect of HF intake and its interaction with VAD, respectively. By 6 weeks of age, HF intake remained the primary driver of hepatic metabolic changes, affecting 43 metabolites, compared to 19 metabolites affected by the interaction and 5 by main effect of VAD. Plasma metabolome at 3 weeks showed significant main effect of HF intake on 26 metabolites. However, these effects diminished by 6 weeks, with fat level affecting 8 metabolites and VAD altering 1. In conclusion, this dissertation presents a mouse model of early-life VAD characterized by clinical symptoms and reduced retinol levels in liver and plasma. This mouse model is not intended to evaluate the role of vitamin A in fetal development or the teratogenic effects of severe vitamin A deficiency, which have been well documented in previous research. Instead, our objective is to induce a less severe form of vitamin A deficiency to provide greater translational insight into the motor, cognitive, and metabolic effects of VAD. These findings provide the first evidence that VAD alters oxylipin production locally in tissues and offers insight into potential molecular mechanisms. Furthermore, the study demonstrates how vitamin A deficiency and a high-fat diet interact in a maternal infant model, revealing unique effects on molecular, metabolomic, and phenotypic outcomes associated with this increasingly prevalent dietary pattern.