Maternal Consumption of Low-Isoflavone Soy Protein Isolate Confers the Increased Predisposition to Alcoholic Liver Injury in Adult Rat Offspring

Abstract

Offspring of female rats fed either a casein (CAS) diet or a low-isoflavone soy protein isolate (SPI) diet were compared in an animal model of chronic ethanol consumption to investigate whether maternal diet regulates the adaptive responses of offspring to postnatal ethanol exposure and potentially affects the development of liver disease in later life. Female rats were fed either a CAS or an SPI diet before mating, and during pregnancy and lactation. Male offspring from the same litter were pair-fed either a control or ethanol diet for six weeks (CAS/CON, CAS/EtOH, SPI/CON, and SPI/EtOH groups). Serum aminotransferase activities and hepatic inflammatory indicators were higher in the SPI/EtOH group than in the CAS/EtOH group. Ethanol consumption increased serum homocysteine levels, hepatic S-adenosylmethionine:S-adenosylhomocysteine ratio, and hepatic endoplasmic reticulum stress only in offspring of SPI-fed female rats. Total and high-density lipoprotein (HDL) cholesterol levels and mRNA levels of hepatic genes involved in HDL cholesterol assembly were reduced in the SPI group in response to ethanol consumption. In conclusion, offspring of SPI-fed female rats were more susceptible to the later development of alcoholic liver disease than offspring of CAS-fed female rats. Furthermore, maternal SPI consumption altered one-carbon metabolism and cholesterol metabolism of offspring fed an ethanol diet.

Keywords: HDL cholesterol; alcoholic liver disease; one-carbon metabolism; rat offspring; soy protein isolate.

Figure 1

Overview of the study design. CAS, casein; SPI, low-isoflavone soy protein isolate.

Figure 2

Effects of maternal diet on the liver injury in adult offspring fed an ethanol diet. (a) Serum glutamic-oxaloacetic transaminase (GOT) levels (n = 8), (b) serum glutamic-pyruvic transaminase (GPT) levels (n = 8), (c) serum bile acid levels (n = 8), (d) serum monocyte chemoattractant protein-1 (MCP1) levels (n = 7‒8), (e) hepatic MCP1 mRNA levels (n = 4‒5), (f) hepatic TNFα mRNA levels (n = 4‒5), (g) correlation between hepatic TNFα mRNA levels and serum GPT levels, and (h) representative H&E staining of liver tissue sections (n = 4). The mRNA levels were analyzed by real-time PCR and normalized to beta-actin (ACTB) as an endogenous control. Data are expressed as means ± SEM. Significant effects of maternal diet (M), offspring ethanol diet (E), and their interaction (M × E) were analyzed by two-way ANOVA (p < 0.05). Means without a common superscript significantly differ based on post-hoc analysis (p < 0.05). Pearson correlation coefficient, r, and p-value are indicated. Data that were not normally distributed were log-transformed before the statistical analysis. CAS, casein; SPI, low-isoflavone soy protein isolate; CON, control; EtOH, ethanol.

Figure 3

Effects of maternal diet on homocysteine and one-carbon metabolism in adult offspring fed an ethanol diet. (a) Serum homocysteine levels (n = 5) and (b) their correlation with serum GPT levels; (c) Hepatic S-adenosylmethionine (SAM):S-adenosylhomocysteine (SAH) ratio (n = 7‒8); Hepatic mRNA levels of genes involved in (d–f) methylation reaction and (g,h) homocysteine metabolism (n = 5). The mRNA levels were analyzed by real-time PCR and normalized to ACTB as an endogenous control. Data are expressed as means ± SEM. Significant effects of maternal diet (M), offspring ethanol diet (E), and their interaction (M × E) were analyzed by two-way ANOVA (p < 0.05). Means without a common superscript significantly differ based on post-hoc analysis (p < 0.05). Pearson correlation coefficient, r, and p-value are indicated. Data that were not normally distributed were log-transformed before the statistical analysis. CAS, casein; SPI, low-isoflavone soy protein isolate; CON, control; EtOH, ethanol.

Figure 4

Effects of maternal diet on the endoplasmic reticulum (ER) stress response in adult offspring fed an ethanol diet. (a,b) Hepatic spliced:unspliced XBP1 mRNA levels were analyzed by semi-quantitative PCR and normalized to ACTB as an endogenous control (n = 4–5). (c) Correlation between hepatic spliced:unspliced XBP1 mRNA levels and serum GPT levels. (d,e) Hepatic p-eIF2α protein levels were analyzed by immunoblotting and normalized to total eIF2α protein levels. (f) Hepatic CCAAT/enhancer binding protein homologous protein (CHOP) mRNA levels were analyzed by real-time PCR and normalized to ACTB as an endogenous control (n = 5). Data are expressed as means ± SEM. Significant effects of maternal diet (M), offspring ethanol diet (E), and their interaction (M × E) were analyzed by two-way ANOVA (p < 0.05). Means without a common superscript significantly differ based on post-hoc analysis (p < 0.05). Pearson correlation coefficient, r, and p-value are indicated. Data that were not normally distributed were log-transformed before the statistical analysis. CAS, casein; SPI, low-isoflavone soy protein isolate; CON, control; EtOH, ethanol.

Figure 4

Effects of maternal diet on the endoplasmic reticulum (ER) stress response in adult offspring fed an ethanol diet. (a,b) Hepatic spliced:unspliced XBP1 mRNA levels were analyzed by semi-quantitative PCR and normalized to ACTB as an endogenous control (n = 4–5). (c) Correlation between hepatic spliced:unspliced XBP1 mRNA levels and serum GPT levels. (d,e) Hepatic p-eIF2α protein levels were analyzed by immunoblotting and normalized to total eIF2α protein levels. (f) Hepatic CCAAT/enhancer binding protein homologous protein (CHOP) mRNA levels were analyzed by real-time PCR and normalized to ACTB as an endogenous control (n = 5). Data are expressed as means ± SEM. Significant effects of maternal diet (M), offspring ethanol diet (E), and their interaction (M × E) were analyzed by two-way ANOVA (p < 0.05). Means without a common superscript significantly differ based on post-hoc analysis (p < 0.05). Pearson correlation coefficient, r, and p-value are indicated. Data that were not normally distributed were log-transformed before the statistical analysis. CAS, casein; SPI, low-isoflavone soy protein isolate; CON, control; EtOH, ethanol.

Figure 5

Effects of maternal diet on serum and hepatic lipid levels in adult offspring fed an ethanol diet. Serum (a) triglyceride, (b) total cholesterol, and (c) HDL cholesterol levels (n = 8); Hepatic (d) triglyceride and (e) cholesterol levels (n = 7‒8). Data are expressed as means ± SEM. Significant effects of maternal diet (M), offspring ethanol diet (E), and their interaction (M × E) were analyzed by two-way ANOVA (p < 0.05). Means without a common superscript significantly differ based on post-hoc analysis (p < 0.05). Data that were not normally distributed were log-transformed before the statistical analysis. CAS, casein; SPI, low-isoflavone soy protein isolate; CON, control; EtOH, ethanol.

Figure 5

Effects of maternal diet on serum and hepatic lipid levels in adult offspring fed an ethanol diet. Serum (a) triglyceride, (b) total cholesterol, and (c) HDL cholesterol levels (n = 8); Hepatic (d) triglyceride and (e) cholesterol levels (n = 7‒8). Data are expressed as means ± SEM. Significant effects of maternal diet (M), offspring ethanol diet (E), and their interaction (M × E) were analyzed by two-way ANOVA (p < 0.05). Means without a common superscript significantly differ based on post-hoc analysis (p < 0.05). Data that were not normally distributed were log-transformed before the statistical analysis. CAS, casein; SPI, low-isoflavone soy protein isolate; CON, control; EtOH, ethanol.

Figure 6

Effects of maternal diet on the gene expression related to cholesterol metabolism in the liver of adult offspring fed an ethanol diet. Hepatic mRNA levels of genes involved in (a–e) cholesterol metabolism and (f–j) HDL cholesterol metabolism (n = 4‒5). The mRNA levels were analyzed by real-time PCR and normalized to ACTB as an endogenous control. Data are expressed as means ± SEM. Significant effects of maternal diet (M), offspring ethanol diet (E), and their interaction (M × E) were analyzed by two-way ANOVA (p < 0.05). Means without a common superscript significantly differ based on post-hoc analysis (p < 0.05). Data that were not normally distributed were log-transformed before the statistical analysis. CAS, casein; SPI, low-isoflavone soy protein isolate; CON, control; EtOH, ethanol.

Figure 6

Effects of maternal diet on the gene expression related to cholesterol metabolism in the liver of adult offspring fed an ethanol diet. Hepatic mRNA levels of genes involved in (a–e) cholesterol metabolism and (f–j) HDL cholesterol metabolism (n = 4‒5). The mRNA levels were analyzed by real-time PCR and normalized to ACTB as an endogenous control. Data are expressed as means ± SEM. Significant effects of maternal diet (M), offspring ethanol diet (E), and their interaction (M × E) were analyzed by two-way ANOVA (p < 0.05). Means without a common superscript significantly differ based on post-hoc analysis (p < 0.05). Data that were not normally distributed were log-transformed before the statistical analysis. CAS, casein; SPI, low-isoflavone soy protein isolate; CON, control; EtOH, ethanol.