1.1. Body weights: The mean body weights of the pups from the IUGR and Sham groups were similar in the 2nd week. However, the mean body weights of the pups from IUGR in the 4th (p < 0.05) and 8th weeks were higher than those of the Sham group in the corresponding weeks. Notably, the catch-up growth was evident as early as the 2nd postnatal week.
1.2. Histopathological analyses: The histopathological analyses revealed some differences between the IUGR and Sham groups. In the sections from the 2-week-old rats in the Sham group, classical hepatic lobules were not distinctly discernible, with central veins and portal areas appearing randomly arranged compared to healthy adult livers. Moreover, the portal areas lacked conspicuous connective tissue, hindering the recognition (Fig. 1A, 1B). Evidence of ongoing hemopoiesis was observed, characterized by dispersed hematopoietic cell groups (Fig. 1A). Similarly, sections from 2-week-old rats in the IUGR group exhibited indistinct classical hepatic lobules with some areas devoid of central veins and the elements of the portal triad including portal vein, portal artery, and bile duct. These areas were occupied by rows of hepatocytes (Fig. 1C). Prominent vacuolization and perinuclear edema were detected in these sections. Additionally, larger hematopoietic cell groups were dispersed compared to the Sham group (Fig. 1D).
In the sections from 4-week-old rats in the Sham group, liver histology exhibited greater similarity to that of healthy adult liver. Classical hepatic lobules were discernible, and the connective tissue elements of portal areas, although not abundant, were detectable (Figs. 2A, B). Hemopoiesis was still ongoing within small hematopoietic cell groups (Fig. 2A). Similarly, sections from 4-week-old rats in the IUGR group displayed discernible classical hepatic lobules (Figs. 2C, D), although some areas were still solely occupied by rows of hepatocytes without central veins or portal spaces. Notably, perinuclear edema was obvious in the cytoplasm of the hepatocytes located in these areas (Fig. 2C). Hemopoiesis persisted, with larger hematopoietic cell groups compared to Sham (Fig. 2D).
In the sections from 8-week-old rats in the Sham group, liver histology closely resembled to that of a healthy adult liver. Classical hepatic lobules were easily identifiable, and the connective tissue connective tissue of portal areas was detectable (Figs. 3A, B). Similarly, sections from 8-week-old rats in the IUGR group displayed predictable classical hepatic lobules, yet some areas remained occupied by only rows of hepatocytes with no sign of central vein or portal elements (Figs. 3C, D). The distribution of central veins and the portal areas appeared more random rather than typical lobule organization. Perinuclear edema was rarely detected (Fig. 3D).
The mean numbers of central veins and portal areas were non-uniform between the Sham and IUGR groups indicating disruption of classical hepatic lobules (Table I). IUGR altered the mean numbers of central veins and portal areas. Specifically, the mean number of the central veins of the 4- and 8-week-old IUGR group was higher than that of the 4- and 8-week-old Sham group (p < 0.001, p < 0.05; respectively). Additionally, the mean number of portal areas in the 2-week-old IUGR group was higher than that in the 2-week-old Sham group (p < 0.001).
1.3. Immunohistochemical analysis: The distribution of Ki-67 +, caspase-8 +, and OV-6 + cells appeared similar across all developmental weeks in both the Sham and IUGR groups. However, statistical analysis revealed significant differences in cell counts between corresponding Sham and IUGR groups (Table I). Specifically, the mean number of Ki-67 + cells in the 8-week-old rats from the IUGR group was significantly higher than that of the Sham group (p < 0.05). Additionally, the mean number of caspase-8 + cells in the IUGR group was higher than those in the Sham group at 2, 4, and 8th weeks of age (p < 0.05, p < 0.05, p < 0.001; respectively). OV-6 cytoplasmic immunopositivity was consistently detected in the epithelium of the bile ducts across all groups, additionally occasionally in the cytoplasm of the hepatocytes in all groups. However, due to strong reactivity obscuring the exact boundaries of epithelial cells, reliable counting of OV-6 + cells were hindered.
1.4. Western blotting analyses: The tissue RIP3/GAPDH level in the 2-week-old IUGR group was found to be lower than that in the 2-week-old Sham group (p < 0.005). Conversely, the tissue NF-kB level of the 2- and 8-week-old IUGR groups was lower than the corresponding Sham groups (p < 0.005, p < 0.001; respectively). However, the NF-kB level in the 4-week-old IUGR group was higher than that in the 4-week-old Sham group (p < 0.05) (Table I). These findings suggest alterations in necroptosis and inflammation patterns between the equivalent weeks of the Sham and IUGR groups, underscoring potential differences in underlying molecular mechanisms associated with IUGR.
1.5. Biochemical analyses of the serum samples: Biochemical analysis of the serum samples from 2-week-old rats was precluded due to insufficient sample volume. However, significant differences were observed in the levels of IDH and MDH between Sham and IUGR groups at 4 and 8 weeks of age. Specifically, the mean IDH level in the 4-week-old IUGR group was higher than that in the 4-week-old Sham group (p < 0.05). Conversely, the mean IDH level in the 8-week-old IUGR group was lower than that in the 8-week-old Sham group (p < 0.05). Additionally, the mean MDH level in the 8-week-old IUGR group was lower than that in the 8-week-old Sham group (p < 0.05) (Table I).
1.6. Biochemical analyses of the tissue samples: Significant differences were observed between IUGR and Sham groups regarding mean tissue levels of TOS, TAS, TThiol, NThiol, glutamate dehydrogenase, glucose-6-phosphate dehydrogenase, NAD-isocitrate dehydrogenase (NAD-IDH), NADP-isocitrate dehydrogenase (NADP-IDH), fumarase, Zn, and Cu (Table I).
1.6.1. Tissue TOS and TAS levels: No significant changes in terms of the mean TOS levels between the IUGR and Sham groups were detected. The mean TAS level of the 2-week-old IUGR group was lower than that of the 2-week-old Sham group (p < 0.05), however, the mean TAS levels of the 4- and 8-week-old IUGR groups were not significantly different from those of the corresponding Sham groups.
1.6.2. Tissue TThiol and NThiol levels: The mean TThiol level in the 2-week-old IUGR group was found to be lower than that of the 2-week-old Sham group (p < 0.05), while the level in the 4-week-old IUGR group was higher than that in the 4-week-old Sham group (p < 0.05). The mean NThiol level in the 2-week-old IUGR group was lower than that in the 2-week-old Sham group (p < 0.005), whereas the level in the 8-week-old IUGR group was higher than that in the 8-week-old Sham group (p < 0.05).
1.6.3. Tissue glutamate dehydrogenase levels: The mean GDH level in the 4-week-old IUGR group was higher than that in the 4-week-old Sham group (p < 0.05).
- Tissue glucose-6-phosphate dehydrogenase levels: The mean G-6-P dehydrogenase level in the 2-week-old IUGR group was lower than that in the 2-week-old Sham group (p < 0.05), while the level in the 4-week-old IUGR group was higher than that in the 4-week-old Sham group (p < 0.05).
1.6.4. Tissue isocitrate dehydrogenase levels: The mean NADP-IDH level in the 2-week-old IUGR group was lower than that in the 2-week-old Sham group, while the level in the 4-week-old IUGR group was higher than that in the 4-week-old Sham group (p < 0.05). Additionally, the mean NAD-IDH level in the 8-week-old IUGR group was higher than that in the 8-week-old Sham group (p < 0.05).
1.6.5. Tissue fumarase levels: The mean fumarase level in the 2-week-old IUGR group was lower than that in the 2-week-old Sham group (p < 0.001), while the levels in the 4- and 8-week-old IUGR groups were higher those in the corresponding Sham groups (p < 0.05).
1.6.6. Tissue zinc and copper levels: The mean zinc level in the 2-week-old IUGR group was higher than that in the 2-week-old Sham group (p = 0.001), whereas the copper level in the 2-week-old IUGR group was lower than that in the 2-week-old Sham group (p < 0.05). The mean zinc level in the 4-week-old IUGR group was lower than that in the corresponding Sham group (p < 0.005). Additionally, the mean zinc and copper levels in the 8-week-old IUGR group were higher than that in the 8-week-old Sham group (p < 0.05, p < 0.005; respectively).
1.7. Metabolomics Analysis: Upon examining the results of the metabolomics in the 2-week-old IUGR and Sham groups, significant alterations were uncovered (Table I). Illustrated in the volcano plot graph (Fig. 4A), the IUGR group exhibited elevated levels of succinate, acetoacetate, isoleucine, cytidine, and formate, while displaying decreased levels of citrate, isocitrate, glucose, glucose-1-phosphate, maltose, and homocysteine to the Sham group. Utilizing Principal Component Analysis (PCA), which is adept at discerning patterns within high-dimensional data, the complexity of the data was effectively mitigated. The PCA score plots derived from NMR-analyzed tissue samples convincingly demonstrated group segregation and intra-group data clustering, accounting for approximately 63.15% of the variables. As depicted in Fig. 4B, the first principal component (PC1) accounted for 47.3% of the variance, while the second principal component (PC2) accounted for 15.8% of the variance. Further insights were gleaned through orthogonal partial least square data analysis (OPLS-DA), a method facilitating detailed examination of multivariate data. The OPLS-DA graph (Fig. 4C) affirmed successful differentiation between the IUGR and Sham groups. Additionally, the VIP score graph (Fig. 4D) showcased metabolites such as succinate, glucose, and acetoacetate as pivotal in distinguishing between groups. These findings underscore the utility of metabolomics in unraveling metabolic distinctions associated with IUGR.
Upon analyzing tissue samples extracted from 4-week-old rat livers, notable disparities emerged in the levels of acetoacetate, ethanol, pyruvate, cadaverine, and ethanolamine, which were found to be significantly elevated in the IUGR group compared to the Sham group (Table I). The volcano plot graph depicting statistically significant differential metabolites and fold change is illustrated in Fig. 5A. While a distinct separation was not readily apparent through PCA (Fig. 5B), the application of OPLS-DA proved instrumental in delineating the IUGR and Sham groups based on their metabolic profiles (Fig. 5C). OPLS-DA results depicted in Fig. 5C underscore the efficacy of this analysis in discerning the metabolic distinctions between the two groups. Figure 5D highlights the metabolites most closely associated with this discrimination.
Upon analyzing the NMR dataset obtained from 8-week-old rats, a distinct group of metabolites including acetoacetate, niacinamide, valine, phenylalanine, glutamate, cytidine, glucose, uracil, cadaverine, isoleucine, 1-methyhistidine, methionine, glucuronate, glycine, and fumarate exhibited an exclusive increase in the IUGR group (Fig. 6A) (Table I) Despite the lack of a clear differentiation between the groups through PCA (Fig. 6B), successful distinction was achieved through OPLS-DA (Fig. 6C). The OPLS-DA illustrated a discernible separation, indicative of distinct metabolic profiles between the IUGR and Sham groups. Further examination revealed the metabolites contributing most significantly to this observed separation (Fig. 6D), underscoring their pivotal role in discriminating between the IUGR and Sham groups. These findings highlight the efficacy of OPLS-DA in uncovering metabolic distinctions associated with IUGR and identifying potential biomarkers warranting further investigation.
1.8 Meta-analysis: To conduct the meta-analysis, all samples were stratified based on their respective groups: IUGR and Sh. The meta-analysis involved samples from animals at three different weeks of age (Table I). Upon meta-analysis of the values of the 2-week-old group, 25 significantly different metabolites were identified. In both the 4-week-old and 8-week-old groups, two metabolites exhibited significant differences (Fig. 7A). Of these metabolites, a total of 25 metabolites were detected specifically in the 2-week-old group, with 23 being unique being unique to this age group. The identified metabolites in the 2-week-old group included 3-amino isobutyrate, acetate, acetoacetate, aspartate, betaine, cadaverine, citrate, cytidine, ethanolamine, formate, glucose, glucose-1-phosphate, glutamate, glycine, homocysteine, isocitrate, maltose, phenylalanine, proline, pyruvate, succinate, threonine, tyrosine, uracil, and valine. Among these metabolites, acetoacetate emerged as the most distinct potential metabolite for selecting IUGR independently of the animal’s age. Its discriminatory ability remained consistent across different age groups, indicating its potential as a reliable marker for distinguishing IUGR from non-IUGR conditions. S-adenosylhomocysteine was identified as another distinctive metabolite in the 4-week-old group. Notably, valine also exhibited predictive potential as a marker in 8-week-old animals. The expression patterns of acetoacetate and valine comparing all 3 groups are depicted in Fig. 7B.
Following the comprehensive analysis of merged data from three different age groups, a volcano plot graph depicting differential metabolite levels between the IUGR group, and the Sham group was generated (Fig. 8A). Notably, acetoacetate, isoleucine, succinate, valine, niacinamide, formate, and pyruvate exhibited increased levels in the IUGR groups compared to the Sham groups. Conversely, serine, nicotinurate, maltose, homocysteine, and citrate levels were decreased in the IUGR groups versus the Sham groups. The heatmap presented in Fig. 8B illustrates the relative abundance of metabolites across the IUGR and Sham groups. Each row represents a metabolite, while each column corresponds to a sample. The color gradient in the heatmap reflects the magnitude of the variable’s abundance with red shades indicating higher values and blue shades indicating lower values. Although PCA did not yield a clear distinction between the groups (Fig. 9A), a discernible differentiation was successfully achieved through OPLS-DA (Fig. 9B). Further analysis pinpointed the metabolites contributing most significantly to the observed separation (Fig. 9C).
1.9. Enrichment and pathway analysis: With the comprehensive dataset analysis, data of all samples were evaluated separately. Figure 10A presents the top 25 biological processes that were found to be most significantly affected in the enrichment analysis obtained with merged data. This analysis employs statistical methods to assess the relationship between metabolites within the data set and relevant pathways. The identified biological processes represent key functional categories that exhibit significant associations with the metabolite data. The inclusion of these processes provides valuable insights into the potential biological mechanisms underlying the observed metabolic changes. Enrichment analysis revealed ketone body metabolism, butyrate metabolism, and methionine metabolism as the most influenced pathways. The most affected pathways and related metabolites revealed by the pathway analysis are depicted in Figs. 10A and 10B.
The visualization of results and enrichment analysis for the subgroups is presented in Fig. 11, Table II. Figure 11A highlights the most affected pathways including the oxidation of branched-chain fatty acids, phytanic acid peroxisomal oxidation, and valine leucine and isoleucine degradation in the 2-week-old group. These pathways demonstrate significant associations with the metabolite profiles observed in this group, indicating notable alterations in these metabolic pathways during early development. In the 4-week-old group betaine metabolism, tryptophan metabolism, and ubiquinone metabolism were found to be the most affected pathways (Fig. 11B). In the 8-week-old group, propanoate metabolism, valine, leucine, and isoleucine degradation, and ketone body metabolism were detected as the most influenced pathways (Fig. 11C). These pathway perturbations highlight the distinctive metabolic signatures present in the aged 8-week group. Overall, the visualization of these results offers valuable insights into the specific pathway dysregulations within each age group, contributing to a comprehensive understanding of the metabolic alterations in this study and providing potential targets for further investigation into mechanisms underlying IUGR.