3.1 Placental Lesions were observed in Dex-induced Cleft Palate in Rabbits
After dexamethasone treatment, approximately 60% of rabbit embryos developed cleft palates, contrasting with the 40% embryos with intact palates (Fig. 2A). Generally, the degree of fetal and placental malformations was not consistent among three groups, and the D-CP had the highest degree of fetal and placental malformations. Histological examination using H&E staining revealed a significant increase in overall placental thickness for both D-CP and D-NCP groups, particularly in the decidua basalis layer (p < 0.01) (Fig. 2B). Pathological changes such as fibrosis, calcification, necrosis, and an increase in immature chromaffin cells were noted in the D-NCP and D-CP groups but absent in controls (Fig. 2C-D). Additionally, Masson staining indicated severe fibrosis in both D-NCP and D-CP groups, contrasting with controls (Fig. 2E).In terms of placental and embryonic weights, significant differences were observed. The placentas in the D-NCP group were notably heavier than those of the control group (p < 0.01), while the D-CP group’s placentas showed no significant weight difference from the controls. Conversely, control group embryos were heavier compared to both D-CP and D-NCP groups (p < 0.01), yet there was no discernible difference in embryo weight between these two Dex-treated groups. Placental efficiency, measured as the ratio of embryo weight to placenta weight, was higher in controls compared to the other groups (p < 0.05), but similar between the D-NCP and D-CP groups (Fig. 2E).
3.2 Transcriptomic profiling of placenta
In placenta, there were 4,744 differentially expressed genes (DEGs) (q-values < 0.05,| log2 (fold change) | >1) between the D-CP and control groups, of which 2,574 were up-regulated, and 2,170 were down-regulated in the D-CP group (Fig. 3A). There were 1118 DEGs (q-values < 0.05,| log2 (fold change) | >1) between the D-NCP and control groups, with 829 up-regulated and 289 down-regulated in the D-NCP group (Fig. 3B). There were 885 same genes between the above two DEGs sets (Fig. 3C). To discern the common pathological pathways in the placenta affected by dexamethasone, enrichment analyses were performed on the 885 DEGs with the Metascape database. The results show hormonal responses, blood vascular development, wound responses, inflammatory reactions, and responses to decreased oxygen levels was significantly enriched (Fig. 3D). Furthermore, the key driving genes (KDG) analysis pinpointed 10 genes – CD44, SPP1, ITGB1, CCN2, MMP10, COL6A6, TNR, TNC, TNN, and PTGER4 – as playing pivotal roles in the observed genetic alterations (Fig. 3E).
To explore the source of the phenotypic varieties in fetus after DEX-treated, the transcriptomics differences of placenta between the D-CP and D-NCP groups were further analysed. There were 97 DEGs (q-values < 0.05,| log2 (fold change) | >1), of which 54 were up-regulated and 43 were down-regulated in the D-CP group compared to the D-NCP group (Fig. 3F). The Gene Ontology (GO) enrichment results of the 97 DEGs included regulation of angiogenesis, fatty acid metabolic process, regulation of cell junction assembly, response to glucocorticoids, regulation of vasculature development, response to nutrient levels, and regulation of excretion (Fig. 3G). The Disease Ontology (DO) enrichment results show hypercholesterolemia, carotid artery diseases, inflammation and maternal hypertension, showed were well enriched (Fig. 3H).
3.3 Metabolomic profiling of placenta
We employed Partial Least Squares Discriminant Analysis (PLS-DA) to distinguish the metabolite features in our study. In the placenta, there were 113 differentially expressed metabolites (DEMs) between the Dexamethasone-Treated Cleft Palate (D-CP) group and the control group (q-value < 0.05, | log2(fold change) | > 1), with 42 being upregulated and 71 downregulated in the D-CP group (Fig. 4A). Between the Dexamethasone-Treated Non-Cleft Palate (D-NCP) group and the control group, there were 92 differentially expressed metabolites (q-value < 0.05, | log2(fold change) | > 1), with 54 upregulated and 38 downregulated in the D-NCP group. And further differentiation of the metabolic features between the D-NCP and D-CP groups was achieved using Orthogonal Partial Least Squares Discriminant Analysis (OPLS-DA). In the placenta, there were 101 significant differential metabolites (VIP > 1, p < 0.05) identified between these two groups. Notable among these were lower levels of alanine, glutamine, L-aspartic acid, and taurine in the D-CP group, while cortisol, xanthosine, and lysoPC levels were higher (Fig. 4C). The pathways enriched in differences between the D-CP and control groups included Purine Metabolism and Arachidonic Acid Metabolism, as well as Galactose Metabolism and Aspartate Metabolism. The pathways that were differentially enriched between the D-CP and D-NCP groups included the Urea Cycle. Commonly highly enriched metabolic pathways between these groups were the Urea Cycle, Aspartate Metabolism, Nicotinate and Nicotinamide Metabolism(Fig. 4B).
3.4 Metabolomic profiling of amniotic fluid
In the amniotic fluid, there were 231 differentially expressed metabolites (DEMs) between the D-CP group and the control group (q-value < 0.05, | log2(fold change) | > 1), with 127 being upregulated and 104 downregulated in the D-CP group. Between the D-NCP group and the control group, there were 248 differentially expressed metabolites (q-value < 0.05, | log2(fold change) | > 1), with 118 upregulated and 130 downregulated. And we identified 34 differential metabolites (VIP > 1, p < 0.05), including D-glucose, taurine, and arginine, which decreased in normalized abundance among the control, D-NCP, and D-CP groups. Additionally, 59 differential metabolites (VIP > 1, p < 0.05), including xanthosine, leukotriene C4, and norepinephrine, were found to increase in normalized abundance among these groups. A comparison in the amniotic fluid revealed 28 differential metabolites (VIP > 1, p < 0.05) between the D-CP and D-NCP groups (Fig. 4D), with significant decreases in pyridoxal 5’-phosphate and PC levels in the D-CP group. The pathways enriched in differences between the D-CP and control groups included Glycine and Serine Metabolism and Spermidine and Spermine Biosynthesis. The differentially enriched pathways between the D-CP and D-NCP groups included Spermidine and Spermine Biosynthesis and Beta-Alanine Metabolism. Commonly highly enriched metabolic pathways in these groups were Glycine and Serine Metabolism, Spermidine and Spermine Biosynthesis, and Methionine Metabolism(Fig. 4B).
3.5 Increased arginase activity in urea cycle is involved in placental pathology
Metabolomics and transcriptomics data were combined analysed that differential metabolites and genes sharing a pathway were identified. The urea cycle emerged as a potential target pathway implicated in dexamethasone-induced placental lesions. Metabolomic data for the placenta showed that several relevant metabolites in the urea cycle were altered after dexamethasone treatment, in particular urea was significantly elevated in the D-CP and D-NCP groups compared to controls (p < 0.05) (Fig. 5A). Interestingly, in the amniotic fluid, the levels of urea and arginine were significantly decreased in the D-NCP and D-CP groups compared to the controls (p < 0.01) (Fig. 5A). The expression of ARG1 and GLB1 in transcriptomics was also significantly increased in D-CP group compared to control group (p < 0.05) (Fig. 5A). The qRT-PCR results showed that the ARG1 and GLB1 mRNA were significantly increased in the D-NCP and D-CP groups compared to the control group (p < 0.05) (Fig. 5H, I). Immunofluorescence results showed that arginase-1 and β-galactosidase expression was increased after dexamethasone treatment in the decidua basali of the placenta (p < 0.05) (Fig. 5B-G). TUNEL staining showed that there was a significant increase in the numbers of apoptotic cells in the placenta of the D-CP and D-NCP groups compared to the control group (p < 0.05) (Fig. 5J-L).