In this study, we examined how four macronutrient-rich dietary regimens high-fat (HFD), high-carbohydrate (HCD), high-protein (HPD), and cafeteria diet (CafD) affect pubertal programming in weaned female Wistar rats through modulation of hypothalamic miRNAs. By integrating small RNA transcriptomics with qPCR validation of key neuroendocrine genes (Kiss1, Gnrh1, Mkrn3, Zeb1, Tbx21, Cebpb, Mapk14), serum hormone profiling (LH, FSH, estradiol), and ovarian histology, we sought to construct a comprehensive mechanistic framework linking diet-induced miRNA shifts to phenotypic markers of pubertal onset. We further employed in-silico target prediction, functional enrichment, and miRNA-mRNA docking analysis to elucidate the regulatory roles of differentially expressed miRNAs. We hypothesized that distinct macronutrient exposures would differentially reprogram hypothalamic miRNA expression, influencing pubertal timing through neuroendocrine pathways. Additionally, the miR-30b/MKRN3 pathway has already been identified as a critical regulator of pubertal timing in both animal and human studies18,19. Building upon this established foundation, our study extends these findings by investigating how macronutrient-specific dietary exposures can modulate this pathway, thereby linking nutritional environments with miRNA-regulated neuroendocrine mechanisms of puberty.
Female Wistar rats (PND 18, 24 ± 0.9 g) were procured from Sync Bio Research Pvt. Ltd. and housed under controlled environmental conditions (22 ± 1 °C, 50-60% humidity, 12:12 h light-dark cycle) with ad libitum access to food and water. All experimental procedures involving animals were reviewed and approved by the Institutional Animal Ethics Committee (IAEC) of the Maharaja Sayajirao University of Baroda, under approval number MSU-Z/IAEC08/13-2024 and complied with ARRIVE guidelines.
Animals (n = 20 per group) were randomly assigned to one of five dietary regimens from PND 21 to 42: standard chow (control), HFD, HCD, HPD, and CafD, a palatable, energy-dense mixture of snack foods (cookies, chips, sweetened milk) mimicking Western diets. Detailed macronutrient compositions are provided in Supplementary Tables 1-5. Daily caloric intake per rat was calculated based on diet-specific consumption and energy density. Body weights were recorded every 7 days and pubertal onset was assessed via daily monitoring of vaginal opening (VO) from PND 25 onward, with VO age recorded and compared across groups, animals were euthanized on PND 42 via intraperitoneal administration of sodium pentobarbital (200 mg/kg) in accordance with the AVMA Guidelines for the Euthanasia of Animals (2020).
Blood samples were collected via retro-orbital puncture on PND 28, 35, and 42. Serum was separated (n = 6) and immediately processed for measuring the levels of follicle-stimulating hormone (FSH) (EA0015Ra, Bioassay Technology Laboratory, Jiaxing, Zhejiang, China), luteinizing hormone (LH) (EA0013Ra, Bioassay Technology Laboratory, Jiaxing, Zhejiang, China), and estradiol (E-OSEL-R0001, Elabscience, Texas, USA) were measured using commercially available ELISA kits, following manufacturer protocols; The sensitivity of kits for LH (0.31mIU/ml), Estradiol (1.17pg/ml) and FSH (0.022mIU/ml) as mentioned on the manufacturer's datasheet. All samples were run in triplicate. For downstream molecular analyses, six animals per group were randomly selected for RNA sequencing and qPCR validation, this allocation was performed at random to minimize potential bias.
Total RNA was extracted from frozen hypothalamic tissue (40 mg) using the Direct-Zol™ RNA MiniPrep Kit (R2050, Zymo Research, USA) according to the manufacturer's protocol. Dissections were guided by stereotaxic landmarks to isolate the medio basal hypothalamus enriched for the ARC, with tissues pooled within groups to ensure adequate RNA yield (n = 6). RNA quantity and purity were assessed using a NanoDrop 2000 spectrophotometer (ND-2000, Thermo Fisher Scientific, USA), and integrity was confirmed using the Agilent 4200 TapeStation (G2991BA, Agilent technologies, USA) with high sensitivity RNA ScreenTape. Only samples with RNA Integrity Numbers (RIN) > 6 were selected for library preparation.
Small RNA libraries were constructed using 1 µg of high-quality RNA per sample and the NEBNext Multiplex Small RNA Library Prep Kit for Illumina (E7300L, New England Biolabs, USA). The protocol involved sequential ligation of 3' and 5' adapters, reverse transcription, and PCR amplification with indexed primers for multiplexing. Adapter-ligated products (120-190 bp) were size-selected using AMPure XP beads (A63880, Beckman Coulter, USA). Library quality was validated via TapeStation analysis to confirm fragment size distribution and absence of adapter dimers. Quantification was performed using the Qubit Fluorometer (Q33238, Thermo Fisher Scientific, USA), and libraries were pooled equimolarly. Sequencing was conducted on the Illumina NextSeq500 platform using single-end 1 × 75 bp reads, targeting ~ 10 million reads per sample.
Raw reads from the Illumina NextSeq500 platform were trimmed using Cutadapt and Trimmomatic to remove adapters, poly(A) tails, low-quality bases (Phred < 20), and ambiguous nucleotides. Reads of 18-30 nucleotide was retained and aligned to the Rattus norvegicus genome (mRatBN7.2, Ensembl) using Bowtie v1.3.1 (https://bowtie-bio.sourceforge.net/index.shtml). Non-miRNA reads (e.g., tRNA, snoRNA, snRNA, piRNA) were filtered using Rfam and Repbase. miRNAs were annotated and quantified using miRDeep2. Known and novel miRNAs were identified based on miRBase alignment and miRDeep2 score (≥ 4) with significant randfold p-values. Expression values were normalized as reads per million (RPM). Differential expression was analysed using DEGseq (Likelihood Ratio Test), with p < 0.05 as the threshold for significance, and Benjamini-Hochberg false discovery rate (FDR) correction was applied to control for multiple testing across all miRNAs. Heatmaps and volcano plots were generated in Rstudio (version: 2025.05.1 + 513).
Target prediction was performed using miRanda with strict 5' seed matching and energy cutoff of - 25 kcal/mol. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment of predicted targets was conducted using ClusterProfiler, focusing on pathways related to hypothalamic signalling and pubertal regulation. Raw data are available in the NCBI SRA (PRJNA1240077) under BioSample accessions SAMN47502131-SAMN47502135. The schematic diagram for this pipeline is detailed in supplementary Fig. 1.
To validate small RNA-seq findings, RT-qPCR was performed on hypothalamic RNA using the miScript II RT Kit (218160, Qiagen, Germany) and SYBR Green PCR Kit (A25741, Applied Biosystems) with miRNA-specific stem-loop primers. Reactions were run on a QuantStudio Real-Time PCR System (4369074, Applied Biosystems), and U6 snRNA was used as an internal control. The thermal cycling conditions were as follows: initial denaturation at 95 °C for 10 min, followed by 40 amplification cycles of 95 °C for 15 s and 60 °C for 60 s. Relative expression of selected miRNAs (miR-30b, miR-29a, miR-375, miR-137, miR-199a, miR-155) was calculated using the 2^-ΔΔCt method.
For mRNA quantification, cDNA was synthesized using the High-Capacity cDNA Reverse Transcription Kit (4374967, Thermo Fisher Scientific, USA), followed by qPCR with PowerUp SYBR Green Master Mix using the Step One PCR (4369074, Applied Biosystems, USA). RT-qPCR was performed using the following thermal cycling conditions: an optional UDG activation step at 50 °C for 2 min, initial denaturation at 95 °C for 2 min, followed by 40 amplification cycles of denaturation at 95 °C for 15 s, annealing at 57-60 °C for 15 s, and extension at 72 °C for 1 min. Target genes included Mkrn3, Kiss1, Gnrh, Tbx21, Zeb1, Cebpb, Mapk14 (p38), Sirt1, Ampk and mTOR normalized to GAPDH. All reactions were run in technical triplicates with no-template controls to ensure specificity. List of primers are detailed in supplementary Tables 6-7.
Ovarian tissue (n = 6) Sections (5 μm thick) were harvested and deparaffinized in xylene and rehydrated through a graded ethanol series. Sections were stained with haematoxylin for 5 min, differentiated in 1% acid alcohol, and blued in ammonia water. Subsequently, sections were counterstained with eosin for 2 min, dehydrated in ascending ethanol concentrations, cleared in xylene, and mounted with DPX. This standard H&E staining technique was employed to assess the general histoarchitecture of the ovarian tissue. For follicle counting, every 5th section was considered for quantitative analysis. Follicles were staged according to established morphological criteria. To avoid repeated counting of the same follicle across adjacent sections, only follicles with a visible oocyte nucleolus were included. Follicles were reported as absolute counts per ovary (mean ± SEM). The observer was blinded to group allocation during analysis.
The thermodynamic stability of miRNA-mRNA interactions was assessed using RNAfold (http://rna.tbi.univie.ac.at/cgi-bin/RNAWebSuite/RNAfold.cgi), which predicts secondary structures and minimum free energy (MFE) values. Mature miRNAs and 3'UTR target sequences were sourced from miRBase and Ensembl. More negative MFE values indicated greater duplex stability, supporting the likelihood of post-transcriptional regulation.
To evaluate interactions between selected miRNAs and their mRNA targets, docking simulations were performed using the HADDOCK 2.4 web server (https://wenmr.science.uu.nl/haddock2.4).Input sequences of mature miRNAs and corresponding mRNA 3'UTRs were retrieved from public databases. Docking involved energy minimization and refinement, and complexes were scored using HADDOCK score, Z-score, and RMSD. Top-ranked interactions, characterized by low HADDOCK scores and favourable Z-scores, were visualized using BIOVIA Discovery Studio (https://www.3ds.com/products/biovia/discovery-studio).
Normality of data distribution was assessed using the Shapiro-Wilk test. Group comparisons for variables such as body weight, age at vaginal opening, and gene expression (RT-qPCR) were performed using one-way analysis of variance (ANOVA) followed by Tukey's multiple comparisons test. For serum hormone levels (LH, FSH, and estradiol), a two-way ANOVA was conducted to evaluate the effects of both dietary group and time point (PND 28, 35, and 42), followed by Bonferroni's post hoc test. All statistical analyses were conducted using GraphPad Prism version 9.0 (GraphPad Software, USA), and data are presented as mean ± standard error of the mean (SEM). A p-value < 0.05 was considered statistically significant.