Introduction

In the last decades, public health and preventive medicine improvements have contributed to more people reaching old age. Nevertheless, many elderly individuals face chronic illnesses that require specialized care. Identifying adjustable risk factors like unhealthy eating, lack of exercise, smoking, stress, and sleep deprivation can lead to a healthier lifestyle, providing an affordable public health strategy for a better quality of life in older age1,2. Intrauterine and early postnatal experiences can shape health throughout a person’s life. David Barker suggested that prenatal development could affect postnatal health in his “Fetal Origin of Adult Diseases” theory. This idea evolved into the “Developmental Origins of Health and Disease” (DOHaD) framework, which considers pre-conception and early postnatal life as critical periods of susceptibility3,4, which may interfere with health or disease during aging.

One model associated with this concept is Maternal Protein Restriction (MPR), which affects many systems and organs. Epidemiological studies elucidated that maternal malnutrition causes a decrease in birth weight, which is a potential factor in the development of chronic diseases with aging3,4. In this context, the prostate is an accessory gland of the male genital system that can be important for reproduction, however, has been recognized for the high incidence of diseases with aging, mainly benign prostatic hyperplasia and PCa5,6, especially the ventral prostate (VP), which is more responsive to hormones.

Several models with male rats elucidated that exposure to severe MPR leads to PCa oncogenic pathways in early life and aging, demonstrating the potential biomarkers of cancer already at the beginning of life7,8. Experimental studies showed that MPR decreases kidney function and increases blood pressure9, reduces insulin, testosterone, and albumin levels in early life7, and causes prostate cancer (PCa) with aging7,8. This condition highlights the severe adverse effects of MPR on hormone production, reproductive factors, and prostate disorders. Epigenetics is a significant factor related to this illness since it involves the effects of early-life exposure on the occurrence of chronic noncommunicable diseases in offspring and future generations10.

Among these mechanisms, the small non-coding RNAs (sRNAs) are key players in the posttranscriptional regulation processes by targeting transcripts. Specifically, the PIWI-Interecting RNAs (piRNAs) are 24–35 nucleotides long processed from precursor transcripts, the piRNA clusters11. These processed small molecules are loaded into PIWI proteins to form the RNA-induced silencing complex (RISC)12 piRNA-RISC captures and slices transposable element transcripts, which are “jumping-genes”, able to create copies of themselves and insert them elsewhere in the genomes. The action of piRNAs controls the TE mobilization, and that is why it is essential to germinative cells, guaranteeing genome integrity13. Although most studies describe the role of epigenetic mechanisms associated with embryonic development and cancer biology, the relationship between MPR, piRNAs, and PCa yet remains to be investigated.

In addition to germinative cells, piRNA expression has also been observed in somatic tissues, such as the prostate, and in certain types of cancer14,15. In this study, we employed small RNA sequencing (sRNAseq) to investigate the presence of piRNA pathways during early life and aging in prostate rats subjected to MPR. Our objective was to characterize piRNAs in the VP of male Sprague Dawley rats exposed to MPR during pregnancy, lactation, early life, and aging. We confirmed the expression of piRNA sequences in the rat prostate and linked this mechanism to dysregulation associated with aging. These findings provided a foundation for pioneering research into the role of piRNA clusters in the prostate of rats exposed to MPR.

Materials and methods

Animals and experimental design

The detailed experimental design is described by Santos et al.7. Briefly, after the determination of pregnancy on gestational day 1 (GD1), pregnant rats were distributed into two experimental groups (n = 6/group): Control (CTR): dams fed a normal protein diet (17% protein) and gestational and lactational low protein (GLLP): dams fed a low protein diet (LPD) during gestational and lactational periods. Normal and LPD diets were provided by PragSoluções (PragSoluções, SP, Brazil). All diets were isocaloric and normosodic7. The male offspring were euthanized on a postnatal day (PND) 21 (weaning) (n = 12/group) and PND 540 (n = 12/group). The offspring, which were euthanized on PND 540, had free access to a normal protein diet after weaning until the end of the experiment.

The animals were euthanized by an overdose of anesthesia (ketamine/xylazine) followed by decapitation and weighing, and the blood and ventral prostate (VP) were collected and processed by a different analysis as described below. The body weight, VP weight, and hormonal levels were analyzed using a Student t-test, and statistical differences were considered when p < 0.05. Throughout the experiment, housing and use of animals were performed accordingly with the appropriate guidelines and regulations. Efforts were made to minimize suffering and to reduce the animal numbers used in the experiments. The acquisition and description of data followed the recommendations set out in the ARRIVE guidelines.

Morphological analysis

VP samples were collected from the CTR and GLLP groups (n = 06 per group) at PND21 and PND540 and fixed in Methacarn solution (70% methanol + 20% chloroform + 10% acetic acid) for 4 h. They then underwent dehydration in ethanol, clearing in xylene, and inclusion in Paraplast (Sigma). For general morphological analysis, 5 μm histological sections were obtained and stained with Hematoxylin/Eosin. Maternal and offspring weight and size assessments were carried out individually, as already described in other works by the group16,17,18 Histological slides were evaluated for the animals’ prostate cells, stained with hematoxylin–eosin (HE) for an overview of glandular morphology.

Total RNA extraction and sequencing

RNA extraction was performed with Trizol (Ambion, USA), following the manufacturer’s instructions. Four samples from the CTR group and 4 samples from the GLLP group were used for RNA sequencing, at both ages. The samples were quantified using the Nanodrop and had their integrity inferred by the Bioanalyzer, keeping only samples with RIN (RNA Integrity Number) values greater than 8.

An aliquot of unfractionated total RNA was submitted for library construction and sequencing. Ribo-Zero was used during library preparation for rRNA depletion. Purification of small RNAs was carried out using the TruSeq Standard mRNA Sample Preparation Kit (Illumina) and TruSeq Small RNA Library Preparation Kit (Illumina), respectively, and construction of the libraries followed the manufacturer’s specifications. Sequencing was performed with the HiSeq Sequencing System. Macrogen (Seoul-South Korea) processed the entire library preparation and sequencing process, and data was downloaded via an HTTP link. The sequencing data is deposited in the Gene Expression Omnibus (GEO DataSets) under accession code GSE180674 (small non-coding RNAs).

Identification and organization of piRNA clusters in the Rattus norvegicus genome

The reading quality of the reads was accessed using the FastQC software (FastQC: A quality control tool for high throughput sequence data. Version 0.11.9, available online at: http://www.bioinformatics.babraham.ac.uk/projects/fastqc)19. Then, the sequences were filtered to remove adapters using the software Cutadapt (Version 3.5.0, available online at: https://cutadapt.readthedocs.io/en/stable/guide.html)20, resulting in the selection of reads with sizes between 24 and 35 nucleotides (Supplementary Table S1). Reads shorter than 24 or longer than 35 nucleotides were removed from the libraries. The prediction of piRNA clusters expressed in the sRNA libraries was performed using the program proTRAC (proTRAC: A software for probabilistic piRNA cluster detection, visualization and analysis)21 based on the rat reference genome (Gold standard version: Rattus_norvegicus mRatBN7.2), as well as the annotations of transposable elements obtained by RepeatMasker (Available at: https://www.repeatmasker.org/RepeatMasker/). HTML and FASTA files were generated for each predicted piRNA cluster and are available in Supplementary Data 1. Detailed information for each cluster, including coordinates, location, normalized hit counts, read numbers, and the sequences of individual piRNAs within the clusters, is provided in Supplementary Table S3. Testis samples from 90-day-old rats (Rattus norvegicus) available at GEO (SRP309332) were used as a comparative model for data found in the prostate (Supplementary Table S4). Then, the genomic coordinates of each piRNA cluster predicted in the prostate and testis were compared using the bedtools intersect package22.

Expression of Piwi genes

The primers were designed using the PrimerExpress 3.0.1® software (Life Technologies, USA) based on sequences from NCBI for Piwil1, Piwil2, and Piwil4 (Supplementary Table S5). The appropriate sample and primer concentration tests were carried out. Total RNA from rats was used, extracted from VP samples (n = 6/group) from both experimental groups at PND 21 and 540, using Trizol reagent (Thermo Fisher), according to the manufacturer’s instructions. The samples were treated with the DNase I Amplification Grade kit (Invitrogen Life Technologies, USA) to eliminate any contaminants from the genomic DNA. According to the manufacturer’s instructions, cDNA conversion was performed with the High-Capacity RNA-to-cDNA Kit (Life Technologies) with 2 μg aliquots in 10 μl reaction. The relative quantification of each gene was performed using the 2-∆∆CT method23. Finally, RT-qPCR reactions were performed in duplicates for each target gene on the QuantStudio 12 K flex Real-Time PCR System (Applied Biosystems, USA) in 96 wells per plate and normalized by the ratio between the reference genes (Gapdh and Gusb). After collecting the results, the appropriate statistical analyses were carried out.

Blast of known piRNA sequences against the clusters expressed in the prostate

We used gold-standard reference sequences from the piRBase database (http://bigdata.ibp.ac.cn/piRBase/index.php, last accessed in February 2024—Supplementary Data 2) to compare with the sequences of piRNAs clusters from our VP samples, using the BLAST (Basic Local Alignment Search Tool) bioinformatics tool24 in BLASTN 2.12.0+ 25 (BLAST Nucleotide) mode to identify similarities between nucleotide sequences (Supplementary Table S1).

Statistical analysis

Statistical analyses were performed in the RStudio® development environment (RStudio for Ubuntu version 9.3-2024) and using the ggplot2 package26. The results were subjected to the normalization test (Shapiro–Wilk), the Student’s t-test was applied, and the others were subjected to the Mann–Whitney test. Data were expressed as mean ± Standard Deviation (SD) and differences were considered statistically significant when p < 0.05.

Ethical approval

This study was performed in line with the principles of ethics. Approval was granted by the Biosciences Institute/UNESP Ethics Committee for Animal Experimentation (Protocol #1178).

Results

Development of the prostate of rats subjected to MRP

There was a delay in glandular development in the early life (postnatal day PND21), with a stromal and epithelial increase and lumen decrease (Fig. 1A,B) in the gestational and lactational low protein (GLLP) group. At PND540 (Fig. 1C,D), there was a luminal decrease and stromal increase in the GLLP group, and an increased incidence of carcinoma in situ, corroborating our previous data7. The morphological results (lumen size, stromal, and epithelial changes) have already been described in detail in other works by the group7,8.

Fig. 1
figure 1

Representative histological sections of the VP lobes from the CTR and GLLP groups on PND 21 and 540, stained with hematoxylin–eosin (HE). (A,C) Glandular growth in the GLLP group on PND 21 was impaired compared to the CTR. (B,D) At PND 540, the carcinoma in situ was highlighted by the dashed circle. S Stroma, L Lumen, E Epithelium, Scale bar: 500 μm.

Full size image

Characterization of piRNAs clusters in prostate and comparison to testis data

piRNA clusters are transcriptional active genomic regions that generate the piRNA precursors to be processed in mature piRNA sequences11. Based on the sRNAseq libraries, we predicted the piRNA clusters potentially expressed in prostate. In total, 123 clusters were predicted in prostate, 23 of which were expressed in the four groups. In young animals, 66 piRNA clusters were identified in the CTR group and 61 in the GLLP group. At PND540, 57 clusters were identified in the control group (CTR540), while 53 clusters were identified in the restricted group (GLLP540). Since piRNAs are well-documented in germ cells, we also compared the predicted clusters found in the prostate with those expressed in the testis to confirm that the characteristics of the predicted sRNA sequences align with piRNAs. A total of 20 clusters from testis are shared with the prostate samples (Fig. 2A). Initially, 7 clusters were detected in the 21-day groups, suggesting a direct impact on the animals’ development and growth. In contrast, 8 clusters were common to the 540-day groups, indicating that aging itself influences cluster expression. Additionally, 10 clusters were found exclusively in the GLLP540 group, linking the combined effects of maternal diet and aging. Another 14 clusters were uniquely identified in the four prostate groups, while 9 clusters were shared between both groups and tissues, highlighting essential clusters for prostate function and other key clusters present across all tissues.

Fig. 2
figure 2

piRNA clusters expressed in prostate from R. norvegicus. (A) The UpsetPlot represents the intersections between the clusters found in the prostate with the testis reference data. (B) Number of reads (standardized in Log10) of the sRNAs by the sequence length distribution.

Full size image

Similarly, we compared the distribution size of the piRNAs expressed in the prostate, miRNAs characterized from the same prostate samples27, and piRNAs generated from the testis28 to verify how the predicted sequences overlap with other sRNAs. miRNAs have a greater abundance of sequences of length between 21 and 23 nucleotides, while prostate and testis piRNAs have a greater abundance of sequences of 24 and 27 nucleotides, respectively (Fig. 2B). In short, this comparison showed that there was no overlap between miRNA from the same RNAseq, meaning that the sequences predicted in the cluster have a profile more similar to piRNA. With all this evidence, we can conclude that VP expresses piRNAs.

Impact of MPR and aging on prostate piRNAs

To identify similarities with previously described sequences, the predicted clusters in the prostate were blasted against known piRNAs from the testis, as deposited in piRBase (Fig. 3). A total of 54 piRNAs found in the prostate showed similarities to those in the database. Specifically, 3 known piRNAs (piR-2047424, piR-2122556, and piR-2131909) have similarity with clusters expressed in CTR21, suggesting an essential role during critical developmental periods. Additionally, 21 known piRNAs showed similarity with sequences in both the CTR21 and GLLP21 groups, indicating a role in early life and growth. A unique sequence in the GLLP21 group was similar to piR-140328, which may be related to the impact of diet. The CTR540 group shared one piRNA (piR-2163700) with the GLLP540 group, suggesting a potential link between aging and diet in regulating this specific piRNA. Lastly, 12 known piRNAs presented similarity with sequences expressed across both groups and age ranges, indicating their activity regardless of age or treatment. Additionally, distinct patterns of piRNA expression were observed across different groups. The Venn Diagram (Fig. 3) displays the distribution of piRNAs across the evaluated groups, highlighting both the piRNAs unique to each group and those shared between them. The cluster and chromosome location of all the known piRNAs can be found in Supplementary Table S1.

Fig. 3
figure 3

Known piRNAs sequences that presented similarity with the clusters predicted in the prostate. The overlapping means that the known piRNA is found simultaneously in more than one cluster from the different groups. The numbers represent the number of piRNAs shared or unique to each condition.

Full size image

Differences in piRNA cluster expression between samples

Based on the normalized reads expressed by cluster, we detailed the variation of the expression across the treatments (Fig. 4). For example, some clusters (IDs 59, 35 and 31) are constantly expressed across the conditions, representing the clusters found in the prostate despite the age or treatments. More specifically, cluster 35 is expressed during the early stage without restriction. Additionally, there are clusters specifically expressed in some groups. For instance, several clusters are only expressed in GLLP21 group, representing the early restriction effect, while other clusters are exclusively expressed during the late stages. Lastly, we detected some clusters that are more expressed in GLLP540, depicting the late effect of maternal restriction in the prostate. Overall, the variation in piRNA cluster expression across the samples suggests distinct regulation of piRNAs, potentially reflecting differences in the biological characteristics of the groups. The impact of differential piRNA expression in the prostate is further explored in the following section which provides an overview of putative piRNA targets (all intersections of piRNA clusters in the samples are available in Supplementary Table S6).

Fig. 4
figure 4

Heatmap of cluster expression across aging and under different experimental conditions. We showed only the sequences presenting at least 10 normalized reads of expression per cluster in at least one sample. The expression is shown in log scale, and 0 represents no expression. Each row depicts a piRNA cluster assigned with an ID that can be cross-checked in Table S6. The chromosome where a certain piRNA is located is also represented by the side column.

Full size image

Composition of piRNA clusters expressed in prostate

Usually, piRNA clusters originate from regions enriched by degenerated TEs. Contradictory, the composition of TEs can indicate the potential target of the piRNA cluster due to the sequence similarity29. Thus, we detailed the composition of piRNA clusters predicted in prostate to investigate their potential targets and TE regulation in this tissue. The rat genome is mainly composed of retrotransposons, such as SINE (45.93%), LTR (26.82%), and LINE (21.27%), while DNA transposons (5.53%) are the least abundant22. Specifically in the prostate, TEs of the order SINE confer half the composition of expressed piRNA clusters, followed by LTR (27.53%) and LINE (13.94%). The least abundant TEs in prostate-expressed clusters were DNA transposons (7.49%), similar to what is found in the entire genome. The piRNA clusters expressed in the testis have the highest composition of LTR (42.37%), followed by SINE (36.16%), LINE (17.51%), and the least abundant, DNA transposons (3.67%). In summary, the general composition of piRNA clusters expressed in the prostate follows the same pattern found in the testis. The differences reflect the number of clusters expressed in each group, and consequently, the abundance of specific TEs within these clusters. This indicates that retrotransposons are potential targets and may be more regulated (Fig. 5). Detailed information on transposable elements is in Supplementary Table S2.

Fig. 5
figure 5

Distribution of TEs present in the piRNA clusters expressed in testis and prostate. The diameter of the circles indicates the abundance of each TE in the piRNA clusters expressed in the samples. The squares highlight a lesser abundance of piRNA derived from retrotransposons L1, Alu, and ERVK expressed in older (PND 540) animals.

Full size image

Expression of PIWI genes in prostate samples

We also evaluated the expression of the main genes involved in the piRNA biogenesis. For instance, Riwi, Rili, and Rili2 are orthologues of PIWIL1, PIWIL2, and PIWIL4 in humans. Therefore, the expression of PIWI protein genes was evaluated in the prostate from both groups and ages. The expression of these genes was detected in both experimental groups, initially in young animals there was no difference in expression for any of them (Fig. 6A–C). In aged animals, there was an increase in Riwi expression in the restricted group. However, Rili expression decreased in the restricted group. The expression of Rili2 showed no differences in any of the groups. Therefore, these findings confirm the activity of these genes (essentially characterized in germinal tissues) in somatic tissues, in addition to highlighting the impact of aging and diet on the expression of these proteins.

Fig. 6
figure 6

Gene expression of PIWI proteins. (A) Riwi (PIWI1) expression at ages 21 and 540 days. (B) Rili (PIWI2) expression at PND 21 and 540. (C) Rili2 (PIWI4) expression at PND 21 and 540. Fold change is indicated by the lateral vertical bars. The asterisk indicates the statistical difference between samples and was applied to Student’s t-test (p < 0.05). The central bars indicate the standard deviation (SD).

Full size image

The Fig. 7 summarizes the workflow depicting the effects of maternal exposure to a low-protein diet on the modulation of piRNA cluster expression in the offspring VP. We further demonstrated differences in the VP piRNAs expression profile in young and old male offspring rats. Overall, these data highlighted the involvement of piRNAs in altering prostate developmental biology with long-lasting consequences for prostatic disorders in old male offspring rats.

Fig. 7
figure 7

We analyzed RNAseq data of the ventral prostate to provide new insights into the association of maternal malnutrition with the deregulation of piRNAs in male offspring rats. Moreover, we identified the differential expression profile of the VP piRNAs between young and old offspring. These analyses revealed that both aging and maternal malnutrition altered epigenetic markers involved in prostate developmental biology and aging, contributing to the increased incidence of prostate disorders in old male offspring rats.

Full size image

Discussion

There is an extensive discussion about how exposure to adverse conditions during the first stages of life can increase the propensity for the development of chronic diseases in offspring in adulthood. Our results demonstrated an increased incidence of carcinoma in situ in the GLLP group, demonstrating the potential of gene expression and epigenetic modulating mechanisms involved. Compiling epigenomic analyses of VP in early life and aging, we found a profile of unique and differentially found piRNA clusters in the prostate of young and old rats exposed to MPR, characterizing piRNAs as potential modulators of prostate carcinogenesis. These results show the epigenetic potential of maternal exposure to nutritional adversities, in addition to providing the characterization of piRNA in prostate tissue, being the first study to highlight the correlation of this pathway between prostate and MPR.

In this context, a global understanding of aspects relating to the DOHaD can provide essential biological and molecular aspects for the discovery of new prognostic and therapeutic targets7,30. These studies corroborate our findings, from biometric data, such as low offspring weight and smaller size visible on the first postnatal day8.

piRNAs are well characterized in the testis, where they regulate transposable elements to preserve genomic integrity. Deactivating piRNA clusters has been shown to cause problems in sperm maturation, even with subtle transcriptome changes14. It is crucial to understand the role of piRNAs in other tissues, including their potential link to diseases like prostate cancer31,32. Similar piRNA clusters found in both testis and prostate confirm that piRNAs are not exclusive to germinal tissue but are also present in other tissues in smaller amounts15. Recent studies have reported piRNAs in various cell types, including the mouse brain, somatic tissues, and human plasma, suggesting that piRNAs have broader functions than previously thought33,34. In a recent study conducted by Ben et al., it was found that the overexpression of a human prostate piRNA called PROPER interacts with the m6A reader protein YTHDF2. This interaction promotes RNA circularization and leads to the degradation of the dual-specificity phosphatase DUSP1. The resulting suppression of DUSP1 translation is associated with enhanced tumor metastasis through the p38 MAPK signaling pathway. These findings underscore the critical role of piRNAs as a key regulator in cancer progression34.

Therefore, our findings highlight the emerging role of piRNAs in somatic tissues, especially in the prostate. The nine clusters commonly found between the prostate and testis indicate that these are specifically essential for regulatory cell functions, as they were not affected independently by diet or aging in the prostate and are present in germinal tissues (testis). This finding correlates to the following, where sequence sizes were compared. Both prostate and testis overlapped in the analysis, which in addition to differing in miRNAs also confirmed that they were piRNAs being expressed. Finally, about reads in the association of a prostate CTR 21 cluster in common with the testis, the number of reads was statistically higher in the prostate, again emphasizing that the role of piRNAs is essentially characterized in gonads but is differentially regulated in somatic cells.

From the perspective of the role of piRNAs in aging, the impacts on the body are demonstrated, in addition to being one of the biggest risk factors for the development of diseases such as PCa35,36. From the perspective of the DOHaD concept, it was demonstrated that there is a decrease in the quantities of piRNA clusters in 540-day-old animals37. These results highlight the impact of aging when correlating the expressed piRNA clusters, where the young control group presents 66 clusters and 58 clusters in aging. Furthermore, when comparing the CTR 21 sample with the GLLP 540 we have an even greater variation of 66 to 53. Therefore, not only aging itself, but the restriction environment was able to modulate this expression. Regarding transposable elements, the correlation with piRNAs favors a deeper understanding of the processes that control development, cell differentiation, and the maintenance of genomic integrity. In the context of diseases, especially cancer, where genomic instability is a common feature, understanding the regulation of transposable elements and piRNAs can lead to possible therapeutic approaches36.

When we analyzed piRNA clusters in the testis, LTR elements, especially LTR/ERVK, were the most abundant, playing a crucial role in cell differentiation and sperm formation. We found more SINE elements in the prostate clusters, mainly SINE/Alu, known for regulating gene expression and genome stability37,38. In the GLLP 540 group, fewer SINE/Alu-derived piRNA clusters are expressed, suggesting that aging and environment impact these piRNAs and potentially their targets (TEs), possibly contributing to tumorigenesis, as carcinoma in situ was observed in these animals. On the contrary, young animals’ prostates have more piRNA clusters derived from SINE/Alu and LINE/L1 elements than the GLLP 540 group, with overall fewer piRNA clusters in the latter, reflecting their role in maintaining TEs. Decreased piRNA in the testis leads to retrotransposon hypomethylation, mainly LINE-1, affecting the expression or regulation of other genes in the genome, resulting in germ cell tumors due to effects on genomic stability. Not only in piRNAs expressed in the testis but also in the genome as a whole, SINE-Alu is the most abundant TEs, which suggests an impact of this element in situations of reactivation38.

The results obtained from RT-qPCR are associated with similar studies in the literature, demonstrating an interface between Prostate-piRNA-DOHaD. In recent studies, Zhang et al. (2020) investigated the role of piRNA sequences in human prostate cancer, revealing that these RNAs reduce the expression of the tumor suppressor protein PCDH9. This suppression leads to increased activity in the AKT signaling pathway, which promotes tumor growth and metastasis15. The research further demonstrated that inhibiting these piRNAs significantly curbs cancer cell growth in both laboratory experiments and animal models39. Furthermore, our results demonstrated that maternal protein restriction combined with aging led to the identification of 10 unique clusters in the GLLP 540 group. These findings highlight the potential of piRNAs in understanding the developmental origins of diseases, including cancer. In other organisms, such as arthropods, using mutated silkworm (Bombyx mori) with loss of function of Siwi (homolog to PIWI), there was a delay in larval growth, and they were unable to become adult moths40. They also exhibited defects in wing development and sexual differentiation. Transcriptome analysis revealed that loss of somatic piRNA biogenesis pathways results in abnormal expression of not only transposons but also host genes, presumably causing severe growth defects39. In particular, the maternal exposure to a toxin called MC-LR (microcystin-leucine arginine) during pregnancy in mice, observed a decrease in the expression of Rili and an increase in Riwi genes, which promoted the activation of the signaling pathway (PI3K/AKT) inducing prostate hyperplasia in the offspring41. In our results, there was also a decrease in Rili expression and an increase in Riwi in aged animals in the restricted group, suggesting not only an effect of aging but one enhanced by maternal protein restriction. This supports the idea that early-life factors affect aging (DOHaD). These findings show that the piRNA pathway works in the prostate and suggest that piRNAs might help detect cancer early and offer new treatment options, improving patient outcomes42. The PI3K/AKT pathway, which controls cell growth and survival, links piRNAs to cancer development41.

In conclusion, for the first time, we presented the piRNA expression profile in rat VP within the context of the DOHaD concept. The young rats had more piRNA clusters than old ones, and diet influenced these cluster expressions. Moreover, the number of transposable elements in piRNA clusters varied with age and the dam’s diet. Overall, these data highlighted the key role of maternal malnutrition in altering epigenetic mechanisms involved in prostate developmental biology early in life, with long-lasting consequences for the incidence of prostate disorders in old offspring rats.