Nutritional restriction during the peri-conceptional period alters the myometrial transcriptome during the peri-implantation period

This study hypothesized that female peri-conceptional undernutrition evokes transcriptomic alterations in the pig myometrium during the peri-implantation period. Myometrium was collected on days 15–16 of pregnancy from pigs fed a normal- (n = 4) or restricted-diet (n = 4) from conception until day 9th of pregnancy, and the transcriptomic profiles of the tissue were compared using Porcine (V2) Expression Microarrays 4 × 44 K. In restricted diet-fed pigs, 1021 differentially expressed genes (DEGs) with fold change ≥ 1.5, P ≤ 0.05 were revealed, and 708 of them were up-regulated. Based on the count score, the top within GOs was GO cellular components “extracellular exosome”, and the top KEGG pathway was the metabolic pathway. Ten selected DEGs, i.e. hydroxysteroid (17β) dehydrogenase 8, cyclooxygenase 2, prostaglandin F receptor, progesterone receptor membrane component 1, progesterone receptor membrane component 2, annexin A2, homeobox A10, S-phase cyclin A-associated protein in the ER, SRC proto-oncogene, non-receptor tyrosine kinase, and proliferating cell nuclear antigen were conducted through qPCR to validate microarray data. In conclusion, dietary restriction during the peri-conceptional period causes alterations in the expression of genes encoding proteins involved i.a. in the endocrine activity of the myometrium, embryo-maternal interactions, and mechanisms regulating cell cycle and proliferation.

excluded that changes in the myometrial potential for DNA methylation, as an effect of an unbalanced diet, may induce transcriptomic alterations in the tissue, determining its potential and activity.
This study hypothesized that undernutrition in females during the peri-conceptional period evokes alterations in the transcriptomic profile of the myometrium during the peri-implantation period. Thus, the current study aimed to determine and compare the transcriptomic profiles of the peri-implantation myometrium collected from females fed a normal-or restricted-diet during the peri-conceptional period. The study was performed on pig model, however, it is intended to imitate deprivation of nutritional habits in women who, during early stages of pregnancy, i.e. during the peri-conceptional period, are not aware of carrying a child. In this study, a particular focus was put on differentially expressed genes (DEGs) that are known to encode proteins crucial for the creation of the proper intrauterine environment, embryo-maternal interactions, and myometrial morphogenesis. Furthermore, there was performed transcription factor enrichment analysis to provide insight into the mechanisms of alterations in the transcriptomic activity of the myometrium in response to dietary restriction. Finally, there were evaluated common DEGs in the myometrium, endometrium 18 , conceptuses 27 of pigs fed a restricted diet during the periconceptional period. This study provides new data concerning molecular alterations in the uterine tissues and conceptuses during the peri-implantation period that occur as the consequence of female peri-conceptional undernutrition.

Results
Differentially expressed genes (DEGs). The transcriptomic analysis evaluated 1021 DEGs (FC ≥ 1.5; P ≤ 0.05) in the myometrium collected during the peri-implantation period from restricted-diet-fed pigs compared to the myometrium collected from normal-diet-fed pigs. Among evaluated DEGs, 69.34% (708 DEGs) were up-regulated and 30.66% (313 DEGs) were down-regulated. The full list of DEGs is presented in Supplementary Table 1, and the top 15 up-and top 15 down-regulated genes are listed in Table 1. Considering the function of the proteins encoded by listed DEGs, based on GeneCards: The Human Gene Database (https:// www. genec ards. org/), three groups of DEGs were evaluated. The first grouped genes which products are involved in the creation of a proper intrauterine environment, i.e.: synthesis of cholesterol, steroid hormones, and prosta- Table 1. The top 15 up-, and down-regulated differentially expressed genes (DEGs) in the myometrium of pigs during the peri-implantation period in response to dietary restriction during the peri-conceptional period.   These were: EP300, MAX, SMC3,  NELFE, REST, SMARCB1, ETS2, KLF4, BDP1, ARID2, ARNTL, TCF7L1, FOXP2, SMAD2, CHD2, CREB1,  ZNF449, RELA, HIF1A, TARDBP, XBP1,  Comparison of the myometrium, endometrium, and embryos transcriptomes of restricted-diet-fed gilts. The dietary restriction during the peri-conceptional period resulted in the alterations in the transcriptomes of the myometrium, endometrium 18 , and embryos 27 collected during the periimplantation period. The Venn diagram is shown in Fig. 2. Among the evaluated DEGs, 19 were common for all studied tissues, 113 were common for myometrium and embryos, 69 were common for myometrium, and endometrium and 113 were common for endometrium 76 and embryos. Details are presented in Supplementary  Table 5. The GO terms evaluated for DEGs common for myometrium, endometrium, and embryos are presented in Table 4. The interactions among DEGs common for the myometrium, endometrium, and embryos are presented in Fig. 3.

Discussion
The present study demonstrated that the short-lasting undernutrition in pigs during the peri-conceptional period alters the transcriptomic profile of the myometrium during the peri-implantation period. Specifically, as a consequence of the females' nutritional restriction, the expression of 1021 genes encoding proteins with known biological functions has been changed ≥ 1.5-fold, and the majority of them, i.e. 69.34%, were up-regulated. Based on the results of TF enrichment analysis, there are 333 TFs that may be involved in the regulation of the expression of DEGs identified in the current microarray experiment. Interestingly, 24 of these TFs were differentially expressed in response to dietary restriction. Thus, the restriction diet used during the peri-conceptional period may alter the transcriptional activity of the myometrium by affecting the expression of TFs. In the current study, the highest number of gene count was within GO CC. Similar findings were demonstrated previously in the endometrium of restricted-diet-fed pigs 18 . Furthermore, among 50 evaluated KEGG biological pathways [28][29][30] assessed in the myometrium of restricted-diet-fed pigs, the highest number of genes was classified to "metabolic pathways", similarly as in the endometrium and conceptuses of restricted-diet-fed pigs 18 www.nature.com/scientificreports/ from restricted-diet-fed pigs, the majority of DEGs were up-regulated (~ 80%), whereas in the endometrium above half of them (58%) were down-regulated 18,27 . Thus, the transcriptomic activity in the myometrium and Table 2. The selected differentially expressed genes (DEGs) in response to peri-conceptional undernutrition in the myometrium of pigs during the peri-implantation period, classified based on the function of encoded products. www.nature.com/scientificreports/ conceptuses during the peri-implantation period in response to dietary restriction during the peri-conceptional period may be mostly up-regulated, while in the endometrium it is rather down-regulated.
In this study, we have selected day 9 as the cut-off of under-nutrition protocol based on epigenetic events, namely methylation profile establishment, occurring during the early stages of pregnancy 31 . Until reaching the blastocyst stage, which occurs approximately at day 9 in this species, the embryonic genome undergoes active and passive demethylation 31 . In consequence of the disruptions in genome-wide methylation and demethylation, embryos are likely to develop with less than fully penetrant phenotypes 31 . It is known that both restricted and Table 3. Selected based on the high gene count gene ontology (GO) terms evaluated in the myometrium of pigs during the peri-implantation period in response to peri-conceptional undernutrition. BP biological processes, CC cellular components, MF molecular functions.  www.nature.com/scientificreports/ excessive maternal dietary protein intake during pregnancy alter the offspring's epigenetic marks and influence the expression of genes 32 . Moreover, DNA demethylation mechanisms act during early embryonic development, and a drastic reduction of methylation levels is expected to be found at later preimplantation stages 31 . Providing that DNA methylation is one of the mechanisms regulating the transcriptomic activity, the selection of a periconceptional period for dietary restriction for further transcriptome profiling during later stages of pregnancy is justified. We believe that the myometrium possesses "epigenetic sensitivity", mainly during intensive changes occurring during early pregnancy, and the epigenetic alterations, including DNA methylation, are one of possible mechanisms underlying the diet-related alterations in the transcriptomic activity of the uterus.
The previous studies determined that the myometrium is an endocrine tissue, and its secretory activity differs regarding the physiological status of the female (pregnant vs. estrous cyclic pigs) and in the following days of pregnancy and the estrous cycle 22,23,26,[33][34][35] . These alterations contribute to the creation of a specific, intrauterine milieu, which is critical for proper embryo-maternal dialog during early pregnancy 36,37 . It is worth highlighting that our previous study showed that restricted-diet fed gilts exhibit a decreased intrauterine concentration of E 2 17 . It is known that in pigs E 2 is one of maternal recognition of pregnancy signal 38 . Thus, the decrease in its concentration may deteriorate the developmental potential of embryos and impact mechanisms of pregnancy establishment. Notably, E 2 is also an important regulator of transcriptional activity of the uterus 39 . Thus, the alterations in the intrauterine concentration of E 2 evoked by dietary restriction 17 could stand for an additional possible mechanism underlying the observed diet-related alterations in the transcriptional activity observed in the current and previous study 18 .
Among the evaluated DEGs in the myometrium of restricted-diet-fed pigs, there was HMGCR , encoding the 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMGCR) classified to some of evaluated KEGG pathways [28][29][30] , i.a. metabolic pathways and also to some GOs, i.e. GO BP: oxidation-reduction process (GO:0055114), GO CC: endoplasmic reticulum membrane (GO:0005789), and GO CC: peroxisomal membrane (GO:0005778). The HMGCR is the rate-limiting enzyme in the biosynthesis of cholesterol and isoprenoids, which are substrates required for post-translational modifications of proteins involved in a signaling pathway that Figure 2. Venn diagram presenting the differentially expressed genes (DEGs) common for the myometrium, endometrium, and embryos collected from pigs during the peri-implantation period that were fed a restrictive diet during the peri-conceptional period compared to pigs fed a normal diet. Diagram generated with the use of online tool available at http:// bioin forma tics. psb. ugent. be/ webto ols/ Venn/, with further modifications. Table 4. Gene ontologies (GO) terms evaluated for DEGs common in the myometrium, endometrium, and conceptuses of pigs during the peri-implantation period that were fed a restrictive diet during the periconceptional period comparing to the myometrium of pigs during the peri-implantation period that were fed a normal diet during the peri-conceptional period. CC cellular components, MF molecular function.  40 ), and for steroid hormones biosynthesis pathway 41 .

GO term
Respecting that porcine myometrium possesses an active steroidogenic pathway [22][23][24]26,34 , the observed increased expression of HMGCR in the myometrium of the restricted diet-fed pigs may lead to a higher supply of cholesterol, for further synthesis of steroid hormones in the myometrium. However, the previous studies demonstrated that the dietary restriction during the peri-conceptional period results in the decrease of the total content of E 2 in the uterine flushings during the peri-implantation period 17 . This finding might be underlaid by previously determined diet-related up-regulation of endometrial HSD17B4 18 , encoding hydroxysteroid (17β) dehydrogenase type 4 (17βHSD4), which converts estradiol-17β (E 2 ) to estrone (E 1 ) 42 . Based on the current results, the expression of HSD17B4 was not altered in the myometrium of restricted-diet-fed pigs, but there was observed an up-regulation of the HSD17B8 gene. This gene codes for 17βHSD type 8 catalyzing the interconversion of androstenedione (A 4 ) to testosterone (T), and E 1 to E 2 . Thus, the diet-related decrease in the uterine content of E 2 is dependent rather on endometrial than myometrial steroidogenic activity. Interestingly, HMGCR is co-expressed and co-localized with HSD11B1, coding for 11β-hydroxysteroid dehydrogenase type 1 (11βHSD). The 11βHSD participates in the conversion of cortisone to cortisol 43,44 . Therefore, the observed in the current study up-regulation of HSD11B1 in the myometrium indicated the potential of the tissue to produce cortisol. Interestingly, it was previously found that cortisol may decrease the release of luteolytic prostaglandin F 2 α (PGF 2 α) from the endometrium 45 . The expression of cyclooxygenase 2 (COX2), prostaglandin F synthase (PGFS), and prostaglandin F receptor (PTGFR) were previously demonstrated in the myometrium as well as the active PGF 2 α release by this tissue in vitro 21,46 . The current study has shown that dietary restriction leads to the decreased expression of COX2 and PTGFR in the myometrium. Thus, the action of PGF 2 α in the myometrium of restricted-diet-fed pigs may be disturbed. The peri-implantation period is accompanied by a molecular dialog between conceptuses and the uterus. This phenomenon can be considered as a semi-inflammatory process 47 , and PGF 2 α is recognized as a proinflammatory and immunostimulatory factor 48 . Moreover, PGF 2 α regulates the expression of genes potentially important for embryo-maternal interactions 49 . Disturbances in the potential of PGF 2 α synthesis and its action in the myometrium of restricted-diet-fed pigs expressed as the decreased COX2 and PTGFR expression may affect the PGF 2 α-mediated process occurring during the peri-implantation period, which can be detrimental for pregnancy success.
The proper embryo-maternal interactions require signals triggered by P 4 50 . Interestingly, the current study documented that the restricted diet resulted in the alterations in the expression of progesterone membrane components, i.e. PGRMC1 (up-regulated) and PGRMC2 (down-regulated). Furthermore, PGRMC1 mediates P 4 -dependent control of the cell cycle, proliferation, and apoptosis, whereas PGRMC2 was found to inhibit cell migration [51][52][53] . Notably, the decreased expression of PGRMC2 was detected also in porcine conceptuses collected from restricted-diet-fed pigs 27 . The alterations of PGRMCs expression in response to the dietary restriction may affect the non-genomic action of P 4, and impact cell cycle, proliferation, and apoptosis of myometrial cells. Uterine development and morphogenesis during early pregnancy are also regulated by homeobox A family  54 . Previous studies demonstrated that the expression of HOXA10 and HOXA13 is increased in the myometrium of early pregnant pigs 46 . Thus, the decreased expression of HOXA3 and HOXA10 in the myometrium of restricted diet-fed pigs may affect cell cycle, myocyte proliferative capacity and differentiation, and development of the myometrium required for proper development of the uterus during pregnancy. Notably, the restricted diet during the peri-conceptional period evokes the alteration in the several genes encoding proteins involved in the regulation of the cell cycle, proliferation, and differentiation. This study evaluated that the expression of genes encoding anaphase-promoting complex subunits, i.a. ANAPC5, ANAPC7, ANAPC16, and the expression of proliferating cell nuclear antigen (PCNA) and SRC proto-oncogene, non-receptor tyrosine kinase (SRC) was significantly increased, while the expression of S-phase cyclin A-associated protein in the endoplasmic reticulum (SCAPER) was decreased. The genes encoding ANAPC is classified to the GO CC term "anaphase-promoting complex" (GO:0005680). The ANAPC regulates the proper chromosome segregation during mitosis, mitotic exit, and the maintenance of the G1 phase of the cell cycle 55 . PCNA may be considered as a regulator of DNA replication, while SRC is involved in the regulation of proliferation, differentiation, and cellular viability 56,57 . It was found that the decreased content of SCAPER contributes to the decreased level of cytoplasmic cyclin A resulting in the delay of G1/S phase transition during the cell cycle 58 . Thus, the dietary restriction in females during the peri-conceptional period may determine the myometrial cells' fate during the peri-implantation period.
The current study demonstrated that the restricted diet during the peri-conceptional period also leads to the up-regulation of ANXA2 (annexin 2) in the myometrium of pigs during the peri-implantation period. The ANXA2 is classified to many GO terms among GO CC, i.a. extracellular exosome (GO:0070062), cytosol (GO:0005829), extracellular matrix (GO:0031012), and membrane raft (GO:0045121), and among GO MF, poly(A) RNA binding (GO:0003723). In general, annexins are a family of proteins involved in many cellular functions, including vesiculation and cell membrane repair, and ANXA2 induces membrane folding and blebbing initiated from membrane structural defects 59 . The previous studies demonstrated that myometrial expression of ANXA2 was greater in pigs during the peri-implantation period than during luteolysis 46 . In the current study, the abundance of the ANXA2 mRNA transcript was found to be greater in the myometrium of the restricteddiet-fed pigs. Further bioinformatical analysis revealed that ANXA2 is co-expressed with genes encoding other annexins (ANXA1 and ANXA11), intercellular adhesion molecules (ICAM1, ICAM2, and ICAM4), and platelet/ endothelial cell adhesion molecule 1 (PECAM1), which exhibited up-regulated expression in the myometrium of restricted-diet-fed gilts. Interestingly, these genes encode for proteins involved in processes of embryo apposition and adhesion to the uterine wall as well in the induction of angiogenesis 59,60 . Thus, the mild nutritional restriction during the peri-conceptional period may affect the mechanisms determining the myometrial integrity and angiogenesis during the peri-implantation period.
This study assessed the transcriptomic alterations in the myometrium (the current study), endometrium 18 , and conceptuses 27 collected from restricted-diet-fed pigs. Among DEGs which are common for uterine tissues and conceptuses there is MCL1 apoptosis regulator, BCL2 family member (MCL1), which expression in restricted diet-fed pigs was increased in the myometrium (the current study) while decreased in the endometrium and embryos 18,27 . The MCL1 is considered an anti-apoptotic protein, and its down-regulation leads to the activation of caspases involved in apoptosis 61 . Thus, the interesting phenomenon of increased gene expression in the myometrium and decreased gene expression in the endometrium and embryos in response to the restricted diet may affect the rate of apoptosis during the peri-implantation period. Furthermore, in the myometrium (the current study), the endometrium 18 , and embryos 27 of the restricted diet-fed pigs there was observed the up-regulation of ACP5, which is involved in the transport of iron from the endometrium to embryos 62 . The ACP5 is classified to the "extracellular exosome" (GO:0070062) GO CC term. This finding indicates that the restricted diet may potentially affect the uterine and conceptuses' capacity to produce the extracellular vesicles that may participate in the embryo-maternal cross-talk 63 . It is worth highlighting that both ACP5 and FTH1, which is the gene encoding ferritin heavy chain 1, were up-regulated in the myometrium (the current study), the endometrium 18 and conceptuses 27 of restricted diet-fed pigs. Interestingly, FTH1 is a major intracellular iron storage protein that may function in the protection from oxidative stress 64 . Thus, the up-regulation of FTH1 and ACP5 genes in uterine tissues and conceptuses may lead to the altered distribution of iron in embryo-uterine interspace. Thus, the described alterations in the expression of ACP5, FTH1, GSTA4 and GPX3, occurring in the uterine tissues and conceptuses of restricted-diet-fed pigs, may also be considered as mechanisms of intrauterine protection from oxidative stress.
For a better understanding of the mechanisms underlying changes in the transcriptional activity of the myometrium occurring in response to a restricted diet, we have identified TFs casually responsible for changes in the expression of the evaluated DEGs. Indication of TFs which may be involved in the regulation of the transcriptional activity of the evaluated DEGs is promising, but still challenging and show that applied undernutrition protocol may evoke a broad spectrum of transcriptomic alterations. Notably, the peri-conceptional undernutrition caused alterations in the transcriptional activity of 24 of the evaluated TFs. Among these differentially expressed TFs, there is KLF4, encoding for Krüppel-like factor 4. KLF4 is involved in the regulation of numerous biological processes including cell development and cell division, apoptosis, and cell programming, and its down-regulation is often coupled with tumor progression 65 . This study showed that nutritional restriction increases the transcriptional activity of KLF4 in the myometrium. Importantly, the previous studies showed that an up-regulation of KLF4 induces cell cycle arrest 65 . Having in mind that this study evaluated diet-related alterations in the transcriptional activity of genes coding for proteins involved in the regulation of cell cycle and determining the proliferative potential of the myometrium, i.a. PGRMC genes, HOXA genes, ANAPC genes, SRC, SCAPER, PCNA, and these changes may be connected with the altered expression of KLF4. Among differentially expressed TFs there was also RELA coding for RELA Proto-Oncogene, NF-KB Subunit. RELA is a part of the nuclear factor kappaB (NF-kappaB) family, regulating the expression of i.a. pro-inflammatory mediators. The Scientific Reports | (2021) 11:21187 | https://doi.org/10.1038/s41598-021-00533-x www.nature.com/scientificreports/ expression of RELA was increased in the myometrium in response to dietary restriction. It is worth highlighting that pregnancy might be recognized as a semi-inflammatory process where the action of cytokines, tumor necrosis factor (TNF), and COX2 is crucial for proper implantation 47 . In this study, we observed diet-related down-regulation in COX2 expression, and this alteration might be connected with the altered expression of RELA. However, the applied undernutrition protocol did not contribute to alterations in the expression of proinflammatory cytokines and TNF what suggests that there must appear additional mechanism controlling the transcriptional activity of the myometrium of restricted-diet-fed pigs. One may not exclude that this mechanism relies on epigenetic mechanisms, including DNA methylation, but this assumption needs further investigation.
In conclusion, the observed in the current study alternations in the transcriptomic profile of the myometrium show that the short-lasting 30% decrease of food intake by pregnant females during the unique peri-conceptional period may disturb the proper course of biological processes occurring in the myometrium during the periimplantation period. Among the observed alterations, increased potential for the synthesis of steroid hormones, decreased potential for prostaglandins synthesis and action, decreased proliferative capacity, and increased adhesion among myocytes were determined. The observed transcriptomic alterations of the myometrium might jeopardize proper implantation and thus, pregnancy success, but this requires further in vivo studies. Moreover, the results documented that the transcriptomes of the myometrium, endometrium, and conceptuses during the peri-implantation period respond differently to female dietary restriction during the peri-conceptional period.

Methods
Ethics statement. All experiments were approved by the Animal Ethics Committee, University of Warmia and Mazury in Olsztyn, Olsztyn, Poland-decision No. 54/2015/DTN. The study was carried out in compliance with the ARRIVE guidelines 66 , and all methods were carried out following relevant guidelines and regulations indicated where appropriate.
Animals, feeding protocol, and collection of myometrial tissue. The post-pubertal pigs (Sus scrofa Domestica L., Polish Landrace × Great White Polish, weighing 95 to 110 kg) were naturally bred on the first and second day of estrus of their second estrous cycle. Gilts assigned to the control group (n = 4) were fed a normal diet, i.e. daily 3 kg/gilt of forage since the first sign of the estrus, during natural mating, and until slaughter. Gilts assigned to the experimental group (n = 4) were fed a restricted diet, i.e. daily 2.1 kg/gilt of forage, which is a 30% food restriction, during the peri-conceptional period (the first sign of the estrus, during mating, and until 9th day of pregnancy), and next fed a normal diet until slaughter.
The composition of forage, feeding procedures and housing of animals were described previously, and the effectiveness of the applied protocol of undernutrition was verified by analyzing the biochemical parameters of blood plasma 17,27,67 . The animals did not lose weight. Briefly, there was observed a diet-related decrease in the concentration of total phosphorus, calcium, and total cholesterol, and a tendency to increase in the concentration of triglycerides in restricted-diet fed pigs 67 . Alterations in the concentration of total cholesterol and triglycerides in blood plasma indicate alterations in lipid metabolism, which is a visible sign of altered nutritional status in pigs fed a restricted diet. Moreover, there was indicated that dietary restriction led to the increase in the concentration of aspartate aminotransferase in blood plasma, a marker of liver damage, which is one more sign that applied dietary restriction affected the wellbeing of these females 67 . We did not observe any diet-related effect on conception success nor the morphology of conceptuses.
Gilts from both groups were slaughtered in a local abattoir (Rozdroże, Poland) during the peri-implantation period, specifically on day 16 after the first mating. Respecting the mating protocol and effectiveness of mating, the animals were on day 15 to 16 of pregnancy. Immediately after slaughter uteri were collected and placed into ice-cold phosphate-buffered saline (PBS) containing 100 IU mL −1 penicillin (Polfa, Poland) and 100 IU mL −1 of streptomycin (Sigma Aldrich, St. Louis, MO, USA), and transported to the laboratory. The day of pregnancy was confirmed by the morphology of ovaries and conceptuses flushed from uterine horns as well as the concentration of P 4 in peripheral blood 33,[68][69][70][71] . Next, uterine horns were opened longitudinally, the perimetrium tissue was discarded by careful scraping, and the myometrial cross-sections were collected with scissors and snap-frozen in liquid nitrogen (− 196 °C). Samples were stored at -80 °C until further analysis. RNA isolation. RNA was extracted from 30 mg-weighting myometrial slices using 500 mL of TRI Reagent (Sigma Aldrich, Germany). Samples were initially homogenized on ice using a homogenizer TissueRuptor (Qiagen, Germany). Then, a 150 µL of chloroform (POCH, Poland) was added, samples were mixed thoroughly and incubated for 10 min at 4 °C. Next, samples were centrifuged 12,000×g, for 10 min at 4 °C and the supernatant was transferred to a 350 µL of propane-2-ol (Sigma Aldrich, Germany) beforehand chilled at − 20 °C. Next, samples were mixed well by vortexing and incubated for 15 min on ice assuring brief mixing every 3 min. Next, samples were centrifuged 12,000×g, for 10 min at 4 °C, and the supernatant was discarded. RNA pellets were washed twice with 1 mL of 75% ethanol (Chempur, Poland) beforehand chilled at − 20 °C. The RNA pellets were suspended in 30 µL of DEPC-treated water (Sigma Aldrich, Germany). The initial analysis of concentration and purity of extracted RNA was determined using Tecan with i-control software (Tecan, Switzerland), and the initial test for RNA integration was performed using gel electrophoresis. Samples with a concentration higher than 100 ng/µL, optical density (OD) from 1.8 to 2.0, and integral electrophoretically were used in the next steps of analysis. The samples that fulfilled these criteria were tested using microcapillary gel electrophoresis in 2100 Bioanalyzer (Agilent Technologies, USA) to determine RNA integrity number (RIN). Samples with RIN ≥ 7.5 were used for microarray analysis. www.nature.com/scientificreports/ Microarray data analysis and differentially expressed genes. The microarray analysis was performed using RNA samples isolated from the myometrium collected from restricted-diet-fed pigs (experimental group, n = 4) in comparison to the normal-diet-fed pigs (control group, n = 4) using Porcine (V2) Expression Microarrays 4 × 44 K (Agilent Technologies, USA) according to the Two-Color Microarray-Based Gene Expression Analysis Protocol (Agilent Technologies, USA). Four independent biological replicates per group, i.e. experimental and control group, allowed to obtain an experiment power of 80% and a false discovery rate (FDR) of 0.1 72 . The specific workflow of the experiment was described previously 18,27,46 . Briefly, 400 ng aliquots of RNA were used to synthesize cRNAs samples labeled with CY3 or CY5. Fluorochrome-labeled cRNA was purified on RNeasy Mini Spin columns (Qiagen, USA). After determination of cRNA concentration and labeling efficiency, samples were conducted through the cRNA fragmentation step and mixed with a hybridization buffer. Then, samples were applied on microarrays following the classic class comparison experiment design, using a dye-swap to prevent dye bias 73,74 . After hybridization (for 17 h at 65 °C), microarrays were washed, dried, and scanned using Agilent High-Resolution C Microarray Scanner (Agilent Technologies, USA). Fluorescence intensity was calculated using Agilent Feature Extraction software v. 10.5.1 (Agilent Technologies, USA), the outlier spots were filtered, and the background correction and dye normalization (linear and LOWESS) were performed. The obtained data were analyzed using GeneSpring GX 13.0 software (Agilent Technologies, USA). The transcripts with fold change (FC) ≥ 1.5 and P ≤ 0.05 (Student's t-test, default statistical model) were aligned with the Sus scrofa transcriptome (txid9823) using the BLAST tool for identification of encoded gene name. When more than one differentially expressed transcript was assigned to one gene name, the FC and P-values of these transcripts were averaged and recognized as one differentially expressed gene (DEG). DEGs that were found in both up-regulated and down-regulated gene lists were omitted in further analyses. Interaction network of selected genes. The list of evaluated DEGs was then used to create groups of DEGs encoding proteins involved in the: 1/The endocrine activity, 2/Embryo-maternal interactions, and 3/ Mechanisms of proliferation and differentiation. For these groups, the gene interaction networks were created with GeneMANIA prediction server 79 , using Homo Sapiens as a reference genome. GO-BP-based weighting was used to determine the interactions within co-expression, co-localization, genetic interaction, and physical interaction categories, and each network was enriched with 10 resultant genes.

Enriched gene ontology (GO) terms and biological pathways.
Validation procedure. RNA samples used for microarray analysis were used for cDNA synthesis using QuantiTec Reverse Transcription Kit (Qiagen, USA) following the protocol provided by the manufacturer and rules described by Bustin et al. 80 . For validation procedure, there were selected five up-regulated and five downregulated genes expressed in the myometrium of restricted-diet-fed pigs. cDNA samples were amplified with POWER SYBR Green PCR Master Mix in AriaMx apparatus with AriaMX Software v. 1.3 (Agilent Technologies, USA). When available, primer sequences were obtained from references [actin beta, ACTB 81 , RNA, 18S ribosomal, RNA18S 19 , hydroxysteroid (17β) dehydrogenase 8, HSD17B8 82 , homeobox A10, HOXA10 83 , cyclooxygenase 2, COX2 33 ], or were designed using BLAST tool provided by NCBI. The reaction parameters were: 1/initial denaturation for 10 min at 95 °C (one cycle), 2/denaturation for 15 s at 95 °C, primer annealing for 1 min, at temperature specific to primers set (Supplementary Table 6), and elongation for 1 min at 72 °C (40 cycles) followed by dissociation step (one cycle). The expression level of amplified genes was calculated using the ∆∆Ct method. For each tested gene, the absolute value of the ratio of the mean expression of the tested gene in the control group i.e. fed a normal diet vs. the mean expression of the tested gene in the experimental group, i.e. fed a restricted diet during the peri-conceptional period were used to calculate FC values (Eq. 1).
Calculation of fold change (FC) values for data obtained with Real-Time PCR (FC RT ).
x 2 −∆∆Ct control-mean expression of the tested gene in the control group, x 2 −∆∆Ct experimental-mean expression of the tested gene in in the experimental group.
The FC values obtained after Real-Time PCR (FC RT ) and FC values obtained after the microarray experiment (FC M ) were used to calculate multiple regression coefficients (R) 84 .
Transcription factor enrichment analysis (TFEA). Transcription factor (TF) enrichment analysis was performed using human TF dataset with rank test Kolmogorov-Smirnov rank test, and the TFEA.CHiP tool (https:// bio. tools/ TFEA. ChIP) 85 , and a GSEA-like analysis method. There were selected default options: source organism: human, plot type: Enrichment Scores ChiP-Seq source database: ReMap2020 & GeneHancer Double Elite. The output list of TFs was screened for duplicates, which were removed from further analysis. Next, the www.nature.com/scientificreports/ VENN diagram was created using lists of the TFs and DEGs to evaluate TFs that were differentially expressed (DE) in the myometrium of restricted-diet-fed pigs.
The comparison of the transcriptomic profiles of the myometrium, endometrium, and embryos of restricted-diet-fed gilts: Venn diagram. The lists of DEGs evaluated in the myometrium (the present study), endometrium 18 , and conceptuses 27 collected from restricted-diet-fed pigs were used to create a Venn diagram. The DEGs common for the myometrium, endometrium, and embryos of restricted diet-fed gilts were then used to create an interaction network using GeneMania software as described in section "Interaction network of selected genes".