Impact of dietary supplementation of beef cows with rumen-protected methionine during the periconceptional period on prenatal growth and performance to weaning

Abstract

Changes in maternal nutrition during the periconceptional period can influence postnatal growth in cattle. This study aimed to identify the impact of supplementing beef cows with rumen-protected methionine (RP-Met) during the periconceptional period on their female progeny. In experiment 1, plasma methionine (Met) levels were analyzed in samples from 10 Angus crossbred, non-lactating beef cows. Cows were randomly assigned to receive 454 g of cottonseed meal with 15 g/d of RP-Met (RPM; Smartamine M, Adisseo) or not (CON) for 5 d and data were analyzed as a completely randomized design with repeated measures. A treatment-by-day interaction was observed (P < 0.001), where plasma Met concentrations increased in the RPM treatment yet remained basal in CON. In experiment 2, 114 cows were fed a roughage-based diet and randomized to receive 454 g/d of corn gluten supplemented with 15 g/d of RP-Met (RPM n = 56) or not (CON n = 58) from days −7 to 7 relative to timed-artificial insemination using sexed semen to obtain females. Amino acids were measured in plasma samples from days −8, 0, and 7 in cows. In the female progeny, body weight, withers height, body length, and heart girth were measured every 60 d from birth through weaning at an average age of 242 ± 5.8 d. Liver, adipose tissue, and longissimus dorsi muscle biopsies were collected at 187.88 ± 5.5 d of age and a subset of 20 random samples (CON = 10; RPM = 10) were selected for RNA-seq on each tissue. Data were analyzed using a generalized randomized block design with repeated measures. Methionine was increased (P < 0.01) in plasma from cows in the RPM treatment on days 0 and 7. After calving, 34 female calves (CON = 16; RPM = 18) remained in the study and no difference was observed in birth weights between treatments. Calves were taller at the withers for RPM than CON (P = 0.03; CON = 92 ± 1.0 cm; RPM = 95 ± 1 cm) but there were no effects of treatment on other measures of body size. A total of 30, 24, and 2 differentially expressed genes (DEGs; P < 0.01) were observed in liver, longissimus dorsi muscle, and adipose tissue respectively. In summary, feeding RP-Met to cows in the periconceptional period resulted in female calves that were taller than CON before weaning. There were DEGs in the tissue samples but no other changes in measurements associated with body size. In conclusion, supplementation of RP-Met to beef cows during the periconceptional period caused minor changes in the female offspring before weaning.

Rumen-protected methionine fed to beef cows during the periconceptional period increased the height of female progeny and modified gene expression in the liver, longissimus dorsi muscle, and adipose tissue.

Introduction

The concept of developmental programming (adaptive developmental plasticity) has been known since Dr. David Barker postulated the hypothesis in 1995, which states that changes during the embryonic and fetal life predispose the fetus to diseases in postnatal life (Barker, 1995, 1998). The effect of developmental programming has been studied in different species, including rodents (Kwong et al., 2000), pigs (Wu et al., 2019), sheep (Sartori et al., 2020), and cattle (Broadhead et al., 2019). Developmental programming can be beneficial or detrimental to the offspring with the specific outcome depending on both the intra-uterine environment and the environment during postnatal life (Reynolds and Caton, 2012).

One approach to program development that has yielded favorable outcomes for the offspring is provision of methyl donors to the preimplantation embryo. Folate, choline, methionine (Met), and betaine are important components of the methionine cycle for generation of S-adenosylmethionine that is essential for different metabolic processes including DNA methylation (Cronje, 2018). The preimplantation embryo undergoes extensive changes in DNA methylation characterized by widespread demethylation followed by the addition of new methylation as the embryo begins to differentiate (Zhu et al., 2021). Feeding rumen-protected methionine (RP-Met) to dairy cows during the periconceptional period (from calving until embryo flushing at 6 d after artificial insemination) resulted in modulations in gene expression in the bovine blastocyst with a total of 276 genes experiencing changes in transcript abundance (Peñagaricano et al., 2013). In another study, feeding RP-Met to postpartum dairy cows caused changes in the lipid composition of recovered embryos (Stella et al., 2024). Moreover, Acosta et al. (2016) reported that using RP-Met during the periconceptional period in Holstein cows decreased blastocyst methylation levels. Another methyl donor, choline, has been found to act on cultured embryos during the first 7 d of development to increase weaning and slaughter weight of beef cattle (Estrada‐Cortés et al., 2021; Haimon et al., 2024). According to Estrada-Cortés et al. (2021) calves exposed to 1.8 mM of choline chloride during in vitro embryo development, resulted in an increase on birth weights and 205-adjusted weaning weights compared to untreated embryos. Such a result leads to the hypothesis that methionine may have similar programming actions. The aim of the current study was to assess the increase of methionine concentrations in plasma upon feeding 15 g of RP-Met and determine the influence of feeding RP-Met during the periconceptional period on prenatal growth and performance to weaning and gene expression of female progeny. We hypothesized that feeding beef cows with 15 g of RP-Met during the periconceptional period will program bovine embryo development in a manner that enhances postnatal growth.

Materials and Methods

This experiment was conducted at the North Florida Research and Education Center, University of Florida, Marianna, FL. The University of Florida Institutional Animal Care and Use Committee reviewed and approved all research procedures (Protocol number 202011174).

Experiment 1

The goal was to determine the increase in plasma Met concentrations upon feeding of 15 g RP-Met (11.25 g of methionine). A total of 10 Angus crossbred, non-lactating beef cows were enrolled in the experiment. The cows were fed Bermuda grass (Cynodon dactylon (L.) Pers) hay ad libitum for 5 d. Cows were brought to individual pens every morning and randomly assigned to 2 treatments (CON n = 5 and RPM n = 5). The CON treatment consisted of feeding 454 g of cottonseed meal and the RPM treatment consisted of feeding 15 g of RP-Met (Smartamine M, Adisseo, Alpharetta, GA, USA) in 454 g of cottonseed meal. Supplementation was delivered in a feed tub every morning after animals were individually sorted.

Blood samples from cows were collected from the jugular vein every day for 6 d. On day 0, blood samples were collected before supplementing the treatments, from days 1 to 6 blood sample was collected 6 h after animals consumed their respective treatment. These times for sampling collections were selected as, according to Toledo et al. (2017) the peak in plasma Met after RPM feeding can be observed after 6 h post feeding. Samples were collected into vacutainer tubes containing EDTA (BD Vacutainer, Becton, Dickinson and Co., Franklin Lakes, NJ) and placed in ice until centrifugation at 3,000 × g for 15 min at 4 °C. Plasma samples were stored at −80 °C until shipping for Met analysis.

Plasma Met was determined using the method described by Hoppmann and Arriola Apelo, (2024). Briefly, the plasma sample (85 µL) was gravimetrically mixed with 12 µL of concentrated 13C labeled internal standard. Protein precipitation was achieved by adding acetonitrile and centrifuging the mixture. The resulting supernatant was buffered and derivatized with Fmoc-Cl (Sigma-Aldrich 23186) in acetonitrile. Solid-phase extraction using a Strata X-A strong anion exchange column was performed, and the eluted AAs were dried under nitrogen and resuspended in liquid chromatography mass spectrometry (LCMS) mobile phase for analysis. A Shimadzu Nexera-i LC-2040C system equipped with a Gemini NX-C18 reversed-phase column was used for separation, and a Shimadzu LCMS-2020 mass spectrometer was employed for detection. Negative electro-spray ionization was utilized for all amino acids except Pro and Arg, which were analyzed in positive ion mode. Data processing was conducted using Lab Solutions (Shimadzu).

Statistical analysis of Met concentration in plasma was analyzed as a repeated measurement within cow using the MIXED procedure of SAS (SAS Institute Inc., Cary, NC, USA). The model included the fixed effects of measurement day, treatment (CON or RPM), the interaction between treatment and measurement day and with cow nested within day as a random variable. The covariance structure used was autoregressive 1, and it was selected by the smallest AICC. One cow from the RPM treatment was excluded from the analysis as she did not consume the treatment.

Experiment 2

Animals and treatments

A total of 114 Angus crossbred beef lactating cows (age = 5.9 ± 2.14 yr; BW = 532.3 ± 68.2 kg) were enrolled in this experiment. The animals were, on average 71 ± 6.9 d postpartum. Animals were blocked in 2 groups for management purposes (group A n = 60 [CON: n = 29; RPM: n = 31]; group B n = 54 [CON: n = 27; RPM: n = 27]). Both groups received a forage-based diet (Table 1), and treatments were supplemented for 14 d, starting 7 d before the timed-artificial insemination (TAI; day 0) to 7 d after TAI (Figure 1). Treatment supplementation was conducted using an automated feeder (Super SmartFeed; C-Lock Inc., Rapid City, SD). Animals were allowed to learn to eat from the automated feeder for 14 d before the beginning of the study. Supplemented treatments consisted of CON: 454 g of corn gluten and RPM: 454 g of corn gluten plus 15 g of RP-Met (Smartamine M, Adisseo). Cows were randomly assigned to 2 treatments within groups (CON n = 56 and RPM n = 58).

Table 1.

Analyzed¹ chemical composition of the ingredients (DM basis) and fed diet to cows

Diet composition, % of DM
Corn silage	30.17
Gin trash	33.67
Cotton seed meal	6.44
Corn gluten feed	28.69
Mineral–vitamin concentrate	1.02
Dietary chemical composition, % of DM 2
CP	14.70
NDF	37.30
ADF	28.10
EE	3.80
Ash	5.96
NEm, Mcal/kg	1.46
NEg, Mcal/kg	0.88

¹Dairy One Forage Testing, Laboratory, Ithaca, NY.

²ADF, acid detergent fiber; CP, crude protein; DM, dry matter; EE, ether extract; NDF, neutral detergent fiber; NE_m, net energy for maintenance; NE_g, net energy for gain.

Figure 1.

Experimental design and estrus synchronization protocol. Treatments (CON and RPM) were supplemented with Super SmartFeed from days −7 to 7 relative to the AI day. Estrus synchronization was done using a 7-d co-synch + CIDR split-time TAI sexed semen protocol. Plasma samples were collected on days −8, 0, and 7 for amino acids (AA) analysis, also at days 30 and 60 for PAG concentrations.

Reproductive management

At the beginning of the estrus synchronization protocol, 46% of the cows were cycling as determined by ultrasound examination of the reproductive tract by the presence of an active CL. Every cow was enrolled in a 7-d Co-synch + CIDR split-time AI sexed semen protocol (Beef reproduction task force, 2023). On d −10, relative to TAI, cows were treated with 100 µg of GnRH (Cystorelin gonadorelin, Boehringer Ingelheim), and an intravaginal progesterone device (Eazi-Breed CIDR, Zoetis) was inserted. On day −3, the intravaginal CIDR device was removed, an injection of 25 mg of dinoprost tromethamine (Prostaglandin (PGF2α); Lutalyse HighCon Injection, Zoetis) was given intramuscularly, and an estrus detection patch (Estrotect) was placed halfway between the hip and tail head per company recommendations on all cows. Estrus detection patch scores were determined 48 h after the intravaginal CIDR removal. Cows in which more than 50% of detection patch was rubbed off were considered in estrus and were inseminated with X-sorted semen 60 h after removing CIDR. The cows that did not display estrus were TAI with X-sorted semen 72 h after removing the CIDR. Three different sires were used for AI.

Blood sample collection and analyses

Blood was sampled from the jugular vein into evacuated tubes containing EDTA (BD Vacutainer, Becton, Dickinson and Co.) in the morning and afternoon on days −8, 0 (TAI), and 7 to evaluate plasma amino acid (AA) concentration. Both samples collected from each animal each day were pooled to represent a single day. Also, a blood sample was collected 30 and 60 d after TAI to measure plasma pregnancy-associated glycoprotein (PAG) concentration. Tubes were placed on ice until centrifugation at 3,000 × g for 15 min at 4 °C. Plasma samples were stored at −80 °C until AA concentration analysis and PAG concentration.

Amino acid concentrations in plasma were determined using the method described by Hoppmann and Arriola Apelo. 2024, described in experiment 1. The concentration of PAG was determined by ELISA (BioPRYN Flex; BioTracking LLC) as per the manufacturer’s guidelines. Briefly, antibodies coated in the wells detected pregnancy-specific protein B, which was then identified by a labeled secondary antibody. A standard curve ranging from 0 to 8 ng/mL was generated and run in duplicate reactions. The intra-assay CV and inter-assay CV were 2.74% and 5.49%, respectively.

Ultrasonography

One technician conducted transrectal ultrasonography (Esaote ultrasound, MyLab Delta Vet, with 10-5 MHz transducer) throughout the experiment. The ovaries of all cows were examined at day −10 relative to TAI to determine cyclicity status (cycling: presence of a corpus luteum; no cycling: absence of a corpus luteum). Pregnancy per artificial insemination (P/AI) was diagnosed by transrectal ultrasonography on day 30 after TAI based on the presence of an amniotic vesicle and an embryo with a heartbeat. Pregnant cows on day 30 after AI were re-evaluated for pregnancy on days 60 and 120 after TAI.

Videos were recorded for embryo and fetal morphology measurements by ultrasonography 30 and 60 d after TAI. Later, 2 trained people that was blinded to experimental treatments analyzed the videos with the My Lab Desk software (Esaote, IN, USA) to determine the correct position to measure the embryo and the fetus. On day 30, amniotic vesicle diameter, circumference, embryo crown-rump length, and abdominal cavity were measured (Figure 2A to D). On day 60, eye cavity diameter, head length, whiter-rump length, and head transversal were measured (Figure 2E to H). Measurements for individual animals were excluded from the analysis when videos provided unclear images to estimate the measurement (day 30: n = 4; day 60: n = 9).

Figure 2.

Embryo morphology measurements at 30 d of pregnancy. (A) Vesicle circumference, (B) vesicle diameter, (C) embryo length, and (D) abdominal cavity length. Fetal morphology at 60 d of pregnancy. (A) Eye cavity length, (B) head length, (C) wither to tail length, and (D) head width.

Postnatal measurements

Birth weights from female calves (n = 34; CON: n = 16; RPM: n = 18) and male calves (n = 6; CON: n = 3; RPM: n = 3) were collected within 24 h of birth. Later, only female calves were considered, as the experiment was not designed to follow male calves. Female calves were kept with their dams until weaning (age = 242 ± 5.8 d). Body weights from female calves were collected every 2 mo until weaning. Also, height at the withers, body length, and heart girth were measured from 60 d of age until weaning. Adjusted 205-d (205-d adjusted) weaning weight was calculated using the formula [weaning weight − birth weight)/days of age at weaning] × 205 + birth weight (BIF, 2010).

Liver, adipose tissue, and muscle biopsies

At 187 ± 5.51 d of age, biopsies samples from the liver, subcutaneous adipose tissue, and longissimus dorsi muscle biopsies were collected from the 34 female calves. Female calves from group A (n = 14 [CON: n = 6; RPM: n = 8]) and group B (n = 20 [CON: n = 10; RPM: n = 10]) were collected on 2 separate days. The disinfecting and anesthesia protocol for the 3 biopsies was as follows: The area was shaved and disinfected with 3 alternating rounds of povidone–iodine scrubbing followed by 70% isopropyl alcohol. Upon disinfecting, 100 cc of lidocaine hydrochloride 2% (Clipper Distributing Company, LLC, St Joseph, MO, USA) was administered in the subcutaneous space and intercostal muscles. An additional disinfection round was performed as described before. An incision was performed on the skin using a sterilized scalpel blade. Liver biopsies were performed using a sterilized marrow aspiration needle (Monoject, Dublin, Ireland) at the 10th intercostal space, as described by (Alfaro et al., 2023). Subcutaneous adipose tissue biopsies were performed over one side of the tail-head area using a biopsy needle (SuperCore Biopsy Instrument 14 ga × 9 cm, Aragon Medical Devices, Frisco, TX, USA). Longissimus dorsi muscle biopsies were performed using a biopsy needle (SuperCore Biopsy Instrument 14 ga × 9 cm, Aragon Medical Devices), as described by (Alfaro et al., 2020). Each sample was rinsed with PBS solution, placed in sterile 2 mL cryotubes, and immediately snapped frozen in liquid nitrogen. Samples were kept at −80 °C freezer until analysis.

RNA extraction

A subset of 20 (CON: n = 10; RPM: n = 10) random samples were selected for RNA extraction on each tissue. TissueRuptor II (Qiagen, Germantown, MD, USA) was used to homogenize tissue fragments from the 3 tissues. Commercial kits were used for RNA extraction. Dynabeads mRNA direct kit (Cat# 61011, Thermofisher, Waltham, MA, USA), RNease lipid tissue kit (Cat#74804, Qiagen), and RNeasy fibrous tissue kit (Cat#74704, Qiagen) were used for liver, adipose tissue, and longissimus dorsi muscle RNA extraction respectively as per manufacturers guidelines. The concentration and purity of total RNA were evaluated using spectrophotometry (NanoDrop One, Thermofisher Scientific) by absorbance at 260 nm. Total RNA was stored at −80 °C until shipping for RNA sequencing (RNA-seq) analysis.

Library generation and RNA sequencing

Samples were sent for RNA Sequencing to Novogene Corporation (Sacramento, CA, USA). Quality control of the samples was assessed before library generation. Samples with an RNA Integrity Number (RIN) below 7 were excluded from the library generation. In total, 15 samples from liver tissue, 19 from adipose tissue, and 20 from longissimus dorsi muscle passed the quality control. Messenger RNA was extracted from total RNA using poly-T oligo-attached magnetic beads. Later, cDNA was synthesized using hexamer primers. cDNA underwent end repair, A-tailing adapter ligation, size selection, amplification, and purification. Illumina NovaSeq platform was used for sequencing. Once the library was ready, the original raw data from the Illumina platform was transformed into Sequenced Reads and recorded in the FASTQ file (Novogene Co).

RNA sequencing analysis

The quality of the sequencing reads was evaluated using FastQC (version 0.11.7, Babraham Bioinformatics, UK). Trimming was performed using Trim Galore (version 0.4.4, Babraham Bioinformatics) with the following parameters: –paired, –length 50, –clip_R1 15, –clip_R2 15, –three_prime_clip_R1 5, and –three_prime_clip_R2 5. After processing, reads were mapped to the bovine reference genome ARS-UCD1.2 using Hisat2 (v2.1.0) (Kim et al., 2015). The number of reads mapped to each annotated gene in the bovine annotation file was obtained using the python script htseq-count (v0.6.1p1) using the intersection-nonempty option (Anders et al., 2015). Differentially expressed genes (DEGs) between treatments were detected using the R package edgeR (Robinson et al., 2010). This software combines the Trimmed Mean of the M-values as a normalization method, an empirical Bayes approach for estimating tagwise negative binomial dispersion values, generalized linear models, and likelihood ratio tests for detecting DEGs.

Pathway analysis was performed to identify groups of functionally related genes. The main goal of this analysis is to identify gene-sets that have a significant enrichment of relevant genes. This analysis is of great importance as it would aid in gaining biological insight from the large volume of data produced by RNA-seq. In a gene-set analysis, the focus is not on single genes, but in a set of genes, applying a stringent threshold in the first step would be overly conservative. Conversely, it is crucial to use a stringent threshold when determining whether a gene-set is significantly enriched. Therefore, P-values from Fisher’s exact test should be adjusted for multiple testing. In this study, we declared genes differentially expressed due to the RPM treatment using an arbitrary threshold P-value < 0.01. Still, we declared gene-sets significantly enriched with DEGs using FDR < 0.25. In other words, we declared genes as differentially expressed, avoiding an overly conservative criterion. Pathway analysis was performed using genes that were categorized as relevant for functional clustering analysis using Database for Annotation, Visualization, and Integrated Discovery (DAVID; Sherman et al., 2022) to determine gene ontologies that were significant based on P-values from Fisher’s exact test after adjustment for multiple testing using the Benjamini-Hochberg procedure.

Statistical analyses

For continuous dependent variables, the normality of the residuals of the models and homogeneity of variances were evaluated. When a variable violated the normality assumption, it was subjected to power transformation according to the BoxCox procedure (Box and Cox, 1964). Eye cavity at day 60 of pregnancy and embryo length, were transformed, and later, the LSM and SEM were back-transformed to present results. Continuous dependent variables (PAG concentrations days 30 and 60, embryo and fetal measurements) were analyzed by linear mixed-effects models using the MIXED procedure of SAS (SAS Institute Inc.). The initial model included the fixed effect of treatment (CON or RPM) and sire, and the random effect of the block (group A or group B). Variables were removed from the model by backward elimination if P > 0.10. Treatment was forced in the final model. The final model included the fixed effect of treatment (CON or RPM) and the random effect of the block (group A or group B). The birth weight model included the fixed effect of treatment (CON or RPM), and block as a random effect.

Linear mixed-effects models for repeated measures were constructed using the MIXED procedure of SAS (SAS Institute Inc.). Repeated measurement within cow analysis was performed for AA concentration in plasma. The covariance structure used was autoregressive 1, and it was selected by the smallest AICC; the model included the fixed effect of the day of measurement (−8, 0, 7), the treatment (CON or RPM), and the interaction between treatment and day, and day was used as the repeated statement. Also, repeated measurement within cow analysis was performed for PAG concentration in plasma. The model included the fixed effect of the day of measurement (days 30 and 60), the treatment (CON or RPM), and the interaction between treatment and day, and day was used as the repeated statement. In addition, body weight, withers height, body length, and heart girth were analyzed using a repeated measures model using the MIXED procedure of SAS. All measurements were determined at 60, 120, 180, and 240 d of age, The complete model accounted for the fixed effect of adjusted age of measurement, treatment (CON or RPM), and the interaction between treatment and age of measurement and with cow nested within treatment as a random variable. Significant differences were reported for all models if P ≤ 0.05, and tendencies were considered when P ≤ 0.10.

Results

Experiment 1

Methionine concentrations in plasma increased after feeding RP-Met on day 1 and were maintained during the 5 d of the experiment (Figure 3). A treatment-by-day interaction was observed (P < 0.001), Met concentrations were the same before treatment (day 0) and diverged thereafter, with RPM being greater than CON.

Figure 3.

Methionine concentration in plasma in non-lactating cows when supplementing (RPM) or not (CON) with 15 g/d of RP-Met for 5 consecutive days.

Experiment 2

Cows that did not consume the treatment were excluded from the analysis. In total, 17 cows (CON = 11, RPM = 6) never approached the automated feeder for supplementation, and 10 (CON = 7, RPM = 3) did not consume the treatment each day. After removing cows that did not consume the treatments, pregnancy per AI was as follows: CON = 53% (20/38) and RPM = 51% (25/48). Cows that were not pregnant from TAI were removed from the study. Also, one cow from the RPM treatment was removed for having an abortion. Following parturition, 4 cows (CON = 1, RPM = 3) rejected their calves, therefore, they were removed from the postnatal analysis. Therefore, a total of 45 (CON: n = 22; RPM: n = 23) pregnant cows were considered for PAGs analysis. Plasma from pregnant cows (n = 34; CON: n = 16; RPM: n = 18) that were inseminated by TAI and had female calves were analyzed for AA concentration. Data from all live calves was used for birth weight analysis and only female offsprings were considered for postnatal measurements and analyses.

Plasma concentrations of AA during the periconceptional period (days −8, 0, and 7 relative to AI) are presented in Table 2. A treatment-by-day interaction was observed for Met (P < 0.0001). Concentrations of Met were not different at day −8 (CON: 13.0 ± 1.4 µM; RPM: 13.7 ± 1.2 µM) but were higher for RPM at day 0 (CON: 14.1 ± 1.4 µM; RPM: 24.3 ± 1.3 µM) and day 7 (CON: 13.1 ± 1.4 µM; RPM: 19.4 ± 1.2 µM). Moreover, day had an effect (P < 0.05) on arginine (Arg; day −8: 79.2 ± 2.8 µM; day 7: 90.2 ± 2.8 µM), isoleucine (Ille; day −8: 116.8 ± 4.3 µM; day 0: 126.9 ± 4.3 µM), phenylalanine (Phe; day −8: 37.9 ± 1.4 µM; day 0: 41.6 ± 1.3 µM), threonine (Thr; day −8: 67.1 ± 2.3 µM; day 0: 60.1 ± 2.2 µM), tryptophan (Trp; day −8: 33.3 ± 1.4 µM; day 0: 28.3 ± 1.4 µM), valine (Val; day 0: 207.6 ± 8.3 µM; day 7: 225.5 ± 8.3 µM) and alanine (Ala; day −8: 297.0 ± 10.6 µM; day 0: 270.4 ± 10.5 µM).

Table 2.

Effect of supplementing rumen-protected methionine (RPM) or not (CON) in plasma AA concentrations of beef cows during the periconceptional period

	Treatment1 × day (CON)			Treatment × day (RPM)			P-value2
AA, µM	−8	0	7	−8	0	7	Trt	Day	Trt × day
EAA
Arg	77.6 ± 4.2	82.5 ± 4.1	91.9 ± 4.2	80.8 ± 3.7	86.7 ± 3.8	88.8 ± 3.7	0.69	0.02	0.55
His	38.9 ± 1.9	38.9 ± 1.8	41.7 ± 1.9	42.4 ± 1.7	40.2 ± 1.7	42.0 ± 1.7	0.37	0.25	0.59
Ile	114.5 ± 6.4	127.9 ± 6.3	125.5 ± 6.5	119.2 ± 5.7	126.0 ± 5.8	119.7 ± 5.7	0.89	0.05	0.59
Leu	59.2 ± 3.1	63.5 ± 3.1	65.3 ± 3.2	63.8 ± 2.8	65.0 ± 2.8	65.3 ± 2.8	0.55	0.31	0.65
Lys	80.5 ± 5.2	78.0 ± 5.0	89.0 ± 5.2	85.4 ± 4.6	81.8 ± 4.7	83.1 ± 4.6	0.85	0.21	0.35
Met	13.0 ± 1.4	14.11 ± 1.2a	13.1 ± 1.4c	13.7 ± 1.2	24.3 ± 1.3b	19.4 ± 1.2d	<0.01	<0.01	0.01
Phe	37.6 ± 2.0	42.6 ± 2.0	38.4 ± 2.0	38.3 ± 1.8	40.7 ± 1.8	36.7 ± 1.8	0.67	0.001	0.59
Thr	61.9 ± 3.4	58.2 ± 3.3	61.9 ± 3.4	72.2 ± 2.3	62.1 ± 3.1	63.0 ± 3.0	0.11	0.11	0.31
Trp	31.3 ± 2.14	27.0 ± 2.1	32.1 ± 2.1	35.3 ± 1.9	29.6 ± 1.9	30.5 ± 1.9	0.47	0.01	0.23
Val	198.1 ± 12.4	206.1 ± 12.1	209.0 ± 11.0	209.0 ± 11.0	209.1 ± 11.2	218.9 ± 11.0	0.99	0.04	0.44
Total EAA	708.4 ± 32.3	727.7 ± 31.3	783.6 ± 32.3	757.4 ± 28.4	762.4 ± 29.0	764.7 ± 28.4	0.50	0.31	0.42
NEAA
Ala	296.3 ± 15.9	278.4 ± 15.5	281.9 ± 15.9	297.8 ± 14.1	262.5 ± 14.3	274.5 ± 14.1	0.67	0.04	0.68
Asn	21.3 ± 1.4	23.2 ± 1.3	22.3 ± 1.4	23.5 ± 1.2	23.3 ± 1.2	21.9 ± 1.2	0.60	0.51	0.54
Asp	4.0 ± 0.8	4.8 ± 0.8	2.4 ± 0.8	3.1 ± 0.7	2.4 ± 0.8	3.2 ± 0.7	0.55	0.15	0.25
Gln	170.2 ± 8.3	174.2 ± 8.0	170.0 ± 8.3	185.3 ± 7.3	178.5 ± 7.5	177.0 ± 7.3	0.28	0.83	0.68
Glu	50.2 ± 4.1	49.2 ± 4.1	46.6 ± 4.1	55.2 ± 3.7	50.1 ± 3.7	48.6 ± 3.7	0.58	0.21	0.66
Gly	552.5 ± 31.8	613.0 ± 31.1	588.3 ± 31.8	572.5 ± 28.2	557.7 ± 28.6	508.7 ± 28.2	0.28	0.12	0.09
Pro	59.7 ± 3.2	63.6 ± 3.1	64.1 ± 3.2	62.1 ± 2.8	57.3 ± 2.9	58.2 ± 2.8	0.28	0.96	0.20
Ser	69.0 ± 4.1	71.8 ± 4.0	71.7 ± 4.1	70.7 ± 3.6	64.0 ± 3.7	64.0 ± 3.6	0.27	0.77	0.24
Tyr	53.2 ± 6.1	55.6 ± 5.6	68.9 ± 5.8	61.2 ± 5.1	56.2 ± 5.3	55.0 ± 5.1	0.74	0.45	0.12
Total NEAA	1275.3 ± 53.6	1338.2 ± 52.3	1319.8 ± 53.6	1333.1 ± 47.3	1253.5 ± 48.2	1212.8 ± 47.3	0.43	0.63	0.10
Total AA	1986.8 ± 76.8	2068.2 ± 74.7	2104.8 ± 76.8	2091.4 ± 67.6	2016.2 ± 69.0	1978.5 ± 67.6	0.75	0.99	0.18

^a,hDifferent lowercase letters indicate P < 0.05 Differences shown are the interaction between Trt × Day.

¹CON = 454 g/d corn gluten feed; RPM = 454 g/d corn gluten feed + 15 g/d RP-Met.

²Trt = treatment; Trt × Day = treatment by day interaction.

No difference (P > 0.05) in PAG concentration due to treatment were observed at day 30 (CON: 16.46 ± 0.6 ng/mL; RPM: 16.5 ± 0.6 ng/mL) or 60 of pregnancy (CON: 10.1 ± 0.7 ng/mL; RPM: 9.9 ± 0.6 ng/mL). However, an effect of time was observed, were PAG concentration decreased (P < 0.001) from day 30 (16.48 ± 0.46 ng/mL) to day 60 (9.96 ± 0.46 ng/mL).

There was a tendency (P = 0.06) for amniotic vesicle circumference to be larger in the RPM treatment but differences in other measurements of fetal size were not significantly different between treatments (Table 3).

Table 3.

Embryo and fetal measurements at 30 and 60 d of pregnancy for cows supplemented (RPM) or not (CON) with RP-Met

	CON	RPM	P-value
Day 30	n = 15	n = 17
Amniotic vesicle diameter, mm	10.96 ± 0.24	11.48 ± 0.23	0.13
Amniotic vesicle circumference, cm2	0.88 ± 0.04	0.99 ± 0.04	0.06
Embryo length, mm	10.97 ± 0.41	10.98 ± 0.40	0.98
Abdominal cavity, mm	5.88 ± 0.17	6.04 ± 0.17	0.51
Day 60	n = 12	n = 15
Eye cavity diameter, mm	5.97 ± 0.29	5.55 ± 0.32	0.28
Wither-tail length, mm	41.47 ± 1.44	41.80 ± 1.22	0.86
Head length, mm	26.77 ± 0.55	27.23 ± 0.53	0.55
Head transversal, mm	19.76 ± 0.49	20.65 ± 0.45	0.34

Postnatal traits

No significant difference was observed in the birth weight of female calves (n = 34; CON: 31.3 ± 1.04 kg [n = 16]; RPM: 33.3 ± 1.0 kg [n = 18]). Three male calves from RPM (43.7 ± 2.8 kg) and 3 from CON (32.6 ± 2.0 kg) were weighted at birth but were removed from the study.

Female calves were considered for further analysis. A treatment effect was observed in withers height (P = 0.03) with taller calves at 60 and 120 d of age in the RPM treatment (Figure 4B). However, there was no interaction (P > 0.05) between treatment and age. There was also no effect of treatment or treatment-by-age on body weight, body length, or heart girth (Figure 4). Furthermore, 205-adj weaning weight was not affected by treatment (P = 0.71; CON: 199.0 ± 5.2 kg; RPM: 201.6 ± 4.6 kg).

Figure 4.

Body weight, whither height, heart girth, and body length in female progeny from cows supplemented (Con) or not (CON) with RP-Met during the periconceptional period.

RNA-Seq analyses

Liver

On average, 24M paired-end reads were generated from the liver samples. Around 93% of the reads were successfully mapped to the ARS-UCD1.2 bovine reference genome. A total of 30 DEGs were observed in the liver (Figure 5), from which 10 genes were upregulated and 20 were downregulated in the RPM treatment (Table 4). Among the 10 upregulated genes in the RPM treatment, 2 genes [retinol binding protein 5 (RBP5) and MID1 interacting protein 1 (MID1IP1)] are related to liver function, and one gene [lipocalin 12 (LCN12)] is an immune response-related gene. Also, several genes upregulated by RPM were involved in important cellular regulation process, such as repair of DNA damage [X-ray repair cross-complementing 3(XRCC3)], cytoskeletal organization, cell adhesion, and migration and signaling [sorbin and SH3 domain containing 3 (SORBS3)], cell cycle progression and differentiation [S100 calcium-binding protein A10 (S100A10)], cell division and growth [glypican 3 (GPC3)], and cell growth, senescence and cell differentiation [inhibitor of DNA binding 1 (ID1)]. Among the 20 downregulated genes in the RPM treatment, several genes are immune response-related genes [C-X-C motif chemokine ligand 10 (CXCL10), T cell immunoreceptor with Ig and ITIM domains (TIGIT), interferon induced with helicase C domain 1 (IFIH1), indoleamine 2,3-dioxygenase 1 (IDO1), interleukin 1 receptor-associated kinase 3 (IRAK3), and actin gamma 1 (ACTG1)]. Also, there are 2 enzymes related genes [phospholipase A2 group IVA (PLA2G4A) and meprin A subunit beta (MEP1B)], and 2 genes related to angiogenesis regulation [secreted frizzled-related protein 1 (SFRP1) and tryptophanyl-tRNA synthetase 1 (WARS1)].

Figure 5.

Volcano plot of the log2 fold change against the -log10(FDR). Illustration of all significant genes (FDR < 0.25) between CON and RPM in liver of female offspring. The plot shows 20 downregulated genes on the left (blue dots) and 10 upregulated genes on the right (red dots) in RPM female offspring.

Table 4.

DEGs in liver samples (FDR < 0.25) detected by RNAseq

ENSemBLe ID	Gene symbol	log2 fold change	P-value	FDR
Upregulated Liver in RPM
ENSBTAG00000009966	XRCC3	0.78	0.00	0.25
ENSBTAG00000014401	SORBS3	0.81	0.00	0.19
ENSBTAG00000015147	S100A10	0.83	0.00	0.22
ENSBTAG00000008127	RBP5	0.98	0.00	0.15
ENSBTAG00000011463	MID1IP1	1.05	0.00	0.06
ENSBTAG00000020406	GPC3	1.11	0.00	0.13
ENSBTAG00000038756	GLB1L3	1.12	0.00	0.22
ENSBTAG00000039531	LCN12	1.31	0.00	0.15
ENSBTAG00000054363		1.35	0.00	0.16
ENSBTAG00000016169	ID1	1.79	0.00	0.13
Downregulated Liver in RPM
ENSBTAG00000023563	LOC786065	−1.63	0.00	0.12
ENSBTAG00000027625	SFRP1	−1.53	0.00	0.13
ENSBTAG00000010954	ART3	−1.40	0.00	0.13
ENSBTAG00000001725	CXCL10	−1.40	0.00	0.12
ENSBTAG00000034662	LOC613316	−1.34	0.00	0.19
ENSBTAG00000011921	TIGIT	−1.26	0.00	0.15
ENSBTAG00000017670	LOC511531	−1.18	0.00	0.13
ENSBTAG00000020391	MEP1B	−1.14	0.00	0.13
ENSBTAG00000050253	TRPM3	−1.07	0.00	0.19
ENSBTAG00000008142	IFIH1	−1.00	0.00	0.22
ENSBTAG00000020602	IDO1	−0.97	0.00	0.16
ENSBTAG00000013298	PLA2G4A	−0.94	0.00	0.22
ENSBTAG00000017740	TUBAL3	−0.93	0.00	0.19
ENSBTAG00000013662	COL8A1	−0.92	0.00	0.19
ENSBTAG00000018424	ACKR3	−0.90	0.00	0.19
ENSBTAG00000054314	LOC786706	−0.85	0.00	0.15
ENSBTAG00000007636	IRAK3	−0.85	0.00	0.16
ENSBTAG00000053354	CYP2C85	−0.79	0.00	0.12
ENSBTAG00000006189	ACTG1	−0.69	0.00	0.19
ENSBTAG00000004679	WARS1	−0.60	0.00	0.13

All 30 DEGs were used to conduct the enrichment analysis in DAVID. To conduct the analysis 10 upregulated and 20 downregulated genes were separated. Upregulated genes did not enhance any GO-term process, while downregulated genes were associated with 7 different enriched processes (Table 5). Additionally, analysis by DAVID indicated that the functional clusters with the highest enrichment scores were those related with intracellular mechanisms processes (9 genes, enrichment score: 1.53).

Table 5.

Group of the gene ontology terms in enriched processes in liver

Tissue treatment	Enriched process	Genes	Benjamini
Liver: downregulated in RPM	Biological process involved in interspecies between organisms	CXCL10, ACKR3, LOC786706, IDO1, IFIH1, IRAK3	0.004
	Response to external stimulus	CXCL10, ACKR3, LOC786706, LOC511531, IDO1, IFIH1, IRAK3	0.004
	Response to other organism	CXCL10, LOC786706, LOC511531, IDO1, IFIH1, IRAK3	0.004
	Catalytic activity	ART3, ACTG1, LOC786706, LOC511531, IDO1, IFIH1, IRAK3, MEP1B, PLA2G4A, WARS1, TUBAL3	0.01
	Immune system process	CXCL10, ACKR3, LOC511531, IDO1, IFIH1	0.01
	Regulation of interleukin-12 production	TGIT, IDO1, IRAK3	0.03
	Positive regulation of cytokine production	TGIT,IDO1, IFIH1, IRAK3	0.05

Longissimus dorsi

On average, 28M paired-end reads were generated from the muscle samples. Roughly 92% of the reads were successfully mapped to the ARS-UCD1.2 bovine reference genome. A total of 24 DEGs were observed (Figure 6), from which 14 genes were upregulated, and 9 genes were downregulated in the RPM treatment (Table S1). Among the 14 upregulated genes in the RPM treatment, 4 genes [CD151 molecule (CD151), zinc finger MIZ-type containing 1 (ZMIZ1), mitogen-activated protein kinase 14 (MAP3K14), CD209 molecule (CD209)] are immune response-related genes. Also, 3 genes are related to cellular regulation process: Cytoskeleton constitution [keratin 19 (KRT19)], membrane regulation [GRAM domain containing 2A (GRAMD2A)], and nuclear-cytoplasmatic connection [Sad1 and UNC84 domain containing 2 (SUN2)]. Additionally, one gene is a lipid metabolic process-related gene [glutathione peroxidase 4 (GPX4)], one gene is a DNA binding transcription factor-related gene [zinc finger protein 500 (ZNF500)], and one gene is Wnt signaling pathway-related [secreted frizzled-related protein 5 (SFRP5)]. Upregulated and downregulated genes in Longissimus dorsi did not enhance any GO-term process with DAVID.

Figure 6.

Volcano plot of the log2 fold change agains the -log10(FDR). Illustration of all significant genes (FDR < 0.25) between CON and RPM in longissimus dorsi muscle of female offspring. The plot shows 9 downregulated genes on the left (blue dots) and 14 upregulated genes on the right (red dots) in RPM female offspring.

Adipose tissue

On average, 24M paired-end reads were generated from the adipose samples. Around 92% of the reads were successfully mapped to the ARS-UCD1.2 bovine reference genome. Only 2 DEGs were observed, from which both are upregulated genes in the RPM treatment (Table S2). Among the 2 upregulated genes, one gene is an immune response-related gene [FOS Like 1, AP-1 Transcription Factor Subunit (FOSL1)]. The other gene [metallothionein-1A (MT1A)] is related to encode for proteins involved in binding heavy metals, in metal ion homeostasis and detoxification in the cell and in decreasing oxidative stress. Upregulated and downregulated genes in adipose tissue did not enhance any GO-term process with DAVID.

Discussion

In this study, our findings contribute to the idea on how maternal nutrition during the periconceptional period can influence the offspring. In the current study, we defined the periconceptional period from days −7 to 7 relative to TAI day. Previously Acosta et al. (2016) reported that using RP-Met during the periconceptional period in Holstein cows decreased blastocyst methylation levels. Additionally, Peñagaricano et al. (2013) showed that feeding dairy cows RP-Met during the periconceptional period (from calving to embryo flush) led to the modulation of gene expression in the bovine blastocyst stage. Whether the modulation of gene expression and decreased methylation in the blastocyst yield advantageous outcomes for the progeny in beef cattle was unknown until we conducted this study. We found that supplementation of RP-Met during this period caused female heifers to be taller than the CON treatment and can have an effect in gene expression of liver and adipose tissue.

Amino acids concentration analysis was conducted in plasma samples to ensure that Met levels in the cows increased when feeding 15 g of RP-Met. In both experiments, we observed increased plasma Met levels after feeding 15 g of RP-Met. Studies in dairy cows have shown increased Met concentration in plasma when supplementing RP-Met. Recently, Toledo et al. (2021) evaluated the effect of supplementing 12 g and 27 g of RP-Met in plasma AA when supplementing to pre- and postpartum Holstein cows, respectively. They observed elevated Met concentrations in plasma 7 d before parturition (CON = 22.9 µM vs. RPM = 25.9 µM), and at 7 d (CON = 21.2 µM vs. RPM = 31.0 µM), 14 d (CON = 19 µM vs. RPM = 30.5 µM), and 21 d (CON = 17.8 µM vs. RPM = 31 µM) after parturition. Additionally, supplementation of RP-Met decreased the concentration of Leu, Val, Phe, Asn, Ser, and Tyr at the periparturient period. In a different study with lactating (from 30 until 126 d in milk) dairy cows Toledo et al. (2017) reported an increase of Met levels in plasma when RP-Met was top fed once a day at 6 (21.4 ± 5.3 μM vs. 34.5 ± 1.1 µM), 9 (24.2 ± 1.8 μM vs. 49.5 ± 2.9 µM), 12 (26 ± 3.3 µM vs. 52.4 ± 2.6 µM) and 18 (25.2 ± 2.5 μM vs. 36.4 ± 2.0) h after supplementing 21.2 g of RP-Met. To the extent of our knowledge, the only study in beef cattle that reports Met plasma concentrations after feeding RP-Met was reported by Waterman et al. (2012). In that study, bred heifers were fed 23.5 g/d of RP-Met individually for 44 d, and they observed an increase in Met levels in plasma (CON: 16.5 µM/L vs. RPM: 26.8 µM/L) comparable to the current experiment. The previous results and our results allowed us to determine that feeding 15 g of RP-Met (9.12 g metabolizable Met) does increase Met levels in plasma in beef cows and does not alter concentrations of other essential and non-essential AAs.

Feeding 15 g of RPM during the periconceptional period did not increase embryonic or fetal size. In the present study, no difference was observed in any measurements at 30 and 60 d of pregnancy. However, a tendency to have a greater amniotic vesicle circumference was observed in the RPM treatment. Toledo et al. (2017) reported an increase (CON = 5.5 mm vs. RPM = 5.8 mm) in fetal abdominal diameter and a tendency of a longer crown-rump length (CON = 10.5 mm vs RPM = 11 mm) on day 33 of the pregnancy in cows supplemented with 21.2 g RP-Met, starting 40 d before AI until 32 d of pregnancy. However, Stangaferro et al. (2021) observed similar results to ours when supplementing Holstein cows with RP-Met starting 4 weeks before parturition (12 g RP-Met) until 147 d postpartum (27 g RP-Met). They did not identify differences in embryo crown-rump length and abdominal diameter on day 32 of the pregnancy. Having contrasting results suggests the need for further research to determine if RP-Met is causing or not to have larger embryos. Additionally, both Toledo et al. (2017) and Stangaferro et al. (2021) used Holstein cows as their model, it is possible that the response in beef cattle is different from dairy cows and is important to consider that the feeding periods are different in both studies and this study.

Feeding 15 g/d of RP-Met in the periconceptional period did not increase birth weight in female offspring, nor was the 205-adj weaning weight. However, although female-sexed semen was used for TAI, a limited number of male calves were born in our experiment. Notably, a numerically difference in birth weight was observed only in male calves, underscoring the importance of further research on male progeny and a possible sexual dimorphism. Silva et al. (2021) reported no differences in birth weight or 205-adj weaning weight in male and female beef calves when supplementing with 10 g/d of a liquid bypass methionine (Metasmart Liquid, Adisseo) to grazing cows during the periconceptional period, which they defined as −64 to 51 d postconception, for a total of 115 d of supplementation. The use of other methyl donors during the first stages of embryogenesis has been assessed by Estrada-Cortés et al. (2021) and Haimon et al. (2024), by using 1.8 mM of choline chloride in embryo culture medium. Estrada-Cortés et al. (2021) reported an increase in birth weight and 205-d adjusted weaning weight for males and females in the choline treatment, while Haimon et al. (2024) reported an increase in 205-d adjusted weaning weight and also a tendency to have taller animals at weaning. It is possible that the varying body weights associated with choline are primarily a result of its direct application to the embryo medium, additionally as postulated by Estrada-Cortés et al. (2021) one of the actions of choline during embryo culture increased gestation length, and it has been associated with higher birth weights. In our study, we presumed that methionine concentrations increase in the uterine luminal fluid, although we have yet to confirm this occurrence. Additionally, it is important to consider that choline and methionine are distinct molecules that may elicit different effects, it still remains unknown if methionine is causing changes in DNA methylation to the embryo when supplementing RP-Met.

Supplementing RP-Met during the periconceptional period caused 30, 24, and 2 DEGs in the liver, Longissimus dorsi muscle, and adipose tissue, respectively in the female offspring from cows supplemented with RP-Met. Despite the absence of analyzing any potential changes in DNA methylation in our study, a noteworthy finding was the identification of 30 DEGs in the liver. Two upregulated genes, RBP5 (Bahar et al., 2007) and MID1IP1 (Higgins et al., 2019), associated with liver function, have previously been influenced by dietary changes. In cattle, Higgins et al. (2019) observed a downregulation in the MID1IP1 gene when feeding a high-concentrate diet and a zero-grazed grass diet to Charolais steers. Conversely, several immune response-related genes were downregulated in the liver in the RPM group, including CXCL10, responsible for chemokine production and induction of migration of natural killer and T cells (Tokunaga et al., 2018), IFIH1, associated with innate immune response, IRAK3, involved in signaling pathways for innate immunity (Freihat et al., 2019), TIGIT, a gene expressed on activated T cells and natural killer cells, it also acts indirectly by suppressing immune responses (Anderson et al., 2016), and IDO1, encoding an enzyme contributing to immune response homeostasis (Van Baren and Van Den Eynde, 2015). Peñagaricano et al. (2013), observed a decreased expression of immune response-related genes in bovine blastocyst collected from cows supplemented with RP-Met from calving until embryo flushing. However, it is important to clarify that the DEGs we observed in liver in the present study did not match with the DEGs from Peñagaricano’s study (2013) observed in the bovine blastocyst.

In the Longissimus dorsi muscle, the transcriptomic analysis revealed more DEGs upregulated in response to the RPM treatment. Notably, these upregulated genes play crucial roles in muscle development. For instance, KRT19 has been identified as a muscle intermediate filaments protein in mice (Muriel et al., 2020). Interestingly, upregulation of KRT19 was observed in Longissimus thoracis muscle in steers supplemented with protein from 100 to 200 d of gestation (Carvalho et al., 2022). CD151, which encodes a protein forming the CD151-integrin complex, was also upregulated in RPM, and studies in rats have linked this complex to angiogenesis and signaling pathways, which are important for muscle development (Liu et al., 2011; Guo et al., 2015). GPX4, also upregulated in RPM, encodes a protein that acts as a cellular defense against oxidative damage in bovine skeletal muscle and it has been identified to be important for the adaptive response of an animal to oxidative stress (Brennan et al., 2009). Moreover, SUN2 is an inner nuclear membrane protein interacting with KASH proteins in the nuclear envelope. In mice, this SUN2 has been identified in the skeletal muscle cells, playing an important role in Syne-1 localization at the nuclear envelope, which is essential for myonuclear positioning (Lei et al., 2009). Additionally, genes associated with immune response were upregulated in the RPM treatment, including ZMIZ1 (Ben Khalaf et al., 2021), identified as important for T cell activation in mice, MAP3K14 (Hamdan et al., 2020), a key mediator in viral infection promoting immune activation, and CD209 (McGreal et al., 2005), which encodes a transmembrane receptor crucial for pathogen recognition on macrophages and dendritic cells. Previously, Liu et al. (2020) and Liu et al.(2021) supplemented 10 g/d of a liquid methionine (Metasmart Liquid, Adisseo) to grazing cows during the periconceptional period (−30 to +90 d postconception). Longissimus dorsi muscle samples were collected from male calves at 1 mo. They reported an impact in splicing patterns, and gene co-expression patterns and were associated with changes in DNA methylation. Furthermore, Amorín et al. 2023 provided evidence that supplementation of 10 g/d of a liquid methionine (Metasmart Liquid, Adisseo) to grazing cows during the periconceptional period (defined as −30 to +90 d relative to breeding) induced changes in DNA methylation in muscle of calves at 30 and 200 d of age. In addition to results supplementing Met in beef cows, Haimon et al. (2024) reported that the addition of 1.8 mM of choline chloride to the embryo medium increased a 15% or greater in CpG methylation when collecting blood samples from calves at weaning. We can speculate that changes in gene expression in our experiment are related to changes in DNA methylation, however, this should be investigated to conclude if this was the case.

In summary, our study has identified the effect of supplementing 15 g of RP-Met during the periconceptional period (7 d before and 7 d after the artificial insemination) on the pre-and postnatal characteristics of female offspring, as well on the gene expression of the liver, Longissimus dorsi muscle, and adipose tissue. The supplementation of 15 g/d of RP-Met was enough to increase the Met concentration in plasma in cows; however, it remains unknown if this effect is also observed in the uterine luminal fluid. Interestingly, in this study, we observed heavier male calves from cows that received RP-Met. This observation should be investigated in future studies aimed at confirming whether males and females respond differently to RP-Met supplementation Furthermore, this study reveals the effect of RP-Met on DEG in the liver, Longissimus dorsi muscle, and adipose tissue in female progeny.

In conclusion, supplementing 15 g/d of RP-Met did not alter embryo development or fetal morphology. Whiter height in female progeny from the RPM treatment until weaning was enhanced. However, the economic impact of having taller animals at weaning in cow–calf operations is minimal. Additionally, different gene pathways were observed to be downregulated in the liver. It remains unknown if DNA methylation is the mechanism causing changes in gene expression. The results presented in this study allow us to conclude that supplementation of 15 g/d RP-Met during the periconceptional period causes postnatal changes in the female offspring.

Glossary

Abbreviations

AA

amino acids

BW

body weight

CIDR

controlled internal drug release

Met

methionine

P/AI

pregnancy per artificial insemination

PAG

pregnancy-associated glycoprotein

RNA-seq

RNA sequencing

RP-Met

rumen-protected methionine

TAI

timed-artificial insemination

US

ultrasonography

Conflict of interest statement

The authors declare the following financial interests/personal relationships, which may be considered as potential competing interests: A.M.G.-D., P.J.H., and N.DL. report that financial support was provided by Adisseo USA Inc. to fund this project. D.L. is employed by Adisseo USA Inc. The remaining authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Author contributions

Daniella Heredia (Data curation, Formal analysis, Investigation, Writing—original draft), Federico Tarnonsky (Formal analysis, Investigation, Writing—review & editing), Maria Lopez-Duarte (Investigation), Mauro Venturini (Investigation), Federico Podversich (Formal analysis, Investigation), Oscar Ojeda-Rojas (Investigation, Methodology), Francisco Peñagaricano (Data curation, Formal analysis, Investigation, Software), Ricardo Chebel (Investigation), Daniel Luchini (Conceptualization, Funding acquisition, Writing—review & editing), Peter Hansen (Conceptualization, Funding acquisition, Writing—review & editing), Nicolas DiLorenzo (Conceptualization, Funding acquisition, Supervision, Writing—review & editing), and Ángela M. Gonella-Diaza (Conceptualization, Data curation, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Writing—review & editing)