Prenatal Exposure to Different Nitrogen Supplementation Strategies in Beef Cows Enhances Foetal Growth and Induces Transcriptional Adaptations in Offspring Skeletal Muscle

ABSTRACT

This study aimed to assess the effects of different nitrogen supplementation strategies during mid‐gestation on postnatal performance, skeletal muscle gene expression and morphology and nutritional and metabolic parameters of beef offspring during both the cow‐calf and backgrounding phases. Thirty Zebu cows (532 ± 11 kg of body weight [BW]) were used. The following treatments were randomly assigned to cows from 127 ± 17 to 227 ± 17 days of gestation: (1) Control (CON; n = 10): basal diet (corn silage + sugarcane bagasse) and a mineral supplement with urea (60 g/100 kg BW); (2) Rumen‐degradable protein (RDP; n = 10): basal diet plus a commercial protein supplement (250 g/100 kg BW) and ground corn (44 g/100 kg BW); and (3) Rumen‐undegradable protein (RUP; n = 10): basal diet plus protected soybean meal (3.5 g/kg BW) and a mineral supplement with urea (60 g/100 kg BW). After supplementation, cows were managed as a single group on pasture. Offspring were evaluated during the cow–calf phase, and in a 60‐day feedlot backgrounding phase. At the end of the supplementation period (227 days of gestation), cows in the RUP treatment had greater BW compared to those in the CON and RDP treatments (p = 0.01). Near parturition (282 days of gestation), no differences in BW were observed among treatments (p = 0.28). Birth weight was greater in the RUP group compared to the CON and RDP groups (p = 0.04), with no differences between the CON and RDP. No effects of treatment were observed on offspring weaning weight, average daily gain, ultrasound carcass traits, nutrient intake, apparent total tract digestibility, ingestive behaviour or muscle fibre number and histomorphometry (p > 0.05). At 346 ± 31 days of age, RUP offspring exhibited increased expression of skeletal muscle genes, including ACACA, FASN, PPARG, CPT2, IGF1R and COL3A1 compared to the CON group (p ≤ 0.05). Additionally, RUP offspring showed greater expression of ACACA, FABP4, PPARA, SCD1, COL3A1 and IGFR1 compared to RDP offspring (p ≤ 0.04). In conclusion, maternal supplementation with rumen‐protected protein during mid‐gestation enhances foetal growth and induces beneficial transcriptional adaptations in skeletal muscle compared to non‐protein or RDP supplementation strategies, without affecting postnatal growth performance.

1. Introduction

In pasture‐based beef production systems across tropical and subtropical regions, pregnant cows often face nutritional challenges that may compromise foetal development and long‐term productivity of their offspring (dos Santos et al. ²⁰²²). Among these constraints, protein deficiency—particularly prevalent when cows graze on low‐quality forages—stands out as a major limiting factor (Reis et al. 2009).

Recently, the concept of foetal programming has emerged as a key biological mechanism linking maternal nutrition to the developmental trajectory of ruminant offspring (dos Santos et al. ²⁰²³; Miranda et al. ²⁰²³). Over the past decade, an expanding body of research has highlighted the central role of maternal protein supplementation in modulating this programming process, with consistent benefits observed in both maternal physiology and offspring performance, particularly in Zebu cattle (Carvalho et al. 2022; Costa et al. ²⁰²¹; Lopes et al. ²⁰²⁰; Mageste De Almeida et al. ²⁰²¹; Marquez et al. ²⁰¹⁷; Meneses et al. ^{2022, 2024}; Nascimento et al. ^{2022, 2024}; Nepomuceno et al. ²⁰¹⁷; Rodrigues et al. ²⁰²¹).

Notably, supplementing gestating cows with protein sources has been shown to restore ruminal nitrogen balance, stimulate microbial protein synthesis and enhance nutrient digestibility and intake (Meneses et al. 2024). These effects collectively improve the maternal nutritional and energetic status and ensure a more consistent supply of amino acids and gluconeogenic substrates to the foetus—an essential support mechanism in scenarios of nutritional restriction (Meneses et al. 2022). Consequently, maternal tissue catabolism is reduced, and postnatal growth trajectories are improved, mitigating the risk of developmental impairments (Nascimento et al. 2024).

Despite these advances, significant knowledge gaps remain, particularly regarding the use of rumen‐undegradable protein (RUP) during gestation. While most studies have focused on rumen‐degradable protein (RDP), evidence from isotopic tracer studies using [¹⁵N¹⁵N]‐urea suggests that RUP supplementation may simultaneously enhance the supply of intestinally available amino acids and stimulate nitrogen recycling, which synergistically benefits microbial protein synthesis in the rumen (Batista et al. 2016). These dual mechanisms position RUP as a promising nutritional strategy for improving maternal‐foetal outcomes in low‐protein environments.

Moreover, the use of RUP sources with targeted amino acid profiles, such as protected soybean meal, is particularly relevant. Rich in arginine, this feedstuff supports key gestational processes including placental angiogenesis, cellular signalling and blood flow regulation (L. C. O. Sousa, Matos, et al. ²⁰²⁴). In line with this, Costa et al. (2022) demonstrated that cows fed a high‐RUP diet during late gestation had greater body weight (BW) and average daily gain, along with improved foetal muscle development and intramuscular adipogenesis—likely a consequence of favourable epigenetic modifications in mesenchymal progenitor cells.

Given this background, the present study aims to investigate the effects of three distinct supplementation programmes during mid‐gestation, comprising low, medium and high protein levels provided by different nitrogen sources, on the performance, metabolic status and physiological development of the offspring. We hypothesize that supplementation with RUP during gestation will result in greater foetal growth and subsequent changes in skeletal muscle gene expression related to lipid metabolism and growth signalling in the offspring.

2. Materials and Methods

2.1. Animal Welfare

This study was performed at the Federal University of Lavras' beef cattle facilities, located in Lavras, Minas Gerais, Brazil. The UFLA Ethics Committee on Animal Use approved all management procedures for this study (protocol number 015/2019).

2.2. Management of Cows and Their Diets During Gestation

Thirty multiparous Zebu beef cows [Bos taurus indicus; 532 ± 11 kg of BW, 6 ± 0.5 years of age] were used in this study. Cows were submitted to a fixed‐time artificial insemination protocol with semen from four Zebu males. At 105 days of gestation, cows were housed in individual stalls (20 m² area, 6 m² of it covered by a roof, with a concrete floor and equipped with individual feeding and watering troughs) and were acclimated to a basal diet consisting of corn silage (82.3% DM) and sugarcane bagasse (17.7% DM) for approximately 15 days. At 127 ± 17 days of gestation, the treatments were randomly assigned to the cows, as follows: (1) Control (CON, n = 10): basal diet + commercial mineral supplement containing urea (Probeef Urea, Cargill Animal Nutrition, Itapira, SP, Brazil) at a level of 60 g/100 kg of BW (to simulate a production system of pregnant beef cows on pastures in tropical countries during the dry season of the year, using low supplementation); (2) Ruminally degradable protein supplement (RDP, n = 10): basal diet + 250 g/100 kg of BW of a commercial protein supplement (Probeef Nutripec Sprint, Cargill Animal Nutrition, Itapira, SP, Brazil) + 44 g of corn finely ground per 100 kg BW (to represent an average level of supplementation); (3) Ruminally undegradable protein supplement (RUP, n = 10): basal diet supplemented with 3.5 g/kg of BW of a commercial undegradable ruminal protein product containing protected soybean meal (Soypass, Cargill Animal Nutrition, Itapira, SP, Brazil) added to a commercial mineral supplement containing urea (Probeef Urea, Cargill Animal Nutrition, Itapira, SP, Brazil) at a level of 60 g per 100 kg of BW (Table 1).

Table 1.

Ingredients, chemical composition and estimated fulfilment of nutrient requirements of beef cows from 100 to 200 days of gestation.

Item	Mid gestation (100−200 days of gestation)
	CON	RDP	RUP
Ingredients
Corn silage, % of DM	79	68.3	65
Sugarcane bagasse, % of DM	20	16.8	15
Ground corn, % of DM	—	2.9	—
Supplement, kg of DM per day
Probeef ureaa	1	—	2
Probeef nutripec sprintb	—	12	—
Soypass	—	—	18
Chemical composition
Dry matter, g/kg of DM	387.71	446.05	482.23
Organic matter, g/kg of DM	956.33	931.13	940.18
Crude protein, g/kg of DM	77.52	116.1	162.43
Rumen degradable protein, g/kg of DM	59.0	98.4	82.2
Rumen undegradable protein, g/kg of DM	18.5	17.6	80.2
Ash and protein‐free neutral detergent fibre, g/kg of DM	545.23	471.63	493.09
Ether extract, g/kg of DM	37.59	53.23	34.01
Total digestible nutrients, g/kg of DM	591.18	585.55	629
Estimated fulfilment of maternal nutrient requirements (%)
Total digestible nutrients	83	104	116
Crude protein	59	104	163
Rumen‐degradable protein	69	120	120
Rumen‐undegradable protein	39	70	251

^a Probeef urea: Assurance level per kilogram of product: calcium, 120−155 g/kg; cobalt (min), 30.6 mg/kg; copper (min), 545 mg/kg; sulfur (min), 15 g/kg; fluoride (max), 330 mg/kg; phosphorus (min), 30 g/kg; iodine (min), 27.2 mg/kg; magnesium (min), 2500 g/kg; manganese (min), 2363 mg/kg; non‐protein nitrogen equivalent (NPN Eq., max), 430 g/kg; selenium (min), 5.90 mg/kg; sodium (min), 80 g/kg; zinc (min), 1636 mg/kg.

^b Probeef nutripec sprint: calcium, 24−40 g/kg; cobalt (min), 7.6 mg/kg; copper (min), 136 mg/kg; sulfur (min), 7500 mg/kg; fluoride (max), 90 mg/kg; phosphorus (min), 8000 mg/kg; iodine (min), 6.8 mg/kg; manganese (min), 590 mg/kg; total digestible nutrients (min), 420 g/kg; non‐protein nitrogen equivalent (NPN Eq., max), 190 g/kg; presence of slow‐release urea; crude protein (min) 350 g/kg; selenium (min), 2.5 mg/kg; presence of organic selenium; sodium (min), 45 g/kg; zinc (min), 409 mg/kg; lasalocid, 160 mg/kg.

Protected soybean meal was used as a source of RUP in the diet of pregnant cows, providing a broad spectrum of amino acids, including a relatively high proportion of arginine. Therefore, the amount of Soypass used in the RUP treatment was calculated to ensure that at least 56% of the arginine in Soypass was protected from ruminal degradation. This proportion was based on a study using beef cows as an animal model (Green et al. 2017). However, it is important to note that arginine availability was not directly quantified in the present study, and therefore, the effects observed should be interpreted in the context of increased metabolizable protein supply rather than attributed specifically to arginine.

Moreover, the diets were not formulated to be isoenergetic because, in order to equalize the energy content among treatments, it would have been necessary to reformulate the supplements, which would alter their ingredient proportions and consequently affect both the level of RUP and the supply of protected arginine. Such changes would prevent the RUP treatment from reaching the level of intestinally available arginine reported in the literature, thereby compromising both the scientific comparability and the applicability of the study under real‐world supplementation conditions. Additionally, the diet containing RUP was formulated to achieve the same proportion of RDP requirements as the RDP treatment, ensuring that any additional protein was supplied exclusively through RUP.

From 127 to 227 days of gestation (the treatment period), the cows were fed twice daily (0700 and 1300 h). The basal diet was provided ad libitum, and refusals were measured daily to adjust feed intake. The cow's BW was monitored to adjust supplement amounts. From 227 days of gestation, pregnant cows were equally fed and moved to a pasture area (Brachiaria brizantha cv. Marandu—DM = 28%; crude protein [CP] = 7%; neutral detergent fibre correct for ash and protein [NDFap] = 58%), in which they were managed as a single group.

The pregnant beef cows were managed under a continuous stocking system and had ad libitum access to a commercial mineral mixture (Probeef 800, Cargill Animal Nutrition, Itapira, SP, Brazil). Assurance level per kilogram of product: 190 g Ca (max); 140 g Ca (min); 61 mg Co (min); 1091 mg Cu (min); 16 g S (min); 880 mg F (max); 80 g P (min); 55 mg I (min); 5000 mg Mg (min); 4727 mg Mn (min); 11.8 mg Se (min); 110 g Na (min); 3273 mg Zn (min).

2.3. Cow‐Calf Management and Measurements

All cows were allowed to give birth naturally. From the 30 pregnant cows utilized, one experienced an abortion, two cows gave birth to twins (and were consequently removed from the study), and four calves died in the postnatal period due to external factors. Thus, the dataset was composed of information from 23 offspring (16 males and 7 females). The calves were categorized as follows: 9 in the CON group (5 males and 4 females), 8 in the RDP group (7 males and 1 female), and 6 in the RUP group (4 males and 2 females).

After birth, the vigour score of the calves was assessed as described by Nascimento et al. (2024). Briefly, a single evaluator assigned a vigour score ranging from 1 to 4 to each calf. A score of 1 indicated extreme weakness, and a score of 4 represented optimal vitality and health. Furthermore, their morphometric measurements were recorded at birth following the same methods also described by Nascimento et al. (2024). The measurements included thoracic and abdominal width; rib and rump depth; distances between the ischial and ileal bones, height at the withers, rump height, body length and girth circumference. Girth circumference was measured using flexible tape, while other measurements were obtained using a hypometer (Walmur, Porto Alegre, Brazil). Moreover, the calves were weighed, identified, dewormed and had their navels dipping in the neonatal period.

Twenty‐four hours from birth, blood samples were obtained by puncturing the jugular vein of the neonates with vacutainer tubes (First Lab, São José dos Pinhais, PR, Brazil). After collection, the blood samples were centrifuged at 3500 × rpm for 15 min at 4°C to separate the serum, which was then stored at −20°C until further analysis. Commercial kits were used to determine serum concentrations of glucose (Glucose Liquiform, Labtest, Lagoa Santa, Brazil), insulin (Monobind Inc., Lake Forest, CA, USA) and insulin‐like growth factor‐1 (IGF‐1) (Bioassay Technology Laboratory, Yangpu Dis., Shanghai, China). Analyses were performed in the Animal Research Laboratory of the Department of Animal Science at UFLA, following the protocols provided in the commercial kits.

During the cow‐calf phase, the cows and their calves were managed in the same pasture area used during the last third of gestation. The cows had ad libitum access to mineral supplementation (Probeef 800, Cargill Animal Nutrition, Itapira, SP, Brazil). At ~30 days of age, the calves were fed a protein‐energetic supplementation (7 g/kg of BW/day) through the creep‐feeding technique, using a commercial supplement (Probeef Bambini Creep, Cargill Nutrição Animal, Itapira, SP, Brazil). The supplement contained the following guaranteed levels per kilogram of product: calcium, 18−33 g/kg; cobalt (min), 3.1 mg/kg; copper (min), 60 mg/kg; organic chromium (min), 1 mg/kg; sulfur (min), 3 g/kg; fluoride (max), 2000 mg/kg; phosphorus (min), 6 g/kg; iodine (min), 2.8 mg/kg; magnesium (min), 1 g/kg; manganese (min), 112 mg/kg; total digestible nutrients (TDN, min), 640 g/kg; non‐protein nitrogen equivalent (NPN Eq., max), 13 g/kg; crude protein (min), 200 g/kg; selenium (min), 0.5 mg/kg; sodium (min), 10 g/kg; zinc (min), 181 mg/kg. The supplement also contained yeast culture and B‐complex vitamins. It was fortified with vitamin A (min), 12,000 IU/kg; vitamin D₃ (min), 1500 IU/kg; vitamin E (min), 50 IU/kg; and monensin sodium, 140 mg/kg.

The calves were weighed at birth and once a month during the calf phase to monitor their development. At 48 ± 10 days of age, biopsies of the Longissimus thoracis muscle were performed between the 12th and 13th ribs following disinfection of the area with 70% alcohol and shaving. A 2% lidocaine was administered for local anaesthesia. A 1 cm incision was made with a scalpel, and approximately 1 cm³ of muscle tissue was extracted for histological analysis. These tissue samples were immediately immersed in formalin, and after 48 h, they were stored in 70% alcohol until further analysis. Additionally, approximately 1 g of tissue was collected for gene expression analysis. These samples were placed in 2 mL cryovials, snap‐frozen in liquid nitrogen and stored at −80°C until analysis. Post‐procedure, the incision sites were sutured and topical antibiotics, healing ointment and anti‐inflammatory agents were applied. The animals were monitored for 3 days, and any adverse reactions were recorded.

Calves were weaned at ~210 days of age. Due to variation in birth dates, calves were weaned in small groups over a short period to ensure similar weaning ages across animals. This approach facilitated experimental management and reduced weaning‐related stress. Upon weaning, the calves were moved to a common pasture area (B. brizantha cv. Marandu), where they remained until all animals had been weaned for the onset of the backgrounding phase.

2.4. Backgrounding Management and Measurements

The average age at the onset of the backgrounding phase was 277 ± 31 days. The young bulls and heifers were housed in individual feedlot pens (4 × 10 m, with a covered trough, cement flooring along the trough line and an earthen remainder). Before conducting the planned measurements, the animals were submitted to a 9‐day adaptation period to both the facilities and the experimental diet. During this period, the animals were fed for 4 consecutive days with a diet containing 75% forage and 25% concentrate, followed by 5 consecutive days with a diet containing 60% forage and 40% concentrate. After this period (at 286 ± 31 days of age), the animals began receiving the final diet for 60 days, which was used throughout the backgrounding and contained 50% forage and 50% concentrate (Table 2). In the feedlot, young bulls and heifers were fed twice daily (0700 and 1300 h) and had free access to clean water. Throughout the entire feedlot period, daily feed refusals were measured to adjust the amount of diet provided accordingly.

Table 2.

Ingredients and chemical composition of feeds used during the backgrounding phase of the offspring.

Item	Inclusion/composition
Ingredients, % of DM
Corn silage	50.0
Ground corn	38.4
SNAP N proteina	11.6
Chemical composition, g/kg of DM
Corn silage
Dry matter	370.5
Organic matter	927.5
Crude protein	77.5
Ash and protein‐free neutral detergent fibre	437.1
Non‐fibrous carbohydrate	385.0
Ether extract	27.9
Ground corn
Dry matter	862.6
Organic matter	986.4
Crude protein	68.1
Ash and protein‐free neutral detergent fibre	113.8
Non‐fibrous carbohydrate	745.6
Ether extract	58.9
Mineral protein supplement a
Dry matter	882.6
Organic matter	829.8
Crude protein	457.7
Ash and protein‐free neutral detergent fibre	107.4
Non‐fibrous carbohydrate	227.4
Ether extract	37.3

^a Protein, mineral and vitamin concentrate: Moisture (Max.) = 120.00 g/kg, crude protein (Min.) = 463.29 g/kg, NPN Equiv. Protein (Min.) = 148.90 g/kg, ether extract (Min.) = 10.60 g/kg, mineral matter (Max.) = 200.00 g/kg, crude fibre (Max.) = 150.00, ADF (Max.) = 130.00 g/kg, calcium (Min.) 36.30 g/kg, calcium (Max.) 38.0 g/kg, phosphorus (Min.) 11.76 g/kg, sodium (Min.) 15.76 g/kg, potassium (Min.) 21.81 g/kg, magnesium (Min.) 15.80 g/kg, sulfur (Min.) 6.930 mg/kg, zinc (Min.) 279.00 mg/kg, copper (Min.) 74.16 mg/kg, fluorine (Max.) 65.40 mg/kg, manganese (Min.) 234.00 mg/kg, cobalt (Min.) 6 39 mg/kg, iodine (Min.) 4.48 mg/kg, selenium (Min.) 1.23 mg/kg, folic acid (Min.) 6.36 mg/kg, biotin (Min.) 4600 µg/kg, pantothenic acid (Min.) 61.10 mg/kg, vitamin A (Min.) 12,870 IU/kg, vitamin B12 (Min.) 523.58 µg/kg, vitamin B6 (Min.) 37.2 mg/kg, vitamin D3 (Min.) 1,788 IU/kg vitamin E (Min.) 193.00 IU/kg and monensin sodium 650.00 mg/kg.

A 5‐day trial was conducted when the animals reached 306 ± 31 days of age to evaluate nutrient intake and apparent total‐tract digestibility of dietary components during the backgrounding. During the trial, the offered diet and feed refusals were weighed daily. In addition, representative samples of forage, concentrate and refusals were collected each day for subsequent chemical analysis. Faecal samples were collected from each animal using the hand grab technique from the rectum. These spot collections were conducted at different times each day of the trial (Day 1 at 0600 h, Day 2 at 0900 h, Day 3 at 1200 h, Day 4 at 1500 h and Day 5 at 1800 h) to ensure that the composite samples were representative of each animal's faecal output throughout the daytime period. All collected samples (feed, refusals and faeces) were stored at −20°C until they were analysed.

In the backgrounding phase, when the animals had 319 ± 31 days of age, a team of evaluators conducted a continuous 48 h assessment of the offspring to observe their behavioural patterns. Feed intake, water consumption, rumination, periods of inactivity (animal standing or sleeping) and other activities (including locomotion, socialization with other animals in adjacent pens and interaction with environmental elements, among others) were systematically recorded at 5 min intervals throughout the ingestive behaviour assessment. Observers were strategically placed to ensure that their presence did not affect the animals' natural behaviour. To minimize disruption, artificial lighting in the facility was turned off during the night and feeding behaviour evaluations were performed using flashlights. Feeding behaviour data were converted to continuous time by multiplying the frequency of each activity observed over 48 h by the 5 min time intervals to provide the total time spent on each activity, expressed in minutes per day.

At the end of the backgrounding period (345 ± 31 days of age), a carcass ultrasonography evaluation was performed to assess the following parameters: Longissimus muscle area (LMA), subcutaneous fat thickness (SFT), rump muscle length (RML) and rump fat thickness. Ultrasonographic evaluations were performed using a high‐resolution ultrasound machine (Aloka 500‐V ultrasound device, Corometrics Medical Systems, Wallingford, CT) equipped with a 3.5 MHz, 17.2 cm linear transducer. For LMA and SFT measurements, the transducer was positioned between the 12th and 13th ribs along the midline of the animal's back. The LMA was delineated on the ultrasound image, and SFT was measured at this site. To assess RML and rump fat thickness, the transducer was placed at the junction of the biceps femoris and gluteus medius, between the ischium and the ilium, oriented parallel to the vertebral column. RML and fat thickness were measured based on the ultrasound images obtained from this location. All images were captured and analysed using ImageJ software to ensure accurate measurement and quantification of the muscle and fat deposition parameters.

In addition, at 346 ± 31 days of age, a biopsy was conducted to obtain samples of the L. thoracis muscle for histological and molecular analysis. The procedure adhered to the same methodology used for biopsies during the cow‐calf phase, as previously described.

2.5. Laboratory Analysis

2.5.1. Muscle Morphology

For determination of muscle fibre number and size, the samples were dehydrated in an ascending alcohol series, cleared in xylene and embedded in histological paraffin. A rotary microtome (RM 2265, Leica Biosystems, Nussloch, Germany) was used to obtain 3−6 sections, each 5.0 μm thick, which were subsequently stained with hematoxylin−eosin, following the method described by Pluske et al. (1996). Photomicrographs were then captured using an OLYMPUS CX31 microscope (Olympus Corp., Tokyo, Japan) at ×40 magnification. Muscle fibre morphometry was analysed using ImageJ software (National Institutes of Health, Baltimore, MD, USA), including measurements of muscle fibre number and muscle fibre cross‐sectional area.

2.6. Gene Expression

For gene expression analysis, muscle tissue samples were pulverized, and total RNA was extracted using the SV Total RNA Isolation System kit (Promega, Madison, WI, USA). RNA concentration was measured with a NanoVue spectrophotometer (GE Healthcare), and RNA integrity was confirmed by 1% agarose gel electrophoresis. The RNA was reverse transcribed into cDNA using the GoScript Reverse Transcription System kit (Promega). Primers for target and reference genes (β‐actin [ACTB] and glyceraldehyde‐3‐phosphate dehydrogenase [GAPDH]) were designed using PrimerQuest software (www.idtdna.com) based on GenBank sequences (Table 3). RT‐qPCR was conducted with a Mastercycler realplex system (Eppendorf) and SYBR Green detection (Applied Biosystems, CA, USA). Gene expression calculations were performed using the ΔΔC _t method.

Table 3.

Sequences (5′ to 3′) of primers used in quantitative real‐time.

Gene	Gene abbreviation	NCBI access code	Primer
Acetyl‐CoA carboxylase 1	ACACA	NM_174224.2	FW: TGAAGAAGCAATGGATGAACACA RV: TTCAGACACGGAGCCAATAA
Acyl‐coenzyme A oxidase 1	ACOX	BC102761.2	FW: GCTGTCCTAAGGCGTTTGTG RV: ATGATGCTCCCCTGAAGAAA
Actin beta	ACTB	NM_173979.3	FW: GTCCACCTTCCAGCAGATGT RV: CAGTCCGCCTAGAAGCATTT
Protein Kinase AMP‐Activated Catalytic Subunit Alpha 2	AMPKa2	NM_001205605.1	FW: TGCGGATCCCCAAATTATGC RV: AAATAACACCGCAGCTCCAG
Carnitine O‐palmitoyltransferase 1	CPT1	NM_001304989.2	FW: TCTGGATGCCGTGGAAAAAG RV: AAACCACCTGTCGAAACACC
Carnitine O‐palmitoyltransferase 2	CPT2	BC105423.1	FW: CATGACTGTCTCTGCCATCC RV: ATCACTTTTGGCAGGGTTCA
Collagen type III	COL3A1	NM_001076831.1	FW: AACCAGAACCGTGCCAAATA RV: TGGGGCAGTCTAATTCTTGG
Enhancer Binding protein alpha constitutive α	C/EBPA	XM_027515988.1	FW: CACGGTGCGTCTAAGATGAG RV: TCCAAGGCACAGGGTTATTC
Fatty acid binding protein 4	FABP4	NM_174314.2	FW: GGATGATAAGATGGTGCTGGA RV: ATCCCTTGGCTTATGCTCTCT
Fatty acid synthase	FASN	U34794.1	FW: ATCAACTCTGAGGGGCTGAA RV: CAACAAAACTGGTGCTCACG
Glyceraldehyde‐3‐Phosphate Dehydrogenase	GAPDH	NM_001034034.1	FW: CGACTTCAACAGCGACACTC RV: R TTGTCGTACACAAGGAAATGAGC
Insulin‐like growth factor 1 receptor	IGF1R	NM_001244512.1	FW: GAGTGGACAACAAGGAGAGAAC RV: CTTCTCAGCCTCATGGTTACAG
Lipoprotein lipase	LPL	NM_001075120.1	FW: CTCAGGACTCCCGAAGACAC RV: GTTTTGCTGCTGTGGTTGAA
Mechanistic target of rapamycin kinase	mTOR	XM_015475105.1	FW: GTCATGGAGGACACGGATTAG RV: GGACCAGTGAGGTAATGAGATG
Myosin heavy chain type I	MyHC I	NM_174117.1	FW: AGGAGAAACACGCCACAGAG RV: CTTTTCCTTGGTCAGCTTGG
Myosin heavy chain type IIa	MyHC IIa	NM_001166227.1	FW: GCCCAAGGAATCTTTTGTCA RV: CTGTCAGAGTCGCTCCTCCT
Myosin heavy chain type IIx	MyHC IIx	AB059399.2	FW: AAGCTGTCAAGGGTCTACGC RV: TCCTGGAGCCTGAGAATGTT
Myogenic differentiation 1	MyoD	NM_001040478.2	FW: CGACGGCATGATGGACTAC RV: CGCCTCGCTGTAGTAAGTGC
Myogenin	MyoG	NM_001111325.1	FW: CCTACAGACGCCCACAATCT RV: TATGGTTTCATCTGGGAAGG
Pyruvate Dehydrogenase Kinase 4	PDK4	NM_001101883.1	FW: AACCAAAGAACCTGGCGAAG RV: ATGGGCAAACGTTCAGGAAG
Peroxisome proliferator‐activated receptor alpha	PPARA	NM_001034036.1	FW: CAATGGAGATGGTGGACACA RV: TTGTAGGAAGTCTGCCGAGAG
Peroxisome proliferator‐activated receptor γ	PPARG	NM_001098905.1	FW: CGACCAACTGAACCCAGAGT RV: TCAGCGGGAAGGACTTTATG
Sterol regulatory element‐binding factor 1	SREBF1	NM_001113302.1	FW: GAGCCACACACTTCAACGAA RV: TGTCTTCTATGTCGGTCAGCA
Sterol regulatory element‐binding protein 2	SREBP2	NM_001040478.2	FW: CGACGGCATGATGGACTAC RV: CGCCTCGCTGTAGTAAGTGC
Stearoyl‐CoA desaturase 1	SCD1	NM_173959.4	FW: TTATTCCGTTATGCCCTTGG RV: TTGTCATAAGGGCGGTATCC
Uncoupling protein 3	UCP3	XM_024974990.2	FW: ACTTTTGACAGCAGCCTTCG RV: TGCTCCTTTGGCACAGTTTC
Zinc finger protein 423	ZFP423	NM_001101893.1	FW: AGACAGGAACAGCGTGACAA RV: CTGACAGTGATCGCAGGTGT

2.7. Chemical Analysis of Feedstuffs, Refusals and Faecal Samples

All samples were dried in a forced ventilation oven (feedstuffs and refusals: 55°C for 36 h; faecal samples: 55°C for 72 h) and ground in a Wiley mill (Willye TE‐680), with screens of 2 and 1 mm, for subsequent chemical analysis. The samples were analysed according to the Brazilian National Institute of Science and Technology (INCT) (Detmann et al. 2021). To determine the faecal production of the animals and thus the apparent total‐tract digestibility of the diet components, indigestible neutral detergent fibre (iNDF) was used as a marker (Valente et al. 2011). The iNDF content in feedstuffs and faecal samples was determined through in situ incubation for 288 h using cannulated beef heifers, with laboratory analyses also conducted according to the protocols from the INCT (Detmann et al. 2021).

2.8. Statistical Analysis

The cow and calf performance traits were analysed according to the following linear model:

$$Y_{\mathit{jklm}}=\mu +T_j+S_k+P_l+{{\beta }_1\mathit{iBW}}_m+{{\beta }_2\mathit{Gen}}_m+{{\beta }_3\mathit{Age}}_m+e_{\mathit{jklm}}$$ (1)

where $Y_{\mathit{jklm}}$ is the observed value; $\mu$ is the intercept; $T_j$ is the fixed effect of the j ^th level of maternal dietary treatment, with j = 1 to 3; $S_k$ is the fixed effect of the k ^th level of sex of the calf, with k = 1 or 2; $P_l$ is the fixed effect of the l ^th level of parity, with k = 1 to 3; ${\beta }_1$ is the partial regression coefficient for the fixed effect of initial body weight ($\mathit{iBW}$) of the cow, and ${\mathit{iBW}}_m$ is the $\mathit{iBW}$ of the m ^th cow; ${\beta }_2$ is the partial regression coefficient for the fixed effect of genetic potential of the cow ($\mathit{Gen}$) of the cow, and ${\mathit{Gen}}_m$ is the $\mathit{Gen}$ of the m ^th cow; ${\beta }_3$ is the partial regression coefficient for the fixed effect of age at measurement ($\mathit{Age}$) of the cow, and ${\mathit{Age}}_m$ is the $\mathit{Age}$ of the m ^th cow; and $e_{\mathit{jklm}}$ is the random error associated with $y_{\mathit{jklm}}$, assuming $e~N(0,I_e{\sigma }_e^2)$ where $I_e$ represents the identity matrix of dimension equal to the number of observations.

The model used for the calf traits measured at birth did not include the covariate Age. In addition, a logistic regression including the same fixed effects described in Eq. [1] was used for the calf trait vigour score, as records for this trait were binary.

The gene expression ($\mathit{CT}$) data were analysed according to the linear mixed model below, following the strategy proposed by Steibel et al. (2009) where the data on the target and endogenous control genes are analysed simultaneously:

$$C{T}_{ijklm}=\mu +G_i+T_j+(G\ast {T)}_{ij}+S_k+(G\ast {S)}_{ik}+P_l+(G\ast {P)}_{il}+{\beta }_1{iBW}_m+{(G\ast {\beta )}_{i1}iBW}_m+{{\beta }_{2}Gen}_m+{(G\ast {\beta )}_{i2}Gen}_m+A_m+e_{ijklm}$$ (2)

where $\mu$, $T_j$, $S_k$, $P_l$, ${\beta }_1$, ${\mathit{iBW}}_m$, ${\beta }_2$ and ${\mathit{Gen}}_m$ are as previously defined in Eq. [2], whereas ${\mathit{CT}}_{\mathit{ijklm}}$ is the observed $\mathit{CT}$, $G_i$ is the fixed effect of the i ^th level of gene, with i = 1 (target) or 2 (endogenous); $(G\ast {T)}_{\mathit{ij}}$ is the fixed effect for the interaction between $G_i$ and $T_j$; $(G\ast {S)}_{\mathit{ik}}$ is the fixed effect for the interaction between $G_i$ and $S_k$; $(G\ast {P)}_{\mathit{il}}$ is the fixed effect for the interaction between $G_i$ and $P_l$; $(G\ast {\beta )}_{i1}$ is the fixed effect for the interaction between $G_i$ and ${\beta }_1$; $(G\ast {\beta )}_{i2}$ is the fixed effect for the interaction between $G_i$ and ${\beta }_2$; $A_{\mathit{jklm}}$ is the random effect of the m ^th animal, assuming $A~N(0,I_A{\sigma }_{S_i}^2)$, where $I_A$ represents the identity matrix with dimensions equal to the number of animals; and $e_{\mathit{ijklm}}$ is the random error associated with ${\mathit{CT}}_{\mathit{ijklm}}$, assuming $e~N(0,I_{\mathit{Ge}}{\sigma }_e^2)$, where $I_{\mathit{Ge}}$ represents the identity matrix with dimensions equal to the number of observations across the two genes. The interactions between $G_i$ with $T_j$, $S_k$, $P_l$, ${\beta }_1$ and ${\beta }_2$, were used to assess the effects of maternal dietary treatment, sex of the calf, parity, iBW and Gen, respectively, through orthogonal contrasts, as in Steibel et al. (2009). Expected $\mathit{CT}$ were computed as $-∆\mathit{CT}$, such that the differences between the levels of the categorical fixed effects of interest treatment were estimated as $-∆∆\mathit{CT}$.

Before final analyses, residuals were evaluated for normality. For each analysis, data points were removed one at a time until absolute Studentized residuals were lower than 3 and had a nonsignificant (p > 0.01) Shapiro−Wilk's test for normality. Expected means were generated from the final models and separated using Fisher's LSD test when significant (p < 0.05). All analyses were performed in SAS Studio 3.81 (Enterprise Edition, SAS Institute Inc., Cary, NC, USA).

3. Results

3.1. Maternal Outcomes During Gestation

According to the Nutritional Requirements of Zebu and Crossbred Cattle (BR‐CORTE) feeding system (Gionbelli et al. 2023), the TDN requirements for the CON, RDP and RUP treatments met 83%, 104% and 116% of the maternal TDN demands, respectively. The CP requirements were fulfilled at levels of 59%, 104% and 163% for the CON, RDP and RUP treatments. The RDP requirements were achieved at 69% for CON and at 120% for both the RDP and RUP treatments. The RUP requirements were met at 39%, 70% and 251% by the CON, RDP and RUP treatments, respectively.

The prenatal feeding regimen affected the intake of DM and diet components (Table 4). Cows in the RUP group exhibited the highest intake of DM, organic matter (OM), CP and TDN, followed by those in the RDP and CON groups (p < 0.01). Neutral detergent fibre (NDF) intake was greater in the RUP group compared to the RDP and CON groups, which did not differ from each other (p < 0.01). Non‐fibrous carbohydrate (NFC) intake was higher in the RDP and RUP groups than in the CON group (p = 0.01). For ether extract (EE), cows in the RDP group had the highest intake, while cows in the RUP and CON groups had lower values (p < 0.01; Table 4).

Table 4.

Effect of prenatal feeding regimen on dry matter and diet's component intakes (measured in kg/day) and performance of beef cows.

Item	Prenatal feeding regimen1			SEM2	p value
	CON	RDP	RUP
Total intake of dry matter and diet compounds
Dry matter	6.67C	7.91B	9.08A	0.436	< 0.01
Organic matter	6.33C	7.37B	8.63A	0.418	< 0.01
Crude protein	1.47C	1.86B	2.42A	0.138	< 0.01
Neutral detergent fibre	3.77B	3.85B	4.79A	0.249	< 0.01
Non‐fibrous carbohydrates	0.921B	1.17A	1.20A	0.082	0.01
Ether extract	0.211C	0.398A	0.270B	0.019	< 0.01
Total digestible nutrients	4.38C	5.32B	6.40A	0.400	< 0.01
Body weight during gestation 3
127 days of gestation	507	550	526	17.4	0.16
227 days of gestation	546B	574AB	611A	16.2	0.01
282 days of gestation	578	580	606	14.3	0.28

^A‐BMeans with different superscripts differ significantly (p < 0.05).

¹ CON (Control): Cows fed a basal diet + commercial mineral supplement containing urea (Probeef Urea, Cargill Animal Nutrition, Itapira, SP, Brazil) at a level of 60 g/100 kg of BW; RDP (Ruminally degradable protein): basal diet + 250 g/100 kg of BW of a commercial protein supplement (Probeef Nutripec Sprint, Cargill Animal Nutrition, Itapira, SP, Brazil) + 44 g of corn finely ground per 100 kg BW; RUP (Ruminally undegradable protein supplement): basal diet supplemented with 3.5 g/kg of BW of a commercial undegradable ruminal protein product containing protected soybean meal (Soypass, Cargill Animal Nutrition, Itapira, SP, Brazil) added to a commercial mineral supplement containing urea (Probeef Urea, Cargill Animal Nutrition, Itapira, SP, Brazil) at a level of 60 g per 100 kg of BW.

² SEM = Standard error of the mean.

³ Days 127, 227 and 282 of gestation represent the onset of the experimental period, the end of the supplementation period and the prepartum period, respectively.

At the beginning of the treatment application, at 127 days of gestation, there was no effect of the prenatal feeding regimen on cow BW (p = 0.16; Table 4). However, at the end of the supplementation period, at 227 days of gestation, cows in the RUP regimen showed higher BW compared to those in the CON and RDP regimens (p = 0.01). Near parturition, at 282 days of gestation, no differences in BW were observed among treatments (p = 0.28).

3.2. Vigour Score at Birth, BW and Average Daily Gain of the Offspring

Maternal treatment had no effect on offspring vigour score (p = 0.20). The vigour scores for RUP, CON and RDP were 3.81, 3.24 and 3.62, respectively (SEM = 0.23; expressed in arbitrary units).

Birth weight differed among treatments (p = 0.04), with calves from cows fed RUP‐enriched diets during gestation exhibiting a greater average birth weight than those from the CON and RDP groups, which were not statistically different from each other (Table 5). There were no differences in weaning weights based on the prenatal feeding regimen (p = 0.84). The ADG during the cow‐calf phase did not differ across the prenatal feeding regimens (p = 0.93; Table 5). iBW at feedlot entry averaged 231 kg for CON, 245 kg for RDP and 255 kg for RUP, with no differences detected among treatments (p = 0.29). Similarly, final BW at the end of the feedlot period were 291 kg for CON, 304 kg for RDP and 319 kg for RUP, with no differences observed (p = 0.24). Average daily gains during the feedlot phase did not differ among treatments due to the prenatal feeding regimen (p = 0.42; Table 5).

Table 5.

Effects of maternal nutrition on offspring growth, morphometry, nutrient deposition and tissue histology.

Item	Prenatal feeding regimen1			SEM2	p value
	CON	RDP	RUP
Performance
Birth weight, kg	29.4B	29.8B	35.3A	1.99	0.043
Weaning weight, kg	206	211	214	11.8	0.837
ADG cow‐calf phase, kg/day	0.850	0.873	0.864	0.05	0.928
Initial BW in the feedlot, kg	231	245	255	11.4	0.288
Final BW in the feedlot, kg	291	304	319	14.2	0.242
ADG in the feedlot, kg/day	1.24	1.18	1.29	0.07	0.419
Morphometric measurements at birth (cm)
Thorax width	13.0B	15.5A	15.4A	0.82	0.015
Abdomen width	13.1	14.9	15.4	0.94	0.084
Rib deep	24.0	23.6	25.2	0.79	0.188
Rump deep	23.2	24.3	26.7	1.41	0.088
Ischial bones distance	8.67	8.85	9.29	0.81	0.758
Ilium bones distance	13.2B	15.5A	14.9A	0.61	0.006
Ilium‐Ischium distance	17.6B	19.5AB	20.9A	0.87	0.011
Height at withers	66.0B	67.5B	71.8A	0.85	< 0.001
Rump height	72.1B	74.6AB	77.9A	1.67	0.019
Body length	58.7	58.2	61.3	3.06	0.599
Body circumference	68.3B	71.5AB	74.4A	1.48	0.006
Nutrient accretion in the backgrounding phase
Longissimus muscle area, cm2	56.7	55.9	56.2	3.42	0.974
Longissimus muscle area, cm2/100 kg	18.6	17.8	17.3	0.48	0.070
Subcutaneous fat thickness, mm	0.286	0.318	0.264	0.03	0.331
Rump muscle area, cm2	7.55	7.84	7.98	0.43	0.637
Rump muscle area, cm2/100 kg	2.49	2.49	2.44	0.11	0.889
Rump fat thickness, mm	0.460	0.527	0.488	0.07	0.723
Histological parameters (Longissimus thoracis)
Fibres number
48 ± 10 days of age	21.1	24.1	26.6	4.39	0.433
346 ± 31 days of age	13.0	11.5	13.3	1.37	0.549
Fibre area, μm2
48 ± 10 days of age	425	394	333	94.8	0.609
346 ± 31 days of age	1560	1981	1470	192	0.105

^A‐BMeans with different superscripts differ significantly (p < 0.05).

¹CON (Control): Cows fed a basal diet + commercial mineral supplement containing urea (Probeef Urea, Cargill Animal Nutrition, Itapira, SP, Brazil) at a level of 60 g/100 kg of BW; RDP (Ruminally degradable protein): basal diet + 250 g/100 kg of BW of a commercial protein supplement (Probeef Nutripec Sprint, Cargill Animal Nutrition, Itapira, SP, Brazil) + 44 g of corn finely ground per 100 kg BW; RUP (Ruminally undegradable protein supplement): basal diet supplemented with 3.5 g/kg of BW of a commercial undegradable ruminal protein product containing protected soybean meal (Soypass, Cargill Animal Nutrition, Itapira, SP, Brazil) added to a commercial mineral supplement containing urea (Probeef Urea, Cargill Animal Nutrition, Itapira, SP, Brazil) at a level of 60 g per 100 kg of BW.

²SEM = Standard error of the mean.

3.3. Morphometric Measurements and Blood Parameters

Thorax width and ilium bone distance differed among treatments (p ≤ 0.01), with both RDP and RUP calves showing greater measurements than those in the CON group (Table 5). Ilium‐ischial distance also differed among treatments (p = 0.01), with RUP calves presenting greater values compared to CON. Calves born to RUP‐fed cows had the greatest height at the withers (p < 0.001). In addition, rump height and body circumference differed among treatments (p ≤ 0.02), with RUP calves exhibiting greater measurements than CON. Conversely, the remaining morphometric traits did not differ among treatments (p ≥ 0.08; Table 5).

No differences were observed in glucose (p = 0.63; CON = 136, RDP = 127 and RUP = 137 mg/dL), insulin (p = 0.48; CON = 9.6, RDP = 9.13 and RUP = 5.16 µIU/mL) or IGF‐1 levels (p = 0.33; CON = 110, RDP = 127 and RUP = 147 ng/mL) among offspring from dams submitted to different supplementation programmes at mid‐gestation.

3.4. Nutrient Accretion and Histological Parameters of L. thoracis Muscle

The LMA did not differ among dietary treatments (p = 0.97; Table 5). When adjusted for BW (cm²/100 kg), LMA likewise did not differ among treatments (p = 0.07). Additionally, the rump muscle area exhibited no variation, whether expressed in cm² (p = 0.64) or adjusted per 100 kg of BW (p = 0.89; Table 5). Measurements of SFT (p = 0.33) and rump fat thickness also failed to demonstrate significant differences (p = 0.72) as a function of the prenatal feeding regimen.

Figure 1 presents photomicrographs of skeletal muscle tissue from offspring at 48 ± 10 days and 346 ± 31 days of age as a function of maternal background. The use of different supplementation programmes during mid‐gestation did not influence fibre number or fibre area at either 48 ± 10 days or 346 ± 31 days of age (p ≥ 0.10; Table 5).

Figure 1.

Photomicrographs of the Longissimus muscle (LM) from bovine offspring at different growth stages, reflecting the influence of maternal dietary treatments during mid‐gestation (Control [CON], ruminally degradable protein [RDP] and ruminally undegradable protein [RUP] supplements). Panels show: (A) CON treatment at 48 ± 10 days of age; (B) CON at 346 ± 31 days of age; (C) RDP treatment at 48 ± 10 days of age; (D) RDP at 346 ± 31 days of age; (E) RUP treatment at 48 ± 10 days of age; and (F) RUP at 346 ± 31 days of age. The representative LM images were stained with hematoxylin−eosin and captured at ×40 magnification. Myocyte count and area did not significantly differ across treatments, as detailed in Table 5 (p ≥ 0.105). [Color figure can be viewed at wileyonlinelibrary.com]

3.5. Intake, Apparent Total Tract Digestibility and Ingestive Behaviour

During the backgrounding phase, total intake of DM and dietary fractions were not affected by maternal supplementation programme, including DM, OM, CP, NDF, ash‐ and protein‐free NDF, NFC, EE and TDN (p ≥ 0.14; Table 6). Apparent total‐tract digestibility values also remained unaffected by prenatal feeding regimen, encompassing digestibility of DM, OM, CP, NDF, ash‐ and protein‐free NDF, NFC and TDN (p ≥ 0.40). Additionally, ingestive behaviour variables, including time spent eating, drinking and ruminating, showed no differences across treatments (p ≥ 0.27; Table 6).

Table 6.

Effects of maternal nutrition on offspring dry matter and nutrient intake, apparent total‐tract digestibility and ingestive behaviour during the backgrounding phase.

Item	Prenatal feeding regimen1			SEM2	p value
	CON	RDP	RUP
Total intake of dry matter and its fractions (kg/day)
Dry matter	6.72	6.68	7.39	0.54	0.456
Organic matter	6.44	6.49	7.01	0.46	0.497
Crude protein	0.889	0.897	0.970	0.06	0.480
Neutral detergent fibre	2.39	2.41	2.60	0.16	0.451
Ash and protein‐free neutral detergent fibre	2.27	2.28	2.47	0.16	0.143
Non‐fibrous carbohydrates	3.15	3.17	3.42	0.23	0.527
Ether extract	0.233	0.235	0.256	0.02	0.421
Total digestible nutrients	4.12	4.12	3.91	0.50	0.892
Apparent total‐tract digestibility (g/kg)
Dry matter	559	576	527	39.2	0.497
Organic matter	540	559	504	51.5	0.653
Crude protein	385	482	453	54.9	0.399
Neutral detergent fibre	330	379	362	40.5	0.642
Ash and protein‐free neutral detergent fibre	295	347	329	42.5	0.639
Non‐fibrous carbohydrates	828	849	830	18.7	0.633
Total digestible nutrients	612	633	575	48.7	0.534
Ingestive behaviour (min/day)
Eating	196	211	206	15.4	0.694
Drinking water	11.7	9.18	8.97	1.63	0.266
Ruminating	386	417	393	22.7	0.410

¹CON (Control): Cows fed a basal diet + commercial mineral supplement containing urea (Probeef Urea, Cargill Animal Nutrition, Itapira, SP, Brazil) at a level of 60 g/100 kg of BW; RDP (Ruminally degradable protein): basal diet + 250 g/100 kg of BW of a commercial protein supplement (Probeef Nutripec Sprint, Cargill Animal Nutrition, Itapira, SP, Brazil) + 44 g of corn finely ground per 100 kg BW; RUP (Ruminally undegradable protein supplement): basal diet supplemented with 3.5 g/kg of BW of a commercial undegradable ruminal protein product containing protected soybean meal (Soypass, Cargill Animal Nutrition, Itapira, SP, Brazil) added to a commercial mineral supplement containing urea (Probeef Urea, Cargill Animal Nutrition, Itapira, SP, Brazil) at a level of 60 g per 100 kg of BW.

²SEM = Standard error of the mean.

3.6. Gene Expression in Skeletal Muscle

The supplementation programme provided during gestation did not influence the expression of genes involved in the regulation of cell differentiation (MYOD, C/EBPA, ZFP423, PPARG, TGFβ1), growth signalling and protein synthesis pathways (IGFR1, mTOR), lipid metabolism (FABP4, PPARG) or collagen synthesis (COL3A1) in the skeletal muscle of offspring at 48 ± 10 days of age (p ≥ 0.075; Table 7).

Table 7.

Effects of maternal nutritional on mRNA expression in Longissimus thoracis muscle.

Gene	CON‐RDPa		CON‐RUPb		RDP‐RUPc
	Log2FC [CI]	p value	Log2FC [CI]	p value	Log2FC [CI]	p value
Cow‐calf phase (48 ± 10 days)
C/EBPA	1.62 [−4.73, 1.49]	0.293	−0.27 [−2.95, 2.40]	0.836	1.35 [−1.53, 4.23]	0.344
COL3A1	−1.92 [−5.28, 1.44]	0.250	−1.40 [−4.29, 1.49]	0.328	0.52 [−2.59, 3.63]	0.733
FABP4	−2.27 [−5.59, 1.05]	0.173	−1.25 [−4.10, 1.61]	0.379	1.02 [−2.05, 4.09]	0.501
IGFR1	−2.03 [−5.31, 1.24]	0.214	0.71 [−2.11, 3.53]	0.611	2.74 [−0.29, 5.77]	0.075
mTOR	−0.34 [−3.49, 2.81]	0.827	−0.32 [−3.03, 2.39]	0.810	0.02 [−2.90, 2.93]	0.990
MYOD	−1.64 [−4.73, 1.44]	0.283	0.11 [−2.54, 2.76]	0.932	1.76 [−1.09, 4.61]	0.217
MYOG	1.10 [−2.21, 4.40]	0.501	0.15 [−2.70, 2.99]	0.916	−0.95 [−4.01, 2.11]	0.529
PPARG	0.03 [−3.18, 3.23]	0.987	−0.71 [−3.47, 2.04]	0.599	−0.74 [−3.70, 2.22]	0.612
TGFB1	−1.18 [−4.38, 2.01]	0.453	−0.05 [−2.80, 2.70]	0.971	1.13 [−1.82, 4.09]	0.438
ZFP423	−2.00 [−5.22, 1.21]	0.212	0.32 [−2.45, 3.09]	0.816	2.32 [−0.65, 5.30]	0.121
Backgrounding phase (346 ± 31 days)
ACACA	1.14 [−1.00, 3.28]	0.285	−3.31 [−5.15, −1.46]	0.001	−4.44 [−6.43, −2.46]	< 0.001
ACOX	−0.69 [−3.11, 1.73]	0.562	−0.38 [−2.46, 1.71]	0.713	0.31 [−1.92, 2.55]	0.775
AMPKa2	−0.31 [−2.73, 2.10]	0.791	1.28 [−0.80, 3.37]	0.217	1.60 [−0.64, 3.84]	0.154
COL3A1	0.91 [−1.47, 3.29]	0.440	−2.03 [−4.07, 0.02]	0.050	−2.94 [−5.14, −0.74]	0.011
CPT1A	−0.58 [−3.36, 2.20]	0.673	−1.67 [−4.07, 0.72]	0.163	−1.09 [−3.67, 1.48]	0.390
CPT2	−0.59 [−2.67, 1.48]	0.562	−1.82 [−3.61, −0.04]	0.046	−1.23 [−3.15, 0.69]	0.201
FABP4	1.66 [−1.19, 4.51]	0.242	−1.91 [−4.37, 0.54]	0.121	−3.57 [−6.21, −0.94]	0.010
FASN	−0.81 [−3.13, 1.52]	0.482	−2.37 [−4.37, −0.37]	0.022	−1.57 [−3.72, 0.59]	0.147
IGFR1	0.03 [−2.07, 2.13]	0.978	−2.23 [−4.04, −0.42]	0.017	−2.26 [−4.20, −0.32]	0.024
INSR	−0.30 [−2.49, 1.88]	0.778	−0.91 [−2.79, 0.97]	0.329	−0.61 [−2.63, 1.41]	0.542
LPL	0.32 [−2,58, 3.22]	0.822	−1.84 [−4.34, 0.66]	0.142	−2.16 [−4.84, 0.52]	0.110
mTOR	0.04 [−2.11, 2.19]	0.970	−1.66 [−3.51, 0.20]	0.077	−1.70 [−3.69, 0.29]	0.091
MYHC1	0.07 [−2.46, 2.60]	0.956	0.74 [−1.44, 2.91]	0.494	0.67 [−1.67, 3.01]	0.564
MYHC2a	0.04 [−2.78, 2.85]	0.980	1.13 [−1.29, 3.55]	0.348	1.09 [−1.51, 3.69]	0.397
MYHC2x	0.10 [−2.39, 2.59]	0.936	0.93 [−1.21, 3.07]	0.382	0.83 [−1.47, 3.13]	0.466
PDK4	−0.47 [−3.91, 2.97]	0.781	0.56 [−2.39, 3.52]	0.698	1.03 [−2.15, 4.21]	0.510
PPARα	1.23 [−0.97, 3.43]	0.260	−1.06 [−2.95, 0.83]	0.261	−2.29 [−4.32, −0.26]	0.029
PPARG	0.07 [−2.41, 2.56]	0.953	−2.30 [−4.44, −0.16]	0.036	−2.37 [−4.67, −0.07]	0.044
SCD1	1.53 [−2.44, 5.49]	0.436	−2.29 [−5.69, 1.12]	0.180	−3.81 [−7.47, −0.15]	0.042
SLC2a4	−0.32 [−2.60, 1.97]	0.779	0.40 [−1.56, 2.37]	0.676	0.72 [−1.39, 2.84]	0.490
SREBF1	1.15 [−1.02, 3.32]	0.286	−0.45 [−2.32, 1.41]	0.623	−1.60 [−3.61, 0.40]	0.113
UCP3	−0.47 [−3.20, 2.26]	0.729	−0.12 [−2.47, 2.23]	0.916	0.34 [−2.18, 2.87]	0.781

^a CON‐RDP = ΔCT Control – ΔCT RDP.

^b CON‐RUP = ΔCT Control – ΔCT RUP.

^c RDP‐RUP3 = ΔCT RDP – ΔCT RUP. The ΔCT values were determined as the average difference between the endogenous control (ACTB and GAPDH) and the target gene for each sample. ΔΔCT values are presented as Log₂ fold change, which quantifies the relative mRNA expression level in a given comparison, where a positive value indicates an upregulation, and a negative value indicates a downregulation, between the two treatments. Values in brackets correspond to the 95% confidence interval (CI), offering a measure of the statistical reliability of each fold change estimate.

At 346 ± 31 days of age, no differences in gene expression were observed between the CON and RDP groups (p ≥ 0.24; Table 7). However, compared to the CON group, offspring from RUP‐supplemented cows showed 3.31‐, 2.37‐ and 2.30‐fold greater expression of ACACA (p ≤ 0.01), FASN (p = 0.02) and PPARG (p ≥ 0.04), respectively. The CPT2 mRNA expression was 1.82‐fold higher in the RUP group compared to the CON group (p = 0.05). Additionally, COL3A1 and IGFR1 expressions were 2.03‐fold (p = 0.05) and 2.23‐fold (p = 0.02) higher, respectively, in the RUP group relative to the control (Table 7). Offspring from cows supplemented with RUP exhibited greater mRNA expression of genes associated with lipid metabolism, specifically ACACA, FABP4, PPARA, PPARG and SCD1, compared to those from cows provided with RDP during mid‐gestation (p ≤ 0.04). Furthermore, COL3A1 mRNA expression was 2.94‐fold higher in offspring from cows fed RUP versus RDP (p = 0.01), while IGFR1 expression was 2.26‐fold upregulated in this same group (p = 0.02). The expression of other evaluated genes was not affected by the prenatal diet (p ≥ 0.07) during the backgrounding phase (Table 7).

4. Discussion

In tropical beef production systems, pregnant cows often experience mid‐to‐late gestation during the dry season, a period characterized by low forage quality, particularly regarding to protein content (Santos et al. 2022). This protein deficiency compromises ruminal microbial activity, reducing fibre digestibility and voluntary feed intake (Carvalho et al. 2024; Lazzarini et al. ²⁰⁰⁹; Meneses et al. ²⁰²⁴). However, there is limited information regarding the comparative effects of increased nitrogen supply—either ruminal or post‐ruminal—on intake and performance parameters of pregnant beef cows.

In the present study, cows supplemented with RUP exhibited the greatest DM and nutrient intakes, followed by those receiving RDP, whereas CON cows had the lowest values. This response may be attributed to the substrate specificity of distinct ruminal microbial populations regarding nitrogen utilization. While some bacterial species rely predominantly on ammonia as their nitrogen source, others require peptides or free amino acids for optimal growth (Dijkstra et al. 2005). Consequently, the provision of true protein sources (RDP or RUP), probably improved microbial growth and fermentation efficiency (Argyle and Baldwin 1989; Cotta and Russell ¹⁹⁸²), thereby improving feed intake in RUP supplemented cows. Additionally, RUP may indirectly contribute to the ruminal nitrogen pool through enhanced urea recycling, as demonstrated in isotopic studies using [¹⁵N¹⁵N]‐urea (Batista et al. 2016). Thus, this mechanism, combined with improved nitrogen synchronization, likely favoured microbial nitrogen capture, fibre degradation and passage rate—factors associated with increased feed intake in cattle fed low‐quality forages. Although not directly measured, the greater NDF intake observed in RUP‐supplemented cows supports this hypothesis.

It is also important to note that, although the supplementation strategies were designed primarily to contrast rumen‐degradable versus rumen‐undegradable nitrogen supply, the RUP diet provided greater TDN than both the CON and RDP diets. Thus, the additional energy supplied by the RUP treatment may have also contributed to the responses observed in this group. In this sense, by 227 days of gestation (the end of the supplementation period), cows fed an RUP‐enriched diet exhibited greater BW than their counterparts. This outcome likely reflects the cumulative effects of higher nutrient intake and improved fulfilment of maternal energy and macronutrient requirements, leading to increased tissue accretion during mid‐gestation. These findings align with evidence from a meta‐analysis including data from 35 studies and 3854 animals (Barcelos et al. 2022), which reported that nutritional strategies meeting a greater proportion of maternal requirements are consistently associated with increased maternal weight gain during gestation.

Nevertheless, no differences in Day 282 of gestation BW were observed among treatments, likely due to the adoption of a common feeding regimen after the supplementation period and the increasing metabolic demands of late gestation. During this phase, physiological adaptations prioritize nutrient partitioning to the gravid uterus through homeorhetic regulation (Gionbelli et al. 2024; Moreira et al. ²⁰²⁵), often leading to maternal tissue mobilization. Thus, cows in the RUP group, which accumulated greater reserves during mid‐gestation, probably mobilized part of these stores to support foetal development, attenuating BW differences near parturition. Similarly, Meneses et al. (2024) observed that protein supplementation during mid‐gestation enhanced maternal body reserves, as evidenced by increased BW gain, body condition score and LMA. These improvements in maternal tissue reserves were mobilized during late gestation, when all cows were subjected to nutritional restriction, helping to sustain the normal growth of the gravid uterus—including foetus, placenta, fluids and udder development.

Despite the convergence in maternal BW near parturition, calves born to RUP‐supplemented cows exhibited greater birth weights, suggesting that maternal gains accumulated during mid‐gestation may have been redirected to support foetal growth during late gestation. This response is most likely associated with the overall increase in metabolizable amino acids and energy supply provided by the RUP treatment, thereby enhancing the nutrient availability required for uteroplacental and foetal development. In addition, previous studies in ruminants have demonstrated that rumen‐protected protein sources or functional analogues can enhance placental development and foetal growth (Zhang et al. ^2016a, ^2016b; Zhang et al. ²⁰²³; Gu et al. ²⁰²¹; L. M. Sousa, de Souza, et al. ²⁰²⁴).

Although calves born to RUP‐supplemented dams exhibited greater birth weights, the lack of treatment effects on the expression of myogenic (e.g., MYOD, MYOG), adipogenic (e.g., C/EBPA, ZFP423, PPARG) and fibrogenic (TGFβ1) genes in the longissimus muscle suggests that maternal protein supplementation during mid‐gestation did not affect the ontogeny of muscle fibres. This interpretation is corroborated by histomorphometric analyses, which showed no differences in muscle fibre density between treatments at 48 or 346 days of age, indicating that muscle fibre hyperplasia was not affected.

No significant differences were detected in neonatal vigour scores or in postnatal performance metrics, including ultrasound‐derived carcass traits, weaning weight, ADG during the cow–calf phase, initial and final feedlot BW or ADG during the feedlot period. These findings suggest that while RUP supplementation programme during mid‐gestation promoted foetal growth, these benefits did not extend into sustained postnatal performance advantages under standardized management conditions. Furthermore, the absence of differences in systemic concentrations of key endocrine biomarkers, such as insulin and IGF‐1, which regulate growth and metabolism, suggests that maternal nutritional interventions for 100 days during mid‐gestation did not influence hormonal pathways critical to anabolic processes during the postnatal phase, thus supporting the lack of differences in offspring performance. Additionally, no differences were observed in total nutrient intake, apparent digestibility or ingestive behaviour among offspring—factors known to influence performance, which remained unaffected in this study. Given the strong association between postnatal growth and these parameters, the absence of improvements in performance outcomes likely reflects the lack of prenatal programming effects on the core mechanisms of nutrient intake and utilization under the conditions of this study.

No differences were observed in the expression of genes related to lipid metabolism, muscle fibre composition, energy homoeostasis or growth signalling pathways between offspring from cows supplemented with RDP and those from the CON group. These results suggest that maternal RDP supplementation during mid‐gestation did not induce transcriptional programming effects in progeny skeletal muscle compared to the basal nutrition regimen.

In contrast, maternal supplementation with RUP elicited long‐lasting programming effects on lipid metabolism in the skeletal muscle of the offspring, as evidenced by distinct molecular signatures detected at 346 days of age. Specifically, offspring born to RUP‐supplemented cows exhibited a transcriptional profile characterized by an upregulation of genes involved in lipogenesis, including ACACA, FASN and PPARG. The ACACA gene encodes acetyl‐CoA carboxylase, the rate‐limiting enzyme of de novo fatty acid synthesis, which catalyzes the conversion of acetyl‐CoA to malonyl‐CoA (Wang et al. 2022). Similarly, FASN encodes fatty acid synthase, a multifunctional enzyme complex involved in the synthesis of long‐chain saturated fatty acids (Ladeira et al. 2016). PPARG acts as a central transcriptional regulator of intramuscular fat deposition. Its activation promotes the coordinated expression of multiple target genes involved in lipid uptake, synthesis and storage, such as FABP4, ACACA, FASN and SCD, collectively enhancing lipogenic capacity in muscle tissue (Graugnard et al. 2010, 2009). Despite the absence of differences in the expression of genes related to peroxisomal (ACOX1) or early mitochondrial (CPT1A, UCP3) fatty acid oxidation, offspring from RUP‐supplemented cows displayed greater expression of CPT2, which encodes carnitine palmitoyltransferase 2—an enzyme critical for the translocation of long‐chain fatty acids into the mitochondrial matrix for β‐oxidation (Kerner and Hoppel 2000). Therefore, collectively, these findings suggest enhanced lipid turnover in the skeletal muscle of RUP offspring, indicative of a more dynamic balance between lipid synthesis and oxidation when compared to CON.

Despite these transcriptional changes, no differences were observed in the expression of genes related to muscle fibre composition (MYH7, MYH2, MYH1) for RUP compared to CON, indicating that maternal RUP supplementation did not affect myofiber phenotypes. Similarly, genes involved in energy sensing (PRKAA2), glucose metabolism (PDK4, SLC2A4), insulin signalling (INSR) and protein synthesis (mTOR) remained unaffected between RUP and CON offspring. In this sense, these findings suggest that the programming effects of maternal RUP supplementation are specifically targeted to pathways governing lipid metabolism, rather than global alterations in energy metabolism or muscle contractile protein expression.

Furthermore, when comparing the RUP‐ and RDP‐supplemented groups, offspring from the RUP treatment exhibited greater expression of ACACA, FABP4, PPARA and SCD1, suggesting an increased capacity for intramuscular lipid synthesis and deposition. These transcriptional adaptations indicate that maternal RUP supplementation, relative to RDP, modulated foetal metabolic programming toward a more lipogenic skeletal muscle phenotype, a shift commonly associated with improved marbling and intramuscular fat accretion. It is also important to recognize that, in addition to the greater supply of rumen‐undegradable amino acids, the RUP diet provided a higher TDN intake than the RDP diet. Thus, the increased energy availability in the RUP regimen may have acted synergistically with the enhanced amino acid supply into the intestine to promote lipogenic programming. Therefore, this combined nutritional stimulus likely contributed to the upregulation of genes governing lipid synthesis and storage observed in RUP offspring.

In addition, increased mRNA expression of IGF1R was observed in the skeletal muscle of RUP offspring compared with CON and also compared to RDP at 346 days of age. This finding suggests enhanced sensitivity of muscle tissue to the IGF axis, which plays a central role in hypertrophy (Glass 2005). Although no phenotypic differences in muscle accretion were detected between groups, the upregulation of IGF1R may reflect a molecular environment more favourable to anabolic signalling and muscle development in RUP offspring. Moreover, greater expression of COL3A1, encoding type III collagen, was also detected in RUP offspring compared to CON and RDP. This gene is essential for maintaining the structural integrity of the extracellular matrix (ECM) (Ricard‐Blum ²⁰¹¹), and its upregulation may indicate increased ECM remodelling activity or enhanced matrix support. These adaptations could facilitate cellular expansion and angiogenesis (Cao 2010)—processes critical for muscle growth and intramuscular fat deposition. Therefore, these findings collectively suggest that maternal RUP supplementation induced transcriptional adaptations in skeletal muscle, consistent with a more metabolically supportive and developmentally responsive microenvironment compared to either the commercial mineral supplement containing urea or the RDP supplementation programme. In addition to the increased supply of rumen‐undegradable amino acids, the higher energy intake associated with the RUP diet may have also contributed to shaping this molecular profile, potentially enhancing the activation of key lipogenic and growth‐related pathways, despite the lack of observable differences in postnatal muscle hypertrophy.

5. Conclusion

In tropical beef production systems, supplementation with RUP during gestation improved maternal performance in mid‐gestation, increasing maternal tissue reserves that could be mobilized under nutrient restriction during late gestation to meet foetal growth demands. Calves from RUP‐supplemented cows had higher birth weights, suggesting improved nutrient transfer to the foetus. Although foetal growth was enhanced, no differences were observed in postnatal calf performance, including weaning weight, average daily gain or carcass traits, possibly due to the lack of direct effects on nutrient intake and utilization post‐birth. Compared to conventional supplementation, RUP supplementation induced molecular changes in the calves' skeletal muscle, with increased expression of genes involved in lipid metabolism, which may contribute to greater intramuscular fat synthesis and improved meat quality. It is also plausible that these transcriptional adaptations were influenced not only by the increased supply of rumen‐undegradable amino acids but also by the higher energy intake associated with the RUP diet, which may have acted synergistically to favour a more lipogenic metabolic profile.

Conflicts of Interest

The authors declare no conflicts of interest.

Animal Welfare Statement

All procedures involving animals were conducted in accordance with ethical standards and approved by the Institutional Animal Care and Use Committee of the Federal University of Lavras (UFLA), under protocol number 015/2019. The study complied with the Brazilian legislation on the scientific use of animals and adhered to the guidelines for the care and use of agricultural animals in research.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.