Effect of supplementation of unprotected or protected arginine to prolific ewes on maternal amino acids profile, lamb survival at birth, and pre- and post-weaning lamb growth

Abstract

This research determined the effects of dietary supplementation with rumen-protected arginine (Pro-Arg) on metabolites and amino acids in maternal plasma and lamb survival rate at birth (LSRAB) in prolific Afec–Assaf ewes. The hypothesis was that Pro-Arg, the precursor for nitric oxide and polyamines, would increase placental development and vascularity, uteroplacental blood flow, and nutrient transport and reduce oxidative stress to increase LSRAB. Ewes were fed either their basal diet, basal diet with Pro-Arg, or basal diet with unprotected arginine (Unp-Arg; 18 g/head/d). The supplemental arginine was about 1% of the dry matter intake from day 40 or 60 of gestation until parturition. Ninety-two of 98 ewes produced live lambs. Ewes fed Pro-Arg had greater (P = 0.002) concentrations of arginine and other amino acids in plasma, whereas Unp-Arg did not affect concentrations of arginine, but decreased (P < 0.05) concentrations of some amino acids. There was no effect of treatments on gestation length (144 ± 2 d), prolificacy (2.65 lambs born per ewe), LSRAB (0.80), body weight (88.8 ± 10.8 kg), and body condition score (2.8 ± 0.6) of ewes, or birth weight and crown-rump length of lambs. The GI (BW/CRL^1.5) was affected by sex of lamb (P = 0.008), parity of ewe (P = 0.002), litter size (P = 0.0001), and lamb status (P = 0.003). Of 229 lambs born, 32 were dead and 16 died before 5 mo of age, leaving 181 lambs with records on weights at birth and 5 mo of age. Interestingly, lambs born to ewes fed the Unp-Arg and Pro-Arg weighed 3.6 kg less at postnatal day 150 than lambs from control ewes.

Introduction

Maximizing lamb survival rates at birth (LSRAB) and weaning (Fogarty, 2009; Dwyer et al., 2016) is limited by under-nutrition (Belkacemi et al., 2010; Igwebuike, 2010) and pregnancies with multiple fetuses (Rhind et al., 2001; Louey et al., 2005; Wu et al., 2006; Gootwine et al., 2007). Introgression of the FecB (Booroola) mutation into the Assaf breed in Israel resulted in prolific Afec–Assaf ewes that average 2.5 lambs born, but LSRAB decreases from 0.95 for singletons to 0.5 for sextuplets (see Gootwine, 2013).

Arginine (Arg) stimulates hypertrophy, hyperplasia, and differentiation of ovine conceptus (embryo/fetus and placental membranes) trophectoderm oTr cells via cell signaling pathways involving nitric oxide (NO) and polyamines (see Bazer et al., 2015). Arg activates mechanistic target of rapamycin cell signaling to stimulate proliferation and migration of oTr cells (Bazer et al., 2015), increase protein synthesis, and reduce protein degradation (Morris et al., 2002; Kim et al., 2011, 2013  Kong et al., 2012). Arg is abundant in ovine uterine fluid (Gao et al., 2009) and allantoic fluid (Kwon et al., 2003). Exogenous Arg improves embryonic survival (Luther et al., 2008) and decreases intra-uterine growth restriction in underfed (Lassala et al., 2010, 2011) and obese ewes (Satterfield et al., 2012). Viagra also increases amino acids and polyamines in fetal fluids and serum, and fetal weights in underfed ewes (Satterfield et al., 2010). However, the degradation of Arg in the rumen may interfere with the use of Arg as a feed additive. Rumen-protected Arg (Pro-Arg) should not be degraded by rumen microbes (Keith et al., 2018) and may increase reproductive performance in ruminants (McCoard et al., 2017; Sun et al., 2018; Berlinguer et al., 2019). This study tested the hypothesis that supplemental dietary Pro-Arg to prolific Afec–Assaf ewes increases available Arg in the circulation, leading to improved LSRAB.

Materials and Methods

Ethical statement

Experimental protocols were approved by the Volcani Center Institutional Animal Care and Use Committee (Permit no. 672/16B IL).

Experimental design

The Afec–Assaf flock of ewes at the Volcani Research Center, Bet Dagan, Israel, was kept in open sheds. Reproductive management of the flock included three mating periods of 40 d each year. During those periods, ewes were hormonally synchronized to estrus and mated to a single ram. Ewes that did not conceive in the first mating were mated again during their next period of estrus. Pregnancies and numbers of fetuses were determined at about 35 d post-mating by transabdominal ultrasonography using an Aquila 5-MHz linear array transducer (Pie Medical, Maastricht, The Netherlands). Ninety-eight Afec–Assaf ewes confirmed to be pregnant were allocated into two cohorts according to their day in pregnancy at the beginning of the nutritional treatments. Treatments began on the same day for both cohorts: relatively late (about day 60) or relatively early (about day 40) in gestation. Ewes of each cohort were assigned randomly, according to their known levels of prolificacy and parity (see Table 1), to three treatments: control (C), diet supplemented daily with Pro-Arg, or diet supplemented daily with unprotected arginine (Unp-Arg). The Pro-Arg was not degraded by ruminal microbes of sheep or steers, based on our analysis of amino acids when the product was incubated with the ruminal fluid of sheep or steers for 2 h, as we described previously (Gilbreath et al., 2020a, 2020b). Ewes in each of the groups (n = 6) were kept in a separate, but adjacent pen. Body weight (BW) and body condition score (BCS) were recorded for each of the ewes at the start of the experiment and one day after lambing. The genotype at the FecB (Booroola) locus of 10, 79, and 9 of the ewes was BB, B+, and ++, respectively.

Table 1.

Averages for prolificacy, parity number, BW, and BCS of ewes at the beginning of treatments, according to Cohort and treatment

Cohort	Late			Early
Treatment1	Control	Pro-Arg	Unp-Arg	Control	Pro-Arg	Unp-Arg
n	15	13	18	18	20	14
Day in pregnancy, d	67 ± 1	60 ± 1	55 ± 0	47 ± 2	36 ± 2	41 ± 2
Parity, n	3.7 ± 2.1	2.3 ± 1.1	3.2 ± 2.5	3.6 ± 2.1	3.1 ± 1.4	3.6 ± 2.3
Mean prolificacy, n	2.3 ± 0.7	2.2 ± 0.7	2.2±0.5	2.3 ± 1.1	2.3 ± 0.7	2.5 ± 0.8
BW, kg2	94 ± 14	92 ± 5	87 ± 13	91 ± 8	85 ± 8	90 ± 7
BCS, unit3	3.5 ± 0.4	3.2 ± 0.4	3.1 ± 0.4	3.2 ± 0.3	3.2 ± 0.3	3.1 ± 0.4

¹Pro-Arg, rumen-protected arginine; Unp-Arg, unprotected arginine.

²BWs of ewes at the time treatments were initiated (means and standard errors).

³Means and standard error for BCS from 1 (very thin) to 5 (very fat).

Diet

Ewes were fed, on a group (pen) basis, a daily diet that included 0.65 kg concentrate (16% crude protein), 1.30 kg corn silage, and free access to oat hay. The ewes consumed an average of 0.57 kg hay/head/d. The Pro-Arg and Unp-Arg (18 g/head/d) dietary supplement was about 1% of the dry matter intake (equivalent to 0.25 g arginine/100 g dietary dry matter) for each of the six subgroups as noted in Table 1. The respective diets were fed until parturition. The use of Pro-Arg in dairy cows was reported previously (Keith et al., 2018). The Pro-Arg was prepared by mixing the amino acid with soybean hydrogenated oil (as a binding material) using the top spray fluidized bed system in the continuous mode. The product was manufactured by Biotechnology Services and Consulting Inc. (Coppell, TX) using methods that are proprietary. The Unp-Arg contained the same components as the Pro-Arg except that the former did not undergo an encapsulation process. During the period of arginine supplementation, the amount of supplemental concentrate provided to the treatment groups was reduced to 0.62 kg/head/d to achieve a diet isonitrogenous to the C diet. Concentrates or daily mixture of concentrates and dietary arginine (protected and unprotected) were divided into three portions, which were provided to ewes at 0700, 1200, and 1600 hours each day. After taking the second blood sample (see below) on about day 130 of gestation, ewes were provided, on a group basis, the molasses-based product ENERGILASS Sheep 15 (Muscatine, IA) free choice until they lambed. The ewes consumed an average of 81 g molasses per day. In addition, the ewes were supplemented daily with 30 mL/head of propylene glycol poured on top of the molasses.

Ewe lambed between October 3, 2016 and November 9, 2016. Average gestation length, litter size, lambs born alive per litter, and weight and body score of ewes post-lambing are summarized in Table 2. Birth weights and crown rump lengths (CRLs) for each lamb were determined within a few hours of lambing. The BW and BCS for each ewe were determined 3 d after lambing. Each ewe reared no more than two lambs. Surplus lambs were removed from their dams and reared in an artificial rearing unit with ad libitum access to milk replacer (Halavit 335, Maabarot Products Ltd., Israel). Following weaning, at about 35 d of age, all lambs were fed a concentrate ration ad libitum (16% crude protein and hay at 0.4 kg/d) and they had free access to water. Lambs were weighed at about 5 mo of age.

Table 2.

Lambing data (mean ± SD)² according to cohort and treatment group

Cohort	Late			Early
Treatment1	Control	Pro-Arg	Unp-Arg	Control	Pro-Arg	Unp-Arg
No. of ewes	15	13	18	18	20	14
Ewe lambing	13	13	17	16	20	13
Ewes aborting	2	—	1	1	—	1
Gestation length, d	144 ± 2.0	144 ± 2.0	144 ± 2	144 ± 3.0	145 ± 2.0	144 ± 2.0
Liter size	2.8 ± 1.1	2.8 ± 1.1	2.5 ± 0.9	2.6 ± 1.0	2.7 ± 0.7	2.5 ± 1.2
Lambs born alive	2.8 ± 1.1	2.3 ± 1.1	2.0 ± 1.1	1.7 ± 1.1	2.1 ± 0.8	2.0 ± 1.0
Post lambing weight, kg	90 ± 10	86 ± 11	86 ± 13	92 ± 11	87 ± 10	93 ± 8
Post lambing BCS	2.7 ± 0.4	2.7 ± 0.6	2.7 ± 0.7	2.9 ± 0.6	2.8 ± 0.5	2.9 ± 0.8

¹Pro-Arg, rumen-protected arginine; Unp-Arg, unprotected arginine.

²Means ± SD are calculated for ewes lambing.

Blood sampling

Blood samples were taken from all ewes on days 100 ± 3 and 130 ± 3 of gestation. The samples were taken at 0700 hours, before feeding, from the jugular vein into vacuum tubes (Becton Dickinson Systems, Cowley, England). The blood samples were placed on ice immediately, and plasma was separated later and stored at −20 °C until analyzed.

Analyses of plasma

Plasma samples were analyzed to determine the concentrations of amino acids (Wu and Meininger, 2008), glucose (Wu, 1995), ammonia (Brock et al., 1994), urea using urease and glutamine (Wu and Knabe, 1994), and d-3-hydroxybutyrate (DHB) (Wu et al., 1991).

Statistical analyses

ANOVA was carried out using the JMP IN computer package (SAS Institute, Inc., Cary, NC, USA).

Statistical analysis I determined the effects of variables on total litter weight. Ewe was considered an experimental unit. The statistical model included the effects of: cohort (n = 2), treatments (n = 3), cohort × treatment interaction (C * T), parity of ewe (n = 5) and litter size (n = 4). Lamb survival rate was used as a covariate.

Statistical analysis II considered the lamb as an experimental unit used chi-square analysis to compare, within litter size (1, 2, 3, 4, or more lambs per litter) effects of treatment on LSRAB.

Statistical analysis III considered the lamb as the experimental unit and determined the effects of variables on birth weight, CRL, GI (BW/CRL^1.5) index (Gootwine, 2013) of lambs, and growth rates (GRs) of lambs from birth to 5 mo of age. The variables in the model included effects of cohort, treatment, sex of lamb, parity of ewe, litter size, and lamb status at birth, alive, or dead (data on dead lambs were excluded from the analysis of postnatal GR of lambs). C * T was not significant and was excluded from the final analysis.

With ewes as the experimental unit, statistical analysis IV used the Repeated Measures procedure to determine the effects of variables on concentrations of metabolites and AA in the serum of ewes at days 100 and 130 of gestation. The fixed effects included: cohort, treatment, C * T, parity of ewes, litter size, day of blood sampling during pregnancy (day 100 ± 3 or 130 ± 3), and sampling day × treatment interaction. Ewe nested within cohort and treatment was considered a random effect. Because they were not normally distributed, data on the concentrations of β-Ala, Arg, Asn, Asp, Cit, Glu, His, Ile, Leu, Met, Orn, Ser, Tau, Thr, Val, urea, ammonia, and beta-hydroxybutyrate were subjected to Johnson transformation before statistical analysis. The statistical analysis of those metabolites was performed on the transformed data. Least-squares means before transformation are presented in Supplementary Table S1.

Data were tested for normality of distribution using the Shapiro–Wilk W Test. P < 0.05 was accepted as significant, and tendencies for significance were at 0.05 ≤ P ≤ 0.10.

Results

Lambing

Out of the 98 ewes included in the experiment, one did not lamb and five aborted (three from control ewes and two from ewes fed Unp-Arg). The remaining 92 ewes (29 control ewes, 33 ewes fed Pro-Arg, and 30 ewes fed Unp-Arg) produced 244 lambs of which 195 were born alive. The average gestation length, litter size, lambs born alive per litter, and BWs and BCS of ewes post-lambing are summarized in Table 2. The overall means and standard deviations were as follows: gestation length was 144 ± 2 d, average prolificacy was 2.65 lambs born/lambing, and average BWs and BCS for ewes post-lambing were 88.8 ± 10.8 kg and 2.8 ± 0.6, respectively. Nine ewes showed clinical symptoms of pregnancy toxemia (five control ewes, two ewes fed Pro-Arg, and two ewes fed Unp-Arg) between days 135 and 138 of pregnancy. All ewes were treated and only one ewe aborted.

The average litter weight at birth (mean ± s.d.) was 9.8 kg ± 2.8. Cohort, treatment, C * T, and parity did not affect litter weight (P > 0.05). However, there was an effect of litter size (P < 0.0001) on litter weight. Litter weights for ewes producing 1, 2, 3, or 4+ lambs (least square mean ± SE) were 5.3 ± 0.7, 9.3 ± 0.4, 10.6 ± 0.4, and 12.3 ± 0.5 kg, respectively. Lamb survival negatively affected litter weight (P = 0.02). LSRAB varied according to treatment and litter size as summarized in Table 3. Chi-Square analysis within each litter size detected no effect of treatment on LSRAB (P > 0.05).

Table 3.

LSRAB according to treatment¹ and litter size

	Control		Pro-Arg		Unp-Arg
Litter size	n	Mean ± SD	n	Mean ± SD	n	Mean ± SD
1	4	1.0 ± 0.0	1	0.0	6	1.0
2	9	0.94 ± 0.16	14	0.96 ± 0.13	9	0.83 ± 0.25
3	8	0.63 ± 0.45	11	0.82 ± 0.23	9	0.66 ± 0.33
4	8	0.84 ± 0.26	7	0.58 ± 0.41	6	0.88 ± 0.14
5	—	—	1	0.8	—	—

¹Pro-Arg, rumen-protected arginine; Unp-Arg, unprotected arginine.

Records for 229 lambs were used to determine the effects of treatment and other variables on birth weight, CRL, and GI of lambs, and 181 records were used to assess the effects of variables on GRs of lambs from birth to 5 mo of age (152 ± 6 d) and an average BW of 54.6 kg (Table 4). Treatment did not affect birth weight, CRL, or GI of lambs, but did affect (P = 0.02) daily GR of lambs during the postnatal period. Lambs born to ewes fed Pro-Arg or Unp-Arg weighed about 3.7 kg less at 5 mo of age than contemporary lambs born to control ewes. The significance of the effect increased to (P = 0.004) when the effect of litter size was replaced in the model with effect of birth weight as a covariate. Lamb mortality at birth (born alive or dead) did not affect CRL but did affect GI (P = 0.004).

Table 4.

Factors affecting BW, CRL, GI index², and daily GR up to 5 mo of age, (mean ± SD) according to cohort and treatment group

	Treatment1				P-value
	Control	Pro-Arg	Unp-Arg	SEM	Cohort	Treatment	C * T	Sex	Parity	Litter size	Status
BW, kg	4.2	4.3	4.1	0.12	0.39	0.19	0.16	0.01	0.005	<0.0001	0.04
CRL, cm	51.2	52.2	51.7	0.57	0.0001	0.3	0.18	0.09	0.004	<0.0001	0.52
G index3	11.1	11.2	10.6	0.22	0.06	0.06	0.03	0.008	0.002	<0.0001	0.003
GR, kg/d	0.36	0.33	0.34	0.007	0.28	0.02	0.25	<0.0001	0.04	<0.0001

¹Pro-Arg, rumen-protected arginine; Unp-Arg, unprotected arginine.

²Born alive or dead.

³G index, GI = BW/CRL^1.5.

Amino acids in plasma of ewes on days 100 and 130 of gestation

The concentrations of amino acids in plasma of ewes on both days 100 and 130 are very similar to those that we reported previously for non-prolific Colombia crossbred ewes on day 135 of gestation Kwon et al. (2003) (see Supplementary Table S1 and Table 5). We correlated the concentrations of metabolites and amino acids in the plasma of ewes between days 100 and 130 of gestation with BW gain (or loss) of the ewes throughout the experiment as an indicator of feed intake. The results indicated that changes in BWs of ewes were poorly correlated with concentrations of metabolites and amino acids in plasma (average r² was 0.03 with a standard deviation of 0.06) indicating that the amount of feed consumed by each ewe, as reflected by changes in BW, had very little effect on the parameters tested, which justifies considering the ewe as the experimental unit in this study. Ewes receiving Pro-Arg had greater concentrations of arginine (P = 0.002) as well as greater concentrations of glutamine, glutamate, glycine, histidine, lysine, methionine, and valine in plasma (see significance in Supplementary Table S1) compared with values for control ewes. Concentrations of arginine in plasma were not different (P > 0.05) between control ewes and ewes receiving Unp-Arg. However, ewes fed Unp-Arg had lower concentrations of alanine, β-alanine, asparagine, glutamine, isoleucine, leucine, phenylalanine, threonine, tryptophan, tyrosine, and valine than those for control ewes.

Table 5.

Effects of Pro-Arg and Unp-Arg on concentrations of amino acids (nmol/mL) in the plasma of pregnant ewes on days 100 and 130 of gestation

		Effect of Treatment vs. Control1
Amino acid	Overall mean concentration	Pro-Arg	Unp-Arg	Effect of parity	Effect of litter size2	Ratio of concentrations: day 130 vs. 1003
Ala	151	=	L	N	0.80	0.87
β-Ala	21	=	L	N	=	=
Arg	177	H	=	Y	0.82	0.70
Asn	45	=	L	N	0.76	0.66
Asp	8.6	=	=	Y	=	1.32
Cit	171	=	=	N	0.65	0.68
Gln	221	=	L	N	=	0.88
Glu	94	H	=	Y	=	=
Gly	276	h	=	N	0.95	0.60
His	40	h	=	N	0.68	0.78
Ile	84	=	=	N	0.85	0.91
Leu	137	=	L	N	0.82	0.76
Lys	129	h	=	N	0.69	0.65
Met	41	H	=	N	0.82	0.72
Orn	55	=	=	N	0.58	0.37
Phe	46	=	L	N	0.83	0.84
Ser	68	=	=	N	=	0.84
Tau	127	=	=	N	=	0.31
Thr	61		L	N	0.67	0.81
Trp	45	=	L	N	0.61	0.75
Tyr	67	=	L	N	0.80	0.77
Val	138	H	L	N	0.65	0.72

¹The amino acid concentrations are noted as being either higher (H) (P < 0.003) or (h) (P < 0.05), or lower (L) (P < 0.003) or (l) (P < 0.05) compared with values for control ewes.

²Ratio of values for ewes with 4+ lambs vs. 1 lamb when the litter size effect was significant (P < 0.05).

³The effect of day of pregnancy was significant (P < 0.003) for concentrations of amino acids in the plasma of ewes. Ratios of amino acids on day 130 vs. 100 of less than 1.00 indicate a decrease in the concentrations of the respective amino acid between day 100 and 130 of gestation.

Parity affected (P < 0.05) the concentrations of aspartic acid (P < 0.007) and glutamic acid (P < 0.002) and also affected concentrations of arginine (P < 0.02) in plasma and increasing parity associated with lower concentrations of amino acids in plasma. Litter size affected (P < 0.05 to P < 0.0002) concentrations of all amino acids in maternal plasma except for β-alanine, aspartic acid, glutamic acid, glutamate, serine, and taurine, and concentrations of glycine tended to be affected (P < 0.05; see Supplementary Table S1). Further, concentrations of amino acids in plasma were less (P < 0.05) at 130 d than at 100 d of gestation, except for asparagine for which concentrations were greater at 130 d of pregnancy. Cohort affected (P > 0.05) concentrations of most amino acids as concentrations were greater for ewes in the “Late” groups of ewes.

Glucose, urea, ammonia, and DHB in plasma of ewes

Treatment affected the concentrations of urea and ammonia in plasma. Ewes fed Unp-Arg had greater concentrations of urea in plasma (P < 0.0001) than ewes fed the C diet and the Pro-Arg (see Supplementary Table S2 and Table 6). Also, concentrations of ammonia were less for ewes fed Pro-Arg and Unp-Arg than those for control ewes (P = 0.008). Stage of pregnancy differentially affected (P < 0.0001) concentrations of glucose, urea, ammonia, and DHB in plasma of ewes. Concentrations of glucose, ammonia, and DHB were greater at 130 d than at 100 d of gestation with a 2.7-fold increase in DHB, whereas concentrations of urea in plasma at day 130 were 86% of values at day 100 of gestation.

Table 6.

Effects of Pro-Arg and Unp-Arg on concentrations glucose (mM), urea (µmol/mL), ammonia (nmol/mL), and DHB (mM) in the plasma of pregnant ewes on days 100 and 130 of gestation

		Effect of Treatment vs. Control1
Amino acid	Overall mean concentration	Pro-Arg	Unp-Arg	Effect of parity2	Effect of litter size3	Ratio of concentrations: day 130 vs. 1004
Glucose	2.7	=	=	N	=	1.24
Urea	3.6	=	H	N	=	0.86
Ammonia	84.5	L	L	Y	=	1.35
DHB	672	=	=	N	1.38	2.69

¹The concentrations are noted as being either higher (H) (P < 0.001) or lower (L) (P < 0.008) compared with values for control ewes.

²N indicates no effect was observed. Y indicates an effect was observed (P = 0.04).

³Ratio of values for ewes with 4+ lambs vs. 1 lamb when the litter size effect was significant (P < 0.005).

⁴The effect of day of pregnancy was significant (P < 0.001) for concentrations of metabolites in the plasma of ewes. Ratios of metabolites on day 130 vs. 100 of less than 1 indicate a decrease in the concentrations of the respective metabolite between day 100 and 130 of gestation.

Litter size tended (P = 0.05) to affect the concentrations of DHB as increases in litter size were associated with increases in the concentrations of DHB in plasma. Cohort affected concentrations of urea and ammonia in plasma as they were greater (P < 0.0001) in the “Late” groups. Parity did not affect (P > 0.05) concentrations of metabolites in plasma, except for ammonia (P = 0.04).

Discussion

There is ample evidence indicating that feeding supplemental Pro-Arg to pregnant ewes may have beneficial effects on the outcome of pregnancy (de Chávez et al., 2015; Peine et al., 2018; Zhang et al., 2016, 2018). The present study is the first, to our knowledge, to examine the effects of dietary supplementation of Pro-Arg and Unp-Arg to small ruminants on pregnancy outcomes. To our knowledge, this study also comprehensively evaluated, for the first time, the effects of dietary supplementation with Pro-Arg on the reproductive performance of the genetically unique prolific Afec–Assaf ewes. In addition, we measured the concentrations of not only amino acids, but also ammonia, urea, ketone bodies, and glucose in the plasma of the ewes. No previous studies using a similar experimental design have presented data on these physiologically significant metabolites. Thus, we believe that our study provides novel and important findings to advance the field of sheep nutrition and reproduction.

Precedents for this study were findings that arginine and its precursor (citrulline) are abundant in ovine uterine fluid (Gao et al., 2009) and allantoic fluid (Kwon et al., 2003) during early-to-mid gestation, and that intravenous administration of arginine improved embryonic survival in pregnant ewes (Luther et al., 2008). Further, as noted previously, administering ewes sildenafil citrate (Viagra) increased fetal weights (Satterfield et al., 2010), while intravenous infusions of arginine increased fetal weight and lamb survival (Lassala et al., 2010, 2011; Satterfield et al., 2010a). There is also extensive experimental evidence that dietary supplementation of arginine enhances embryonic/fetal and neonatal survival in pigs (Ramaekers et al., 2006; Mateo et al., 2007; Campbell, 2009; De Blasio et al., 2009; Bérard and Bee, 2010; Wu, 2010; Li et al., 2011; Wu et al. 2011a, 2011b, 2013; Gao et al., 2012), rodents (Zeng et al., 2008, 2012, 2013), and humans (Xiao and Li, 2005; Shen and Hua, 2011; Gui et al., 2014).

Prolificacy of Afec–Assaf ewes is about 2.5 lambs born per ewe lambing, but an average of only 2.1 lambs are born alive (Gootwine et al., 2008). Afec–Assaf were selected for this study to test our hypothesis that supplemental dietary arginine improves LSRAB, because stillborn lambs are exposed to asymmetrical intrauterine growth restriction resulting in fetal deaths 7 to 10 d before lambing (see Gootwine et al., 2013). As fetal programming resulting in intrauterine growth restriction may have adverse effects on animal health and productivity during the post-lambing period and into adulthood for sheep (Kenyon and Blair, 2014; Brien et al., 1994; Sinclair et al., 2016), we also determined effects of diet on growth of lambs to 5 mo of age.

The working hypothesis for this study was that LSRAB would be improved in highly prolific Afec–Assaf ewes in response to dietary supplementation with Pro-Arg due to increases in the production of nitric oxide, polyamines, and agmatine. Those molecules are associated with placental development and vascularity, uteroplacental blood flow, and placental functions for nutrient transport, as well as reducing oxidative stress to increase LSRAB.

Prolificacy and LSRAB for Afec–Assaf ewes in the present study were 2.6 and 0.82, respectively. The administration of Pro-Arg significantly increased concentrations of arginine in plasma of the pregnant ewes, as well as for some essential amino acids (histidine, lysine, methionine, and valine) and some nonessential amino acids (glutamine, glutamate, glycine, and ornithine) available for transport across the placenta to support the growth and development of the fetus. Arginine, citrulline, ornithine, and methionine are precursors for the synthesis of polyamines, whereas arginine is also used to generate nitric oxide that increases angiogenesis and vasodilation to increase uterine blood flow. The amino acids such as histidine, methionine, threonine, and valine that cannot be synthesized in the body were, except for tryptophan, present in greater concentrations in plasma of ewes fed Pro-Arg. Amino acids that are synthesized in sheep include alanine, arginine, asparagine, aspartic acid, cysteine, glutamic acid, glutamine, glycine, proline, serine, and tyrosine. Of those, arginine, glutamine, and glycine were present in greater concentrations in the plasma of ewes fed Pro-Arg. Collectively, many amino acids were present in greater concentrations in plasma of ewes fed Pro-Arg, and each of them is important for the growth and development of the conceptus. Also, note that concentrations of all of the amino acids analyzed, except for aspartic acid, were less in plasma of ewes on day 130 of gestation compared with day 100 of gestation, suggesting greater transfer of most amino acids across the placenta to meet the demands of the fetus(es) for development, growth, and survival in the latter stages of pregnancy.

Ewes fed Unp-Arg compared with Pro-Arg had lower concentrations of several essential amino acids (histidine, isoleucine, leucine, lysine, methionine, phenylalanine, threonine, tryptophan, taurine, and valine) and six nonessential amino acids (alanine, arginine, asparagine, glutamine, glycine, and tyrosine).

The results of this study revealed that neither Pro-Arg nor Unp-Arg improved LSRAB, birth weight of lambs, or morphometric parameters of lambs. It is not clear why feeding Pro-Arg did not improve LSRAB, but the excellent nutritional status of the ewes was such that the effects of further dietary intervention were not detectable. Alternatively, the supplemental dose of arginine may have been insufficient to influence the fetal growth and survival in Afec–Assaf ewes that had higher basal concentrations of arginine in their plasma (Tables 5 and 6) compared with those reported previously for Suffolk ewes (Satterfield et al., 2010).

Although there was no effect of treatment on LSRAB, concentrations of urea in plasma of ewes were lower in ewes fed Pro-Arg and the C diet compared with values for ewes fed Unp-Arg, and the concentrations of ammonia were lower in the plasma of ewes fed both Pro-Arg and Unp-Arg. Those results are consistent with the fact that arginine is key to the urea cycle that is necessary to reduce the production of ammonia and prevent hyperammonemia that is lethal.

Concentrations of DHB in plasma of ewes were greater (P < 0.0001) on day 130 than day 100 of gestation, indicating an increase in ketosis with advancing stages of pregnancy. Although control ewes tended (P = 0.15) to have lower concentrations of DHB in their plasma than ewes treated with either Pro-Arg or Unp-Arg, supplemental dietary arginine did not ameliorate ketosis in ewes having multiple fetuses. Again, this may be related to the long-chain saturated fatty acid binder used to manufacture the arginine product. Oxidation of long-chain fatty acids can produce ketone bodies in sheep (Wu, 2018) and, in this study, with increases in litter size, the concentrations of DHB increased (P < 0.001).

Of particular interest regarding the postnatal development of lambs, there was evidence that maternal treatment affected weights of lambs at 5 mo of age. Lambs born to ewes fed Unp-Arg or Pro-Arg weighed about 3.7 kg less at 5 mo of age than contemporary lambs born to control ewes. The basis for this epigenetic effect on the growth of lambs during the postnatal period is unknown. However, it may be related to a long-term response to the hydrogenated soybean oil binder used to manufacture the arginine product. There is a report that maternal dietary supplementation with 7% or 21% hydrogenated soybean oil during the gestation negatively affected epigenomics and impaired vascular function in rat offspring, in comparison with the control (no supplementation) group (Kelsall et al., 2012). In our study, the amount of supplemental hydrogenated soybean oil as the binder was 1% of the maternal diet. Ruminants tolerate much less dietary fat (maximum 7% fat in diets on a dry matter basis) than nonruminants (up to 30% fat in diets on a dry matter basis; Wu, 2018). It is unknown whether 1% hydrogenated soybean oil in the diet may affect metabolism (DNA modifications and gene expression) in ruminal microbes and tissues of sheep. Another possible explanation is that a long-term alteration in the amino acid profile during gestation may have epigenetic effects (Dwyer et al., 2016) that require further investigation.

Conclusions

The results of this study indicated that dietary supplementation with Pro-Arg increased the availability of arginine in the plasma of ewes but did not improve the reproductive performance of Afec–Assaf ewes. There were effects of Pro-Arg to increase concentrations of some essential and nonessential amino acids, as well as reduce concentrations of urea and ammonia in the plasma of ewes. Future studies of dietary supplementation to increase nitric oxide and polyamines will take advantage of the recent discovery that citrulline, an immediate precursor for synthesis of arginine, is not metabolized by rumen microbes (Gilbreath et al., 2019, 2020a). Thus, citrulline can be fed, without binders, for protection in the rumen, to increase the availability of arginine for tissues to synthesize nitric oxide, polyamines, and agmatine to enhance the growth and development of the conceptus. Rumen microbes also have a limited ability to metabolize extracellular glutamate. Those findings refute the traditional view that all dietary amino acids are extensively catabolized by rumen microorganisms (Gilbreath et al., 2019, 2020b). Future studies of dietary supplementation of ruminants with citrulline are expected to yield novel effects on reproductive performance and lactational performance as reported for beneficial effects of dietary arginine supplementation in nonruminant species, such as pigs, rodents, and humans. The effects of dietary supplementation with citrulline on reproductive performance and lactation will, as for dietary supplementation of arginine in nonruminants, likely be most advantageous under conditions in which protein supplementation is limited. The goal of these nutritional strategies with dietary citrulline may be particularly important in litter-bearing small ruminants. For example, Moallem et al. (2012) reported increased susceptibility of ewes with multiple fetuses to pregnancy toxemia and suggested that appropriate nutritional strategies should be developed to mitigated pregnancy toxemia in ewes that conceive more than three fetuses.

Glossary

Abbreviations

BCS

body condition score

BW

body weight

CRL

crown rump length

DHB

d-3-hydroxybutyrate

GR

growth rate

LSRAB

lamb survival rate at birth

Conflict of interest statement

The authors declare no conflicts of interest.