Does early pregnancy nutrition affect the performance of triplet-bearing ewes and their progeny to weaning?

Abstract

Triplet-bearing ewes were offered one of three pastoral-based nutritional regimens from day 21 to day 50 of pregnancy (P21 to P50): su3b-maintenance (SM), maintenance (M) or ad lib (AD). From P50 to term, ewes were managed with the aim of ensuring total liveweight increased with expected gravid uterus mass. AD ewes were heavier (P < 0.05) and had greater (P < 0.05) body condition scores than both SM and M ewes at P50, with the latter two treatments not differing (P > 0.05). Ewe nutritional treatment had no effect (P > 0.05) on lamb birthweight, indices of colostrum uptake or lamb weaning weight. In conclusion, the data suggest that the nutrition of triplet-bearing ewes in the early pregnancy period can be controlled without any deleterious effects to offspring. However, it is important to note that in mid to late pregnancy the ewes were managed to ensure gravid uterus development was not impeded.

Keywords:

• lambs
• early pregnancy
• lamb survival
• birthweight
• lamb growth

Introduction

The effects of differing herbage or concentrate feeding levels in mid and late pregnancy on triplet-bearing ewes and their offspring's performance has been examined in several studies (Everts 1990; Kleemann et al. 1993; Hinch et al. 1996; Morris & Kenyon 2004; Kenyon et al. 2005 Kenyon et al. 2010a Kenyon et al. 2010b; Everett-Hincks et al. 2005; Corner et al. 2008 Corner et al. 2010; Kerslake 2009 Kerslake 2010). Under pasture-based conditions, under-nutrition in mid to late pregnancy can result in reduced ewe liveweight, ewe body condition and lamb liveweight (Morris & Kenyon 2004; Corner et al. 2008), poorer ewe and lamb behaviour post-birth (Everett-Hincks et al. 2005; Corner et al. 2010) and reduced indices of colostrum uptake (Kenyon et al. 2005).

Everts (1990) reported that in late pregnancy, dry matter (DM) intakes dramatically decreased in multiple- compared with single-bearing ewes. They attributed this decrease in intake to a reduction in abdominal space as the uterus increased in size. In support of this, Forbes (1968) reported that rumen volume was reduced in the last 5 weeks of pregnancy. At term, the gravid uterus weight of a triplet-bearing ewe can weigh up to 22 kg (Grazul-Bilska et al. 2006; Kenyon et al. 2007). The results of Morris & Kenyon (2004) suggest multiple-bearing ewes may fail to consume their theoretical nutritional requirements in late pregnancy under unrestricted perennial ryegrass/white clover grazing conditions (Morris & Kenyon 2004; Nicol & Brookes 2007). When ewes fail to meet their nutritional requirements in late pregnancy they must rely on their body reserves to buffer against this shortfall. In single- and twin-bearing ewes, only limited beneficial effects have been reported from nutritional regimens in the early pregnancy period (Everitt 1967; Parr et al. 1986; Krausgrill et al. 1999; Gopalakrishnan et al. 2004, 2005; Gardner et al. 2007).

To date, the potential effects of early pregnancy nutrition of triplet-bearing ewes under pastoral-based conditions have not been examined. However, given the greater nutritional demand of triplet-bearing ewes in late pregnancy (Nicol & Brookes 2007), any body reserves they can gain in early pregnancy, when their nutritional requirements are relatively low (Rattray et al. 1974; Robinson 1977; Nicol & Brookes 2007), may have a greater impact on their later performance and that of their offspring than previously reported in both single- and twin-bearing ewes. The objective of this study was therefore to examine the effects of ewe nutrition treatments in early pregnancy on the performance of triplet-bearing ewes and their offspring to weaning.

Materials and methods

Experimental design

The present study utilised 86 triplet-bearing 3- to 5-year-old multiparous Romney ewes (67.53±1.07 (SE) kg; condition score 3.04±0.06) that conceived to an artificial breeding programme, over a 2-day period, using five sires and were part of a larger study involving 1169 ewes (average liveweight 66.3±0.2 (SE) kg; condition score 2.96±0.02; Kenyon et al. 2011). Five days after insemination (P5), 12 crayon-harnessed entire Romney rams were joined with the 1169 ewes and they remained together until P21. Ewes and rams were offered herbage with a minimum post-herbage mass of 1200 kg DM/ha until P21. At P21, any ewes displaying harness marks on their rumps indicating returns to service were removed. The remaining ewes (n = 879; Kenyon et al. 2011) were randomly allocated to one of three nutritional regimes until P50: sub-maintenance (SM), maintenance (M) or ad lib (AD). The aim of the SM treatment was to achieve a loss in total ewe liveweight of 100 g/day while the aim of the M treatment was to ensure no change in total ewe liveweight. To achieve these targets, the length of the grazing period within each paddock was altered depending on herbage mass and liveweight measurements. The AD treatment aimed to provide unrestricted herbage under grazing conditions. Target pre- and post-grazing herbage masses for the SM treatment were 1000 and 700 kg DM/ha respectively and for the M treatment 1400 and 900 kg DM/ha, respectively. Morris & Kenyon (2004) have previously shown that ewe intakes do not differ above a minimum herbage cover of approximately 1200 kg DM/ha. Therefore, AD ewes were offered a herbage mass above this level. At P48, all ewes were pregnancy scanned via transabdominal ultrasonography and triplet-bearing ewes in the three groups were identified: SM (n = 33), M (n = 27), AD (n = 26). At P50, all triplet-bearing ewes were merged into one group and managed with the aim of achieving a total ewe liveweight increase during mid to late pregnancy at a level similar to that of the expected gravid uterus mass at term (i.e. maternal gravid uterine free weight to be maintained). Previous studies have indicated that the total gravid mass at term is in the range of 18–22 kg (Rattray et al. 1974; Grazul-Bilska et al. 2006; Kenyon et al. 2007). The duration of the grazing period within each paddock was dependent on herbage mass and ewe liveweight change previously measured.

On P139, ewes were set-stocked at 11.7 ewes/ha onto paddocks with an average herbage mass of 1357±111 kg DM/ha. Ewes remained in their treatment paddocks until 23 days after the midpoint of the lambing period (L23) and subsequently managed as one group, which was rotationally grazed with a minimum pre-grazing herbage mass of 1200 kg DM/ha until weaning (L92).

Herbage measurements

Pre- and post-grazing herbage masses were measured with a rising-plate meter (Ashgrove Pastoral Products, New Zealand; 50 readings per paddock) during the pregnancy period and herbage mass was measured monthly during lactation. The herbage mass was determined using the following formula:

where MR is the average meter reading (Hodgson et al. 1999).

During the period P21 to P50, pre-grazing herbage grab samples were collected at twice-weekly intervals, to mimic ewe intake; from P50 to P139, grab samples were collected fortnightly. At set-stocking, pasture plucks were taken and on a further four occasions during the lactation period. Samples were immediately frozen at -20 °C. The levels of ash, crude protein, lipid, neutral detergent fibre (NDF), acid detergent fibre (ADF), organic matter digestibility (OMD), metabolisable energy (ME) and starch and soluble sugars were determined by near-infrared reflectance spectrometry (NIR, FeedTech, Palmerston North, New Zealand). A Bruker MPA NIR spectrophotometer (Ettlingen, Germany) was used to scan the samples and the resulting NIR spectra were analysed using Optic user software (OPUS) version 5.0.

Ewe measurements

Liveweights of ewes were recorded on P21, P30, P50, P90, P137 and L92. Ewe condition score (scale 0–5 including half units (Jefferies 1961)) was determined on days P21, P50, P137 and L92.

At P131, 10 ml blood samples were collected from 48 ewes (SM, n = 17; M, n = 14; AD, n = 17) via jugular venipuncture (Heparin Becton Dickinson Vacutainer Systems, USA). Further blood samples were collected from 50 ewes at P139 (SM, n = 17; M, n = 14; AD, n = 19). All blood samples were immediately placed on ice before being centrifuged at 3500 rpm for 15 min. The plasma samples were frozen at -20 °C until analysis via radioimmunoassay (RIA) kits for glucose using a hexokinase assay method (Roche, Basel, Switzerland, sensitivity 0.11 mmol/l) and β-hydroxybutyrate (BOHB) concentrations using a Randox kit (Crumlin, Antrim, UK, 0.1 mmol/l). Russel et al. (1984) reported that B-OH concentrations of less than 0.8 nmol/l and/or a loss of body condition score of less than 0.5 units in late pregnancy indicated adequate nutrition.

Lamb measurements

Lambs were identified to their dam, tagged, their sex determined, weighed and had their girth and crown–rump, foreleg and hindleg lengths measured within 12 h of birth (L0). Lambs were re-weighed at L23, L50 and L92 (weaning). Causes of lamb deaths or dates of deaths were not recorded.

A random cohort of complete triplet-litters alive at 24–36 h of age were blood sampled via jugular venipuncture: SM (n = 45), M (n = 39), AD (n = 27). Blood samples were immediately placed on ice before being centrifuged at 3500 rpm for 15 min. The plasma samples were frozen at -20 °C until analysis via RIA kits for glucose (Roche Diagnostics, Basel, Switzerland, sensitivity 0.11 mmol/l) and gamma glutamyl transferase (GGT) activities (Roche Diagnostics, sensitivity 3 U/L).

The study was conducted at the Massey University Keeble Sheep and Beef farm, 5 km south of Palmerston North, New Zealand and was approved by the Massey University Animal Ethics Committee.

Data analysis

Ewe and lamb data were subjected to analyses of variance using the GLM procedure in Minitab 13.1 (Minitab Inc, Pennsylvania, USA) unless otherwise stated. The models used to analyse ewe liveweight, body condition score, metabolite concentrations and lamb liveweight, body dimensions and metabolite concentrations included the fixed effect of nutritional treatment during P21 to P50. Day of artificial insemination and sire were fitted as fixed effects. Rearing rank (number of lambs alive within a litter) and lambing paddock were fitted as fixed effects where appropriate. In addition, the potential interaction between ewe treatment and rearing rank was tested and, where significant (P< 0.05), the data are presented in the tables. Date of birth was fitted as a covariate, and sex of lamb as a fixed effect in the models for lamb liveweights, body dimensions and metabolite concentrations. The data for lamb liveweights, sizes, glucose and GGT were re-analysed with dam as a random effect. The results found did not differ from the above-mentioned models and thus are not reported.

The total lamb birth weight, a combination of the live weight of each individual lamb at birth was calculated for each ewe. In addition, the variation in lamb birth weight, the difference in live weight between the lightest and heaviest siblings within a set was calculated. Total weight of lambs per ewe at L92 was also calculated. Lambs not present at L92 were given a nominal weight of 0 kg. These three parameters were analysed as described above. These three parameters for each ewe were then analysed using models described previously.

Lamb survival from birth to L92, based on the presence of a lamb at this later time point, was analysed as a binomial trait via logit transformation (PROC GENMOD, SAS v5, SAS Institute Inc., Cary NC, USA) for categorical data modelling. Lamb GGT concentration data were not normally distributed and to achieve normality the data were transformed by square root and then analysed.

Results

Herbage mass and quality

During the period P21 to P50, pre- and post-grazing herbage mass of the AD treatment was greater (P < 0.05) than that of the M treatment, which in turn was greater (P < 0.05) than the SM treatment (Table 1). The range of pre-grazing masses for the SM, M and AD treatments were 778–1204, 1157–1710 and 1751–3500 kg DM/ha respectively, while the range of post-grazing masses for the SM, M and AD treatments were 607–1034, 800–1494 and 1299–1969 kg DM/ha respectively. During P50 to P139, when ewes were grazed together, the average pre- and post-grazing herbage mass was 1450±84 and 1011±33 kg DM/ha respectively. The range for pre-grazing masses was 951–1953 and the range of post-grazing masses was 774–1489 kg DM/ha. Average herbage mass during the period from set-stocking to L23 was 1490±79 kg DM/ha. During L23 to L92 pre- and post-grazing herbage mass was 1763±84 and 1524±84 kg DM/ha respectively.

Table 1  Effect of ewe nutritional treatment from P21 to P50 on pre- and post-herbage grazing mass (kg DM/ha) and metabolisable energy (ME) (MJ/kg) (means±SE). Means (± SE) within columns with letters in common or without superscripts are not significantly different (P > 0.05).

	Herbage mass (kg DM/ha)
	Pre-grazing	Post-grazing	ME (MJ/kg)
Sub-maintenance (SM)	996^a±89	814^a±54	12.35±0.26
Maintenance (M)	1479^b±107	1112^b±59	12.67±0.26
Ad lib (AD)	2331^c±82	1649^c±54	12.14±0.23

During P21 to P50, pre-gazing herbage ME values of the three treatments did not differ (P > 0.05; Table 1). The range of ME values for SM, M and AD treatments was 11.7–12.7, 12.4–12.7 and 11.7–13.0 MJ/kg respectively. During the period P51 to P139, the average ME value was 12.67±0.11 MJ/kg (range 11.9–13.0 MJ/kg) and during lactation it was 12.65±0.26 MJ/kg.

Ewe metabolic status

Ewe treatment had no effect (P > 0.05) on BOHB concentration at either time point (0.41, 0.51 and 0.56 and 0.61, 0.73 and 0.92 mmol/l for SM, M and AD ewe groups at P131 and P139 respectively). The square root of glucose concentrations of AD ewes (2.24±0.04 mmol/l) at P131 tended to be greater (P = 0.05) than M ewes (2.09±0.05 mmol/l) with SM ewes not differing from either (2.14±0.04 mmol/l, P > 0.05). Ewe treatment had no effect (P > 0.05) on glucose concentration at P139 (3.92, 3.87 and 4.01 mmol/l for SM, M and AD ewe groups respectively).

Ewe body condition score

At P50, AD ewes had higher (P < 0.05) condition scores than SM ewes (Table 2). AD ewes tended (P = 0.08) to have higher condition scores than M ewes at P50 (P = 0.08). There was no effect (P > 0.05) of ewe treatment on ewe condition score at P21, P137 or L92. At L92, ewes rearing just one lamb had greater (P < 0.05) condition scores (3.4±0.1) than those rearing either two (2.8±0.1) or three lambs (2.8±0.1), which did not differ (P > 0.05) from each other.

Table 2  Effect of ewe nutritional treatment from P21 to P50 on ewe condition score (ECS) at P21, P50, P137 and L92 (P, pregnancy; L, lactation) (means±SE). Means (± SE) within columns with letters in common or without superscripts are not significantly different (P > 0.05).

		ECS				ECS
	n	P21	P50	P137	n	L92
SM	33	3.0±0.1	2.3^a±0.1	2.9±0.1	30	3.1±0.1
M	27	2.9±0.1	2.9^b±0.1	3.0±0.1	24	3.0±0.1
AD	26	2.9±0.1	3.2^b±0.1	3.0±0.1	23	3.0±0.1

Ewe liveweight

AD ewes were heavier (P < 0.05) than both SM and M ewes at P50, with the latter two treatments not differing (P > 0.05; Table 3). There was no effect (P > 0.05) of ewe treatment on ewe liveweight at P21, P30, P90 P137 or L92. At L92, ewes that reared one lamb were heavier (P < 0.05) than those that reared two or three (data not shown). During the period P21 to P50, the change in liveweight per day differed (P < 0.05) between all treatments (-0.158±0.014, -0.026±0.016 and 0.146±0.016 kg/day for SM, M and AD treatments respectively).

Table 3  Effect of ewe nutritional treatment from P21 to P50 on ewe liveweight (kg) in pregnancy (P) and lactation (L) (means±SE). Means (± SE) within columns with letters in common or without superscripts are not significantly different (P > 0.05).

		Ewe liveweight (kg)						Ewe liveweight (kg)
	n	P21	P30	P50	P90	P137	n	L92
SM	33	69.1±1.4	66.3±1.1	64.5^a±1.0	69.1±1.1	86.9±1.3	30	75.5±1.6
M	27	67.6±1.3	68.2±1.2	66.9^a±1.2	71.0±1.3	87.7±1.5	24	72.0±1.8
AD	26	68.1±1.3	68.9±1.2	69.5^b±0.5	72.5±1.3	87.3±1.5	23	72.3±2.0

Lamb liveweight

Ewe treatment had no effect on lamb birthweight (Table 4). At L23, there was an interaction (P < 0.05) between lamb rearing rank and ewe treatment, such that in lambs born to SM ewes, triplet-reared lambs were lighter (P < 0.05) than those twin-reared. No such relationship (P > 0.05) was observed in lambs born to either M or AD ewes.

Table 4  The effect of ewe rearing rank (single-, twin-, triplet-rearing) and nutritional treatment from P21 to P50 and on lamb liveweight (LL) (kg) at birth, L23, L50 and L92 (means±SE). Means (± SE) within nutritional treatments, birth/rearing ranks and columns with letters in common or without superscripts are not significantly different (P > 0.05).

	n	LL at birth (kg)	n	LL at L23 (kg)	n	LL at L50 (kg)	n	LL at L92 (kg)
Rearing rank^1
Singleton	—	—	11	11.16^b±0.56	16	16.85^b±0.64	21	28.04^b±0.97
Twin	—	—	74	10.03^b±0.30	80	15.50^ab±0.36	72	26.08^ab±0.63
Triplet	—	—	96	9.28^a±0.29	60	14.50^a± 0.41	60	24.16^a±0.71
P21-50
SM	99	4.22±0.09	69	9.26^a±0.38	59	15.30±0.47	56	25.49±0.62
M	81	4.33±0.10	54	10.95^b±0.38	48	16.01±0.45	51	26.52±0.73
AD	78	4.27±0.11	58	10.53^ab 0.21	49	15.54±0.47	46	26.29±0.83
RR * P21–50^2
Singleton * SM	—	—	4	8.56^ab±0.93	—	—	—	—
Singleton * M	—	—	4	12.89^b±0.98	—	—	—	—
Singleton *AD	—	—	3	12.03^ab±1.12	—	—	—	—
Twin * SM	—	—	26	10.59^b±0.43	—	—	—	—
Twin * M	—	—	26	10.32^b±0.42	—	—	—	—
Twin *AD	—	—	22	10.00^ab±0.44	—	—	—	—
Triplet * SM	—	—	39	8.64^a±0.38	—	—	—	—
Triplet * M	—	—	24	9.43^ab±0.46	—	—	—	—
Triplet *AD	—	—	33	9.55^ab±0.41	—	—	—	—
^1Number of lambs still alive within litter.
^2Rearing rank by nutrition treatment interaction.

At both L50 and L92, lambs reared as singletons were heavier (P < 0.05) than those reared as triplets. Ewe treatment had no effect on lamb liveweight at either L50 or L92.

Lamb dimensions

Ewe treatment had no effect (P > 0.05) on lamb thoracic girth, crown–rump, foreleg and hindleg lengths (data not shown).

Lamb indices of colostrum uptake

Ewe treatment had no effect (P > 0.05) on glucose (5.24±0.33, 6.03±0.36 and 4.89±0.43 mmol/l for SM, M and AD treatments respectively) or the square root of GGT concentrations (36.66±2.72, 40.84±3.09 and 38.07±3.72 IU for SM, M and AD treatments respectively) of triplet lambs in which full sets were alive at 24–36 h of age.

Lamb survival to weaning

Ewe treatment had no effect (P > 0.05; 56.7, 62.7 and 57.7% for SM, M and AD treatments respectively) on lamb survival.

Difference in birthweights within litter, total lamb birthweight per ewe and total lamb liveweight at L92 per ewe

Ewe treatment had no effect (P > 0.05) on total lamb birthweight per ewe (12.67±0.24, 13.00±0.29 and 12.80±0.32 kg for SM, M and AD treatments respectively). Ewe treatment had no effect (P > 0.05) on the weight of the ‘lightest’ lamb (3.55±0.12, 3.59±0.14 and 3.54±0.16 kg for SM, M and AD treatments respectively), ‘middle’ lamb (4.30±0.11, 4.31±0.13 and 4.21±0.14 kg for SM, M and AD treatments respectively) or ‘heaviest’ lamb (4.82±0.09, 5.11±0.11 and 5.05±0.12 kg for SM, M and AD treatments respectively) within triplet litters. There was no effect (P > 0.05) on the difference in weight between the heaviest and lightest lamb within each litter (data not shown). Ewe treatment had no effect (P > 0.05) on total weight of lamb per ewe at L92 (41.25±4.11, 48.57±4.90 and 47.37±5.43 kg for SM, M and AD treatments respectively).

Discussion

The three nutritional treatments were successful in altering ewe liveweight during the feeding period with the liveweight changes achieved being close to those aimed for. This success allowed for the testing of the potential effects of early pregnancy nutrition on triplet-bearing ewes and their progeny.

The nutritional treatments in early pregnancy failed to have a major impact on the productive performance of the progeny to weaning, a result that concurs with the previously reported minimal or no effects in single- and twin-bearing ewes and their progeny (Everitt 1967; Parr et al. 1986; Krausgrill et al. 1999; Gopalakrishnan et al. 2004, 2005; Gardner et al. 2007, Kenyon et al. 2011). This indicates that triplet-bearing ewes can be managed under conditions that result in the ewe losing up to 0.15 kg/day liveweight, post-breeding prior to day 50 of pregnancy, without compromising their offspring's productive performance to weaning. This type of management matches the traditional approach of New Zealand farmers, when ewes were less fecund, which allowed for ewes to maintain or even lose liveweight during the early pregnancy stage in order to preserve feed for the lambing period (Coop 1986). However, it is important to remember that all ewes in the present study were managed in the mid to late pregnancy period, post-day 50 of pregnancy, with the aim of ensuring total ewe liveweight change was equivalent to expected gravid uterus weight at term. This resulted in no differences between treatments in ewe metabolic indices in late pregnancy or liveweight and body condition in late pregnancy or lactation. At P139, AD ewe BOHB concentrations may suggest these ewes were nutritionally stressed. However, given their minor change in body condition score in the mid to late pregnancy period and the fact that they had been managed with other ewes in this period that did not differ in BOHB concentrations nor display elevated BOHB, it is difficult to suggest the AD ewes were nutritionally stressed in late pregnancy. In addition, as aimed for, the total liveweight change observed in the ewes during the period from breeding to late pregnancy was approximately 21 kg, which is similar to the previously reported weight of the gravid uterus in late pregnancy of triplet-bearing ewes (Grazul-Bilska et al. 2006; Kenyon et al. 2007).

However, some caution is required as it is unknown if the general lack of effects observed in the ewes and their lambs would have been found if the ewes had been restricted in the mid to late pregnancy period, which can often occur under New Zealand's pastoral winter conditions. Further studies are required to investigate the effects of early pregnancy nutrition under conditions in which mid to late pregnancy nutrition is not optimal.

Conclusion

The present study successfully manipulated triplet-bearing ewe liveweight change in early pregnancy. However, these changes had no effect on progeny performance. Therefore, the data suggest that the nutrition of triplet-bearing ewes in early pregnancy can be controlled without any deleterious effects on the offspring to weaning. However, it is important to note that, in mid to late pregnancy, the ewes were managed to ensure gravid uterus development was not impeded. Further studies are required in which ewes are fed restricted levels in early pregnancy and then either restricted or non-restricted levels in mid to late pregnancy to determine if similar results are observed.

Acknowledgements

The authors wish to acknowledge the funding provided by Massey University and the National Research Centre for Growth and Development and the technical help provided by Massey University technical staff. The author P.R. Kenyon is partly funded by the National Research Centre for Growth and Development.