Intake, Digestibility, Milk Yield and Indicators of the Metabolic Status of Native Ewes Fed Supplemented Diet Under Grazing System

Abstract

This study aimed to evaluate the intake, digestibility, milk yield and indicators of the metabolic status of native lactating ewes supplemented under grazing system. In the present study 28 lactating ewes were selected: 14 Morada Nova (MN) and 14 Santa Inês (SI), distributed in a completely randomized experimental design, with a 2x2x2 factorial arrangement of two breeds, two supplementation levels and two experimental periods. Blood samples were collected from lambing to 70 days of lactation, with 14 days interval. Laboratory analyses consisted in determining serum glucose, triglycerides, cholesterol, urea and creatinine. Concentrate intake (g/day) during the lactation differed (P<0.05) between breeds. Total dry matter intake of ewes was affected (P<0.05) by breed and treatment. Morada Nova ewes presented lower intake than SI due to their smaller size. Serum glucose, triglycerides, cholesterol, urea and creatinine differed between breeds (P<0.05) with greater levels in SI ewes compared with MN. Concentrate supplementation did not affect (P>0.05) the serum levels of these metabolites. Indicators of the metabolic status were not affected by concentrate supplementation, which was effective for the maintenance of normal serum concentrations. However, these values differ between breeds within the permissible standards, indicating a high adaptation to postpartum.

Key Words:

• Concentrate
• Metabolites
• Milk production
• Morada Nova
• Santa Inês

Introduction

The establishment of small ruminant production is essential for the human survival in semiarid regions. Breeds adapted to the edaphic climatic conditions of these regions can improve the profitability of local livestock. As an example, the breeds Morada Nova (MN) and Santa Inês (SI), which are native animals from northeastern Brazil and are used for meat and skin production, are both highly valued in the international market (Souza et al., 2011). Despite the great cattle, rates of productivity, production and profitability show that the semi-arid northeastern Brazil has much to advance in this segment. Currently, production systems are extensively based on pastures of native Caatinga vegetation. Maintained in these conditions, the animals showed lower rates of weight gain, high mortality and low reproductive efficiency (Moreira et al., 2008). One decisive practice for the advance of these production systems is related to feed handling animals with the aim to provide feed and nutrients in quantity and quality to reduce slaughter age and improve product quality. As the use of tropical pastures alone cannot supply the nutritional requirements of animals (mainly the categories of higher nutritional requirements), the concentrate supplementation may be an important alternative. Meanwhile, its use for animals at grazing must comply with basic conditions, such as genetic potential of animals, the quality and quantity of forage available, concentrate price and final product price (Voltolini et al., 2008; Pompeu et al., 2009). This management is important because supplementation during early lactation aims to improve the availability of glucogenic precursors in the rumen, resulting in increased weight gain as well as body condition, impacting positively on milk production and production index, thus increasing the reproductive efficiency. In this case, the animal with insufficient energy intake may launch compensation means of metabolism by mobilizing its corporeal reserves (Ramin et al., 2005). This study aimed to evaluate the intake, digestibility, milk yield and indicators of the metabolic status of native lactating ewes supplemented under grazing system.

Materials and methods

Animals, feeding plans and management

The experiment was conducted at the Research Unit in Small Ruminants of Campus Cinobelina Elvas of the Federal University of Piauí, Bom Jesus, Brazil (09º04’28” South latitude and 44º21’31” West longitude), at an altitude of 277 m above sea level. The experiment was conducted with the approval of the ethical committee for animal experimentation of the Federal University of Piauí (ECAE/UFPI) under the number 091/2010.

In the present study twenty-eight pluriparous sheep were used, being 14 SI and 14 MN, with an average initial body weight (BW) (±SD) of 52.6±6.54 and 31.3±3.74 kg, respectively. Oestrus was synchronized using an intravaginal sponge containing 60 mg of medroxiprogesterone acetate (Progespon®; Syntex S.A, Luis Guillón, Argentina) for 9 days. After sponge removal, at the beginning of the FSH treatment, 200 U of equine chorionic gonadotropin (Folligon 5000®; Intervet, Kempton Park, South Africa) and 0.1 mg of cloprostenol sodium (Ciosin®; Intervet) were given intramuscularly to synchronize the follicular wave and to stimulate ovulations. After oestrous synchronization, ewes were artificially inseminated at fixed time with Dorper Breed semen. On day 60 post-artificial insemination, the pregnancy diagnosis was performed through trans-abdominal ultrasound and pregnant ewes from each genetic group were selected. These ewes remained until the middle of pregnancy in paddocks formed by andropogon pasture grass (Andropogon gayanus), being collected in the late afternoon and allocated in collective pens with cement floor, covered and surrounded with a screen containing feeders and having water ad libitum until the next day.

The experiment began at lambing and lasted 70 days, when the lambs were weaned. The sanitary management consisted of the deworming and vaccination against clostridial diseases, in addition to routine preventive care. During experimental period, the ewes remained in pasture paddock of andropogon grass (Andropogon gayanus), under continuous grazing, in an area with 1.8 ha and approximately 0.44 animal unit/ha, from 07:00 to 17:00 h. They were collected in the late afternoon and allocated in individual pens measuring 1.75 m². Sheep received supplementation with concentrated ground corn (70%), soybean meal (25%) and mineral supplement (5%), based on the dry matter (DM) (Table 1) established according to the weight of ewes (0.5% and 1.5% of BW), being adjusted weekly. The treatments were formed based on the percentage of concentrate supplementation, according to the recommendations by the National Research Council (2007) for animals in reproduction.

Body weight and body condition score

The BW of the ewes was registered weekly, the body condition score (BCS) was evaluated at the moment of the weekly weighing, being performed according to the method described by Thompson and Meyer (2006). Later, the averages were calculated for the periods 30 and 60 days postpartum. The measurements were made on a scale from 1 to 5 (1=under-weight and 5=overweight).

Determination of intake and digestibility

For estimation of intake and digestibility of nutrients we used as internal indicator the indigestible dry matter (iDM). In order to evaluate the levels of indigestible components of feed (grass: 0.4 g and concentrate: 0.4 g) and faeces (0.4 g), they were packed in 50-µm nylon bags (8x12). The iDM was obtained after the bags were kept for 264 h of incubation in situ of the supplied feed and faeces according to Casali et al. (2009), being used one animal cannulated in the rumen under the same production system.

After the removal of the rumen, the bags were washed with water until full clearance and immediately transferred to the forced-ventilation greenhouse (60°C) during 48 h. Then, they were dried in a non-ventilated oven (105°C for 45 min), put in a desiccator and weighed (Detmann et al., 2001) to obtain undigested DM. The determination of neutral and acid detergent fibre (NDF and ADF) followed the methods described by Van Soest et al. (1991).

The content of iDM was obtained by the difference of the dry weight of the material before incubation and the dry weight of the residue after incubation.

Faecal production was obtained from an external indicator, the hydroxyphenylpropane - LIPE^® (Saliba et al., 2003). The LIPE^® was given to animals at dose of 250 mg, before the first meal of the day, with two days for adaptation and five days to collect the faeces. The faeces were collected directly in the rectum once a day, always at the same time (07:00 h), after a composite sample was formed based on the pre-dry weight per animal. During the collection period, the daily samplings of feed were carried out. These samples were properly frozen and then dried (60°C) processed in a grinder with a 1-mm sieve and submitted to laboratory analysis. The analyses for the determination of DM, crude protein (CP), NDF and ADF were conducted according to the methodology described by Silva and Queiroz (2002).

The calculations of digestibility and DM intake were made according to the following equations:

DMD (%) = 100 – [100 x (indicator in the diet/indicator in the faeces)]

Faecal output (g/day) = [ingested indicator (g)/indicator in the faeces (g/g DM)]

DMI (g/day) = {[(DMfaecal x % indicator in the faeces) - nidicator in the diet]} + DMIsuppl/% indicator in the forage

where, DMD is dry matter digestibility, DMI is dry matter intake, DMIsuppl is dry matter intake of supplementation.

Milk yield

Milk yield was determined using the protocol described by Doney et al. (1979). Once a week, after being separated from their lambs, the ewes were injected intramuscularly with 5 U of synthetic oxytocin (0.5 mL) and milked immediately by hand milking. After the first milking, the ewes were returned to their paddock, while their lambs were kept an in separate pen, out of sight from their mothers. After 2 h, the ewes received a second intramuscular injection of synthetic oxytocin (5 U) and the milk output at the second milking was recorded using a graduated cylinder. The yielded volume was then multiplied by 12 to calculate daily milk production and, consequently, multiplied by 7 to calculate week milk production.

Determination of blood parameters

Blood samples were collected from lambing, every 14 days, to 70 days of lactation. Blood was collected always in the morning (before the sheep were released to graze) by jugular venipuncture using needles coupled to vacuum tubes (10 mL) without anticoagulant. The serum was obtained after the centrifugation at 3500 rpm during 15 min in a refrigerated centrifuge and stored in micro tubes at -20°C until the time of laboratory analysis, which consisted in determining serum glucose, triglycerides, cholesterol, urea and creatinine.

All the biochemical analysis were performed in semi-automatic biochemical analyzer (Espectrum BS- 001 000 013), using commercial kits of Labtest® (Milan, Italy), following the manufacturer’s recommendations.

Statistical procedures

To evaluate the intake, digestibility, milk yield and variation in weight gain as well as in the BCS, the experimental design was a randomized 2x2x2 factorial arrangement with two breeds (MN and SI), two supplementation levels (0.5 nd 1.5% BW) and two experimental periods (30 and 60 days postpartum). The data were subjected to normality test and homoscedasticity was verified.

Factorial analysis was carried out using data of supplementation level, BW, BCS, intake, digestibility, week of lactation and milk yield. The factorial analysis performs a variety of common factor and component analyses, with measures carried out using as multivariate data. The first two factors that explained the greater proportion of data variation were selected. The results of the factors, therefore, show the relationship between the variables in the study. The statistical analyses were carried out using MIXED and FACTOR procedures of the SAS (1999) and the means were compared by Tukey’s test at a 5% probability.

For the blood variables, the analysis took into account data from the same animals across time. Factorial arrangement was performed for analyzing the data. Effect of fixed factors (supplementation level of 0.5, and 1.5% BW; MN and SI breeds) and days (longitudinal time over which the experiment was carried out, i.e. Day 0, Day 14, Day 28, Day 42, Day 56 and Day 70) were also considered. The results were submitted to ANOVA using the GLM procedure of SAS (SAS, 1999) and means were compared by SNK’s test at a 5% probability.

Results and discussion

The average BW (kg) and BCS of the ewes during lactation were influenced (P<0.05) by period, breed and supplementation level (Table 2). There was no influence of the lactation period on BW of MB ewes. However, SI ewes subjected to 0.5% of supplementation showed statistical difference (P<0.05) between periods, which did not occur with SI ewes supplemented with 1.5%, mainly due to the intense energy and protein metabolism that always occurs during lactation by the demand for nutrients to yield milk.

The BCS differed between periods in MN ewes supplemented with 0.5% and SI ewes supplemented with 1.5% of concentrate supplementation and, in both treatments, an increased body condition occurred at 60^th day of lactation. In this period, SI ewes supplemented with 1.5% – which had probably recovered from postpartum stress – presented a greater DMI, promoting good recovery of body condition (Torreão et al., 2008).

In the immediate postpartum, MN ewes supplemented with 1.5% exhibited better BCS (P<0.05) compared to treatment of 0.5%, as well as in relation the SI ewes. Also, in the second period, they kept their score and SI ewes receiving 1.5% presented similar scores (P>0.05).

The concentrate intake in g/day (DMI) during lactation is shown in Table 3. Difference (P<0.05) between breed and supplementation level was verified, with higher values in SI ewes and ewes subjected to 1.5% of concentrate supplementation. This superiority in concentrate intake in SI ewes reflects the greater need for intake due to their larger size. As SI ewes are more demanding than MN ewes, they demand better forage quality and/or higher amount of concentrate supplementation (Silva Sobrinho, 2006; Facó et al., 2008).

In the current study the ewes were submitted to similar experimental conditions and pasture quality. Thus, SI ewes tried to supply the deficiency of this type of pasture with a higher (P<0.05) concentrate intake by metabolic size unit. This low pasture quality was verified by Menezes et al. (2010), who compared andropogon with other grasses. He verified increased intake of concentrate supplementation in animals fed with andropogon, possibly due to its lower nutrient quality, especially in this critical period which demands a high quality diet to avoid big losses arising from the negative energy balance (NEB).

The DMI was not influenced (P>0.05) by lactation advance, indicating that the concentrate levels met the nutritional requirements, once there is an increase in nutritional requirements during this period, due to the gradual increase in milk production towards the peak of production, maintenance requirements and possible NEB recovery. Another factor would be the return of the total capacity of the rumen fill, which was prevented during the final third of pregnancy, characterized by 80% of fetal growth (Robinson et al., 1999).

The DMI based on iDM was affected (P<0.05) by breed and concentrate supplementation levels to which these ewes were submitted (Table 4). Also, significant interaction between period×supplementation, breed×treatment, breed×period, and triple interaction (breed×treatment×period) was found.

Evaluating the intake, an increase (P>0.05) in the total DMI by the ewes that received the highest supplementation level was verified, corroborating findings by Rodrigues et al. (2007). In general, these ewes showed unexpected ingestive behaviour (i.e., with a high concentrate intake a high forage intake occurred). This result was not expected in ewes subjected to 1.5% of concentrate supplementation, but it may have occurred due to excess of rumen degradable protein. Thus, they increased the intake to compensate excess of protein and maintain a good relation between proteins: fermentable carbohydrate to maintain ruminal homeostasis. Therefore, DMI is controlled by physiological factors of short and long term, in which the control is done by the nutrient balance diet, specifically related to the energy balance maintenance (Mertens, 1994; Van Soest, 1994).

This current study shows that there was a negative effect (P<0.05) on intake with the advance of postpartum. A curious fact is that uterine involution and capacity of maximum rumen fill showed a decrease in roughage intake for both breeds in the two treatments, without compensation in the concentrate intake (P>0.05) in the same periods.

The breed factor was decisive (P<0.05), where SI ewes consumed a higher amount of forage as well as concentrate intake, due to their larger size.

Triple interaction (Table 4) showed that DMI (g/day), DMI (g/kg^0.75) and DMI (%BW) was influenced. This result explains that intake was affected by multiple factors such as concentrate supplementation, breed and period of lactation.

The DM digestibility (Table 4) was higher (P<0.05) in SI ewes, However, the supplementation influenced only MN ewes, with higher digestibility in animals feed high concentrate, due to increased availability of nitrogen compounds in the rumen and, consequently, greater proliferation of microorganisms that degrade fibres. In addition, Van Soest (1994) reported the gradual replacement of NDF of corn silage by non-fibre carbohydrates of the concentrate, due to rapid and high digestion in the digestive tract. This result is in agreement with that of Macedo Junior et al. (2007), who stated that concentrate level, low-fibre quality diet in sheep (especially ewes lactating) can affect intake and digestibility of dry matter.

In the current study, SI ewes produced more (P<0.05) kg of milk than MN ewes (Table 5) due to their larger size, higher production capacity and higher intake. The supplementation affected (P<0.05) milk production of the SI ewes, with greater values in ewes submitted to 1.5% BW. Milk yield in the present study was superior to some previous studies (Ribeiro et al., 2007; Araujo et al., 2008; Pulina et al., 2012). For this production level, even supplemented, these ewes may have used body reserves reflected by smaller BCS at that stage (Table 2).

Evaluating the lactation progress, a significant increase (P<0.05) in SI ewes is verified. This behaviour reflects a normal lactation curve, that shows an increase to reach peak yield and according to Ribeiro et al. (2007) the peak yield of the SI ewes was reached at 37 days of lactation. According to the results concerning the metabolic parameters evaluated during lactation, it is verified that serum glucose, triglycerides, cholesterol, urea, creatinine differ (P<0.05) according to the genetic group (Table 6).

There was no influence (P>0.05) of concentrate supplementation on serum levels of these metabolites during the postpartum in ewes. At this stage, negative energy balance becomes more pronounced (Caldeira et al., 2007; Kaneko et al., 2008; Scarpino et al., 2014). Therefore, the sheep needs to produce milk in sufficient quantities to their offspring. In the synthesis of milk, the glucose (main molecules of energy metabolism) is involved. This mechanism occurs as consequence of the main objective of the adaptation process to the negative energy balance: to limit the mobilization of muscle proteins for the gluconeogenesis maintenance (Herdt, 2000; Kaneko et al., 2008). Thus, during the NEB, the muscles get the energy derived from fat, including non-esterified fatty acids (NEFA) and ketone bodies, although the ruminant may be less efficient in the use of NEFA than are other species. Changing the energy sources, the utilization of glucose is reduced, helping the maintenance of the plasma concentrations (Lemor et al., 2009).

There was no effect (P>0.05) of supplementation on serum glucose. An opposite result was reported by Balaro et al. (2012), who described an elevation of levels according to supplementation, and according to Van Soest (1994), who showed a consequent increase in the synthesis of propionate in the rumen, indicating higher consumption and availability of protein coming from the diet.

Evaluating different breeds, a significant effect (P<0.05) was found for the glucose levels that remained with higher serum levels for SI ewes, with values above those cited by Kaneko et al. (2008). Opposite results were reported by Bizelis et al. (2000), who described that the demand for glucose by mammary gland is so great that decreased serum concentration of this metabolite can occur when compared with the phase of pregnancy. However, the current study showed that higher serum glucose values compared to Silva et al. (2015) reported on indigenous breeds during pregnancy.

In addition, the uterine involution and milk yield are determinant factors for return of their full capacity of feed intake. Even if there is a correct balance between the ingress and egress of nutrients, as well as metabolites, can also occur nutrient mobilization of body tissues to supply the energy demand (Scarpino et al., 2014).

The interaction between breed and days of lactation (P<0.05) for the metabolites evaluated in this study are presented in Figures 1, 2, 3, 4 and 5.

Table 6 shows higher average blood glucose for SI ewes when compared to MN ewes during the postpartum period. However, significant difference (P<0.05) was found at 42 days of lactation (Figure 1). Rising levels up to 28 days of lactation were verified; this was probably due to the greater availability of glucose for the ewes to reach peak milk yield (Cardoso et al., 2010).

Based on data on the concentration of triglycerides, effect of breed (P>0.05) was observed with the highest concentrations for MN ewes. Considering what described above, it is perceived that the highest values were found at lambing stabilizing at day 14^th of lactation. Highest concentrations were found in MN ewes at lambing due to highest accumulation capacity and their adaptation to experimental conditions. The SI ewes probably used this metabolite at higher intensity during pregnancy to supply their needs for maintenance and growth fetal.

The analysis of cholesterol (Table 6 and Figure 3) showed how the breed factor influenced the concentrations of this metabolite. It is known that in the synthesis of triacylglycerols, the circulating NEFA can be esterified, mainly in triacylglycerols, phospholipids and esters of cholesterol, resulting in rapid accumulation of these substances in the organ (Caldeira et al., 2007). Thus, it can be perceived that the MN ewes exhibited lower cholesterol values than SI ewes in the different periods (P<0.05), which was possibly due to their lower demand of utilization and which caused the lower mobilization of hepatic triglycerides.

The serum urea showed relationship inversely proportional to the concentrate supplementation levels, although there was no difference (P>0.05) between treatments (Ziguer et al., 2012). When breed effect was evaluated, higher values were found for SI ewes (P<0.05). According to Vosooghi-Poostindoz et al. (2014), the urea can be a good indicator of the protein status in sheep and it varies with the level of element intake. However, in this study, the effect of nutritional plans on the urea concentration was not observed.

During lactation low variation in serum urea of SI ewes was found (Figure 4). An opposite result was found in MN ewes at lambing (Contreras et al., 2000), probably due to the higher protein catabolism to supply the appropriate levels in this period (lambing) and at 56 days of lactation.

A deficient nutrition can cause an increase in serum urea (Andrews et al., 1996), due to increased catabolism of endogenous nitrogen compounds in order to reverse energy deficiency. It should be reported that overnutrition may exhibit high concentrations of urea and cause an increase in ammonia production in the rumen with excess of nitrogenous compounds absorbed in the intestine. In contrast, a decrease in serum urea can be observed when there is drastic reduction in feed intake or a minimum mobilization of protein reserves (Caldeira et al., 2007).

In evaluating creatinine levels, higher concentrations were observed for SI ewes (P<0.05), which resulted in higher concentrations of serum glucose (Table 6). In animals under energy deficiency condition and under severe conditions of mobilization of body protein, mobilization of lipid reserves may occur. In addition, an increase of serum creatinine to maintain the homeostasis and normal blood glucose (Rodrigues et al., 2007) may occur. According to Brito et al. (2006), the body reserves are considered mobile portion of the BW. Therefore, in situations of deficiency or imbalanced production, they can be mobilized to supply this deficiency.

The factorial analysis (Figure 6) shows that the first factor indicates a positive correlation among BW, DMIc, DDM and milk production, being verified that animals of larger size had higher forage intake and concentrate with a consequent increase in the protein-carbohydrate interaction and, consequently maximization of microbial synthesis and the efficiency of utilization of dietary nutrients (Van Soest, 1994; Voltolini et al., 2008), high degradation rates (Mertens, 1994) and high digestibility (Medeiros et al., 2007).

According to the second factor, a high supplementation level promoted a decrease in DM intake (g/day), maybe in function of the substitution effect with high concentrate intake (Voltolini et al., 2011).

Figure 1. Plasma concentration of glucose in lactating ewes. Differences for each time point between breeds are shown with *P<0.05 (means±standard deviations represented as vertical bars).

Figure 2. Plasma concentration of triglycerides in lactating ewes. Differences for each time point between breeds are shown with *P<0.05 (means±standard deviations represented as vertical bars).

Figure 3. Plasma concentration of cholesterol in lactating ewes. Differences for each time point between breeds are shown with *P<0.05 (means±standard deviations represented as vertical bars).

Figure 4. Plasma concentration of urea in lactating ewes. Differences for each time point between breeds are shown with *P<0.05 (means±standard deviations represented as vertical bars).

Figure 5. Plasma concentration of creatinine in lactating ewes. Differences for each time point between breeds are shown with *P<0.05 (means±standard deviations represented as vertical bars).

Figure 6. Two first factors of supplementation level, week of lactation and dry matter intake (g/day), body weight (kg kg^0.75), dry matter intake (%BW and g/kg^0.75), dry matter intake of concentrate (g/day and g/kg^0.75), dry matter digestibility (%) and milk yield. The first two factors explained 81.73% of the variance among variables.

Table 1. Chemical composition of experimental ingredients, concentrate supplement and andropogon (Andropogon gayanus).

Ingredients	DM, %	CP, %	EE, %	TDN, %	NDF, %	ADF, %	Ca, %	P, %
Corn meal	87.19	9.98	5.19	67.50	24.55	5.87	0.05	0.49
Soybean meal	88.48	48.76	1.75	80.73	15.37	9.64	0.33	0.57
Mineral mixture^°	97.91	-	-	-	-	-	18.00	13.00
Supplement	88.04	19.17	4.06	67.43	21.02	6.51	0.90	1.13
Andropogon	27.45	7.50	2.02	53.56	74.70	41.97	0.33	0.11

DM, dry matter; CP, crude protein; EE, ether extract; TDN, total digestible nutrients; NDF, neutral detergent fibre; ADF, acid detergent fibre; Ca, calcium; P, phosphorus.

°Mineral mixture provided the following per kg of diet: zinc, 1600 mg; copper, 600 mg; manganese, 1500 mg; iron, 1100 mg; cobalt, 10 mg; iodine, 27 mg; selenium, 22 mg.

Table 2. Body weight and body condition score of Morada Nova and Santa Inês ewes based on breed, supplementation level and period of lactation

	MN		SI
	0.5% BW	1.5% BW	0.5% BW	1.5% BW
BW, kg
Initial	28.1±1.9	29.4±1.1	45.3±1.4	53.6±1.3
30 days	28.9±1.1^a^C^#	30.1±1.3^a^C^#	45.1±0.9^a^B^°	54.1±1.3^Footnotea A°
60 days	28.7±1.5^a^C^#	29.5±1.1^a^C^#	43.5±1.2^b^B^°	54.2±1.5^Footnotea A°
BCS, 1-5
Initial	2.1±0.2	2.9±0.2	2.1±0.1	2.7±0.2
30 days	1.8±0.1^b^D#	2.9±0.1^a^A^°	2.3±0.2^a^C^°	2.5±0.3^b^B^#
60 days	2.2±0.1^a^B#	2.9±0.2^a^A^°	2.2±0.1^a^B#	2.9±0.2^a^A^°

MN, Morada Nova; SI, Santa Inês; BW, body weight; BCS, body condition score (measures were on a scale from 1=underweight to 5=overweight).

^a,bMeans followed by the same lowercase letter in columns do not differ (period effect) from Tukey’s test at a 5% of probability;

^A-Dmeans followed by the same uppercase letter in rows do not differ (treatment effect) from Tukey’s test at a 5% of probability;

^#,°means followed by the same symbol in rows do not differ, treatment effect, from Tukey’s test at a 5% of probability.

Table 3. Dry matter intake of ewes fed a supplemented diet under grazing system at different periods of lactation

Parameters	0.5% BW		1.5% BW		Effect
	30 days	60 days	30 days	60 days	T	P	B	T×P	B×T	B×P	B×T×P
MN (n=14)
DMI, g/day	142.66±1.32^B^b^#	142.41±1.23^B^b^#	448.96±1.23^B^a^#	443.76±1.23^B^a^#	^**	ns	^*	ns	ns	ns	ns
DMI, g/kg^0.75	11.54±0.84^B^b^#	11.55±0.49^B^a^#	35.03±0.84^B^b^#	34.91±0.49^B^b^#	^**	ns	^*	ns	ns	ns	ns
DMIc, g/day	131.49±0.68^B^b^#	131.26±0.75^B^b^#	413.80±0.68^B^a^#	409.01±0.75^B^a^#	^**	ns	^*	ns	^*	ns	ns
DMIc, g/kg^0.75	10.64±0.32^B^b^#	10.64±0.46^B^b^#	32.28±0.32^B^a^#	32.18±0.46^B^a^#	^**	ns	^*	ns	^*	ns	ns
SI (n=14)
DMI, g/day	225.0±1.1^A^b^°	217.70±1.21^A^b^°	806.61±1.1^A^a^°	813.05±1.21^A^a^°	^**	ns	^*	ns	ns	ns	ns
DMI, g/kg^0.75	12.91±1.32^A^b^°	12.81±1.24^A^b^°	40.35±1.24^A^a^°	40.53±1.24^A^a^°	^**	ns	^*	ns	ns	ns	ns
DMIc, g/day	207.38±1.24^A^b^°	200.65±1.24^A^b^°	743.45±1.1^A^a^°	749.39±1.1^A^a^°	^**	ns	^*	ns	^*	ns	ns
DMIc, g/kg^0.75	11.90±1.1^A^b^°	11.81±1.02^A^b^°	37.19±1.1^A^a^°	37.35±1.1^A^a^°	^**	ns	^*	ns	^*	ns	ns

BW, body weight; T, treatment; P, period; B, breed; T×P, interaction between treatment and period; B×T, interaction between breed and treatment; B×P, interaction between breed and period; B×T×P, interaction among breed, treatment and period; MN, Morada Nova; SI, Santa Inês; DMI, dry matter intake; DMIc, dry matter intake of concentrate.

^a,bMeans followed by the same lowercase letter in rows do not differ (treatment effect) from Tukey’s test at a 5% probability; ^A,Bmeans followed by the same uppercase letter in rows do not differ (period effect) from Tukey’s test at a 5% probability;

^#,°means followed by the same symbol in columns do not differ (breed effect) from Tukey’s test at a 5% of probability.

*P<0.05;

**P<0.01; ns, not significant.

Table 4. Total dry matter intake and digestibility based on indigestible dry matter in ewes fed a supplemented diet under grazing system at different periods of lactation.

Parameters	0.5% BW		1.5% BW		Effect
	30 days	60 days	30 days	60 days	T	P	B	T×P	B×T	B×P	B×T×P
MN (n=14)
DMI, g/day	600.79±0.56^B^a^#	394.96±0.56^B^b^#	820.77±0.68^A^b^#	691.28±0.68^A^b^#	^*	^*	^*	^*	^*	^*	^*
DMI, g/kg^0.75	49.28±0.32^B^b^#	32..13B±0.68^c^c^#	64.20±1.21^A^b^#	54.71±1.23^B^a^b^#	^*	^*	^*	^**	^*	^*	^*
DMI, %BW	2.14±1.22^A^b^#	1.39±0.98^A^a^#	2.75±0.76^B^b^#	2.35±0.56B^a^b^#	^*	^*	^*	^*	^*	^*	^*
DMD, %	56.66±0.12^b^#	57.50±0.0#	68.00±0.0^a^#	63.33±0.0#	^*	ns	^*	^**	^*	ns	ns
SI (n=14)
DMI, g/day	1191.74±0.43^B^a^°	878.91±1.21^B^b^°	1623.82±1.21^A^a^°	1352.02±1.32^A^b^°	^*	^*	^*	^*	^*	^*	^*
DMI, g/kg^0.75	69.73±0.58^B^b^°	52.19±0.0^B^b^°	82.47±0.21^A^b^°	68.29±0.21^B^b^°	^*	^*	^*	^**	^*	^*	^*
DMI, %BW	2.71±0.21^B^a^°	2.04±0.1^B^a^°	3.07±0.1^A^a^°	2.53±0.1^A^b^°	^*	^*	^*	^*	^*	^*	^*
DMD, %	70.00±0.23^°	65.50±0.3^°	76.00±0.3^°	78.00±0.1^°	ns	ns	^*	ns	ns	ns	ns

BW, body weight; T, treatment; P, period; B, breed; T×P, interaction between treatment and period; B×T, interaction between breed and treatment; B×P, interaction between breed and period; B×T×P, interaction among breed, treatment and period; MN, Morada Nova; SI, Santa Inês; DMI, dry matter intake; DMD, dry matter digestibility.

^a-dMeans followed by the same lowercase letter in rows do not differ (treatment effect) from Tukey’s test at a 5% of probability;

^A,Bmeans followed by the same uppercase letter in rows do not differ (period effect) from Tukey’s test at a 5% of probability;

^#,°means followed by the same symbol in columns do not differ (breed effect) from Tukey’s test at a 1 and 5% of probability.

*P<0.05;

**P<0.0; ns, not significant.

Table 5. Milk yield (kg/week) of native ewes fed a supplemented diet under grazing system during the maternal-dependent phase.

Week of lactation	MN		SI
	0.5% BW	1.5% BW	0.5% BW	1.5% BW
1	5.73±1.38^A^a^°	5.18±1.28^A^a^°	7.66±0.91^C^b^#	10.78±1.18^B^a^#
2	6.02±1.38^A^a^°	5.60±1.28^A^a^°	9.29±0.91^B^b^#	11.37±1.18B^A^a^#
3	5.64±1.38^A^a^°	6.23±1.28^A^a^°	10.59±0.91^B^A b^#	11.96±1.18^A^a^#
4	3.46±1.38^A^a^°	5.32±1.28^A^a^°	11.46±0.91^A^b^#	12.93±1.18^A^a^#
Total	20.85±1.38^°	22.33±1.28^°	39.00±0.91b^#	47.04±1.18^a^#

MN, Morada Nova; SI, Santa Inês; BW, body weight.

^a,bMeans followed by the same lowercase letter in rows do not differ (treatment effect) from Tukey’s test at a 5% of probability;

^A-Cmeans followed by the same uppercase letter in rows do not differ (week of lactation effect) from Tukey’s test at a 5% of probability;

^#,°means followed by the same symbol in columns do not differ (breed effect) from Tukey’s test at a 1 and 5% of probability.

Table 6. Average serum concentrations of metabolic parameters of lactating ewes according to breed and supplementation level.

Parameters	Reference values°	Breed		Supplementation level		Effect
		MN	SI	0.5%	1.5%	T	P	B	T×P	B×T	B×P	B×T×P
GLU, mmol/L	1.7-3.6	3.55±0.24^b	4.08±0.3^a	3.80±0.92	3.75±0.81	ns	^*	^*	ns	ns	^*	ns
TRI, mmol/L	-	0.28±0.02^a	0.22±0.0^b	0.24±0.10	0.26±0.16	ns	^*	^*	ns	ns	^*	ns
CHO, mmol/L	1.05-1.50	1.23±0.17^b	1.47±0.0^a	1.37±0.09	1.33±0.12	ns	^*	^*	ns	ns	^*	ns
URE, mmol/L	3.0-10	5.24±0.3^b	6.47±0.25^a	6.31±0.44	5.90±0.56	ns	^*	^*	ns	ns	^*	ns
CRE, µmol/L	70-105	66.50±1.05^b	75.64±1.26^a	69.07±1.97	63.14±1.21	ns	^*	^*	ns	ns	^*	ns

MN, Morada Nova; SI, Santa Inês; T, treatment; P, period; B, breed; T×P, interaction between treatment and period; B×T, interaction between breed and treatment; B×P, interaction between breed and period; B×T×P, interaction among breed, treatment and period; GLU, Glucose; TRI, triglycerides; CHO, cholesterol; URE, urea; CRE, creatinine.

°Kaneko et al. (2008).

^a,bMeans followed by the same letter in columns are not significantly different from test Student-541 Newman Keuls.

*P<0.05; ns, not significant.

Conclusions

Milk yield was affected by breed, with higher values in SI ewes. In addition, there was an increase in milk yield in ewes supplemented with 1.5% of concentrate. Metabolic parameters were not affected by concentrate supplementation, which was effective for the maintenance of normal serum concentrations. However, these values differ between breeds within the permissible standards, indicating a high adaptation to postpartum.

Acknowledgments

The dissertation of the first author Master’s Degree is financially supported by Northwestern Brazil Bank (BNB) and National Counsel of Technological and Scientific Development (CNPq). The authors would like to thank BNB and CNPq for financial support, as well as Brazilian Federal Agency for the Support and Evaluation of Graduate Education (CAPES) for concession of the scholarship.