Selenium (Se) is essential in the antioxidant defense system in animals and humans. Among its functions are to serve as part of the enzyme glutathione peroxidase (GSH-Px), which destroys free radicals in the cytoplasm and protects tissues from oxidative stress¹. Se has been studied for its functions in the immune system and DNA protection². On the other hand, deficiency of the mineral is associated with diseases such as white muscle, and repression of immunity in lambs. The negative effects of the deficiency are explained by the relationship between the mineral and the hormones produced by the thyroid^3,4.

Sodium selenite (Na₂SeO₃) has been the preferred source of inorganic Se in ruminant feeding. However, with the appearance of new sources of organic Se⁵, the question as to which is better arose. Most of the mineral is found as GSH-PX and selenoproteins that are produced in the liver and distributed in serum. However, there are not many reports of the relationships between dietary Se and serum concentrations of albumin, α-, ß- and ∂-globulins in sheep.

Selenite is mainly used for the formation of selenoenzymes and differs from organic forms of greater availability than Se + yeasts^6,7,8. Others have shown that it is possible to increase the Se content of meat with Sel-Plex^®(9,10,11 compared to SS. Both forms increase in animal tissues, improving the consumption of Se by humans who ingest meat from animals supplemented with Se.

Based on the above, the objective was to study the effects of supplementation of the source of Se on feed intake, changes in body weight, feed efficiency, carcass weights, selenium in serum, muscles, albumin and globulins of growing Pelibuey lambs.

The study was carried out in the sheep module of the Experimental Farm of the Chapingo Autonomous University, in Chapingo, Mexico (98° 29 ́23 ́ ́N and 98° 53 ́27 ́ ́ W); at 2,250 m altitude with temperate subhumid climate. The temperature varies from 12 to 18 °C, with 645 mm of annual precipitation, distributed from July to September¹².

The study used 50 newly weaned Pelibuey lambs (LW=23 ± 1.1 kg; five to six months old), which were stratified and randomly assigned to one of five treatments (n=10): 1) basal diet (BD, C); 2) C+0.30 mg of Se kg^-1 DM, of SS, 30SS; 3) C+0.90 mg of Se kg^-1 DM, of SS, 90SS; 4) C+0.30 mg of Se kg^-1 DM (Sel-plex™ OSe; Alltech, Inc., Nicholasville, KY), of SP, 30SP; and 5) C+0.90 mg of Se kg^-1 DM, of SP, 90SP. The diet was formulated according to the recommendations of the NRC¹⁰. In the final composition, diet C contained 0.1 mg of Se per kg of DM, while 03SS and 03SP contained 0.4 mg. In the same sense, diets 09SS and 09SP contained 1.0 mg of Se per kg of DM (Table 1).

The lambs received feed twice a day. Fifty (50) percent of the feed offered was served at 0700 h and the rest at 1500 h. These animals were trained to eat in individual Calan door-type feeders (American Calan, Inc.; Northwood, NH, US), equipped with a container of approximately 15 kg. Feed was offered ad libitum (15 % more than the previous day’s intake). The portion was weighed, recorded and deposited in the feeders. Uneaten feed was removed, weighed and recorded. For each animal, a sample of the consumed feed and the non-consumed feed was obtained weekly.

The total DM was determined in an oven at 100 ºC for 24 h and incinerated in a muffle at 500 ºC to quantify the OM and ash content. DM intake was estimated by multiplying daily feed intake by its DM content.

The NDF and ADF contents of the diets were quantified following the procedures of Goering and Van Soest¹³; while the protein was obtained by Kjeldahl¹⁴. Changes in live weight (LW) were recorded weekly and used to calculate DWG. The lambs were slaughtered at the end of the fattening period following official slaughter procedures¹⁵.

The carcass yield, expressed as a percentage, was calculated as the proportion of the weight of the hot carcass, divided by the LW and multiplied by 100. The weight of the cold carcass was obtained 24-30 h after cold storage at approximately 2 ± 2 ºC. Carcass yield was recalculated and reported as cold carcass yield.

The research protocols and management procedures were carried out following the Official Mexican Standard (NOM-051-ZOO-195). During mobilization, the animals were treated in accordance with the standard NOM-024-ZOO-195.

Every 14 d, a blood sample, approximately 10 ml, was taken by puncture of the jugular vein, in vacutainer tubes without anticoagulant (Beckton-Dickinson, Franklin Lakes, NJ). The samples were kept in the environment for 60 min in order for them to coagulate and then they were refrigerated. The blood was centrifuged at 1,000 xg for 25 min at 4 ºC. The serum was stored in vials at -20 ºC and later sent to the laboratories of the Faculty of Veterinary Medicine and Zootechnics of UNAM, for the determination of albumin, α-, ß-, ∂-globulin and Se. The globulins were determined by the procedures of Connell et al¹⁶; while the Se in serum was quantified with a spectrofluorometer (Perkin Elmer model LS30), following the procedures of Tamari et al¹⁷.

After 48 h of slaughter, the Longissimus dorsi, Gluteus maximus and Musculus deltoideus muscles were removed from each carcass according to the procedures of Covenin¹⁸. From each muscle and carcass, three cuts approximately 2.54 cm thick were obtained and packed individually. All cuts were frozen at -30 ºC and stored at -20 ºC until the corresponding analyses. After thawing, the thickness of the dorsal fat layer between the 12th and 13th rib was taken¹⁸.

The meat samples were partially thawed at 4 ºC (to avoid fluid loss). Subsequently, the visible adipose tissue was removed, mixed with a Black and Decker™ food processor (Model HC3061, New Britain, CT, USA), packed in bags (Whirl-Pak Bags, Nasco, Fort Atkinson, WI), and stored at -20 ºC until the final analyses. Se was quantified with an atomic absorption spectrophotometer (SpectrAA 220^®, New Britain, CT, USA), following the procedures of the manufacturers.

Data were analyzed using the statistical package SAS¹⁹ (version 9.2, SAS Institute Inc., Cary, NC, US). DMI, LW changes, and FC were analyzed with the Mixed procedure of SAS in a completely randomized design with measures repeated over time¹⁹. The model included fixed effects of treatment, week and the treatment×week interaction. The random effect of animal was nested in treatments and was taken as the repeated term. The statistical model is described below, after removing the double or triple interactions that were not significant:

Where:

Y_ikjl is the response variable, μ is the overall mean;

T_i is the fixed effect of treatment_i (i = 1, 2, ..., 5);

T_j is the fixed effect of time (j = 1, 2, ..., 4);

T×T_ij is the fixed effect of the interaction of i-th treatment_i and j-th time_j;

Lk_(i) is the random effect of the animal;

E_ijkl is the random effect of experimental error.

A model similar to the previous one was used to study the levels of albumin, α-, ß- and ∂-globulins. Se concentrations in muscles were analyzed with Proc Mixed in a completely randomized design with a classification criterion¹⁹. The results were declared significant where it was observed that P<0.05. When differences between treatments were detected, the means were compared with the Tukey procedure with α=0.05. The covariance structure that produced the lowest Akaike²⁰ criteria was that of composite symmetry in all the variables studied, except for Se levels in muscles, which better adapted to the autoregressive of order (1).

Table 2 shows the results obtained for DMI, DWG and FC of growing lambs supplemented with SS and SP for 56 d. The supplementation with Se did not influence (P>0.05) the DMI of the sheep. This may be because the addition of Se increases the digestibility of OM, NDF and N in the total tract and possibly facilitates the absorption of the mineral in the abomasum. However, it was not enough to increase the DMI.

The results obtained in this study agree with Alimohamady et al⁴, who observed an improvement in the digestibility of dietary components. Other studies reported different results. Domínguez-Vara et al²¹ observed no differences in DWG and feed conversion in Rambouillet lambs fed 0.30 mg of Se per kg^-1 of DM of SS compared to the control. The non-difference was attributed to the state of Se and its low availability in the diets.

In the present study, supplementation with 0.30 and 0.90 mg kg^-1 DM of SS or SP impacted DWG. This may be due to an improvement in feed digestibility. On the contrary, the superiority in feed conversion with 90SP, possibly, is explained because the Se from SP has been shown to have a higher bioavailability with respect to SS^9,21.

Table 2 shows the final LWs, the weights of the hot and cold carcasses, and the thickness of the dorsal fat layer of sheep fed SS and SP. There was no effect of the level and source of Se (P>0.05) on the aforementioned variables. The lack of effect is explained by the similarity in the DMI of the animals. As is known, the LW of animals depends on feed intake and in the present case, this intake was similar between treatments, although the FC was different.

Vignola et al⁶ indicated that supplementation with 0.3 and 0.9 mg of Se from SS or SP did not affect the Longissimus dorsi area, dorsal fat thickness and the weights of hot and cold carcasses of growing lambs. The loss of effect was attributed to the similar DMI among the different sources of the mineral.

As shown in Table 3, the effect of the treatments influenced (P≤ 0.05) the content of Se in the blood serum. As the Se in the diet increased, the concentration in blood serum increased and then it showed a decreasing return with 0.9 mg of Se kg^-1 DM.

According to Puls²², adequate levels of Se in blood serum of growing sheep are 0.08 to 0.50 mg L^-1, in order to balance internal homeostasis. In the present study, all treatments were within the indicated range, except for C and 30SS, which showed concentrations below 0.07 mg L^-1.

As shown in Table 3, supplementation with different levels of SS or SP influenced (P<0.05) the Se level in the Longissimus dorsi, with no apparent effects (P>0.05) in the Gluteus maximus and Musculus deltoideus. The Se in skeletal muscle increases as the diet is richer in the mineral. Based on the report by Puls²², the observed values of Se in the muscles are within the appropriate range for animals of the same characteristics as in the present study. The highest response to supplementation was 30SP. Perhaps due to the greater availability of the mineral to be incorporated into tissues, however, with the highest level, its response tends to decrease.

The results obtained in the present study agree with others previously published. Juniper et al⁹ indicated that 0.35 mg of SS or SP increased the mineral in the muscles, in a manner dependent on the diets of finishing bovines. The difference between sources was remarkable. The Se of SS produced 0.31 mg of Se, while that of SP yielded 0.46 mg in the Longissimus dorsi, which are similar to those observed in the present study.

The levels of albumin, α-, ß- and ∂-globulins in the blood serum of sheep are presented in Table 3. The levels and sources of Se only increased (P<0.05) albumin and α-globulins. On the contrary, they did not affect the levels of ß- and ∂-globulins. Serum albumin originates in hepatocytes, from where it passes into the bloodstream (which represents approximately 13 % of the total protein produced by the liver)²³.

In the present study, the increased presence of albumin in lambs that consumed 90SS and 30SP was not due to the increase in the synthetic capacity of hepatic albumin. Nor was it because the animals had a high capacity for synthesis. The difference is explained because Se, as an antioxidant, improves the activity of hepatocytes, so perhaps it improved overall protein production.

Based on the report by Kaneko et al²⁴, the albumin concentration obtained in the present study was approximately twice the minimum recommended levels, and in all cases, it exceeded the maximum indicated. The maximum level was reached with 30SP and subsequently, it tended to decrease. The values observed in the present study are consistent with those observed by De Paula Silva et al²⁵ with several sheep breeds created in tropical conditions.

Immunoglobulins act as membrane receptors on β lymphocytes and are used by the immune system to identify and neutralize viruses and bacteria²⁶. In the present study, the highest concentration of α-globulins was found in lambs that consumed 30SS and 90SP. This behavior was related to the antioxidant capacity of the mineral included in the diet.

In conclusion, the level of 0.90 mg of Sel-Plex™ improved DWG and FC. Selenite and SP increased serum Se in 30SS, 90SS and 30SP and it decreased with 0.90 mg per kg of SP. In the Longissimus dorsi, Se was improved in 30SP with respect to 90SP and C and was similar to 90SS and 30SS. The organic Se of 90SP improved the level of albumin and α-globulins.