General conditions

The studies were carried out in the state of Querétaro, México (20^o 42’ N; 100^o 01’ W). The experimental protocols and procedures were approved by the Institutional Subcommittee for Caring and Use of Animals for Experimentation (SICUAE), College of Veterinary Medicine and Animal Husbandry, National Autonomous University of Mexico, and according to the “International Guiding Principles for Biomedical Research Involving Animals” (http://www.cioms.ch/images/stories/CIOMS/IGP2012.pdf). Ram lambs were born between March 2 and 29 (Study 1) or between October 29 and November 14 (Study 2). Before weaning (70 d of age), lambs were creep fed a diet containing 16.9 % CP and 2.74 Mcal ME/kg DM, and ewes were fed a diet containing 13.04 % CP and 2.33 Mcal ME/kg DM. After weaning, lambs were offered a diet with 14.05 % CP and 2.65 Mcal ME/kg DM, and water ad libitum. One day before each lamb reached 2, 4, 8, 12, 16, 20 and 24 wk of age, body weight (BW) was recorded, scrotal circumference (SC) was measured as described previously¹⁵, and lambs were fitted with indwelling jugular catheters. At that time and at the end of samplings, lambs were administered vitamin B complex (thiamin 0.15 g, riboflavin 0.2 g, nicotinamide 0. 3 g, pyridoxine 0.2 g, cyanocobalamin 90 µg and D-panthenol 0.15 g), to prevent anemia, and a long -acting oxytetracycline solution (20 mg•kg^-1 BW) in a prophylactic manner, as the catheter remained in the vein for at least one day.

Study 1. Development of LH and testosterone profiles

Ram lambs were allotted in three homogeneous genotype groups, based on birth weight, type of birth and age of the mother: Blackbelly (n=4), Pelibuey (n=5) and Crossbred (n=8). At 2, 4, 8, 12, 16, 20 and 24 wk of age, blood samples were collected every 15 min, during 6 h (0700 to 1300 h), and centrifuged at 720 xg for 15 min at 4 ^oC; plasma was obtained and kept frozen at -20 ^oC until assayed. The profile variables evaluated for LH and T were mean and basal concentrations (ng•ml^-1), frequency (pulses•6h^-1), amplitude and height of the pulses (ng•ml^-1). Pulse detection was performed using computer algorithms¹⁶ (Pulsar software, modified for IBM-PC by JF Gitzen and VD Ramirez, Urbana, IL). G(n)-values (G(n): cut-off parameters to determine if one point value or a series of consecutive values belong to a pulse-¹⁶), for LH were G(1)=10.0, G(2)=2.6, G(3)=1.9, G(4)=1.5 and G(5)=1.1; for T were G(1)=3.8, G(2)=2.6, G(3)=1.9, G(4)=1.5 and G(5)=1.1. The standard deviation of the assays used in the algorithm was modeled as a quadratic function of the dose for each hormone.

Study 2. Development of LH and testosterone response to GnRH administration

Experimental plan, GnRH administration and blood sample collection. Ram lambs were allotted in three homogeneous genotype groups, based on birth weight, type of birth and age of the mother: Blackbelly (n=4), Pelibuey (n=4) and Crossbred (n=4). The acetate salt of the natural sequence of GnRH was used (Sigma Chemical Co., St Louis, MO.). At the same ages as in the study 1, animals were challenged in consecutive days with graded doses of GnRH: 3, 9 and 27 ng•kg^-1 BW, via i.v. (small, medium and large dose, respectively), starting with the small dose at each age. The small dose was established based on a previous study with adult rams, where it was calculated that the administration of 3 ng•kg^-1 generated LH pulses with characteristics similar to non-induced pulses¹⁷; therefore, it was initially considered as the physiological dose. Blood samples were collected at -15, 0, 15, 30, 60, 120 and 180 min from GnRH administration (0700 to 1015 h), and processed as explained for study 1.

Analysis of the hormonal response to GnRH administration. The variables of LH and T response to GnRH administration were the following: peak height (the highest hormone concentration subsequent to GnRH administration); area under the curve of response (AUC), which was calculated using the trapezoid rule from the initial concentration (average of the -15 and 0 min samples) until 3 h after GnRH administration; magnitude of the response (ΔLH and ΔT), calculated by subtracting the average initial concentration from the peak height, and using the initial concentration as a covariable for each hormone.

Hormone analysis

LH concentrations were determined by a double antibody RIA¹⁸, sensitivity was 95 pg•ml^-1, the intra - and inter- assay CV were 8.6 and 9.5 %, and 14.2 and 7 % for studies 1 and 2, respectively. Testosterone concentrations were determined by a solid phase RIA (Coat-A-Count®, Siemens Medical Solutions Diagnostic; Los Angeles CA, USA); sensitivity was 40 pg•ml^-1; the intra - and inter-assay CV were 4.9 and 7.1 %, and 5 and 5.9 % for studies 1 and 2, respectively.

Statistical analyses

Hormone data were analyzed as a completely randomized design, using the procedure MIXED of SAS (SAS Institute Inc. Cary NC, USA) for repeated measures, with the statement “repeated”, the option sub=animal (genotype) and the autoregressive (1) covariance structure within animal. Genotype, age (and dose in study 2) and the corresponding interactions were included in the model. The endocrine variables were transformed to Log_n (Y+1), so the assumptions of the analysis of variance were valid; after analysis, data were transformed back to actual values for tables and figures. Data of BW and SC were pooled across studies before the analysis, and each study was included in the model as a block. Differences were considered significant with P<0.05, and tended to be significant with P values between 0.05 and 0.1.

Development of LH profiles

The effect of genotype and genotype by age interaction were not significant for any of the variables of LH profiles, whilst the effect of age was significant for all variables (Table 1). The smallest mean concentration was observed on wk 2, an elevation was detected on wk 4 (P<0.05); then, there was a progressive increase, reaching the greatest concentration on wk 12 to fall on wk 16 (P<0.05), with an additional increase observed on wk 24 (P<0.05). Basal concentration did not change from 2 to 8 wk; then, increased at wk 12 (P<0.05), when reached a plateau until wk 20, with an additional increase on wk 24 (P<0.05). Pulse frequency was greatest on wk 8, but no differences were detected from 2 to 20 wk, whilst the least frequency was detected on wk 24 (P<0.05). Pulses of low amplitude were detected on wk 2; then, amplitude increased drastically on wk 4 (P<0.05) to reach the greatest value on wk 12, with a decreasing trend thereafter.

Development of testosterone profiles

The effect of genotype and genotype by age interaction were not significant for any of the variables of T profiles, whilst the effect of age was significant for all variables (Table 1). Mean T concentration increased constantly from wk 2 to 24 (P<0.05). Basal concentration was maintained low until wk 8; thereafter, a sustained increase was observed from wk 12 to 24 (P<0.05). Pulses of low magnitude were detected in some animals at 2 wk of age, but the average frequency was very small and was different from the rest of the ages. The highest frequency was observed on wk 12, but it was not different from wk 8, 16, 20 and 24. Pulse amplitude was small between wk 2 and 8; then, significant increases were observed on wk 16 and 20 (P<0.05), with no further changes.

Development of LH and testosterone responses to GnRH administration

The effects of the main factors and the corresponding interactions on LH and T response to the administration of GnRH are shown in Table 2.

LH secretion after GnRH administration. The triple interaction genotype by age by dose was not significant for any of the variables of LH response (Table 2). Genotype by age interaction was or tended to be significant for peak height of LH secretion (P=0.046) and for ΔLH (P=0.063; Tables 2 and 3). There were not differences between genotypes on wk 2, 4, 16 and 20, whereas on wk 8, 12 and 24, LH response was greatest in Pelibuey lambs. Pelibuey and crossbred lambs reached the highest response on wk 8 (P<0.05), which was maintained until wk 16, whereas Blackbelly lambs did not reach the highest response until wk 16 (P<0.05); after wk 16, peak height showed a decreasing trend for all the genotypes. Genotype by dose interaction was or tended to be significant for peak height (P=0.006), AUC (P=0.06) and ΔLH (P=0.034; Table 2). No differences between genotypes were detected for the small dose (Table 4; peak height and ΔLH), but for the medium and large doses, Pelibuey had the highest and Blackbelly the lowest LH response to GnRH, with crossbred lambs being intermediate. Dose by age interaction was significant for all the variables of LH response (Table 2); differences between doses were detected at every age in a dose-dependent manner (Table 5), except for AUC at wk 2. The effects of genotype, GnRH dose and age on the three variables of LH response were significant (Table 2). Pelibuey ram lambs always showed the greatest response, as compared with Blackbelly and crossbred lambs (Table 6). The response to GnRH administration was in a dose -dependent manner (Table 6). Regarding the effect of age, the greatest LH response was observed between 8 and 16 wk of age; thereafter, the response decreased (P<0.05) for the three variables evaluated (Figure 1).

Figure 1 Effect of age on peak height (panel A), area under the curve of response (panel B) and magnitude of the response (panel C) of LH and testosterone, after administration of GnRH 

Testosterone secretion after GnRH administration. The triple interaction of genotype by age by dose, as well as the interactions of genotype by age and genotype by dose were not significant for any of the variables of T response (Table 2). Dose by age interaction was significant for the three variables of T response (Table 2). No differences among doses (Table 7) were detected for peak height (2 and 4 wk) and ΔT (2 wk); AUC was greater for the large dose, as compared with the small dose (2 wk); AUC and ΔT were greater in the large dose, as compared with the other doses. From wk 8 onwards, T response to the small dose was lower than T response to the medium and large doses, with no difference between the last for all variables, except for AUC at wk 24, when the response to all doses was different from each other.

Within dose, T response increased as the lambs were older (Table 7); however, differences were not always significant. For the small dose, peak height did not change from 2 to 12 wk of age, but then it increased until wk 24; for the medium and large doses, peak height was not different between 2 and 4 wk, but from wk 8 it increased constantly until the end of the study. Regarding AUC, when the small and medium doses were used, there was no difference between 2 and 4 wk, then it increased between wk 8 and 20; the AUC to the large dose increased continuously from 2 to 24 wk. ΔT was no different between 2 and 4 wk, but from wk 8 onwards it increased until wk 24, for all doses.

The effect of genotype on the three variables of T response tended to be significant (Table 2), Blackbelly ram lambs always showed a lower response than crossbred lambs (Table 6), with Pelibuey being intermediate. The effects of GnRH dose and age were significant (Table 2); T response to the small dose was lower than the response to the medium and large doses (P<0.05; Table 6), with no difference between the larger doses (peak height and ΔT); AUC was different among the three doses administered (P<0.05). The response of T increased as the lambs were older for the three variables evaluated (Figure 1).

Body weight and scrotal circumference

The interaction of genotype by age on BW was significant (P=0.01; Figure 2 A ); from wk 12 onwards, Blackbelly lambs had usually the lightest and crossbred lambs had the heaviest BW (P<0.05). Regarding SC, the effect of the genotype and genotype by age interaction were not significant. The effect of age was significant (P<0.001), and SC augmented as the age increased (Figure 2 B ).

Figure 2 Mean ± SE of body weight (panel A) and scrotal circumference (panel B) of Blackbelly and Pelibuey ram lambs and their crosses, from 2 to 24 wk of age