Seasonal Patterns in Growth Hormone, Insulin and Insulin-like Growth Factor-1 in Female Muskoxen

Over a year, mean serum growrh hormone concentrations ranged from 0.5 to 3 ng/ml in tame muskoxen with the lowest values and fewest pulses between October and January. Serum insulin ranged from 0.25 to 0.92 ng/ml, with minima in summer. There were marked seasonal changes in serum insulin-like growth factor-1 in both tame and wild muskoxen but the late summer peak was higher and more disrinct in the tame animals with levels reaching 160-250 ng/ml in September.

Large annual changes in weight and body composition of female muskoxen (Ovibos moschatus) are partly due to lactation and decreased nutrient availability in winter (Thing et al, 1987;White et al, 1989). However, tame well-fed non-breeding muskoxen also lose weight in late winter and gain weight in autumn and early winter (Groves, 1992;Adamczewski et al., 1994), suggesting an effect of photoperiod, independent of nutrition. Growth processes in ruminants are regulated by hormones such as growth hormone (GH), insulin-like growth factor-1 (IGF-1) and insulin (Pearson & Dutson, 1991). To investigate the seasonal regulation of weight change in muskoxen, we measured seasonal patterns in GH, IGF-1, and insulin in muskox cows offered a diet of constant medium quality, and IGF-1 in hunter-killed wild muskoxen. A summary of preliminary IGF-1 data was presented previously (Adamczewski etal, 1992).
The tame muskoxen were part of a research herd kept near Saskatoon, Saskatchewan, Canada (52°N), and were offered brome-alfalfa hay [9-12% crude protein (CP)] and a pelleted supplement (13-14% CP; 20% of diet; Adamczewski et al, 1994) during each blood-sampling period and for at least 12 days previously, in 1990 and 1991. We studied 3 cows which bred annually and 2 non-breeding hysterectomized cows with intact ovaries, all > 4 years old. Jugular catheters were installed 36-48 hours prior to sampling, using local anaesthetic and mild sedation (Tedesco, 1996). During sampling, the muskoxen were kept loosely tied in shaded outdoor stalls with access to hay and water or snow. Blood samples of 5-10 cc were taken every 15 min for 24 h  under ambient light augmented by dim red lights at night. Cortisol analysis of samples taken periodically showed that the muskoxen were unstressed at these times (Tedesco, 1996). Wild muskox cows > 3 years old were shot by Inuit hunters near Cambridge Bay during 15 hunts in April, May, 132 July, August, September and November from 1989 to 1993 (Adamczewski, 1995). Blood was collected just after death from a jugular vein. Serum GH, IGF-1 and insulin were assayed by radioimmunoassays described by Van Kessel (1992). The GH assay was developed for ovine serum and validated for muskox serum by showing parallelism to the standard curve in a serial dilution of a concentrated sample. GH was assayed in representative 24-h sequences from 5 muskox cows, and GH pulsatility was analysed using the pulse-detection algorithm PC-PULSAR (J. Gitzen and V. Ramirez, Univ. Illinois), following Merriam & Wachter (1982). The IGF-1 assay was developed and tested for sera from various species and validated for muskox serum as above. The insulin assay was developed for bovine and ovine serum, and similarly validated. IGF-1 and insulin were assayed in samples pooled from 24-h profiles from tame muskoxen, and IGF-1 was assayed in individual samples from wild muskoxen. Monthly values from tame muskoxen were compared for all three hormones by split-plot analysis-of-variance (ANOVA) and values from wild muskoxen by one-way ANOVA, using season as the main effect.
Serum GH was pulsatile in the tame muskoxen ( Fig. 1), with highly individual profiles. Mean GH ranged from 0.5 to 3 ng/ml and GH pulse frequency ranged from 6 to 18 pulses/24 h. Both the frequency of pulses and mean GH were lowest between October and January, with significant (P=0.04) or near-significant (P=0.06) monthly variation, respectively (Table 1). Serum insulin means ranged from 0.25 to 0.92 ng/ml, with significant monthly variation (P<0.001) and the lowest concentrations in June and July. Individual variation in insulin was high and a clear seasonal progression was not apparent. There was a strong seasonal progression (P<0.001) in serum IGF-1 in the tame muskox cows, with mean concentrations increasing during summer, a clear peak in September (160-250 ng/ml) and lowest concentrations in November (60-120 ng/ml) ( Table 2). There was a strong (P<0.001) but different seasonal pattern in serum IGF-1 in wild muskoxen, with a plateau of high values (80-120 ng/ml) in July, August and September, and the lowest values (30-50 ng/ml) in November. Serum IGF-1 was lower throughout the year in wild muskox cows than in tame ones.
These endocrine patterns are consistent with the tendency of muskox cows to gain weight in autumn and early winter, and with a large seasonal variation in intake and diet quality in free-ranging muskoxen. Although the insulin values varied widely in tame muskoxen, they were lowest in early summer, a period when fat deposition is limited (White et al, 1989;Adamczewski, 1995). The autumn peak in IGF-1 in tame muskox cows, despite a diet of Table 2. Seasonal changes in serum concentrations (ng/ml) of insulin-like growth factor-1 (IGF-1) in tame muskox cows offered a constant-quality diet in Saskatoon, Saskatchewan, 1990Saskatchewan, -1991 (repeated samples from the same animals) and from wild muskox cows on Victoria Island, NT (single samples from individual animals).

Means with standard errors in parenrheses.
Month Tame  constant quality, is also consistent with lean tissue growth at this time of year (Adamczewski, 1995) and may account for the high IGF-1 in August and September in wild muskox cows at a time of decreasing forage quality. The proximate cause of this peak is unknown but may include seasonal changes in melatonin secretion (Tedesco, 1996) or increased estrogen due to reproductive cycling (cf. Sauerwein et al., 1992). These results are consistent with the strong seasonality in growth and intake of temperate and arctic cervids (Suttie et al., 1993) but suggest that the effects of season on growth are subtly different in muskoxen and caribou, as has been shown for the effects of season on reproduction in muskoxen (Tedesco, 1996). The late IGF-1 peak, in particular, appears to contrast with the earlier peak of IGF-1 in red deer (Suttie et al., 1992) and reindeer (Suttie et al., 1993), where the IGF-1 rise appears to anticipate lean growth. Female muskoxen appear predisposed to an apparently simultaneous surge in growth of lean and adipose tissue toward the end of summer, well past the period of peak forage quality (White et al, 1989;Adamczewski 1995). This timing underscores the association of weight gain in female muskoxen with reproductive events, as suggested in reindeer by Tyler (1987). The elevated IGF-1 and insulin in the autumn also suggest that serum concentrations of these hormones, intimately involved in ovarian physiology (Spicer & Echternkamp, 1995), influence the highly variable productivity of female muskoxen.