Restricted nutrient intake does not alter serum-mediated measures of implant response in cell culture

Background During nutritional stress, reduced intake may reduce the efficacy of anabolic implants. This study was conducted to evaluate basic cellular responses to a growth promotant implant at two intake levels. Methods Sixteen crossbred steers (293 ± 19.3 kg) were used to evaluate the impact of anabolic implants in either an adequate or a restricted nutritional state. Steers were trained to individual Calan gates, and then randomly assigned to 1 of 4 treatments in a 2 × 2 factorial arrangement. Treatments consisted of: presence or absence of an anabolic growth implant (Revalor-XS, 200 mg TBA and 40 mg estradiol; IMPLANT or CONTROL) and a moderate energy, pelleted, starting cattle diet fed at either 2.0 × or 1.0 × maintenance energy (NEM) requirements (HIGH or LOW). Serum (d 0, 14, and 28) was used for application to bovine muscle satellite cells. After treatment with the serum (20% of total media) from the trial cattle, the satellite cells were incubated for 72 h. Protein abundance of myosin heavy chain (MHC), phosphorylated extracellular signal-related kinase (phospho-ERK), and phosphorylated mammalian target of rapamycin (phospho-mTOR) were analyzed to determine the effects of implant, intake, and their interaction (applied via the serum). Results Intake had no effect on MHC (P = 0.85) but IMPLANT increased (P < 0.01) MHC abundance vs. CONTROL. Implant status, intake status, and the interaction had no effect on the abundance of phospho-ERK (P ≥ 0.23). Implanting increased phospho-mTOR (P < 0.01) but there was no effect (P ≥ 0.51) of intake or intake × implant. Conclusions The nearly complete lack of interaction between implant and nutritional status indicates that the signaling molecules measured herein respond to implants and nutritional status independently. Furthermore, results suggest that the muscle hypertrophic effects of anabolic implants may not be mediated by circulating IGF-1.


Background
Anabolic implants increase average daily gains 15 to 20%, and improve feed efficiency by 6 to 14% in feedlot cattle [1]. However, when animals first arrive in a feedlot many are in a state of stress and have not yet become accustomed to the environment, feeding routine, other animals, and other factors, especially if they have traveled long distances or have recently been weaned. These cattle are considered at high risk to develop bovine respiratory disease (BRD). Many of these high-risk cattle have reduced intakes for up to three weeks after arrival at the feedlot [2].
Stress and infection can lead to a hypermetabolic state, characterized by weight loss, increased resting metabolic rate, and muscle protein catabolism [3]. In cattle infected with BRD, decreased feed intake, weight loss, and decreased body condition are often observed [3]. Although ruminants may have a greater capacity to store digesta because of their larger digestive tracts, this weight loss and decrease in body condition may be comparable to the hypermetabolic state that develops in non-ruminants [3]. Munson et al. [4] reported that delaying implant administration in high-risk calves did not alter feedlot performance or the risk of respiratory disease, suggesting that stress brought on by the transition to the feedlot did not diminish the anabolic signals from implanting early in the finishing phase.
Anabolic implants impact growth rate at least in part by stimulating muscle hypertrophy [5]. Androgenic and estrogenic steroids increase in vitro bovine satellite cell proliferation, and these effects are mediated in part by the extracellular signal-regulated kinase (ERK) and phosphatidylinositide 3-kinase/AKT/mammalian target of rapamycin (mTOR) signaling pathways [6]. Activation of these pathways, however, is dependent on adequate nutritional status, which is sensed through a complex array of signals [7,8]. Therefore, animals in a depleted nutritional state, accompanied by alterations in hormonal signals, may not have the expected muscle growth response to anabolic implants.
The objectives of this study were to examine the effects of anabolic implants on measures of nutrient balance, metabolic status, and growth factors in animals consuming nutrients either adequate or inadequate to support growth. Moreover, effects of serum-derived signals on in vitro proliferation of bovine muscle satellite cells were also assessed.

Animals
All procedures conducted in the present study were approved by the Kansas State University Institutional Animal Care and Use Committee (Protocols 2745.0 and 2745.1).
Sixteen crossbred steers (293 ± 19.3 kg) were purchased locally and transported to Kansas State University. They were weighed, tagged, and vaccinated with a 5-way respiratory viral vaccine and a 7-way clostridial vaccine. The steers were trained to individual Calan® gates over a period of 4 wk.
After training was complete, the steers were randomly assigned to 1 of 4 treatments in a 2 × 2 factorial arrangement. Main effect treatments were as follows: administration or absence of an anabolic implant (Revalor-XS, 200 mg TBA and 40 mg estradiol 17-β [Merck Animal Health, Whitehouse Station, NJ]; IMPLANT or CONTROL, respectively) and either 2.0 × or 1.0 × maintenance energy (NE M ) requirements of a moderate energy, pelleted, starting cattle diet (HIGH or LOW, respectively; Table 1). Energy content of the diet was calculated from ingredient composition and tabular estimates of individual ingredient energy values [9].
Steers were weighed on d 0, 14, and 28 of the trial. At each of these time points, maintenance energy requirements were calculated for each steer [10], and the diet was fed based these calculations and the treatment assignment for each animal.

Biochemical assays
All assays were conducted in duplicate. Satellite cell isolation was conducted as described by Johnson et al. [5].
Using sterile techniques, 500 g of the semimembranosus muscle was dissected from 2 steers which were not part of the present study (18 mo of age). After removal of connective tissue, the muscle was passed through a sterile meat grinder. The ground muscle was incubated with 0.1% pronase in Earl's Balanced Salt Solution (EBSS) for 1 h at 37°C with frequent mixing. After incubation the mixture was centrifuged at 1,500 × g for 5 min, the pellet was suspended in phosphate buffered saline (PBS; Ca-and Mg-free PBS; Life Technologies, Grand Island, NY), and the suspension was centrifuged at 500 × g for 10 min. The supernatant was centrifuged at 1,500 × g for 10 min to pellet the mononucleated cells. The PBS wash and differential centrifugation were repeated 2 more times. Cell viability was over 90%. The resulting mononucleated cell preparation was suspended, pooled (density 3 × 10 4 cells/mL) in cold DMEM containing 10% fetal bovine serum (FBS) and 10% dimethylsulfoxide and frozen in liquid nitrogen. Cells were counted using a hemocytometer and 0.4% Trypan Blue solution.
Blood samples (serum and plasma) were collected from each animal at d 0, 14, and 28 of the trial and used in determining levels of total IGF-1, and additional serum samples were collected from each animal for application to satellite cells [11][12][13] which had been previously isolated. Total IGF-1 concentration was determined in plasma samples using a commercial ELISA kit (Octeia IGF-1, catalog #AC27F1, Immunodiagnostic Systems Inc., Scottsdale, AZ). The kit includes reagents for pretreatment of samples to release IGF-1 from binding proteins [14]. Frozen satellite cells were re-suspended in DMEM supplemented with 5% FBS and 1% penicillinstreptomycin mixture (containing 10,000 IU/mL penicillin and 10,000 μg/mL streptomycin). Cells were passaged once, then at 70% confluence, cells were incubated in DMEM for 72 h with serum (20% of total media) taken from individual animals on d 0, 14, or 28 and with an additional control containing 20% FBS and no serum from trial cattle. Halting incubation at 72 h prevents overconfluency of the cells, which can lead to excessive differentiation; since these are skeletal muscle cells and would contract or twitch after differentiation, excessive differentiation would lead the cell layers to come off the cell culture plate. All media contained 1% penicillin-streptomycin mixture, but FBS was not added to wells containing treatment serum. Protein abundance of myosin heavy chain (MHC; d 0, 14, and 28), phosphorylated ERK (phospho-ERK; d 0 and 28), and phosphorylated mTOR (phospho-mTOR; d 0 and 28) was analyzed in differentiated satellite cells to determine the effects of implant, intake, and their interaction (applied via the serum). MHC was used as a marker of myotube formation, and phospho-ERK and phospho-mTOR are regulators of cell proliferation.

Statistical analysis
Animal was the experimental unit for all measures. Data were analyzed using the MIXED procedure of SAS® (v. 9.3; SAS Institute, Cary, NC); IGF-1 and MHC were analyzed using repeated measures and d 0 values were included in the model as a covariate; phospho-ERK and phospho-mTOR were analyzed as a completely randomized design but included d 0 values as a covariate. The independent variables included the main and interactive effects of implant status and intake level. Effects were considered significant with a protected F-test with α < 0.05; effects were considered a trend if P < 0.10.

Results and discussion
Level of feeding achieved different growth rates (0.76 vs. 0.02 kg/d for 2.0 × vs. 1.0 × maintenance intake, respectively; P < 0.01); implant status did not affect 28-d growth rates (0.36 vs. 0.41 kg/d for implanted vs. control, respectively; P = 0.38). There was no interaction between level of intake and implant status for growth rate (P = 0.38).
The IGF-1 concentrations in implanted steers were not different from those in control cattle in this study. Bryant et al. [18,19] reported no effects of implants on IGF-1 in feedlot heifers until d 42 after implant administration; however, Johnson et al. [18,19] and Reinhardt [20] reported that circulating IGF-1increased in implanted steers by d 21 and d 28, respectively, compared to non-implanted steers. Anderson et al. [21] reported that exogenous estradiol supplementation in post-menopausal women increased circulating growth hormone 2.1-fold but actually decreased circulating IGF-1 by 27%.
When serum was applied to the cultured satellite cells from the semimembranosus muscle of non-study steers, level of intake had no effect (P = 0.85; Figure 2) on MHC abundance but implant increased MHC abundance (P < 0.01), with no interaction observed between intake level and implant status (P = 0.23). In this study, MHC was used as a marker of myotube formation, reflecting satellite cell proliferation and differentation. Baxa et al. [22] found that implanting increased MHC type IIa mRNA in vivo. In a smaller study, Baxa [23] found that implanting with a combined estradiol-17β (E 2 )/trenbolone acetate (TBA) led to numerical increases in MHC type I and IIa mRNA. Chung et al. [24] found no effect of E 2 /TBA implants on MHC type I, IIa or IIx mRNA, but did find that MHC IIa increased over time which was also seen in the present study (effect of day P < 0.01). Implanting with E 2 /TBA increases muscle fiber diameter [25,26], due to an initial increase in DNA transcription followed by an increase in nuclei within the muscle fiber which support hypertrophy, but ultimately due to an increase in total protein produced by the cell [27], of which MHC is a key constituent.
Extracellular signal-regulated kinase plays a role in regulating insulin-and potentially IGF-I-mediated glucose uptake by skeletal muscle cells [28], and certain fatty acids reduce muscle cell proliferation via ERK [29]. Implant status did not affect the phospho-ERK abundance (P = 0.46; Figure 3). Intake had no effect (P = 0.65) on phospho-ERK, and there was no interaction between the two main effects (P = 0.41). Because ERK is phosphorylated (activated) by IGF-I signaling [6,30], the lack of phospho-ERK response to implant in the present study is understandable, given the lack of effect on circulating IGF-1. However, intake level also failed to influence phospho-ERK despite the fact that HIGH increased serum IGF-1 compared to LOW.
The protein kinase mTOR plays an essential part in transcription, translation, muscle cell size, and protein stability [31,32]. To help regulate these processes, mTOR integrates information about nutrient status, energy status, and many growth factors and may be a central component in the regulation of cellular processes that determine cell growth and proliferation, and even ultimate overall animal size [33]. Phosphorylation of mTOR can occur at both Ser 2448 [34] and Ser 2481 [35]. We investigated Ser 2481 phosphorylation of mTOR. This site is responsive to nutrient supply and Ser 2481 phosphorylation status is directly correlated with activity of mTOR complex 1, which is responsible for promoting protein synthesis through activation of S6 kinase 1 and eukaryotic translation initiation factor 4E-binding protein 1 [35]. There was no interaction between intake and implant status (P = 0.88; Figure 4) for phospho-mTOR, and intake had no effect (P = 0.51); however, serum from implanted steers (P < 0.01) stimulated greater phospho-mTOR levels vs. serum from CONTROL steers. Phosphorylated mTOR normally increases when nutrition is adequate for growth and decreases when the nutrient environment is not favorable for growth [31,36]. Although we did not see this response, it is important to recognize that our experimental design only partially subjected satellite cells to the milleu that muscle tissue of LOW animals would have seen. Our protocol provided cells with essentially unlimited nutrient supply via the DMEM that comprised 80% of the media, thereby negating effects of intake level on nutrient supply to the satellite cells. The lack of intake effect on phospho-mTOR, therefore, suggests only that endocrine signals (or compounds not supplied in DMEM) that differed by intake failed to alter satellite cell mTOR signaling. Yang et al. [32] reported that mTOR is highly sensitive to upregulation by IGF-1 and that the stress caused by low nutrient availability can inhibit mTOR signaling. In the present study, although we saw no effect of implant on IGF-1, we did measure a strong response of implant on phospho-mTOR. We saw a significant effect of increased intake level on IGF-1 but no corresponding effect of high nutrient intake stimulating phospho-mTOR. Johnson et al. [18,19] reported increased mitogenic activity of serum from implanted vs. control steers when applied directly to satellite cell cultures, but that study also reported increased levels of circulating IGF-I in the implanted steers. Results from the present study suggest that elevated IGF-I is not the sole serum-mediated factor which increases signalling for increased mitogenic activity in vitro, given that serum from nutrient abundant and nutrient restricted calves, which differed greatly in IGF-I concentration, did not alter MHC or phospho-mTOR. The phospho-mTOR and MHC response to implant, in the absence of a circulating IGF-I response, could be explained by a number of additional, nonsomatomedin-mediated pathways associated with growth regulation. Binding proteins specific for growth hormone and some, although not all, for IGF-I increase in plasma in response to elevated nutrient availability independent of treatment with exogenous growth hormone [37], and the same study indicated that elevated nutrient status actually reduced growth hormone in plasma. Concentration of IGF-I mRNA is elevated in longissimus muscle cells from animals previously administered exogenous E 2 [38,39], and muscle mRNA concentrations for the adipogenic regulators PPARγ and stearoyl CoA desaturase are reduced after administration of exogenous TBA [24]. In addition, both E 2 and TBA, which were used in combination in the present study, stimulate protein synthesis and reduce protein degradation in a dose-dependent manner when directly applied to fused satellite cell cultures [40,41]. This action is at least partially modulated through stimulation of muscle IGF-I mRNA by E 2 and TBA [42]. Treatment of proliferating bovine satellite cells cultures with either E 2 or TBA resulted in increases in IGF-1 mRNA as well as increased 3 H-thymidine incorporation rate [42]. Direct treatment of fused bovine satellite cell cultures with E 2 increases protein synthesis and decreases protein degradation rates, and blocking of estradiol binding suppresses estradiol-induced modifications in protein synthesis and degradation [40]. Testosterone also increases satellite cell activation in rats [43]. These documented effects indicate a direct mechanism for both estradiol and TBA on satellite cell proliferation and protein synthesis.
Implanting with a combination of E 2 and TBA increases circulating estradiol and TBA, but only exogenous E 2 increases circulating growth hormone [44][45][46]. Circulating estradiol is correlated to circulating growth hormone in humans [47], and exogenous E 2 supplementation increased growth hormone secretion in postmenopausal women [48] and ovariectamized ewes [49]. Sotiropoulos et al. [50] reported that growth hormone stimulates myoblast fusion with myotubes. Ge et al. [51] reported that growth hormone did not stimulate myoblast fusion or proliferation, but growth hormone did increase protein synthesis.
Hepatocyte growth factor increases satellite cell activation [52][53][54] and fibroblast growth factor enhances proliferation of satellite cells in culture [55,56]. Although hepatocyte growth factor [57][58][59] and fibroblast growth factor [60,61] are elevated in vitro in response to treatment with E 2 , liver and muscle mRNA for hepatocyte growth factor are not affected in vivo after implanting cattle with E 2 and TBA [62].
The satellite cell results, where serum from implanted cattle activated mTOR signaling, make it tempting to speculate that mTOR is a more critical mediator of muscle hypertrophy in response to implants than ERK. Certainly, mTOR has been shown to be a key regulator of protein synthesis [8] and is activated by testosterone treatment [63]. The primary upstream regulators of mTOR are AMP-activated protein kinase (AMPK), AKT, and ERK. Blockade of either the AKT and ERK signalling pathways ablated the mitogenic response to trenbalone acetate in bovine satellite cells, suggesting that at least basal activation of both of these pathways is required [6]. Conversely, activation of AMPK, which is triggered by nutrient deficiency, impairs mTOR phosphorylation , and this pathway has been implicated in loss of muscle mass in nutrient-restricted cows [36]. Given that implant status did not enhance phospho-ERK abundance in this study, we propose that basal ERK signaling was likely necessary to permit activation of mTOR, but that activation of the AKT pathway was probably more directly involved in stimulating mTOR signaling. This is also consistent with the unique roles of ERK and mTOR signaling pathways. Although ERK does contribute to mTOR activation, it is probably more important as a regulator of cell proliferation [64]. Proliferation of satellite cells is important for maintainance of future hypertrophic potential in vivo, but myotube formation is driven by cell differentation and activation of protein synthesis, which is more closely related to the function of mTOR.

Conclusions
No interactions were found between nutrient restriction and implant status in muscle satellite cell culture treated with serum from calves either implanted or not implanted and fed at either 1× or 2 × maintenance. This suggests that the effects of exogenous anabolic agents and nutrient status on growth regulation, at least insofar as it is regulated by the molecules measured in this study, are independent of one another. Serum from implanted cattle promoted increased abundance of the active phospho-mTOR regulatory protein as well as myotube formation, measured by MHC abundance, in bovine satellite cells. These finding suggest the presence of circulating factors in implanted cattle that influence cell signals and promote hypertrophy of muscle tissue.