Citrulline and muscle protein homeostasis in three different models of hypercatabolism.

Supplementation of enteral nutrition (EN) by specific amino acids (AAs) has been proposed to prevent muscle protein loss in intensive care unit (ICU) patients. Citrulline (Cit), which has been shown to stimulate muscle protein synthesis in other situations, may be of interest in this setting. Our aim was to assess the effect of Cit in three catabolic models relevant to critical illness: endotoxemia (LPS), traumatic brain injury (TBI), and TBI with infectious complications (TBI-Ec), which are characterized by different alterations in protein homeostasis. Fifty-eight male Sprague-Dawley rats (200-220 g) were randomized to receive a standard diet ad libitum (CON, n = 9) or to undergo catabolic injuries on day 0 (D0, n = 49), and EN (Sondalis HP energy® 290 kcal/kg/d) from day 1 (D1) combined with Cit (2 g/kg/d) or isonitrogenous non-essential AAs (NEAAs) until day 3 (D3). Endotoxemia was induced by IP injection of LPS from E. coli (3 mg/kg), TBI by hydraulic percussion, and infectious complications (TBI-Ec) by administration of luminescent E. coli on D1. Nitrogen balance (ΔN) and 3-methylhistidine (3-MHis) were measured daily. Muscle protein synthesis (MPS, measured by the SUnSET method) and mTORC1 activation (S6K-1 and 4E-BP1 phosphorylation) were measured on D3 2 h after the arrest of enteral nutrition in soleus, extensor digitorum longus (EDL), gastrocnemius and tibialis muscles. ΔN was lower (p < 0.001) in all three models of injury compared with basal and CON from D1 to D3, and more negative in the LPS-CIT (p < 0.05) than in the LPS group. The 3-MHis/creatinine ratio was significantly increased on D1 in all groups compared with CON, and on D2 only in the LPS and TBI groups (p < 0.0001, LPS and TBI vs. CON). MPS was similar in all groups in soleus and tibialis but significantly higher in EDL in LPS-CIT [LPS-CIT: 4.5 ± 1.7 (mean ± SD) vs. CON: 2.3 ± 1.2; and vs. LPS-NEAA: 3.1 ± 2.3] and in gastrocnemius (LPS-CIT vs. CON; p = 0.05). S6K-1 phosphorylation in the EDL was also higher in LPS-CIT vs. CON (LPS-CIT: 0.94 ± 0.51 CON: 0.42 ± 0.28), but not in gastrocnemius. IL-6 plasma level was significantly higher in all the catabolic groups vs. CON (p < 0.005) with no difference between treatments (Cit or NEAAs). In conclusion, the TBI model showed only a rise in muscle proteolysis, whereas the LPS model displayed a rise in both protein synthesis and proteolysis. Secondly, our results show that the Cit effect varies according to the type of injury and to the muscle under study. The stimulation of MPS rate and the mTOR pathway in LPS-treated rats contrasts with degraded ΔN, suggesting that the Cit effect on protein metabolism in critically ill rats is limited at the whole-body level.