Insulin and glucosamine infusions increase O-linked N-acetyl-glucosamine in skeletal muscle proteins in vivo

O-linked N-acetylglucosamine (O-GlcNAc) is an abundant posttranslational modification of serine/threonine residues of nuclear and cytoplasmic proteins. We determined whether insulin or coinfusion of glucosamine (GlcN) with insulin alters O-GlcNAc of skeletal muscle proteins. Three groups of conscious fasted rats received 6-hour infusions of either saline (BAS), insulin 18 mU/kg.min and saline (INS), or insulin and GlcN 30 mu mol/kg.min (GLCN) during maintenance of normoglycemia. At 6 hours, the concentrations of muscle UDP-GlcNAc, UDP-N-acetylgalactosamine (UDP-GalNAc), UDP-glucose (UDP-Glc), UDP-galactose (UDP-Gal), glycogen, and N- and O-linked GlcNAc (galactosyltransferase labeling followed by beta elimination) were measured in freeze-clamped abdominis muscle. Insulin increased whole-body glucose uptake from 49 +/- 5 to 239 +/- 8 mu mol/kg.min (P <.001) and glycogen in abdominis muscle from 138 +/- 11 to 370 +/- 26 mmol/kg dry weight (P <.001). Insulin increased the amount of cytosolic N- and O-linked GlcNAc by 56% from 362 +/- 30 to 564 +/- 45 dpm/mu g protein.100 min (P <.02), and O-GlcNAc from 221 +/- 16 to 339 +/- 27 dpm/mu g.100 min (P <.02). Glycogen content was positively correlated with the amount of total (r =.90, P <.005) and O-linked GlcNAc in insulin-infused animals. Coinfusion of GlcN with insulin increased muscle UDP-GlcNAc about fourfold (100 +/- 6 nmol/g) compared with insulin (27 +/- 1, P <.001) or saline (25 +/- 1, P <.001) infusion, GlcN also decreased glucose uptake over 6 hours by 30% to 168 +/- 8 mu mol/kg.min (P <.001 for GLCN v INS) and muscle glycogen to 292 +/- 24 mmol/kg dry weight (P <.05 for GLCN v INS). Both total (635 +/- 60 dpm/mu g.100 min, P <.002) and O-linked GlcNAc (375 +/- 36 dpm/mu g.100 min, P <.002) in the cytosol were significantly higher in GLCN rats (635 +/- 60 dpm/mu g) versus BAS rats (P <.002). As in INS rats, muscle glycogen and O-GlcNAc were positively correlated in GLCN rats (r =.54, P <.05). Variation in total and O-linked GlcNAc in GLCN rats was due both to GlcN (P <.02) and to variation in the glycogen content (P <.005). Copyright (C) 1998 by W.B. Saunders Company.