Rs, covering the full range of BMI and M-values. As with western blot we found that there was a great deal of inter-individual variability in the basal level of activity. There was no significant correlation between either basal activity or post-insulin p42/p44 MAPK activity levels, and M-value or BMI (Figure 5). However there was an inverse correlation between fold-induction of p42/44 MAPK activity by insulin and body mass index (r = 0.73; p = 0.0009) (Figure 5A) and 15857111 a significant correlation between p42/44 MAPK activity in response to insulin and M value (r = 0.52; p = 0.04) (Figure 5B). Thus, whether 4EGI-1 biological activity measured against the degree of obesity or IR, the data indicates a close relationship between defective response to insulin of p42/44 MAPK activity in muscle and the clinical measures of pre-diabetes. This suggests that abnormal p42/p44 MAPK response to insulin in skeletal muscle is a better marker of whole body insulin resistance than the response of the PI3K-PKB pathway, at least in obese non-diabetic individuals. FOXO, GSK3 and ribosomal S6. There were no correlations between the basal or insulin-induced levels of phosphorylation of FOXO, GSK3 and ribosomal S6 protein with either BMI or M value (data not shown).Phosphorylation statusPKB. The induction of PKB phosphorylation by insulin was apparent in most volunteers (Figure 3 A and B). There was a tendency for the degree of insulin-induced phosphorylation of 
PKB to reduce with increasing BMI (r = 2.38; p = 0.09) (C) and to increase with increasing M value (r = 0.4; p = 0.08) (D) but these failed to reach significance. In contrast to the analysis of p42/p44 MAPK, direct assay of PKB activity rather than western blotting of phosphorylation failed to improve the correlation between PKB activity and insulin sensitivity (data not shown). 
p42/44 MAPK. There were no significant correlations between basal p42/44 MAPK phosphorylation and either BMI or M value (Figure 4). There was a tendency for p42/44 MAPK phosphorylation following insulin exposure to correlate with BMI (Spearman r = 0.4; p = 0.07) (C) or with M value (Spearman r = 0.59; p = 0.08) (D) but these both failed to reach significance.Figure 2. Relationship of IRS1 expression with body mass index or M value. Relative IRS1 protein expression according to body mass index (A) or to M value (B) and fold increase in IRS1 expression according to body mass index (r = 20.36; p = 0.10) (C) or to M value (r = 0.27; p = 0.23) (D). doi:10.1371/journal.pone.0056928.gSkeletal Muscle Signalling Defects in ObesityFigure 3. Relationship of PKB phosphorylation with body mass index or M value. Relative PKB phosphorylation according to body mass index (A) or to M value (B) and fold increase in PKB phosphorylation by insulin according to body mass index (r = 2.38; p = 0.09) (C) or to M value (r = 0.4; p = 0.08) (D). doi:10.1371/journal.pone.0056928.gSummary of signalling analysis (Table 1)The study group was stratified incrementally according to their whole body insulin resistance, determined by the M value, and the Peptide M site responses of each individual signalling protein to insulin were ranked and the four individuals with the greatest (Green numbers, ranking 1 to 4)) or least (Red numbers, ranking 1 to 4) responses for each protein were noted. Representative blots are shown (Figure 6). The responses of interest were insulin-induced changes in IRS1 protein expression, in PKB or p42/p44 MAP kinase phosphorylation or in p42/p44 MAP kinase activity. We observed a.Rs, covering the full range of BMI and M-values. As with western blot we found that there was a great deal of inter-individual variability in the basal level of activity. There was no significant correlation between either basal activity or post-insulin p42/p44 MAPK activity levels, and M-value or BMI (Figure 5). However there was an inverse correlation between fold-induction of p42/44 MAPK activity by insulin and body mass index (r = 0.73; p = 0.0009) (Figure 5A) and 15857111 a significant correlation between p42/44 MAPK activity in response to insulin and M value (r = 0.52; p = 0.04) (Figure 5B). Thus, whether measured against the degree of obesity or IR, the data indicates a close relationship between defective response to insulin of p42/44 MAPK activity in muscle and the clinical measures of pre-diabetes. This suggests that abnormal p42/p44 MAPK response to insulin in skeletal muscle is a better marker of whole body insulin resistance than the response of the PI3K-PKB pathway, at least in obese non-diabetic individuals. FOXO, GSK3 and ribosomal S6. There were no correlations between the basal or insulin-induced levels of phosphorylation of FOXO, GSK3 and ribosomal S6 protein with either BMI or M value (data not shown).Phosphorylation statusPKB. The induction of PKB phosphorylation by insulin was apparent in most volunteers (Figure 3 A and B). There was a tendency for the degree of insulin-induced phosphorylation of PKB to reduce with increasing BMI (r = 2.38; p = 0.09) (C) and to increase with increasing M value (r = 0.4; p = 0.08) (D) but these failed to reach significance. In contrast to the analysis of p42/p44 MAPK, direct assay of PKB activity rather than western blotting of phosphorylation failed to improve the correlation between PKB activity and insulin sensitivity (data not shown). p42/44 MAPK. There were no significant correlations between basal p42/44 MAPK phosphorylation and either BMI or M value (Figure 4). There was a tendency for p42/44 MAPK phosphorylation following insulin exposure to correlate with BMI (Spearman r = 0.4; p = 0.07) (C) or with M value (Spearman r = 0.59; p = 0.08) (D) but these both failed to reach significance.Figure 2. Relationship of IRS1 expression with body mass index or M value. Relative IRS1 protein expression according to body mass index (A) or to M value (B) and fold increase in IRS1 expression according to body mass index (r = 20.36; p = 0.10) (C) or to M value (r = 0.27; p = 0.23) (D). doi:10.1371/journal.pone.0056928.gSkeletal Muscle Signalling Defects in ObesityFigure 3. Relationship of PKB phosphorylation with body mass index or M value. Relative PKB phosphorylation according to body mass index (A) or to M value (B) and fold increase in PKB phosphorylation by insulin according to body mass index (r = 2.38; p = 0.09) (C) or to M value (r = 0.4; p = 0.08) (D). doi:10.1371/journal.pone.0056928.gSummary of signalling analysis (Table 1)The study group was stratified incrementally according to their whole body insulin resistance, determined by the M value, and the responses of each individual signalling protein to insulin were ranked and the four individuals with the greatest (Green numbers, ranking 1 to 4)) or least (Red numbers, ranking 1 to 4) responses for each protein were noted. Representative blots are shown (Figure 6). The responses of interest were insulin-induced changes in IRS1 protein expression, in PKB or p42/p44 MAP kinase phosphorylation or in p42/p44 MAP kinase activity. We observed a.
