Peroxisome proliferator-activated receptor-gamma agonist improves skeletal muscle insulin signaling in the pregestational intrauterine growth-restricted rat offspring

Abstract

The effect of early intervention with a peroxisome proliferator-activated receptor-gamma (PPARgamma) agonist on skeletal muscle GLUT4 translocation and insulin signaling was examined in intrauterine (IUGR) and postnatal (PNGR) growth-restricted pregestational female rat offspring. Rosiglitazone [11 mumol/day provided from postnatal day (PN)21 to PN60] improved skeletal muscle insulin sensitivity and GLUT4 translocation in prenatal nutrient restriction [50% calories from embryonic day (e)11 to e21; IUGR] with (IUGR+PNGR) and without (IUGR) postnatal nutrient restriction (50% calories from PN1 to PN21; PNGR) similar to that of control (ad libitum feeds throughout; Con) (n = 6 each). This was accomplished by diminished basal and improved insulin-responsive GLUT4 association with the plasma membrane in IUGR, IUGR+PNGR, and PNGR mimicking that in Con (P < 0.005). While no change in p85-phosphatidylinositol 3-kinase (PI3-K) and phosphatase and tensin homolog deleted on chromosome 10 (PTEN) was observed, a decrease in protein tyrosine phosphatase 1B (PTP1B; P < 0.0002) and SH2-containing protein tyrosine phosphatase 2 (SHP2; P < 0.05) contributing to the rosiglitazone-induced insulin sensitivity was seen only in IUGR+PNGR. In contrast, an increase in phosphorylated 5'-adenosine monophosphate kinase (pAMPK; P < 0.04) and insulin responsiveness of phosphorylated phosphoinositide-dependent protein kinase-1 (pPDK1; P < 0.05), pAkt (P < 0.01), and particularly pPKCzeta (P < 0.0001) and its corresponding enzyme activity (P < 0.005) were observed in all four experimental groups. We conclude that early introduction of PPARgamma agonist improved skeletal muscle activation of AMPK and insulin signaling, resulting in insulin-independent AMPK and insulin-responsive GLUT4 association with plasma membranes in IUGR, IUGR+PNGR, and PNGR adult offspring, similar to that of Con. These findings support a role for insulin sensitizers in preventing the subsequent development of gestational or type 2 diabetes mellitus in intrauterine and postnatal growth-restricted offspring.

Fig. 1.

Experimental design demonstrating the 4 experimental groups obtained by cross-fostering postnatal rat pups: Con, control mothers rearing control pups; IUGR, control mothers rearing in utero semi-nutrient-restricted pups; IUGR+PNGR, semi-nutrient-restricted mothers rearing in utero semi-nutrient-restricted pups; PNGR, semi-nutrient-restricted mothers rearing control pups. Semi-nutrient-restricted mothers received 50% of daily nutrient intake from mid- to late pregnancy [embryonic day (e)11 to e21] through lactation [postnatal day(PN)1 to PN21]. Female offspring were divided into 2 groups, non-Rosi and Rosi, based on receipt of 0.1% rosiglitazone pretreatment from PN1 to PN60. At PN60 the female offspring in each group received vehicle (basal, −) or insulin (insulin stimulated, +).

Fig. 2.

Schematic representation of the postreceptor insulin signaling pathway and insulin-independent 5′-adenosine monophosphate kinase (AMPK) beginning from insulin binding to its receptor to GLUT4 translocation via vesicular trafficking. Ins, insulin; IR, insulin receptor; IRS, insulin receptor substrate; P, phosphorylation; PPARγ, peroxisome proliferator-activated receptor-γ; PTP1B, protein tyrosine phosphatase 1B; SHP2, SH2-containing protein tyrosine phosphatase 2; PTEN, phosphatase and tensin homolog deleted on chromosome 10; PIP₂, phosphatidylinositol 4,5-bisphosphate; PIP₃, phosphatidylinositol 1,4,5-trisphosphate; PDK1, phosphoinositide-dependent protein kinase-1.

Fig. 3.

Total, plasma membrane (PM), and low-density microsomal (LDM) GLUT4 protein concentrations. Top: representative Western blots of total GLUT4 with vinculin (Vin; internal control) in non-Rosi (−)- and Rosi (+)-pretreated groups (n = 6 per pretreatment and per experimental group) (A) and PM GLUT4 (B) and LDM GLUT4 (C) protein concentrations in the Rosi-pretreated groups under basal (−Ins) and insulin-stimulated (+Ins) states (n = 3 per state in each experimental group). Bottom: densitometric quantification of the protein band of interest shown as a ratio to vinculin (A) and as % of the corresponding either non-Rosi (A) or basal (B, C) Con; depicted as means ± SD. B: *P < 0.0005 vs. corresponding basal state (−Ins), ^ψP < 0.0025 vs. Con basal state (−Ins); ^ΩP < 0.0025 vs. Con insulin-stimulated (+Ins) state. C: *P < 0.005 vs. corresponding basal (−Ins) state; ^ψP < 0.04 vs. Con basal (−Ins) state; ^ΩP < 0.0005 vs. Con insulin-stimulated (+Ins) state.

Fig. 4.

Skeletal muscle total and phosphorylated (p) PKCζ concentrations and enzyme activity. Top: representative Western blots demonstrating total PKCζ and vinculin (internal control) in non-Rosi (−) and Rosi (+) pretreatments (A) and pPKCζ and vinculin (internal control) in the basal (−Ins) and insulin-stimulated (+Ins) states of the 4 Rosi-pretreated experimental groups (B). Bottom: densitometric quantification of the protein bands of interest/vinculin shown as % of either the non-Rosi-pretreated Con (A) or basal state of Con (B); presented as means ± SD (n = 6 per Rosi or non-Rosi treatments, per basal or insulin-stimulated states per experimental group in A and B). C: PKCζ activity in basal (−Ins) and insulin-stimulated (+Ins) states of non-Rosi and Rosi pretreatments is shown as cpm/min (means ± SD; n = 6 per basal or insulin-stimulated state and per experimental group). *P < 0.005 vs. corresponding basal state (−Ins) within the same experimental group; ^ψP < 0.02 vs. non-Rosi- and Rosi-pretreated Con insulin-stimulated state (+Ins).

Fig. 5.

Total (Akt1, 2, 3), Akt2, and phosphorylated (p)Akt. Top: representative Western blots demonstrating total Akt1, 2, and 3 and vinculin (internal control) in the non-Rosi (−)- and Rosi (+)-pretreated 4 experimental groups (n = 6 per treatment and per group) (A), Akt2 and vinculin (internal control) in the non-Rosi (−)- and Rosi (+)-pretreated 4 experimental groups (n = 3 per treatment and per group) (B), and pAkt and vinculin (internal control) in the basal (−) and insulin-stimulated (+)states of the Rosi-pretreated 4 experimental groups (n = 6 per state and per group) (C). Bottom: corresponding densitometric analyses of the protein bands of interest/vinculin shown as % of the non-Rosi-pretreated Con; presented as means ± SD. *P < 0.005 vs. corresponding basal (−Ins) state within the same experimental group.

Fig. 6.

Total and phosphorylated PDK1. Top: representative Western blots demonstrating total PDK1 and vinculin (internal control) (A) and pPDK1 and vinculin (internal control) (B). Bottom: corresponding densitometric quantification of the protein bands of interest/vinculin shown as % of non-Rosi pretreatment or basal state of Con; presented as means ± SD (n = 6 per group and per pretreatment/state). *P < 0.05 vs. corresponding basal state (−Ins) within the same experimental group.

Fig. 7.

p85 subunit of phosphatidylinositol 3-kinase (PI3-K) and total PTP1B, SHP2, and PTEN concentrations. Top: representative Western blots demonstrating p85 subunit of the PI-3-K enzyme (A), PTP1B (B), SHP2 (C), PTEN (D), and vinculin (internal control) in non-Rosi and Rosi groups. Bottom: densitometric quantification of the protein bands of interest/vinculin shown as % of non-Rosi pretreatment of Con; presented as means ± SD (n = 6 per pretreatment and per experimental group). PTP1B: *P < 0.05 vs. corresponding non-Rosi pretreatment within the same experimental group; ^ψP < 0.05 vs. non-Rosi pretreated Con group. SHP2: **P < 0.03 vs. non-Rosi treatment within the same experimental group; ^ψP < 0.03 vs. non-Rosi-treated Con group.

Fig. 8.

Total and phosphorylated AMPK. Top: representative Western blots demonstrating total AMPK (A) and pAMPK (B) along with vinculin (internal control) in non-Rosi and Rosi groups. Bottom: corresponding densitometric quantification of the protein bands of interest/vinculin shown as % of non-Rosi pretreatment of Con; presented as means ± SD (n = 6 per pretreatment and per experimental group). *P < 0.01 (AMPK) or <0.008 (pAMPK) vs. non-Rosi pretreatment within the same experimental group; ^#P < 0.02 (AMPK) or <0.001 (pAMPK) vs. non-Rosi pretreatment of Con (AMPK and pAMPK) and PNGR (AMPK); ^ψP < 0.01 (AMPK) or <0.03 (pAMPK) vs. non-Rosi pretreatment of IUGR. C: ratio between pAMPK and total AMPK/vinculin each shown as % of non-Rosi pretreatment of Con and presented as means ± SD (n = 6 per pretreatment and per experimental group). *P < 0.04 vs. corresponding non-Rosi pretreatment of the same experimental group; ^#P < 0.04 vs. non-Rosi pretreatment of Cοn; ^ψP < 0.002 vs. non-Rosi pretreatment of IUGR and PNGR; ^ΩP < 0.02 vs. Rosi pretreatment of Con and IUGR.