Gene deletion of γ-actin impairs insulin-stimulated skeletal muscle glucose uptake in growing mice but not in mature adult mice

Abstract

The cortical cytoskeleton, consisting of the cytoplasmic actin isoforms β and/or γ-actin, has been implicated in insulin-stimulated GLUT4 translocation and glucose uptake in muscle and adipose cell culture. Furthermore, transgenic inhibition of multiple actin-regulating proteins in muscle inhibits insulin-stimulated muscle glucose uptake. The current study tested if γ-actin was required for insulin-stimulated glucose uptake in mouse skeletal muscle. Based on our previously reported age-dependent phenotype in muscle-specific β-actin gene deletion (^-/- ) mice, we included cohorts of growing 8-14 weeks old and mature 18-32 weeks old muscle-specific γ-actin^-/- mice or wild-type littermates. In growing mice, insulin significantly increased the glucose uptake in slow-twitch oxidative soleus and fast-twitch glycolytic EDL muscles from wild-type mice, but not γ-actin^-/- . In relative values, the maximal insulin-stimulated glucose uptake was reduced by ~50% in soleus and by ~70% in EDL muscles from growing γ-actin^-/- mice compared to growing wild-type mice. In contrast, the insulin-stimulated glucose uptake responses in mature adult γ-actin^-/- soleus and EDL muscles were indistinguishable from the responses in wild-type muscles. Mature adult insulin-stimulated phosphorylations on Akt, p70S6K, and ULK1 were not significantly affected by genotype. Hence, insulin-stimulated muscle glucose uptake shows an age-dependent impairment in young growing but not in fully grown γ-actin^-/- mice, bearing phenotypic resemblance to β-actin^-/- mice. Overall, γ-actin does not appear required for insulin-stimulated muscle glucose uptake in adulthood. Furthermore, our data emphasize the need to consider the rapid growth of young mice as a potential confounder in transgenic mouse phenotyping studies.

Keywords: glucose uptake; skeletal muscle; γ-actin.

FIGURE 1

Verification of the γ‐actin^−/− genotype. (a) Immunoblotting against γ‐actin in isolated soleus and EDL muscles from wild‐type and muscle specific γ‐actin^−/− mice. (b) Quantification of the flox allele in quadriceps and soleus muscles of wild‐type and γ‐actin^−/− mice. Two‐way repeated measures ANOVA was performed followed by a Sidak's post hoc test. ¤¤¤p < 0.001 ANOVA effects. ME = main effect. ***p < 0.001 effect of genotype. For wild‐type and γ‐actin^−/− n = 10 and 13 animals respectively

FIGURE 2

Impaired insulin‐stimulated glucose uptake in soleus and EDL muscles from growing muscle‐specific γ‐actin^−/− mice. Isolated soleus and EDL muscles excised from growing (8–14 weeks old) wild‐type and muscle specific γ‐actin^−/− mice were incubated in constantly oxygenated Krebs–Ringer–Henseleit buffer at 30°C for 30 min and then stimulated with insulin (60 nM) or kept in the unstimulated state for 10 min before a final 10 min of [³H]‐2‐Deoxy‐D‐glucose (2‐DG) tracer accumulation. (a) glucose transport into soleus and EDL muscles. (b) insulin‐stimulated muscle glucose uptake calculated as the difference between insulin and basal‐stimulated transport. (c) phospho(p)‐Akt Ser473, (d) p‐Akt Thr308, (e) Akt2, (f) p‐p70S6K Thr389, (g) p‐ULK1 Ser757, (h) p70S6K, (i) GLUT4 and (j) Rac1 levels quantified from soleus and EDL. (k) representative blots of the quantified proteins in (c–j). Ser = serine, Thr = threonine, ME = main effect, Int. = interaction, geno. = genotype, ins. = insulin. For (a, c, d, f and g) two‐way ANOVA was performed followed by a Tukey's post hoc test in case of ANOVA effect. For (e, h, i, and j), a t test was performed. ¤/¤¤/¤¤¤p < 0.1/0.01/0.001 ANOVA effect. */**/***p < 0.05/0.01/0.001 effect of insulin, #p < 0.05 different from corresponding wild‐type group. For soleus wild‐type basal, γ‐actin^−/− basal, wild‐type insulin and γ‐actin^−/− insulin n = 17, 14, 12, 11 respectively and for EDL in the same order n = 17, 14, 12, 12

FIGURE 3

Similar insulin‐induced glucose uptake in soleus and EDL muscles from adult wild‐type and muscle‐specific γ‐actin^−/− mice. Isolated soleus and EDL muscles excised from adult (18–32 weeks old) wild‐type and muscle specific γ‐actin^−/− mice were incubated in constantly oxygenated Krebs–Ringer–Henseleit buffer at 30°C for 30 min and then stimulated with insulin (60 nM) or kept in the unstimulated state for 10 min before a final 10 min of [³H]‐2‐Deoxy‐D‐glucose (2‐DG) tracer accumulation. (a) glucose transport into soleus and EDL muscles. (b) insulin‐stimulated muscle glucose uptake calculated as the difference between insulin and basal‐stimulated transport. (c) phospho(p)‐Akt Ser473, (d) p‐Akt Thr308, (e) Akt2, (f) p‐70S6K Thr389, (g) p‐ULK1 Ser757, (h) p70S6K, (i) GLUT4 and (j) Rac1 levels quantified from soleus and EDL. (k) Representative blots of the quantified proteins in (c–j). Ser = serine, Thr = threonine, ME = main effect, Int. = interaction, geno. = genotype, ins. = insulin. For (a, c, d, f and g) two‐way ANOVA was performed followed by a Tukey's post hoc test in case of ANOVA effect. For (e, h, i and j) a t test was performed. ¤/¤¤¤p < 0.1/0.001 ANOVA effect. **/***p < 0.01/0.001 effect of insulin, ##p < 0.01 different from corresponding wild‐type group. For soleus wild‐type basal, γ‐actin^−/− basal, wild‐type insulin and γ‐actin^−/− insulin n = 12, 18, 12, 19 respectively and for EDL in the same order n = 12, 17, 12, 19