Skeletal Muscle Atrophy Induced by Diabetes Is Mediated by Non-Selective Channels and Prevented by Boldine

Abstract

Individuals with diabetes mellitus present a skeletal muscle myopathy characterized by atrophy. However, the mechanism underlying this muscular alteration remains elusive, which makes it difficult to design a rational treatment that could avoid the negative consequences in muscles due to diabetes. In the present work, the atrophy of skeletal myofibers from streptozotocin-induced diabetic rats was prevented with boldine, suggesting that non-selective channels inhibited by this alkaloid are involved in this process, as has previously shown for other muscular pathologies. Accordingly, we found a relevant increase in sarcolemma permeability of skeletal myofibers of diabetic animals in vivo and in vitro due to de novo expression of functional connexin hemichannels (Cx HCs) containing connexins (Cxs) 39, 43, and 45. These cells also expressed P2X7 receptors, and their inhibition in vitro drastically reduced sarcolemma permeability, suggesting their participation in the activation of Cx HCs. Notably, sarcolemma permeability of skeletal myofibers was prevented by boldine treatment that blocks Cx43 and Cx45 HCs, and now we demonstrated that it also blocks P2X7 receptors. In addition, the skeletal muscle alterations described above were not observed in diabetic mice with myofibers deficient in Cx43/Cx45 expression. Moreover, murine myofibers cultured for 24 h in high glucose presented a drastic increase in sarcolemma permeability and levels of NLRP3, a molecular member of the inflammasome, a response that was also prevented by boldine, suggesting that, in addition to the systemic inflammatory response found in diabetes, high glucose can promote the expression of functional Cx HCs and activation of the inflammasome in skeletal myofibers. Therefore, Cx43 and Cx45 HCs play a critical role in myofiber degeneration, and boldine could be considered a potential therapeutic agent to treat muscular complications due to diabetes.

Keywords: calcium atrophy; connexins; hemichannel blocker; sarcolemma permeability.

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Figure 1

Boldine prevents the reduction in cross-sectional area of skeletal myofibers and partially prevents the elevated glycaemia in diabetic rats. (A) Representative hematoxylin–eosin-stained preparations of TA muscles from control and diabetic (STZ) rats treated or not with boldine. (B) Quantifications of images such as those shown in (A) (around 100 myofibers per animal were analyzed). Muscles of control (white bar), control treated with boldine (grey bar), diabetic (STZ, black bar), and diabetic treated with boldine (STZ + Boldine; 50 mg/kg, dashed bar) animals were processed for hematoxylin–eosin staining. The CSA was measured by offline analysis with ImageJ software. (C) Glycaemia levels were analyzed from tail blood of control and STZ rats treated or not with boldine. *** p < 0.001; ** p < 0.01; and * p < 0.05. Scale bar: 50 µm. (N = 5 animals per group, 10 images were taken per animal and were analyzed (around 500 fibers per group)).

Figure 2

Boldine reduces the sarcolemma permeability of skeletal myofibers in diabetic rats. (A), Control and diabetic (STZ) rats treated or not with boldine for 3 weeks were injected i.p. with Evans blue (80 mg/kg). Tibialis anterior muscles were dissected 5 h post-injection. Evans blue (EB4−) was detected as red staining. Scale bar: 50 μm. (B), Quantification in arbitrary units (AU) of EB4− inside the fibers (red signal) from images like show in (A), Boldine totally corrected the sarcolemma permeability. The red staining depicts the presence of the dye surrounding myofibers (control, control + boldine, and STZ + boldine) or within the myofibers (STZ). p < 0.001. (N = 5 animals per group and ten images per animal were analyzed (around 500 fibers per group)).

Figure 3

Boldine reverts the sarcolemma permeabilization of freshly isolated skeletal myofibers from diabetic rats. (A) Immunofluorescence against Cx39, Cx43, Cx45, P2X7, and Panx1 proteins was evaluated in freshly isolated skeletal myofibers from control and diabetic (STZ) rats treated or not with boldine (scale bar: 20 μm). (B) Time lapse of ethidium (Etd+) uptake assay in myofibers of flexor digitorum brevis muscles from control (circles), control + boldine (black circles), diabetic (STZ, squares) and diabetic animals treated with boldine (50 mg/kg) for 3 weeks (STZ + boldine, triangles). (C) Quantification of Etd+ uptake rate (slopes of curves like in (A). In some experiments, the acute effect of 150 mM La3+ on the uptake rate of myofibers from control and diabetic animals was evaluated. *** p < 0.001. (N = 5 animals per group and from each animal 10 myofibers were analyzed).

Figure 4

Boldine blocks P2X7 receptors of cells cultured in high glucose. Hela cells were transiently transfected with a P2X7R-EGFP vector in 22.5 mM glucose and 24 h after being loaded with Fura-2 to evaluate the P2X7R activity in response to boldine. (A) Treatment with 2 mM ATP evoked a rapid and significant increase in Ca2+ signal and showed a nadir for at least 10 min. (B). When 50 µM boldine was applied after 5 min of ATP stimulation, the Ca2+ signal decreased drastically. Each plotted point corresponds to the average ± SEM of the Ca2+ signal. In each experiment, ~20 EGFP-positive cells were recorded over time (n = 3).

Figure 5

Deletion of Cx43 and Cx45 prevents skeletal muscle atrophy and partially prevents the glycaemia elevation induced by STZ in mice. (A) Representative hematoxylin–eosin-stained preparations of TA muscles from control (Cx43fl/flCx45fl/fl), control treated with STZ (STZ + Cx43fl/flCx45fl/fl), Cx43/Cx45 muscle-expression-deficient mice (Cx43fl/flCx45fl/fl:M-C), and Cx43fl/flCx45fl/fl: M-C + STZ mice. (B) Quantification of all sets of images collected as shown in (A). (C) Glycaemia levels were analyzed from tail blood of Cx43fl/flCx45fl/fl; STZ + Cx43fl/flCx45fl/fl mice treated or not with boldine. The cross-sectional area (CSA) was evaluated to analyze the presence of atrophy. Control animals of the same age were used as controls (N = 5 animals; each value corresponds to the average ± SEM, ** p < 0.01 and *** p < 0.001. Scale bar: 50 µm.

Figure 6

Lack of connexin43 and connexin45 expression prevents increased membrane permeability induced by diabetes in mice skeletal myofibers. Freshly isolated skeletal muscle fibers from flexor digitorum brevis muscles were isolated from control (Cx43fl/flCx45fl/fl) and Cx43/Cx45 muscle-expression-deficient (Cx43fl/flCx45fl/fl: Myo-Cre) mice, which were treated with streptozotocin (+STZ) or not (−STZ) to induce diabetes in these mice. The myofibers were used to measure membrane permeability by an ethidium (Etd+) uptake assay. The graph shows the Etd+ uptake rate obtained from slopes of fluorescent intensity over time curves. The white bar represents myofibers from control mice without STZ treatment (Cx43fl/flCx45fl/fl). The horizontal dashed bar represents myofibers from mice from which Cx43 and Cx45 genes were removed in early differentiation stages with the Cre-LoxP method (Cx43fl/flCx45fl/fl: Myo-Cre). The black bar represents myofibers from control mice (Cx43fl/flCx45fl/fl) treated with STZ, and the diagonal dashed bar represents myofibers from Cx43fl/flCx45fl/fl: Myo-Cre mice treated with STZ. Each bar represents the mean ± SE of n = 4 animals for each condition in independent experiments. ** p < 0.01.

Figure 7

High glucose increases the sarcolemma permeability of cultured skeletal myofibers. (A) Etd+ uptake rate of myofibers cultured for 24 h in DMEM/F12 culture medium plus different glucose concentrations. Each value is the mean  ±  SEM of nuclear Etd uptake rate. * p <  0.05 and ** p  <  0.01 by ANOVA with Tukey′s multiple comparisons test. N = 3–6. (B) Representative fluorescence intensity curve of Etd+ uptake of myofibers cultured for 24 h with 8 mM glucose (NG), 8 mM glucose plus 50 µM boldine (Boldine + NG), 22.5 mM glucose (HG), or 22.5 mM glucose (HG) plus 50 µM boldine (Boldine + HG). At about 300 s recording, cells were treated with 200 µM La3+, a known HC blocker.