HEPES, leupeptin, pepstatin, phenylmethylsulphonyl fluoride (PMSF), N-benzyl-ptoluene (BTS), sulfonamide, collagenase type I, suramin, and ethidium (Etd⁺) bromide were obtained from Sigma-Aldrich (St. Louis,MO, United States). The DMEM/F12 culture medium and fetal bovine serum albumin were from Gibco (Mary-land, United States). Tramadol was obtained from Genfar (Gentilly, Francia). Previously characterized anti-Cx39 and anti-Panx1 antibodies were used (Riquelme et al., 2013). Anti-atrogin1, anti-Murf, anti-Cx45, anti-Cx43, and P2X7R antibodies were obtained from Abcam (Cambridge, UK). Cy2-and Cy3-conjugated goat anti-rabbit IgG antibodies were purchased from Jackson Immuno Research (Indianapolis, IN, United States).

The present study was performed with approval from the bioethics committee of Pontificia Universidad Católica de Chile (No. 150323001). We used C57BL6 skeletal-muscle-deficient mice for Cx43 and Cx45 generated from breeding Cx43^fl/fl mice and Cx45^fl/fl mice with Myo-Cre mice, which express Cre recombinase under the control of a myogenin promoter and the MEF2C enhancer (Cea et al., 2013; Cisterna et al., 2016). The Cx43^fl/flCx45^fl/fl mice (hereafter referred to as Cre-) normally express Cx43 and Cx45, and the Cx43^fl/flCx45^fl/fl:Myo-Cre mice are double-inducible deficient in Cx43 and Cx45 (hereafter referred to as Cx3/45 deficient mice). The mice were kept in standard housing conditions with a 12-h/12-h dark/light cycle and food and water ad libitum. To minimize variability, only male mice were used in this study, focusing on sex-based differences in skeletal muscle physiology (Haizlip et al., 2015).

The tibialis anterior (TA) muscle and diaphragm were used for hematoxylin-eosin staining and immunofluorescence. For isolated fiber measurements, the flexor digitorum brevis (FDB) muscle was chosen due to the ease of isolating intact fibers without damage. Using different mouse muscles for distinct methodologies helps minimize the number of animals required. Importantly, both the FDB and TA muscles are composed predominantly of fast-twitch fibers, sharing similar metabolic and contractile properties, which supports the comparability of findings across these muscles.

The mice were placed under CLP for six or 12 h. For this, the mice were anesthetized using isoflurane (Baxter Healthcare, Guayama, Puerto Rico). The abdomen was shaved, and a midline incision (0.5 cm) was made in the intraperitoneal area. The cecum was isolated, tied, and punctured once with a 21-gauge needle (0.8 × 38 mm), and a small amount of cecal material was extruded. The cecum was returned to the abdomen, and 200 μL of PBS 1X was added to ensure sepsis. The incision was closed with a 20 mm nylon surgical suture (Tagumedica S.A., Santiago, Chile). Sham controls were subjected to the same surgical protocol and cecal isolation, but the cecum was neither tied nor punctured. The control mice consisted of the sham group. A sham procedure was performed for each time point (0, 6, and 12 h). However, no significant differences were observed between the sham groups at different time points for the parameters analyzed. Therefore, to minimize the number of mice used and to increase the sample size of the control group, all the sham groups were combined to form a single control group.

The mouse IL-6 content was measured in 35 μL of plasma using the Mouse IL-6 Quantikine ELISA Kit (R&D Systems, Inc; Minneapolis, MN, United States), according to the manufacturer’s instructions.

Freshly dissected TA muscles were embedded in a tissue-mounting OCT solution (Andes import, Chile) and fast frozen in liquid-nitrogen-cooled isopentane (Merck, Germany). Serial cryostat sections with a 16 μm thickness were obtained, placed on glass slides (75 × 25 mm, B&C, Germany), and fixed for 10 min with 4% (wt/vol) paraformaldehyde (Electron Microscopy Sciences, United States) for immunofluorescence or cross-sectional area (CSA) analyses.

The muscle fiber size was quantified by evaluating the cross-sectional area (CSA) of the fibers, measured as described previously (Cea et al., 2013). Briefly, cross-sections were stained with hematoxylin and eosin (H&E). The CSA was evaluated in multiple fibers from three images per mouse (4 mice per group) by using offline analyses with the ImageJ software (National Institutes of Health).

To detect different proteins in cross-sections (16 μm), the TA muscles sections were blocked for 1 h in a blocking solution (1% BSA, 50 mM NH4Cl, 0.01% Triton x-100, 1X PBS, and a pH of 7.4) and then incubated at 4°C overnight with diluted primary anti-atrogin1 (1:300), anti-Murf 1 (1:300), anti-Cx39 (1:300), anti-Cx43 (1:300), anti-Cx45 (1:300), anti-Panx1 (1:300), or anti-P2X7 (1:200) antibodies (Balboa et al., 2018). The samples were washed 4 times with PBS 1X and then incubated with an appropriate dilution of Cy2-or Cy3-conjugated goat anti-rabbit IgG antibodies (1:300) for 1 h in a dark room. The samples were washed 4 times with PBS 1X and one time with distilled water, and then mounted using Fluoromount-G™ with DAPI (Electron Microscopy Science, Hatfield, PA, United States) on glass slides. The images were generated using a Nikon Eclipse Ti micro-scope, and the fluorescence intensity was quantified with ImageJ. For each condition, histological sections from at least four mice were analyzed, with a minimum of three histological sections per mouse. For each section, three photographs were taken, and the fluorescence intensity of each photograph was quantified. The average fluorescence intensity for each image was plotted using a violin plot.

The myofibers of the mouse FDB muscles were isolated from anesthetized and sacrificed mice, as described previously (Riquelme et al., 2013). The muscles were immersed in a cultured medium (DMEM/F12 supplemented with 10% fetal bovine se-rum) containing 0.2% collagenase type I, and incubated for 2.5 h at 37°C. Then, they were transferred to a 15 mL tube containing 3 mL of Krebs buffer (in mM: 145 NaCl, 5 KCl, 3 CaCl₂, 1 MgCl₂, 5.6 glucose, 10 HEPES-Na, and a pH of 7,4) plus 10 μM BTS (contraction inhibitor, to reduce muscle damage). The muscle tissue was incubated for 2 min with 200 μM suramin, a P2 receptor inhibitor, to prevent the effects of ATP release during the procedure. Then, the tissue was gently triturated by passing it 7 times through a wide-tip Pasteur pipette. The myofibers were left to decant for 7 min, and the sediment was washed with HEPES-buffered Krebs saline solution containing 10 μM BTS. The myofibers were gently triturated again by passing them 15 times through a narrow tip to dissociate single myofibers. The dissociated myofibers were left to decant again for 7 min, after which the sediment was washed, resuspended in HEPES-buffered Krebs saline solution containing 10 μM BTS, and placed in 1.5 mL Eppendorf tubes until use for further analyses.

The myofiber permeability was evaluated by assessing the ethidium (Etd⁺) uptake using time-lapse measurements, as described previously (Riquelme et al., 2013). In brief, freshly FDB myofibers plated onto plastic culture dishes were washed twice with Krebs buffered solution. For the time-lapse measurements, the myofibers were incubated in a recording Krebs solution containing 5 μM Etd⁺ and recorded for 5 min in a control solution. They were subsequently recorded for 5 min after an HC blocker (lanthanum ion (La³⁺), 200 µM) was added to the Krebs solution (22). The Etd⁺ fluorescence was recorded in regions of interest (ROI) corresponding to the nuclei of the myofibers, using a Nikon Eclipse Ti inverted microscope and the NIS-Element software (Nikon, Tokio, Japan).

The oxygen consumption of muscle fibers was measured as described previously (Kuznetsov et al., 2008). Fiber bundles from the soleus muscle (0.2–0.8 mg dry/wet) were separated along their longitudinal axis with a pair of needle-tipped forceps under magnification with an Olympus SZ61 microscope. The bundles were then treated with 50 μg/mL of saponin for 30 min in ice and subsequently washed in an MIR05 medium. High-resolution O² consumption measurements were conducted at 30°C with the OROBOROS Oxygraph-2K (OROBOROS Instruments, Innsbruck, Austria).

The mice were anesthetized with 70 µL of a mixture consisting of xylazine and ketamine (1:3 proportion), and a conventional in vivo microelectrode recording technique was used (Cisterna et al., 2020). The flexor digirotium brevis (FDB) muscle was completely exposed and submerged in Krebs solution (pH of 7.4), and the RMP measurement was set to “current clamp” at 37°C. The Ag-Cl reference electrode was inserted in a 3 M KCl agarose bridge and submerged in Krebs solution to record superficial fibers from the FDB muscle. Each electrode was fabricated from a borosilicate capillary (A-M System, Inc., Cat # 596800) with the horizontal stretcher P97 (Sutter instruments, CA, EUA), and filled with a 3 M KCl solution with electrode resistances within the range of 40–50 MΩ. The recorded RMP corresponded to the potential value measured when the microelectrode accessed the sarcoplasm, crossed the sarcolemma, and remained stable for 5 s after impelling. All the experiments used an Olympus IX 51 inverted microscope with an Axopatch1-D amplifier (Molecular Devices, CA, EUA), a Digidata 1,322 digitalizer, and the Clampex 9.1 acquisition software.

Based on the literature and our previous studies, we calculated the required sample size to detect significant differences in various parameters with a significance level of 0.05 and a power of 0.8, ensuring the minimum number of animals needed to observe significant effects, which was determined to be five in most cases.

The data for each condition are summarized in different type graphs as means ± standard error (SEM), where the number of mice per treatment group is indicated in the legend of each figure. The data were evaluated using non-parametric tests, specifically the Kruskal–Wallis test followed by Dunn’s multiple comparison test. Analyses were carried out using the GraphPad software. Significant differences between groups were considered when P < 0.05.

Abundant evidence indicates that sepsis induces a myopathy characterized by reductions in muscle mass (Rocheteau et al., 2015). However, the mechanisms of this are not well understood. Previously, we observed that WT mice showed an almost 17% reduction in body weight 2 days after CLP, and this percentage remained until day 5, finally recovering at day 7 (Balboa et al., 2018). This weight reduction was attributed to the loss of muscular mass due to the documented muscular atrophy. However, at early time points, a decrease in mouse weight could have been due to a postoperative effect stemming from reduced water or food intake as a result of postoperative discomfort. The observed weight changes were similar between the control (sham) animals and those subjected to CLP in Cre- (Cx43^fl/flCx45^fl/fl) and Cx43/45-deficient (Cx43^fl/flCx45^fl/fl: Myo-Cre) mice at 6 and 12 h (Figure 1A). At 48 h, the mice begin to experience significant weight loss, which is attenuated in Cx43/45-deficient mice (Supplementary Figure 1).

As previously demonstrated, skeletal muscle IL-6 production contributes to the overall circulating IL-6 levels during exposure to a septic peritoneal infection (Laitano et al., 2021). We aimed to evaluate whether differences in the IL-6 levels existed in our model. As shown in Figure 1B, a tenfold increase in the IL-6 levels was observed in Cre-mice at 6 h post-CLP compared to the controls, and this increased approximately tenfold at 12 h. Conversely, in Cx43/45-deficient mice, only a fivefold increase in the IL-6 levels was observed at 6 and 12 h post-CLP, suggesting that there could be less muscle inflammation in Cx43/45-deficient mice. Hematoxylin and eosin staining revealed immune cell infiltration in muscles of Cre-mice at 12 h post-CLP, with multiple infiltration foci observed. In contrast, muscles of Cre + mice exhibited very few foci of immune cell infiltration, further supporting the observation of reduced muscle inflammation in these mice (Supplementary Figure 2C).

As we have previously shown that sepsis induces a decrease in CSA at 7 days (Balboa et al., 2018), we wanted to evaluate the CSA at early times. At 6 and 12 h post-sepsis induction, we observed no significant changes CSA (Supplementary Figure 2); however, levels of atrogins and MuRF, proteins associated with muscle atrophy (Tiao et al., 1994), were increased in Cre-mice. Notably, this increase was not observed in deficient CXs43/45 mice (Figure 2A). On the other hand, the diaphragm muscles showed an increase in the Murf immunoreactivity at 6 and 12 h after CLP in Cre-mice and at 12 h in Cx43/45 deficient mice (Figure 2B).

Previous studies have described that denervated skeletal muscles show the de novo expression of Cx39, 43, and 45 HCs, in addition to P2X7Rs, and the upregulation of pannexin1 (Panx1), which play a relevant role in muscle atrophy (Cea et al., 2013; Cisterna et al., 2016). Nevertheless, there is currently no evidence regarding the early expression of membrane channel-forming proteins in the skeletal muscles of mice subjected to CLP. To study this possibility, we evaluated the reactivity of Cx39, 43, and 45, Panx1, and P2X7Rs in slices of the TA and diaphragm muscles from Cre- and Cx43/45-deficient mice subjected or not to CLP.

In the TA muscles of Cre-mice Cx45 increased at 6 h, and Cx43 increased at 12 h. Meanwhile, Panx1 showed no significant changes, and P2X7Rs decreased at 12 h. In contrast, no changes were observed in Cx43/45-deficient mice in the expression of Cx39, Cx43, and Cx45 at both 6 and 12 h post-CLP. However, Panx1 and P2X7Rs increased at 6 h post-CLP (Figures 3A–E). We found similar changes in the diaphragm muscles to those observed in the TA muscles (Supplementary Figure 3). All the proteins analyzed were mainly detected in the contours of the myofibers, and were likely associated with the sarcolemma.

Given that the de novo expression of connexins after CLP was observed in the myofibers of Cre-animals, we aimed to assess whether the expression of these connexins on the sarcolemma had an impact on the resting membrane potential (RMP) of myofibers due to an increase in the ion flux through the HCs, as described before (Cisterna et al., 2020). Furthermore, the sarcolemma of the skeletal myofibers in mice exposed to sepsis becomes electrically unexcitable (Trojaborg et al., 2001). Accordingly, we evaluated whether or not the RMP is altered during early sepsis in Cre- and Cx43/45-deficient mice subjected to CLP. We observed that the RMP in Cre-animals decreased significantly at 6 h and 12 h after CLP. In contrast, Cx43/45 animals were protected, and their RMP was kept stable over time (Figure 4A).

Cx HCs allow for the flow of ions such as Na⁺ and Ca²⁺ [24], and given our previous results indicating that Cx HCs were expressed in the muscles and the RMP was altered in CLP Cre-mice, we decided to evaluate the Cx HC activity in freshly isolated FDB myofibers using the ethidium (Etd+) uptake technique. We found that the Etd⁺ uptake increased in septic Cre-animals as early as 6 h compared with control animals, but it did not increase in septic Cx43/45-deficient animals. In addition, the Etd⁺ uptake was reverted by adding La³⁺, a non-selective Cx HC blocker (Shoji et al., 2014). These results suggest that sepsis increases the sarcolemma permeability through Cx HCs (Figure 4B).

Changes induced by acute inflammatory conditions, such as sepsis, are linked to the dysfunction of mitochondrial components in muscle [30]. Moreover, alterations in the sarcolemma permeability and RMP secondary to ion influx, such as Ca²⁺, are important for the correct functioning of mitochondria (Ruiz-Meana et al., 2009; Contreras et al., 2010). To determine the mitochondrial function in CLP mice, we measured the mitochondrial oxygen consumption rate (OCR) using permeabilized fibers from the soleus muscles. The mitochondrial oxygen consumption rate was similar between Cre- and Cx43/45-deficient mice across all the evaluated parameters (Supplementary Figure 4). CLP at 6 and 12 h did not affect the mitochondrial function in our model. This suggests that mitochondria can withstand the changes observed in the other results.