Secondary Logo

Journal Logo

Basic Science Aspects

Contractility and Myofibrillar Content in Skeletal Muscle are Decreased During Post-Sepsis Recovery, But Not During the Acute Phase of Sepsis

Crowell, Kristen T.∗,†; Lang, Charles H.∗,‡

Author Information
doi: 10.1097/SHK.0000000000001555
  • Free
  • Editor's Choice



While multiple lines of evidence demonstrate that severe sepsis acutely decreases muscle protein synthesis leading to a reduction in mass when the septic insult persists (1), there is considerably less understanding of the metabolic defects in skeletal muscle during the post-sepsis-recovery phase and how these changes may translate into functional changes. This convalescence phase in rodents, studied 10 days after induction of sepsis and the subsequent surgical removal of the septic nidus and treatment with antibiotics, has been characterized by normal food intake and, when compared with the acute phase, a >98% reduction in plasma inflammatory cytokine levels (2, 3). Moreover, muscle atrophy persists during recovery and is independent of any difference in food consumption. As muscle protein synthesis is more than 2-fold higher during sepsis recovery, the blunted restoration of muscle mass during this phase appears to result from the sustained elevation in ubiquitin-proteasome activity (2). However, this previous study was limited in that it determined the rate of global mixed protein synthesis and did not directly assess the synthesis of proteins in the myofibrillar versus sarcoplasmic pools. This imbalance between protein synthesis and breakdown is integral to both the initial reduction in protein mass during the acute septic insult as well as the accretion of muscle during convalescence which, in turn, is responsible for the recovery of muscle contractile function. In humans, the convalescence phase may proceed over a period of weeks to months, and can be extended even longer in some types of critical illness myopathies (4, 5). This chronic post-sepsis reduction of muscle mass and strength during convalescence leads to functional limitations, loss of independence, increased frailty, and an impaired quality-adjusted survival (6, 7).

While numerous studies have focused on unraveling the effect of sepsis on protein balance in muscle (1), there are surprisingly few studies on the how sepsis alters skeletal muscle function (8). For example, early studies reported that low-dose endotoxin (used as a model of sepsis) produced transient weakness of voluntary quadriceps contraction in humans 3 to 5 h after its administration, a defect that was independent of altered nerve conduction (9). Moreover, endotoxin when administered in preclinical rodent models also acutely decreased isometric force generation and the force–frequency relationship under in vitro conditions (10, 11). Tumor necrosis factor-α was identified as one possible mediator for these endotoxin-induced changes as direct in vitro incubation of limb muscle with this proinflammatory cytokine also acutely impaired contractility, as evidenced by a downward shift in the force–frequency relationship and depression of tetanic force generation in isolated fibers (12). Using a more clinically relevant model of sepsis, Rossignol et al. (13) reported contractile defects in the extensor digitorum longus (EDL) at 10 days post-cecal ligation and puncture (CLP). It is noteworthy that rats in this study were in a state of chronic sepsis and inflammation, and were cachexic at the time muscle function was assessed. However, muscle function in this earlier report was studied in vivo by stimulation of the sciatic nerve so that contractile defects could not be unequivocally localized to muscle per se as opposed to impaired nerve conduction. Finally, in studies of humans with sepsis exceeding 7 days, there was a greater than 50% decrease in specific force generation in abdominal muscle, compared with muscle from control patients (14). To our knowledge, only the recent study by Owen et al. (15) has examined muscle function during convalescence after sepsis when the inflammatory milieu had greatly diminished. Their work reported a reduction in maximal specific force by the EDL at 2 weeks (−25%) and 1 month (−17%) in 16 month old (e.g., aged) male mice.

Given the abovementioned gaps in knowledge, the purpose of the current study was to examine muscle contractility during the acute phase of sepsis (24 h post-CLP) and during the recovery phase (10 days post-CLP, cecal resection and antibiotics) using a clinically relevant model of sepsis and convalescence (2, 3, 16). By assessing in vitro muscle function we can separate contractile defects that are related to a reduction in muscle mass and/or cross-sectional area (CSA) versus intrinsic defects in the contractile apparatus that are independent of nerve conduction, by assessing isometric and tetanic-twitch force generation under basal conditions and after fatiguing stimuli. Finally, we also assessed the relative abundance of representative contractile proteins and energy status as they are primary determinants of force generation in skeletal muscle.


Animal care

All experiments were approved by the Institutional Animal Care and Use Committee at the Pennsylvania State University College of Medicine (#46-946-17) and adhered to National Institutes of Health (NIH) guidelines. Information contained in Materials and Methods addresses the ARRIVE guidelines for the use of animals in research (17). Male C57BL/6 mice (Charles River; Wilmington, Mass) aged 10 weeks (≈27 g) were received by the animal care facility at the Pennsylvania State College of Medicine and acclimated for 7 to 10 days. Mice were maintained in a controlled environment (i.e., 23 C, 12 h/12 h light/dark cycle, and 30% to 70% humidity) and singly housed in plastic cages with corncob bedding. Mice were provided ad libitum water and standard rodent chow (Teklad Global 2018, Envigo; Indianapolis, Ind) that has an average nutrient profile of 19% protein and 9% fat. At the end of acclimation, mice were randomly divided into sepsis or pair-fed sham controls (details below), with the exact number of mice in each group provided in the legends to the respective figures and tables. Only male mice were used in the current study because of the known difference in lean body and muscle mass between sexes.

Experimental design

Mice were randomly assigned to either the septic or time-matched control groups. Polymicrobial peritonitis was induced using CLP and separate groups of mice were studied in the acute phase of sepsis (i.e., 24 h post-CLP) and during the post-sepsis recovery phase (i.e., day 10), as we previously characterized in detail (2, 3, 16). On Day 0, mice were anesthetized with isoflurane (3–4% induction with 2% to 3% maintenance; Vedco, St. Joseph, Mo) with oxygen. Sterile surgical technique was used and a midline laparotomy was performed after the abdomen was shaved and cleaned with betadine. The cecum was ligated using surgical silk (4-0, Covidien, Minneapolis, Minn) approximately 0.8 cm from the distal end, and punctured twice using a 25-gauge needle. Patency was ensured by extruding a small amount of feces from the puncture sites before replacing the cecum back in the abdominal cavity. The abdominal wall was closed using 5-0 silk prior to closing the skin incision with wound clips. Resuscitation consisted of 1 mL of ≈37 C sterile 0.9% saline containing 0.05 mg/kg buprenorphine (Reckitt Benckiser Pharmaceuticals, Richmond, Va) and was administered subcutaneously 15 min prior to surgery and once again 12 h later. Sham-control mice underwent the same surgical procedure except the cecum was not ligated or punctured. Septic mice consume little or no food during the first night after CLP (2); therefore, control and septic mice were fasted (with water ad libitum) to ensure the same basal nutritional state. Mice were euthanized 24 ± 2 h after laparotomy ± CLP and this is referred to as the “acute sepsis” group.

A separate group of mice were used to investigate alterations in the recovery phase post-sepsis. While the CLP procedure was the same as above, septic mice had free access to food and water for the entire experiment, and control animals were pair-fed. Starting 24 h post-CLP (Day 1), antibiotic (0.5 mg meropenem; Fresenius Kabi, Lake Zurich, Ill) and buprenorphine were injected subcutaneously (total volume 1 mL of sterile saline) every 12 h for the next 5 days (Days 1–5). Forty-eight hours post-CLP, mice were anesthetized using isoflurane as above. The original incision was reopened, the ischemic portion of the cecum distal to the original suture ligation was resected, and the abscess around the cecum was removed. The peritoneal cavity was irrigated with warm saline (≈5 mL) until the debris and abscess was cleared. The abdominal incision was closed as described above. Body temperature was maintained during the procedure by keeping mice on a warming pad until they regained consciousness, and another 1 mL of saline was administered for resuscitation. Time-matched control mice received the same volume of fluid resuscitation with buprenorphine, but did not receive antibiotic, and the cecum remained intact. Mice in this “sepsis-recovery” group were observed through Day 10. Our current model, where the cecum is removed and antibiotics administered, permits the study of convalescence from sepsis, and differs from models where the ligated and punctured cecum is not removed and a more chronic low-grade inflammatory state develops (see Discussion for details). This latter group was not included in the current study to minimize the number of experimental groups for statistical purposes. Death was not an endpoint in this study. Mice were checked three times per day, and mice were euthanized if they were found to be moribund, immobile, unable to consume food and/or water, or had a rectal temperature < 30 C for more than 8 h. Mice not meeting these humane endpoints were anesthetized with isoflurane and euthanized by cardiac excision and exsanguination. Both the acute sepsis and sepsis-recovery groups with their respective control groups were repeated twice.

Muscle function

Mice were anesthetized with isoflurane and the EDL muscles carefully excised by the same technician blinded to animal treatment. The contraction protocols used are those previously described (18, 19), and modified by our laboratory (20). The EDL in C57Bl/6J mice has been reported to have a fiber type distribution of approximately 70% type IIb, 3% type IIx, and 1% type I fibers (21). Muscle contraction data were obtained from isolated EDL muscles using a four-chamber MyoDynamics muscle strip myograph system (model 840MD; DMT, Denmark). Muscles were incubated at 25 C in a bath containing Krebs-Henseleit buffer with 10 mM glucose and continuously oxygenated with 95% O2/5% CO2. All determinations were made on muscles for which the optimal length (Lo) was determined (20), and which did not differ between treatment groups. Time to peak tension (ms) was the time from the onset of contraction to maximum tension, whereas the half-relaxation time (1/2RT; ms) equals the time from peak tension to 50% of peak tension. Thereafter, maximum twitch tension (Pt; mN) and maximal tetanic tension (Po; mN) were determined. Force–frequency curves were generated by exposing muscle to a train of stimulations of increasing frequency (15, 30, 45, 60, 75, 90, 105, 125, 140, and 160 Hz) and muscle fatigue was produced by low-frequency tetanic muscle stimulation. The resultant time–fatigue curve yields an early peak response followed by a decay in contractile force. All data acquisition and analysis were completed by Powerlab/LabChart (ADInstrucments; Colorado Springs, Colo). Data were expressed as absolute values and, where appropriate, data were also normalized to the physiological cross sectional area (PCSA; mm2) of each muscle, using the previously validated relationship: PCSA = muscle mass (mg)/[Lo (mm) × (L/Lo) × 1.06 mg/mm3)], where 0.45 was used for the fiber-to-muscle length (L/Lo) ratio of EDL. For example, specific twitch force = Pt/PCSA and specific tetanic force = Po/PCSA.

Protein synthesis, and separation of myofibrillar and sarcoplasmic proteins

In vivo protein synthesis was measured using the nonisotopic SUnSET method as originally validated (22), and previously reported by our laboratory (2, 23). Briefly, mice were injected intraperitoneally with puromycin (0.04 μmol/g body weight) dissolved in sterile saline 30 min prior to euthanasia. At this relatively low-dose, puromycin is incorporated into elongating peptide chains and the relative rate of protein synthesis estimated by determining the relative amount of puromycin-labeled peptides formed. Mice were then deeply anesthetized with isoflurane and the EDL from both legs was excised and weighed. One muscle was homogenized in ice-cold buffer and the contralateral ELD was freeze-clamped and stored at −80°C; sarcoplasmic and myofibrillar fractions were then isolated from skeletal muscle as previously described by our laboratory (24). Western blotting was performed using an antipuromycin antibody for the immunological detection of puromycin-labeled peptides that are synthesized during the labeling period (see Table 1 for details). All four experimental groups were run on the same gel and data expressed as a percentage of the control value in the acute septic state. Direct comparison of myofibrillar and sarcoplasmic protein synthesis is not possible as the two pools were determined on separate gels.

Table 1 - Antibodies and gel details used for Western blot analysis
Antibody Vendor Catalog no. Dilution Molecular weight (kDa) SDS-PAGE gel (%) Protein loaded (μg)
α-Actinin 3 Invitrogen PA5-27862 1:1,000 103 7.5 20
Actin CST 8457 1:5,000 45 10 50
Desmin CST 5332 1:1,000 53 15 20
GAPDH CST 2118 1:10,000 37 15 20
Myomesin-2 Invitrogen PA5-51017 1:1,000 165 7.5 20
Myosin Light Chain 1 and 3f DSHB F310 1:4,000 18 15 15
OXPHOS-CI Abcam ab110413 1:1,000 20 12.5 20
OXPHOS- CV,CIV,CIII,CII Abcam ab110413 1:2,000 30, 40, 48, 55 12.5 20
Sarcomeric Actin Clone 5C5 Sigma A2172 1:2,000 43 15 15
Skeletal Myosin (FAST) Clone MY-32 9 Sigma M4276 1:10,000 223 7.5 10
Tropomyosin, striated muscle DSHB CH1 1:1,000 39 10 20
Troponin I—fast skeletal Abcam ab 184554 1:1,000 21 15 20
Troponin T, fast skeletal muscle specific DSHB JLT12 1:1,000 26 15 20
Tubulin Clone B-5–1–2 Sigma T6074 1:4,000 50 10 15
Vimentin CST 3932 1:1,000 57 10 20
Puromycin Millipore MABE343 1:4,000 15-250 15 20
Abcam (Cambridge, UK); CST, Cell Signaling Technology (Danvers, Mass), DSHB, Developmental Studies Hybridoma Bank (Iowa City, Iowa); Invitrogen/ThermoFisher (Carlsbad, Calif); Sigma (St. Louis, Mo).

Western blotting

Western blot analysis was performed as previously described (2, 16, 23) with specific details presented in Table 1. Briefly, a portion of muscle was homogenized in ice-cold homogenizing buffer (20 HEPES (pH 7.4), 2 EGTA, 0.2 EDTA, 100 KCl, 50 β-glycerophosphate, 50 NaF, 0.5 sodium orthovanadate, 1 benzamidine, 0.1 PMSF and 1 DTT). The protein concentration of each tissue homogenate was quantified (Bio-Rad Protein Assay, Hercules, Calif) and SDS-PAGE was performed using equal amounts of total protein per sample. To verify equal protein loading the PDVF membranes were stained with Ponceau S (Aqua Solutions, Deer Park, Tex). Blots were blocked in 5% nonfat dry milk and incubated overnight with primary antibody at 4 C. Blots were washed with an appropriate secondary antibody and developed with enhanced chemiluminescence reagents (Pierce Chemical, Rockford, Ill) according to manufacturer's instruction. Blots were imaged using FluorChem (ProteinSimple, San Jose, Calif) and densities in the linear range were quantified using Image J (NIH, Bethesda, Md). For each protein of interest, all four experimental groups were run on the same gel and data expressed as the percent change from the control value determined in the acute septic condition.

High-energy phosphate content

Intracellular concentrations of ATP, ADP, AMP, and phosphocreatine (PCr) were determined on EDL muscles obtained from mice each of the four experimental groups. Frozen tissue was powdered under liquid nitrogen and homogenized at 4 C in 6% (w/v) perchloric acid. After centrifugation, the supernatant was neutralized with KOH. The potassium perchlorate precipitate was removed by centrifugation, and the neutralized supernatant was used to measure high-energy phosphate metabolite concentrations by standard spectrophotometric methods, as previously described (25).

Statistical analysis

All data were analyzed using Graph Pad (San Diego, Calif) by two-way analysis of variance (ANOVA; sepsis × time) with Student–Newman–Keuls post hoc when appropriate, and values considered statistically significant when P < 0.05. Where appropriate, figures have individual data points presented as a box-and-whisker plot with the mean, upper and lower quartiles, and upper and lower values identified. The sample size for each group is presented in the legend to each figure or table.


General characteristics of muscle from septic mice

Mortality for the acute sepsis group at the time of euthanasia was ≈16%. For the sepsis-recovery group of mice, morality was ≈12% at 24 h and ≈45% at the 10-day time point. The starting body weight did not differ between mice in the four experimental groups (Table 2). The decrement in body weight seen in the acute septic group was approximately 3 g, and this decrease was comparable to that observed in control mice that were time-matched and pair-fed. By the Day 10 time point, both the pair-fed control and septic mice had regained body weight such that weights were not statistically different from their respective starting weights on Day 0. The weight of the EDL did not differ between control and septic mice at 24 h, but there was a 10% decrease in the mass of the EDL from mice in the sepsis-recovery phase, compared with time-matched control mice. This decrease was evident whether data were expressed as absolute values or normalized to each animal's body weight (Table 2). The length of the isolated EDL averaged 11.9 ± 0.12 mm and did not differ between control and septic groups during either the acute or recovery phase. The PCSA of the EDL did not differ between in the four groups, although the PSCA of the EDL from sepsis-recovery mice tended to be lower than time-matched control values (Table 2). As control mice were pair-fed, there was no difference in food consumption throughout the experimental protocol between control and septic mice (days 1–2: 0.5 ± 0.2 g; days 3–4: 2.1 ± 0.4 g; days 5–6: 4.6 ± 0.5 g; days 7–8: 4.5 ± 0.3 g; and days 9–10: 4.3 ± 0.3 g). Food intake in the current study at days 5 to 10 did not differ from published data from our laboratory obtained in mice that had no surgical intervention (4.1 ± 0.2 g/d) (2).

Table 2 - General characteristics of EDL muscles isolated from control and septic mice during the acute and recovery phases
Control Acute sepsis Control Sepsis-recovery
Body weight, g
 Preweight, Day 0, g 27.4 ± 1.6 26.9 ± 1.1 26.8 ± 2.1 27.2 ± 1.3
 Decrement in BW, g −3.1 ± 0.3a −2.9 ± 0.4a −0.2 ± 0.5b −0.4 ± 0.7b
Absolute EDL weight, mg 15.2 ± 0.8a 15.0 ± 0.9a 15.1 ± 0.7a 13.7 ± 0.6b
Normalized EDL/BW, mg/g 0.57 ± 0.01a 0.55 ± 0.01a 0.56 ± 0.02a 0.50 ± 0.02b
PCSA, mm2 2.67 ± 0.15 2.64 ± 0.17 2.64 ± 0.13 2.41 ± 0.12
Protein synthesis, % control
 Myofibrillar pool 100 ± 9a 47 ± 10b 104 ± 12a 113 ± 18a
 Sarcoplasmic pool 100 ± 11a 52 ± 8b 97 ± 11a 163 ± 22c
Values are means ± SEM; n = 16 mice per group. Data are from male mice at either 24 h (acute) after sepsis or during the recovery phase (10 day post-CLP). Decrement in body weight was calculated as the body weight at Day 0 minus body weight at the time of euthanasia.
Protein synthesis in myofibrillar and sarcoplasmic pools was determined by the SUnSET method and data in each group are compared to their respective acute control value. Values with different superscripted letters are statistically different (P < 0.05), while values with the same letter are not statistically different.EDL indicates extensor digitorum longus.

Despite the matched food intake between control and septic mice at each time point, we detected marked differences between the relative rate of protein synthesis in skeletal muscle (Table 2). The synthesis of both myofibrillar and sarcoplasmic proteins was reduced ≈50% during the acute septic state. The synthesis of proteins in both pools increased during the post-sepsis recovery phase, compared to values during acute sepsis. However, while the synthesis of sarcoplasmic proteins was elevated ≈60% above time-matched control values during the recovery period, there was no significant difference in the synthesis of myofibrillar proteins between sepsis-recovery and control mice.

Contractile function

After determining Lo for each muscle, in vitro contractile function was assessed. No contractile defects were detected when assessed in the acute period of sepsis (e.g., 24 h post-CLP) (Fig. 1, first two bars in (A–D)). However, there was a ≈50% sepsis-induced decrease in both the specific maximum isometric twitch force (Fig. 1A) and the specific maximum tetanic force (Fig. 1B) during the sepsis-recovering period, compared with time-matched pair-fed control values. In contrast, there was no difference in the time to peak tension or half-relaxation rate in EDL from control or septic mice either during the acute period or sepsis-recovery (Fig. 1, C and D). The EDL from both control and septic mice demonstrated a typical frequency-dependent increase in tension. The force–frequency curve was not altered 24 h after sepsis (Fig. 1E), but there was a downward shift in the curve in the sepsis-recovery mice, compared with time-matched control values, such that the absolute force generated by the EDL during sepsis-recovery was 25% to 30% lower when the stimulus frequency was 90 Hz to 160 Hz (Fig. 1F).

Fig. 1
Fig. 1:
Effect of sepsis during the acute and recovery phase on isometric force and tetanic force production in the extensor digitorum longus (EDL) from male mice at either 24 h after sepsis (acute) or during sepsis-recovery (10 days post-CLP). Box-and-whisker plots represent the mean, upper and lower quartiles, and upper and lower values identified. Sample size was 16 for each group, unless otherwise noted. Data were analyzed by two-way ANOVA (sepsis × time) with Student–Newman–Keuls post hoc tests when appropriate. P < 0.05 compared to value from all other groups. CLP indicates cecal ligation and puncture.

We also assessed whether sepsis altered resistance to fatigue and the ability to recovery from previous contractile activity. Two-way ANOVA indicated a significant sepsis effect compared to control values for the time necessary to reach 50% of maximal force; however, only the decrease seen during the sepsis-recovery period was statistically significant by post-hoc analysis (Fig. 2A). Likewise specific twitch force generation after recovery from the fatigue protocol was also reduced in septic mice (Fig. 2B). Moreover, there was time-dependent effect of sepsis, with fatigue force generation being reduced 34% during the acute phase and 71% during sepsis-recovery. Finally, sepsis also impaired the recovery profile of tetanic contraction in EDL (i.e., sepsis effect); however, again, contractile recovery was only significantly reduced compared to time-matched control values in EDL from rats during the post-sepsis recovery period.

Fig. 2
Fig. 2:
Effect of sepsis during the acute and recovery phase on fatigability and force generation post-recovery in the extensor digitorum longus (EDL) from male mice at either 24 h after sepsis (acute) or during sepsis-recovery (10 days post-CLP). A, B, Box-and-whisker plots representing the mean, upper and lower quartiles, and upper and lower values identified. These data represent the average values from each muscle determined twice during the 3 to 4 min time point during the fatigue protocol described in Methods. C, Values are means ± standard deviation for the recovery profile of tetanic contraction in all four experimental groups. Sample size was 16 for each group, unless otherwise noted. Data were analyzed by two-way ANOVA (sepsis × time) with Student–Newman–Keuls post hoc tests when appropriate. A, P < 0.05 compared to value from all other groups. B, Groups with different letters are statistically different from each other at P < 0.05. C, P < 0.05 compared to value from time-matched control group.

Contractile proteins

The relative abundance of the thick filaments myosin heavy chain and myosin light chain was reduced more than 80% and 50%, respectively, in muscle from the sepsis-recovery mice, compared with time-matched pair-fed control values (Fig. 3, A and B, respectively). In contrast, there was no significant difference in the relative amount of these two proteins between acute sepsis and their respective control values. The decrease in the protein abundance in muscle during the recovery phase of sepsis did not result from differential protein loading on the gel as α-tubulin (loading control) was comparable between all groups (see bottom band on top Western blot). Moreover, as illustrated in the representative Ponceau S gel (Fig. 3C), total protein per lane did not differ between treatment groups.

Fig. 3
Fig. 3:
Effect of sepsis during the acute and recovery phase on muscle thick filament proteins. Representative Western blot of extensor digitorum longus (EDL) from male mice at either 24 h after sepsis (acute) or during sepsis-recovery (10 days post-CLP). All proteins ran at or near their predicted molecular weight (identified at the margin of each blot). A total of n = 8 muscles per group were run and a box and whisker plot presented for each with the mean, upper and lower quartiles, and upper and lower values identified. Data were analyzed by two-way ANOVA (sepsis × time) with Student–Newman–Keuls post hoc tests when appropriate. A, B, P < 0.05 compared to value from all other groups.

The relative amount of the thin filament proteins α-sarcomeric actin, tropomyosin, and troponin I and T were all decreased in EDL during sepsis-recovery, compared to control values (Fig. 4, A–D). During the acute phase, the amount of the first three of these proteins did not differ between septic and control mice; however, there was a small albeit statistically significant increase in troponin T in the ELD from septic mice at this time (Fig. 4C).

Fig. 4
Fig. 4:
Effect of sepsis during the acute and recovery phase on muscle thin filament proteins. Representative Western blot of extensor digitorum longus (EDL) from male mice at either 24 h after sepsis (acute) or during sepsis-recovery (10 days post-CLP). All proteins ran at or near their predicted molecular weight (identified at the right of blot). A total of n = 8 muscles per group were run and a box and whisker plot presented for each with the mean, upper and lower quartiles and upper and lower values identified. Data were analyzed by two-way ANOVA (sepsis × time) with Student–Newman–Keuls post hoc tests when appropriate. P < 0.05 compared to value from all other groups. (C) Groups with different letters are statistically different from each other, P < 0.05.

Representative proteins in the Z-disc (i.e., α-actinin 3; Fig. 5A) and the M-band (i.e., myomesin-2; Fig. 5B) also demonstrated a similar relative reduction (≈-60%) in abundance during the recovery phase. In contrast, during post-sepsis recovery, the relative amount of the intermediate filament desmin was unaltered (Fig. 5C), while the protein abundance for vimentin was increased ≈50%, compared to control values (Fig. 5D).

Fig. 5
Fig. 5:
Effect of sepsis during the acute and recovery phase on representative muscle proteins present in the Z-disc, M-band, or intermediate filament proteins. Representative Western blot of extensor digitorum longus (EDL) from male mice at either 24 h after sepsis (acute) or during sepsis-recovery (10 days post-CLP). All proteins ran at or near their predicted molecular weight (identified at the right of each blot). A total of n = 8 muscles per group were run and a box and whisker plot presented for each with the mean, upper and lower quartiles and upper and lower values identified. Data were analyzed by two-way ANOVA (sepsis × time) with Student–Newman–Keuls post hoc tests when appropriate. P < 0.05 compared to value from all other groups.

OXPHOS proteins and high-energy phosphate content

There was no statistically significant treatment (i.e., control vs sepsis) or time (i.e., acute vs recovery) effect in the relative content for the OXPHOS complex-1, -II, -III, -IV, or –V proteins in EDL (Table 3 and representative blot in Fig. 6). There was also no statistical difference in the ATP, AMP, or PCr concentrations in EDL from mice during the acute sepsis phase or during the recovery period, compared with their respective time-matched pair-fed control values (Table 3). The only statistically significant change observed was for the AMP:ATP ratio during acute sepsis that was the result of changes in ATP (decreased) and AMP (increased) that individually did not achieve statistical significance.

Table 3 - Skeletal muscle content of OXPHOS complex proteins and high-energy phosphates in control and septic mice during the acute and recovery phase
Acute sepsis Sepsis recovery
Control Septic Control Septic
OXPHOS complexes
 Complex I-NDUFB8 100 ± 18 159 ± 33 179 ± 32 103 ± 10
 Complex II-SDHB 100 ± 12 149 ± 21 158 ± 13 161 ± 21
 Complex III-UQCRC2 100 ± 10 128 ± 18 117 ± 12 101 ± 6
 Complex IV-MTCO1 100 ± 7 107 ± 7 111 ± 8 124 ± 11
 Complex V-ATP5A 100 ± 8 110 ± 9 100 ± 7 96 ± 11
High-energy phosphates
 ATP, nmol/mg protein 65.2 ± 4.8 51.4 ± 3.6 62.5 ± 3.3 66.7 ± 4.2
 AMP, nmol/mg protein 1.5 ± 0.2 1.9 ± 0.3 1.6 ± 0.2 1.8 ± 0.2
 AMP:ATP ratio (x 103) 23.1 ± 3.3a 37.2 ± 2.6b 25.6 ± 2.9a 26.9 ± 3.6a
 PCr, nmol/mg protein 227 ± 29 191 ± 21 236 ± 35 241 ± 28
 PCr:ATP ratio 3.5 ± 0.3 3.7 ± 0.4 3.8 ± 0.4 3.5 ± 0.4
For the various OXPHOS complex proteins, values are means ± SEM; n = 8 per group, where the control value was set at 100 AU for each individual protein. NDUFB8 indicates NADH dehydrogenase ubiquinone 1 beta subcomplex subunit 8; SDHB, succinate dehydrogenase iron-sulfur subunit; UQCRC2, cytochrome b-c1 complex subunit 2; MTCO1, mitochondrial encoded cytochrome c oxidase I; ATP5A, ATP synthase F1 subunit alpha. There were no statistical differences detected in muscles from control and septic mice during either the acute or recovery phase by 2-way ANOVA (P > 0.05). Representative blot is presented in Figure 6. For high-energy phosphate concentrations, values are means ± SEM; n = 8 per group. PCr, phosphocreatine. Means with different superscripted letter (a, b) are significantly different (P < 0.05).

Fig. 6
Fig. 6:
Sepsis does not alter mitochondrial proteins in the OXPHOS complex during the acute or recovery phases of sepsis. Representative Western blot of the extensor digitorum longus (EDL) from male mice at either 24 h after sepsis (acute) or during sepsis-recovery (10 days post-CLP). A total of n = 8 per group were run and mean ± SEM values for all groups are presented in Table 3; there were no statistically significant differences for any of the five proteins between the four experimental groups. NDUFB8 indicates NADH dehydrogenase ubiquinone 1 beta subcomplex subunit 8; SDHB, succinate dehydrogenase iron-sulfur subunit; UQCRC2, cytochrome b-c1 complex subunit 2; MTCO1, mitochondrial encoded cytochrome c oxidase I; ATP5A, ATP synthase F1 subunit alpha. Molecular weights are noted in the left margin of the representative blot.


Our results demonstrate profound skeletal muscle contractile defects in the post-sepsis recovery phase that are not observed acutely after sepsis. This impaired contractility during convalescence was evidenced by a reduction in basal isometric and tetanic-specific force generation as well as an increased fatigability. These defects were intrinsic and independent of mass, and were associated with a marked reduction in the abundance of several thick and thin filament proteins as well as proteins present in the M-band and Z-disc.

Previous studies have examined the acute effect of sepsis or sepsis-like insults on skeletal muscle contractility, often generating equivocal conclusions. For example, although recognized as a relatively poor model of sepsis, endotoxin decreased force generation in flexor halluces longus from Syrian Golden hamsters (11) and in the hindlimb of rabbits (26) that were independent of muscle atrophy. In contrast, others have reported there was little or no acute decrease in the force-generating capacity of the gastrocnemius in rats administered endotoxin (10) or the proinflammatory agent zymosan which produces a sterile peritonitis (27). Our data are more consistent with the findings of these latter studies as there was no acute effect of CLP-induced sepsis on EDL contractility, with regards to specific twitch force and specific tetanic force generation as well as absolute force generated with increasing frequency of stimulation. As a consequence, derived parameters such as time to peak tension and half-relaxation time also did not differ from time-matched control values. It is important to note that in our study there was no difference in muscle mass between control and septic rats during the acute phase as both groups were fasted after surgery to compensate for sepsis-induced cachexia. In contrast, sepsis acutely decreased the ability of the EDL to generate force after a fatigue-inducing stimulus and tended to blunt the post-fatigue recovery profile, although the later decrement did not achieve statistical significance. These data suggest muscle fatigue is an early occurring endpoint of muscle dysfunction that has not been previously reported.

The post-sepsis recovery phase in our study was characterized by food intake that did not differ from basal levels seen in mice without surgery, and where the early sepsis-induced decreases in body weight as well as the inflammatory cytokines in blood and muscle have returned to presurgery/sepsis levels (2, 3). As opposed to acute sepsis, this post-sepsis recovery phase was characterized by a consistent reduction in specific twitch and tetanic force as well as a decrease in the force–frequency relationship. Moreover, the convalescence phase was also characterized by a decreased time to muscle fatigue. As all of these parameters have been normalized for muscle mass, they appear independent of any sepsis-induced muscle atrophy that has not been fully compensated for during recovery. Our data are consistent with a recent report (15) that also demonstrated a reduction in specific force in response to increasing stimulus frequency in surviving mice studied 2 weeks after induction of sepsis. We note that although both the method of producing sepsis (CLP vs cecal slurry) and the age of the mice (4 vs 16 month) differed between our study and this earlier work, respectively, the general conclusion reached is similar. However, our current data expand and highlight the scope of the contractility defects observed during the post-sepsis recovery phase and demonstrate an inability to fully recover normal contractility after a fatiguing stimuli. Although not as sophisticated as the indices of muscle function performed in these prior studies, multiple preclinical and clinical data have indicated that grip strength is decreased for weeks to months after the recovery from sepsis and inflammatory states (28–30), suggesting the observed contractile defects have translational significance.

Contractile dysfunction can result from alterations in sarcomeric structure and the relative abundance of various myofibrillar proteins that impart specific twitch characteristics (31). However, there are surprisingly few reports on sepsis-induced changes in the relative abundance of specific myofibrillar proteins in skeletal muscle per se, as opposed to cardiac muscle or diaphragm. In this regard, Moarbes et al. (32) recently reported time-dependent changes in myofibrillar proteins in skeletal muscle from 1 to 4 days post-CLP. This work demonstrated marked (>50%) decreases in total myosin heavy chain as well as myosin light chain 1/3f in the tibialis anterior (TA) muscle at 1, 2, and 4 days after induction of sepsis. This group also saw an early (1–2 days) decrease in tropomyosin as well as troponin T and C (but not troponin I) that returned to basal levels by day 4 (32). However, in our current study, we were unable to confirm these previous observations as we failed to detect a significant change in any of the abovementioned thick or thin filaments at 1 day post-CLP. The apparent discrepancy between these two studies reporting changes in the acute phase of sepsis may be related to a differential response between the TA and EDL muscles and/or to differences in experimental design where control mice were ad libitum fed and food intake was not matched to that of septic mice.

Our data collected during the convalescence phase are more consistent with the sepsis-induced changes in thick and thin filament abundance at 4 days post-CLP described by Moarbes et al. (32). That is, both studies showed a reduction in isoforms for myosin heavy and light chain proteins, tropomyosin and troponin T. In addition, we extended these observations by showing that α-sarcomeric actin and troponin I are also decreased during recovery. Furthermore, we now report that during sepsis recovery there are sustained decreases in α-actinin-3 (fast) and myomesin-2 which are representative proteins in the Z-disc and M-band, respectively, but no change in the intermediate filament proteins desmin and vimentin. It is not altogether unexpected that our results should differ from that of Moarbes et al. (32) as we studied animals during the post-sepsis recovery period while the earlier study used a more chronic sepsis model in which CLP mice were likely still septic at the time of study as the source of sepsis, the punctured cecum, was not removed and they were not treated with antibiotics. This assertion is further supported by the continued reduction in whole-body oxygen consumption and locomotor activity in septic mice at the 4 day time point (32). Collectively, our data on myofibrillar protein abundance is consistent with RNA sequencing data of muscle from critically ill patients who were mechanically ventilated and immobilized for 2 weeks that demonstrated a downregulation of multiple genes central to muscle contraction, including myosin light chain, actinin-3, myomesin-2, tropomyosin, and troponin T (33). Finally, it is important to note that during sepsis-recovery the relative synthetic rate of myofibrillar proteins was not increased above control levels as was seen for the synthesis of sarcoplasmic proteins. Hence, the inability of skeletal muscle to fully replete myofibrillar proteins appears be mediated by two distinct mechanisms: the blunted recovery of myofibrillar protein synthesis (not previously described) as well as an increase in the protein degradation via the ubiquitin-proteasome pathway during recovery that was previously reported (2).

In addition to changes in the content of myofibrillar proteins, muscle contraction is dependent on the availability of high-energy phosphates to maintain optimal function (34). Our current data suggest that steady-state levels of ATP and PCr are well maintained during the post-sepsis recovery period and there were no detectable decreases in the abundance of the proteins constituting the OXPHOS complex. Our data during acute sepsis are largely confirmatory of previous reports (35, 36). In contrast, our results from the recovery phase do not confirm previous studies showing a reduction in mitochondria bioenergetics (e.g., maximal ADP phosphorylation rate and Complex I-driven electron transport) when assessed in surviving mice 2 weeks after induction of peritonitis (15).

In summary, our results unequivocally reveal a constellation of contractile defects in skeletal muscle during post-sepsis recovery, but not the acute phase of sepsis, that are evidenced by a reduction in basal isometric and tetanic specific force generation as well as an increased fatigability. These defects were intrinsic and not solely attributable to the loss of muscle mass. Finally, contractile defects occurred concomitantly with alterations in myofibrillar protein synthesis and composition, as evidenced by the reduced abundance of several thick and thin filament proteins as well as proteins present in the M-band and Z-disc, but under conditions where the steady-state high-energy phosphate content is unchanged. Collectively, our current data in combination with published work highlight the precept that the recovery phase post-sepsis represents a distinct stage—separate from the acute and even chronic phases of sepsis— that requires further study to better understand the underlying molecular mechanisms to limit deconditioning and speed recovery.


The authors thank Anne Pruznak for her excellent technical assistance.


1. Frost RA, Lang CH. mTor signaling in skeletal muscle during sepsis and inflammation: where does it all go wrong? Physiology (Bethesda) 2011; 26 (2):83–96.
2. Crowell KT, Soybel DI, Lang CH. Restorative mechanisms regulating protein balance in skeletal muscle during recovery from sepsis. Shock 2017; 47 (4):463–473.
3. Crowell KT, Soybel DI, Lang CH. Inability to replete white adipose tissue during recovery phase of sepsis is associated with increased autophagy, apoptosis, and proteasome activity. Am J Physiol Regul Integr Comp Physiol 2017; 312 (3):R388–R399.
4. Friedrich O, Reid MB, Van den Berghe G, Vanhorebeek I, Hermans G, Rich MM, Larsson L. The sick and the weak: neuropathies/myopathies in the critically ill. Physiol Rev 2015; 95 (3):1025–1109.
5. Chao T, Herndon DN, Porter C, Chondronikola M, Chaidemenou A, Abdelrahman DR, Bohanon FJ, Andersen C, Sidossis LS. Skeletal muscle protein breakdown remains elevated in pediatric burn survivors up to one-year post-injury. Shock 2015; 44 (5):397–401.
6. Cree MG, Fram RY, Barr D, Chinkes D, Wolfe RR, Herndon DN. Insulin resistance, secretion and breakdown are increased 9 months following severe burn injury. Burns 2009; 35 (1):63–69.
7. Branski LK, Herndon DN, Pereira C, Mlcak RP, Celis MM, Lee JO, Sanford AP, Norbury WB, Zhang XJ, Jeschke MG. Longitudinal assessment of Integra in primary burn management: a randomized pediatric clinical trial. Crit Care Med 2007; 35 (11):2615–2623.
8. Callahan LA, Supinski GS. Sepsis-induced myopathy. Crit Care Med 2009; 37: (10 suppl): S354–S367.
9. McNicol FJ, Hoyland JA, Cooper RG, Carlson GL. Skeletal muscle contractile properties and proinflammatory cytokine gene expression in human endotoxaemia. Br J Surg 2010; 97 (3):434–442.
10. Giannesini B, Izquierdo M, Dalmasso C, Le Fur Y, Cozzone PJ, Verleye M, Le Guern ME, Gillardin JM, Bendahan D. Endotoxemia does not limit energy supply in exercising rat skeletal muscle. Muscle Nerve 2008; 37 (4):496–504.
11. Supinski G, Nethery D, Stofan D, DiMarco A. Comparison of the effects of endotoxin on limb, respiratory, and cardiac muscles. J Appl Physiol 1996; 81 (3):1370–1378.
12. Reid MB, Lannergren J, Westerblad H. Respiratory and limb muscle weakness induced by tumor necrosis factor-alpha: involvement of muscle myofilaments. Am J Respir Crit Care Med 2002; 166 (4):479–484.
13. Rossignol B, Gueret G, Pennec JP, Morel J, Rannou F, Giroux-Metges MA, Talarmin H, Gioux M, Arvieux CC. Effects of chronic sepsis on contractile properties of fast twitch muscle in an experimental model of critical illness neuromyopathy in the rat. Crit Care Med 2008; 36 (6):1855–1863.
14. Lanone S, Mebazaa A, Heymes C, Henin D, Poderoso JJ, Panis Y, Zedda C, Billiar T, Payen D, Aubier M, et al. Muscular contractile failure in septic patients: role of the inducible nitric oxide synthase pathway. Am J Respir Crit Care Med 2000; 162 (6):2308–2315.
15. Owen AM, Patel SP, Smith JD, Balasuriya BK, Mori SF, Hawk GS, Stromberg AJ, Kuriyama N, Kaneki M, Rabchevsky AG, et al. Chronic muscle weakness and mitochondrial dysfunction in the absence of sustained atrophy in a preclinical sepsis model. Elife 2019; 8:e49920.
16. Crowell KT, Moreno S, Steiner JL, Coleman CS, Soybel DI, Lang CH. temporally distinct regulation of pathways contributing to cardiac proteostasis during the acute and recovery phases of sepsis. Shock 2018; 50 (6):616–626.
17. Reynolds P, Wall P, van Griensven M, McConnell K, Lang C, Buchman T. Shock supports the use of animal research reporting guidelines. Shock 2012; 38 (1):1–3.
18. Oishi PE, Cholsiripunlert S, Gong W, Baker AJ, Bernstein HS. Myo-mechanical analysis of isolated skeletal muscle. J Vis Exp 2011; 48:e2582.
19. Moorwood C, Liu M, Tian Z, Barton ER. Isometric and eccentric force generation assessment of skeletal muscles isolated from murine models of muscular dystrophies. J Vis Exp 2013; (71):e50036.
20. Crowell KT, Laufenberg LJ, Lang CH. Chronic alcohol consumption, but not acute intoxication, decreases in vitro skeletal muscle contractile function. Alcohol Clin Exp Res 2019; 43 (10):2090–2099.
21. Luedeke JD, McCall RD, Dillaman RM, Kinsey ST. Properties of slow- and fast-twitch skeletal muscle from mice with an inherited capacity for hypoxic exercise. Comp Biochem Physiol A Mol Integr Physiol 2004; 138 (3):373–382.
22. Goodman CA, Hornberger TA. Measuring protein synthesis with SUnSET: a valid alternative to traditional techniques? Exerc Sport Sci Rev 2013; 41 (2):107–115.
23. Steiner JL, Crowell KT, Kimball SR, Lang CH. Disruption of REDD1 gene ameliorates sepsis-induced decrease in mTORC1 signaling but has divergent effects on proteolytic signaling in skeletal muscle. Am J Physiol Endocrinol Metab 2015; 309 (12):E981–E994.
24. Vary TC, Lang CH. Assessing effects of alcohol consumption on protein synthesis in striated muscles. Methods Mol Biol 2008; 447:343–355.
25. Vary TC, Lynch CJ, Lang CH. Effects of chronic alcohol consumption on regulation of myocardial protein synthesis. Am J Physiol Heart Circ Physiol 2001; 281 (3):H1242–H1251.
26. Gutierrez G, Hurtado FJ, Fernandez E. Inhibitory effect of Escherichia coli endotoxin on skeletal muscle contractility. Crit Care Med 1995; 23 (2):308–315.
27. Minnaard R, Drost MR, Wagenmakers AJ, van Kranenburg GP, Kuipers H, Hesselink MK. Skeletal Muscle wasting and contractile performance in septic rats. Muscle Nerve 2005; 31 (3):339–348.
28. Hill NE, Murphy KG, Saeed S, Phadke R, Chambers D, Wilson DR, Brett SJ, Singer M. Impact of ghrelin on body composition and muscle function in a long-term rodent model of critical illness. PLoS One 2017; 12 (8):e0182659.
29. Custodero C, Wu Q, Ghita GL, Anton SD, Brakenridge SC, Brumback BA, Efron PA, Gardner AK, Leeuwenburgh C, Moldawer LL, et al. Prognostic value of NT-proBNP levels in the acute phase of sepsis on lower long-term physical function and muscle strength in sepsis survivors. Crit Care 2019; 23 (1):230.
30. Gardner AK, Ghita GL, Wang Z, Ozrazgat-Baslanti T, Raymond SL, Mankowski RT, Brumback BA, Efron PA, Bihorac A, Moore FA, et al. The development of chronic critical illness determines physical function, quality of life, and long-term survival among early survivors of sepsis in surgical ICUs. Crit Care Med 2019; 47 (4):566–573.
31. Henderson CA, Gomez CG, Novak SM, Mi-Mi L, Gregorio CC. Overview of the muscle cytoskeleton. Compr Physiol 2017; 7 (3):891–944.
32. Moarbes V, Mayaki D, Huck L, Leblanc P, Vassilakopoulos T, Petrof BJ, Hussain SNA. Differential regulation of myofibrillar proteins in skeletal muscles of septic mice. Physiol Rep 2019; 7 (20):e14248.
33. Llano-Diez M, Fury W, Okamoto H, Bai Y, Gromada J, Larsson L. RNA-sequencing reveals altered skeletal muscle contraction, E3 ligases, autophagy, apoptosis, and chaperone expression in patients with critical illness myopathy. Skelet Muscle 2019; 9 (1):9.
34. Baker JS, McCormick MC, Robergs RA. Interaction among skeletal muscle metabolic energy systems during intense exercise. J Nutr Metab 2010; 2010:905612.
35. Peruchi BB, Petronilho F, Rojas HA, Constantino L, Mina F, Vuolo F, Cardoso MR, Goncalves CL, Rezin GT, Streck EL, et al. Skeletal muscle electron transport chain dysfunction after sepsis in rats. J Surg Res 2011; 167 (2):e333–e338.
36. Brealey D, Karyampudi S, Jacques TS, Novelli M, Stidwill R, Taylor V, Smolenski RT, Singer M. Mitochondrial dysfunction in a long-term rodent model of sepsis and organ failure. Am J Physiol Regul Integr Comp Physiol 2004; 286 (3):R491–R497.

Contractile filaments; convalescence; muscle contraction; myosin; protein synthesis

Copyright © 2020 by the Shock Society