Syndecan-4-/- mice have smaller muscle fibers, increased Akt/mTOR/S6K1 and Notch/HES-1 pathways, and alterations in extracellular matrix components

Background Extracellular matrix (ECM) remodeling is essential for skeletal muscle development and adaption in response to environmental cues such as exercise and injury. The cell surface proteoglycan syndecan-4 has been reported to be essential for muscle differentiation, but few molecular mechanisms are known. Syndecan-4-/- mice are unable to regenerate damaged muscle, and display deficient satellite cell activation, proliferation, and differentiation. A reduced myofiber basal lamina has also been reported in syndecan-4-/- muscle, indicating possible defects in ECM production. To get a better understanding of the underlying molecular mechanisms, we have here investigated the effects of syndecan-4 genetic ablation on molecules involved in ECM remodeling and muscle growth, both under steady state conditions and in response to exercise. Methods Tibialis anterior (TA) muscles from sedentary and exercised syndecan-4-/- and WT mice were analyzed by immunohistochemistry, real-time PCR and western blotting. Results Compared to WT, we found that syndecan-4-/- mice had reduced body weight, reduced muscle weight, muscle fibers with a smaller cross-sectional area, and reduced expression of myogenic regulatory transcription factors. Sedentary syndecan-4-/- had also increased mRNA levels of syndecan-2, decorin, collagens, fibromodulin, biglycan, and LOX. Some of these latter ECM components were reduced at protein level, suggesting them to be more susceptible to degradation or less efficiently translated when syndecan-4 is absent. At the protein level, TRPC7 was reduced, whereas activation of the Akt/mTOR/S6K1 and Notch/HES-1 pathways were increased. Finally, although exercise induced upregulation of several of these components in WT, a further upregulation of these molecules was not observed in exercised syndecan-4-/- mice. Conclusions Altogether our data suggest an important role of syndecan-4 in muscle development.

6 10 min and one time for 5 min in TBS-T. Blots were developed using ECL Prime (RPN 2232, 2 2 6 GE Healthcare, Il, USA). The chemiluminescence signals were detected by Las-4,000 (GE 2 2 7 Healthcare). Membranes were re-probed after stripping using the Restore Western Blot 2 2 8 Stripping buffer for 5 min at room temperature (21059, Thermo Fisher Scientific, MA, USA).

9
Antibodies and conditions were: anti-pSer235/236-RPS6 (1:1000, 5% BSA, #4858, (1:500, 1% casein, #AB5702, Millipore, Merck, Darmstadt, Germany), anti-LRP6 (1:1000, Cell Signaling), anti-pThr308-Akt (1:500, #5106, 5% BSA, Cell Signaling), anti-Akt (1:500,  All data were expressed as mean ±SEM. Comparisons between two groups were analyzed were included as random effects. The reason for including magnification was that image 2 5 7 generation and fiber identification could be affected by the magnification. Muscle image was 2 5 8 included to separate individual mouse effects from the true fiber area effects. Analyses were 2 5 9 conducted in MATLAB 2015a, The MathWorks, Inc., Natick, Massachusetts, United States. To investigate the skeletal muscle of syndecan-4 -/mice, we analyzed the tibialis anterior (TA) showed no obvious postnatal abnormalities, however, they had a significantly lower body 2 6 7 weight and TA weight compared to WT ( Fig. 1A and 1B, respectively). Closer inspection showed that the syndecan-4 -/muscle fibers had smaller cross-sectional area compared with 2 6 9 those of the WT mice (Fig. 1C). More specifically, a larger fraction of the syndecan-4 -/-2 7 0 muscle fibers scored less than 2000 µm 2 and there were hardly any larger muscle fibers (6000 2 7 1 µm 2 and over) (Fig. 1D). Statistical analyses showed that the average size of the syndecan-4 -/-2 7 2 muscle fibers was 468 µm 2 smaller compared to WT (p=0.0009), which was not accompanied (myoD) and myogenin (this study) were significantly lower in syndecan-4 -/compared to WT 2 7 6 ( Fig. 1F). Consistent with the previous report (20), we did not observe any centrally-nucleated 2 7 7 myofibers in the syndecan-4 -/-(Suppl. Fig. 1A), a characteristic often associated with muscle 2 7 8 disorders (20, 25) and impaired maintenance of MuSC quiescence (11). We did neither 2 7 9 observe any differences in the protein levels of the paired box transcription factor Pax7,  To further investigate the role of syndecan-4, syndecan-4 -/and age-matched WT littermates 2 8 5 were subjected to a two weeks-long treadmill exercise protocol (ET mice) (10-12 weeks at 2 8 6 harvest). Sedentary syndecan-4 -/and WT littermates of same age were used as controls (SED 2 8 7 mice). As shown in Fig. 2A, the syndecan-4 -/-TA muscles were still significantly smaller than  To detect possible differences between syndecan-4 -/and WT at the molecular level, 2 9 0 we measured mRNA and protein levels of different ECM and signaling molecules we and fibromodulin showed a tendency to increase in WT after exercise, whereas myoD, 2 9 7 biglycan, collagen 1, collagen 3 and the collagen cross-linking enzyme lysyl oxidase (LOX)   Notably, the biglycan and LOX protein levels increased in syndecan-4 -/mice in response to  Taken together, the syndecan-4 -/muscle had increased mRNA levels of syndecan-2, 3 1 3 collagen 1 and 3, decorin, fibromodulin, biglycan, and LOX. These three latter ECM 3 1 4 components were reduced at protein level, suggesting that they might be more susceptible to  We next analyzed Cleaved Notch, as Notch signaling associates with increased 3 4 1 syndecan-2 levels (32), and is known to increase upon exercise and be involved in Notch and HES-1 levels were increased in sedentary syndecan-4 -/mice and did not further  Finally, since syndecan-4 has been shown to associate with Wnt (34), calcineurin- signaling (36), we also analyzed proteins in these pathways. Immunoblotting showed that the SED and KO-ET versus WT-SED). However, no differences were detected in dishevelled 3 5 2 (Dvl), β-catenin, frizzled-7, low-density lipoprotein receptor-related protein 6 (LRP6) or  Taken together, our data indicate that the syndecan-4 -/muscle had increased activation 3 5 8 of the Akt/mTOR/S6K1 and Notch-HES-1 pathways and that these pathways were not induced any further by exercise. Except for an also reduced TRPC7 level, few other changes 3 6 0 were observed. In this study we have characterized the TA muscle from syndecan-4 -/mice and analyzed conditions and in response to exercise. Compared to WT, we found that syndecan-4 -/mice 3 6 6 had reduced body weight, reduced muscle weight, muscle fibers with a smaller cross-sectional 3 6 7 area, and reduced expression of myogenic regulatory transcription factors. The syndecan-4-/-3 6 8 mice had also increased activation of the Akt/mTOR/S6K1 and Notch-HES-1 pathways, a 3 6 9 9 reduced TRPC7 level, increased syndecan-2 expression, and altered expression of several 3 7 0 ECM components. These molecular changes were observed under steady state conditions, 3 7 1 and in contrast to WT mice, were mostly not affected further by exercise.

7 2
Closer inspection of the syndecan-4 -/muscle showed that the muscle fibers were on 3 7 3 average smaller than wild type fibers, which probably accounts for the reduced TA weight.

7 4
Consistent with previous findings (20), we found that myoD and myogenin were reduced in  syndecan-3 was up-regulated after long-term exercise (39). However, in another study using 3 8 5 rats, exercise did not change the syndecan-4 level (40). In our study using mice subjected to 3 8 6 two weeks treadmill running, syndecan-2, but not syndecan-4 was increased, suggesting a 3 8 7 potential involvement of syndecan-2 in the muscle response to exercise. Interestingly, we also 3 8 8 observed higher levels of syndecan-2 in syndecan-4 -/compared to WT (illustrated in Fig.   3 8 9 6B). Upregulation of syndecan-2 has been reported as a compensatory mechanism during 3 9 0 cartilage development in syndecan-4 -/-(41). However, syndecan-2 was not higher in the left 3 9 1 ventricle from syndecan-4 -/compared to WT (42), suggesting that compensatory upregulation 3 9 2 of syndecan-2 is tissue dependent. Although the mRNA levels of fibromodulin, decorin, biglycan, collagen 1, collagen 3 3 9 4 and the collagen cross-linking enzyme LOX were increased in syndecan-4 -/mice, the protein 3 9 5 levels of fibromodulin, biglycan and LOX were reduced (illustrated in Fig. 6B). This finding 3 9 6 might suggest that some ECM components are more susceptible to degradation or less  Whether the increased mRNA expression of these ECM components, including collagens and 4 1 5 LOX, represent a compensatory mechanism to improve ECM function and trigger myoblast 4 1 6 differentiation in the syndecan-4 -/muscle, has to be determined in future work. In contrast to 4 1 7 our findings, the syndecan-4 -/heart has rather reduced mRNA levels of collagen 1, collagen 3 4 1 8 and LOX following pressure overload, leading to an impaired cross-linking and collagen 4 1 9 1 0 assembly (19). Our group has also previously shown that syndecan-4 is essential for cardiac 4 2 0 myoblast differentiation (50) and that the syndecan-4 -/hearts do not develop concentric 4 2 1 hypertrophy, but rather left ventricle dilatation and dysfunction following pressure overload 4 2 2 (35). Although a syndecan-4-calcineurin-NFAT pathway appeared essential for these events 4 2 3 in the heart (35, 50, 51), we found no changes in the calcineurin-NFAT pathway in the 4 2 4 syndecan-4 -/-TA muscle. Thus, although syndecan-4 is involved in both differentiation and 4 2 5 cell growth of the heart and skeletal muscle, different pathways seem to be involved.

2 7
The Akt/mTOR pathway is an important signaling pathway promoting muscle growth and activation of the Akt/mTOR/S6K1 pathway in syndecan-4 -/is illustrated in Fig. 6C. Notably, 4 3 7 decorin, which was also increased in syndecan-4 -/-, has been shown to activate Akt  was also higher in exercised compared to sedentary WT mice. Notch is a regulator of 4 4 3 myogenesis and is increased upon physiological stimuli such as exercise (see review (31)).

4
Notch signaling is indispensable for maintaining quiescence and self-renewal of MuSCs (30, in the MuSC, and this interaction has been suggested to be important for Notch cleavage (57).

5 2
Loss of syndecan-3 impairs Notch signaling, alters MuSC homeostasis and leads to 4 5 3 progressively increased myofiber size, which is especially noticed in repeatedly injured and 4 5 4 dystrophic mice (11, 57). Syndecan-2 is reported highly abundant in early-differentiated has to be determined by future experiments (illustrated in Fig. 6D).

6 0
Desmin was the only protein in our analyses that was increased in exercised WT, but showed an impaired performance on endurance tests (58). Whether the syndecan-4 -/mice 4 6 7 also fatigues more easily must be determined in larger endurance tests in future. Conclusively, our data show that syndecan-4 -/mice had a lower body weight, lower TA 4 7 2 muscles weight, smaller muscle fibers, reduced myoD and myogenin expression, changes in 4 7 3 several ECM components, an increased syndecan-2 expression, a reduced TRPC7 level, and 4 7 4 increased activation of the Akt/mTOR/S6K1 and Notch-HES-1 pathways. These molecular Writing the original draft. We would like to thank Anita Kaupang for some of the immunoblot analyses and Per Kristian 5 1 5 Lunde for help with muscle dissection. dishevelled; LRP6: Low-density lipoprotein receptor-related protein 6; LOX: Lysyl oxidase.              o  s  o  m  a  l  p  r  o  t  e  i  n  g  e  n  e  R  P  1  3  2   M  m  0  2  5  2  8  4  6  7  _  g  1   7  5  2