NS-398, a cyclooxygenase-2-specific inhibitor, delays skeletal muscle healing by decreasing regeneration and promoting fibrosis

Abstract

Nonsteroidal anti-inflammatory drugs are often prescribed after muscle injury. However, the effect of nonsteroidal anti-inflammatory drugs on muscle healing remains primarily controversial. To further examine the validity of using these drugs after muscle injury, we investigated the working mechanism of NS-398, a cyclooxygenase-2-specific inhibitor. In vitro experiments showed that NS-398 inhibited the proliferation and maturation of differentiated myogenic precursor cells, suggesting a detrimental effect on skeletal muscle healing. Using a mouse laceration model, we analyzed the in vivo effect of NS-398 on skeletal muscle healing at time points up to 4 weeks after injury. The in vivo results revealed delayed muscle regeneration at early time points after injury in the NS-398-treated mice. Compared to controls, lacerated muscles treated with NS-398 expressed higher levels of transforming growth factor-beta1, which corresponded with increased fibrosis. In addition, transforming growth factor-beta1 co-localized with myostatin, a negative regulator of skeletal muscle growth. We also found reduced neutrophil and macrophage infiltration in treated muscles, indicating that the delayed skeletal muscle healing observed after NS-398 treatment could be influenced by the anti-inflammatory effect of NS-398. Our findings suggest that the use of cyclooxygenase-2-specific inhibitors to treat skeletal muscle injuries warrants caution because they may interfere with muscle healing.

Figures

Figure 1

A: NS-398 (100 μmol/L) significantly inhibited the proliferation of myogenic precursor cells at day 3 and day 4. B and C: Western blot results show that NS-398 (10 μmol/L and 100 μmol/L) decreased the expression of MHC-d and myogenin but not of MyoD. D: The addition of all tested concentrations of NS-398 in vitro severely reduced the expression of prostaglandins by myogenic precursor cells. The asterisks indicate a significant difference (P < 0.05) between the marked groups and the 0 μmol/L group.

Figure 2

H&E staining of muscle sections harvested 7 days (A and B), 14 days (C and D), and 28 days (E and F) after injury are shown. Seven and fourteen days after injury, muscle sections in the control group contained a significantly higher percentage of centronucleated myofibers than did muscle sections in the NS-398-treated groups. Furthermore, the minor axis diameters of the centronucleated myofibers in the control group were significantly larger than those of the centronucleated myofibers in the NS-398-treated groups. However, follow-up analysis of these parameters conducted 28 days after injury revealed no significant difference between the control and treated groups. The asterisks indicate a significant difference (P < 0.05) between the marked groups and the control group. Original magnifications: ×200 (A–F); ×100 (insets).

Figure 3

Immunostaining results of the injured area reveal the expression of MHC-d (red) and MHC-m (red) 3 days (A–D), 5 days (E–H), and 7 days (I–L) after injury. Collagen IV and cell nuclei are stained green and blue, respectively. During this period, the percentage of MHC-d-expressing myofibers in the muscle sections from the control group dropped from 1.42 to 1.11 to 0%, while the percentage of MHC-m-expressing myofibers increased from 0 to 5.34 to 6.65%. During the same period, the percentage of MHC-d-expressing myofibers in the muscle sections from the NS-398-treated group (10 mg-3 days) changed from 3.69 to 5.16 to 4.95%, while the percentage of MHC-m-expressing myofibers increased from 0 to 1.23 to 2.19%. Although we observed an isoform transition from MHC-d to MHC-m in both the NS-398-treated group and the control group during muscle regeneration, the transition was delayed in the NS-398-treated group. The asterisks in M and N indicate a significant difference (P < 0.05) between the control group and the NS-398-treated group (10 mg-3 days) in terms of the MHC-d-expressing myofibers and MHC-m-expressing myofibers. N depicts the densities of the main cell types (cells per unit area) found within the muscle 5 days after injury, including myogenic cells (MHC-d, MHC-m), immune cells (F4/80, CD-11b), and fibroblastic cells (α-smooth muscle actin). Original magnifications, ×200.

Figure 4

Trichrome staining was used to observe fibrosis after injury. Fibrous tissue is stained blue. At one of the earlier time points (14 days after injury), the control group contained less fibrous tissue (less stained area) than did either of the NS-398-treated groups (P < 0.05). The low-dose, short-duration group (5 mg-3 days) contained less fibrous tissue than the high-dose (10 mg-3 days) and longer duration (5 mg-5 days) groups (P < 0.05). However, at the last time point (28 days after injury), no significant differences were observed. The asterisks indicate a significant difference (P < 0.05) between the marked groups and the control group. Original magnifications, ×100.

Figure 5

Immunostaining results for TGF-β1 (red) and myostatin expression (green). Cell nuclei are blue. There was a significantly higher percentage of TGF-β1- or myostatin-expressing myofibers in the muscle sections from the NS-398-treated group than in the muscle sections from the control group 3 days after injury. The expression of myostatin was highly co-localized with the expression of TGF-β1 in both the control (E and F) and the NS-398-treated groups (10 mg-3 days; G and H). The asterisks indicate a significant difference (P < 0.05) between the marked groups and the control group. Original magnifications, ×200.

Figure 6

Immunostaining results for TGF-β1 expression. TGF-β1, collagen IV, and cell nuclei are red (arrows), green, and blue, respectively. Both the control and the NS-398-treated groups (5 mg-3 days) exhibited relatively low TGF-β1 expression 7 days after injury (A and D). However, 10 days and 14 days after injury the control group showed hardly any expression of TGF-β1 (B and C), whereas, relative to the control group, the NS-398-treated groups exhibited significantly higher TGF-β1 expression (E and F). Asterisks in the graph indicate that the TGF-β1-expressing area was significantly less in the control group than in the various NS-398-treated groups (P < 0.05). Original magnifications: ×200 (A–F); ×100 (insets).

Figure 7

Neutrophils (A) and macrophages (B) infiltrated the injury site as early as 12 hours after injury. The percentages of these cells (in terms of the entire muscle cell population) peaked 48 hours after injury. At all time points, the NS-398-treated groups contained lower percentages of inflammatory cells than did the control group, but we observed significant differences (P < 0.05) only at 48 hours after injury for neutrophils and at 24 hours after injury for macrophages.

Figure 8

Proposed mechanism of myogenic differentiation. Different markers distinguish various stages of myogenic differentiation. MyoD and myogenin denote the early and late stages of muscle differentiation, respectively. The detrimental effects of NS-398 on myogenic cells likely occur via inhibition of both cell proliferation and maturation of differentiated myogenic cells.