Pharmacological elevation of sphingosine-1-phosphate by S1P lyase inhibition accelerates bone regeneration after post-traumatic osteomyelitis

Abstract

Posttraumatic osteomyelitis and the ensuing bone defects are a debilitating complication after open fractures with little therapeutic options. We have recently identified potent osteoanabolic effects of sphingosine-1-phosphate (S1P) signalling and have now tested whether it may beneficially affect bone regeneration after infection. We employed pharmacological S1P lyase inhibition by 4-deoxypyrodoxin (DOP) to raise S1P levels in vivo in an unicortical long bone defect model of posttraumatic osteomyelitis in mice. In a translational approach, human bone specimens of clinical osteomyelitis patients were treated in organ culture in vitro with DOP. Bone regeneration was assessed by μCT, histomorphometry, immunohistology and gene expression analysis. The role of S1P receptors was addressed using S1PR3 deficient mice. Here, we present data that DOP treatment markedly enhanced osteogenesis in posttraumatic osteomyelitis. This was accompanied by greatly improved osteoblastogenesis and enhanced angiogenesis in the callus accompanied by osteoclast-mediated bone remodelling. We also identified the target of increased S1P to be the S1PR3 as S1PR3^-/- mice showed no improvement of bone regeneration by DOP. In the human bone explants, bone mass significantly increased along with enhanced osteoblastogenesis and angiogenesis. Our data suggest that enhancement of S1P/S1PR3 signalling may be a promising therapeutic target for bone regeneration in posttraumatic osteomyelitis.

Keywords: bone regeneration; osteomyelitis; sphingosin-1-phosphate.

FIGURE 1

DOP treatment leads to increased S1P Plasma Levels and formation of callus during bone healing in C57Bl/6‐J mice 2 weeks after a uni‐cortical defect was created. (A) S1P levels in plasma after treatment with 180 mg/L DOP and control animals 2 weeks after a unicortical defect was created measured with mass spectrometry (n = 10 control, n = 9 DOP), and (B) Callus formation (A) and representative images (whitebone, Blue‐Callus) these mice mice analysed using micro‐computer tomography (n = 10 control, n = 13 DOP). Data are presented as mean ± SD and tested with two‐tailed t‐ test.

FIGURE 2

Representative aniline blue staining (A) and immunohistochemistry for (B) Osteocalcin, (C) TRAP and (D) CD31 with histomorphometrical quantification (E–H) in murine tibial defects 2 weeks postoperatively without (Co; n = 10) and with DOP treatment (DOP; n = 13). In all pictures the bony defect is shown with intact cortical bone at the bottom and monocortical defect at the top of each image. Black arrows in TRAP staining indicate stained osteoclasts and white arrows in CD‐31 staining indicate bloodvessels in the defect area. Results are shown as mean ± SEM. Scale bar: 200 μm.

FIGURE 3

Quantitative RT‐PCR analysis of osteoblast related genes Bglap (A), Runx2 (B) and Wnt5a (C); osteogenic genes Spp1 (D), Noggin (E) and Alpl (F); angiogenesis related genes Cd31 (G) and Vegfa (H); osteoclast related genes Tnfsf11 (I), Tnfrsf11b (J) Tgfb1 (K), Mmp9 (L), Nfatc1 (M); adipogenic differentiation genes Pparγ (N), Fabp4 (O) and Glut4 (P), and inflammatory markers Tnf‐alpha (Q), Il6 (R), Il1a (S), and Il1rn (T) (n = 7 control and DOP each). Data are presented as mean ± SEM. p‐value: * < 0.05, ** < 0.01, *** < 0.001.

FIGURE 4

DOP treatment leads to increased formation of callus during bone healing in S1P3+/+ mice 2 weeks after uni‐cortical bone defect was created. This effect is prevented in S1P3−/− receptor KO mice. (A) Callus formation (n = 9/10/10/11) and (B) representative images (white‐ bone, blue‐callus) of WT and KO mice treated with 180 mg/L DOP for 2 weeks and according controls were analysed using micro‐ computer tomography. Data are presented as mean ± SEM and tested with two‐way anova.

FIGURE 5

Stainings for aniline blue staining (A) and immunohistochemistry for (B) TRAP and (C) CD31 and according histomorphometrical quantification (D–F) of S1P3 receptor deficient knockout mice (S1P3−/−) (n = 11) and wildtype control (S1P3+/+) (n = 10) 2 weeks postoperatively without (Co) and with DOP treatment (DOP). Results are shown as mean ± SEM. Scale bar: 200 μm in aniline blue and TRAP staining and 100 μm in CD‐31 staining.

FIGURE 6

(A) Percentage of changes in bone volume of control and 0.2 mM DOP treated human osteomyelitic bone samples after 4 weeks of tissue culture analysed using micro‐computer tomography (n = 5) and immunohistochemistry for (B) Osteocalcin, (C) Runx2 and (D) CD‐31 (E‐G) (n = 7). Results are shown as mean ± SEM. ns, not significant, p‐value: * < 0.05, *** < 0.001. Scale bar: 200 μm in Osteocalcin staining and 100 μm in Runx2 and CD31 staining.