OPN isoforms differentially mediate pro-inflammatory cytokine production from macrophages and myoblasts Open Access
Downloadable ContentDownload PDF
Background: Osteopontin (OPN) is a pleiotropic extracellular matrix protein that promotes immune cell chemotaxis and is associated with acute and chronic pro-inflammatory states. The ability of OPN to act as a pro-inflammatory protein has been inconsistently reported in the literature, as it has been documented to also function as an anti-inflammatory protein. Genetic polymorphisms in the OPN gene are associated with muscle remodeling in both healthy volunteers and Duchenne muscular dystrophy patients. Therefore, understanding the role of the OPN in muscle remodeling would be significant. In humans there are three spliced isoforms (OPN -a, -b, and -c) that are suggested to differentially regulate immune cell chemotaxis and tumor cell behaviors. To our knowledge, the effects of these spliced isoforms on inflammatory signaling in macrophages and myogenic cells has not been investigated outside of a tumor cell specific context. We thus sought to define the pro-inflammatory activities of human isoforms on cytokine secretion from human macrophages and myoblasts.Methods: OPN isoform precursor transcripts where quantified with specific RT-PCR primers in canine (CXMD) and human (DMD) dystrophin-deficient skeletal muscle. Primary human macrophages and immortalized human myoblasts were stimulated with E. coli derived recombinant human OPN spliced isoforms (OPN -a, -b, and -c), or stromal fibroblast derived OPN-a proteins containing either an intact or mutated RGD integrin-binding domain (OPNa-RGD or ΔRGD→KAE [OPNa-KAE], respectively). Changes in inflammatory signaling were assessed via Nanostring expression profiling (transcriptional responses) and via ELISA, FACS or Luminex (protein secretion). To determine the mechanisms of OPN-a mediated signaling, macrophages were pretreated with TAK-242, a small molecule inhibitor of TLR4 signaling. Phagocytosis was assessed by monocyte uptake of fluorescently conjugated E. coli particles via FACS and luminometry. Results: OPN was very highly expressed in remodeling (dystrophic) skeletal muscle, with human DMD muscle showing all three isoforms (OPN-a>OPN-b>OPN-c), whereas dog CXMD dystrophic muscle showed two isoforms (OPN-a=OPN-c). Further, in DMD and BMD muscle biopsies, when accounting for OPN-c expression, the expression of other forms of OPN was more closely associated with pathological severity. Human OPN-a induced pro-inflammatory cytokine production from both primary macrophages and myoblasts (IL-1β, CCL5, IL-6, TNFα in macrophages; IL-6, IL-8, and CCL5 in myoblasts). Comparison of human and murine macrophages showed that murine macrophages require IFNγ priming to show potent response to OPN. Human OPN-b and OPN-c isoforms were considerably less pro-inflammatory than OPN-a. OPN-a induction of cytokines was RGD- and TLR4- dependent. All OPN isoforms increased macrophage phagocytosis in an RGD-independent manner. OPN did not affect TGFβ production from macrophages or myoblasts. Conclusions: OPN isoforms differentially induce pro-inflammatory cytokine production in an RGD- and TLR4- dependent manner from macrophages and myoblasts, with the effects of OPN-a>OPN-b>OPN-c. All human isoforms induced macrophage phagocytosis in an RGD-independent manner. Our data suggests that human OPN isoforms differentially alter the balance of pro-inflammatory state in remodeling (dystrophic) muscle.