Osteopontin mediates macrophage chemotaxis via α4 and α9 integrins and survival via the α4 integrin

Abstract

Osteopontin (OPN) is highly expressed by macrophages and plays a key role in the pathology of several chronic inflammatory diseases including atherosclerosis and the foreign body reaction. However, the molecular mechanism behind OPN regulation of macrophage functions is not well understood. OPN is a secreted molecule and interacts with several integrins via two domains: the RGD sequence binding to α(v) -containing integrins, and the SLAYGLR sequence binding to α(4) β(1), α(4) β(7), and α(9) β(1) integrins. Here we determined the role of OPN in macrophage survival, chemotaxis, and activation state. For survival studies, OPN treated-bone marrow derived macrophages (BMDMs) were challenged with growth factor withdrawal and neutralizing integrin antibodies. We found that survival in BMDMs is mediated primarily through the α(4) integrin. In chemotaxis studies, we observed that migration to OPN was blocked by neutralizing α(4) and α(9) integrin antibodies. Further, OPN did not affect macrophage activation as measured by IL-12 production. Finally, the relative contributions of the RGD and the SLAYGLR functional domains of OPN to leukocyte recruitment were evaluated in an in vivo model. We generated chimeric mice expressing mutated forms of OPN in myeloid-derived leukocytes, and found that the SLAYGLR functional domain of OPN, but not the RGD, mediates macrophage accumulation in response to thioglycollate-elicited peritonitis. Collectively, these data indicate that α(4) and α(9) integrins interacting with OPN via the SLAYGLR domain play a key role in macrophage biology by regulating migration, survival, and accumulation.

Figure 1.. OPN promotes macrophage survival via an α₄ integrin pathway.

(A) WT BMDMs were pre-incubated with 10 ug/ml of neutralizing α_4, α_V, or α₉ integrin antibody or control rat non-immune IgG for 15 minutes. Cells were then plated on OPN-coated surfaces in serum-free medium for 24 hours. Cells were stained with Hoechst 33258 and percent apoptosis was determined by nuclear fragmentation. Data are presented as mean ± SEM for triplicate wells representative of 3 independent experiments. *p<0.05 vs untreated. (B) WT BMDMs were pre-incubated with 100 uM RGD or RAD peptide for 15 minutes. Cells were then plated on OPN-coated surfaces in serum-free RPMI for 24 hours and percent apoptosis was determined. Data are presented as mean ± SEM.

Figure 2.. OPN-induced macrophage migration is mediated by integrins α₄ and α₉.

Migration of WT BMDMs to OPN was determined using a Transwell migration assay. (A) BMDMs were pre-incubated with 1 ug/ml α₄ integrin neutralizing antibody or control rat IgG prior to Transwell migration assay to OPN. (B) BMDMs were pre-incubated with anti-integrin α₉ antibody or hamster IgG isotype control prior to assessing migration to OPN in a Transwell migration assay. (C) BMDMs were incubated with α_V integrin neutralizing antibody prior to assessing migration to OPN. (D) BMDMs were pre-incubated with RGD peptide or RAD control peptide prior to migration assay. Data expressed as cell number per high power field (HPF). Data are presented as mean ± SEM of triplicate wells representative of 3 independent experiments. *p<0.05 vs basal.

Figure 3.. OPN does not affect IL-12 production in resident peritoneal macrophages.

Resident peritoneal macrophages were stimulated with IFN-γ + LPS, 5 nM OPN, or 100 nM OPN and the concentration of IL-12p70 was determined by ELISA after 24 hours (black bars) or 48 hours (gray bars) of treatment. ND: not detectable. Data presented as mean ± SD of triplicate wells. Data from a single experiment.

Figure 4.. The SLAYGLR domain of OPN contributes to leukocyte recruitment to the peritoneal cavity in response to thioglycollate.

Sterile peritonitis was elicited in OPN hematopoeitic chimeric mice by the injection of thioglycollate intraperitoneally. After 72 hours, peritoneal leukocytes were harvested by peritoneal lavage. (A) Cell concentration in the peritoneal lavage fluid was determined by manual counting with a hemocytometer. (B) Monocytes recruited to the peritoneal cavity in response to thioglycollate elicitation was determined by flow cytometry by staining for CD115. The number of CD115 positive cells was determined in WT and OPN-null chimeric mice expressing eGFP or OPN from a macrophage-selective retroviral vector. Additionally, monocyte recruitment was assessed in chimeric mice in which the RGD domain (RAD), the SLAYGLR domain (SLAAGLR), or both (R/S) had been mutationally inactivated. Data are expressed as mean ± SEM of between 5–7 individual animals. *p<0.05 versus WT.