Toll-like receptor 2 mediates mesenchymal stem cell-associated myocardial recovery and VEGF production following acute ischemia-reperfusion injury

Abstract

Toll-like receptor 2 (TLR2), a key component of the innate immune system, is linked to inflammation and myocardial dysfunction after ischemia-reperfusion injury (I/R). Treatment of the heart with mesenchymal stem cells (MSCs) is known to improve myocardial recovery after I/R in part by paracrine factors such as VEGF. However, it is unknown whether TLR2 activation on the MSCs affects MSC-mediated myocardial recovery and VEGF production. We hypothesized that the knockout of TLR2 on the MSCs (TLR2KO MSCs) would 1) improve MSC-mediated myocardial recovery and 2) increase myocardial and MSC VEGF release. With the isolated heart perfusion system, Sprague-Dawley rat hearts were subjected to I/R and received one of three intracoronary treatments: vehicle, male wild-type MSCs (MWT MSCs), or TL2KO MSCs. All treatments were performed immediately before ischemia, and heart function was measured continuously. Postreperfusion, heart homogenates were analyzed for myocardial VEGF production. Contrary to our hypothesis, only MWT MSC treatment significantly improved the recovery of left ventricular developed pressure and the maximal positive and negative values of the first derivative of pressure. In addition, VEGF production was greatest in hearts treated with MWT MSCs. To investigate MSC production of VEGF, MSCs were activated with TNF in vitro and the supernatants collected for ELISA. In vitro basal levels of MSC VEGF production were similar. However, with TNF activation, MWT MSCs produced significantly more VEGF, whereas activated TLR2KO MSC production of VEGF was unchanged. Finally, we observed that MWT MSCs proliferated more rapidly than TLR2KO MSCs. These data indicate that TLR2 may be essential to MSC-mediated myocardial recovery and VEGF production.

Fig. 1.

Male wild-type (MWT) and Toll-like receptor 2 knockout (TLR2KO) mesenchymal stem cell (MSC) expression of cell surface markers. Plastic-adherent MSCs isolated from the bone marrow of adult MWT and TLR2KO mice are >95% negative for CD45 (FITC) and CD90 (phycoerythrin), 2 non-MSC markers by flow cytometry (A and B, and C and D). MSCs isolated from MWT and TLR2KO mice expressed both stem cell antigen-1 (Sca-1; E and G) and CD44 (F and H). The shaded areas denote the isotype controls. The intensity of Sca-1 staining (x-axis) was greater in the TLR2KO MSC group compared with the MWT MSC group.

Fig. 2.

Multipotency of MWT and TLR2KO MSCs. MWT and TLR2KO MSCs after incubation with adipogenic or osteogenic differentiation media expressed fatty acid-binding protein 4 or osteopontin markers of adipogenesis and osteogenesis respectively [Northern Lights 577 (NL557)-positive red staining]. Nuclei were counterstained with 4,6-diamidino-2-phenylindole (DAPI). There were no visible vacuoles present in the TLR2KO adipogenesis group. Representative fields at ×200 magnification are shown. (Per APS Ethical Policy on figure reproduction, contrast adjustment was applied to the entire image in Adobe Photoshop because of interface issues with QCapture and Adobe Photoshop software.)

Fig. 3.

Effects of preischemic MSC infusion on myocardial function. Ischemia-reperfusion injury resulted in markedly decreased left ventricular developed pressure (LVDP) in all groups. Postischemic recovery of LVDP was significantly greater from 20 min postischemia to end reperfusion in hearts infused with MWT MSCs compared with vehicle hearts (A). Recovery of maximal positive and negative values of the first derivative of pressure (±dP/dt) was also significantly increased in the MWT MSC-treated hearts compared with the vehicle (B and C). Infusion of TLR2KO MSCs did increase LVDP and ±dP/dt, however, this trend did not reach significance (A–C). The end-diastolic pressure (EDP; D) was not significantly different among the groups. There was no significant difference in the end equilibrium (Eq) measurements of the groups (n = 8 vehicle, 12 MWT, and 13 TLR2KO-infused hearts). *P < 0.02, MWT MSC-treated hearts vs. vehicle hearts; +P < 0.03, MWT vs. TLR2KO MSC-treated hearts (ANOVA with post hoc Holms-Sidak).

Fig. 4.

TLR2 effects on VEGF production. Preischemic treatment of the heart with MWT MSCs increased production of myocardial VEGF following ischemia-reperfusion injury (A). VEGF results are expressed as pg/mg of protein of myocardial homogenate. *P < 0.05, MWT vs. vehicle-treated hearts; n = 6–8 hearts/group. In culture, basal production of VEGF is similar between MWT and TLR2KO MSCs (B). However, with TNF activation, MWT MSCs produced significantly more VEGF (∼2.6 × control), whereas activation of TLR2KO MSCs did not significantly change VEGF production (∼1.3 × control). ***P < 0.001, MWT + TNF vs. MWT; +++P < 0.001, MWT + TNF vs. TLR2KO + TNF (ANOVA with post hoc Holm-Sidak).

Fig. 5.

TLR2 effects on MSC proliferation. TLR2KO MSCs proliferated less rapidly compared with MWT MSCs after 48 h of incubation in complete media as measured by 5-bromo-2′-deoxy-uridine (BrdU) incorporation. Absorbance is measured at a wavelength of 405 nm with a reference wavelength of 490 nm. **P < 0.01, MWT vs. TLR2KO MSCs (Student's t-test).