Human mesenchymal stem cells stimulated by TNF-α, LPS, or hypoxia produce growth factors by an NFκB- but not JNK-dependent mechanism

Paul R. Crisostomo, Yue Wang, Troy A. Markel, Meijing Wang, Tim Lahm, Daniel R. Meldrum

Abstract

Understanding the mechanisms by which adult stem cells produce growth factors may represent an important way to optimize their beneficial paracrine and autocrine effects. Components of the wound milieu may stimulate growth factor production to promote stem cell-mediated repair. We hypothesized that tumor necrosis factor-α (TNF-α), endotoxin (LPS), or hypoxia may activate human mesenchymal stem cells (MSCs) to increase release of vascular endothelial growth factor (VEGF), fibroblast growth factor 2 (FGF2), insulin-like growth factor 1 (IGF-1), or hepatocyte growth factor (HGF) and that nuclear factor-κB (NFκB), c-Jun NH2-terminal kinase (JNK), and extracellular signal-regulated kinase (ERK) mediates growth factor production from human MSCs. To study this, human MSCs were harvested, passaged, divided into four groups (100,000 cells, triplicates) and treated as follows: 1) with vehicle; 2) with stimulant alone [24 h LPS (200 ng/ml), 24 h TNF-α (50 ng/ml), or 24 h hypoxia (1% O2)]; 3) with inhibitor alone [NFκB (PDTC, 1 mM), JNK (TI-JIP, 10 μM), or ERK (ERK Inhibitor II, 25 μM)]; and 4) with stimulant and the various inhibitors. After 24 h incubation, MSC activation was determined by measuring supernatants for VEGF, FGF2, IGF-1, or HGF (ELISA). TNF-α, LPS, and hypoxia significantly increased human MSC VEGF, FGF2, HGF, and IGF-1 production versus controls. Stem cells exposed to injury demonstrated increased activation of NFκB, ERK, and JNK. VEGF, FGF2, and HGF expression was significantly reduced by NFκB inhibition (50% decrease) but not ERK or JNK inhibition. Moreover, ERK, JNK, and NFκB inhibitor alone did not activate MSC VEGF expression over controls. Various stressors activate human MSCs to increase VEGF, FGF2, HGF, and IGF-1 expression, which depends on an NFkB mechanism.

  • vascular endothelial growth factor
  • fibroblast growth factor 2
  • hepatocyte growth factor
  • insulin-like growth factor 1
  • mitogen-activated protein kinase

more than a decade of research in animal models suggests that stem cell treatment restores function, remodels injured tissue, and reverses cellular damage (2, 8, 25). Recent phase I and II clinical trials also indicate that stem cell treatment is safe, practical, and may repair damaged tissue (27). Despite this rapid translation to the bedside, the mechanism of stem cell reparative effects is still not well characterized. It was initially hypothesized that immature stem cells differentiated into the phenotype of injured tissue, repopulated the diseased organ with healthy cells, and subsequently improved function (10). However, the levels of engraftment and survival have been consistently <5%, too few to be therapeutically relevant (32). Furthermore, acute stem cell mediate improvement within days or even hours precludes differentiation as a cause (5, 36). Importantly, much of the functional improvement and attenuation of injury afforded by stem cells can be replicated by cell-free conditioned media derived from stem cells (23). Thus, rather than differentiating into target tissue, stem cells may mediate their beneficial effects by complex paracrine actions.

We and others (7, 20, 28, 35) have previously demonstrated that stem cells orchestrate the repair process by secreting a large number of angiogenic growth factors, anti-apoptotic factors, and anti-inflammatory cytokines. Although the particular growth factors contributing to these reparative effects remain to be defined, angiogenic factors may include vascular endothelial growth factor (VEGF), hepatocyte growth factor (HGF), and basic fibroblast growth factor (FGF2). Stem cells have also been shown to release insulin-like growth factor-1 (IGF-1) in response to injury (35), which when transplanted into ischemic tissue may subsequently activate resident stem cells and mobilize circulating progenitor cells. We have previously demonstrated that stem cells increase the release of paracrine factors by a p38 mitogen-activated protein kinase (MAPK) and STAT3-dependent mechanism in response to tumor necrosis factor-α (TNF-α) (35, 37). It remains unknown whether stem cells increase paracrine factor release in response to other injuries, and if so, whether this depends on activation of other members of the MAPK family or the transcriptional factor nuclear factor-κB (NFκB) (15). Delineating these mechanistic pathways may allow one to better target injured tissue and promote stem cell-mediated repair (11, 13).

Stem cell paracrine effects have been observed in hematopoietic stem cells (HSCs) and mesenchymal stem cells (MSCs). However, MSCs have advantages over the well-characterized HSC population, including availability from small aspirates of donor bone marrow, ease of expansion in an in vitro cell culture, simple isolation via plastic adherence, and an innate ability to evade rejection (24). We hypothesize that TNF-α, endotoxin (LPS), or hypoxia activates human mesenchymal stem cells (MSCs) to increase release of VEGF, FGF2, IGF-1, or HGF, and that NFκB, c-Jun NH2-terminal kinase (JNK), and extracellular signal-regulated kinase (ERK) mediate paracrine growth factor expression in human MSCs.

MATERIALS AND METHODS

Mesenchymal stem cells.

Poietics human mesenchymal stem cell were purchased from Cambrex BioScience (Walkersville, MD). Mesenchymal stem cells were harvested and cultured from normal human bone marrow. Thawing of cells and initiation of culture process were performed based on the manufacturer's instructions. Human MSCs (hMSCs) were plated in T-225 tissue culture flasks (Corning, Coring, NY) and cultured with Mesenchymal Stem Cell Basal Medium (Cambrex BioScience) containing 10% fetal bovine serum (FBS), Mesenchymal Cell Growth Supplement (Cambrex BioScience), 4 mM l-glutamine, and penicillin-streptomycin at 37°C, 5% CO2 and 90% humidity. Medium was changed every 3 days.

Experimental groups.

After cells were 70% confluent, hMSCs were plated in 12-well plates (Corning) in a concentration of 0.1 × 106 cells·well−1·ml−1. Cells were divided into the following groups: 1) without intervention; 2) with stimulant alone [24 h LPS (200 ng/ml), 24 h TNF-α (50 ng/ml), or 24 h hypoxia (1.0% O2)]; 3) with inhibitor alone [NFκB (PDTC 1 mM), JNK (TI-JIP10 μM), or ERK (ERK Inhibitor II 25 μM)]; and 4) with stimulant and inhibitor. Inhibitors were obtained from Calbiochem (San Diego, CA). After 24 h incubation, supernatants were harvested for VEGF, FGF2, HGF, and IGF-1 assay. With the use of a Nikon TE2000U microscope at ×200 magnification, cell morpohology and cell viability were assessed using fluorescein diacetate and propiodium iodide staining (Sigma Aldrich, St. Louis, MO).

VEGF, FGF2, IGF-1 HGF, and NFkB ELISA.

Production of VEGF, FGF2, IGF-1, and HGF from hMSCs were determined by enzyme-linked immunosorbent assay (ELISA) using a commercially available ELISA kit (R&D Systems, Minneapolis, MN and BD Biosciences, San Diego, CA). Activation of NFκB p65 was determined by ELISA using a commercially available ELISA kit (Active Motif, Carlsbad, CA). ELISA was performed according to the manufacturer's instructions. All samples and standards were measured in duplicate.

Protein isolation and Western blot analysis.

Western blot analysis was performed to measure the ERK and JNK kinase pathways. Cells were collected in cold buffer containing 20 mM Tris (pH 7.5), 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM β-glycerophosphate, 1 mM Na3VO4, 1 μg/ml leupeptin, 1 mM PMSF, and centrifuged at 12,000 rpm for 10 min. The protein extracts (10 μg/lane) were electrophoresed on a 12% Tris·HCl gel from Bio-Rad and transferred to a nitrocellulose membrane, which was stained by naphthol blue-black to confirm equal protein loading. The membranes were incubated in 5% dry milk for 1 h and then incubated with the following primary antibodies: ERK antibody, phospho-ERK (Thr202/Tyr204) antibody, JNK antibody, and phospho-JNK (Thr183/Tyr185) antibody (Cell Signaling Technology, Beverly, MA), followed by incubation with horseradish peroxidase-conjugated goat anti-rabbit IgG secondary antibody and detection using supersignal west picostable peroxide solution (Pierce, Rockford, IL).

Presentation of data and statistical analysis.

All reported values are means ± SE (n = 3/group). Data were compared using Student's t-test. A two-tailed probability value of <0.05 was considered statistically significant.

RESULTS

Human mesenchymal stem cells morphology.

Mesenchymal stem cells viewed under a Nikon TE2000U Microscope at ×200 magnification demonstrated no differences in morphology between control and experimental groups (Fig. 1, AD). Fluorescent staining also demonstrated no differences in viable (green) or nonviable (red) cells between control and experimental groups (Fig. 1, EH).

Fig. 1.

Representative human mesenchymal stem cells (MSCs) viewed under Nikon TE2000U microscope at ×200 magnification. No differences in morphology were observed between control and experimental groups. A: unstimulated MSCs; B: tumor necrosis factor-α (TNF-α) stimulated; C: LPS stimulated; D: hypoxia stimulated. No differences in viable (green) or nonviable (red) cells were observed between control and experimental groups. E: unstimulated MSCs; F: TNF-α stimulated; G: LPS stimulated; H: hypoxia stimulated.

Effect of TNF-α on hMSCs or hAPCs production of growth factors.

Over a 24-h period under TNF-α, LPS, or hypoxic exposure, hMSCs secreted significantly increased VEGF and FGF2 compared with nonstimulation groups (Fig. 2, A and B). Over a 24-h period under TNF-α or LPS exposure, hMSCs also secreted significantly increased HGF and IGF-1 compared with nonstimulation groups (Fig. 2, C and D).

Fig. 2.

Effect of stimuli on human mesenchymal stem cells (hMSCs) production of paracrine growth factors. A: TNF-α, LPS, or hypoxia significantly increased secretion of VEGF. B: TNF-α, LPS, or hypoxia significantly increased secretion of FGF2. C: TNF-α, LPS, or hypoxia significantly increased secretion of IGF-1. D: TNF or LPS significantly increased secretion of hepatocyte growth factor (HGF). Results are means ± SE; n = 3–5/group, *P < 0.05 vs. control.

NFκB, ERK, and JNK activation.

TNF-α, LPS, and hypoxic stimulation resulted in increased activation of NFκB in hMSCs (Fig. 3A). TNF-α provoked a significant 75% increase, LPS nearly 40% increase, and hypoxia 35% increase in NFκB p65 activation. TNF-α, LPS, and hypoxia also provoked significant activation of ERK (Fig. 3B) and JNK (Fig. 3C).

Fig. 3.

Expression of nuclear factor-κB (NFκB), extracellular signal-regulated kinase (ERK), and c-Jun NH2-terminal kinase (JNK). A: TNF-α, LPS, or hypoxia stimulation increased NFκB p65 activation. B: TNF-α, LPS, or hypoxia stimulation increased ERK activation. C: TNF-α, LPS, or hypoxia stimulation increased JNK activation. Representative immunoblots.

Regulation of growth factor secretion by NFκB and ERK, but not JNK.

A concentration of 1 mM NFκB inhibitor (PDTC) resulted in decreased production of VEGF, FGF2, and HGF in response to TNF-α, LPS, or hypoxia (Figs. 47). There was no effect of NFκB inhibitor alone on production of growth factors. The concentration of 25 μM ERK inhibitor (ERK Inhibitor II) resulted in no significant change in production of FGF2, HGF, or IGF-1 in response to all stimuli. The concentration of 10 μM JNK inhibitor (TI-JIP) resulted in no significant change in production of VEGF, FGF2, HGF, or IGF-1 in response to all stimuli.

Fig. 4.

Production of vascular endothelial growth factor (VEGF) in hMSC response to TNF-α, LPS, or hypoxia with or without NFκB inhibitor, ERK inhibitor, or JNK inhibitor. Results are means ± SE; n = 3–5/group, *P < 0.05 vs. control.

Fig. 5.

Production of fibroblast growth factor 2 (FGF2) in hMSC response to TNF-α, LPS, or hypoxia with or without NFκB inhibitor, ERK inhibitor, or JNK inhibitor. Results are means ± SE; n = 3–5/group, *P < 0.05 vs. control.

Fig. 6.

Production of insulin-like growth factor 1 (IGF-1) in hMSC response to TNF-α, LPS, or hypoxia with or without NFκB inhibitor, ERK inhibitor, or JNK inhibitor. Results are means ± SE; n = 3–5/group, *P < 0.05 vs. control.

Fig. 7.

Production of HGF in hMSC response to TNF-α, LPS, or hypoxia with or without NFκB inhibitor, ERK inhibitor, or JNK inhibitor. Results are means ± SE; n = 3–5/group, *P < 0.05 vs. control.

DISCUSSION

These results of this study represent the first demonstration that: 1) TNF-α, LPS, or hypoxia stimulates MSC production of VEGF, FGF2, HGF, and IGF-1; 2) NFκB is involved in stress-induced production of VEGF, FGF2, and HGF in MSCs; 3) ERK and JNK are not involved in stress-induced production of paracrine growth factors.

Stem cells transplanted into injured tissue express several paracrine signaling factors, including cytokines, chemokines, and growth factors, which are involved in stem cell-mediated repair. A critically important part of this process may be their ability to improve perfusion and enhance angiogenesis to chronically ischemic tissue. Although the particular local signaling molecules contributing to this neovascular effect remain to be defined, the list most likely includes VEGF and FGF2 (6, 34, 40). VEGF is a strong promoter of angiogenesis. FGF2, a specific member of the FGF signaling family, has been shown to be intimately involved with endothelial cell proliferation and may be a more potent angiogenic factor than VEGF (17). The results of this study demonstrate that MSC in cell culture exposed to both specific stimuli such as TNF-α or LPS as well the heterogeneous stimuli hypoxia significantly increase release of VEGF and FGF2. This increase in angiogenic stem cell factors may improve regional blood flow as well as promote autocrine self survival (12). Increased perfusion due to the stem cell angiogenic growth factor production has also been associated with improved end organ function (19). Furthermore, VEGF overexpressing bone marrow stem cells demonstrate greater protection of injured tissue than controls (9, 38). Thus VEGF and FGF2 may be important paracrine signaling molecules in stem cell-mediated angiogenesis, protection, and survival.

Stem cell beneficial paracrine effects may also include salvage of malfunctioning cell types and activation of nearby circulating and resident stem cells. MSCs may activate an anti-apoptosis signaling system at the infarct border zone, which effectively protects ischemia-threatened cell types from apoptosis. In fact, injection of MSC into a cryo-induced infarct has been shown to reduce myocardial scar width 10 wk later (14). HGF may play an important role in this decrease in cell apoptosis (26). Recent work has also demonstrated the existence of endogenous stem cell-like populations in the circulation as well as the heart, liver, brain, and kidney (1, 3). These resident stem cells may possess growth factor receptors that can be activated to induce their migration, proliferation, and promote restoration of dead tissue and improved function in damaged tissue. HGF and IGF-1 have been shown to activate circulating and resident stem cells tissue (30, 33). The results of our study demonstrate that in response to various injury (TNF-α, LPS, or hypoxia), mesenchymal stem cells significantly increase the release HGF and IGF-1. Further understanding of the mechanisms by which these paracrine growth factors are released may enable us to maximize stem cell paracrine effects when transplanted into ischemic tissue.

It remains unknown by what mechanisms injury stimuli induce MSCs to release growth factors. NFκB is an important rapid acting transcription factor found in all cell types and is involved in cellular responses to stimuli such as stress, cytokines, free radicals, ultraviolet irradiation, and bacterial or viral antigens. In stem cells, recent data suggest that NFkB plays a role in proliferation, migration, and differentiation (39). HGF has been implicated in as an important upstream component of the NFκB proliferation process in neural and hepatic stem cells (41). The results of this study shed further light on the role of NFκB in stem cell-mediated protection. TNF-α, LPS, or hypoxia stimulation resulted in significant NFκB activation. Furthermore, NFκB inhibition (PDTC) significantly decreased the production of VEGF, FGF2, and HGF. Interestingly, umbilical cord blood, which has a higher proliferation rate than adult stem cells, was recently found to have higher expression of NFκB compared with that of adults (22). Thus pharmacological activation of NFκB to enhance growth factor production and proliferation may be a strategy to improve their paracrine effects before cell transplant; however, this remains to be determined.

We have previously shown that the increase in MSCs production of VEGF, HGF, and IGF-1 was associated with the increase of p38 MAPK activation exposure to TNF-α. p38 MAPK inhibitor administration also resulted in a decreased release of growth factors in hMSCs response to TNF-α. MAPKs are critically involved in regulatory signaling pathways that ultimately lead to inflammation (4, 21, 31). Activation of p38 MAPK and JNK is a critical step in the generation of deleterious inflammatory cytokines after ischemia-reperfusion injury, whereas ERK activation has been found to improve functional recovery after ischemia (16, 18, 29, 31). The results of this study demonstrate that although TNF-α, LPS, or hypoxia exposure increases ERK and JNK activation, inhibition of ERK or JNK does not significantly affect stem cell production of VEGF, FGF2, HGF, or IGF-1. Thus it appears that stress-induced stem cell production of paracrine growth factors depends on p38 MAPK activation but does not depend on increased ERK or JNK phosphorylation.

GRANTS

This work was supported in part by National Institutes of Health Grants R01GM-070628, R01HL-085595, K99/R00 HL-0876077, F32HL-085982 and by an American Heart Association (AHA) Grant-in-aid, and AHA Postdoctoral Fellowship 0725663Z.

Footnotes

  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

REFERENCES

View Abstract