|Home | About | Journals | Submit | Contact Us | Français|
Jump-starting and subsequently maintaining epidermal and dermal cell migration are essential processes for skin wound healing. These events are often disrupted in nonhealing wounds, causing patient morbidity and even fatality. Currently available treatments are unsatisfactory. To identify novel wound-healing targets, we investigated secreted molecules from transforming growth factor α (TGFα)-stimulated human keratinoytes, which contained strong motogenic, but not mitogenic, activity. Protein purification allowed us to identify the heat shock protein 90α (hsp90α) as the factor fully responsible for the motogenic activity in keratinocyte secretion. TGFα causes rapid membrane translocation and subsequent secretion of hsp90α via the unconventional exosome pathway in the cells. Secreted hsp90α promotes both epidermal and dermal cell migration through the surface receptor LRP-1 (LDL receptor-related protein 1)/CD91. The promotility activity resides in the middle domain plus the charged sequence of hsp90α but is independent of the ATPase activity. Neutralizing the extracellular function of hsp90α blocks TGFα-induced keratinicyte migration. Most intriguingly, unlike the effects of canonical growth factors, the hsp90α signaling overrides the inhibition of TGFβ, an abundant inhibitor of dermal cell migration in skin wounds. This finding provides a long-sought answer to the question of how dermal cells migrate into the wound environment to build new connective tissues and blood vessels. Thus, secreted hsp90α is potentially a new agent for wound healing.
Transforming growth factor α (TGFα) belongs to the epidermal growth factor (EGF) family, which has 10 members in humans. The mature form of TGFα is a 50-amino-acid polypeptide originally isolated from conditioned medium of virally transformed cells and tumor cells (9, 39). Active TGFα shows a heterogeneous molecular mass, ranging from 5 to 20 kDa in sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), due to variable degrees of N- and O-linked glycosylation (46). Although the amount of TGFα in human circulation (plasma) is low or undetectable, it is widely expressed in developing embryos and a number of adult tissues. Thus, TGFα is mainly an autocrine/paracrine growth factor. TGFα elicits all its biological effects by binding to and activating a 170-kDa cell surface tyrosine kinase receptor (EGF receptor [EGFR] or c-erbB1), a member of the EGFR subfamily. This receptor family also includes c-erbB2/HER2/Neu, c-erb3/HER3, and c-erbB4/HER4 (13). A major well-characterized effect of TGFα on cells of the epidermal origin is enhancement of cell proliferation and migration, which often cannot occur simultaneously. In vivo studies showed that increased local concentrations of TGFα or overexpression of EGFR plays an important role in both physiological processes, such as wound healing and hair follicle development, and pathological disorders, such as psoriasis (hyperproliferation of keratinocytes) and skin tumorigenesis (15, 32). However, ablation of TGFα or EGFR genes in mice only resulted in some visible abnormalities in the outer root sheath and hair follicle architecture, likely due to the multiple family members present at both the ligand and receptor levels of TGFα (27, 28). While the EGFR → Grb2/Sos → Ras-GTP → Raf → Mek1 → extracellular signal-regulated kinase 1/extracellular signal-regulated kinase 2 (ERK1/2) pathway is widely accepted as the mechanism promoting cell proliferation, the motility signaling pathway remains poorly understood.
During human skin wound healing, a critical step is the initiation of the resident epidermal and dermal cells at the wound edge to migrate into the wound bed (29, 43). Human keratinocytes (HKCs) laterally migrate across the wound bed from the cut edge to eventually close the wound in the process known as reepithelialization. The dermal cells, including dermal fibroblasts (DFs) and human dermal microvascular endothelial cells (HDMECs), start to move into the wound following the keratinocyte migration, a process whereby these cells deposit matrix proteins, contract and remodel the newly closed wound, and subsequently build new blood vessels. The HKC migration is largely driven by TGFα from human serum (HS) (25) and is unaffected by TGFβ family cytokines copresent in the wound (3). In contrast, the presence of TGFβ blocks both DF and HDMEC migration even in the presence of their growth factors, platelet-derived growth factor BB (PDGF-BB) and vascular endothelial growth factor (2). Therefore, while it is clear why HKC migration jump-starts ahead of the DF and HDMEC migration during wound healing, it has remained a long-time puzzle how DFs and HDMECs move into the wound bed in the presence of TGFβ.
The heat shock protein (hsp) families include chaperon proteins that either are constitutively expressed, such as members of the hsp90 family, or are the product of stress-induced expression, such as members of the hsp70 and hsp27 families. Historically, their function is to interact with and facilitate proper folding and intracellular trafficking of target proteins to maintain cellular homeostasis and to promote cell survival (16). However, recent studies showed that hsp proteins could also be secreted by the cells. The secreted hsp proteins then carry out important extracellular functions, including stimulation of immunological cytokine production, activation of antigen-presenting cells, and anticancer functions (6, 41). We recently showed that hypoxia causes hsp90 secretion, which in turn mediates hypoxia-induced dermal fibroblast migration during wound healing (24). Since hsp proteins lack any signal sequences at the amino terminus, these proteins cannot be secreted via the classical endoplasmic reticulum/Golgi transport pathway. Instead, they are secreted via a discrete population of nanovesicles (30 to 90 nm in diameter) called exosomes (8, 18, 21, 31, 51). The exosome secretion mechanism, therefore, constitutes a potential mode of intercellular communication and opens up new therapeutic and diagnostic strategies (30). In the present study, we provide evidence that TGFα “pushes” hsp90α out of HKCs via the exosome pathway, which in turn promotes migration of both the epidermal and dermal cells through cell surface receptor LRP-1 (LDL receptor-related protein 1)/CD91. The physiological significance of these findings is also discussed.
Melanocytes (MCs) were purchased from Clonetics (San Diego, CA). HKCs were cultured in EpiLife medium with added human keratinocyte growth supplements. MCs were maintained in medium 154 supplemented with human melanocyte growth supplement (Cascade Biologics, Portland, OR). DFs were cultured in Dulbecco's minimal Eagle's medium supplemented with 10% fetal bovine serum. HDMECs were cultured in growth factor-supplemented medium 131 (Cascade Biologics). The fifth to seventh passages of the cultured cells were used in conditioned medium preparation, and the third or fourth passage was used in cell migration assays. HS samples, collected from a variety of donors, were purchased from Sigma-Aldrich (St. Louis, MO). Human recombinant TGFα and TGFβ3 were purchased from R&D Systems (Minneapolis, MN). A HMW gel filtration calibration kit (catalog no. 28-4038-42) was obtained from GE Healthcare (Uppsala, Sweden). Anti-hsp90α antibody (SPA-840) for Western analysis and anti-hsp90α neutralizing antibody (SPS-771 [blocking the target-binding site of hsp90]) were from Stressgen (Victoria, British Columbia, Canada). Recombinant hsp-90α was from Stressgen and produced in our laboratory. The glutathione S-transferase(GST)-human 14-3-3σ construct was a gift from Mong-Hong Lee (University of Texas, Houston). Recombinant autocrine motility factor (AMF)/phosphoglucose isomerase/neuroleukin, brefeldin A (BFA), and dimethyl amiloride (DMA) were purchased from Sigma-Aldrich (St. Louis, MO). Rhodamine-conjugated phalloidin was from Sigma. Rat type I collagen was purchased from BD Biosciences (Bedford, MA). Anti-β-actin antibody and anti-glyceraldehyde-3-phosphate dehydrogenase (anti-GAPDH) antibody were from Cell Signaling Technology.
The colloidal gold migration assay and the in vitro wound-healing (“scratch”) assay were modified as described previously in detail by us (22). To measure DNA synthesis, indicated cells were plated in 24-well tissue culture plates precoated with collagen in triplicate, starved in serum-free media for 24 h, and incubated with either indicated growth factors or conditioned medium of HKC (HKC-CM) for 14 h. Induced DNA synthesis was measured by incorporation of [3H]thymidine into chromosomal DNA, as described previously (7). Data from four independent experiments (n = 4) were averaged and calculated by severalfold induction (with means ± standard deviations determined) in response to corresponding stimulus (P < 0.05).
For protein purification details, see Table S2 in the supplemental material.
Immunostaining cells or human skin with antibodies was performed as described previously (3, 12). The results were analyzed using either a Nikon TE-200U microscope with a green (fluorescein isothiocyanate), blue (DAPI [4′,6′-diamidino-2-phenylindole]), or red (phalloidin) filter or a Nikon Eclipse 80i confocal microscope. Usually, 80 to 120 cells per set of conditions were analyzed.
HKC-CM (10× concentrated) was preincubated with the optimized amount of anti-hsp90α rabbit polyclonal antibody (7 μg immunoglobulin G [IgG]/2 ml of 10× HKC-CM to remove all hsp90α) for 16 h at 4°C. The immune complexes were precipitated by incubation and agitation with protein G-Sepharose (Pharmacia) (25 μl of packed beads) for 2 h. After centrifugation, the hsp90α-free supernatants were removed and used for cell migration assays.
The cDNAs for wild-type (wt) hsp90α and three mutants (E47D, E47A, and D93N) were kindly provided by Ulrich Hartl (Max Planck Institute for Biochemistry, Germany). The coding regions of these cDNAs were subcloned into His-tagged pET15b vector (EMD Biosciences, Inc., San Diego, CA) at BamHI by use of a PCR-based cloning technique. The primers for PCR were as follows: 5′-GGATCCGATGCCTGAGGAAACCCAG-3′ and 5′-ACTGTCGGATCCTTAGTCTACTTCTTCCAT-3′. The pET15b-hsp90α constructs were transformed into BL21-codonPlus (DE3)-RP-competent cells (Stratagene, La Jolla, CA) following the manufacturer-provided protocol. Protein synthesis was induced by the addition of 0.25 mM IPTG (isopropyl-β-d-thiogalactopyranoside) to the bacterium culture (optical density 0.8) and incubation for 5 h at 25°C. The His-tagged proteins were first purified by use of a nickel-nitrilotriacetic acid (Ni-NTA) column with a HisBind purification kit (EMD Biosciences, Inc.) according to the manufacturer's procedure. The purified proteins were concentrated in a Centricon YM-100 apparatus (Millipore, Billerica, MA) to 4 ml and loaded onto a Superdex 200 HiLoad gel filtration column (GE Healthcare, Piscataway, NJ) and separated by fast protein liquid chromatography (FPLC). Proteins were eluted by use of Dulbecco's phosphate-buffered saline buffer with a flow speed of 1.2 ml/min. The fractions with hsp90α were further concentrated in a Centricon YM-100 apparatus to achieve a final concentration of 1 mg/ml. Proteins were stored in 10% glycerol-Dulbecco's phosphate-buffered saline at −70°C. Generally, 20 mg of hsp90α could be generated from 1 liter of the bacterium culture.
Five distinct domains (N′, M-1, M-2, C′-1, and C′-2) of hsp90α were incorporated into the His-tagged pET15b as described above. After Ni-NTA column purification, each of the domain proteins was concentrated in a Centricon YM-50 or YM-10 apparatus to a volume of 4 ml, depending on the size of the domains. These domains were then purified by FPLC in a Superdex 75 HiLoad gel filtration column (GE Healthcare, Piscataway, NJ) followed by being concentrated in a Centricon YM-50 or YM-10 apparatus to a final concentration of 1 mg/ml.
The cDNA for wt hsp90α was amplified using PCR and subcloned in frame into the 3′ end of a green fluorescent protein (GFP) gene. This fusion gene was inserted into the lentivirus-derived vector pRRLsinhCMV at BamHI and EcoRI. To identify possible target sequences for small interfering RNA (siRNA) against CD91, we used an RNA interference (RNAi) selection program (12). Four potential sites were selected and synthesized. The effectiveness of synthetic double-stranded siRNA in down-regulation of CD91 was first measured in 293 cells by transfection, and the cell lysates were blotted with corresponding antibodies. The two most effective examples of RNAi were then cloned into the lentiviral RNAi delivery vector FG-12 (12). The selected RNAi sequences (sense) with respect to human CD91 for FG12 cloning were GACCAGTGCTCTCTGAATA (RNAi-1) and GGAGTGGTATTCTGGTATA (RNAi-2). These constructs were used to transfect 293T cells together with two packaging vectors, pCMVΔR8.2 and pMDG, to produce virus stocks as previously described (12).
The ATPase assay was based on a regenerating coupled enzyme assay (1) in which the phosphorylation of ADP by pyruvate kinase (PK) at the expense of phosphoenol pyruvate is coupled to the reduction of the resulting pyruvate by lactate dehydrogenase (LDH) at the expense of NADH. Oxidation of NADH to NAD+ produces a loss of optical density at the NADH absorbance maximum of 340 nm in a relationship of direct stoichiometry to the amount of ADP phosphorylated. The ATPase activity was expressed as the turnover of NADH per minute per microgram of protein. Each 400-μl assay contained 50 mM HEPES-KOH (pH 8.0), 5 mM MgSO4, 0.6 mg ATP (Sigma), 30 μg NADH (Sigma), 2 mM phosphoenol pyruvate (Sigma), 1.4 units of PK, 2 units of L-LDH (PK/LDH mix from Sigma), and 200 μg of wt or mutant hsp90α proteins.
TGFα is a potent mitogen and motogen in HS for epidermal HKCs (25). Migrating cells are not proliferating and vice versa. We wanted to identify downstream and motility signal-specific targets that satisfy the criteria of being (i) primary targets of TGFα signaling, (ii) secreted proteins, and (iii) factors that selectively promote migration but not proliferation. We focused on serum-free conditioned media of TGFα-primed (and subsequently removed during incubation and medium collection) HKCs, according to the step-by-step procedure depicted schematically in Fig. Fig.1A.1A. The starting control medium and the conditioned medium (HKC-CM) were independently subjected to migration and DNA synthesis assays. As shown in Fig. Fig.1B,1B, 10× HKC-CM (panel d), but not 10× control medium (panel c), was able to duplicate the promotility effect of TGFα (panel b versus panel a) in the colloidal gold migration assays (quantitated as migration index values) (22). This observation was confirmed by using the in vitro “scratch” assay in the presence of mitomycin C, a DNA synthesis inhibitor. As shown in Fig. Fig.1C,1C, the control medium only slightly closed the “wounded” area following overnight incubation (panel c′ versus panel a′), likely due to coated collagen-driven cell migration (23). Incubation with HKC-CM, however, caused complete closure of the wounded area (panel d′), in similarity to the results seen with TGFα-stimulated migration (panel b′). The data were quantitated as the “average gap” of the unclosed space (22). A 10× concentration of HKC-CM was chosen, because it was the lowest HKC-CM concentration that resulted in plateau promotility activity (see Fig. S1 in the supplemental material).
Interestingly, unlike the dual effects of TGFα on cell migration and growth, HKC-CM showed little mitogenic effect on either epidermal or dermal cells. As shown in Fig. Fig.1D,1D, TGFα strongly stimulated DNA synthesis in HKCs, as expected (bar 4 versus bar 1). In comparison, HKC-CM showed little stimulation of DNA synthesis (bar 7). Similar effects of HKC-CM on HDMECs (bar 8 versus bar 2) and DFs (bar 9 versus bar 3) were observed. In comparison, PDGF-BB and fetal bovine serum (FBS) strongly stimulated DNA synthesis in these cells (bars 5 and 6), as expected. In contrast, we found that HKC-CM was also able to stimulate migration of DFs and HDMECs. As shown in Fig. Fig.1E,1E, in comparison to the promotility effects of PDGF-BB and FBS, respectively (bars 5 and 6), HKC-CM showed equivalent levels of stimulation of HDMEC (bar 8) and DF (bar 9) migration. Therefore, this yet-to-be-identified promotility activity in the HKC-CM satisfied two of the three proposed criteria, i.e., it represented (i) a secreted factor(s) with (ii) a motogenic but not mitogenic effect on all the major types of human skin cells involved in wound healing.
What is the factor(s) in HKC-CM that promotes skin cell migration? SDS-PAGE slab gel and silver staining of HKC-CM, as shown in Fig. Fig.2A,2A, revealed multiple proteins with a wide range of molecular masses (left column). Initially, 19 of the clearly definable protein bands were excised (as indicated in Fig. S2 in the supplemental material). The peptides were extracted and subjected to mass spectrometry analyses (see Table S1 in the supplemental material). Contrary to our expectations, among the 13 known proteins identified, none was related to any growth factors, cytokines, or other well-known soluble promotility peptides. Most of the 13 proteins, however, have previously been reported in studies of conditioned media of various types of cells (see Table S1 in the supplemental material). Therefore, we were left with only the choice of protein purification to identify which of these proteins is responsible for the promotility activity found in HKC-CM.
The protein contents of 10 liters of HKC-CM were concentrated 100-fold and subjected to FPLC, as diagrammed schematically in Fig. Fig.2A2A (middle column) (for a representation of a detailed protein purification procedure, see Table S2 in the supplemental material). Fractions 17 and 18 of the final chromatography procedure (Superdex 200), which contained a peak of promotility activity, were pooled, concentrated, and resolved by SDS-PAGE. Silver staining of the SDS gel revealed four major polypeptides. Mass spectrometry analyses of the excised individual bands unveiled four previously known gene products. In order of molecular mass from high to low, the products were hsp90α, AMF/neuroleukin, fibronectin isoform 6 preproprotein, and 14-3-3σ. To narrow down whether any of the proteins are responsible for the motogenic activity in HKC-CM, the recombinant forms of these four proteins were individually tested. As shown in Fig. Fig.2B,2B, we surprisingly discovered that recombinant hsp90α stimulated HKC migration in a dose-dependent manner equivalent to TGFα stimulation (bars 3 to 6 versus bar 2). In contrast, neither AMF nor 14-3-3σ showed any significant promotility effect on HKCs (bars 8 to 11 or bars 12 to 15 versus bar 1). We reported previously that fibronectin minimally stimulates HKC migration (22).
To prove the physical presence of hsp90α protein in HKC-CM and to estimate its working concentration in comparison to that of recombinant hap90α, 10× HKC-CM, together with a series of known amounts of recombinant hsp90α proteins, was subjected to anti-hsp90α Western blotting analyses. As shown in Fig. Fig.2C,2C, based on a standard curve generated from the results seen with increasing amounts of recombinant hsp90α (lanes 1, 2, and 3), 10× HKC-CM contained 0.06 to 0.08 μM hsp90α, which is close to the optimal concentration for the promotility effect seen with 0.1 μM recombinant hsp90α (see Fig. Fig.2B,2B, bar 5). To verify that hsp90α is the promotility factor in HKC-CM, we depleted hsp90α from HKC-CM by anti-hsp90α antibody immunoprecipitation and then tested the hsp90α-free HKC-CM with respect to HKC migration. As shown in Fig. Fig.2D,2D, anti-hsp90α antibody depletion eliminated the promotility activity of HKC-CM in an antibody-dose-dependent manner (bars 6 to 10). In contrast, neither a control IgG nor neutralizing antibodies against AMF or 14-3-3σ showed any inhibitory effect on HKC-CM-stimulated HKC migration (bars 3, 4, and 5 versus bar 2). These results clearly indicated that the promotility activity in HKC-CM is due to hsp90α.
We were curious whether secretion of hsp90α is a general event in other types of skin cells stimulated with growth factors, i.e., DFs, HDMECs, and MCs. As shown in Fig. Fig.3A,3A, we found that hsp90α was undetectable in CM of PDGF-BB-stimulated DFs (lane 1), vascular endothelial growth factor-stimulated HDMECs (lane 4), or FBS-stimulated MCs (lane 3). However, a 90-kDa protein was clearly detected in TGFα-stimulated HKC-CM (lane 2) that comigrated with the recombinant hsp90α control (lane 5). We then tested whether hsp90α also satisfies our third criterion, i.e., that it is not a mitogen. As shown in Fig. Fig.3B,3B, consistent with the lack of mitogenic effect in HKC-CM, recombinant hsp90α was unable to cause any significant increase in DNA synthesis in the three primary human skin cell types studied (bars 7 to 9).
As far as the intracellular hsp90α levels were concerned, we observed that HKCs and DFs had similar levels and HDMECs showed a slightly lower level of intracellular hsp90α proteins (Fig. (Fig.3C,3C, panels a and b). Hsp90β was included as a specificity control. HKCs showed lower expression, and DFs and HDMECs showed similar levels of hsp90β (Fig. (Fig.3C,3C, panels c and d). In examining the expression of hsp90α versus that of hsp90β (control) in vivo with monoclonal antibodies specifically against either hsp90α or hsp90β, as shown in Fig. Fig.3D,3D, we found that hsp90α staining is significantly stronger in the epidermis than in the dermis (panel b versus panel a). In contrast, hsp90β staining was strongly detected in the dermis but was only weakly detected in the epidermis (panel e versus panel d).
Based on the findings presented above, we speculated that TGFα should stimulate membrane translocation and secretion of intracellular hsp90α in HKCs but not in dermal cells. To test this hypothesis, we undertook two independent approaches: (i) direct immunostaining of endogenous hsp90α with a monoclonal anti-hsp90α antibody and (ii) detection of secreted hsp90α from the conditioned media. As shown in Fig. Fig.4A,4A, TGFα stimulation caused rapid and robust membrane relocation and cell surface clustering of hsp90α in a time-dependent manner in HKCs (left column, panels b to d versus panel a). The accumulation of hsp90α, particularly between 15 min and 60 min following TGFα stimulation, was so overwhelming that it formed “budding structures” toward the outside of the cell membrane (panels c and d; see inserted enlargement). In fact, this translocation could occur as rapidly as 2 min following TGFα stimulation (see Fig. S3 in the supplemental material). The presence of cycloheximide, a common protein synthesis inhibitor, exerted no inhibitory effect on the translocation (data not shown). Interestingly, however, no detectable hsp90α relocation occurred in DFs in response to PDGF-BB, a potent DF promotility factor (right column, panels a′ to d′). More surprisingly, as shown in Fig. Fig.4B,4B, although TGFα can also stimulate DF migration, it was unable to induce any detectable membrane relocation of hsp90α in DFs (panel d), although it did so in HKCs in a side-by-side experiment (panel b). The reason for this disparity in results remains unknown.
To verify directly that TGFα stimulation causes further secretion of hsp90α from HKCs, two identical dishes of serum-starved HKCs were either left untreated or treated with TGFα for 4 h. Both conditioned media and total cell lysates were comparatively analyzed by Western blotting with an anti-hsp90α-specific antibody. As shown in Fig. Fig.4C,4C, TGFα stimulation resulted in a detectable decrease in the intracellular hsp90α level (panel a, lane 2 versus lane 1). An anti-β-actin blot was used as the loading control (panel b). Consistently, TGFα stimulated a concomitant increase in secretion of hsp90α in the conditioned medium over a basal level (panel c, lane 2 versus lane 1). Thus, these data provided direct evidence that TGFα stimulates secretion of hsp90α from its preexisting intracellular pool. Estimation of the secreted hsp90α versus the total intracellular hsp90α at both 4-h and 16-h (overnight) time points revealed 16 to 20% secretion from the intracellular hsp90α pool.
We confirmed the observation by constructing and introducing GFP-linked wt and ATP-binding and ATPase mutants of human hsp90α fusion genes. The results of the experiments clearly showed that TGFα stimulation increased the amount of extracellular GFP-hsp90α. Mutations in ATP-binding and ATPase domains of hsp90α decreased the efficiency of membrane translocation in response to TGFα (see Fig. S4A in the supplemental material). Furthermore, TGFα stimulation caused time-dependent secretion of the GFP-hsp90α fusion protein (see Fig. S4B in the supplemental material).
Finally, to investigate the mechanism by which hsp90α was exported, we asked whether secretion of hsp90α is mediated by the classical endoplasmic reticulum/Golgi transport pathway or via the nonconventional “exosome traffic pathway.” We took advantage of two inhibitors, BFA and DMA, which block classical and exosome-mediated protein secretion, respectively (21, 40). In addition, confocal microscopy was used to analyze hsp90α membrane translocation at a specific layer of the cells. As shown in Fig. Fig.5A,5A, TGFα stimulation caused clear membrane translocation of hsp90α (panel f versus a). DMA inhibited the membrane translocation in a dose-dependent fashion (panels g to i versus panel f). In contrast, BFA, even at its reported high inhibitory concentration, showed little effect (panel j versus panel f). Cells treated with vehicle alone were included as positive controls (panel f versus panel a). Furthermore, inhibition of TGFα-stimulated secretion of hsp90α by DMA was also observed. As shown in Fig. Fig.5B,5B, DMA (lane 6), but not BFA (lane 4), almost completely blocked hsp90α secretion into the extracellular environment by HKCs in response to the presence of TGFα (panel a, lane 6 versus lanes 2 and 4). As positive controls, as shown in Fig. Fig.5B,5B, DMA blocked secretion of hsp70, a well-known exosome-mediated secretion (panel b, lane 6) (8, 21), but not that of matrix metalloproteinase 10 (MMP10), which is secreted by the classical endoplasmic reticulum/Golgi transport pathway (panel c, lanes 5 and 6). In fact, DMA increased MMP10 secretion for unknown reasons (lane 6 versus lane 2). BFA blocked the classical endoplasmic reticulum/Golgi transport pathway-mediated secretion of MMP10 (panel c, lane 4) but not shp70 secretion (panel b, lane 4 versus lane 2). These data indicated that TGFα stimulates secretion of hsp90α in HKCs via the exosome pathway.
Hsp90α is an ATP-binding protein and has intrinsic ATPase activity, and both are required for its intracellular chaperon function in the so-called hsp90 chaperon cycle (35, 38, 50). Therefore, we asked whether these activities are also required for hsp90α's extracellular function to promote cell migration. His-tagged recombinant hsp90α-wt, hsp90α-E47A, hsp90α-E47D, and hsp90α-D93N proteins were produced in bacteria, isolated by use of an Ni-NTA column, and purified by FPLC (Fig. (Fig.6A).6A). To verify the effects of the mutations, we subjected the purified proteins to an in vitro ATPase assay. As shown in Fig. Fig.6B,6B, hsp90α-wt showed expected ATPase activity (bar 2) greater than that seen with a bovine serum albumin (BSA) control (bar 1). Hsp90α-E47D showed about one-half of the ATPase activity seen with hsp90α-wt (bar 3). Neither hsp90α-E47A nor hsp90α-D93N had any significant ATPase activity (bars 4 and 5) greater than that seen with the BSA control (bar 1). When these proteins were tested for their effects on HKC migration, however, we surprisingly found that none of the mutations affected the ability of the proteins to stimulate HKC migration. As shown in Fig. Fig.6C,6C, the three mutant hsp90α proteins (bars 6 to 14) showed dose-dependent stimulation of HKC migration, just as seen with the wt hsp90α (bars 3 to 5) or TGFα (bar 2) stimulation. These results indicate that, unlike its intracellular action as an ATP-dependent chaperon, extracellular hsp90α promotes cell motility without the need to bind any cofactors or ATP.
What part(s) of Hsp90α promotes migration, then? Hsp90α is composed of an N′-terminal domain, a charged sequence, a middle domain, and a C′-terminal domain (Fig. (Fig.7).7). To narrow down the identity of the domain that carries out the promotility function of hsp90α, we constructed each of the individual domains, expressed each in bacteria as His-tagged proteins, and purified them by FPLC (see Materials and Methods). Equal molarities of the proteins were then tested for promotility effects on HKCs. As shown in Fig. Fig.7,7, full-length (wt) hsp90α showed a remarkable promotility activity in comparison to the control medium results. The middle domain plus the charged sequence (M-1) showed a degree of activity similar to that seen with the wt hsp90α. However, the middle domain lacking the charged sequence (M-2) showed significantly decreased activity, although the charged sequence plus the entire N′-terminal domain (N′) showed no stimulating activity. The two C′-terminal domains (C′-1 and C′-2, constructed for the purpose of ensuring the accuracy of the results) both showed moderate promotility activity. Therefore, hsp90α promotes HKC migration mainly through its middle sequence plus the charged sequence, which is consistent with their surface location in hsp90α (34).
To investigate how hsp90α promotes skin cell migration from outside of the cells, we focused on a reported common receptor for heat shock proteins, LRP-1/α2-macroglobulin receptor/CD91 (5, 49). CD91 consists of a 515-kDa extracellular subunit and a membrane-anchoring 85-kDa subunit, which are formed from proteolytic products of a common 600-kDa precursor (19). We undertook two independent approaches to study the role of CD91. First, we designed and delivered two sequence-independent siRNAs to HKCs by lentiviral infection. As shown in Fig. Fig.8A,8A, CD91-RNAi-1 and CD91-RNAi-2 almost completely down-regulated the endogenous CD91 protein (lanes 2 and 3 versus lane 1). When these cells were subjected to migration assays, as shown in Fig. Fig.8D,8D, hsp90α strongly stimulated HKC migration (bar 2). However, down-regulation of CD91 completely blocked HKC cell migration in response to hsp90α (bars 4 and 6 versus bar 2). Second, we used a monoclonal neutralizing antibody to block the cell surface CD91, as schematically shown in Fig. Fig.8E,8E, and similar results were obtained from the experiments. As shown Fig. Fig.8F,8F, hsp90α strongly stimulated HKC migration (bar 2 versus bar 1). Addition of a control IgG showed little effect (bar 3). However, addition of even 3 μg/ml of anti-CD91 antibodies completely blocked hsp90α-induced HKC migration in a dose-dependent manner (bars 4 to 6 versus bar 2).
We confirmed the importance of CD91 in hsp90α signaling using other human skin cell types. In comparison to HKCs, DFs and HDMECs not only are CD91 positive but express even higher levels of CD91 (Fig. (Fig.8B,8B, lanes 3 and 4 versus lane 2). Dendritic cells were included as a positive control (lane 1). When CD91 in DFs was down-regulated by siRNA (Fig. (Fig.8C)8C) and tested in migration assays, we found that hsp90α was no longer able to stimulate migration of DFs (Fig. (Fig.8G,8G, bar 4). In contrast, hsp90α still stimulated migration of the DFs infected with vector alone (bar 2).
Finally, we studied whether hsp90α is able to bind CD91. In GST-hsp90α pull-down assays, as shown in Fig. Fig.8H,8H, GST alone was unable to bind any CD91 from the lysates of HKCs (lane 1). However, GST-hsp90α pulled down CD91 in a dose-dependent manner (lanes 2 to 5). Using His-tagged domains of hsp90α on beads, as shown in Fig. Fig.8I,8I, we observed that, of the N′-terminal ATP binding and ATPase domain, the middle domain, and the C′-terminal domain, the middle domain of hsp90α (lane 3) bound more strongly to CD91 than the C′-terminal and the N′-terminal domains (lanes 4 and 2). Full-length hsp90α (lane 5) and empty beads (lane 1) were included as positive and negative controls, respectively. Taken together, these findings indicated that CD91 is a receptor for secreted hsp90α, whose extracellular function is to promote migration of all CD91-positive (CD91+) skin cell types during wound healing.
To investigate whether the extracellular function of hsp90α is required for mediating TGFα's promotility signaling in HKCs, we needed to avoid using small-molecule inhibitors, such as geldanamycin (17-AGG), which would simultaneously inhibit both extracellular and intracellular hsp90α. Therefore, anti-hsp90α neutralizing antibodies were used to selectively nullify the extracellular hsp90α. We tested whether the continued presence of the antibodies in culture medium would block TGFα-stimulated HK migration. In the same experiments, the effect of the same antibodies on PDGF-BB-stimulated DF migration was included as a cell type specificity control. As shown in Fig. Fig.9A,9A, TGFα dramatically stimulated HK migration on collagen (panel b versus panel a). However, the presence of increasing amounts of the anti-hsp90α neutralizing antibody (far right column) inhibited TGFα-stimulated HK migration in a dose-dependent manner (panels c to f), ultimately reducing the migration to the background level (panel f versus panel a). More intriguingly, as shown in Fig. Fig.9B,9B, the addition of the same antibody blocked even HS-stimulated HK migration (panels i to l). In contrast, the same antibody showed little inhibitory effect on PDGF-BB-stimulated DF migration (Fig. (Fig.9C,9C, panels o to r), which is consistent with the previous observation that DFs do not secrete hsp90α in response to PDGF-BB (see Fig. Fig.3A,3A, lane 1). The addition of even the highest amount of a control IgG did not result in any detectable effect (panels b and h).
In human skin, the main body of cells in epidermis is represented by HKCs (>90%) and the two main types of cells in the dermis are DFs and HDMECs. HKC migration is initiated immediately following wounding. Dermal cells, however, move into the wound approximately 4 days after the injury, when they deposit extracellular matrices and remodel the wound after keratinocytes close the wound (43). It has long remained a puzzle how the dermal cells could have migrated into the wound, since the high concentration (30 to 50 ng/ml) of TGFβ present in human serum (the main soluble environment in the wound) would completely block their migration (39). In contrast, HKC migration is insensitive to TGFβ due to the lower expression level of the type II TGFβ receptor in the cells (3). Thus, a clinically relevant issue was whether or not the hsp90α-induced migration of dermal cells is sensitive to TGFβ. Consistently, as shown in Fig. 10A, HKC-CM and recombinant hsp90α promoted migration of all three cell types (bars 7 to 12), in similarity to the results seen with respect to growth factor stimulation (bars 4 to 6). Based on these data, we conclude that hsp90α is a general promotility factor for all the major skin cell types involved in wound healing. As shown in Fig. 10B, human plasma (HP) (containing little TGFβ) stimulated DF migration (bar 2 versus bar 1), whereas HS (with high concentrations of TGFβ) completely blocked DF migration (bar 3), as is consistent with our previous findings (3). As expected, PDGF-BB and hsp90α equally stimulated DF migration (bar 4 versus bar 5). However, in the copresence of TGFβ-, HP-, and PDGF-BB-induced DF, migration was completely shut down (bars 6 and 8). Intriguingly, TGFβ was unable to inhibit hsp90α-stimulated DF migration (bar 9 versus bar 5). In contrast, when purified hsp90α was used to “rescue” the inhibition of DF migration from HS or TGFβ inhibition, the addition of hsp90α overrode the blockage of HS or TGFβ3 on PDGF-BB-induced DF migration (bars 11 and 12 versus bars 3 and 7). Therefore, the secretion of hsp90α by HKCs may subsequently serve as a major promotility factor in the wound bed.
In the current report, we provide evidence that TGFα triggers HKCs to secrete hsp90α and that extracellularly located hsp90α in turn acts as a promotility factor for all major human skin cell types. Among the major cell types in skin, only HKCs secrete hsp90α. Mechanistically, TGFα triggers rapid membrane translocation of hsp90α and its secretion to extracellular environment via the exosome pathway. Among the four domains of hsp90α, only the middle domain (plus the charged sequence) is able to fully duplicate the promotility effect of the full-length hsp90α, independently of its ATP-binding and ATPase functions. Furthermore, extracellular hsp90α promotes migration of both epidermal and dermal cells though the cell surface receptor LRP-1/CD91. This report presents the physiologically significant finding that the extracellular hsp90α-induced skin cell migration could no longer be inhibited by TGFβ, which is abundantly present in skin wounds. TGFβ potently inhibits canonical growth factor-stimulated migration of DFs and HDMECs; therefore, our finding provides for the first time an explanation of how the dermal cells move into the wound bed in the presence of TGFβ. A schematic representation of these findings is shown in Fig. Fig.11.11. In support of this model in vivo, we recently reported that topical application of purified hsp90α enhanced skin wound healing in nude mice (24). Taking all of these results together, we propose that following wounding, increased TGFα from the serum stimulates HKCs to secrete hsp90α. After reaching its threshold, extracellular hsp90α not only promotes HKC migration for reepithelialization but also recruits dermal cells to migrate into the wound for new connective tissue formation, new blood vessel formation, and wound remodeling. Moreover, hsp90α's TGFβ-insensitive signaling has a clear advantage over that of the canonical growth factors in promotion of cell migration under the conditions of the TGFβ-rich microenvironment of skin wounds. Considering the fact that TGFβ potently inhibits dermal cell migration, such as that of DFs and HDMECs (3), extracellular hsp90α may represent a bona fide physiological driving force for dermal cell migration into the wound bed to participate in wound healing.
Historically, the hsp90 family chaperons regulate folding, transport, maturation, and degradation of a diverse, but selective, set of client proteins, in particular, signaling molecules (33, 38). Recently, extracellularly located hsp90 proteins have also been reported. Liao and colleagues showed that oxidative stress causes sustained release of hsp90α, which in turn stimulates activation of ERK1/2 in rat vascular smooth muscle cells (26). Jay and his colleagues identified hsp90α, but not hsp90β, in the conditioned media of tumor cells (11). Moreover, two groups showed that heat stress causes increased secretion of hsp90 and hsp70 (8, 21). Yu and colleagues showed that γ radiation stimulates secretion of hsp90β into conditioned medium in a p53-dependent manner (51). As previously mentioned, these proteins have to undergo the exosome-mediated exocytosis that results in export of cellular proteins that lack a signal sequence. In contrast, these proteins cannot be secreted via the conventional endoplasmic reticulum/Golgi transport pathway. Exosomes, also called “intraluminal vesicles,” are contained within multivesicular bodies (MVBs), whose known function is to act as an intermediate in the degradation of proteins internalized from the cell surface or sorted from the trans-Golgi organelle (10, 47). However, MVBs and their exosomes can also fuse with the plasma membrane to release cargo proteins into the extracellular space (14, 44). In fact, all of the proteins that have been identified in exosomes are located in the cell cytosol or endosomal compartments and never in the endoplasmic reticulum, Golgi apparatus, mitochondria, or nucleus (44). Hsp90α has been identified in exosomes (8, 18). In this paper, our data link EGFR activation to the exosome pathway, leading to membrane translocation and secretion of hsp90α. How EGFR communicates with the MVB protein-trafficking pathway remains an unanswered question.
How does extracellular hsp90α promote cell migration? Jay and his colleagues showed that extracellular hsp90α, but not hsp90β, interacts with and acts as an activator of MMP2 and that functional inhibition of extracellular hsp90α inhibited tumor invasion (11). However, unlike the results seen with hsp90α, which is able to duplicate the full promotility effect of TGFα on HKCs, there is no evidence that the addition of any recombinant MMP alone can promote cell motility. In addition, the presence of MMP inhibitors, such as GM6001 or MMP Inhibitor III, had little inhibitory effect on hsp90α-stimulated HKC migration (unpublished results). Instead, we provided evidence that CD91 is the cell surface receptor that mediates hsp90α signaling to promote cell migration. CD91 is found in monocytes, hepatocytes, fibroblasts, and keratinocytes (19, 20, 45, 48, 49). We showed that the three major skin cell types, HKCs, DFs, and HDMECs, were all CD91 positive and that all responded to the promotility effect of hsp90α. Basu and colleagues have shown that released hsp90α binds to CD91, triggering macrophages to secrete cytokines and causing dendritic cells to express antigen-presenting and costimulatory molecules (4, 5). In addition to hsp90α, other extracellular heat shock proteins, including gp96, hsp60, hsp70, and calreticulin, bind CD91 (5, 17, 36, 48). However, while CD91 is a common receptor for heat shock proteins, not every CD91-binding heat shock protein was able to stimulate cell migration. In our hands, for instance, recombinant calreticulin showed no detectable promotility effect on HKCs, which contrasts with the results seen with hsp90α or TGFα (unpublished data). Therefore, the outcomes of CD91 activation may depend upon binding to its 515-kDa extracellular domain by specific heat shock proteins, such as hsp70 for antigen presentation (6) and hsp90α for cell migration. Interestingly, RNAi down-regulation of CD91 did not block TGFα-stimulated HKC migration in vitro (C.-F. Cheng and W. Li, unpublished data), again supporting our hypothesis that HKC-secreted hsp90α is physiologically important for recruiting dermal cells into wounds.
Hsp90α is composed of an N′-terminal domain, a charged sequence (linker), a middle domain, and a C′-terminal domain. Monoclonal antibody screening analyses suggest that the middle and C′-terminal domains are exposed at the surface of hsp90α protein (34). Analysis of the crystal structures of yeast and Escherichia coli hsp90 revealed that the middle domain forms a highly conserved surface loop, suggesting a common role as a potential client protein binding site (2, 42). Our study showed that the combination of the middle domain plus the changed sequence strongly binds CD91 and exerts the same full-promotility effect as full-length hsp90α. Confirmation of these findings in in vivo wound healing models could further reveal the therapeutic potential of hsp90α.
Finally, the observation that TGFα selectively stimulates secretion of hsp90α in HKCs, but not in DFs, was unexpected, especially since TGFα can promote DF migration as well (23). This finding suggests a previously unrecognized signaling mechanism: a single growth factor can use different mechanisms in two geographically close skin cell types to induce the same cellular response, i.e., motility. While the physiological implication of such TGFα signaling specificity remains to be studied, it is conceivable that a possible “linker” for communication between TGFα-bound EGFR and hsp90α-containing exosomes is selectively present in HKCs but absent in the dermal cell types. The nature of such a linker could be one of several possibilities by which intracellular signaling molecules interpret upstream signals. First, the signaling molecules can be differentially organized by scaffolding proteins, resulting in a variety of combinations. Second, distinct locations of the same signaling molecules can dictate specific biological responses. Third, the strength and duration of a given signal/pathway can affect the outcomes of the responses. The choice of these mechanisms and their responses may differ from cell type to cell type (37). The bottom line is that differences in the cellular context of HKCs and DFs are responsible for determining the signaling specificity of TGFα.
We thank Ulrich Hartl for the cDNAs of wt and mutant hsp90 and the USC Liver Diseases Confocal Microscopy Core.
The study was supported by NIH grants GM/AR066193-01 (to W.L.) and AR46538 (to D.T.W.).
We declare that there was no commercial interest or conflict of interest for this study.
Published ahead of print on 10 March 2008.
†Supplemental material for this article may be found at http://mcb.asm.org/.