Abstract
We recently described a novel population of blood-borne cells, termed fibrocytes, that display a distinct cell surface phenotype (collagen+/CD13+/CD34+/CD45+), rapidly enter sites of tissue injury, and contribute to scar formation. To further characterize the role of these cells in vivo, we examined the expression of type I collagen and cytokine mRNAs by cells isolated from wound chambers implanted into mice. Five days after chamber implantation, CD34+ fibrocytes but not CD14+ monocytes or CD90+ T cells expressed mRNA for type I collagen. Fibrocytes purified from wound chambers also were found to express mRNA for IL-1β, IL-10, TNF-α, JE/MCP, MIP-1α, MIP-1β, MIP-2, PDGF-A, TGF-β1, and M-CSF. The addition of IL-1β (1–100 ng/ml), a critical mediator in wound healing, to fibrocytes isolated from human peripheral blood induced the secretion of chemokines (MIP-1α, MIP-1β, MCP-1, IL-8, and GROα), hemopoietic growth factors (IL-6, IL-10, and macrophage-CSF), and the fibrogenic cytokine TNF-α. By contrast, IL-1β decreased the constitutive secretion of type I collagen as measured by ELISA. Additional evidence for a role for fibrocytes in collagen production in vivo was obtained in studies of livers obtained from Schistosoma japonicum-infected mice. Mouse fibrocytes localized to areas of granuloma formation and connective matrix deposition. We conclude that fibrocytes are an important source of cytokines and type I collagen during both the inflammatory and the repair phase of the wound healing response. Furthermore, IL-1β may act on fibrocytes to effect a phenotypic transition between a repair/remodeling and a proinflammatory mode.
The host response to tissue injury requires a complex interplay of cellular, humoral, and connective tissue elements (1, 2). Platelets play an early role in this response by releasing chemotactic factors that act to recruit peripheral blood leukocytes into the injured site. Peripheral blood leukocytes then produce a variety of mediators that serve to combat infection and coordinate successive steps of the tissue repair response (1, 2, 3). Fibrogenic cytokines such as TGF-β1 and TNF-α stimulate connective tissue cells to proliferate and to secrete extracellular matrix proteins (4). The normal outcome of this cascading series of events is elimination of the invasive stimulus followed by connective tissue scar formation and, over time, by remodeling of the injured site.
We recently described a population of novel, blood-borne fibroblast-like cells that rapidly enter sites of tissue injury and contribute to connective tissue scar formation (5, 6, 7). Termed fibrocytes, these cells comprise ∼0.5% of peripheral blood leukocytes and display an adherent, spindle-shaped morphology (Ref. 5 and J. Chesney, unpublished observations). Fibrocytes obtained from blood express the fibroblast products collagen I, collagen III, and fibronectin, as well as the leukocyte common Ag CD45RO, the pan-myeloid Ag CD13, and the hemopoietic stem cell Ag CD34. Fibrocytes do not synthesize epithelial (cytokeratin), endothelial (von Willebrand factor VIII-related protein), or smooth muscle (α-actin) cell markers and are negative for nonspecific esterases as well as the monocyte/macrophage-specific markers CD14 and CD16 (5). Fibrocytes also do not express proteins produced by dendritic cells or their precursors (CD25, CD10, and CD38) or by the pan-B cell Ag CD19 (5, 8, 9, 10).
In the present study, we report that fibrocytes isolated from wound chambers or peripheral blood express cytokines and type I collagen. Mouse fibrocytes participating in tissue repair in vivo were found to express type I collagen and inflammatory cytokine mRNAs and were identified to localize to areas of matrix deposition in Schistosoma japonicum-infected, granulomatous livers. In vitro, IL-1β and TNF-α stimulated inflammatory cytokine secretion by human peripheral blood fibrocytes, while IL-1β inhibited type I collagen synthesis. IL-1β, an important mediator of wound healing responses, may play an important role in effecting a fibrocyte transition from a “repair” to a “proinflammatory” phenotype.
Materials and Methods
Mice
BALB/c and C57BL/6 mice (18–20 gm) of both sexes were purchased from The Jackson Laboratory, Bar Harbor, ME.
Fibrocyte isolation from peripheral blood
Fibrocytes were harvested and cultured as previously described (5, 6). Briefly, total PBMCs first were isolated from human or murine blood by centrifugation over Ficoll-Paque (Pharmacia, Uppsala, Sweden) following the manufacturer’s protocol. After overnight culture on fibronectin-coated plates (6-well plates, 5 × 106 PBMCs/well) (Becton Dickinson Labware, Bedford, MA) in DMEM (Life Technologies, Gaithersburg, MD) supplemented with 20% FCS (HyClone Labs, Logan, UT), the nonadherent cells were removed by a single, gentle aspiration. Following 10 days of continuous culture, the adherent cells were lifted by incubation in cold 0.05% EDTA/PBS and were depleted by immunomagnetic selection of contaminating T cells (Dynabeads M-450 pan-T, anti-CD2, Dynal, Lake Success, NY), monocytes (Dynabeads M-450 anti-CD14, Dynal), and B cells (Dynabeads M-450 Pan-B, anti-CD19, Dynal). Fibrocyte purity was verified to be >95% (70–80% before depletion of contaminating cells) by FACS analysis (described below). Cell viability was determined to be >90% by trypan blue exclusion.
FACS analysis
2 × 105 purified fibrocytes/sample were washed twice in PBS containing 0.1% sodium azide (Sigma Chemical Co., St. Louis, MO) and 1% BSA (Sigma). The cells were resuspended in 25 μl of diluted Ab (in PBS) and incubated for 30 min on ice (11). The cells then were washed twice and resuspended in 200 μl of PBS. At least 10,000 cells were analyzed on a FACScan instrument (Becton Dickinson). Cells were stained using both phycoerythrin-conjugated anti-CD34 mAb (clone 8G12) (Becton Dickinson) and fluorescein-conjugated anti-collagen I mAb (clone MAB1340) (Chemicon, Temecula, CA). For controls, directly conjugated isotype controls and cell-only samples were analyzed with each Ab.
Protein secretion studies
Purified human fibrocytes were cultured in six-well fibronectin-coated flat-bottom plates (1 × 105 cells/well) in DMEM supplemented with 2% heat-inactivated FCS (HyClone). Cells were incubated for 16 h alone as control or in the presence of IL-1β, IFN-γ, platelet-derived growth factor (PDGF-BB),212, 13, 14n = 3 assays of duplicate wells).
Cell proliferation assay
Purified human fibrocytes were cultured in 96-well flat-bottom plates (5 × 103 cells/well) in DMEM supplemented with 2% heat-inactivated FCS (HyClone). Cells were incubated for 32 h alone as control or in the presence of IL-1β, IFN-γ, PDGF-BB, TGF-β1, or TNF-α (each at 1, 10, or 100 ng/ml). The proliferative activity was measured by direct cell enumeration and by the incorporation of [3H]thymidine (4 μCi/ml) into DNA over the last 12 h of incubation as measured by liquid scintillation counting. Data are expressed as the mean ± SD (n = 3).
Cell migration studies
IL-1β or TNF-α (each at 1, 10, and 100 ng/ml) diluted in DMEM supplemented with 2% heat-inactivated FCS were added to the lower chambers of a 24-well chemotaxis insert plate (8-μm pores) (Becton Dickinson Labware). 2 × 104 fibrocytes in the same medium with or without cytokines were then added to the upper wells. After incubation for 4 h at 37°C, the nonmigrating cells were washed from the upper surface of the filters. The membranes were then removed, fixed in 10% buffered neutral formalin, and stained with Diff-Quik (Sigma). Fibrocytes that migrated through the membrane were enumerated by light microscopy in five 200× fields. Data are expressed as the mean ± SD.
Isolation of wound chamber cells
BALB/c mice were anesthetized by i.p. injection of sodium pentobarbital (40 mg/kg), and an 0.5-cm dorsal midline incision was made using aseptic technique. After dissecting a s.c. pocket, a single sterile wound chamber consisting of a perforated 3.5-cm piece of silicone tubing (Dow Corning, Midland, MI) containing a strip of polyvinyl alcohol sponge (Unipoint Inc., High Point, NC) was inserted along each flank. After 5 days, the mice were killed by CO2 asphyxiation, and the wound chambers were removed under strict aseptic conditions. The cells that had infiltrated the wound chambers were dispersed by trituration in ice-cold PBS containing 0.05% EDTA, centrifuged, and purified into three distinct cell populations by immunomagnetic separation following the manufacturer’s protocol: 1) fibrocytes: biotinylated anti-CD34, clone RAM34, PharMingen, San Diego, CA; Dynabeads M-280 Streptavidin, Dynal; 2) T cells: pan-T, anti-CD90, Dynal; 3) macrophages: anti-CD14, rat IgG1, clone rmC5-3, PharMingen; Dynabeads M-450 sheep anti-rat IgG, Dynal. To verify the purity of each cell fraction, a subset of cells was cultured in RPMI, 10% FCS, for 12 h, separated from the magnetic beads, and examined using spot immunofluorescence analysis. The CD34 fraction was found to be 83 ± 2% CD34+ (anti-CD34, clone RAM34, PharMingen), 79 ± 6% CD45+ (anti-CD45, clone 30F11.1, PharMingen), and 94 ± 2% collagen I+ (anti-collagen I, clone T40775R; Biodesign, Kennebunk, ME). The CD90 fraction was found to be 76 ± 3% CD3+ (αCD3, clone 145–2C11, PharMingen) and the CD14 fraction was found to be 87 ± 5% CD14+ (αCD14, clone rmC5–3, PharMingen). Overall cell viability was >90%.
mRNA expression by wound chamber cells
Total cellular RNA was isolated by a modified guanidinium isothiocyanate method (RNAzol, Cinna Biotecx, Friendswood, TX) within 20 min of the start of the wound chamber cell isolation procedure. cDNA was prepared from 1.0 μg of total RNA using 0.25 ng of oligo(dT)12–18 and Superscript II following the manufacturer’s protocol (Life Technologies, Grand Island, NY). Two-microliter aliquots of cDNA were then amplified by PCR in a Perkin-Elmer model 9600 thermal cycler (Norwalk, CT) using the primers listed below and the following cycling program: denaturation for 15 s at 95°C, annealing for 20 s at 55°C, and extension for 30 s at 72°C for 28 cycles (except β-actin, 22 cycles) with a final extension for 5 min at 72°C. Amplimer sets consisting of presynthesized PCR primer pairs were used for the following mouse mRNAs: IFN-α (294 bp), IFN-γ (365 bp), IL-3 (496 bp), IL-6 (638 bp), IL-10 (455 bp), M-CSF (397 bp), TGF-β1 (525 bp), and TNF-α (354 bp) (Clontech, Palo Alto, CA). The following mouse mRNA primer pairs were custom synthesized: β-actin, 5′GTGGGCCGCTCTAGGCACCA3′, 5′TGGCCT-TAGGGTGCAGGGGG3′ (236 bp); CD3, 5′TCTATCCAGCACCCAGAATC3′, 5′AGTAGGGGGCACTCTGTAAA3′ (269 bp); CD14, 5′GTCCTTAAAGCGGCTTACGG3′, 5′GCGCTAAAACTTGGAGGGCT3′ (399 bp); CD34, 5′AGACTCAGGGAAAGGCCAAT3′, 5′TGAAGGCAGCATGAAGTCAG3′ (168 bp); IL-12, 5′GAAGACGGCCAGAGAAAAAC3′, 5′GCAGAGTCTCGCCATTATGA3′ (357 bp); JE/MCP, 5′ACCAGCCAACTCTCACTGAA3′, 5′TGAAGACCTTAGGGCAGATG3′ (360 bp), KC, 5′CTGTCAGTGCCTGCAGACCA3′, 5′CCAAGGGAGCTTCAGGGTCA3′ (132 bp); PDGF-A, 5′CTCCGTAGGGGCTGAGGATG3′, 5′CTCCTCCTCCCGATGGTCTG3′, (404 bp); PDGF-B, 5′GCCGCCAGCGCCCATCTTAT3′, 5′CGGGCAGGGAGAGGTGCAAA3′ (332 bp); IL-1β, 5′GCTGGAGAGTGTGGATCCCA3′, 5′GCTGTCTGCTCATTCATGAC3′ (301 bp); MIP-1α, 5′TTGTCACCAGACGCGGTGTG3′, 5′GGCAATCAGTTCCAGGTCAG3′ (282 bp); MIP-1β, 5′CTCTGCGTGTCTGCCCTCTC3′, 5′GACCATCTCCATGGGAGAGA3′ (490 bp); and MIP-2, 5′TCTTCCTCGGGCACTCCAGA3′, 5′GGACAGCAGCCCAGGCTCCT3′ (378 bp) (Life Technologies).
Immunohistochemistry
C57BL/6 mice were infected by the s.c. route with 25 cercariae of a Philippine strain of S. japonicum cercariae at Lowell University, Lowell, MA (National Institute of Allergy and Infectious Diseases Supply Contract AI-02636) (15). After 14 wk of infection, the livers were removed, fixed in 10% buffered neutral formalin, embedded in paraffin, sectioned, and processed for immunohistochemical analyzes. After blocking endogenous peroxidases with H2O2 (3%), the deparaffinized sections were incubated with an anti-CD34 mAb (1:50 dilution) (clone QB-END/10) (Accurate) or an IgG1 isotype control. The sections were then washed, and an immunoperoxidase-linked secondary Ab (Dako, Copenhagen, Denmark) was added, followed by diaminobenzidene as substrate. Control sections were stained with an isotype control or without primary Ab and showed no immunoreactivity. Adjacent sections also were stained with Masson’s Trichrome (Sigma) to visualize connective tissue matrix.
Results
To more closely examine fibrocyte function in vivo, we first studied the expression of cytokine and of collagen mRNA in various cell types actively participating in tissue repair. Wound chambers were implanted into mice and, 5 days later, the infiltrating cells were collected and fractionated by positive immunoselection into CD90+ T cells, CD14+ monocytes, and CD34+ fibrocytes. The expression of different mRNAs by these purified cell populations then was analyzed by reverse transcription PCR. Type I collagen mRNA was found to be expressed by CD34+ fibrocytes but not CD14+ monocytes or CD90+ T cells (Fig. 1⇓). Fibrocytes also expressed detectable levels of mRNAs for the pro-inflammatory cytokines, IL-1β and TNF-α, and the anti-inflammatory cytokine, IL-10. Chemokine mRNAs were expressed by all 3 cell types. By comparing the intensities of the DNA amplification products, fibrocytes appeared to express the highest levels of MIP-1α and MIP-2. Fibrocytes also expressed high levels of mRNA for the fibrogenic growth factors, PDGF-A and TGF-β1, and the hemopoietic growth factor M-CSF when compared with monocytes or T cells. Fibrocytes did not express detectable levels of IFN-γ, IL-3, IL-6, or PDGF-B (Fig. 1⇓ and data not shown).
Wound chamber cell mRNA expression. Wound chambers were implanted s.c. into mice, and 5 days later the infiltrating inflammatory cells were collected and fractionated into CD90+ T cells, CD14+ monocytes, and CD34+ fibrocytes (11 ± 2%, 36 ± 5%, and 16 ± 3% of total cells, respectively). Total cellular RNA was isolated from each cell type and analyzed by reverse transcription PCR as described in Materials and Methods.
We next investigated the secretory profile of human fibrocytes isolated from peripheral blood and stimulated in vitro with mediators known to be present during the wound-healing response (1, 4). To verify that cultures of fibrocytes purified from human blood were homogenous, we first examined these cells by flow cytometry for coexpression of type I collagen and CD34, a phenotype unique to fibrocytes (5). After negative immunoselection of T cells, monocytes, and B cells, peripheral blood fibrocytes were found to be >95% pure (Fig. 2⇓). Purified peripheral blood fibrocytes thus were found to constitutively secrete the α chemokines, MIP-1α, MIP-1β, MCP-1, and the β chemokines: IL-8 and GROα (Fig. 3⇓). The addition of the proinflammatory cytokines, IL-1β or TNF-α, to fibrocyte cultures increased the secretion of these mediators.
Type I collagen and CD34 coexpression by human peripheral blood fibrocytes. Purified human fibrocytes were incubated with a FITC-conjugated anti-type I collagen mAb and a phycoerythrin-conjugated anti-CD34 mAb, then analyzed by flow cytometry as described in Materials and Methods. The horizontal and vertical lines mark fluorescence intensity greater than the background that was observed with isotype control mAbs.
Chemokine secretion by human peripheral blood fibrocytes. MIP-1α (A), MIP-1β (B), MCP-1 (C), IL-8 (D), and GROα (E) release were analyzed by ELISA of supernatants collected from unstimulated human fibrocytes (□) and human fibrocytes stimulated for 16 h with IL-1β (▧) or TNF-α (▪). Measurements were performed in triplicate cultures; the data shown are the mean ± SD of two separate experiments.
The hemopoietic growth factors M-CSF and IL-6 regulate macrophage differentiation and lymphocyte proliferation, respectively, and are known to be released during the early phase of tissue repair (16, 17, 18). Peripheral blood fibrocytes constitutively secrete substantial amounts of M-CSF and are induced to secrete even higher levels by the addition of IL-1β or TNF-α (Fig. 4⇓A). IL-6 also is produced by fibrocytes upon stimulation with IL-1β or TNF-α (Fig. 4⇓B). IL-10 is an “anti-inflammatory” cytokine and may play an important role in the wound-healing response by acting as a potent mitogen for mast cells (17). IL-10 was found to be secreted by peripheral blood fibrocytes when induced by IL-1β or TNF-α (Fig. 4⇓C). IL-3 synergizes with IL-6, M-CSF, or IL-10 to induce stem cell differentiation and/or proliferation (19, 20, 21). However, peripheral blood fibrocytes did not secrete IL-3, either constitutively or when induced by IL-1β, TNF-α, PDGF-BB, TGF-β1, or IFN-γ (detection limit = 24 pg/ml) (data not shown).
Hemopoietic growth factor secretion by human peripheral blood fibrocytes. M-CSF (A), IL-6 (B), and IL-10 (C) release were analyzed by ELISA of supernatants collected from unstimulated human fibrocytes (□) and human fibrocytes stimulated for 16 h with IL-1β (▧) or TNF-α (▪). Measurements were performed in triplicate cultures; the data shown are the mean ± SD of two separate experiments. p < 0.01 and p < 0.05 for M-CSF and TNF-α secretion, respectively, for fibrocytes stimulated with 100 ng IL-1β vs control (no stimulation) (two sample t test, assuming unequal variances).
IL-1β, TNF-α, PDGF-AB/AA, and TGF-β1 each have been found to be present in the tissue repair microenvironment and have been implicated in the stimulation of fibrogenic responses in vivo (4). Peripheral blood fibrocytes constitutively secrete appreciable levels of TGF-β1 and secrete TNF-α upon induction with IL-1β (Fig. 5⇓). However, fibrocytes do not secrete IL-1β (detection limit = 3.9 pg/ml), PDGF-AB (detection limit = 8.4 pg/ml), PDGF-AA (detection limit = 1.5 ng/ml), or MIF (detection limit = 0.15 ng/ml) either constitutively or after stimulation with IL-1β, TNF-α, PDGF-BB, TGF-β1, or IFN-γ (data not shown).
Fibrogenic cytokine secretion by human peripheral blood fibrocytes. TGF-β1 (A) and TNF-α (B) release were analyzed by ELISA of supernatants collected from unstimulated human fibrocytes (□) and human fibrocytes stimulated for 16 h with IL-1β (▧) or TNF-α (▪). Measurements were performed in triplicate cultures; the data shown are the mean ± SD of two separate experiments.
Previous studies have shown that IFN-γ inhibits proliferation and collagen synthesis by connective tissue fibroblasts in vitro and that IL-1β, TNF-α, PDGF-BB, and TGF-β1 can stimulate these activities (4). IFN-γ did not inhibit fibrocyte proliferation or constitutive collagen expression and, of the fibrogenic cytokines tested (IL-1β, TNF-α, PDGF-BB, and TGF-β1), only IL-1β was found to affect collagen expression and proliferation (Figs. 6⇓ and 7 and data not shown). The proliferative response of fibrocytes to IL-1β in vitro was confirmed by direct cell enumeration (control, 5.6 ± 0.33 × 103 cells; IL-1β-stimulated (100 ng/ml at 12 h), 8.7 ± 0.45 × 103 cells, p < 0.05). IL-1β also stimulated fibrocyte migration; however, the potent tissue fibroblast chemoattractants PDGF-BB and TGF-β1 were not found to affect fibrocyte movement (Fig. 8⇓ and data not shown).
Type I collagen secretion by human peripheral blood fibrocytes. Type I collagen release was analyzed by ELISA of supernatants collected from unstimulated human fibrocytes (□) and human fibrocytes stimulated for 16 h with IL-1β (▧) or TNF-α (▪). Measurements were performed in triplicate cultures; the data shown are the mean ± SD of two separate experiments.
Proliferation by human peripheral blood fibrocytes. Unstimulated human fibrocytes (□) and human fibrocytes stimulated for 20 h with IL-1β (▧) or TNF-α (▪) were pulsed for 12 h with 4 μCi/ml [3H]thymidine and cell proliferation analyzed by liquid scintillation counting. Data are expressed as mean ± SD and are representative of one experiment that was performed three times.
Migration by human peripheral blood fibrocytes. IL-1β (▧), TNF-α (▪), or media alone (□) were added to the lower chamber of 24-well chemotaxis insert plates and fibrocytes were added to the upper wells. Four hours later, fibrocytes that migrated through the membrane were enumerated in five 200× fields by light microscopy. Data are expressed as the mean ± SD and are representative of one experiment that was performed three times. No significant difference was observed when the stimulant was place in both the upper and lower wells.
The parasitic disease schistosomiasis is characterized by a fibrosing, granulomatous reaction directed against parasite eggs that become entrapped in the hepatic and pulmonary circulations (22). To obtain evidence for collagen synthesis by fibrocytes in vivo, we examined liver tissue from S. japonicum-infected mice for the presence of CD34+ fibrocytes. Figure 9⇓ shows a section of liver obtained 14 wk after infection in which numerous CD34+ cells can be seen to localize to areas of connective tissue matrix deposition (compare Fig. 9⇓, A and B to E and F). No CD34+ fibrocytes could be detected in normal, uninfected livers (data not shown). These data suggest that fibrocytes contribute to the fibrotic pathology that occurs as a consequence of S. japonicum infection.
CD34+ fibrocytes localize with areas of matrix deposition in S. japonicum-infected mouse liver. Liver specimens were fixed, sectioned, and stained with Masson’s trichrome to visualize connective tissue matrix (cyan blue), (A) 200×, (B) 400×. Adjacent sections were labeled with an isotype control mAb (C) 200×, (D) 400X, or an anti-mouse CD34 mAb (E) 200×, (F) 400×, and binding was detected with an immunoperoxidase-linked secondary Ab. Numerous CD34+ fibrocytes can be seen scattered in the granuloma depicted in E, and certain cells are indicated by the arrows in the higher power view shown in F.
Discussion
Connective tissue fibroblasts are a quiescent cell population that under normal circumstances remain sparsely distributed throughout the extracellular matrix (23). As a consequence of injury, fibroblasts can enter and proliferate within the injured site (4). The precise origin of the fibroblast-like cells within wounds has been controversial since the original microscopic studies of developing connective tissue performed by Paget in 1863 (24, 25). That wound fibroblasts appeared by migration from adjacent tissue was supported by experiments showing the apparent ingrowth of fibroblasts from local areas and by the observation that India ink-tagged monocytes failed to develop into tissue fibroblasts in vivo (25, 26). Other studies, however, reported evidence for the differentiation of leukocytes into fibroblasts within s.c. diffusion chambers and the apparent in vitro transformation of peripheral blood mononuclear cells into collagen-producing cells (27, 28).
Studies leading to the identification and purification of the fibrocyte, a blood-borne cell that expresses collagen, together with the hemopoietic/leukocyte cell surface markers CD34 and CD45RO, suggest that this cell type constitutes a previously unrecognized circulating, fibroblast-like cell population (5). The isolation and culture of these cells in high purity (>95% positivity for both CD34 and collagen I) provides evidence for a cell type distinct from that described previously (29, 30) including a population originally identified by Ross and Lillywhite in 1965 (31), who traced the appearance of a blood-borne, connective tissue-like cell type to contamination from local tissue sources. Peripheral blood fibrocytes account for as many as 10% of the inflammatory cells that infiltrate sites of acute injury and have been identified to be present in cutaneous scar tissue (5).
In the current study, mouse fibrocytes participating in tissue repair were found to express the mRNAs for type I collagen and inflammatory cytokines and to be present in significant numbers in areas of matrix deposition in S. japonicum-infected granulomatous livers. Human peripheral blood fibrocytes constitutively secrete type I collagen and are a particularly abundant source of the chemokines MIP-1α, MIP-1β, MCP-1, IL-8, and GROα and the hemopoietic growth factors IL-6, IL-10, and M-CSF. Both cytokine and collagen expression were found to occur in a regulated fashion. IL-1β and TNF-α, two early mediators of the wound-healing response (4), stimulate chemokine and hemopoietic growth factor secretion by peripheral blood fibrocytes in vitro, whereas IL-1β inhibits type I collagen synthesis.
IL-1β is expressed within 24 h of an injury and is essential for the generation of inflammatory and reparative responses in vivo (32). IL-1β up-regulates the expression of adhesion molecules expression by endothelial cells (ELAM-1, VCAM-1, ICAM-1) and stimulates endothelial cell and fibroblast proliferation and matrix production (4, 33). Studies employing IL-1 receptor antagonists in vivo have established that IL-1β is an absolute requirement forconnective tissue scar formation (34). IL-1β may function to maintain peripheral blood fibrocytes in a proinflammatory state early in wound repair, resulting in the increased production of molecules that recruit and expand the inflammatory cell population within the wound environment. The infiltration of CD4+ T cells into areas of tissue damage is considered a critical requirement for the generation of Ag-specific immunity (1). The fibrocyte products MIP-1α, MIP-1β, and MCP-1 are potent T cell chemoattractants and may act to specifically recruit CD4+ T cells into the tissue repair microenvironment (35). Recently, fibrocytes have been identified to be especially potent stimulators of Ag-specific CD4+ T cell proliferation in vivo (6). Fibrocytes thus may function not only to recruit but also to activate CD4+ T cells during tissue repair and may play an important role in the initiation of Ag-specific immunity.
The ability of fibrocytes to both recruit and activate T cells and to secrete type I collagen suggests that these cells may play a critical role in certain connective tissue disorders. A persistent fibrocyte:T cell activation response, for instance, may lead to pathologically significant fibrosis. T cells are essential for the development of schistosomiasis, and the localization of fibrocytes to areas of matrix deposition suggests that these cells may be an important source of collagen production in schistosome-infected liver. Fibrocytes also may participate in the generation of excessive fibroses in certain autoimmune disorders that involve persistent T cell activation, such as scleroderma and graft vs host disease. Further investigation into the function of this novel cell population may provide important insight into the control of the fibrotic responses associated with inflammation.
Acknowledgments
We thank Cathy Rapelje for expert assistance with the FACS analyses and the late Dr. Ken Warren for helpful discussions.
Footnotes
-
↵1 Address correspondence and reprint requests to Dr. Richard Bucala, Department of Medical Biochemistry, The Picower Institute for Medical Research, 350 Community Drive, Manhasset, NY 11030. E-mail address: rbucala{at}picower.edu
-
2 Abbreviations used in this paper: PDGF, platelet-derived growth factor; MCP-1, monocyte chemotactic protein 1; M-CSF, macrophage colony-stimulating factor; MIP, macrophage inflammatory protein; MIF, macrophage migration inhibitory factor; GRO, growth stimulatory activity..
- Received May 29, 1997.
- Accepted September 16, 1997.
- Copyright © 1998 by The American Association of Immunologists
References
In this issue
