Human Mast Cell Development from Hematopoietic Stem Cells in a Connective Tissue-Equivalent Model

Abstract

Mast cells (MCs) play critical roles in the pathogenesis of IgE- and non-IgE-mediated immune responses, as well as host defense against parasites, bacteria, and viruses. Due to the effect of extracellular matrix components on tissue morphogenesis and cell behavior, utilizing a tissue model that mimics MC microenvironmental conditions in vivo has greater relevance for in vitro studies. For this work, MCs were developed within a connective tissue-equivalent model and cell function was examined in response to an allergen. MCs are located in proximity to fibroblasts and endothelial cells (ECs) that play a role in MC development and maturity. Accordingly, MC progenitors isolated from human peripheral blood were co-cultured with human primary fibroblasts in a 3D collagen matrix to represent the connective tissue. The matrix was coated with type IV collagen and fibronectin before seeding with primary human ECs, representing the capillary wall. The stem cell-derived cells demonstrated MC characteristics, including typical MC morphology, and the expression of cytoplasmic granules and phenotypic markers. Also, the generated cells released histamine in IgE-mediated reactions, showing typical MC functional phenotype in an immediate-type allergenic response. The created tissue model is applicable to a variety of research studies and allergy testing.

Impact Statement

Mast cells (MCs) are key effector and immunoregulatory cells in immune disorders; however, their role is not fully understood. Few studies have investigated human ex vivo MCs in culture, due to the difficulties in isolating large numbers. Our study demonstrates, for the first time, the generation of cells exhibiting MC phenotypic and functional characteristics from hematopoietic stem cells within a connective tissue-equivalent model with ancillary cells. Utilizing the 3D matrix-embedded cells can advance our understanding of MC biological profile and immunoregulatory roles. The tissue model can also be used for studying the mechanism of allergic diseases and other inflammatory disorders.

Introduction

Since the discovery of mast cells (MCs) by Paul Ehrlich in 1878,¹ researchers have attempted to understand the biological and functional characteristics of MCs to elucidate their role in the pathogenesis of innate and adaptive immune responses.² MCs are localized in tissues exposed to the external environment, accordingly positioned to be among the first immune cells that interact with invading pathogens.³ Upon activation, MCs can selectively release the contents of their secretory granules and vesicles, including preformed mediators, such as histamine and neutral proteases, lipid mediators, and an array of de novo synthesized cytokines and chemokines.⁴ Through the expression of such diverse molecules and receptors, MCs exert their effector and immunomodulatory functions on other immune and nonimmune cells, such as fibroblasts and endothelial cells (ECs).

MC histamine, tryptase, and leukotrienes can induce vascular permeability, collagen synthesis, and the influx of effector cells to the inflamed areas.⁵ However, the local microenvironmental factors and the signals they receive from neighboring cells and molecules within the extracellular matrix (ECM) can alter their secretory profile and magnitude. For example, the epithelial cell-derived thymic stromal lymphopoietin, only in synergy with interleukin (IL)-1 and tumor necrosis factor, induces the release of Th2 cytokines by MCs.⁶ Therefore, to fully understand the multifaceted roles of MCs, they should be examined within the context of microenvironmental conditions that closely recapitulate the in vivo physiology.

MCs are derived from circulating stem cells that migrate to vascularized tissues, differentiate, and mature into MCs under the regulation of the specific tissue milieu.⁷Based on their protease contents, human MCs have been traditionally classified into two main subtypes: MC_TC, which contains both tryptase and chymase, and MC_T containing only tryptase.^2,7

Altered site specificity of MC subpopulations when infiltrated during inflammatory reactions along with their intrinsic differences in cytokine content and response to secretagogues and pharmacological agents indicates their distinct pathological roles.^8,9 Although it is not clear whether certain factors direct the heterogeneous commitment of MC precursors, it is postulated that local and systemic environmental factors define MC phenotype and specialize them for particular biological and pathological functions.^10,11 Furthermore, MCs express certain integrins and receptors for extracellular components of connective tissue, including laminin, fibronectin, and vitronectin, which can regulate their localization, distribution, and proliferation in specific tissues.⁷ Consequently, establishing a tissue-equivalent matrix has greater relevance for studying MC ontogeny and plasticity in vitro.

Due to the pivotal roles of in vivo microenvironmental factors, it has been challenging to define a method or approach to study MCs. Although murine models provide valuable information, fundamental differences between human and animal anatomy, cellular biology, and functions limit the applicability of the findings from murine models to human.¹² For example, murine MCs constitutively express FcγRI, while its expression by human MCs requires interferon-γ preincubation.¹³

Due to difficulties in isolating human MCs ex vivo, researchers have shown that the synergy of essential growth factors, including stem cell factor (SCF) and Th2-derived cytokines, or the presence of a layer of feeder cells, such as fibroblasts, renders stem cell differentiation into MC lineage in vitro.^14–16 However, due to the phenotypic and functional plasticity of MCs governed by tissue milieu, the generalization of the findings from a particular population that has been examined under suspension cell culture (2D model) can be misleading and requires further investigation.

Therefore, in this study, the objective was to establish a new culture condition for MC precursors that mimics their microlocalization within the connective tissue. In our previous work, we showed that MCs can be developed from monoculture of their progenitors within a 3D matrix.¹⁷ However, MC precursors reside in proximity to fibroblasts and ECs in tissue, where they receive signaling molecules that contribute to their maturation and emergence of their distinct phenotype.^18,19 Consequently, in this work, their incorporation into the tissue model would be a step toward replicating the MC in vivo microenvironment. Utilizing such a tissue model that provides the condition for direct cell-cell and cell-ECM interactions for MC in vitro studies would shed more light on MC ontogeny, and physiological and pathological roles.

Materials and Methods

Antibodies and reagents

StemSpan media were purchased from STEMCELL Technologies (Vancouver, Canada). Human recombinant SCF, IL-6, and IL-3 were from PeproTech (Rocky Hill, NJ). Defined HyClone fetal bovine serum (FBS) was purchased from GE Health care Life Sciences (Logan, UT). Human dermal fibroblasts treated with mitomycin-C were purchased from Merck Millipore (Billerica, MA), while human umbilical vein ECs were from Promocell (Heidelberg, Germany). Anti-human APC CD117/c-kit (Catalog# 313206, clone 104D2), FITC FcɛRI (Catalog# 334608, clone CRA-1), PE CD31 (Catalog# 303106, clone WM59), PerCP-Cy5.5 CD90 (Catalog# 328118, clone 5E10), and the fluorochrome-conjugated isotype controls, Ms IgG1 (Catalog# 400120, 400112, 400150, clone MOPC-21) and Ms IgG2b (Catalog# 400310, clone MPC-11) were from BioLegend (San Diego, CA). Anti-human PE CD133/2 (Catalog# 130-113-748, clone 293C3) was purchased from MACS Miltenyi Biotec (San Diego, CA). For immunocytochemical staining, mouse anti-human tryptase (Catalog# ab2378, clone AA1) was purchased from Abcam (Cambridge, MA), mouse anti-human chymase (Catalog# MAB1254, clone B7) was from Chemicon International (Temecula, CA), and the secondary antibody PE goat anti-mouse IgG1 (Catalog# SC-3764, polyclonal) was from Santa Cruz Biotechnology (Dallas, TX).

Cell culture

Preparation of samples is shown in Figure 1. Human buffy coats from healthy donors were obtained from Oklahoma Blood Institute (Oklahoma City, OK). MC progenitors, CD133⁺ cells, were isolated from peripheral blood mononuclear cells (PBMCs) using density gradient centrifugation method and magnetic separation technology (MACS Miltenyi Biotec). A type I collagen (Type-1 bovine, Advanced BioMatrix, CA) was used to prepare the gel solution.¹⁷ Progenitor cells from peripheral blood (70,000–75,000 cell/mL) and fibroblasts (mitomycin-C treated to prevent proliferation), at a ratio of 1:0.6 (for the cell activation with the allergen) and 1:1 (for all other studies), were mixed with the collagen solution and added to the wells of a 48-solid well plate (Greiner Bio-One; NC) at 220 μL/cm². After 45 min of incubation at 37°C, 5% CO₂ (defined as “standard conditions”) for the collagen to gel, StemSpan media (300 μL/well) were added to the top of the matrix. Media were supplemented with 100 ng/mL SCF, 50 ng/mL IL-6, and 1 ng/mL IL-3 for the first 3 weeks of culture, and 100 U/mL penicillin/streptomycin (Gibco; CA). Samples were incubated for 7 weeks to allow cell differentiation, proliferation, and maturation. Six weeks postseeding, spent media were aspirated and the gel was coated with 200 μL/cm² mixture of type IV collagen (Advanced BioMatrix; CA) and fibronectin (Alfa Aesar; MA) with final concentrations of 50 and 20 μg/mL, respectively, in media. After at least 2 h of incubation and aspiration of the coating solution, 45,000–55,000 cell/cm² of ECs were mixed with the described medium supplemented with FBS (20%, v/v) and seeded on top of the matrix. From day 1 of seeding, media were changed and the morphology of the cells was monitored by microscopy weekly.

FIG. 1.

Generation of MCs from stem cells within a connective tissue-equivalent matrix. MC, mast cell.

Cell characterization

Cytology, histology, and immunohistochemistry

Metachromatic staining of MC cytoplasmic granules was evaluated cytologically following Giemsa stain using an automated stainer (Ames HemaTek 2000 Stainer). Cell sizes were obtained using micrographs and ImageJ image processing software. Histological evaluation for spatial observations of the tissue model was performed following careful removal of the 3D matrix from the well and fixation in 10% neutral buffered formalin. Following 24 h of fixation, the matrix was routinely processed, paraffin embedded, and sectioned at 4 μm. Sections were either stained with hematoxylin:eosin or submitted to immunohistochemistry for identification of CD117 and CD31. Immunostains were performed on a fee-for-service basis by the Histology Laboratory at North Carolina State University, College of Veterinary Medicine (Raleigh; NC).

Immunophenotype and proliferation

Expression of CD117/c-kit and FcɛRI by the generated MCs was determined by immunofluorescence staining and flow cytometry (BD Accuri C6; BD Biosciences, CA). Cells were collected from the matrix after gel digestion by using 2.2 mg/mL collagenase D (Roche Applied Science; IN) and stained with appropriate antibodies or isotype controls. To sensitize MCs, 15 μg/mL human myeloma IgE (Athens Research & Technology, Athens, GA) was added to the culture media for 24 h before assessing FcɛRI expression. To identify MCs, fibroblasts, ECs, or progenitor cells that did not differentiate were excluded by using their phenotypic markers, CD90 and CD31, respectively.¹⁷ Cell numbers were determined after applying gates to the cell scatters based on the expression of phenotypic markers.

Immunocytochemistry

Intracellular tryptase and chymase were identified by immunocytochemical staining and flow cytometry. The cells were collected from the collagen matrix after gel digestion, added to a V-bottom 96-well plate (Greiner Bio-One; NC), and fixed/permeabilized using a commercial kit (Catalog# 554714; BD Bioscience, CA). Cells were blocked with a buffer containing 1% bovine serum albumin (BSA; Sigma-Aldrich, MO) and 10% goat serum (v/v; Gibco) in permeabilization solution for 1 h. Then, the samples were incubated with the primary antibodies against tryptase and chymase before the secondary antibody, all at 10 μg/mL in the blocking buffer for 30 min at room temperature. For microscopy, the cells were added to a glass slide and labeled with DAPI (Life Technologies, CA). For flow cytometry, cells were incubated in staining buffer containing 0.2% BSA and 0.09% sodium azide in phosphate-buffered saline for 1 h, and stained with anti-CD117, anti-CD90, and anti-CD31 antibodies or appropriate isotype controls for 45 min at 4°C.

Cell activation

Generated cells were sensitized with 15 μg/mL human myeloma IgE for 24 h at 37°C. The free IgE was removed after three rinses with warm Tyrode's solution (Boston BioProducts, MA) supplemented with 100 μg/mL SCF and 50 μg/mL IL-6 before addition of 40 μg/mL monoclonal anti-human IgE (clone G7–18; BD Biosciences, CA) in Tyrode's supplemented solution for 1 h at 37°C. The Tyrode's solution consisted of 10 mM Hepes, 135 mM NaCl, 2.8 mM KCl, 1 mM MgCl₂, 2 mM CaCl₂, 12 mM NaHCO₃, 0.4 mM NaH₂PO₄, 5.5 mM Glucose, and 0.25% BSA. The supernatant and the digested collagen gel were collected, and cellular histamine was measured after cell lysis in water by three freeze and thaw cycles and sonication for 5 min. Commercial ELISA kit (Catalog# BA E-1700; Labor Diagnostika Nord, Nordhorn, Germany) was used for quantifying histamine. Identical samples, but without IgE and anti-IgE, were used as negative control. Release percentage was calculated as the ratio of histamine in media and the digested gel to the total histamine content. In addition, the spontaneous amount of histamine released by the cells in the control group was subtracted from the total amount released.

To investigate the response of the generated MCs to an antigen, Dermatophagoides pteronyssinus (Der p) was selected. Nonallergic and allergic patients with a positive skin prick or intradermal test to Der p extract were included in this part of the study. A wheel size of 3 mm or more in diameter, developed within 15 min was considered a positive response to saline and 1% histamine as negative and positive controls, respectively. The study protocol was approved by Oklahoma State University Institutional Review Board and written informed consent was obtained from each subject. MCs were generated by using the leukocyte preparations of the patients according to the protocol mentioned above. The cells were passively sensitized with 20% (v/v) serum of allergic (Phadia, MI) or nonallergic individuals (PlasmaLab International, WA) for 2 h at 37°C. The specific IgE in the allergic and nonallergic serum was 18–35 and 0.021 kU/L, respectively, validated by ImmunoCAP analysis done by the vendors. The same procedure, as mentioned above for activation of the cells with anti-IgE, was followed afterward, except that the Tyrode's solution used for rinses was not supplemented with cytokines and the Der p extract of mite (HollisterStier, WA) at various concentrations was added to the supplemented Tyrode's solution to induce cell activation. The concentrations and incubation times of the serum and allergen were selected based on the studies done on passive sensitization of MCs and leukocytes.^20,21 A quantitative histamine plate kit (RefLab ApS, Copenhagen N, Denmark) was used to determine the concentration of histamine in the samples. Group of samples sensitized with the serum, but not activated with the allergen, served as a negative control.

Statistical analysis

Experimental results are expressed as mean ± SD of data from three donors with triplicate samples from each donor unless otherwise stated. Student's t-test or Tukey-Kramer method was applied to determine significant difference among the groups. A value of p < 0.05 was considered significant. GraphPad Prism (GraphPad Software, CA) was used for nonlinear regression (curve fit) for histamine measurements and statistical analyses.

Results

Morphology of the generated cells within the connective tissue-equivalent matrix

After seeding, the cells were evenly distributed within the matrix, while the MC precursors formed colonies as early as the first week of culture (Fig. 2A). The generated cells remained in colonies and spread within the matrix as the cell proliferation progressed. The cells were round or oval, as well as tailed, showing their mobility within the matrix, as highlighted in Figure 2A, week 1. The thickness of the matrix at the well center was 1.6 ± 0.1 mm and allowed the diffusion of the media to the bottom of the matrix, as evidenced by the survival and growth of the cells at the bottom of the well (Fig. 2A, Week 7, bottom micrograph). Seven weeks postseeding, around 80% of the round cells were 9–15 μm in diameter, which is in the range of in vivo MC size (data from four donors, n = 60 ± 20 cells per donor).^22,23 Furthermore, the ECs appear to form a confluent layer on the matrix within 36 h postseeding, as observed by microscopic examination (Fig. 2B). The histologic analysis of the collagen matrix confirmed a connective tissue-equivalent matrix (CTEM), with a monolayer of CD31⁺ ECs on the apical surface and the CD117⁺ cells along with the fibroblasts spread within the subendothelium (Fig. 2C–E). Although progenitor cells are not granular, the generated cells exhibited granular morphology with 70–90% of the granular cells mononucleated, as is the normal morphology of mature MCs (Fig. 2F, data from two donors, n = 220 and 500 cells).

FIG. 2.

Morphology of the seeded cells within the connective tissue-equivalent matrix. (A) The generated cells and fibroblasts 2 days and 1–7 weeks post seeding within the matrix. (B) The seeded matrix with ECs. Histologic analysis of the matrix stained with (C) hematoxylin and eosin; (D) anti-CD117; and (E) anti-CD31 antibodies. Images shown are representative of the model. (F) Wright-Giemsa staining (left) of freshly isolated cells from peripheral blood that are devoid of granules and (right) generated cells after 7 weeks in co-culture that acquired the granules. Some of the MC precursors and fibroblasts are highlighted by black and white arrows, respectively, and ECs are highlighted by the arrowheads. EC, endothelial cell.

Immunophenotype of the generated cells within the CTEM

CD133 antigen has been used as a marker of hematopoietic progenitor and stem cells.²⁴ The frequency of isolated cells from human PBMCs was 0.8% ± 0.5% (from 20 donors). The purity of CD133⁺ cells within the magnetically isolated cells was less than 50%. As shown in Figure 3A, the CD133⁺ cells before culture were more than 80% and around 40% positive for CD31 and CD117, respectively, while only less than 10% were expressing FcɛRI. After excluding fibroblasts and ECs or nondifferentiated progenitor expressing CD31 (Fig. 3B), the CD117⁺ CD31⁻ CD90⁻ cells were 3.2 ± 1.3- to 8.4 ± 0.1-fold higher in number than the seeded cells. Also, the expression of FcɛRI was enhanced after 24 h of incubation with the IgE antibody, with 5.8 ± 1.3-fold higher mean fluorescence intensity (MFI, p < 0.01), which reflects the upregulation of the surface receptor density. In fact, IgE incubation resulted in the expression of FcɛRI by 31.9% ± 3.6% of CD117⁺ cells (Fig. 3C, data selected from three independent donors, n = 3 in each experiment). Furthermore, almost all the generated CD117⁺ cells were stained for tryptase and chymase granules (98.7% ± 0.8% and 96.7% ± 2.6%, respectively), exhibiting MC_TC phenotype (Fig. 3C).

FIG. 3.

Expression of MC phenotypic markers by the generated cells within the connective tissue-equivalent matrix. (A) Representative density plots of the expression of the markers used to identify MCs by CD133⁺ cells isolated from peripheral blood before co-culture. (B) Gating scheme and representative density plots of marker expressions by the generated cells after co-culture. To identify the CD117-expressing cells, anti-CD90 and CD31 antibodies were used to gate out fibroblasts and ECs or progenitor cells, respectively. (C) The expression of FcɛRI, tryptase, and chymase (gray histogram) by the CD117-positive cells is compared with the FMO control (black histogram). (D) Expression of (top) CD117 and (bottom) tryptase determined by immunofluorescence staining. Cells were stained with anti-tryptase or CD117 and PE-secondary antibody. Nuclei were stained with DAPI. The figure shows the mixed population of the cell types and only MCs stained positive. FMO, fluorescence minus one.

Activation of the generated cells within the CTEM in an IgE-mediated challenge

After 7 weeks in co-culture, histamine content of the generated cells was 4.4 ± 0.8 pg/cell and their histamine spontaneous release was 2.35% ± 0.9% of their total histamine content. Function of the generated MCs was examined after challenge with IgE and anti-IgE, which is known to cause histamine release upon FcɛRI trigger.²⁵ Upon sensitization and activation with IgE and anti-IgE, the cells released 31.2% ± 2.0% of their total histamine content, which was 13.3 ± 5.3-fold higher than the spontaneous release of histamine (p < 0.001). Around 80–90% of the released histamine was detected within the matrix. When passively sensitized with allergic serum, the cells dose dependently released histamine in response to Der p allergen (Fig. 4). Since the extract of the allergen contained glycerol as a preservative, identical concentrations of the glycerol were also tested, but did not induce histamine release (data not shown). The histamine release in response to 15 AU/mL of the allergen, which showed the highest average, was 18.09 ± 4.10-fold higher than the control group that was not activated with the allergen (p < 0.001). In contrast, the cells did not respond to the allergen when serum from a nonallergic individual that did not contain specific IgE to Der p allergen was used.

FIG. 4.

Histamine release by the generated MCs in response to Der p allergen within the connective tissue-equivalent matrix. The cells were passively sensitized with human serum before activation with the allergen. All the data from allergic serum are significantly higher than the nonactivated samples (p < 0.01). Data are mean ± SD. *Indicates p < 0.05.

Discussion

CD34⁺ and CD133⁺ stem cells or mononuclear cells have been previously used to generate MCs under 2D culture conditions for in vitro studies.^26–28 Depending on the cell source and culture conditions, specifically the culture media supplements, the characteristics of the obtained MCs varied,^14,16,27 for example, in contrast to IL-10, IL-4 enhanced FcɛRI expression and cell function.^27,29 Given that the cells within a matrix are surrounded with the ECM components that can alter the distribution and morphological and physiological characteristics of the cells, 2D models may lack the needed geometry and components to mimic the in vivo.^30,31

A 3D matrix can regulate commitment of stem cells to a specific lineage and enhance their maturity.^32,33 Therefore, in this study, we have established a novel culture method for the generation of functional MCs from peripheral blood stem cells within a CTEM that comprised multiple cell types. We have used type I collagen, which is the most predominant fibrillar component of the ECM found in the dermis and interstitial tissue, to create a matrix.³⁴ The matrix was coated with type IV collagen and fibronectin to mimic the basement membrane, enhancing the EC growth and attachment to the matrix.^35,36 The microarchitecture of the matrix allowed the cell motility within the collagen matrix, as evidenced by colony formation and tailed-like cells (Fig. 2A).

We have previously determined that StemSpan medium supports the generation of MCs from monoculture of CD133⁺ cells in a matrix and growth of fibroblasts and ECs.¹⁷ Serum was a prerequisite for EC survival, while its addition to the media from the beginning of culture suppressed MC development.¹⁷ When added in the seventh or eighth week of culture, serum induced FcɛRI expression and histamine release by the generated cells.^37,38 Therefore, for the co-culture model in this work, ECs were added in the seventh week, which was the same time that media were supplemented with serum.

MCs are distributed near blood vessels and fibroblasts that are abundant in connective tissue. Both fibroblasts and ECs release growth factors, such as SCF and IL-6, which support MC survival and growth.^17,39,40 When 3T3 fibroblasts were in co-culture with human cord blood and human or mice bone marrow-derived MCs, the differentiation and proliferation of MC precursors were enhanced.^41–43 In addition, ECs promote MC proliferation more effectively than SCF-supplemented media through interaction between CD117 and very late antigen-4 on MCs and membrane form of SCF (mSCF) and vascular cell adhesion protein-1 on ECs, respectively.¹⁹

When compared with the 3D tissue model without ancillary cells,¹⁷ even with seven times lower progenitor cell seeding concentration, in co-culture with fibroblasts and ECs, not only was the cell yield significantly higher but also the histamine content was augmented (p < 0.01). Similar results have been observed for mice bone marrow-derived MCs when co-cultured with 3T3 fibroblasts in a culture medium containing IL-3.⁴¹ Although fibroblasts were shown to support the survival of MCs, even when separated, the direct cellular interactions induced the maturity and regulated the effector functions of MCs.^44–47 As an example, activated MCs isolated from the human lung in co-culture with fibroblasts released elevated levels of histamine in a cell contact-dependent manner.⁴⁶ The results in this work were similar to the previous findings as the generated cells in co-culture released approximately two-fold higher histamine compared with the 3D model in the absence of ancillary cells,¹⁷ in response to anti-IgE (p < 0.001).

Soluble SCF in synergy with mSCF expressed by fibroblasts and other fibroblast-derived proteins, such as IL-33, may modulate MC maturity and function.^44,47 The mSCF also induces MC attachment to fibroblasts, ECs, and basement membrane proteins and may play a role in MC migration and localization of progenitor cells bearing its agonist, CD117, in connective tissue.^48,49 Furthermore, when in direct contact with fibroblasts, MCs were responsive to eosinophil major basic protein and increased eotaxin, an eosinophil chemotactic factor, all indicating the importance of cellular interactions in leukocyte infiltration during inflammatory responses and utilizing co-culture models for in vitro studies.^45,46

In vitro studies have shown that stem cells can differentiate into MC_T or MC_TC phenotype based on the chymase expression. When human or mice cord blood- and mice fetal liver-derived MCs were in co-culture with fibroblasts, they appeared as MC_TC and MC_T, respectively,^18,50,51 suggesting that progenitor cells might be committed to a specific MC phenotype. However, human cord blood and peripheral blood CD34⁺ cells in the presence of SCF, IL-6, and IL-3 were shown to differentiate into both subtypes,^52,53 while the presence of fibroblasts and IL-4 favored the generation of MC_TC,^40,54 indicating that cell microenvironment and specific growth factors can influence the phenotype of the generated MCs. Therefore, the ultimate MC phenotype in vivo may be regulated under the influence of local tissue milieu that specializes the generated MCs to perform certain functions. For example, the proliferation of mucosal MCs (MMC) in mice, which are mainly chymase negative, caused by helminth infection suggests their role in defense against parasites.⁵⁵

In our previous work, in the absence of ancillary cells, the developed MCs within the 3D matrix appeared as MC_TC subtype.¹⁷ In this work, in co-culture with fibroblasts and ECs, their phenotype was preserved, indicating that factors besides the cellular interactions, such as ECM proteins, can determine the fate of MC phenotype. In addition, previous findings suggest that human MC subtypes are not restricted to a specific anatomical location and may be developed, be recruited, or dynamically change as a result of inflammation, infection, interaction with neighboring cells, and released cytokines.⁵⁰ As an example, MCs in the respiratory system are site-specifically heterogeneous,⁵⁶ and in asthmatic patients, recruitment of MCs, particularly of chymase-positive phenotype to airway smooth muscle and small airway, has been observed.⁵⁷ Furthermore, human MCs generated from cord blood after co-culture with epithelial cells were shown to undergo a transition from MC_TC to MC_T type,⁵⁸ while isolated intestine MCs in co-culture with ECs upregulated the expression of chymase, indicating the development of MC_TC.¹⁹ This transition matches with the population of MCs located in proximity to epithelial cells in alveoli and bronchi (MC_T) and around blood vessels (MC_TC).⁵⁹

The result of our study is also in line with these findings as the morphological phenotype of the developed cells within the CTEM matched with their counterparts in skin and submucosa (MC_TC).² This has also been confirmed in vivo where MMCs generated from bone marrow were transferred into the peritoneal cavity of mice and acquired the features of resident connective tissue MCs. Also, when intravenously injected, MMCs could develop into both subtypes in MC-deficient mice in accordance with their anatomical location,⁶⁰ all of which signifies the complexity of MC ontogeny and plasticity that must be clarified within the context of in vitro tissue-equivalent models.

To compare the characteristics of the cells obtained in this study with previous protocols, a 2D culture method that generated MCs in a more similar condition than other studies to our 3D model was selected.^27,38 In this study, considering that the cells were expanded and maintained within a matrix, using more than seven times lower concentration of progenitors, the yield of generated CD117⁺ MCs (4.6 ± 2.4) was higher on average than the selected 2D method (3.2 ± 1) for granular cells and CD117 expression of 88.3% ± 2.2%.

The size and the histamine content of MCs developed within the CTEM were within the range of in vivo MCs isolated from human skin, intestine, or lung, with around 2–5 pg histamine/cell,^23,61,62 while it was higher for the selected 2D culture method (23.3 ± 3.3 pg/cell). This indicates that the cells in 3D culture condition can be morphologically different from cells in 2D.

Although FcɛRI has been considered a phenotypic marker of MCs, human MC subtypes can express IgE receptor at different levels.⁵⁶ Given that IgE antibody upregulates the expression of FcɛRI, as has also been observed in this work, it can facilitate the detection of IgE receptors.⁶³ After sensitization with IgE, the expression of FcɛRI for the generated cells within the CTEM was comparable with the selected 2D model.

To examine the function of the generated cells, the release of histamine in an anti-IgE-mediated reaction was measured. As shown in previous studies, the response of MCs depends on the concentration of anti-IgE used for their activation and the activation protocol.^27,37,64 For this study, the response of the generated cells to a selected concentration of anti-IgE (40 μg/mL) is shown.

According to our previous observation,¹⁷ the selected concentration of anti-IgE is not high enough to induce the maximum histamine release within a matrix. Therefore, the histamine released (31.2% ± 2.0%) seems to be lower than the selected 2D model (52.9% ± 2.5%) that utilized an optimal concentration of anti-IgE to obtain the maximum histamine release. It is of great importance to highlight that when we examined the function of the cells that were generated within the matrix, but removed from it before activation, the histamine degranulation was comparable to the cells developed under 2D culture conditions.¹⁷This indicates that the reduced histamine release in the matrix was not because the generated cells were functionally impaired, but rather the effect of anti-IgE was attenuated by the cell microenvironment.

Reduced levels of histamine release in response to anti-IgE for MCs in lung fragments compared with dispersed cells from the same tissue have also been observed.^65,66 Furthermore, the response varies with the tissue type that MCs were isolated from, possibly due to their functional heterogeneity.^61,67 Therefore, one cannot expect the same level of histamine release from MCs tested ex vivo or generated in vitro, besides the discrepancies in the activation protocols. Nevertheless, our main objective was to demonstrate that the generated cells can also degranulate in an immediate hypersensitivity reaction.

In contrast with serum of a nonallergic patient, the MCs, regardless of whether they were generated from peripheral blood of allergic or nonallergic individuals, were passively sensitized with allergic serum and released histamine in response to Der p allergen. The response curve is analogous to the response of passively sensitized MCs isolated from human tissue to dust mite or grass pollen allergen.^64,68 With increasing the concentration of the antigen, the histamine release reaches a maximum level at 15 AU/mL. Due to the saturation of antibody-binding sites or formation of clusters of crosslinked receptors (patching and capping), the histamine release significantly decreases in response to 1500 AU/mL (p < 0.05), resulting in a bell-shaped response curve.⁶⁹

It is noteworthy that in the control group that was also sensitized with the serum, but not activated with the allergen, when the allergic serum was used, the histamine release was marginally higher than nonallergic serum (p < 0.01). In fact, previous studies have shown that some type of IgEs called cytokinergic IgEs can induce histamine and cytokine release,^70,71 and the results of this study indicate that human allergic serum can activate MCs in the absence of an antigen.

MCs may play critical roles in the pathogenesis of IgE- and non-IgE-mediated immune responses, as well as host defense against parasites, bacteria, and viruses.^2,50 In this study, we have shown that progenitor-derived MCs within a CTEM are able to respond in IgE-mediated reactions, specifically to an allergen, and therefore may serve as a novel tool in clinical settings for diagnosis of allergic diseases⁷² or pharmacological studies. Indeed, previous studies on immunotherapy had shown that cells in 3D culture conditions exhibit a comparable resistance to drugs seen in humans than 2D culture.⁷³ Therefore, 3D models may be widely used in clinical trials due to a higher potential in predicting in vivo cellular responses.

Furthermore, the developed tissue model can be used to study the mechanisms of allergic diseases and other inflammatory disorders. As an example, MC histamine and tumor necrosis factor-α can stimulate ECs to release proinflammatory cytokines and express adhesion molecules,^74,75 suggesting the role of MCs in the late phase of inflammatory reactions, which can be demonstrated using the developed tissue model in this work. Furthermore, the release of fibrogenic, fibrolytic, and proangiogenic factors, such as tryptase, transforming growth factor-β, heparin, and vascular endothelial growth factor, suggests MC role in fibrosis, tissue remodeling, and vascular diseases.^76,77

MC co-culture with fibroblasts or smooth muscle cells was shown to induce collagen synthesis as well as collagen contraction, which may modulate the infiltration of inflammatory cells in asthmatic airways and fibrotic tissues.^39,78,79 ECs alone or in co-culture with fibroblasts can also induce collagen contraction.^80,81 Likewise, we have observed contraction of the collagen matrix depending on the number of seeded fibroblasts and generated MCs, especially after the gel was seeded with ECs. Therefore, the co-culture of all three cell types may be required for a more comprehensive study of the mechanism of tissue remodeling and wound healing. Many of the findings in this area have not been demonstrated by using primary human cells or direct cellular interactions, which occur continuously in the human body. Therefore, with the advent of 3D culture method, which provides the condition for cell motility and migration, interaction with multiple cells types, and ECM proteins, numerous hypotheses that remained unanswered or controversial can be examined.

Footnotes

Acknowledgments

The authors are grateful to Dr. Amy L. Darter and her laboratory at the Oklahoma Institute of Allergy, Asthma & Immunology for performing skin allergy tests on patients. This work was supported by a grant from National Science Foundation CAREER Award 1150831.

Authors' Contribution

T.D. designed and performed the experiments, analyzed and interpreted the data, and wrote the original draft. R.B. and J.H.M. advised on data presentation. J.H.M. and J.W.R. performed the cytological, histological, and immunohistochemical observations. H.F. designed and supervised the research. All the authors reviewed and edited the article.

Disclosure Statement

The authors declare no conflict of interests.

References

Ehrlich

Select contributions to the theory and practice of histological staining [In German]. The Collected Papers of Paul Ehrlich. Elsevier, 1956, p. 29.

Reber

L.L.

, Sibilano

, Mukai

, and Galli

S.J.

Potential effector and immunoregulatory functions of mast cells in mucosal immunity. Mucosal Immunol, 8, 444, 2015.

Galli

S.J.

, Nakae

, and Tsai

Mast cells in the development of adaptive immune responses. Nat Immunol, 6, 135, 2005.

Theoharides

T.C.

, Alysandratos

K.-D.

, Angelidou

, et al. Mast cells and inflammation. Biochim Biophys Acta, 1822, 21, 2012.

Galli

S.J.

, Grimbaldeston

, and Tsai

Immunomodulatory mast cells: negative, as well as positive, regulators of immunity. Nat Rev Immunol, 8, 478, 2008.

Allakhverdi

, Comeau

M.R.

, Jessup

H.K.

, et al. Thymic stromal lymphopoietin is released by human epithelial cells in response to microbes, trauma, or inflammation and potently activates mast cells. J Exp Med, 204, 253, 2007.

Metcalfe

D.D.

, Baram

, and Mekori

Y.A.

Mast cells. Physiol Rev, 77, 1033, 1997.

Irani

A.M.

, and Schwartz

L.B.

Human mast cell heterogeneity. Allergy Proc, 15, 303, 1994.

Bradding

, Okayama

, Howarth

P.H.

, Church

M.K.

, and Holgate

S.T.

Heterogeneity of human mast cells based on cytokine content. J Immunol (Baltimore, Md: 1950), 155, 297, 1995.

10.

Shea-Donohue

, Stiltz

, Zhao

, and Notari

Mast cells. Curr Gastroenterol Rep, 12, 349, 2010.

11.

Marshall

J.S.

Mast-cell responses to pathogens. Nat Rev Immunol, 4, 787, 2004.

12.

Bischoff

S.C.

Role of mast cells in allergic and non-allergic immune responses: comparison of human and murine data. Nat Rev Immunol, 7, 93, 2007.

13.

Okayama

, Hagaman

D.D.

, and Metcalfe

D.D.

A comparison of mediators released or generated by IFN-γ-treated human mast cells following aggregation of FcγRI or FcɛRI. J Immunol, 166, 4705, 2001.

14.

Rennick

, Hunte

, Holland

, and Thompson-Snipes

Cofactors are essential for stem cell factor-dependent growth and maturation of mast cell progenitors: comparative effects of interleukin-3 (IL-3), IL-4, IL-10, and fibroblasts. Blood, 85, 57, 1995.

15.

Yanagida

, Fukamachi

, Ohgami

, et al. Effects of T-helper 2-type cytokines, interleukin-3 (IL-3), IL-4, IL-5, and IL-6 on the survival of cultured human mast cells. Blood, 86, 3705, 1995.

16.

Liu

, Liu

, Li

, and Wu

Molecular regulation of mast cell development and maturation. Mol Biol Rep, 37, 1993, 2010.

17.

Derakhshan

, Bhowmick

, Ritchey

J.W.

, and Gappa-Fahlenkamp

Development of human mast cells from hematopoietic stem cells within a 3D collagen matrix: effect of stem cell media on mast cell generation. Stem Cells Int, 2018, 14, 2018.

18.

Irani

A.A.

, Craig

S.S.

, Nilsson

, Ishizaka

, and Schwartz

L.B.

Characterization of human mast cells developed in vitro from fetal liver cells cocultured with murine 3T3 fibroblasts. Immunology, 77, 136, 1992.

19.

Mierke

C.T.

, Ballmaier

, Werner

, Manns

M.P.

, Welte

, and Bischoff

S.C.

Human endothelial cells regulate survival and proliferation of human mast cells. J Exp Med, 192, 801, 2000.

20.

Levy

D.A.

, and Osler

A.G.

Studies on the mechanisms of hypersensitivity phenomena. XVI. In vitro assays of reaginic activity in human sera: effect of therapeutic immunization on seasonal titer changes. J Immunol (Baltimore, Md: 1950), 99, 1068, 1967.

21.

Nolte

, Kruse

, Stahl Skov

, and Schiøtz

P.O.

Passive sensitization of human intestinal mast cells. Agents And Actions, 27, 93, 1989.

22.

Schulman

E.S.

, Kagey-Sobotka

, MacGlashan

D.W.

Jr. , et al. Heterogeneity of human mast cells. J Immunol (Baltimore, Md: 1950), 131, 1936, 1983.

23.

Benyon

R.C.

, Lowman

M.A.

, and Church

M.K.

Human skin mast cells: their dispersion, purification, and secretory characterization. J Immunol (Baltimore, Md: 1950), 138, 861, 1987.

24.

CD133: a stem cell biomarker and beyond. Exp Hematol Oncol, 2, 17, 2013.

25.

Krishnaswamy

, and Chi

D.S.

Mast Cells: Methods and Protocols. Methods in Molecular Biology. Totowa, NJ: Humana Press, 2005.

26.

Yin

, Bai

, Olivera

, Desai

, and Metcalfe

D.D.

An optimized protocol for the generation and functional analysis of human mast cells from CD34(+) enriched cell populations. J Immunol Methods, 448, 105, 2017.

27.

Andersen

H.B.

, Holm

, Hetland

T.E.

, et al. Comparison of short term in vitro cultured human mast cells from different progenitors - Peripheral blood-derived progenitors generate highly mature and functional mast cells. J Immunol Methods, 336, 166, 2008.

28.

Iida

, Matsumoto

, Tomita

, et al. Selective down-regulation of high-affinity IgE receptor (FcepsilonRI) alpha-chain messenger RNA among transcriptome in cord blood-derived versus adult peripheral blood-derived cultured human mast cells. Blood, 97, 1016, 2001.

29.

Kennedy Norton

, Barnstein

, Brenzovich

, et al. IL-10 suppresses mast cell IgE receptor expression and signaling in vitro and in vivo. J Immunol, 180, 2848, 2008.

30.

G.N.

, Livi

L.L.

, Gourd

C.M.

, Deweerd

E.S.

, and Hoffman-Kim

Genomic and morphological changes of neuroblastoma cells in response to three-dimensional matrices. Tissue Eng, 13, 1035, 2007.

31.

Sun

, Jackson

, Haycock

J.W.

, and MacNeil

Culture of skin cells in 3D rather than 2D improves their ability to survive exposure to cytotoxic agents. J Biotechnol, 122, 372, 2006.

32.

Engler

A.J.

, Sen

, Sweeney

H.L.

, and Discher

D.E.

Matrix elasticity directs stem cell lineage specification. Cell, 126, 677, 2006.

33.

Fong

A.H.

, Romero-López

, Heylman

C.M.

, et al. Three-dimensional adult cardiac extracellular matrix promotes maturation of human induced pluripotent stem cell-derived cardiomyocytes. Tissue Eng Part A, 22, 1016, 2016.

34.

Henriksen

, and Karsdal

M.A.

Chapter 1 - Type I Collagen. In: Karsdal

M.A.

, ed. Biochemistry of Collagens, Laminins and Elastin. Amsterdam, The Netherlands: Academic Press, 2016, p. 1.

35.

Jayadev

, and Sherwood

D.R.

Basement membranes. Curr Biol, 27, R207, 2017.

36.

Heo

, Shin

Y.M.

, Lee

Y.B.

, Lim

Y.M.

, and Shin

Effect of immobilized collagen type IV on biological properties of endothelial cells for the enhanced endothelialization of synthetic vascular graft materials. Colloids Surf B Biointerfaces, 134, 196, 2015.

37.

Dahl

, Saito

, Nielsen

H.V.

, and Schiøtz

P.O.

The establishment of a combined serum-free and serum-supplemented culture method of obtaining functional cord blood-derived human mast cells. J Immunol Methods, 262, 137, 2002.

38.

Holm

, Andersen

H.B.

, Hetland

T.E.

, et al. Seven week culture of functional human mast cells from buffy coat preparations. J Immunol Methods, 336, 213, 2008.

39.

Montier

, Lorentz

, Krämer

, et al. Central role of IL-6 and MMP-1 for cross talk between human intestinal mast cells and human intestinal fibroblasts. Immunobiology, 217, 912, 2012.

40.

Nakahata

, and Toru

Cytokines regulate development of human mast cells from hematopoietic progenitors. Int J Hematol, 75, 350, 2002.

41.

Levi-Schaffer

, Austen

K.F.

, Gravallese

P.M.

, and Stevens

R.L.

Coculture of interleukin 3-dependent mouse mast cells with fibroblasts results in a phenotypic change of the mast cells. Proc Natl Acad Sci U S A, 83, 6485, 1986.

42.

Levi-Schaffer

, Kelav-Appelbaum

, and Rubinchik

Human foreskin mast cell viability and functional activity is maintained ex vivo by coculture with fibroblasts. Cell Immunol, 162, 211, 1995.

43.

Jozaki

, Kuriu

, Hirota

, et al. Bone marrow-derived cultured mast cells and peritoneal mast cells as targets of a growth activity secreted by BALB/3T3 fibroblasts. Exp Hematol, 19, 185, 1991.

44.

Gurish

M.F.

, and Austen

K.F.

Developmental origin and functional specialization of mast cell subsets. Immunity, 37, 25, 2012.

45.

Piliponsky

A.M.

, Gleich

G.J.

, Nagler

, Bar

, and Levi-Schaffer

Non-IgE-dependent activation of human lung- and cord blood-derived mast cells is induced by eosinophil major basic protein and modulated by the membrane form of stem cell factor. Blood, 101, 1898, 2003.

46.

Hogaboam

, Kunkel

S.L.

, Strieter

R.M.

, et al. Novel role of transmembrane SCF for mast cell activation and eotaxin production in mast cell-fibroblast interactions. J Immunol (Baltimore, Md: 1950), 160, 6166, 1998.

47.

Dvorak

A.M.

, Mitsui

, and Ishizaka

Stimulation of partial development of human mast cells by supernatant fluid from mouse fibroblast cultures. Clin Exp Allergy, 24, 649, 1994.

48.

Adachi

, Ebi

, Nishikawa

, et al. Necessity of extracellular domain of W (c-kit) receptors for attachment of murine cultured mast cells to fibroblasts. Blood, 79, 650, 1992.

49.

Dastych

, and Metcalfe

D.D.

Stem cell factor induces mast cell adhesion to fibronectin. J Immunol (Baltimore, Md: 1950), 152, 213, 1994.

50.

Moon

T.C.

, St Laurent

C.D.

, Morris

K.E.

, et al. Advances in mast cell biology: new understanding of heterogeneity and function. Mucosal Immunol, 3, 111, 2010.

51.

Rubinchik

, and Levi-Schaffer

Mast cells and fibroblasts: two interacting cells. Int J Clin Lab Res, 24, 139, 1994.

52.

Lappalainen

, Lindstedt

K.A.

, and Kovanen

P.T.

A protocol for generating high numbers of mature and functional human mast cells from peripheral blood. Clin Exp Allergy, 37, 1404, 2007.

53.

Ahn

, Takai

, Pawankar

, et al. Regulation of chymase production in human mast cell progenitors. J Allergy Clin Immunol, 106, 321, 2000.

54.

Toru

, Eguchi

, Matsumoto

, Yanagida

, Yata

, and Nakahata

Interleukin-4 promotes the development of tryptase and chymase double-positive human mast cells accompanied by cell maturation. Blood, 91, 187, 1998.

55.

Madden

K.B.

, Urban

J.F.

Jr. , Ziltener

H.J.

, Schrader

J.W.

, Finkelman

F.D.

, and Katona

I.M.

Antibodies to IL-3 and IL-4 suppress helminth-induced intestinal mastocytosis. J Immunol (Baltimore, Md: 1950), 147, 1387, 1991.

56.

Andersson

C.K.

, Mori

, Bjermer

, Löfdahl

C.G.

, and Erjefält

J.S.

Novel site-specific mast cell subpopulations in the human lung. Thorax, 64, 297, 2009.

57.

Balzar

, Chu

H.W.

, Strand

, and Wenzel

Relationship of small airway chymase-positive mast cells and lung function in severe asthma. Am J Respir Critic Care Med, 171, 431, 2005.

58.

Hsieh

F.H.

, Sharma

, Gibbons

, Goggans

, Erzurum

S.C.

, and Haque

S.J.

Human airway epithelial cell determinants of survival and functional phenotype for primary human mast cells. Proc Natl Acad Sci U S A, 102, 14380, 2005.

59.

Gilfillan

A.M.

, and Metcalfe

D.D.

Mast Cell Biology: Contemporary and Emerging Topics/[Electronic Resource]. New York; Austin, TX: Springer Science+Business Media, 2011.

60.

Nakano

, Sonoda

, Hayashi

, et al. Fate of bone marrow-derived cultured mast cells after intracutaneous, intraperitoneal, and intravenous transfer into genetically mast cell-deficient W/Wv mice. Evidence that cultured mast cells can give rise to both connective tissue type and mucosal mast cells. J Exp Med, 162, 1025, 1985.

61.

Lowman

M.A.

, Rees

P.H.

, Benyon

R.C.

, and Church

M.K.

Human mast cell heterogeneity: histamine release from mast cells dispersed from skin, lung, adenoids, tonsils, and colon in response to IgE-dependent and nonimmunologic stimuli. J Allergy Clin Immunol, 81, 590, 1988.

62.

Schulman

E.S.

, Post

T.J.

, and Vigderman

R.J.

Density heterogeneity of human lung mast cells. J Allergy Clin Immunol, 82, 78, 1988.

63.

Yamaguchi

, Lantz

C.S.

, Oettgen

H.C.

, et al. IgE enhances mouse mast cell Fc(epsilon)RI expression in vitro and in vivo: evidence for a novel amplification mechanism in IgE-dependent reactions. J Exp Med, 185, 663, 1997.

64.

Gospos

, Dreikhausen

, Dartsch

D.C.

, Szamel

, and Hockertz

Development of an allergy test model: activation of human mast cells with potentially allergenic substances. Toxicology, 166, 91, 2001.

65.

Ennis

Histamine release from human pulmonary mast cells. Agents Actions, 12, 60, 1982.

66.

Paterson

N.A.

, Wasserman

S.I.

, Said

J.W.

, and Austen

K.F.

Release of chemical mediators from partially purified human lung mast cells. J Immunol (Baltimore, Md: 1950), 117, 1356, 1976.

67.

Church

M., RC., B., PH., R., MA., L.

, and Campbell

A.M.

Functional Heterogeneity of Human Mast Cells. In: Austen

S.J. G

. K.F., eds. Mast Cell and Basophil Differentiation and Function in Health and Disease. New York: Raven Press, 1989, p. 161.

68.

Coleman

J.W.

, and Godfrey

R.C.

The number and affinity of IgE receptors on dispersed human lung mast cells. Immunology, 44, 859, 1981.

69.

Dembo

, Goldstein

, Sobotka

A.K.

, and Lichtenstein

L.M.

Histamine release due to bivalent penicilloyl haptens control by the number of cross-linked IgE antibodies on the basophil plasma membrane. J Immunol, 121, 354, 1978.

70.

Kitaura

, Song

, Tsai

, et al. Evidence that IgE molecules mediate a spectrum of effects on mast cell survival and activation via aggregation of the FcɛRI. Proc Natl Acad Sci USA, 100, 12911, 2003.

71.

Bax

H.J.

, Keeble

A.H.

, and Gould

H.J.

Cytokinergic IgE Action in Mast Cell Activation. Front Immunol, 3, 229, 2012.

72.

Bahri

, Custovic

, Korosec

, et al. Mast cell activation test in the diagnosis of allergic disease and anaphylaxis. J Allergy Clin Immunol, 2018.

73.

Edmondson

, Broglie

J.J.

, Adcock

A.F.

, and Yang

Three-dimensional cell culture systems and their applications in drug discovery and cell-based biosensors. Assay Drug Dev Technol, 12, 207, 2014.

74.

Zhang

, Alcaide

, Liu

, et al. Regulation of endothelial cell adhesion molecule expression by mast cells, macrophages, and neutrophils. PLoS One, 6, e14525, 2011.

75.

, Chi

, Stechschulte

D.J.

, and Dileepan

K.N.

Histamine-induced production of interleukin-6 and interleukin-8 by human coronary artery endothelial cells is enhanced by endotoxin and tumor necrosis factor-alpha. Microvasc Res, 61, 253, 2001.

76.

Krishnaswamy

, Ajitawi

, and Chi

D.S.

The human mast cell: an overview. Methods Mol Biol, 315, 13, 2006.

77.

Krystel-Whittemore

, Dileepan

K.N.

, and Wood

J.G.

Mast cell: a multi-functional master cell. Front Immunol, 6, 2016.

78.

Abel

, and Vliagoftis

Mast cell-fibroblast interactions induce matrix metalloproteinase-9 release from fibroblasts: role for IgE-mediated mast cell activation. J Immunol (Baltimore, Md: 1950), 180, 3543, 2008.

79.

Ceresa

C.C.

, Knox

A.J.

, and Johnson

S.R.

Use of a three-dimensional cell culture model to study airway smooth muscle-mast cell interactions in airway remodeling. Am J Physiol Lung Cell Mol Physiol, 296, L1059, 2009.

80.

Vernon

R.B.

, and Sage

E.H.

Contraction of fibrillar type I collagen by endothelial cells: a study in vitro. J Cell Biochem, 60, 185, 1996.

81.

Villaschi

, and Nicosia

R.F.

Paracrine interactions between fibroblasts and endothelial cells in a serum-free coculture model. Modulation of angiogenesis and collagen gel contraction. Lab Invest, 71, 291, 1994.