FASEB J. Pierce now sold as Thermo Scientific
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published as doi: 10.1096/fj.06-7051com.
 (The FASEB Journal. 2007;21:1515-1526.)
© 2007 FASEB
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
fj.06-7051comv1
21/7/1515    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Methe, H.
Right arrow Articles by Edelman, E. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Methe, H.
Right arrow Articles by Edelman, E. R.

Matrix adherence of endothelial cells attenuates immune reactivity: induction of hyporesponsiveness in allo- and xenogeneic models

Heiko Methe*,1, Adam Groothuis*, Mohamed H. Sayegh{dagger} and Elazer R. Edelman*,{ddagger}

* Harvard-MIT Division of Health Sciences and Technology, Massachusetts Institute of Technology, Cambridge, Massachusetts, USA;

{dagger} Transplantation Research Center, Brigham and Women’s Hospital and Children’s Hospital, Harvard Medical School, Boston, Massachusetts, USA, and;

{ddagger} Cardiovascular Division, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts, USA

1Correspondence: Harvard-MIT Division of Health Sciences and Technology; Massachusetts Institute of Technology, 77 Massachusetts Ave., Bldg 56–322, Cambridge, MA, 02139 USA. E-mail: hmethe{at}mit.edu


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Endothelial integrity regulates vascular tone, luminal patency, and the immune reactivity to tissue grafts. Endothelial dysfunction is the first marker and site of disease initiation and severity. It has long been known that endothelial biochemical function is density dependent, and we have recently shown that endothelial immunobiology is anchorage dependent. Matrix-embedded endothelial cells (EC) establish a controlled anchorage state and are not only immune protected but also induce a system immune protective state. We now define this aspect of vascular and immune biology in detail. The in vitro immune response of allogeneic splenocytes (proliferation, lytic activity, and cytokine expression) on exposure to aortic EC was significantly reduced if EC were embedded within three-dimensional collagen matrices (3D-EC; P<0.005) to an even greater extent than EC that had reached confluence as monolayers on tissue culture plates (EC-TCPS). Splenocyte reactivity was enhanced with repeated exposure to EC-TCPS but minimally if preexposed to 3D-EC (P<0.002). 3D-EC induced significantly greater differentiation of splenocytes into CD4+CD25+Foxp3+ regulatory T cells than EC-TCPS (P<0.02). The reduced response to 3D-EC and potential protective effect to subsequent exposure were confirmed in vivo. Repeated exposure of immune-competent mice to injections of xenogeneic EC-TCPS induced vigorous host immunity. In contrast, prior implantation of 3D-EC induced hyporesponsiveness toward subsequent injection of EC-TCPS with reduced humoral response, decreased lytic activity, and lower frequency of effector splenocytes (P<0.001). EC interaction with its matrix determines phenotype, viability, and biosecretory potential. We now show that this microenvironmental interaction also influences endothelial-mediated activation of allo- and xenogeneic immune cells. 3D matrix-embedding limits the ability of EC to initiate adaptive immunity, and initial exposure to 3D-EC confers hyporesponsiveness to subsequent exposure to immunogeneic EC. These effects transcended the traditional control that confluence imposes on EC and reflects perhaps even higher order control. Our findings might offer novel insights to endothelial-mediated diseases and potential cell-based therapies.—Methe, H., Groothuis, A., Sayegh, M. H., Edelman, E. R. Matrix adherence of endothelial cells attenuates immune reactivity: induction of hyporesponsiveness in allo- and xenogeneic models.


Key Words: three-dimensional collagen matrices • T cells • tolerance • suppression • anergy • rodent


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
AS THE INTACT CELL POSSESSES the full force of physiological responsive regulatory mechanisms, it also bears the potential to restore health and halt disease. Cell-based therapies might therefore offer hope for a number of diseases, in particular those that single pharmacological agents cannot control. The challenges have been in directing cells to a site of injury, in a timely fashion, without loss of biochemical function or induction of an immune response. The technology of tissue engineering might provide such a vehicle and allow one to exam simultaneously the factors that drive cell-based control of repair. Tissue engineered endothelial cell constructs are fascinating in this regard. Endothelial cell biology is density dependent (1) . Confluent endothelial cells impart vascular quiescence, and subconfluent cells promote local growth. It has been assumed that tissue engineered constructs must provide the former and avoid the latter. Moreover, it is assumed that in general autologous cells are optimal for cell implantation, as it is expected that these cells will retain the greatest biochemical functionality and engender the least immune response on introduction into hosts. Yet, both the assumption that confluence alone drives efficacy and that autologous cells are optimal implants must be reconsidered in light of studies with tissue engineered endothelial cells.

Three-dimensional cell culture systems offer a milieu to study biosecretory, migratory, and proliferative functionality (2 3 4 5) . Embedding of endothelial cells in three-dimensional collagen-based matrices allows these cells to grow to confluence in a controlled environment. Such constructs allow endothelial cells to retain endothelial quiescence, secretion of essential regulatory factors, and the associated potential for vasoregulatory control, within vehicles that can be stored, manipulated, functionally validated, and implanted at will at sites protected from environmental forces (6 7 8 9) . However, as important as confluence may be it cannot alone explain the profound modulatory effect matrix-embedded endothelial cells have on vascular repair. Matrix-embedded endothelial cells are far more potent regulatory systems than monolayers of endothelial cells that are absolutely confluent. In particular, matrix embedding confers biochemical and immune regulation of cell implants (10) .

The importance of matrix architecture and cell surface interactions for cellular immune behavior is less well characterized, and it appears that such interactions may force reconsideration of notions of optimal cell sources for cell therapy. Cell implantation engenders a host response and potentially immunizes hosts for future cell implants, and it is accepted that autologous cells minimize this effect. In many disease states, however, the very cell that might best be used for implantation is most affected by illnesses. Autologous cells in diseased patients are often dysfunctional and may have already established an immune reaction against the diseased cells/tissue (11 , 12) . This is especially the case in vascular disease where endothelial cell dysfunction is a forme fruste of disease, and circulating antiendothelial cell antibodies can abound. As such endothelial cells from patients with established vascular disease are neither optimal bioreactors nor least immunogenic and may offer only limited potential for successful isolation and reimplantation for restoration of physiological control. Allogeneic or even xenogeneic cells are readily available for implantation, and though they are foreign cells they offer advantages well beyond autologous cells. These foreign cells can be harvested in advance of possible implantation, grown to desired density and even genetically manipulated and reinforced, but most importantly their health and biosecretory functionality can be validated before implantation. Thus, if these cells could be implanted with minimal immune response, the potential viability of cell-based therapies might well be extended.

Endothelial cells embedded in a physiological three-dimensional environment display reduced expression of positive costimulatory and adhesion molecules in vitro and minimal cell infiltration of allo- and xenogeneic implant sites in vivo (10) . We therefore investigated if matrix-imposed modifications on endothelial cell immune behavior would translate into alterations of endothelial-mediated activation of immune cells in vitro and in vivo. As positive costimulatory signals are pivotal for efficient T cell priming, we also explored secondary immune responses to nonembedded endothelial cells after initial exposure to endothelial cells embedded within a three-dimensional matrix.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Isolation and matrix-embedding of endothelial cells
Porcine aortic endothelial (PAE) cells were isolated from large white adult swine aortae by collagenase treatment (10) and grown in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 2 mM L-glutamine, 10% FBS (HyClone, UT, USA), 100 U/ml penicillin G, and 100 µg/ml streptomycin (Life Technologies, Grand Island, NY, USA). PAE were grown to confluence on tissue culture polystyrene plates (TCPS) or embedded within Gelfoam blocks (Pharmacia & Upjohn, Kalamazoo, MI, USA; ref. 10 ). Initial seeding density and the dimensions of the Gelfoam blocks were selected to achieve the final cell numbers needed for subsequent coculture experiments. EC surface attachment and confluence on Gelfoam matrices were demonstrated by confocal microscopy (data not shown). EC density on the TCPS and within the Gelfoam was determined with a Neubauer’s counting chamber. Experiments were performed on at least two Gelfoam blocks from each fabrication run. The blocks were washed with HBSS (Life Technologies) and digested with collagenase (1 mg/ml, type I, Worthington Biochemical Corp., Lakewood, NJ, USA). Detachment of EC from TCPS was achieved by brief trypsin treatment (0.5% trypsin in normal saline and 5.3 mM EDTA), which was interrupted with the addition of optimized EGM. Neither trypsin nor collagenase affects endothelial expression of immunoregulatory molecules (10) . Unaltered biosecretory function of surface-adherent endothelial cells has previously been demonstrated (6 , 10) . Cell viability was determined by trypan blue exclusion and a LIVE/DEAD viability/cytotoxicity kit (Molecular Probes, Eugene, OR, USA).

Splenocyte isolation
Spleens of large white adult swine (allogeneic to pigs PAE were isolated from) or B6 mice were harvested and cut in several pieces under sterile conditions. Clumps were immersed in solution and further dispersed by drawing and expelling the suspension several times through a sterile syringe with a 19-G needle. The suspension was filtered through a 200 µm mesh nylon screen to remove debris. Erythrocytes were lysed by treatment with ACK buffer (Cambrex, Walkersville, MD, USA) for 5 min at room temperature. Remaining cells were washed twice with RPMI (containing 2 mM L-glutamine, 0.1 M HEPES, 200 U/ml penicillin G, 200 µg/ml streptomycin, and 5% heat-inactivated calf serum; Life Technologies) and immediately used.

Enzyme-linked immunosorbent spot (ELISPOT) assay
Immunospot plates (Millipore, MA, USA) were coated with 5 µg/ml of anti-mouse (BD Pharmingen, San Jose, CA, USA) or antiporcine (Biosource, Camarillo, CA, USA) IFN-{gamma}, interleukin (IL)-2, IL-4, or IL-10 monoclonal antibodies in sterile PBS overnight. The plates were then blocked for 2 h with complete RPMI-medium without phenol red, containing 10% heat-inactivated calf serum; 5 x 105 endothelial cells isolated after grown to confluence on TCPS or within matrices were placed in immunospot wells with 5 x 105 splenocytes in a total volume of 400 µl RPMI-medium, completely covering the matrices. Plates were cultured for 48 h at 37°C/5% CO2 under gentle rocking conditions (10 cycles/min). After being washed with deionized water followed by PBS containing 0.05% Tween (PBST, Sigma, St. Louis, MO, USA), 2 µg/ml of biotinylated anti-mouse (BD Pharmingen) or antiporcine (Biosource) IFN-{gamma}, IL-2, IL-4, or IL-10 monoclonal antibody (mAb) were added and incubated overnight. The plates were then washed three times in PBST, followed by 1 h of incubation with horseradish peroxidase-conjugated streptavidin (BD Pharmingen). After being washed four times with PBST followed by PBS, the plates were developed using 3-amino-9-ethyl-carbazole (BD Pharmingen). The resulting spots were counted on a computer-assisted enzyme-linked immunospot image analyzer (Cellular Technology, Cleveland, OH, USA). To account for background in data analysis, the number of spots in negative control wells (medium, splenocytes, or endothelial cells alone) were subtracted from those in responsive wells.

IFN-{gamma} and IL-4 secretion by splenocytes
PAE matrix-embedded or TCPS-cultured PAE were seeded in 96-well plates at 5 x 105 cells/well, and 5 x 105 porcine splenocytes were seeded per well in complete RPMI-medium without phenol red, containing 10% heat-inactivated calf serum. Supernatants after 2 days of cocultures were separated by centrifugation and porcine IFN-{gamma} (detection limit<2.0 pg/ml), and IL-4 (detection limit<3.0 pg/ml) concentrations were quantified by ELISA (Biosource). Supernatants were stored at –80°C and measured at the same time by the same ELISA to avoid variations of assay conditions.

Lymphocyte proliferation assay
PAE matrix-embedded or TCPS-cultured PAE were seeded in 96-well plates at 5 x 104 cells/well and stimulated with 40 ng/ml porcine IFN-{gamma} (R&D Systems, Minneapolis, MN, USA) for 48 h, followed by mitomycin C treatment (50 µg/ml for 30 min; Sigma) to prevent background proliferation. After being washed, porcine splenocytes were added at 2 x 105 cells/well in complete RPMI-medium without phenol red, containing 10% heat-inactivated calf serum. For proliferation assays, 3[H]thymidine incorporation was measured on day 6 by 16h pulse (1 µCi/ml, Amersham, Piscataway, NJ, USA). Thymidine uptake of mitomycin-treated endothelial cells, medium, or splenocytes alone was used as negative controls. In some experiments, a murine antibody directed against human leukocyte antigen (HLA)-DP,DQ,DR was used to block activation via endothelial MHC class II molecules (10 ng/ml, DakoCytomation, Carpentina, CA, USA).

Calcein-acetoxymethyl release assay
Matrix-embedded or TCPS-cultured PAE (2x104) were incubated with 15 µM calcein-acetoxymethyl (calcein-AM, Molecular Probes) for 40 min at 37°C with occasional agitation. After two washes with complete medium, porcine splenocytes were added for 3 h at 37°C/5% CO2 at effector:target cell ratios of 50:1 to 1:1. Calcein-AM release was measured using a Fluoroskan Ascent FL dual-scanning microplate luminofluorimeter (Thermo Electron Corporation, Waltham, MA, USA). Specific lysis was calculated according to the formula [(test release–spontaneous release)/(maximum release–spontaneous release)] x 100. Spontaneous release represents calcein-AM release from target cells in medium alone, and maximum release is the calcein-AM release from target cells lysed in medium plus 2% Triton X-100, each measured in at least six replicate wells. Data presented within this manuscript are effector:target ratios of 25:1 as calcein-AM release reached a plateau at this ratio.

Secondary culture and rescue
For secondary cultures, 1 x 106 naive porcine splenocytes were cocultured with 1 x 106 PAE embedded in Gelfoam or TCPS-cultured in complete RPMI-medium without phenol red, containing 10% heat-inactivated calf serum in vitro. On day 7, porcine splenocytes were harvested, purified by isolation on a Ficoll-Paque gradient, washed, and reconstituted in fresh RPMI-medium. After 3 days of rest at 37°C/5% CO2, splenocytes from the two groups were reexposed to PAE in solution in complete RPMI-medium isolated after grown to confluence on TCPS to assess lytic activity (5x105 splenocytes cocultured with 2x104 PAE for another 3 h), frequency of alloreactive splenocytes, secretion of cytokines (5x105 splenocytes cocultured with 5x105 PAE for another 48 h), and splenocyte proliferation (2x105 splenocytes cocultured with 5x104 PAE for another 6 days). Recombinant porcine IL-2 at 10 IU/ml (Biosource) was added to certain wells for rescue experiments.

Foxp3 mRNA expression in murine CD4+CD25+ T lymphocytes
Murine CD4+-T lymphocytes were purified from isolated splenocytes with a negative CD4 isolation kit (Miltenyi Biotec, Auburn, CA, USA) following the manufacturer’s instructions. 5 x 105 CD4+-T lymphocytes were cultured with 1 x 105 PAE prestimulated with 40 ng/ml porcine IFN-{gamma} for 48 h (13) . After 5 days in culture, CD4+CD25+ T cells were isolated with a CD4+CD25+ regulatory T Cell Isolation Kit (Miltenyi Biotec). The purity was consistently >95% as revealed by flow cytometry (data not shown). Total RNA was extracted from isolated CD4+CD25+ T cells using the RNeasy Mini Kit (Qiagen, Valencia, CA, USA) according to the manufacturer’s instructions. As controls, RNA was extracted from CD4+CD25+ T cells without coculture with PAE. Complementary DNA was synthesized using TaqMan reverse transcription reagents from Applied Biosystems (Foster City, CA, USA). Real-time polymerase chain reaction (PCR) analysis was performed with an Opticon Real Time PCR Machine (MJ Research, Waltham, MA, USA) with SYBR Green PCR Master Mix (Applied Biosystems) using previously published primer sequences (13) . Data from the reaction were collected and analyzed by the complementary Opticon computer software. Relative quantification of gene expression was calculated with standard curves and normalized to GAPDH and are presented as relative units (RU).

In vivo experiments
All animal procedures were reviewed and approved by the local ethics committee on animal care and were conducted in accordance to the principles expressed in the Helsinki Declaration. Fifty-four B6-mice received 5 x 105 PAE implants in the subcutaneously dorsal space as matrix-embedded cells or saline-suspended cell pellets after grown to confluence on TCPS. After dorsal incision, a small subcutaneous cavity was created with blunt dissection and matrices carefully inserted into this space. To evaluate the impact of matrix embedding on formation of immunological memory, the same groups of mice were rechallenged with injections of free PAE on day 100. Sera were collected serially from 0 to 90 days after each implantation-procedure, aliquoted, and stored at –70°C. Six mice of each group were sacrificed on day 28 and day 128, respectively, for splenocyte isolation.

PAE-specific immunoglobulins
The levels of immunoglobulins specific for the implanted PAE in serum of the experimental mice were measured by flow cytometry. PAE (2 x 105), from the same strain as the implanted cells, were detached from cell culture plates with 0.25% trypsin/0.04% EDTA, pelleted, washed, and resuspended in FACS buffer (PBS, 1% FBS, and 0.1% sodium azide, Sigma Chemicals). These cells were then incubated with serum from recipient mice for 60 min at 4°C (diluted 1:10 in FACS buffer). After being washed twice with FACS buffer, cells were incubated with FITC-conjugated rat anti-mouse IgG2a (clone H106.771; Southern Biotechnology, Birmingham, AL, USA), IgG1 (clone A85–1), or IgM (clone R6–60.2, BD Pharmingen), respectively. After 30 min incubation at 4°C, samples were again washed twice with cold FACS buffer, fixed in 0.25 ml 1% paraformaldehyde, and 104 cells were analyzed by flow cytometry using a FACScalibur instrument and CellQuest software (Becton-Dickinson, Franklin Lakes, NJ, USA). Control samples included sera from naive mice and antibody incubation of PAE without serum to account for non-specific binding of the secondary antibodies (background). Data are presented as mean fluorescence intensity (MFI) per PAE with background subtraction for all cells analyzed.

Characterization of host effector cells
Splenocytes recovered from endothelial cell implant recipients were resuspended in FACS buffer at a concentration of 2 x 106/ml. Cells were stained with anti-CD4 FITC (clone L3T4), anti-CD8 FITC (clone Ly-2), anti-CD44 R-PE (clone Ly-24), and anti-CD62L allophycocyanin (clone Ly-22), and isotype controls (all BD PharMingen). CD4+ and CD8+ effector cells expressing CD44high and CD62Llow were enumerated, as described previously (14 , 15) .

Statistical analysis
All statistical analyses were performed with JMP software (SAS Institute, USA 2002). Data were normally distributed and are expressed as mean ± SD. Comparisons between two groups were analyzed by Student’s t test, and comparisons between more than two groups were analyzed by ANOVA. A Spearman correlation determined relations between spleen size and splenocyte density. A value of P < 0.05 was considered statistically significant.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Induction of Th1-polarized allogeneic splenocytes is reduced when PAE are matrix-embedded
The significant increase that TCPS-cultured PAE induced in T-helper cell (Th)1 (IL-2, IFN-{gamma} by ELISPOT) cytokine expression in allogeneic porcine splenocytes was muted when splenocytes were cocultured with matrix-embedded endothelial cells as described in the method section for ELISPOT analysis (P<0.01; Fig. 1 A). There was no significant difference in Th2-cytokine production between the two groups (Fig. 1B ). Reexposure of splenocytes to TCPS-cultured endothelial cells after purification from first coculture with TCPS-cultured endothelial cells (secondary culture) increased the Th1 response further and even elicited a substantial Th2 response, but there remained no elevation in either cytokine family with exposure to the same endothelial cells when first exposure was to matrix-embedded PAE (P<0.002; Fig. 1A, B ). IL-2 and IFN-{gamma} expression significantly exceeded IL-4 and IL-10 expression in all samples studied. IFN-{gamma} secretion and the resultant effects on splenocyte proliferation followed suit. As analyzed by ELISA, IFN-{gamma}-concentration in culture supernatants of porcine splenocytes exposed to free PAE was 7.4-fold greater than in splenocytes exposed to matrix-embedded PAE (187.9±39.3 vs. 25.4±22.3 pg/ml; P<0.005; Fig. 1C ). Reexposure to TCPS-cultured PAE increased IFN-{gamma} secretion even further by 57% to 295.2 ± 31.8 pg/ml, while second exposure to TCPS-cultured endothelial cells after initial exposure to matrix-embedded endothelial cells actually decreased secretion of this critical cytokine (5.2±0.4 pg/ml; P<0.002 vs. secondary exposure to free PAE; P<0.05 vs. matrix-embedded PAE; Fig. 1C ). There were no significant differences in IL-4 secretion between the two groups (Fig. 1C ).


Figure 1
View larger version (12K):
[in this window]
[in a new window]

  		Figure 1. Induction of Th1-cytokine-expressing and secreting splenocytes is reduced when endothelial cells are matrix embedded; 5 x 105 splenocytes were cocultured with 5 x 105 PAE isolated after grown to confluence on TCPS plates (free PAE) or within Gelfoam matrices for 48 h and analyzed by ELISPOT (A, B) and ELISA (C) for production and secretion of Th1- and Th2 cytokines. Reexposure was performed with splenocytes purified from coculture experiments with free and matrix-embedded PAE for 7 days and subsequent coculture with 5 x 105 free PAE for an additional 48 h. Data are mean percentage ± SD. *P < 0.01 free PAE vs. matrix-embedded PAE; #P < 0.002 primary exposure to free PAE vs. matrix-embedded PAE; {dagger}P < 0.005 free PAE vs. matrix-embedded PAE; {ddagger}P < 0.05 vs. primary exposure to matrix-embedded PAE.

Analysis of 3[H]thymidine incorporation and calcein-AM release determined that only TCPS-cultured but not matrix-embedded endothelial cells induced a significant splenocyte proliferation and lytic activity. Splenocytes exposed to matrix-embedded endothelial cells grew at 40-fold lower rate than splenocytes exposed to TCPS-cultured PAE (441±56 vs. 17215±662 cpm; P<0.005; Fig. 2 A) and demonstrated reduced lytic ability with lower calcein-AM release from coresident endothelial cells (Fig. 2B ; P<0.05). Reexposure of splenocytes to TCPS-cultured PAE after purification from first coculture with TCPS-cultured endothelial cells (secondary culture) amplified this difference in proliferation induction similar to stimulation of cytokine production (53.5% increase to 26421±983 cpm; P<0.01; Fig. 2A ) and with an even greater increase in lytic ability (Fig. 2B ). Once again when the initial exposure was between splenocytes and matrix-embedded PAE secondary exposure to PAE induced no greater splenocyte proliferation and lytic ability than that observed with the primary exposure (Fig. 2) . The addition of exogenous porcine IL-2 to mixed endothelial cell-splenocyte cultures after purification of splenocytes from first coculture with TCPS-cultured or matrix-embedded PAE (secondary culture) did not induce a proliferative response nor increase in lytic ability in the latter cells when first exposure was to matrix-embedded endothelial cells (P<0.005; Fig. 2 ).


Figure 2
View larger version (17K):
[in this window]
[in a new window]

  		Figure 2. Splenocyte proliferation (A) and lytic activity (B) are reduced when endothelial cells are matrix embedded. A) 2 x 105 porcine splenocytes were incubated 6 days with IFN-{gamma} stimulated PAE (5x104) matrix-embedded or cultured in TCPS (free). Splenocyte proliferation at the termination of the experiment was determined through incorporation of 3[H]thymidine administered to the cells 16 h before harvest. Thymidine uptake of mitomycin-treated endothelial cells, medium, or splenocytes alone served as negative controls. In some experiments, a murine antibody directed against HLA-DP,DQ,DR was used to identify the role of MHC II signaling for subsequent splenocyte proliferation. B) Endothelial cell lysis was monitored through calcein-AM release; 2 x 104 PAE matrix-embedded or TCPS-cultured were incubated with 15 µM calcein-AM for 40 min at 37°C. After being washed, porcine splenocytes were added to target cells for 3 h spanning a 50:1 to 1:1 effector:target cell ratio. Calcein-AM release from PAE was analyzed luminofluorimetrically and is presented as specific lysis. Data are splenocyte:PAE ratios of 25:1. Subsequent proliferation of splenocytes to free PAE and calcein-AM release by free PAE was analyzed after purification of splenocytes from primary culture with free or matrix-embedded PAE and reculture with free PAE. Rescue of splenocyte proliferation and lytic activity was tested by addition of exogenous IL-2. Data are mean percentage ± SD. *P < 0.005 free PAE vs. matrix-embedded PAE; #P < 0.002 primary exposure to free PAE vs. matrix-embedded PAE; {dagger}P < 0.05 free PAE vs. matrix-embedded PAE; {ddagger}P < 0.005 primary exposure to free PAE vs. matrix-embedded PAE primary exposure vs. re-exposure to free PAE.

In contrast, completely blocking MHC II signaling on matrix-embedded PAE by adding anti-MHC II antibodies restored the full proliferation capacity of splenocytes on secondary culture of splenocytes with TCPS-cultured PAE (18,354±1107 vs. 17,215±662 cpm at primary contact of splenocytes to TCPS-cultured PAE; P=0.07; Fig. 2A ).

Matrix embedding increases endothelial induction of xenogeneic Foxp3-expressing CD4+CD25+ T lymphocytes in vitro
In vitro coculture of IFN-{gamma} activated free PAE with murine CD4+ T cells resulted in a significant expansion of Foxp3 mRNA expressing CD4+CD25+ subsets of T lymphocytes [reverse transcriptase (RT)-PCR, 0.0014±0.0005 vs. 0.00045±0.0002 RU in naive CD4+CD25+ T cells; P<0.02]. Expansion of Foxp3 mRNA expressing CD4+CD25+ T cells increases 2-fold when in place of TCPS grown PAE coculture was performed with cytokine-activated matrix-embedded PAE (0.0028±0.0007 RU; P<0.02; Fig. 3 ).


Figure 3
View larger version (8K):
[in this window]
[in a new window]

  		Figure 3. RT-PCR analysis determined Foxp3 mRNA expression in naive CD4+CD25+ T cells and in CD4+CD25+ T cells after coculture with IFN-{gamma} activated matrix-embedded and free PAE for 5 days. Foxp3 mRNA expression was normalized to GAPDH mRNA expression and is expressed as relative units ± SD from 6 different experiments. *P < 0.02 vs. naive CD4+CD25+ T cells and vs. coculture with matrix-embedded PAE.

Matrix embedding attenuates humoral and cellular immune reactions against xenogeneic endothelial cells in vivo
Matrix embedding of endothelial cells had similar effects in vivo as in culture. Humoral and cellular immune responses in B6 mice after subcutaneous implantation of free PAE, matrix-embedded PAE, and PAE pellets adjacent to empty matrices were analyzed via flow cytometry, ELISPOT and cytotoxicity assays. Circulating anti-PAE immunoglobulin (Ig)G1 and IgM levels were similar in mice implanted with PAE pellets alone or PAE adjacent to empty matrices, and both groups had significantly higher antibody titers than were seen in recipients of matrix-embedded PAE (Fig. 4 A, B). There was a transient and minor elevation in anti-PAE IgG2a 12 days after implantation of matrix-embedded PAE (P<0.05) but no formation of porcine-specific IgG2a-antibodies in the other two groups (Fig. 4C ). No significant PAE-specific antibody concentrations were detectable in sera from naive mice (Fig. 4) . The lytic ability of splenocytes from these mice tracked with immune response. Lytic ability was 69% lower in splenocytes isolated from mice implanted with matrix-embedded PAE compared to free PAE or PAE implanted adjacent to the three-dimensional matrix (P<0.001; Fig. 5 ).


Figure 4
View larger version (14K):
[in this window]
[in a new window]

  		Figure 4. Effect of matrix-embedding endothelial cells on the formation of circulating antiendothelial cell antibodies in B6 mice was assessed by flow cytometry, including PAE-specific IgG1 (A), IgM (B), and IgG2a (C). Mice were challenged with sc injections of free PAE, implantation of Gelfoam-embedded PAE, or concomitant injection of PAE adjacent to empty Gelfoam matrices (n=18 per group to day 28, n=12 per group day 56–100 postimplantation). On day 100, the remaining 12 mice in each group were rechallenged with sc injection of free PAE (n=12 per group to day 128, n=6 per group day 156–190 postimplantation). Control mice did not receive PAE (naive mice). To analyze PAE-specific antibody titers, PAE were cultured with serum from treated and naive mice for 60 min in vitro, washed, and incubated with fluorophor-labeled mice-specific anti-immunoglobuline antibodies. Data are expressed as mean fluorescence intensity± SD. *P < 0.00001, {dagger}P < 0.0001, {ddagger}P < 0.005, and §P < 0.05.


Figure 5
View larger version (9K):
[in this window]
[in a new window]

  		Figure 5. Mice received sc injection of free PAE, implantation of matrix-embedded PAE, or concomitant injection of PAE adjacent to empty Gelfoam matrices. Six mice of each group were sacrificed on day 28 for splenocyte isolation. Remaining mice from each treatment group were rechallenged with injections of free PAE on day 100 (second implantation) and 6 of these mice were sacrificed on day 128 for splenocyte isolation. To test the ability of splenocytes from challenged mice to lyse endothelial cells in vitro, 5 x 105 splenocytes were cocultured with 2 x 104 calcein-AM-labeled PAE for 3 h. Calcein release as the dependent variable of endothelial cell lysis was measured by luminofluorimeter. *P < 0.05 first vs. second implantation, {dagger}P < 0.001 vs. free PAE and PAE with Gelfoam first implantation; {ddagger}P < 0.0001 vs. free PAE and PAE with Gelfoam first implantation.

Significant numbers of Th2 cytokine-producing cells were isolated from splenocytes of mice receiving free PAE or PAE adjacent to empty matrices at levels significantly higher than in mice with matrix-embedded endothelial cells (IL-4: P<0.0001; IL-10: P<0.001; Fig. 6 A, B). There was no difference in the frequency of Th1 cytokine-producing cells isolated from splenocytes in the three groups 28 days postimplantation. IL-2- and IFN-{gamma}-ELISPOT reactivity were comparable in the three groups (IL-2, free PAE 297±127, matrix-embedded 253±101, PAE adjacent to matrix 256±87 spots, P=0.62; IFN-{gamma}, free PAE 181±114, matrix-embedded 229±123, PAE adjacent to matrix 274±139 spots, P=0.71). Although xenoreactive Th2-producing cells significantly outnumbered Th1-producing cells in mice receiving free PAE or PAE adjacent to empty matrices, there was no clear shift toward Th1- or Th2 cytokine-producing cells in mice receiving matrix-embedded PAE. Two sets of controls validated the specificity of the effect. Neither Th1 nor Th2 cell cohorts were elevated in splenocytes from naive mice never exposed to endothelial cell implants and subsequently challenged in culture with PAE. Similarly, PAE from pigs different than the source of implanted cells evoked only a modest increase in cytokine-producing splenocytes isolated from the three groups of mice (data not shown).


Figure 6
View larger version (48K):
[in this window]
[in a new window]

  		Figure 6. Frequency of xenoantigen-specific cytokine-producing cells in recipients after sc implantation of PAE in mice. Mice received sc injection of free PAE, implantation of matrix-embedded PAE, or concomitant injection of PAE adjacent to empty Gelfoam matrices. Six mice of each group were sacrificed on day 28 for splenocyte isolation. Remaining mice from each treatment group were rechallenged with sc injections of free PAE on day 100 (second implantation) and 6 of these mice were sacrificed on day 128 for splenocyte isolation. A) Representative ELISPOT wells for 1 mouse from each treatment group 28 days after first implantation and 28 days after second stimulation, respectively. IL-4 production in response to PAE was measured. B) Recipients of free PAE exhibited a significant increased frequency of xenoreactive IL-4 and IL-10 producing T cells compared to recipients of matrix-embedded PAE on day 28. Rechallenge with free PAE on day 100 increased frequency of xenoantigen-specific T cells but not in mice receiving matrix-embedded PAE. *P < 0.05 d 28 vs. day 128; {dagger}P < 0.0001 vs. free PAE and PAE with Gelfoam; {ddagger}P < 0.001 vs. free PAE and PAE with Gelfoam.

Matrix-embedded endothelial cells inhibit in vivo generation of effector T cells
CD4+ and CD8+ effector cells have been reliably identified as CD62LlowCD44high cells (14 , 15) . To determine the effect of matrix embedding on the generation and function of xenoreactive T-cells, we measured the number of effector T-cells in spleens of mice from the three treatment groups 28 days after implantation and in naive mice (Fig. 7 ). The percentage of CD62LlowCD44high cells increased significantly in free PAE-recipients and mice receiving PAE adjacent to empty matrices compared with mice receiving matrix-embedded PAE. The frequency of CD4+CD62LlowCD44high T cells outnumbered CD8+ effector cells in all groups (ratio 1.7–2.3; Fig. 7B ). In concert with these data, spleen size and splenocyte density were elevated in all animals except those that received matrix-embedded endothelial cells (112.7±16.9, 102.5±18.8, and 62.9±9.6 mm3, P<0.01; and 2.76±0.37, 2.47±0.4, 1and .61±0.33 106 splenocytes/ml; P<0.01; r=0.82, P<0.0001).


Figure 7
View larger version (27K):
[in this window]
[in a new window]

  		Figure 7. Evaluation of CD4+ and CD8+ effector cells in naive mice and mice receiving sc implants of PAE. Splenocytes isolated 28 days after first implantation and second implantation respectively were subjected to FACS. A) Representative FACS analysis of splenic lymphocytes gated on CD4+ cells from mice receiving free (a), matrix-embedded (b), or PAE adjacent to Gelfoam (c) 28 days after implantation, and from naive mice (d). The percentage of CD62LlowCD44high cells within the total CD4+ cell population is shown in the right lower corner. B) Column graphs represent frequency of splenic CD62LlowCD44highCD4+ and CD8+ lymphocytes in naive mice and 6 mice from each group 28 days after first and second implantation of PAE. *P < 0.05 d 28 vs. day 128; {dagger}P < 0.001 vs. free PAE and PAE with Gelfoam.

Immune memory is muted in mice by matrix-embedded endothelial cells
One hundred days after first implantation 12 mice in each group were rechallenged with pellets of saline-suspended PAE that had been grown to confluence on TCPS. Mice first receiving free PAE pellets or pellets adjacent to empty matrices showed a significant secondary humoral immune response, with PAE-specific IgG1-antibody levels exceeding the response observed after the first course of implantation (Fig. 4A ). Only a weak IgM-antibody release was seen (Fig. 4B ). In marked contrast, there was no significant rise in PAE-specific IgG1 or IgM antibodies in mice with matrix-embedded PAE on single or repeated exposure to saline-suspended PAE (Fig. 4A, B ). There was a mild and transient increase of PAE-specific IgG2a antibodies 12 days after reimplantation throughout all treatment groups (Fig. 4C ), yet the IgG2a antibody response was an order of magnitude weaker than the PAE-specific IgG1 antibody response throughout the study. Rechallenge to matrix-embedded PAE was without effect on host humoral and cellular immunity (data not shown).

The lytic ability of splenocytes from mice followed anti-PAE levels. Whereas the lytic ability of splenocytes from mice prechallenged to matrix-embedded PAE decreased over time, the ability of splenocytes from mice receiving a secondary implant of pellets of saline-suspended PAE after first implantation of free PAE or PAE injections adjacent to empty Gelfoam blocks was significant and significantly higher than after the first implantation (Fig. 5) . In addition, the frequency of xenoreactive IL-4 and IL-10 producing T cells increased significantly in mice after reimplantation of free PAE, while remaining unchanged in mice after initial implantation of matrix-embedded PAE (Fig. 6A, B ). Finally, the percentage of CD4+ effector cells seen 28 days after rechallenge further increased in mice receiving free PAE and increased significantly in mice receiving PAE adjacent to empty Gelfoam implants but remained unchanged in mice that had first receiving matrix embedded PAE. The same pattern was evident for CD8+ effector T cells (Fig. 7B ).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Matrix-embedding shields pancreatic islet cells from immune response (16 , 17) and reduces the immunogenicity of fibroblasts and smooth muscle cells (18 , 19) . We now show that this form of protection extends to endothelial cells but in a manner that involves far more than physical isolation of embedded cells. Matrix-embedding significantly reduced endothelial cell expression of costimulatory, adhesion, and MHC II molecules on cytokine stimulation in vitro, restoring cells to a state resembling quiescent, nonactivated cells (8 , 10) . Such manipulation of seeding environment also drives T cell differentiation along regulatory phenotype development pathways and by modulating the balanced induction of effector and regulatory T cells from splenocytes. Matrix embedding may therefore recapitulate a natural quiescent environment and establish significant control over endothelial T cell interactions, rather than simply serve as a convenient vehicle for endothelial cell transportation or isolation.

Differentiation of CD4+ T cells and induction of effector T cells
The reaction to allo- and xenografts begins with contact between graft endothelial cells and host immune cells. Cross activation of the graft endothelium induces adhesion, costimulatory, and MHC molecule expression (20) . Adhesion molecules enhance the capacity of endothelial cells to capture T cells and formation of contact regions between the two cell types. MHC and costimulatory molecule expression enables endothelial cells to provide antigen-dependent signals to naive CD8+ and CD4+ T cells through the direct pathway (21 22 23) . Matrix embedding reduces the expression of these molecules (10) , thereby limiting activation and differentiation of allo- and xenogeneic immune cells and promoting hyporesponsiveness to secondary contact with nonembedded endothelial cells in vitro and in vivo. Peripheral tolerance can in this way be attained through control of MHC II levels (24 25 26 27) . The significantly increased humoral and cellular immune response observed with repeated exposure to saline-suspended endothelial cell pellets was not evident if initial exposure was to matrix-embedded endothelial cells (Figs. 1 , 4 , and 6) and was related to reduction in endothelial MHC II expression (Fig. 2) . The protective effect of matrix-embedded endothelial cells was retained over time and may reflect an influence on the immune systems that is of far longer duration than the presence of the cells. The significant induction of effector T cells and lytic activity that is expected when splenocytes are exposed to nonembedded endothelial cells was significantly limited when first contact was to matrix-embedded endothelial cells (Figs. 2 , 5 , and 7) . This protective effect was observed for xeno- and allogeneic endothelial cell-splenocyte crosstalk.

The seemingly similar immunomodulatory influence of cross- and interspecies cell interactions does not necessarily imply identical mechanism, and indeed our data would suggest that they are governed by subtly different forms of Th1 and Th2 responses. It is accepted that Th1-mediated immunity controls acute and chronic allograft rejection (Fig. 1 ; refs. 28 , 29 ). The cellular immune response in xenotransplantation, however, is less well characterized, as hyperacute and acute xenorejection is largely attributed to humoral host immune responses. Our nonvascularized xenograft model allowed us to examine the in vivo xenoresponse in more detail. In line with previous studies (30 31 32 33) , our cytokine and antibody isotype analyses suggest that nonembedded xenogeneic endothelial cells evoke a Th2-polarized immune response (Fig. 6) . The complexity of the Th1/Th2 paradigm is amplified by the overlapping cross activation during memory responses to repeated exposure to nonembedded endothelial cells. Th2 cytokines do dominate the memory response in the in vivo xenograft setting when mice were reexposed to nonembedded PAE (Fig. 6) . However, the significant increase in Th1 response with in vitro allogeneic interactions is accompanied by a lesser, but nonetheless evident, increases in Th2 cytokines on reexposure of splenocytes to endothelial cells (Fig. 1) . It remains possible that these differences are simply from in vivo in vitro experimental discordance.

T regulatory cell differentiation
The controlled induction of Th-cell activation and of effector T cells reflects a cytokine-mediated process of T cell differentiation. There is another differentiation pathway that primarily involves TGF-ß influence and in particular over induction of T regulatory cells. Endothelial cells can activate and induce allogeneic Foxp3+CD4+CD25+ T cells that have been shown to inhibit proliferation of alloreactive T cells in vitro and in vivo (13) . Our data now demonstrate that matrix embedding significantly enhances endothelial-induced expansion of Foxp3+CD4+CD25+ regulatory T cells (Fig. 3) . Induction of T regulatory cells by endothelial cells is dependent on programmed death ligand 1 (PD-L1), which is expressed on endothelial cells (13) . We have previously demonstrated that matrix embedding decreases cytokine-induced endothelial expression of positive costimulatory molecules (e.g., B7–1 and B7–2) but maintains the up-regulation of negative costimulatory molecules (e.g., PD-L1, PD-L2) (10) . In line with these results, LaGier and Pober (34) recently demonstrated that overexpressing PD-L1 on endothelial cells limited their ability to induce proliferation of allogeneic memory cells.

The combined effect of these processes, limited differentiation of CD4+ T cells, muted induction of effector T cells, and increased induction of regulatory T cells, control T cell proliferation in the face of immune challenge (Fig. 2) but without permanent T cell toxicity. This is in marked contrast to the reduced T cell response in allografts when graft endothelial cells were transfected with Fas-ligand to induce apoptosis of alloreactive lymphocytes (35) .

Shielding or phenotypic modulation
Matrix embedding of some cells primarily serves to isolate them from immune identity or immune attack (16 , 17) . Our data suggest that matrix seeding is a far more complex phenomenon than the creation of a physical barrier to endothelial-T cell contact and may well involve recapitulation of physiological control. Matrix embedding significantly reduced the ability of splenocytes to proliferate on contact with endothelial cells (Fig. 2) , and secondary response to nonembedded endothelial cells was muted when splenocytes first had seen matrix-embedded endothelial cells. Activation of naive lymphocytes requires both presentation of antigen by MHC molecules and costimulatory signals. Matrix-embedding reduces but does not entirely eliminate MHC II responsiveness and costimulatory molecule expression and has no effect on MHC I or PD-L1 expression. For this reason, there is hyporesponsiveness but not anergy, and this alone should support the existence of endothelial T cell contact. The presence of a reduced and controlled active interaction, rather than passive isolation, of splenocytes and embedded endothelial cells is further substantiated by the manipulation of MHC II signaling. When antibodies entirely eliminate the MHC II response, matrix-embedded endothelial cells render splenocytes naive and fully responsiveness to subsequent exposure to nonembedded endothelial cells (Fig. 2) . The retention of PD-L1 signaling also indicates profound phenotypic modulation of endothelial cells inducing immunomodulation rather than simply shielding of endothelial cells from immune recognition by matrix embedding. Induction of regulatory T cells involves signaling via PD-L1 (13) , and there may well be a correlation between intact PD-L1 signaling in matrix-embedded endothelial cells (10) and their enhanced ability to induce Foxp3+CD4+CD25+ regulatory T cells (Fig. 3) .

Immunological aspects of cell-matrix interactions
Our results add to the understanding of cell:matrix contact as a modulatory force in cell biology. It has long been established that endothelial cell function is density-dependent (1) . Contiguous endothelial cells elicit a series of biochemical mediators that induce local quiescence; disrupted monolayers stimulate growth of self and adjacent cells. Yet, two-dimensional surface contiguity is only a part of the powerful regulatory aspect of endothelial function. Intact endothelial cells reside in a polarized state where there is not only an up and down but complex matrix interactions. While the apical aspect of endothelial cells is exposed to flow, the basal aspect of the endothelial cell is in intimate contact with the basement membrane. This anisotropy is well mimicked by embedding endothelial cells in a three-dimensional matrix (6 7 8 , 10 , 36) in a manner that transcends the reproduction of cell monolayers that tissue culturing in flat plates can provide. Such a finding might explain the importance of endothelial cell continuity for vascular health and vascular disease progression with endothelial disruption. Disintegration of the physiological matrix architecture often heralds disease states from diabetes to glomerulopathy to atherosclerosis (37 38 39) . These pathological states are also associated with immune cell attraction and activation via increased expression of adhesion, costimulatory, and MHC molecules on endothelial cells. Anti-endothelial cell antibodies can be detected after endothelial disruption, and changes in matrix architecture by mechanical, metabolic, and immune stimuli can elicit an immune cascade on the one hand and immune memory on the other with profound potential of immune-mediated cell destruction as seen in acute and chronic graft rejection (21 , 23 , 40) .

The idea of endothelial cell seating integrity as determinant of immune activation may help link aberrant matrix:cell interactions with diseases such as atherosclerosis and vasculitis (41 , 42) , where adhesion and activation of circulating immune cells seem pivotal for disease initiation and progression. We are cognizant just the same of the limitations of our study. Critical questions remain as to whether endothelial cell seating integrity alone or only in concert with the chosen matrix material (collagen) and/or the three-dimensional state contributes to reduced endothelial immunogenicity when embedded within three-dimensional collagen-based matrices. Further detailed investigations on the specific molecular mechanisms are currently underway. Furthermore, in aiming to imitate in vivo extracellular matrix:endothelial cell interactions, our three-dimensional model lacks other important physiological factors such as shear stress and exposure to the bloodstream.

Moreover, matrix embedding might provide a survival benefit for engrafted endothelial cells. When freed from their underlying matrix, necrotic cells might induce immune recognition and memory response. Although we could not demonstrate significant differences in cell death between TCPS-cultured and matrix-embedded PAE in vitro, a survival benefit of matrix-embedded cells might partially explain the reduced immune response on initial exposure and subsequent hyporesponsiveness toward secondary implantation of nonembedded endothelial cells in our in vivo model. Our findings might continue the evolution of thought in vascular disease and guide design of tissue-engineered devices for allogeneic or even xenogeneic cells.


   ACKNOWLEDGMENTS
 
This work was supported in part by a Philip Morris External Research Program Postdoctoral Fellowship to H. Methe and a grant from the National Institutes of Health to E. R. Edelman (HL-49039). The authors have no conflicting financial interests.

Received for publication August 9, 2006. Accepted for publication December 14, 2006.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Nugent, M. A., Karnovsky, M. J., Edelman, E. R. (1993) Vascular cell-derived heparan sulfate shows coupled inhibition of basic fibroblast growth factor binding and mitogenesis in vascular smooth muscle cells. Circ. Res. 73,1051-1060[Abstract/Free Full Text]
  2. Lukashev, M. E., Werb, Z. (1998) ECM signalling: orchestrating cell behaviour and misbehaviour. Trends. Cell Biol. 8,437-441[CrossRef][Medline]
  3. Weaver, V. M., Petersen, O. W., Wang, F., Larabell, C. A., Briand, P., Damsky, C., Bissell, M. J. (1997) Reversion of the malignant phenotype of human breast cells in three-dimensional culture and in vivo by integrin blocking antibodies. J. Cell Biol. 137,231-245[Abstract/Free Full Text]
  4. Sasaki, T., Takagi, M., Soma, T., Yoshida, T. (2002) 3D culture of murine hematopoietic cells with spatial development of stromal cells in nonwoven fabrics. Cytotherapy 4,285-291[CrossRef][Medline]
  5. Edelman, D. B., Keefer, E. W. (2005) A cultural renaissance: in vitro cell biology embraces three-dimensional context. Exp. Neurol. 192,1-6[CrossRef][Medline]
  6. Nathan, A., Nugent, M. A., Edelman, E. R. (1995) Tissue engineered perivascular endothelial cell implants regulate vascular injury. Proc. Natl. Acad. Sci. U. S. A. 92,8130-8134[Abstract/Free Full Text]
  7. Nugent, H. M., Rogers, C., Edelman, E. R. (1999) Endothelial implants inhibit intimal hyperplasia after porcine angioplasty. Circ. Res. 84,384-391[Abstract/Free Full Text]
  8. Nugent, H. M., Edelman, E. R. (2001) Endothelial implants provide long-term control of vascular repair in a porcine model of arterial injury. J. Surg. Res. 99,228-234[CrossRef][Medline]
  9. Nugent, H. M., Edelman, E. R. (2003) Tissue engineering therapy for cardiovascular disease. Circ. Res. 92,1068-1078[Abstract/Free Full Text]
  10. Methe, H., Nugent, H. M., Groothuis, A., Seifert, P., Sayegh, M. H., Edelman, E. R. (2005) Matrix embedding alters the immune response against endothelial cells in vitro and in vivo. Circulation 112,I89-I95[Medline]
  11. Papagianni, A., Kalovoulos, M., Kirmizis, D., Vainas, A., Belechri, A. M., Alexopoulos, E., Memmos, D. (2003) Carotid atherosclerosis is associated with inflammation and endothelial cell adhesion molecules in chronic haemodialysis patients. Nephrol. Dial. Transplant. 18,113-119[Abstract/Free Full Text]
  12. Turesson, C. (2004) Endothelial expression of MHC class II molecules in autoimmune disease. Curr. Pharm. Des. 10,129-143[CrossRef][Medline]
  13. Krupnick, A. S., Gelman, A. E., Barchet, W., Richardson, S., Kreisel, F. H., Turka, L. A., Colonna, M., Patterson, G. A., Kreisel, D. (2005) Murine vascular endothelium activates and induces the generation of allogeneic CD4+25+Foxp3+ regulatory T cells. J. Immunol. 175,6265-6270[Abstract/Free Full Text]
  14. Zhai, Y., Shen, X. D., Gao, F., Coito, A. J., Wasowska, B. A., Salama, A., Schmitt, I., Busuttil, R. W., Sayegh, M. H., Kupiec-Weglinski, J. W. (2002) The CD154-CD40 T cell costimulation pathway is required for host sensitization of CD8(+) T cells by skin grafts via direct antigen presentation. J. Immunol. 169,1270-1276[Abstract/Free Full Text]
  15. Harada, H., Salama, A. D., Sho, M., Izawa, A., Sandner, S. E., Ito, T., Akiba, H., Yagita, H., Sharpe, A. H., Freeman, G. J., et al (2003) The role of the ICOS-B7h T cell costimulatory pathway in transplantation immunity. J. Clin. Invest. 112,234-243[CrossRef][Medline]
  16. Lee, D. Y., Nam, J. H., Byun, Y. (2004) Effect of polyethylene glycol grafted onto islet capsules on prevention of splenocyte and cytokine attacks. J. Biomater. Sci. Polym. Ed. 15,753-766[CrossRef][Medline]
  17. Siebers, U., Horcher, A., Bretzel, R. G., Federlin, K., Zekorn, T. (1997) Alginate-based microcapsules for immunoprotected islet transplantation. Ann. N. Y. Acad. Sci. 831,304-312[Abstract/Free Full Text]
  18. Kern, A., Liu, K., Mansbridge, J. (2001) Modification of fibroblast gamma-interferon responses by extracellular matrix. J. Invest. Dermatol. 117,112-118[CrossRef][Medline]
  19. Stegemann, J. P., Nerem, R. M. (2003) Altered response of vascular smooth muscle cells to exogenous biochemical stimulation in two- and three-dimensional culture. Exp. Cell Res. 283,146-155[CrossRef][Medline]
  20. Valujskikh, A., Heeger, P. S. (2003) Emerging roles of endothelial cells in transplant rejection. Curr. Opin. Immunol. 15,493-498[CrossRef][Medline]
  21. Choi, J., Enis, D. R., Koh, K. P., Shiao, S. L., Pober, J. S. (2004) T lymphocyte-endothelial cell interactions. Annu. Rev. Immunol. 22,683-709[CrossRef][Medline]
  22. Gould, D. S., Auchincloss, H., Jr (1999) Direct and indirect recognition: the role of MHC antigens in graft rejection. Immunol. Today 20,77-82[CrossRef][Medline]
  23. Denton, M. D., Davis, S. F., Baum, M. A., Melter, M., Reinders, M. E., Exeni, A., Samsonov, D. V., Fang, J., Ganz, P., Briscoe, D. M. (2000) The role of the graft endothelium in transplant rejection: evidence that endothelial activation may serve as a clinical marker for the development of chronic rejection. Pediatr. Transplant. 4,252-260[CrossRef][Medline]
  24. Rosemblatt, M., Bono, M. R. (2004) Functional consequences of immune cell adhesion to endothelial cells. Curr. Pharm. Des. 10,109-120[CrossRef][Medline]
  25. Perez, V. L., Henault, L., Lichtman, A. H. (1998) Endothelial antigen presentation: stimulation of previously activated but not naive TCR-transgenic mouse T cells. Cell Immunol. 189,31-40[CrossRef][Medline]
  26. Marelli-Berg, F. M., Scott, D., Bartok, I., Peek, E., Dyson, J., Lechler, R. I. (2000) Activated murine endothelial cells have reduced immunogenicity for CD8+ T cells: a mechanism of immunoregulation?. J. Immunol. 165,4182-4189[Abstract/Free Full Text]
  27. Limmer, A., Ohl, J., Kurts, C., Ljunggren, H. G., Reiss, Y., Groettrup, M., Momburg, F., Arnold, B., Knolle, P. A. (2000) Efficient presentation of exogenous antigen by liver endothelial cells to CD8+ T cells results in antigen-specific T-cell tolerance. Nat. Med. 6,1348-1354[CrossRef][Medline]
  28. Koglin, J., Glysing-Jensen, T., Gadiraju, S., Russell, M. E. (2000) Attenuated cardiac allograft vasculopathy in mice with targeted deletion of the transcription factor STAT4. Circulation 101,1034-1039[Abstract/Free Full Text]
  29. Methe, H., Wiegand, D., Welsch, U., Krombach, F., Meiser, B., Nabauer, M., Koglin, J. (2005) Peripheral expansion of circulating T-helper 1 cells predicts coronary endothelial dysfunction after cardiac transplantation. J. Heart Lung Transplant. 24,833-840[CrossRef][Medline]
  30. Wren, S. M., Wang, S. C., Thai, N. L., Conrad, B., Hoffman, R. A., Fung, J. J., Simmons, R. L., Ildstad, S. T. (1993) Evidence for early Th 2 T cell predominance in xenoreactivity. Transplantation 56,905-911[Medline]
  31. Simeonovic, C. J., Townsend, M. J., Karupiah, G., Wilson, J. D., Zarb, J. C., Mann, D. A., Young, I. G. (1999) Analysis of the Th1/Th2 paradigm in transplantation: interferon-gamma deficiency converts Th1-type proislet allograft rejection to a Th2-type xenograft-like response. Cell Transplant. 8,365-373[Medline]
  32. Singh, N. P., Guo, L., Mhoyan, A., Shirwan, H. (2003) Predominant expression of Th2 cytokines and interferon-gamma in xenogeneic cardiac grafts undergoing acute vascular rejection. Transplantation 75,586-590[CrossRef][Medline]
  33. Wang, H., DeVries, M. E., Deng, S., Khandaker, M. H., Pickering, J. G., Chow, L. H., Garcia, B., Kelvin, D. J., Zhong, R. (2000) The axis of interleukin 12 and gamma interferon regulates acute vascular xenogeneic rejection. Nat. Med. 6,549-555[CrossRef][Medline]
  34. LaGier, A. J., Pober, J. S. (2006) Immune accessory functions of human endothelial cells are modulated by overexpression of B7–H1 (PDL1). Hum. Immunol. 67,568-578[CrossRef][Medline]
  35. Cappellesso, S., Thibault, G., Hoarau, C., Herault, O., Iochmann, S., Bardos, P., Lebranchu, Y. (2002) FasL-transfected endothelial cells decrease the proliferative response of allogeneic PBL. Transpl. Immunol. 10,293-302[CrossRef][Medline]
  36. Centra, M., Ratych, R. E., Cao, G. L., Li, J., Williams, E., Taylor, R. M., Rosen, G. M. (1992) Culture of bovine pulmonary artery endothelial cells on Gelfoam blocks. FASEB J. 6,3117-3121[Abstract]
  37. Raines, E. W. (2000) The extracellular matrix can regulate vascular cell migration, proliferation, and survival: relationships to vascular disease. Int. J. Exp. Pathol. 81,173-182[CrossRef][Medline]
  38. Schleicher, E., Kolm, V., Ceol, M., Nerlich, A. (1996) Structural and functional changes in diabetic glomerulopathy. Kidney Blood Press Res. 19,305-315[Medline]
  39. McMillan, D. E. (1997) Development of vascular complications in diabetes. Vasc. Med. 2,132-142[Medline]
  40. Robson, S. C., Candinas, D., Hancock, W. W., Wrighton, C., Winkler, H., Bach, F. H. (1995) Role of endothelial cells in transplantation. Int. Arch. Allergy Immunol. 106,305-322[Medline]
  41. Hansson, G. K. (2005) Inflammation, atherosclerosis, and coronary artery disease. N. Engl. J. Med. 352,1685-1695[Free Full Text]
  42. Cid, M. C., Segarra, M., Garcia-Martinez, A., Hernandez-Rodriguez, J. (2004) Endothelial cells, antineutrophil cytoplasmic antibodies, and cytokines in the pathogenesis of systemic vasculitis. Curr. Rheumatol. Rep. 6,184-194[Medline]



This article has been cited by other articles:


Home page
 Proc. Natl. Acad. Sci. USAHome page
B. G. Zani, K. Kojima, C. A. Vacanti, and E. R. Edelman
Tissue-engineered endothelial and epithelial implants differentially and synergistically regulate airway repair
PNAS, May 13, 2008; 105(19): 7046 - 7051.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
fj.06-7051comv1
21/7/1515    most recent