| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,1
* Department of Pathology and
Veterinary Service, University of Regensburg, Regensburg, Germany; and
Department of Pathology and
Department of Physiology, Immunology Programme, National University of Singapore, Singapore
1 Correspondence: Department of Physiology, National University of Singapore, 2 Medical Dr., MD 9, Singapore 117597. E-mail: phssh{at}nus.edu.sg
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: extravasation • vascular biology
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2
3)
. Costimulation through CD137 enhances T cell activity, enabling the immune system to reject tumors or transplants (4
, 5)
.
The ligand for CD137 is expressed by antigen presenting cells as a transmembrane protein, and it too can deliver signals into the cells it is expressed on. Therefore, bidirectional signaling exists for the CD137 receptor/ligand system (6)
.
Signaling through CD137 ligand enhances proliferation and immunoglobulin secretion of B cells (7)
, and the expression of costimulatory molecules, cytokines, and cellular adherence in dendritic cells (8
9
10)
. In peripheral monocytes it induces activation, prolongation of survival, and proliferation (11
12
13
14)
. Monocytes are generated in the bone marrow. They enter the circulation from where they migrate into the tissues and differentiate to tissue macrophages. There they participate in the regulation of inflammatory and immune reactions, and in physiological processes such as wound healing (15)
.
Leukocyte extravasation begins with binding of carbohydrate moieties on circulating leukocytes to selectins, adhesion molecules on the vascular endothelium. This slows the cells down to a "rolling". Tight binding is mediated by leukocyte integrins and endothelial cell molecules, such as intracellular adhesion molecule. Interations of integrins, molecules of the immunoglobulin family and the extracellular matrix (ECM), mediate the final step: the passage through the endothelium and the vessel wall. While the general principle of leukocyte extravasation is well established, relatively little is known about how the selective extravasation of leukocyte subsets is regulated (16
, 17)
.
Here we show that (1) CD137 can be strongly expressed by blood vessel walls at sites of inflammation; (2) expression of CD137 is induced by proinflammatory cytokines in endothelial cells; and (3) CD137 induces monocyte migration in vitro and in vivo. These results imply that CD137 as an important regulator of monocyte extravasation in inflammatory tissues.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Human IgG1 Fc protein was purchased from Accurate Chemical and Scientific Corporation, (Westbury, NY, USA). Anti-CD14 antibody (Ab) was ordered from Immunotech, Marseille, France. Anti-CD137 Ab (clone BBK-2) and its isotype control, MOPC21, were obtained from Bioscource (Ratingen, Germany) and Sigma (Deisenhofen, Germany), respectively. Anti-CD137 Ab (clone M127) was a gift from Immunex Corp. (Seattle, WA, USA). Growth factor-reduced murine matrigel was obtained from Becton Dickinson (Mountain View, CA, USA).
Cells and cell culture
Human peripheral blood mononuclear cells (PBMC) were prepared by percoll gradient density centrifugation as described previously (7)
. Primary monocytes were isolated by elutriation (19)
. Elutriated monocytes were more than 95% pure, and contaminating T cells were less than 3% as estimated by morphology and antigenic phenotype (CD14, CD3, CD4, CD8 expression). Cells were cultured in polystyrene dishes (Becton Dickinson, Franklin Lakes, NJ, USA) in RPMI 1640 supplemented with 5% FCS at a concentration of 106/ml.
Primary human umbilical vein endothelial cells (HUVEC) were isolated from umbilical cord vein. The umbilical cord vein was cannulated and flashed with copious amounts of HBSS to remove red blood cells and cell debris. The vein was filled with 0.1% type 1 collagenase (Sigma) and incubated at 37°C for 10 min. Next, the cord was gently messaged to dislodge the endothelial layer, and the endothelial cells-collagenase mixture collected into a 20 ml syringe. The vein was rinsed with more HBSS, and the pooled cell suspension centrifuged at 400 g for 10 min at 4°C. The cell pellet was resuspended in medium M199 supplemented with 20% FCS, 2 mM L-glutamine, 100 U/100 mg/ml penicillin/streptomycin, 1 mg/ml heparin (Sigma) and 1x endothelial growth factor (Sigma) and plated out in a 60 mm culture dish. The cells were fed with fresh medium once every 3 d and routinely subcultured at a split ratio of 1:4. Cells were used between passages 3–6 for the experiments described below.
Murine peritoneal exudate cells were isolated by injecting a mouse intraperitoneally (i.p.) with 1 ml of PBS. After 16 h, the mouse was sacrificed and cells that had migrated into the peritoneum were recovered by injection of 3 ml of ice-cold PBS.
RT-polymerase chain reaction (RT-PCR)
Total RNA was isolated from PBMC using RNAzol B (Tel-Test, Inc., Friendswood, TX, USA) and up to 5 µg RNA was reverse-transcribed in a 20 µl vol, using random hexanucleotide primers (50 µg/ml), 25 µM dNTP, 10 mM DTT, 200 U SuperScriptTMII RNaseH-RT (Life Technologies, Inc., Eggenstein, Germany) and 20 U RNAsin (Roche, Mannheim, Germany) for 60 min at 42°C.
The RT reaction (2 µl) served as template for the subsequent PCR, which was performed in a 20 µl vol with 1 U TaqDNA polymerase (Roche); 200 µM dNTPs; 1.5 mM MgCl2; 10 mM Tris, pH 8.3; 50 mM KCl; and 10 µM of each primer. After a 5 min denaturation step at 94°C, the reaction proceeded in 30 cycles of 30 s at 94°C, 30 s at 58°C, and 1 min at 72°C, followed by 10 min at 72°C.
Primers used were:
CD137 sense: 5' ATCATG GGAAACAGCTGTTACAAC
CD137 antisense: 5' TGGTCCACAGACCACGTCCCTCTC
cyclophilin sense: 5' GTCCAGCATTTGCCATGGAC
cyclophilin antisense: 5' GACAAGGTCCCAAAGACAGC
Real-time quantitative PCR
Total RNA was extracted from endothelial cells using RNAsy® Mini Kit (Qiagen). Total RNA (1 µg) was used for first-strand cDNA synthesis with RevertAidTM First-strand cDNA Synthesis Kit (Fermentas) on Mastercycle® Gradient (Eppendorf, Hamburg, Germany) PCR machine.
Quantitative real-time PCR was performed on Lightcycler® 1.2 System (Roche) using LightCycler® FastStart DNA MasterPLUS SYBR Green I (Roche). Human cyclophilin A was used as the normalizing gene. The following primers, which have an amplicon size of 195 bp for both CD137 and Cyclophilin A, were used. Cyclophilin A forward: 5' CCATGGCAAATGCTGGACCC; Cyclophilin A reverse: 5' CGAGTTGTCCACAGTCAGCA; CD137 forward: 5' CCCTGCGAGA-GAGCCAGGA; CD137 reverse: 5' GCCATCTTCCTCTTGAGTAGTTT.
The assay was run according the manufacturer’s instructions with the following changes. The reaction volume was halved to 10 µl per reaction, and primer concentration was adjusted to 0.4 µM for normalizing reactions and 0.3 µM for CD137 reactions. cDNA (15 ng) were used from cyclophilin A reactions, and 2 µg of cDNA was used for CD137 reactions after optimization with the standard curve. The PCR conditions used in all reactions were Preincubation at 95°C for 10 min (Hot start); Amplification for 45 cycles of 5 s at 95°C, 5 s at 57°C, 8 s at 72°C; Melting Curve Analysis at 95°C for 20 s, 60°C for 20 s, and temperature gradient from 60°C to 95°C at 0.1°C/s. Data obtained were analyzed with LightCycler® Software 3.5.3 (Roche).
Generation of spheroids
COS cells (106) were transfected with an expression vector for full-length CD137 or an empty vector using the Dextran-diethylaminoethyl (DEAE) method. After 24 h, the formation of spheroids was initiated. Cells were trypsinized and resuspended at a concentration of 104/ml in Dulbecco’s modified Eagle medium (DMEM) + 10% FCS (Gibco BRL, Eggenstein, Germany). Cells 103/well were seeded into a 96-well microtiter plate. Each well was coated with 100 µl of 1.5% agarose to prevent cell attachment. After 24 h of culture, the medium was changed to RPMI + 10% FCS and 104 peripheral monocytes in 100 µl were added to the spheroids with a diameter of 
100 µm. Two days later spheroid cocultures with a final size of 200–300 µm were embedded in tissue-freezing medium (Leica Instruments, Nussloch, Germany) and frozen in liquid nitrogen. Sections of the spheroids were stained with anti-CD14 and anti-CD11c for the presence of monocytes.
Immunohistochemistry
Tissues for immunohistochemistry were obtained from the tissue bank of the Department of Pathology, at the University of Regensburg, Germany, or from archival samples from the Department of Pathology, of the National University of Singapore. Frozen tissue sections were fixed with 2% paraformaldehyde for 10 min. Endogenous peroxidases were inactivated by 6.5% hydrogen peroxide in methanol for 15 min. Unspecific staining was blocked by 3% dry milk in PBS for 30 min. Anti-CD137 (2 µg/ml; clone BBK-2, Bioscource, Ratingen, Germany) or an isotype control Ab (MOPC21, Sigma) in 3% dry milk were added overnight. Staining was continued with the avidin-biotin complex (ABC) kit (Dako, Hamburg, Germany) by using diaminobenzidine as substrate. The entire procedure was performed at room temperature, and after each step the samples were washed three times with PBS. Tissue sections were stained with hematoxylin and embedded in Entellan (Merck, Darmstadt, Germany).
Embedded spheroids were cut with a cryostat (Reichert, Heidelberg, Germany) at 5 µm, mounted on poly-L-lysin (Sigma) coated glass slides and air-dried according to a standard alkaline phosphatase antialkaline phosphatase (APAAP)-technique. Briefly, the specimens were fixed for 15 min in acetone and chloroform each at room temperature and incubated with mouse monoclonal antibodies against CD11c (Dianova, Hamburg, Germany) or CD14 (My 4, Coulter, Krefeld, Germany). After rinsing with TBS-buffer, rabbit anti-mouse immunoglobulin (Dakopatts, Hamburg, Germany) was applied as secondary Ab followed by incubation with the APAAP complex (Dianova, Hamburg, Germany). The incubation with rabbit anti-mouse Ig and APAAP complex was repeated once. Subsequent incubation with new fuchsin substrate resulted in a red precipitate. Finally, the specimens were counterstained with hematoxylin and mounted.
Mice
Female NMRI mice of 8–12 wk of age were purchased from Charles River (Sulzfeld, Germany).
Histology
At the end of the experiment mice were killed by rapid cervical dislocation. Immediately after death, tissue samples, including matrigel; adhering subcutanous tissue; and parts of the skin were carefully prepared. The specimens were fixed in Bouin’s solution and embedded for routine paraffin histology. Deparaffinized 5 µm sections were stained with hematoxilin and eosin as well as by the method of Masson and Goldner, respectively, for histological examination (20)
.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A). CD137 expression on blood vessel walls was not detected in healthy tissues but was a feature of inflamed tissues.
|
We could identify CD137-positive blood vessels in the skin of a patient who suffered from vasculitis caused by Borreliosis, in the nasal septum of a patient suffering from chronic rhinitis polyposa, in the colon of a Crohn’s disease patient, and in the inflamed thyroid gland of a patient with Morbus Basedow’s (Grave’s disease). Among the CD137-positive vessels were arteries, microvessels, and venules. Both, the endothelial cell lining, as well as the much thicker layer of VSMC, stained positive for CD137.
To exclude cross-reactivity of the monoclonal anti-CD137 Ab to an unrelated antigen, we confirmed expression of CD137 on blood vessel walls by staining with a different Ab, a polyclonal, affinity-purified chicken anti-CD137 serum. Also this Ab detected expression of CD137 on blood vessel walls in the nasal septum of the chronic rhinitis polyposa patient (Fig. 1A
).
The inducibility of CD137 mRNA was also verified in pure cultures of HUVEC using quantitative PCR. The purity of the endothelial cell population was verified by staining for CD31, an endothelial cell-specific marker (Fig. 1B
).
HUVEC (106) were activated with 10 ng/ml of IL-1 + 10 ng/ml of TNF for 24 h. Untreated cells were used for base line reference. The theoretical crossing point of a single copy per reaction is at cycle 37 in the Roche lightcycler system. The crossing point for CD137 in untreated endothelial cells was at 35.5 cycles with 2 µg of cDNA used. This indicates that CD137 is expressed at very low levels or not at all in resting endothelial cells. In the IL-1 + TNF-activated endothelial cells, CD137 expression was increased 79-fold relative to the untreated cells. The primers for the quantitative PCR were designed to span an intron in order to avoid amplification of genomic DNA. The inducibility of CD137 expression by proinflammatory stimuli corresponds well to the presence of CD137-positive vessels in inflamed tissues and their absence in healthy tissues.
CD137 initiates migration of monocytes in vitro
The expression of CD137 on blood vessel walls at sites of inflammation and its induction by proinflammatory stimuli, together with earlier described activities that CD137 induces activation, survival, and proliferation in monocytes, suggested that CD137 may be involved in monocyte extravasation.
We tested this hypothesis by culturing peripheral human monocytes on dishes, which were coated with a fusion protein consisting of the extracellular domain of human CD137, fused to the constant domain of human IgG1 (CD137-Fc). Fc control protein-coated and untreated plates were used as negative controls. Coating was performed with a solution of 5 µg/ml protein in PBS at 4°C overnight. Monocytes were isolated by elutriation and used without prior treatment or culture. Monocyte movements on the coated dishes were measured by time-lapse microscopy under tissue culture conditions (37°C, 5% CO2). Only 5.5 ± 0.5% of the monocytes, which were cultured in wells coated with the Fc control protein had migrated a distance of 
20 µm in the culture dishes. This number was significantly (P=0.034) increased to 12.9 ± 2% when dishes were coated with CD137-Fc protein (Fig. 2
).
|
To reflect more closely the physiological situation of monocyte extravasation from CD137-positive vessels, we generated spheroids expressing CD137. COS cells were transiently transfected with an empty expression vector (pCDNA3, Invitrogen, San Diego, CA, USA) or with a vector, expressing the full-length CD137 protein (pCDNA3-CD137). About 25% of the pCDNA3-CD137 transfected COS cells stained positive for CD137 by flow cytometry after 24 h (not shown). Spheroids were generated from CD137-expressing and control cells by a 24 h incubation on agar-coated, nonadherent 96-well plate culture dishes to which the cells could not attach. Primary monocytes were then added and after 48 h cryosections were made from the spheroids. Sections from the middle of the spheroids were stained with an Ab directed against the monocyte-specific protein CD14 (Fig. 3
A). The numbers of monocytes, which had migrated into the spheroids, were determined by image analysis of four equally sized sections (median sections) of each of five spheroid/monocyte cocultures. About six-times more monocytes were found in the CD137-expressing spheroids than in the control spheroids (Fig. 3B
). Identical results were obtained when the monocytes were stained for CD11c, another monocyte marker (Fig. 3B
).
|
Migration of monocytes into CD137-expressing spheroids more closely resembles the in vivo situation when monocytes migrate through blood vessel walls. It also demonstrates that CD137 is sufficient to induce monocyte migration and that the Fc domain of the CD137-Fc fusion protein is not required.
Functional immobilization of CD137
Next we wanted to determine whether CD137 also enhances migration of monocytes in vivo. As only immobilized but not soluble CD137 induces activation and cytokine secretion in monocytes (11)
, we evaluated whether murine matrigel would be suitable as a carrier and matrix for immobilization of CD137. Matrigel (80 µl) was pipetted into wells of a 96-well plate and incubated at 37°C until solidified. The matrigel was coated with 20 µl of CD137-Fc solution at 2.5 µg/ml or an equimolar amount of Fc protein (20 µl of 1.25 µg/ml). As controls, CD137-Fc protein (2.5 µg/ml) and Fc protein (1.25 µg/ml) were immobilized directly in different wells of the plate. Peripheral human monocytes (105) were added per well, and the IL-8 concentrations in 24 h supernatants were determined by ELISA. Both, CD137 protein, which was immobilized onto matrigel, and CD137 protein, which was immobilized onto the plate, induced strong and comparable activation of monocytes as evidenced by IL-8 secretion, demonstrating that matrigel-bound CD137 is functional (Fig. 4
A).
|
Further, we tested whether human CD137 can cross-link murine CD137 ligand. Human CD137-Fc was almost as effective in inducing IL-6 release from murine peritoneal exudate cells, which are mainly monocytes and macrophages, as murine CD137-Fc protein (Fig. 4B
).
CD137 initiates migration of monocytes in vivo
CD137-induced monocyte migration was studied in vivo by loading matrigel with CD137-Fc or Fc protein and implanting it into mice. A 10 µg/ml CD137-Fc protein solution (20 µl) or equimolar solution (20 µl) of Fc control protein (5 µg/ml) was mixed with 80 µl murine and growth factor-reduced matrigel and allow to solidify, which resulted in 100 µl of matrigel that contained either 200 ng of CD137-Fc protein or the equimolar amount of Fc control protein.
Five NMRI mice were anesthetized, and each mouse received a 5-mm-long cut on the right and left flank. CD137-Fc and Fc control protein-containing pieces of matrigel were introduced into the cuts on the right and left flank of each mouse, respectively, and the wounds were sealed with metal clips. Introduction of CD137-Fc-coated and Fc-coated matrigel on either flank of the same mouse served as a control for potential differences between individual animals. Tissues at wound sites were harvested 7 d later and embedded in paraffin. Monocytes were identified by Masson-Goldner staining. Few monocytes were present at the Fc-containing matrigel, and some adhered to the surface of the matrigel (Fig. 5
). In contrast, a large number of monocytes had accumulated around the CD137-Fc containing matrigel and had massively infiltrated the matrigel (Fig. 5)
. This experiment confirmed the migration-inducing activity of CD137, which was observed by time-lapse microscopy, and in CD137-expressing spheroids and demonstrated that CD137 exerts this activity also in vivo.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 22
, 6)
. In this study we have identified CD137 as a molecule that can be expressed by cells of inflamed blood vessel walls and induces monocyte migration. Blood vessel walls are sites of essential biological processes, such as blood coagulation and leukocyte extravasation. These processes are generally restricted to sites of injuries and are regulated by the inducible and restricted expression of molecules, such as cell adhesion molecules, on endothelial cells. CD137 has been shown previously to induce activation, prolonged survival, and proliferation of monocytes. Therefore, it was tempting to speculate that the most likely function of CD137 on blood vessel walls would be a facilitation of monocyte recruitment to sites of inflammation. We could support this assumption by demonstrating an enhancement of the in vitro migratory activity of monocytes by CD137 and by showing increased monocyte migration into CD137-expressing spheroids. The latter more closely resembles an in vivo situation when monocytes migrate through blood vessel walls. The experiment with CD137-expressing spheroids also clearly demonstrates that CD137 is sufficient to induce monocyte migration and that the Fc domain of the CD137-Fc fusion protein is not required.
Since human CD137 protein also activates murine monocytes, it was possible to confirm induction of monocyte migration by CD137 in vivo. Introduction of CD137-Fc-coated and Fc-coated matrigel on either flank of the mice served as an ideal control to account for potential differences between individual animals. Monocytes accumulated at and infiltrated into the CD137-Fc containing matrigel, while these processes did not occur at the Fc-containing matrigel.
These data suggest that CD137 may be haptotactic for monocytes. Attraction and recruitment of monocytes is a typical feature of chemokines, typically soluble mediators. CD137 was expressed as a membrane-bound molecule in the spheroids, which raises the question how CD137 could induce migration of monocytes into the spheroids. Most likely, monocytes were induced to migrate by CD137-expressing COS cells at the spheroid surface and kept on migrating as long as they had contact to neighboring CD137-expressing COS cells. This possible mechanism would explain the even distribution of the monocytes in the spheroids. The situation was similar in the experiment in which CD137-Fc protein was immobilized on matrigel. CD137-Fc protein was mixed into the matrigel before it solidified, ensuring an even distribution. Also in vivo CD137 appears to be homogeneously expressed throughout the entire blood vessel wall. This would ensure the continuous migration of monocytes through the vessel walls, while chemokine gradients could determine the direction.
Selectins, integrins, and molecules of the immunoglobulin family have been identified as key regulators for leukocyte extravasation, raising the question of why CD137 would be required in addition (23)
. CD137 induces activation and migration of monocytes (11
12
13
14)
, while it delivers inactivating signals to T cells (18
, 24)
. Therefore, it is possible that CD137 expression on inflamed blood vessels favors recruitment of monocytes over recruitment of T cells. Similarly, CD134 (OX40) ligand is expressed on endothelial cells at inflammatory sites and favors recruitment of CD134-expressing CD4-positive T cells (25)
. Vascular adhesion protein 1 (VAP-1) has also been shown to be expressed activation-dependently on the vascular endothelium and to facilitate extravasation of granulocytes (26)
. CD137, CD134 ligand, and VAP-1 may just be examples of a much larger group of cell-type specific extravasation signals, which selectively recruit leukocyte subpopulations and thereby influence the nature and course of the inflammatory process.
CD137 is a very promising target for cancer immunotherapy. CD137 is expressed as a costimulatory receptor on activated T cells, and CD137 agonists such as antibodies have shown remarkable therapeutic effect in a range of murine tumor models and are currently being developed for human therapy (27
28
29
30
31
32)
. Data from this study imply that anti-CD137 antibodies that would be administered for cancer therapy could also interfere with monocyte recruitment to sites of inflammation. This is an aspect that should be addressed before anti-CD137 antibodies are being tested in patients.
In this study we have identified expression of CD137 on cells of the blood vessel wall at sites of inflammation, and an enhancement of monocyte migratory activity in vitro and in vivo, which suggests a novel function for CD137, i.e., the recruitment of monocytes in inflammatory tissues. These data, together with earlier ones, which demonstrate induction of activation, survival, and proliferation of monocytes by CD137 substantiate CD137 as an important regulator of monocyte activities (11
12
13
14)
.
|
ACKNOWLEDGMENTS |
|---|
Received for publication January 30, 2006. Accepted for publication February 17, 2006.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. S. Olofsson, L. A. Soderstrom, D. Wagsater, Y. Sheikine, P. Ocaya, F. Lang, C. Rabu, L. Chen, M. Rudling, P. Aukrust, et al. CD137 Is Expressed in Human Atherosclerosis and Promotes Development of Plaque Inflammation in Hypercholesterolemic Mice Circulation, March 11, 2008; 117(10): 1292 - 1301. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Sun and P. J. Fink A New Class of Reverse Signaling Costimulators Belongs to the TNF Family J. Immunol., October 1, 2007; 179(7): 4307 - 4312. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
