| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
From the Department of Renal Immunobiology,*
Division of
Medical Sciences, and the Department of Pathology,
The
Medical School, University of Birmingham, Birmingham, United Kingdom
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Leukocyte adhesion is an early event for
the subsequent migration into extravascular tissues.8
T
lymphocyte interactions are mediated by the ß1 integrin
VLA-4 (
4ß1) and the ß2 integrin LFA-1 (CD11a/CD18).
Integrins can convert between low- and high-affinity states, depending
on the local environment (ie, presence of cytokines and chemokines), to
mediate leukocyte adhesion.9
LFA-1 interacts with
tissue-expressed ICAM-1, whereas VLA-4 can interact with both VCAM-1
and an alternatively spliced 25-mer CS-1-peptide-containing isoform of
the extracellular matrix glycoprotein fibronectin
(CS-1Fn).10
VCAM-1 and ICAM-1 are constitutively present in the normal kidney. There is distinct VCAM-1 expression on the glomerular parietal epithelium and also on occasional tubular epithelial cells, whereas ICAM-1 expression is prominent in the glomeruli and sparse in peritubular regions. In vasculitic GN and acute allograft rejection, both ICAM-1 and VCAM-1 are heavily expressed in peritubular regions and on tubular epithelial cells11-14 ; however, glomerular expression of VCAM-1 associated with areas of segmental necrosis is a discerning feature of vasculitic GN.15,16 Increased expression of fibronectin has been reported in experimental immune complex glomerulonephritis and rejecting renal allografts.17-19
Here, we describe the first study using peripheral blood T lymphocytes in a modified Stamper-Woodruff adhesion assay20 to investigate the functional roles of ICAM-1, VCAM-1, and CS-1Fn in human renal inflammatory states. In previous studies, this technique has been applied to demonstrate adhesion molecule interactions in a variety of inflammatory conditions, including ICAM-1-dependent T cell adhesion in a rat model of acute renal allograft rejection,21 VCAM-1- and ICAM-1-mediated adhesion of T cell and macrophage-derived cell lines in murine lupus nephritis,22 CS-1Fn-mediated T cell adhesion in rheumatoid arthritis,23 ICAM-1- and VCAM-1-dependent T cell adhesion in cervical intraepithelial neoplasia,24 VCAM-1-mediated monocyte adhesion in simian encephalitis,25 and P-selectin-dependent eosinophil adhesion in asthma and nasal polyposis,26 and a role for multiple adhesion molecules in mediating monocyte adhesion to human atherosclerotic plaques.27
We have studied the adhesive properties of peripheral blood T lymphocytes to renal biopsies taken from patients with vasculitic GN and acute allograft rejection. Furthermore, we have used an LFA-1-expressing KL/4 cell line28 and a VLA-4-expressing Jurkat cell line in this assay system to demonstrate differential roles of ICAM-1/LFA-1-, VCAM-1/VLA-4-, and CS-1Fn/VLA-4-mediated interactions within the kidney in these disease states.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Renal biopsy specimens from 18 patients with anti-neutrophil cytoplasmic autoantibody (ANCA)-positive vasculitis (Wegener's granulomatosis or microscopic polyangiitis with appearance of segmental glomerular lesions) and 18 patients with acute cellular allograft rejection (scoring 4(IA) to 4(IIA) on the 1997 Banff diagnostic categorization system) were used in these studies. Studies with leukocytic cell lines were performed on biopsies from 8 patients with vasculitis and 7 patients with acute rejection, whereas studies with peripheral blood T lymphocytes were performed on an additional 10 vasculitis and 11 acute rejection patient biopsies. Each biopsy was embedded in Brights cryo-m-bed (Merck, Dorset, UK), snap-frozen in liquid nitrogen, and stored at -70°C. Control tissue comprised cortical fragments from the unaffected pole of five kidneys removed for renal carcinoma. Serial 7 to 10-µm cryostat sections collected onto Superfrost Plus slides (Merck), air dried and fixed in acetone (Merck) for 10 minutes, were used in all subsequent studies. Approval for the study of human material was obtained from the local research ethics committee.
Cell Culture
A Jurkat human T lymphoblastoid cell line (a gift from the Department of Cancer Studies, University of Birmingham) and the K562 human erythroleukemic cell line transfected with LFA-1, KL/4 (a gift from Dr. M. Robinson, Celltech, Slough, UK) were used in these studies. Both cell lines were maintained in RPMI-1640 supplemented with 10% fetal calf serum (FCS), 100 U/ml penicillin, 100 µg/ml streptomycin, and 2 mmol/L glutamine. For the KL/4 cells the medium was additionally supplemented with 0.2 mg/ml G418 (GIBCO, Paisley, UK).
Antibodies
The following mouse monoclonal antibodies were used: anti-human ICAM-1, RR1 (a gift from Dr. R. Rothlein; Boehringer-Ingelheim R&D, Ridgefield, CT); anti-human VCAM-1 cell culture supernatant, Ig11 (a gift from Professor D. Haskard; Imperial College, London, UK); anti-human VCAM-1, BBIG-VI (R&D Sytems, Oxon, UK); anti-human CD11a, BCA1 (R&D Systems); anti-human VLA-4, Max68, and anti-human LFA-1, KIM185 (gifts from Dr. M. Robinson); anti-human CS-1Fn cell culture supernatant, 90.45 (gift from Dr. M. Elices; Cytel Corp. San Diego, CA); and anti-human CD3 (DAKO, High Wycombe, UK) conjugated to horseradish peroxidase (HRP; conjugate synthesized within the laboratory). All monoclonal antibodies to adhesion molecules were blocking except KIM185 (ß2 integrin activating antibody). RR1 and BBIG-VI were suitable for immunohistochemistry. We also used biotinylated rabbit anti-mouse IgG (DAKO) and HRP-conjugated streptavidin ABC complex (DAKO). Irrelevant mouse IgG (Serotec, Oxford, UK) and mouse IgM (Binding Site, Birmingham, UK) were used as controls.
Tissue Preparation
All incubations were performed at room temperature in a humidified box. Tissue endogenous peroxidase activity was blocked by treating the sections with Tris-buffered saline (TBS; pH 7.4) containing 0.3% H2O2 and 0.1% NaN3, for 10 minutes.
Immunohistochemistry
The sections were treated sequentially with 0.1% avidin and 0.01% biotin to block endogenous biotin, followed by a 10-minute blocking stage with 10% rabbit serum. Three-stage indirect immunohistochemistry was performed with primary anti-human monoclonal antibodies (ICAM-1 (RR1) or VCAM-1 (BBIG-VI)), a biotin-labeled rabbit anti-mouse secondary antibody and streptavidin ABC/HRP complex. Binding of the tertiary complex was visualized by the addition of 3,3'-diaminobenzidine (DAB; Sigma, Dorset, UK). The sections were counterstained with Mayer's hematoxylin (Merck), mounted in dibutyl polystyrene xylene (DPX) (Merck), and coverslipped.
Adhesion Studies
Before the adhesion assay, cells in suspension were stimulated (peripheral blood T lymphocytes and Jurkat cells with 1 ng/ml phorbol 12-myristate 13-acetate (PMA; Sigma) and KL/4 cells with 25 µg/ml KIM185 for 30 minutes at 37°C. Acetone-fixed serial sections and preactivated cell suspensions were co-incubated in an agitated system in the presence of irrelevant control antibody or blocking monoclonal antibody. The binding of peripheral blood T lymphocytes isolated from patients showing evidence of acute rejection to their own biopsies was assesed. Due to the lower incidence of vasculitis as compared with acute rejection and the limited availability of blood before initiation of therapy in patients with vasculitis, T lymphocytes obtained from control volunteers were used to address peripheral blood T cell adhesion processes in vasculitic GN. We also investigated specific LFA-1- and VLA-4-mediated interactions using an LFA-1-transfected KL/4 cell line and a VLA-4-expressing Jurkat cell line, respectively. The cell numbers (see below) used in the studies were carefully optimized to delineate specific localization of adherent cells while obtaining an appreciable level of adhesion to determine the effect of blocking antibodies.
Peripheral Blood T Lymphocyte Adhesion
Peripheral blood from patients with acute cellular allograft rejection and control volunteers was collected and heparinized. Peripheral blood mononuclear cells were isolated by centrifugation over Ficoll/Hypaque. Mononuclear cells were collected from the interface, washed, and incubated in RPMI-1640 with 10% FCS in a petri dish (20 x 106 cells in 10 ml per dish) for 1 hour at 37°C in 5% CO2. T lymphocytes (nonadherent cells) were collected by gentle washing with assay medium.
Preactivated T lymphocytes suspended at 5 x 106 cells per ml were overlaid on the tissue sections in a volume of 100 µl and incubated on a rotating platform at 70 revolutions per min (rpm) for 30 minutes at room temperature. Nonadherent cells were washed off with cold TBS, and sections were fixed in acetone for 10 minutes. Bound cells were labeled with the anti-CD3 HRP conjugate, and the cells were visualized by developing their peroxidase activity with DAB for 30 minutes. The sections were counterstained with Mayer's hematoxylin, mounted in DPX, and coverslipped.
KL/4 Cell Adhesion
Preactivated KL/4 cells, suspended at 5 x 106 cells per ml in RPMI-1640 containing 10% FCS, were overlaid on tissue sections in a volume of 100 µl and incubated on a rotating platform at 70 rpm for 30 minutes at 37°C. The slides were then washed in cold TBS to remove any nonadherent cells. Adherent cells were fixed in acetone for 10 minutes. Bound KL/4 cells were visualized by incubation with DAB for 30 minutes, using the endogenous peroxidase activity of K562 cells. The sections were counterstained with Mayer's hematoxylin, mounted in DPX, and coverslipped.
Jurkat Cell Adhesion
Preactivated Jurkat cells suspended at 5 x 106 cells per ml in RPMI-1640 containing 10% FCS were overlaid on the tissue sections in a volume of 100 µl and incubated on a rotating platform at 70 rpm for 30 minutes at room temperature. Nonadherent cells were washed off with cold TBS, and sections were fixed in acetone for 10 minutes. Bound cells were labeled with the anti-CD3 HRP conjugate, and the cells were visualized by developing their peroxidase activity with DAB for 30 minutes. The sections were counterstained with Mayer's hematoxylin, mounted in DPX, and coverslipped.
Blocking Studies
Before the co-incubation, serial sections and the cell suspension were incubated for 30 minutes with neutralizing antibodies directed against VCAM-1 and ICAM-1 at room temperature and LFA-1 or VLA-4 at 37°C. In preliminary studies, a concentration of 20 µg/ml consistently provided maximal blocking for all antibodies tested.
Combined Tissue Adhesion Assay and Immunohistochemistry
To co-localize leukocyte adhesion and tissue adhesion molecule expression, after the adhesion assay, sections were fixed in acetone for 10 minutes and stained for VCAM-1 and ICAM-1 using the immunoperoxidase staining technique described above. This order of performing the combined assay precluded any intervention by the antibodies in the initial adhesion assay.
Data Analysis
We analyzed glomerular and tubulointerstitial adhesion in turn. For each section the number of adherent cells in 5 x 625-µm2 fields of tubulointerstitium were counted using a graticule and expressed as number of adherent cells per field. The total number of glomeruli in each section was counted, and the total number of adherent cells on the glomerular tuft and the parietal epithelium was expressed as number of adherent cells per glomerulus.
Data are quoted as mean and standard error of mean (mean ± SEM). Significance of differences in adhesion between control and neutralizing antibodies was determined using a one-tailed two-sample t-test. Nonparametric statistical analyses using the Mann-Whitney two-sample test confirmed the results of the t-test. Statistical advice was supplied by Dr. Roger L. Holder from the Department of Mathematics and Statistics at the University of Birmingham.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Immunohistochemical (IHC) analysis confirmed the presence and
up-regulation of the tissue ligands VCAM-1 and ICAM-1 in both disease
states. In both vasculitic GN and acute rejection there was marked
VCAM-1 expression in the parietal epithelium and tubular epithelium and
at peritubular regions of the interstitium. In addition, there was
distinct intraglomerular expression of VCAM-1 in vasculitic GN, which
was notably absent in acute allograft rejection (Figure 1, A and B)
.
|
.
|
Peripheral Blood T Lymphocyte Adhesion in Vasculitic Glomerulonephritis
Adhesion of resting T lymphocytes isolated from peripheral blood to cryostat sections of renal biopsy material was sparse at both glomerular and tubulointerstitial sites. Hence, T cell integrin molecule activation with PMA was necessary to achieve optimal levels of adhesion and evaluate the inhibitory effects of blocking monoclonal antibodies. T lymphocytes did not adhere to normal tissue sections irrespective of their activation state.
T lymphocyte adhesion at glomerular and tubulointerstitial sites was
differentially attenuated using the anti-ICAM-1 monoclonal antibody
(Figure 3)
. Inhibition at glomerular
sites was greater than at tubulointerstitial sites, reducing T cell
adhesion by 58% (from 14.55 ± 2.4 to 6 ± 0.79) compared
with 37% (from 22.1 ± 2.67 to 13.76 ± 1.4), respectively
(Figure 4)
. In additional studies, the
combination of anti-LFA-1 with anti-ICAM-1 had no effect over and above
those seen with anti-ICAM-1 alone. Inhibitory effects of anti-VCAM-1
monoclonal antibody were also more prominent within the glomerular
compartment. Glomerular T cell adhesion was reduced by 56% (from
14.55 ± 2.4 to 6.31 ± 2.29) as compared with 47% (from
22.1 ± 2.67 to 11.7 ± 1.56) at tubulointerstitial sites
(Figures 4 and 5)
. In additional studies
using biopsies from an additional five patients, anti-VCAM-1 combined
with anti-VLA-4 produced a small increase in inhibition of adhesion at
both glomerular and tubulointerstitial sites. Thus, in these studies
anti-VCAM-1 showed 63% and 46% inhibition at glomerular and
tubulointerstitial sites, respectively, anti-VLA-4 showed 50% and 30%
inhibition at each site, whereas the antibodies in combination
inhibited by 73% and 54% at each site.
|
|
|
. T cell adhesion was further reduced at both
glomerular and tubulointerstitial regions when anti-CS-1Fn and
anti-VCAM-1 monoclonal antibodies were used in combination, producing
maximal inhibition of up to 72% (Figures 4 and 5)
. This level of
inhibition was similar to that observed with combined anti-VLA-4 and
anti-VCAM-1. KL/4 Cell Adhesion in Vasculitic Glomerulonephritis
We used an LFA-1-transfected K562 cell line (KL/4) that was 97.1%
LFA-1 positive (determined by FACS analysis), to isolate LFA-1-mediated
adhesion processes. Adhesion of KL/4 cells was ß2
integrin activation dependent as no adhesion was observed by resting
KL/4 cells in the absence of integrin activation by KIM185. Although
normal tissue sections were able to support activated KL/4 cell
adhesion (7.2 ± 3.7 and 4.7 ± 2 at glomerular and
tubulointerstitial sites, respectively), this was significantly greater
in the inflamed tissue (Table 1)
.
|
. Jurkat Cell Adhesion in Vasculitic Glomerulonephritis
We used a functional VLA-4-expressing Jurkat cell line, which was
97.66% VLA-4 positive (determined by FACS analysis) to isolate VLA-4-
from LFA-1-mediated interactions. As with KL/4 cells, resting Jurkat
cells were unable to adhere to either normal or inflamed tissue
sections. PMA-activated Jurkat cell adhesion to normal tissue was
negligible, but there was substantial adhesion in vasculitic GN. There
was strong adhesion at intraglomerular, periglomerular, and
tubulointerstitial sites. Anti-VCAM-1 significantly inhibited
glomerular (from 8.19 ± 3.2 to 0.89 ± 0.89) and
tubulointerstitial (from 23.34 ± 4.37 to 3.54 ± 2.38)
Jurkat cell adhesion by up to 89% in both compartments (Figure 6)
. Thus, using this naturally
VLA-4-expressing cell line to isolate VLA-4-mediated adhesion, effects
of the anti-VCAM-1 blocking monoclonal antibody were more marked than
with peripheral blood T lymphocytes where 56% and 47% inhibition of
adhesion was achieved in glomerular and tubulointerstitial
compartments, increasing to 73% and 54%, respectively, with combined
anti-VCAM-1 and anti-VLA-4.
|
. However, when anti-VLA-4 and anti-CS-1Fn were
used in combination, Jurkat cell adhesion was significantly reduced by
88% (from 8.19 ± 3.2 to 1 ± 0.8) and by 65% (from
23.34 ± 4.37 to 8.15 ± 3.12) at glomerular and
tubulointerstitial sites, respectively (Figure 6)
. Peripheral Blood T Lymphocyte Adhesion in Acute Rejection
Adhesion of autologous resting T lymphocytes to cryostat sections
of acute allograft rejection renal biopsies was sparse at both
glomerular and tubulointerstitial sites, necessitating previous
treatment with PMA for the adhesion studies. Activated T cells bound to
glomerular and tubulointerstitial sites. Compared with vasculitic GN,
there was equivalent T cell adhesion at tubulointerstitial sites but
less adhesion at glomerular sites (Figures 4
and
7).
In allograft rejection, 57% inhibition of T lymphocyte adhesion was
achieved at tubulointerstitial sites (from 20.02 ± 3.3 to
8.58 ± 2.04) by anti-ICAM-1 monoclonal antibody compared with
49% (from 9.08 ± 1.22 to 4.59 ± 1.25) at glomerular sites
(Figure 7)
. The inhibition by anti-ICAM-1 at tubulointerstitial sites
was greater than the 37% seen in vasculitic GN, but inhibition at
glomerular sites was less than the 58% seen in vasculitic GN (Figure 4)
. Combining anti-ICAM-1 and anti-LFA-1 did not further increase
inhibition over that seen with anti-ICAM-1 alone.
|
, similar to the pattern seen
with the anti-ICAM-1 monoclonal antibody. Tubulointerstitial inhibition
was greater and glomerular inhibition less than seen in vasculitis
(Figure 4)
, reflecting the differences in the glomerular and
tubulointerstitial distribution of VCAM-1 in the two inflammatory
conditions. Combining anti-VCAM-1 and anti-VLA-4 reduced T cell
adhesion by a further 7% at glomerular sites but had no further
inhibitory effect at tubulointerstitial sites.
T cell adhesion was significantly reduced in the presence of
anti-CS-1Fn by 43% (9.08 ± 1.22 to 5.19 ± 2.3) in
glomerular compartments and by 67% (20.02 ± 3.3 to 6.5 ±
1.6) in tubulointerstitial compartments (Figure 7)
. These levels of
inhibition were comparable to the 39% and 62% seen in the glomerular
and tubulointerstitial compartments in vasculitis (Figure 4)
.
Anti-CS-1Fn and anti-VCAM-1 used in combination further reduced
adhesion in tubulointerstitial regions to 81% (from 20.02 ± 3.3
to 3.72 ± 0.72). In the glomerulus, however, there was less
marked enhancement of inhibition of T cell adhesion using the two
antibodies; anti-CS-1Fn and anti-VCAM-1 inhibited by 56% (9.08 ±
1.22 to 3.98 ± 1.6) compared with 43% (9.08 ± 1.22 to
5.19 ± 2.3) with anti-CS-1Fn used alone (Figure 7)
, indicating a
lesser role for VCAM-1 at glomerular sites in acute rejection where
VCAM-1 is mainly expressed along the glomerular parietal
epithelium.
KL/4 Cell Adhesion in Acute Rejection
Adhesion of activated KL/4 cells at glomerular sites was
comparable between acute rejection and vasculitic GN. However, there
was enhanced adhesion of KL/4 cells to the tubulointerstitium in acute
rejection (Table 1)
. Only modest inhibition of adhesion was achieved at
both glomerular and tubulointerstitial sites by blocking antibodies
against ICAM-1 and its leukocyte-expressed ligand LFA-1 (Table 1)
; only
anti-ICAM-1 achieved marginally significant inhibition of 34% at
tubulointerstitial sites.
Jurkat Cell Adhesion in Acute Rejection
Adhesion of Jurkat cells was of similar magnitude to vasculitic GN
at glomerular sites (Figures 6, A and 8
,
A), whereas at tubulointerstitial sites
adhesion was nearly twice as great as in vasculitic GN (Figures 6, B and 8
, B). Tubulointerstitial adhesion was significantly inhibited by
anti-VCAM-1 monoclonal antibody by 70% (from 38.4 ± 6.9 to
11.4 ± 2.8; Figure 8
). Inhibition of glomerular adhesion with
anti-VCAM-1 was 49% (13.71 ± 1.5 to 6.5 ± 1.8), which was
considerably less pronounced than at tubulointerstitial regions (Figure 8)
; this represented inhibition of VCAM-1-dependent adhesion to the
parietal epithelium as there was little or no VCAM-1 expression at
intraglomerular sites. (Figure 8)
. Anti-VLA-4 and anti-CS-1Fn
monoclonal antibodies alone produced significant inhibition of similar
magnitudes in glomerular and tubulointerstitial compartments.
Additional increases in inhibition were observed when the two
antibodies were used together, reaching maximal inhibitions of 64%
(13.71 ± 1.5 to 4.9 ± 3.3) and 72% (38.4 ± 6.9 to
10.8 ± 2.5) at glomerular and tubulointerstitial sites,
respectively (Figure 8)
.
|
We further analyzed the co-localization of VLA-4- and
LFA-1-mediated adhesion to their respective tissue ligands, VCAM-1 and
ICAM-1. LFA-1-transfected KL/4 cells specifically adhered to
ICAM-1-positive regions only (Figure 9A)
.
Jurkat cells, however, adhered to both VCAM-1-positive and -negative
sites (Figure 9B)
. Co-localization of Jurkat cells and VCAM-1 confirmed
VLA-4/VCAM-1 interactions, whereas adhesion at VCAM-1-free sites
supports the involvement of other VLA-4-mediated interactions, such as
VLA-4 binding to CS-1-containing fibronectin.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Immunohistochemical studies on serial sections confirmed previously reported differential patterns of VCAM-1 and ICAM-1 staining in the two disease states.9-14 The T-lymphocyte-expressed integrins LFA-1 and VLA-4 mediate T cell adhesion to tissue-expressed ICAM-1 and VCAM-1, respectively. The activation state of the integrins is a critical determinant of leukocyte adhesion processes.29-31 In our system, low levels of peripheral blood T cell adhesion was possible to inflamed tissues in the absence of any previous activation, but to achieve optimal adhesion for subsequent evaluation of inhibition, the cells had to be pretreated with PMA. In contrast, activation of both the Jurkat and KL/4 cell lines was an absolute prerequisite for any leukocyte adhesion.
Up-regulation of tissue-expressed adhesion molecules and activation of
their leukocyte-expressed integrin ligands promotes T lymphocyte
recruitment to sites of inflammation. This process is closely regulated
by local stimuli such as cytokines and chemokines. In vitro
studies have shown that pro-inflammatory cytokines, such as tumor
necrosis factor (TNF)-, interleukin (IL)-1, and interferon
(IFN)-
, stimulate expression and up-regulation of adhesion molecules
such as ICAM-1 and VCAM-1.32,33
Expression of TNF- and
IL-1 has been demonstrated in vasculitic GN,34
and
cytokines are up-regulated to varying extents in allograft
rejection.35,36
In addition, locally produced chemotactic
mediators or chemokines, such as MCP-1, RANTES, and MIP-1, regulate
the rapid switching of integrins from a low-affinity to a high-affinity
state.37,38
We have shown expression of these chemokines
as well as MIP-1ß in vasculitic glomerulonephritis.39
In vivo, these chemokines may be the true physiological
trigger for integrin activation, and in future studies our system could
be adapted to test this by replacing PMA or the KIM185 antibody with a
relevant chemokine.
Activated peripheral blood T lymphocytes did not adhere to normal tissue sections, but adhesion was seen in both disease states to capillary tuft and parietal regions of the glomerulus as well as to tubular and interstitial regions. Adhesion to glomerular sites was greater in vasculitic GN compared with acute rejection, whereas comparable levels of adhesion were seen to tubulointerstitial sites in both diseases. Adhesion was inhibitable by blocking tissue-expressed adhesion molecules ICAM-1, VCAM-1, and CS-1 domain fibronectin. The pattern and degree of inhibition was variable between disease states and between glomerular and tubulointerstitial compartments within the kidney. T cell adhesion at glomerular sites in vasculitic GN and at tubulointerstitial sites in acute rejection was most sensitive to inhibition by both anti-ICAM-1 and anti-VCAM-1. This observation suggests a major role for ICAM-1 and VCAM-1 in mediating T cell adhesion at glomerular sites in vasculitis and at tubulointerstitial sites in acute rejection. In both diseases, CS-1Fn appeared to play a greater adhesive role within the tubulointerstitium than within glomeruli.
Using VLA-4-expressing (Jurkat) and LFA-1-expressing (KL/4) cell lines the differential roles of ß1 and ß2 integrins were investigated in isolation. KL/4 cells, but not Jurkat cells, were able to adhere to normal kidney sections at glomerular and peritubular sites corresponding to the markedly higher levels of ICAM-1 in normal tissue as compared with VCAM-1. This suggests the potential for constitutively expressed ICAM-1 to mediate T cell trafficking in the normal kidney; however, this would be conditional on the presence of other permissive factors such as appropriate chemoattractant factors, which may not be available in the normal kidney. Adhesion of both cell types was substantial in both diseases, although there were some differences in the patterns of localization. Thus, adhesion of Jurkat cells and KL/4 cells was greater at tubulointersitial sites in acute rejection compared with vasculitic GN, which is consistent with the strong ability of antibodies to ICAM-1, VCAM-1, and CS-1Fn to block binding of peripheral blood T lymphocytes from patients with acute rejection at this site. This may reflect a greater density of VLA-4 adhesion ligands within the tubulointerstitial regions in acute rejection. There was little difference in adhesion of Jurkat cells or KL/4 cells at glomerular sites in the two diseases.
Inhibition of adhesion of both the KL/4 and Jurkat cell lines was achieved using blocking antibodies against LFA-1 and VLA-4 and their tissue ligands ICAM-1, VCAM-1, and CS-1-containing fibronectin. The Jurkat cell line proved the most susceptible to adhesion inhibition using anti-VCAM-1 monoclonal antibody or a combination of anti-VLA-4 and anti-CS-1Fn monoclonal antibodies. Inhibition of KL/4 cell adhesion was less satisfactory, but nonetheless patterns of inhibition seen with anti-ICAM-1 were similar to those seen with peripheral blood T lymphocytes. The expression of ICAM-1 on KL/4 cells40 may contribute to KL/4 cell adhesion to LFA-1 expressed on tissue leukocyte infiltrates. This would not be inhibited by anti-LFA-1 treatment of KL/4 cells in suspension and may explain the low levels of inhibition observed with this antibody.
Furthermore, we have demonstrated adhesion-molecule-specific
localization of KL/4 and Jurkat cells in our modified Stamper-Woodruff
assay system. Curiously, unlike KL/4 cells, which bound solely to
ICAM-1-rich areas, Jurkat cell binding was also noted at VCAM-1-free
sites. This observation and the inhibitory effect of anti-CS-1Fn
suggest an important role for VLA-4/CS-1Fn interactions. Jurkat cells
express a second ß1 integrin, VLA-5,41
which
is also a receptor for the fibronectin receptor. Another 4 integrin
(4ß7) has also been reported on leukocytes. Like VLA-4, this
molecule binds to both VCAM-1 and the CS-1 domain of fibronectin as
well as a mucosal addressin cell adhesion molecule-1,
MAdCAM-1.42,43
It is possible that adhesion molecules such
as MAdCAM-1 and VLA-5 are responsible for additional VCAM-1-independent
adhesion processes, although we have not observed MAdCAM-1 expression
in the kidney (unpublished).
Binding of peripheral blood T cells and Jurkat cells to the glomerular tuft in acute rejection was of particular interest for two reasons. First, the lack of glomerular VCAM-1 expression in acute rejection suggests a role for VLA-4/CS-1Fn or possibly VLA-5/fibronectin interactions. This is supported by the inhibitory effects of anti-CS-1Fn. Second, despite this in vitro evidence showing ability of peripheral blood T cells and specific integrin-molecule-expressing cell lines to bind to glomeruli in acute allograft rejection, little T cell infiltration is observed in vivo.2 This highlights the necessity for other cofactors in the microenvironment. We have found little chemokine expression at this location (unpublished), so the ability to activate T cell integrin ligands may be limited in allograft glomeruli. The reasons for this are unclear but suggest a degree of protection of glomerular sites from allogeneic T cell attack.
To date, a number of studies on animal models have demonstrated that monoclonal antibodies against adhesion molecules can suppress T-lymphocyte-mediated immunological reactions. Anti-ICAM-1 and anti-LFA-1 monoclonal antibodies have been used to protect against the onset of acute cellular rejection and restore deteriorating renal function in experimental models of renal transplant with variable success.44,45 Furthermore, an anti-ß3-integrin antibody for the prevention of arterial restenosis is available, and several other integrin-based drugs are under development.46 This study provides further justification for the development of targeted therapeutic agents that will modify T cell retention and recruitment within the inflamed kidney.
|
Footnotes |
|---|
Supported by the Medical Research Council of Great Britain and the Special Trustees of Former United Birmingham Hospitals.
Accepted for publication October 28, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
expression in 15 cases of anti-neutrophil cytoplasmic autoantibody-associated renal vasculitis. Am J Kidney Dis 1996, 27:48-57[Medline]
4ß1 integrin interactions. Am J Pathol 1994, 144:27-40[Abstract]
, IL-1ß and IL-2R in ANCA-positive glomerulonephritis. Kidney Int 1993, 43:682-692[Medline]
4ß1 and 5ß1 integrin avidity by CC chemokines in monocytes: implications for transendothelial chemotaxis. J Cell Biol 1996, 134:1063-10734ß7 mediates human eosinophil interaction with MAdCAM-1, VCAM-1 and fibronectin. Immunology 1996, 89:112-119[Medline]
This article has been cited by other articles:
 |
|
 |
|
  P. Keshavan, T. L. Deem, S. J. Schwemberger, G. F. Babcock, J. M. Cook-Mills, and S. D. Zucker Unconjugated Bilirubin Inhibits VCAM-1-Mediated Transendothelial Leukocyte Migration J. Immunol., March 15, 2005; 174(6): 3709 - 3718. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. E. Dixon, J. B. Mandac, P. J. Martin, R. C. Hackman, D. K. Madtes, and J. G. Clark Adherence of adoptively transferred alloreactive Th1 cells in lung: partial dependence on LFA-1 and ICAM-1 Am J Physiol Lung Cell Mol Physiol, September 1, 2000; 279(3): L583 - L591. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
