Download Human Leukocyte Antigen-Class II-Positive

Human Leukocyte Antigen-Class II-Positive Human
Corneal Epithelial Cells Activate Allogeneic T Cells
Mitsuhiro Iwata,*^ Atsuhito Yagihashi,%% Melvin I. Roat* Adrian Zeevi,%
Yuichi Iwaki,X and Richard A. ThoftW
Purpose. To achieve a better understanding of the mechanism of corneal immune diseases,
including corneal allograft rejection, the authors examined the potential of human corneal
epithelial (HCE) cells to activate allogeneic T lymphocytes.
Methods. The mixed lymphocyte-HCE cell reaction (MLCER) was performed as follows: HCE
cells from primary cultures, with or without treatment with interferon-7 (IFN-y), were treated
with mitomycin C and then mixed with peripheral blood lymphocytes (PBL) from normal
volunteers. Triplicate cultures were incubated for 7 days. Interleukin-1-or (IL-l-a) was added
to some cultures to examine its effect on MLCER. The lymphocyte responses were measured
by 3H-thymidine uptake for the last 18 hours of incubation in MLCER.
Results. IFN-y-treated, HLA-class-II-bearing HCE cells stimulated allogenic lymphocytes,
whereas IFN-y nontreated, class-II-negative HCE cells did not. The stimulation by IFN-ytreated HCE cells was blocked by anti-HLA class II monoclonal antibody. In addition, exogenous IL-l-a reduced the lymphocyte response in MLCER. This effort was inhibited by indomethacin.
Conclusions. This study demonstrates that HLA-class-II-bearing HCE cells can activate allogeneic PBL by a major histocompatibility complex class II-dependent mechanism. In addition,
HCE cells may regulate immune reactions, probably through prostaglandin production caused
by IL-1. Invest Ophthalmol Vis Sci. 1994; 35:3991-4000.
x lymphocytes recognize antigenic peptides in the
context of major histocompatibility complex (MHC)
gene products.1 The initiation of T-cell responses requires an antigen-presenting cell (APC). The MHC
class II antigens expressed on the surface of APC play
a pivotal role in the primary T-cell response.2 Originally, the MHC class II antigens were thought to be
expressed only on the surface of the cells of the immune system. However, with inflammation such as
that found with autoimmune diseases or allograft reFrom the Departments of * Ophthalmology, The Eye and Ear Institute of Pittsburgh,
Pennsylvania; %Pathology, University of Pittsburgh School of Medicine;
"[Ophthalmology, Nihon University School of Medicine, Tokyo; and §Surgery,
Sapporo Medical University, Sapporo City, japan. II Deceased.
Supported by National Institutes of Health grant EY06I86 (RAT) and by an
unrestricted grant to the Department of Ophthalmology from Research to Prevent
Blindness, Inc., New York, New York.
Submitted for publication February 17, 1993; revised June 7, 1994; accepted June
16, 1994.
Proprietary interest category: N.
Reprint requests: Mitsuhiro Iwala, MD, The Department of Ophthalmology, Nihon
University School of Medicine, 30-1 Oyaguchi Kamimachi, Itabashi-ku, Tokyo 173,
Japan.
jection, class II antigen expression has been detected
on many cell types that do not normally express class
II antigens.3"6 Aberrant class II antigen expression on
the cell surface may be critical to the pathogenesis of
such diseases.3'4'7 Cytokines such as interferon-y (IFNy) can induce class II antigen expression on various
kinds of cells in vitro.8"11 The function of IFN-y-induced, class II-bearing cells to present antigen to T
cells has been studied in such cells as renal tubular
cells,12 vascular endothelial cells,13 liver sinusoidal lining cells,14 epidermal keratinocytes,15'16 and fibroblasts.17'18 According to the results of those studies,
the ability of these cells to function as APCs varies
and depends on cell type origin. For example, IFN-ytreated, class II-bearing fibroblasts are an effective
APC for alloprimed CD4-positive T cells. This is in
contrast to their inability to stimulate freshly isolated
CD4-positive T cells.17 It is suggested that the initiation
of the primary T-cell responses requires another interaction with APC, in addition to the interaction
through antigen peptide-MHC class II complex of
Investigative Ophthalmology & Visual Science, November 1994, Vol. 35, No. 12
Copyright © Association for Research in Vision and Ophthalmology
Downloaded From: http://jov.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933402/ on 05/05/2017
3991
3992
Investigative Ophthalmology 8c Visual Science, November 1994, Vol. 35, No. 12
APC and T-cell receptor-CD3 molecules.1920 Recently, adhesion molecules, such as ICAM-1, ICAM2, or LFA-3 expressed on APC and LFA-1 or LFA-2
expressed on T cells, have been shown to play a role
in the APC-T-cell interaction.21 Adhesion molecules
such as LFA-1 and LFA-2 produce important signals
inside T cells20'22"24 and may be especially important
to the primary T-cell response.
In the cornea, it is thought that Langerhans cells,
which exist in limbal and peripheral cornea, can be
potent APCs. It has been shown that rabbit corneal
endothelium, expressing class II antigen after treatment with IFN-y, could stimulate fresh allogenic lymphocytes.25 On the other hand, human corneal fibroblasts, treated with IFN-y and expressed class II antigen, were not capable of inducing the primary
response of allogenic lymphocytes.26'27 The ability of
human corneal epithelial cells to function as APCs
and to initiate T cell-responses has not been studied.
It has been demonstrated that there is no class II
antigen expression on cultured human corneal epithelial (HCE) cells and that the expression of the human leukocyte antigen (HLA) class II could be induced on HCE cells by IFN-y.28-30
To investigate the potential of human corneal epithelial cells to function as APCs, we examined the
ability of cultured HCE cells to stimulate allogenic
lymphocytes in the primary mixed lymphocyte-HCE
cell reaction (MLCER).
It has been shown that IL-1 is necessary and increases lymphocyte proliferation in the primary mixed
lymphocyte reaction (MLR).31 IL-1 can be produced
by corneal epithelial cells.32 Therefore, we examined
IL-1 production by cultured HCE cells, and we also
examined the effect of the addition of exogenous recombinant human IL-1 on MLCER.
MATERIALS AND METHODS
Human Corneal Epithelial Cell Culture
Human donor eyes were provided by the Medical Eye
Bank of Western Pennsylvania (Pittsburgh, PA).
Primary cultures of HCE cells were prepared as
described.30 Briefly, limbal explants without endothelium were incubated with modified SHEM, which consists of Ham's F12 and Dulbecco's modified Eagle's
medium (1:1), supplemented with mouse epidermal
growth factor (10 ng/ml), bovine insulin (5 //g/ml)
(Collaborative Research, Bedford, MA), cholera toxin
(0.1 //g/ml) (Sigma, St. Louis, MO), L-glutamine (1
//g/ml), dimethylsulfoxide (0.5%), gentamicin (40
fjLg/ml), and 10% fetal calf serum (Irvine Scientific,
Santa Ana, CA) in 35-mm tissue culture dishes (Falcon
3001, Becton Dickinson, Franklin Lakes, NJ). The cul-
tures were incubated at 37°C in 5% CO 2 , and the
medium was changed twice a week. The explants were
removed after confluence was reached.
Induction of the HLA Class II Antigens on
HCE Cells
HCE cell cultures were treated with human recombinant human IFN-y (500 U/ml) (Genzyme, Boston,
MA) for the last 4 days of each incubation period.30
Flow Cytometric Analysis
Primary HCE cell cultures were treated with 0.25%
trypsin and 0.5% ethylenediaminetetraacetic acid
(EDTA, Sigma) and were converted to single-cell suspension. After the cells were settled in 10% fetal calf
serum at room temperature for 1 to 2 hours for cellular recovery, the cells were stained by the following
indirect immunofluorescence technique. The cells
were incubated with anti-HLA-DR monoclonal antibody (Becton Dickinson, Mountain View, CA) or antiHLA class I monoclonal antibody (DAKO, Carpenteria, CA) or control mouse IgG (Organ Tecknika
Cappel, Westchester, PA) and then with fluorescein
isothiocyanate-conjugated goat anti-mouse IgG second antibody (Organ Tecknika Cappel) for 30
minutes at 4°C. The cells were analyzed by flow cytometry using FACStar (Becton Dickinson, San Jose, CA).
Keratin Expression in Primary HCE Cell
Cultures
Primary cultures from limbal explants grown in LabTec chamber slides (Nunc, Naperville, IL) were prepared as described by Kiritoshi et al.33 Also frozen
OCT sections from scraped primary cultures were prepared as described.30 The cells and OCT sections were
examined by immunoperoxidase staining with AE1
and AE3 (gifts of Tung-Tien Sun, PhD), using a previously described procedure.30
Isolation of Lymphocytes
Peripheral blood was collected from normal, healthy
volunteers and then heparinized, and lymphocytes
were isolated by density step gradient centrifugation
with lymphocyte separation medium (LSM, Organ
Teknika, Durham, NC). Peripheral blood lymphocytes
(PBL) were washed and suspended in RPMI 1640
(Gibco, Grand Island, NY) with 10% heat-inactivated
(at 56°C, for 60 minutes) human AB serum (NABI,
Miami, FL).
Mixed Lymphocyte-HCE Cell Reaction
We examined the function of HCE cells from five
different donors to stimulate responder lymphocytes,
which were isolated from two different individuals.
HCE cells from primary cultures, with or without treat-
Downloaded From: http://jov.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933402/ on 05/05/2017
Lymphocyte Activation by Human Cornea! Epithelial Cells
3993
ment with IFN-y, were treated with mitomycin C
(Sigma) 20 Mg/ml for 30 minutes. After treatment
with mitomycin C, HCE cells were treated with 0.25%
trypsin and 0.5% EDTA and converted to cell suspension. HCE cells were resuspended in RPMI 1640 with
10% heat-inactivated human AB serum. Eighty percent to 90% of the cells were viable by trypan-blue
exclusion. HCE cells, as a stimulator, were mixed with
1 X 105 responder PBL in 96-well flat or U-bottomed
microtiter plates (Corning Glass Works, Corning, NY)
to a total volume of 200 fj\. Triplicate cultures were
incubated at 37°C in 5% CO 2 . One //Ci of 3H-thymidine was added to the cultures, and, after 18 hours,
the cultures were harvested on glass fiber filters. 3Hthymidine (NEN, Boston, MA) incorporation was measured by a liquid scintillation counter (Beckman, Fullerton, CA).
ene-diamine was added and incubated for 30 minutes.
The enzyme reactions were stopped with 1 N H2SO4.
Allo Mixed Lymphocyte Reaction
Allo mixed lymphocyte reaction (allo MLR) was established with mitomycin C-treated (20 /itg/ml, for 30
minutes) stimulator allogenic lymphocytes (1 X 105)
and an equal number of responder lymphocytes in 96well microtiter plates. Triplicate cultures were incubated in RPMI 1640 with 10% heat-inactivated human
AB serum at 37°C in 5% CO 2 . The lymphocyte proliferation was assessed in the same fashion as with the
MLCER.
Blocking of MLCER and Allo MLR With
Monoclonal Antibodies
Blocking assays with monoclonal antibodies (rnAb)
were performed as follows: Anti-HLA-DR mAb (clone
L243, Becton Dickinson, Mountain View, CA) (0.5
Mg/ml), anti-HLA class I mAb (clone W6/32, DAKO)
(0.5 //g/ml), or control mouse IgG (Organ Tecknica
Cappel, Westchester, PA) (1.0 /zg/ml) were added to
the wells at the initiation of MLCER and allo MLR
and remained throughout the culture period.
Sandwich Enzyme Immunoassays for Human
Interleukin-1
The enzyme-linked immunosorbent assay for human
interleukin-l-(hIL-l-)a and hIL-1-/? were performed
with hIL-1 ELISA kit (Otsuka Pharmaceutical, Tokushima, Japan) as follows: Mouse anti-hIL-1-a or -(5 mAb
was immobilized on 96-well microplates. Recombinant
hIL-1-o: or -/5, or culture supernatants of HCE cells,
were added to the 96-well plate to a total volume of
200 [A and then incubated overnight at 4°C. After
washing the plates, rabbit anti-IL-1-a or -/3 antiserum
was added and incubated for 2 hours at room temperature. The plates were incubated with anti-rabbit IgGperoxidase conjugate for 2 hours at room temperature. After washing the plates, the substrate o-phenyl-
Effects of the Addition of Recombinant hIL-1
and Indomethacin
Recombinant human interleukin 1 (rIL-1) (Genzyme,
Boston, MA) or indomethacin (INDO, Sigma), or the
combination of rIL-1 and indomethacin, were added
in some cultures of MLCER and allo MLR for the
whole incubation period.
RESULTS
Characterization of Cell Cultures
The cell shape of the primary cultures of HCE cells
consisted of small polygonal cells. All the HCE cells
showed strong reactivity with AE1-AE3 by immunohistochemical staining of the cells cultured in Lab-Tec
chamber slides (Nunc) and frozen sections of scraped
HCE cell cultures (data not shown).
Expression of the HLA Antigens
By flow cytometric analysis, as shown in Figure la,
there was no positive staining with anti-HLA-DR
monoclonal antibody on HCE cells without IFN-y
treatment. This finding supports our previous report.30 Approximately 50% of HCE cells treated with
human recombinant IFN-y 500 U/ml for 4 days expressed HLA-DR antigen after 0.25% trypsin and 0.5%
EDTA treatment to detach the cells from culture
dishes (Fig. lb). HLA class I expression on HCE cells
was detected without IFN-y treatment. IFN-y (500 U/
ml, for 4 days) enhanced HLA class I expression on
HCE cells (Fig. lc). As we previously demonstrated,30
HLA class II induction is nearly maximized by incubation with IFN-y (500 U/ml) for 4 days.
We examined the effect of mitomycin C treatment
on HLA expression by HCE cells. We compared HLA
class I or IFN-y-induced class II antigen expressed on
HCE cells treated with mitomycin C (20 /Ltg/ml) for
30 minutes to the expression on HCE cells without
mitomycin C treatment. We could not find any major
difference between the HLA antigen expression on
HCE cells before and after mitomycin C treatment
(data not shown).
Ability of HCE Cells to Stimulate Allogenic
Lymphocytes in MLCER
After stimulation by HCE cells, nearly maximum lymphocyte responses were obtained at day 7 of incubation (Fig. 2). A high level of response was obtained
when 0.6 X 104 to 2.5 X 104 HCE cells, as stimulators,
were mixed with responder lymphocytes (Table 1).
The response of lymphocytes declined when the num-
Downloaded From: http://jov.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933402/ on 05/05/2017
Investigative Ophthalmology & Visual Science, November 1994, Vol. 35, No. 12
3994
(a) control IgG
L«J2
IB1
IB2
IB3
Fluorescence Intensity
IB'
FIGURE l. Flow cytometric analysis of the
HLA-antigen expression on cultured
HCE cells, treated or not treated with
IFN-y (500 U/ml). (a) HCE cells were
stained with control IgG. (b) HCE cells
were stained with anti-HLA-DR mAb. (c)
HCE cells were stained with anti-HLA
class I mAb. HLA = Human leukocyte
antigen; HCE = human corneal epithelial; IFN = interferon; IgG = immunogobulin G; mAb = monoclonal antibody.
(c) anti-HLA class I MoAb
(b) anti-HLA-DR MoAb
Control medium
Control medium
IFN-gamma
IFN-gamma
IB 1 *
IB'
IB'
Fluorescence Intensity
ber of stimulator HCE cells exceeded more than 2.5
X 104 per well. In most cases, IFN-y-treated HCE cells
stimulated allogenic lymphocytes two to four times as
much as HCE cells without IFN-y treatment, which
induced litde or no allo lymphocyte proliferation (Table 2).
As shown in Table 3, the stimulation by IFN-ytreated HCE cells could be blocked by anti-class II
(HLA-DR) mAb to the level of stimulation of HCE
cells without IFN-y treatment in MLCER (P < 0.005,
Student's Mest), whereas anti-class I mAb did not significantly (P < 0.2, Student's Mest) block stimulation
by IFN-y-treated HCE cells. Ninety percent of the proliferation response of lymphocytes to allogenic PBL
in allo MLR could be blocked by the same concentration of the anti-class II mAb. In donor 4 (see Table
2), HCE cells without IFN-y treatment stimulated lym-
phocytes but much less than did IFN-y-treated HCE
cells. The stimulation by HCE cells without IFN-y
treatment could not be blocked either by anti-class II
or anti-class I mAb (data not shown).
IL-1 Effect on MLCER
As shown in Table 4, with the sandwich enzyme immunoassays for IL-1, we detected IL-l-a in the supernatants of HCE cell cultures, which supports the results
of other studies.34 Lipopolysaccaride added to the culture medium enhanced IL-l-a production by HCE
cells. We did not detect IL-1-/? in the supernatants of
HCE cell cultures, either with or without lipopolysaccaride stimulation.
As shown in Table 5, exogenously added recombinant human IL-l-a (rIL-l-a) enhanced the lymphocyte proliferation in allo MLR. However, the addition
Downloaded From: http://jov.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933402/ on 05/05/2017
Lymphocyte Activation by Human Cornea! Epithelial Cells
7000
8000
5000
4000
ffN-gamma HCE
3000
HCE
2000
1000
Day?
Dav 9
3
FIGURE 2. Time course of H-thymidine incorporation by responder allogeneic lymphocytes (1.0 X 103) cultured with
IFN-y-treated HCE cells (IFN-y HCE) or HCE cells without
IFN-y treatment (HCE) (2.0 X 104). Triplicate cultures were
incubated for 5, 7, and 9 days. Experimental counts-background counts (PBL alone + HCE cells alone) are expressed
as the mean cpm of triplicate determination. IFN = Interferon; HCE = human corneal epithelial; PBL = peripheral
blood lymphocytes; cpm = counts per minute.
of 40 U/ml (400 pg/ml) of rlL-l-a reduced the response significandy in MLCER (Table 6).
We examined the effect of indomethacin, an inhibitor of prostaglandin production. One microgram
per milliliter of indomethacin was added to the well
at the initiation of MLCER. The addition of indomethacin blocked the inhibition of rIL-l-a in MLCER completely (Table 6, Experiment 2).
DISCUSSION
Langerhans cells are thought to be potent antigenpresenting cells35 found in the peripheral and limbal
cornea, which is also the source of the primary HCE
cell cultures. By using immunohistochemical staining,
we and others have found HLA-DR positive, but CD
l(a Langerhans cells marker)-negative Langerhans
cells in fresh normal limbal corneal sections.36 Also,
we examined the cultures from limbal explants after
1 to 4 weeks of incubation with immunohistochemical
staining. We did not detect any HLA-DR-positive or
CDl-positive cells in the cultures. In addition, using
flow cytometric analysis, we did not detect any HLADR-positive cells in the HCE cell cultures. These findings indicate that Langerhans cells neither migrate
with HCE cells from limbal explants nor survive in the
culture. Therefore, we concluded that the lymphocyte
proliferation in MLCER was not the result of contamination by Langerhans cells. In addition, there were
no fibroblasts in our culture, because 100% of the
cultured cells showed strong reactivity with AE1-AE3
mAb. AE1-AE3 mAb recognizes acidic and basic cytokeratins,37 characteristic of epithelial cells and not of
3995
fibroblasts. In some experiments, we used pretreated
HCE cells with 0.02% EDTA for a short time after
removal of the explants and before the addition of
IFN-y, a method described by Sun et al38 to remove
fibroblasts selectively from epithelial cell cultures.
However, we did not find any difference in the ability
of HCE cells pretreated and not pretreated with 0.02%
EDTA to stimulate lymphocytes (data not shown).
We demonstrate that IFN-y-treated, HLA class IIbearing cultured HCE cells stimulated freshly isolated
allogeneic lymphocytes in the primary MLCER,
whereas without IFN-y treatment, class II-negative
HCE cells induced little or no allogeneic lymphocyte
proliferation. Blocking assays with mAb showed that
anti-class II (HLA-DR) mAb blocked the stimulation
by IFN-y-treated HCE cells to the level of non-IFN-ytreated HCE cells, whereas neither anti-class I mAb
nor control mouse IgG blocked the stimulation by
IFN-y-treated HCE cells. These data demonstrate that
IFN-y, in leading to class II antigen expression on
HCE cells, can result in these cells initiating allogeneic
lymphocyte responses by an MHC-dependent mechanism. In addition, these findings suggest that class IIbearing HCE cells can participate in corneal alloimmune responses.
The function of class II-bearing nonlymphoid cells
as APC is variable and depends on cell type. Keratinocytes, renal tubular cells, and fibroblasts from several
kinds of tissue17'27 do not stimulate allogenic lymphocytes in the primary mixed lymphocyte reaction, even
when HLA class II antigen expression has been induced by IFN-y. On the other hand, IFN-y-treated,
class II-bearing vascular endothelial cells are capable
of inducing the primary allogenic lymphocyte re-
1. Stimulation of Allogeneic
Lymphocytes by HCE Cells or
IFN-y-Treated HCE Cells
TABLE
Lymphocyte Proliferation
(mean CPM ± SD)
Stimulator Cells*
Number of
Stimulator Cells
0.6
1.2
2.5
5.0
X
X
X
X
104
104
104
104
HCE
950
1250
660
390
±
±
±
±
320
180
250
80
IFN-y HCE
3780
4400
3380
620
±
±
±
±
1490
550
670
130
Lymphocyte proliferation represented by 3H-thymidine
incorporation by responder allogeneic lymphocytes (1.0 X 105)
cultured with a variable number of HCE cells without IFN-y
treatment (HCE) or IFN-7-treated HCE cells (IFN-7 HCE).
Triplicate cultures were incubated for 7 days.
* Stimulator cells were pretreated with mitomycin C.
CPM = Counts per minute; SD = standard deviation.
Downloaded From: http://jov.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933402/ on 05/05/2017
Investigative Ophthalmology 8c Visual Science, November 1994, Vol. 35, No. 12
3996
2. Ability of HCE Cells to Stimulate Allogeneic Lymphocytes
in MLCER
TABLE
Lymphocyte Proliferation
(mean CPM ± SD)
Stimulator Cells*
Donor
Responder
HCE
A
1260 ± 440
850 ± 150
1070 ± 260
1530 ± 200
1090 ± 210
2350 ± 640
1250 ± 180
<200
<1000
1
2
A
B
A
B
A
A
3
4
5
HCE stimulator only
Responder (A, B) only
IFN-y HCE
4850
3160
2500
4120
2470
5050
4400
±
±
±
±
±
±
±
270
790
130
700
250
530
550
Lymphocyte proliferation represented by sH-thymidine incorporation by responder allogeneic
lymphocytes (1.0 X 105), isorated from indivisuala A or B, cultured with HCE cells without IFN--y
treatment (HCE) or IFN-y-treated HCE cells (IFN--y HCE) (1.2 X 104 to 2.0 X 104), Triplicate
cultures were incubated for 7 days.
* Stimulator cells were pretreated with mitomycin C.
CPM — Counts per minute; SD = standard deviation.
sponses. The reason for these differences has not been
found conclusively. It has been suggested17 that class
II antigen is necessary in the primary T-cell response,
but it may not be sufficient. The initiation of T-cell
activation may require co-signals through other molecules in addition to the T-cell receptor-CD3 complex,39 There is little doubt that adhesion molecules,
such as ICAMs and Leu-CAMs, play a key role in the
TABLE 3. Effect of Anti-Class I or Class II
Monoclonal Antibodies on Lymphocyte
Proliferative Responses in MLCER
Lymhocyte Proliferation {mean CPM ± SD)
Stimulator Cells*
Monoclonal
Ab
None
Control IgG
Anti-class I
Anti-class II
HCE
2140
2890
2810
1740
±
±
±
±
660
300
230
110
IFN-y HCE
7350
7270
6170
1690
±
±
±
±
1030
1120
680f
50t
s
AUoPBL
63230
62240
59890
5910
±
±
±
±
6070
7540
2430
1910
Lymphocyte proliferation represented by H-thymidine
incorporation by responder allogeneic lymphocytes (1.0 X 105)
cultured with HCE cells without IFN-7 treatment (HCE) or IFN-y
treated HCE cells (IFN-y HCE) (2.0 X 104). Allo MLR was
established with stimulator lymphocytes (1.0 X 105) and
responder allogeneic lymphocyte (1.0 X 105). Anti-class II (HLADR) MAb (0.5 Aig/ml) or anti-HLA class I MAb (0.5 /ig/ml), or
mouse IgG control (1.0 ^g/ml) were added at the initiation of
the culture and remained through the incubation period.
Triplicate cultures were incubated for 7 days.
* Stimulator cells were pretreated with mitomycin C.
f P < 0.2, for different from control IgG; Student's Hest.
X P < 0.005, for different from control IgG; Student's r-test.
CPM = Counts per minute; SD = standard deviation.
initiation of T-cell responses. MHC class II antigen
expression is induced or increased at sites of inflammation, such as those resulting from autoimmune disease or allograft rejection. It is still not clear whether
MHC class II antigen expression is the cause or the
result. In rejected corneal buttons, HLA class II antigen (HLA-DR) was detected on corneal cells.40 In corneal allograft rejection, it is possible that IFN-y is produced locally by the interaction of Langerhans cells
in donor tissue with recipient lymphocytes. The subsequent production of IFN-y induces class II antigens
on HCE cells, which may provoke corneal allograft
rejection. It is also possible that IFN-y, secreted in the
anterior chamber or in tears due to other inflammatory events (suture removal, bacterial keratitis, herpes
simplex keratitis, uveitis)41 may affect the corneal epithelium. In addition, similar mechanisms would begin
in corneal endothelial cells,42'43 which may induce
class II antigen expression on corneal epithelial cells.
TABLE 4.
Interleukin 1 (IL-1) Production by
Cultured HCE Cells
Experiment
IL-l-a (pg/ml)
+LPS
+LPS
120
90
190
85
135
IL-l-fi (pg/ml)
ND
ND
ND
ND
ND
(<7.8)
(<7.8)
(<7.8)
(<7.8)
(<7.8)
Interleukin 1 (IL-1) in culture supernatant production by
primary cultures of HCE cells (2 x 105 per 35-mm culture dish)
incubated in 10% PCS with or without LPS (20 fig/ml).
Downloaded From: http://jov.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933402/ on 05/05/2017
Lymphocyte Activation by Human Corneal Epithelial Cells
TABLE 5.
Effect of Exogenous Interleukin 1
(IL-1) on Allo MLR (Allogeneic PBL as
Stimulator Cells)
Addition
Lymphocyte Proliferation
(mean CPM ± SD)
none
rIL-1 (10 U/ml)
rlL-1 (40 V/m\)
22590 ± 3560
30560 ± 4220*
40600 ± 5330f
Lymphocyte proliferation represented by 3H-thymidine
incorporation by responder lymphocytes (1.0 X 10ft) mixed with
mitomycin C treated (20 /ig/ml, 30 minutes) allogeneic
stimulator lymphocytes (1.0 X 105) and incubated with
recombinant human IL-1-alpha (10 U/ml [100 pg/ml] or 40 U/
ml [400 pg/ml]). Triplicate cultures were incubated for 7 days.
* P < 0.02, for different from addition: none, Student's (-test.
f P < 0.01, for different from addition: none, Student's Rest.
CPM = Counts per minute; SD = standard deviation.
In some experiments, HCE cells without IFN-y
treatment stimulated allogeneic lymphocytes, but always less than IFN-y-treated HCE cells did. The stimulation by non-IFN-y-treated HCE cells was observed
when large numbers of stimulator HCE cells (2 X 104/
well) were employed. Regardless of the treatment with
IFN-y, lymphocytes cultured with an excess of 2.5 X
104 stimulator HCE cells did not proliferate in
MLCER. This may be due to either, the excessive consumption of nutrition because of the high metabolic
rate of epithelial cells or the prevention of sufficient
interaction between lymphocytes and HCE cells by the
aggregation of large numbers of HCE cells in the well.
3997
The stimulation of HCE cells without IFN-y treatment could not be blocked by anti-class I or anti-class
II (HLA-DR) mAb. Therefore, it is unlikely that, within
the wells, T cells that responded to donor HLA class
I antigen would produce IFN-y. Similarly, it is unlikely
that, during the incubation period, IFN-y would induce HLA class II antigen expression on class II-negative HCE cells. It is possible that small numbers of
lymphocytes in the Gl phase might respond to IL-1,44
which could be produced by HCE cells.
Interleukin 1 is a powerful inflammatory cytokine.
It can be produced not only by leukocytes but by various cell types of nonhematopoietic lineage.44 IL-1 produces its effect by binding to an IL-1 receptor. Two
different isoforms of IL-1 have been identified, IL-1a and IL-1-/?.45 Both isoforms bind the same receptor,
and their function is nearly the same.46 It has been
shown that IL-1 is necessary and increases lymphocyte
proliferation in the primary MLR by means of induction of IL-2 production and IL-2 receptor expression.44 Large amounts of IL-1 have been detected in
the synovial fluid of patients with rheumatoid arthritis.47 In addition, it has been shown that rheumatoid
ardiritis synovial mononuclear cells and dendritic cells
produce IL-1 activity in vitro.48 IL-1, produced abnormally, may be critical in the progressive destruction
of joints of patients with rheumatoid arthritis. IL-1 can
be produced by HCE cells32 and fibroblasts.49 Therefore, in corneal immune responses, it is possible that
IL-1 may be produced by corneal cells at die inflammatory sites as well as by infiltrating leukocytes.
TABLE 6. Effect of Exogenous Interleukin 1 (IL-1) on MLCER
Lymphocyte Proliferation
(mean CPM ± SD)
Stimulator Cells*
Experiment
Addition
1
None
rIL-1 (10 U/ml)
rIL-1 (40 U/ml)
None
rIL-1 (10 U/ml)
rIL-1 (40 U/ml)
rIL-1 (40 U/ml)
+ INDO (1 //g/ml)
INDO (1 Mg/ml)
HCE
2350
2060
1330
2560
2970
1840
±
±
±
±
±
±
640
310
110
340
280
490
2560 ± 60
2500 ± 50
IFN-y HCE
5050
4750
3140
6840
5540
3950
±
±
±
±
±
±
530
390
520t
1210
790
260f
6340 ± 1510
5590 ± 430
Lymphocyte proliferation represented by 3H-thymidine incorporation by HCE cells without IFN-y
treatment (HCE) or IFN-y treated HCE cells (IFN-y HCE) (2.0 X 10") cultured with responder
allogeneic lymphocytes (1.0 X 10') and incubated with recombinant human IL-l-a (10 U/ml [100
pg/ml] or 40 U/ml [400 pg/ml]) or indomethacin (INDO) (1 ^g/ml), or combination with IL-l-a
(40 U/ml [400 pg/ml]) and INDO (1 /ig/ml). Triplicate cultures were incubated for 7 days.
* Stimulator cells were pretreated with mitomycin C.
t P < 0.02, for different from addition: none, Student's (-test.
CPM = Counts per minute; SD = standard deviation.
Downloaded From: http://jov.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933402/ on 05/05/2017
3998
Investigative Ophthalmology & Visual Science, November 1994, Vol. 35, No. 12
To investigate the effect of IL-1 in the corneal
immune reaction, we examined the effect of exogenous recombinant human IL-l-a (rIL-l-a) in MLCER.
Forty units per milliliter (400 pg/ml) of rIL-l-a significantly enhanced the lymphocyte response in the
allo MLR. Unexpectedly, the addition of rIL-l-a at the
same dose in MLCER reduced lymphocyte proliferation. Indomethacin, an inhibitor of prostaglandin production, completely blocked the inhibitory effect of
rIL-l-a to MLCER. These findings suggest that the
inhibitory effect of rIL-l-a was caused by prostaglandin E synthesis from HCE cells. Indomethacin had
little effect in MLCER without the addition of rIL-l-a
or with a low concentration of exogenous rIL-l-a 10
U/ml (100 pg/ml). These data suggest that the immune response in corneal epithelial cells could be
regulated by an IL-1-rich environment capable of inducing a sufficient amount of prostaglandin E production from corneal epithelial cells to inhibit lymphocyte
proliferation.
For a better understanding of the mechanism of
corneal immune diseases, it is important to elucidate
the immunologic function of the main components
of the cornea, such as the epithelial cells, stromal fibroblasts, and endothelial cells. Langerhans cells have
been thought to be the only potent APC in the cornea,
because only Langerhans cells express the MHC class
II antigen constitutively.50 However, it has been demonstrated51'52 clearly, using animal skin graft models,
that the presence of allogenic Langerhans cells within
a skin graft is not sufficient to induce or cause rejection of the entire graft.
The corneal endothelium can stimulate allogeneic lymphocytes after IFN-y treatment.25 In addition, we demonstrate in this study that the IFN-ytreated, class II-bearing corneal epithelium can induce allogeneic lymphocyte responses. Previous research26 has shown that corneal stromal fibroblasts
are incapable of inducing the primary T-cell response. However, corneal stromal fibroblasts, like
dermal fibroblasts, may be able to produce several
cytokines, which may regulate the function of other
corneal components.
In conclusion, corneal cells as well as Langerhans
cells can participate in corneal immune responses. In
addition, human corneal epithelial cells may regulate
corneal immune reaction by secreting IL-1 and prostaglandins.
Key Words
HLA class II antigens, interferon-7, corneal epithelial cells,
mixed lymphocyte-corneal epithelial cell reaction, interleukin-1
Acknowledgments
The authors thank Tung-Tien Sun, PhD, for providing the
monoclonal antibodies to keratins used in this study. They
also thank the staff at the Department of Immunopathology,
University of Pittsburgh, for technical assistance.
References
1. Roth bard JB, Gefter ML. Interactions between immunogenic peptides and MHC proteins. Ann Rev Immunol. 1991;9:527-565.
2. Rosenthal AS, Shevach EM. Function of macrophages
in antigen recognition by guinea pig T lymphocytes:
I: Requirement for histocompatible macrophages and
lymphocytes./£*/> Med. 1973; 138:1194-1212.
3. Hall BM, Bishop GA, Duggin GG, Horvath JS, Philips
J, Tiller DJ. Increased expression of HLA-DR antigens
on renal tubular cells in renal transplants: Relevance
to the rejection response. Lancet. 1984;2:247-251.
4. Volc-Platzer B, Madjic O, Knapp W, et al. Evidence of
HLA-DR antigen biosynthesis by human keratinocytes
in disease. / Exp Med. 1984; 159:1784-1789.
5. MostJM, Knapp W, Wick G. Class II antigens in Hashimoto thyroiditis: I: Synthesis and expression of HLADR and HLA-DQ by thyroid epithelial cells. Clinical
Immunol Immunopathol. 1986;41:165-174.
6. Chan C-C, Detric B, Nussenblatt RB, Palestine AG,
Fujikawa LS, Hooks JJ. HLA-DR antigen on retinal
pigment epithelial cells from patients with uveitis. Arch
Ophthalmol. 1986; 41:165-174.
7. Jansson R, Karlsson A, Forsum U. Intrathyroidal
HLA-DR expression and T lymphocyte phenotypes
in Grave's thyrotoxicosis, Hasimoto's thyroiditis
and nodular colloid goiter. Clin Exp Immunol.
1984;58:264-272.
8. Basham TY, Nickoloff BJ, Merigan TC, Morhenn VB.
Recombinant gamma interferon induces HLA-DR expression on cultured human keratinocytes. JlnvestDermatol. 1984; 83:88-90.
9. Todd I, Pujol-Borrell R, Hammond LJ, Bottazzo GF,
Feldmann M. Interferon-7 induces HLA-DR expression by thyroid epithelium. Clin Exp Immunol.
1985;61:265-273.
10. Bishop GA, Hall BM, Suranyi MG, Tiller DJ, Horvath
JS, Duggin GG. Expression of HLA antigens on renal
tubular cells in culture: I: Evidence that mixed lymphocyte culture supernatants and gamma interferon
increase both class I and class II HLA antigens. Transplantation. 1986; 42:671-679.
11. LiversidgeJM, Sewell HF, Forrester JV. Human retinal
pigment epithelial cells differentially express MHC
class II (HLA DP, DR, and DQ) antigens in response
to in vitro stimulation with lymphokine or purified
IFN-y. Clin Exp Immunol. 1988; 73:489-494.
12. Bishop GA, Waugh JA, Hall BM. Expression of HLA
antigens on renal tubular cells in culture: II: Effect
of increased HLA antigen expression on tubular cell
stimulation of lymphocyte activation and on their vulnerability to cell-mediated lysis. Transplantation.
1988;46:303-310.
Downloaded From: http://jov.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933402/ on 05/05/2017
Lymphocyte Activation by Human Cornea! Epithelial Cells
13. Markus BH, Colson YL, Fung JJ, Zeevi A, Duquesnoy
RJ. HLA antigen expression on cultured human arterial endothelial cells. Tissue Antigens. 1988; 32:241253.
14. Rubinstein D, Roska AK, Lipsky PE. Liver sinusoidal
lining cells express class II major histocompatiblity
antigens but are poor stimulators of fresh allogeneic
T lymphocytes. JImmunol. 1986; 137:1803-1810.
15. Nickoloff BJ, Basham TY, Merigan TC, Torseth JW,
Morhenn VB. Human keratinocyte-lymphocyte reactions in vitro./Invest Dermatol. 1986;87:11-18.
16. Niederwieser D, Aubock J, Troppmair J, et al. IFNmediated induction of MHC antigen expression on
human keratinocytes and its influence on in vitro alloimmune responses. J Immunol. 1988; 140:2556-2564.
17. Geppert TD, Lipsky PE. Antigen presentation by interferon-y treated endodielial cells and fibroblasts: Differential ability to function as antigen presenting cells
despite comparable la expression. J Immunol.
1985; 135:3750-3762.
18. Maurer DH, Hanke JH, Mickelson E, Rich RR, Pollack
MS. Differential presentation of HLA-DR, DQ, and
DP restriction elements by interferon-y-treated dermalfibroblasts.J Immunol. 1987; 139:715-723.
19. Bretscher P, Cohn M. A Theory of self-nonself discrimination. Science. 1970; 169:1042-1049.
20. van Seventer GA, Shimizu Y, Horgan KJ, Shaw S. The
LFA-1 ligand ICAM-1 provides an important costimulatory signal for T cell receptor-mediated activation of
resting T cells. / Immunol. 1990; 144:4579-4586.
21. Springer TA, Dustin ML, Kishimoto TK, Marlin SD.
The lymphocyte function-associated LFA-1, CD2, and
LFA-3 molecules: Cell adhesion receptors of the immune system. Ann Rev Immunol. 1987;5:223-252.
22. Alcover A, Weiss MJ, Daley JF, Reinherz EL. The Til
glycoprotein is functionally linked to a calcium channel in precursor and mature T-lineage cells. Proc Natl
Acad Sd USA. 1986;83:2614-2618.
23. June CH, Ledbetter JA, Rabinovitch PS, Martin PJ,
Beatty PG, Hansen JA. Distinct patterns of transmembrane calcium flux and intracellular calcium mobilization after differentiation antigen cluster 2 (E rosette
receptor) or 3 (T3) stimulation of human lymphocytes. J Clin Invest. 1986; 77:1224-1232.
24. O'Flynn K, Knott LJ, Russul-Saib M, et al. CD2 and
CD3 antigens mobilize Ca(2 + ) independendy. EurJ
Immunol. 1986; 16:580-584.
25. Donnelly JJ, Li W, Rockey JH, Prendergast RA. Induction of class II (la) alloantigen expression on corneal
endodielium in vivo and in vitro. Invest Ophthalmol Vis
Sd. 1985; 26:575-580.
26. Donnelly JJ, Chan LS, Xi M-S, Rockey JH. Effect of
human corneal fibroblasts on lymphocyte proliferation in vitro. Exp Eye Res. 1988;47:61-70.
27. Young E, Stark WJ. In vitro immunological function
of human corneal fibroblasts. Invest Ophthalmol Vis Sd.
1988;29:1402-1406.
28. Dreizen NG, Whitsett CF, Stulting RD. Modulation
of HLA antigen expression on corneal epidielial and
3999
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
stromal cells. Invest Ophthalmol Vis Sd. 1988; 29:933939.
El-Asrar AMA, van Den Oord JJ, Billiau A, Desmet V,
Emarah MH, Missotten L. Recombinant interferongamma induces HLA-DR expression on human corneal epithelial and endothelial cells in vitro: A preliminary report. Br J Ophthalmol. 1989; 73:587-590.
Iwata M, Kiritoshi A, Roat MI, Yagihashi A, Thoft RA.
Regulation of HLA class II antigen expression on cultured corneal epithelium by interferon-gamma. Invest
Ophthalmol Vis Sd. 1992;33:2714-2721.
Scala G, Oppenheim JJ. Antigen presentation by human monocytes: Evidence for stimulant processing
and requirement for interleukin 1. J Immunol.
1983;131:1160-1166.
Grabner G, Luger TA, Smolin G, Oppenheim JJ. Corneal epidielial cell-derived diymocyte-activating factor
(CETAF). Invest Ophthalmol Vis Sd. 1982;23:757-763.
Kiritoshi A, SundarRaj N, Thoft RA. Differentiation
in cultured limbal epithelium as defined by keratin
expression. Invest Ophthalmol Vis Sd. 1991; 32:30733077.
Sakamoto S, Seino K, Tazawa Y, Inada K, Yoshida M.
Human corneal epithelial cell production of interleukin 1 a (IL-1 a). Folia OphthalmolJpn. 1990;41:505510.
Stingl G, Katz SI, Green I, Shevach EM. The functional
role of Langerhans cells. J Invest Dermatol. 1980; 74:315318.
Seto SK, Gillette TE, Chandler JW. HLA-DR+/T6"
Langerhans cells of die human cornea. Invest Ophthalmol Vis Sd. 2987;28:1719-1722.
Tseng SCG, Jarvinen MJ, Nelson WG, Huang J-W,
Woodcock-Mitchell J, Sun T-T. Correlation of specific
keratins widi different types of epidielial differentiation: Monoclonal antibody studies. Cell. 1982; 30:361372.
Reinwald JG, Green H. Serial cultivation of strains of
human epidermal keratinocytes: The formation of keratinizing colonies from single cells. Cell. 1975;6:331-334.
Inaba K, Steinman R. Resting and sensitized T lymphocytes exhibit distinct stimulatory (antigen-presenting cell) requirements for growth and lymphokine
release. J Exp Med. 1984; 160:1717-1735.
Pepose JS, Gardner KM, Nestor MS, Foos RY, Pettit
TH. Detection of HLA class I and class II antigens
in rejected human corneal allografts. Ophthalmology.
1985;92:1480-1484.
Abi-Hanna D, McCluskey P, Wakefield D. HLA antigens in the iris and aqueous humor gamma interferon
level in anterior uveitis. Invest Ophthalmol Vis Sd.
1989; 30:990-994.
Kusuda M, Gaspari AA, Chann C-C, Gery I, Katzt SI.
Expression of la antigen by ocular tissues of mice
treated widi interferon gamma. Invest Ophthalmol Vis
Sd. 1989; 30:764-768.
Brandt CR, Knupfer PB, Boush GA, Gausas RE, Chandler JW. In vivo induction of la expression in murine
cornea after intravitreal injection of interferon-y. Invest Ophthalmol Vis Sd. 1990; 31:2248-2253.
Downloaded From: http://jov.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933402/ on 05/05/2017
4000
Investigative Ophthalmology & Visual Science, November 1994, Vol. 35, No. 12
44. Oppenheim JJ, Kovacs EJ, Matsushima K, Durum SK.
There is more than one interleukin 1. Immunol Today.
1986; 7:45-46.
45. March CJ, Mosley B, Larsen A, et al. Cloning, sequence and expression of two distinct human interleukin-1 complementary DNAs. Nature. 1985; 315:641647.
46. Dower SK, Kronheim SR, Hopp TP, et al. The cell
surface receptors for interleukin-1 a and interleukin1 0 are identical. Nature. 1986;324:266-268.
47. Fontana A, Hengartner H, Weber E, Fehr K, Grab PJ,
Cohen G. Interleukin 1 activity in the synovial fluid
of patients with rheumatoid arthritis. Rheumatol Int.
1982;2:49-53.
48. Duff GW, Forre O, Waalen K, Dickens E, Nuki G.
49.
50.
51.
52.
Rheumatoid arthritis synovial dendritic cells produce
interleukin 1. BrJRheumatol. 1985;24(suppl l):94-97.
Iribe H, Koga T, Kotani S, Kusumoto S, Shiba T. Stimulating effect of MDP and its adjuvant-active analogues on
guinea pigfibroblastsfor the production of thymocyteactivating factor. J Exp Med. 1983; 157:2190-2195.
Treseler PA, Foulks GN, Sanfilippo F. The expression
of HLA antigens by cells in the human cornea. Am]
Ophthalmol. 1984; 98:763-772.
Rosenberg AS, Katz SI, Singer A. Rejection of skin
allografts by CD4+ T cells is antigen-specific and requires expression of target alloantigen on Ia~ epidermal cells. J Immunol. 1989; 143:2452-2456.
Lerner-Tung MB, Hull BE. The role of la antigen+
epidermal cells in rejection of rat skin equivalent
grafts. Transplantation. 1990;49:1181-1184.
Downloaded From: http://jov.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933402/ on 05/05/2017