Download Subretinal space and vitreous cavity as immunologically

Subretinal Space and Vitreous Cavity as Immunologically
Privileged Sites for Retinal Allografts
Luke Qi Jiang, Marianela Jorquera, and J. Wayne Streilein
Purpose. Because immune rejection is likely to be a major barrier to successful retinal transplantation, it is important to determine whether immune privilege for allogeneic retinal grafts
is a feature of the subretinal space and vitreous cavity.
Methods. Newborn neural retinas of C57BL/6 mice were implanted into the subretinal space,
vitreous cavity, or subconjunctival space of eyes of adult BALB/c (disparate from C57BL/6 at
major and minor histocompatibility loci). At postimplantation day 12, the recipients were
evaluated for donor-specific delayed hypersensitivity and examined clinically and histologically
for evidence of rejection.
Results. Newborn neural retina allografts in the subconjunctival space were destroyed by postimplantation day 12 and these recipients displayed intense donor-specific delayed hypersensitivity. By contrast, grafts in the subretinal space and vitreous cavity at postimplantation day 12
were found to be well differentiated and with no evidence of inflammation; these recipients
failed to display donor-specific delayed hypersensitivity. Moreover, their spleens contained
regulatory T cells that suppressed donor-specific delayed hypersensitivity in naive syngeneic
recipients.
Conclusions. Allogeneic newborn neural retinal grafts implanted in the subretinal space and
vitreous cavity experience immune privilege and induce deviant immune responses resembling
anterior chamber associated immune deviation. Invest Ophthalmol Vis Sci. 1993; 34:33473354.
Urthotopic retinal transplantation holds promise as
a means of restoring vision to eyes blinded by destructive retinal disease. Experimental studies have demonstrated that neural retina or retinal pigment epithelium implanted into the subretinal space can display
function to a limited extent, and can survive for a variable period of time.1 Clinical use of retinal grafts, however, must be associated with full function and longterm graft survival. To achieve these objectives, researchers must overcome two major obstacles. The
aggregate limits of mammalian neural regeneration,
plasticity, and maintenance constitute the first obstacle. These limits must be stretched if retinal grafts are
to be fully functional. Even if grafts are able to survive,
immune rejection represents the second obstacle, and,
it is this issue that is addressed here.
From the Department of Microbiology and Immunology, University of Miami School
of Medicine, Miami, Florida.
This ivork was supported by USPHS Grant EY 09595 and a grant from the
National Relinilis Pigmentosa Foundation.
Submitted for publication March 12, 1993; accepted April 30, 1993.
Property interest category: N.
Reprint requests: Luke Qi Jiang, Schepens Eye Research Institute, Harvard School
of Medicine, 20 Stamford Street, Boston, MA 02114.
Two factors appear to influence the potential of
immune rejection of retinal grafts placed in the eye:
the inherent immunogenicity of the retinal grafts, and
the anatomic site of implantation. Although histocompatibility antigens (known to be potent inducers of systemic immunity) are normally expressed sparsely in
the neural retina and other neural tissue, the expression of such molecules is upregulated after transplantation.2 We have demonstrated in mice that neural retinal grafts are immunogenic, and that the immunity
induced by neural retinal cells is directed at both
transplantation antigens and retinal restricted autoantigens.3 However, the type of immunity elicited depends on the sites at which the retina is implanted.
Implantation of neural retinal grafts into the subconjunctival space (SCon) has been found to induce antigen-specific delayed hypersensitivity (DH), whereas
implantation of similar retinal grafts into the anterior
chamber (AC) induce a deviant form of immunity (anterior chamber associated immune deviation, ACAID).45
In this deviant response, antigen-specific suppressor T
cells are generated and DH reactivity is selectively impaired. Thus, grafts placed in the AC are protected by
ACAID, that is, they enjoy immune privilege.
Investigative Ophthalmology & Visual Science, November 1993, Vol. 34, No. 12
Copyright © Association for Research in Vision and Ophthalmology
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933171/ on 05/06/2017
3347
3348
Investigative Ophthalmology & Visual Science, November 1993, Vol. 34, No. 12
From a therapeutic point of view, the subretinal
space (SR) and vitreous cavity (VC)—both of which
are adjacent to the retina—are the most likely potential sites for therapeutic retinal grafts. Therefore, determining whether these posterior compartments of
the eye are also immunologically privileged sites is essential.
In this study, we implanted allogeneic newborn
neural retinal (NNR) tissue into the eyes of adult mice,
either into the SR or VC. We subsequently conducted
clinical, histologic, and immunologic examinations to
determine the characteristics of the systemic immune
responses evoked by these implants.
MATERIALS AND METHODS
Animals
Donor retinal tissue was obtained from newborn
C57BL/6 mice (aged < 24 hours). Adult male BALB/c
mice (aged 7 to 12 weeks) served as recipients. For
retinal transplantation into the VC, we repeated two
experiments switching the donor and recipient strains.
All experimental mice were obtained from our breeding colony at the University of Miami School of Medicine, Miami, Florida. Mice were maintained in a common room of the vivarium where an overhead fluorescent light provided 12-hour cycles of light and dark.
Inoculations, clinical examinations, and enucleations
of grafted eyes were performed under anesthesia induced by intramuscular injections of ketamine (Ketalar, Parke Davis, Shawnee, KS) 0.075 mg/g body
weight, and xylazine (Rompun, Haver-Lockhart,
Morris Plains, NJ), 0.006 mg/g body weight. All experimental procedures conformed to the ARVO Resolution on the Use of Animals in Research.
Preparation of Donor Neural Retina
Donor C57BL/6 newborn mice were decapitated.
Their eyes were immediately enucleated and placed in
ice-cold calcium-magnesium-free Hanks balanced salt
solution (HBSS). Each eye was cut open along the edge
of the cornea using microsurgical scissors. After the
lens and iris were removed, the neural retina was
gently separated from the eyecup and transferred to a
new petri dish containing cold calcium-magnesiumfree HBSS. After two changes to fresh medium, the
entire neural retina was halved and each half was used
for transplantation.
Intraocular or Subconjunctival Implantation of
Newborn Neural Retina Tissue
Recipient mice received general anesthesia. For implantation of retina tissue into the SR or VC, the eyelids were kept open and the eyeballs were held steady
with forceps. A 0.3-mm penetrating wound was made
at the posterior portion of the wall of the eye using a
microsurgical knife with a 15° angle (Edward Week
and Company, Inc., Research Triangle Park, NC). For
retinal implantations into the SCon, a penetrating
wound was made in the fornix portion of the conjunctiva. The retinal tissue was then drawn into a glass needle made from a glass bore of a 1O-jul micropipetter
(diameter of 200 /mi). The retinal tissue, along with
about 1 /il of HBSS, was slowly injected via the wound
into the SR, VC, or SCon of the eye.
Clinical Examination
The mice were anesthetized and their eyes examined
with a dissecting microscope on designated days. The
pupils were dilated with 0.5% Mydriacyl (Alcon Inc.,
Humacao, PR) and 2.5% phenylephrine hydrochloride and the fundus was visualized through a contact
lens so the posterior segments of the eyes could be
viewed.
Histologic Examination
Sectioning Procedure. The graft was localized by clinical examinations and sections were cut through the
portion of the eye containing the graft. For hematoxylin and eosin staining, the eyes were fixed with 10%
buffered formalin embedded with paraffin and cut 5
/xm thick. For immunohistochemical staining, the eyes
were fixed with 4% paraformaldehyde in 0.1 M phosphate-buffer, pH 7.2. After fixation for 24 hours, the
eyes were immersed in 30% phosphate-buffered sucrose and embedded with Tissue-Tek II O.C.T. compound (Lab-Tek Products, Elkhart, IN). Cryostat sections of 7 ixm were cut transversely, mounted on gelatin-coated slides, and stored at 4°C.
Immunocytochemical Staining for S-Antigen.
Cryostat sections were washed in phosphate-buffered
saline (PBS) pH 7.2 for 1 minute for immunofluorescent staining, then incubated with primary antibody
for 30 minutes. Guinea pig anti-S antigen antiserum
(donated by Dr. Carolyn Kalsow, University of Rochester, Rochester, NY), which cross reacts with mouse
S-antigen, was diluted with PBS at 1:10 and used as the
primary antibody. For negative controls, sections were
incubated with PBS only. The sections were washed in
PBS and incubated for 30 minutes with the secondary
antibody, fluorescein-conjugated F(ab')2 fragments of
goat anti-guinea pig immunoglobulin G (Rockland,
Gilbertsville, PA) diluted with PBS at 1:50. After three
washes in PBS, sections were mounted with polyvinyl
alcohol in PBS and glycerol, and examined under an
Olympus research microscope (Olympus Corp., Lake
Success, NY) using epifluorescence and a Blue Filter
block (excitation light wavelength 380 to 490 nm).
Photographs
Photographs of the fundus were taken with a 35-mm
camera system attached to an Olympus dissection mi-
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933171/ on 05/06/2017
3349
Immune Privilege and Retinal Allografts
croscope (using Kodak Plus-X pan 125 black-andwhite print film; Eastman Kodak Co., Rochester, NY).
Micrographs of sections stained with hematoxylin and
eosin technique and immunocytochemical technique
were taken with a 35-mm camera system attached to an
Olympus research microscope (using Kodak Plus-X
pan 125 black-and-white print film). For immunofluorescent staining, micrographs were taken with a 35mm camera system attached to an Olympus research
microscope with epifluorescence and a Blue Filter
block (excitation light wavelength 380 to 490 nm). Kodak Tri-X pan 400 black-and-white printfilmwas used.
DH Response to Retinal Grafts
DH was measured based as ear swelling, as previously
described.6 Briefly, half of the neural retina of newborn donor mice was implanted into the SR of recipient mice. For positive controls, mice received similar
retinal grafts into the SCon. Normal BALB/c mice
without transplants of neural retina served as negative
controls. On day 12 after transplantation, both ear
pinnae of each mouse were challenged and measured
for DH responses. A Mitutoyo engineer's micrometer
(Mitutoyo Corp., Tokyo, Japan) was used to measure
the thickness of both ears immediately before challenge. For challenge, 2 X 106 (BALB/c X C57BL/6)
Fl spleen cells (irradiated) were suspended in 10 ^1 of
HBSS and injected into the subcutaneous tissue of the
left ear pinnae. The right ear served as an untreated
control. The difference in measured ear thickness
after 24 hours was used as a measurement of DH intensity. Results were expressed as specific ear swelling
= (24 hour measurement — 0 hour measurement) experimental ear — (24 hour measurement — 0 hour
measurement) negative control ear X 10~3 mm. A twotailed Student's t test was performed on the data
presented and significance was assumed to exist if
P<0.05.
Adoptive Transfer of Capacity to Induce
Suppression of Alloantigen-Specific DH
Spleens from BALB/c mice bearing C57BL/6 retinal
grafts in the SR were collected aseptically. Single cell
suspensions were prepared by pressing whole spleens
through stainless steel screens (60-mesh) as described
elsewhere.6 Spleen cells were then washed and resuspended in HBSS. Each naive BALB/c recipient received 5 X 107 spleen cells in 100 v\ of HBSS via the
tail vein. For positive control, a similar number of
naive BALB/c spleen cells was infused intravenously
into naive BALB/c recipients. Negative controls received no infusion. Within 2 hours, all experimental
mice received implants in SCon of neural retina grafts
from immature C57BL/6 mice (aged 8 to 14 days).
Twelve days later, DH reactivity, as described earlier,
was assayed by ear challenge.
RESULTS
Clinical Appearance and Course of Retinal
Graft Implanted into the Subretinal
Space or the Vitreous Cavity
Entire neural retinas isolated from the eyes of newborn C57BL/6 mice were halved and each half was
implanted into the SR or the VC of adult BALB/c
mice. As the medium containing the retinal grafts was
injected into the SR, detachment of the retina could
be observed as a white reflection through the dilated
pupil. On examination of the fundus with a contact
lens, the detached retina with a retinal implant underneath appeared as a white drapelike projection,
smooth surfaced with a well-defined edge. Within 3
days, the projection gradually flattened (possibly because of reabsorption of the medium), and the retinal
graft became recognizable through the translucent ret-
Graft-
Retina] Vessel
-V
B
FIGURE 1. (A) Fundus appearance of the eye of adult BALB/
c mouse 12 days after receiving allogeneic C57BL/6 newborn neural retina graft into SR. The graft (arrow) appears
white with no sharp limit. The arrowhead indicates the vessels of the host retina, which is superior to the graft. (B) A
schematic diagram showing the relationship between the
vessel and the SR graft.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933171/ on 05/06/2017
Investigative Ophthalmology & Visual Science, November 1993, Vol. 34, No. 12
3350
ina. Subretinal grafts had a white color with an irregular shape and no clearly denned edge. Since the portion of the host retina that covered the graft was virtually detached, a climbing pattern of vessels was often
observed at the edge of the graft (Fig. 1, A and B).
Retinal folds usually surrounded the grafts. No hemorrhage or inflammatory exudate was in the AC, VC, SR
or the retina itself during the time course of these
experiments.
When retinal tissues were implanted into the VC,
the implant could be observed directly under the microscope. Through the dilated pupil, the white reflection of the retinal graft appeared immediately behind
the lens. Within a few days, the cloudlike graft tissue
gradually aggregated and formed a white mass with an
irregular shape and a clearly defined edge. The graft
typically attached to the retinal surface superior to the
retinal vessels (Fig. 2, A and B). Because no neovessels
FIGURE 3. (A) Histologic appearance of an allogeneic
C57BL/6 newborn neural retinal graft implanted into SR of
the eye of an adult BALB/c mouse. The graft (G) consists
mainly of photoreceptor cells, which form rosettes (Rs), and
it is adjacent to the host retina (H). (B) Inimunofluorescent
staining for S antigen reveals a majority of cells in the SR
graft to be positive. Photoreceptor cells (Ph) of the host retina (H) are positive with S antigen.
Graft
Retinal Vessel—
B
2. (A) Fundus appearance of the eye of an adult
C57BL/6 mouse 12 days after receiving allogeneic BALB/c
newborn neural retina graft into VC. The graft (arrow) has a
white translucent appearance with the edge clearly denned.
The graft is attached to the surface of the retina and is superior to the retinal vessels (arrowhead). (B) A schematic diagram showing the relationship between the vessels and the
VC graft.
FIGURE
developed on the VC surface of the graft, the relationship of the graft to the retinal vessels (i.e., beneath or
superior; Figs. IB, 2B) became an important clinical
sign for differentiating VC grafts from SR grafts, particularly the flat-type graft. No inflammatory exudate
or hemorrhage was detected in media of the eye, retinal grafts, or host retina. On deep illumination, grafts
appeared translucent and they retained this appearance for the remainder of the experiment.
Histologic Examination
SR Retinal Crafts. The eyes of five animals that received retinal grafts into the SR 12 days earlier were
enucleated and processed for hematoxylin and eosin
staining and for inimunofluorescent staining for S an-
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933171/ on 05/06/2017
3351
Immune Privilege and Retinal Allografts
tigen. Examination of the hematoxylin and eosin sections revealed that the grafts were well developed and
differentiated, and located between the host neural
retina and retinal pigment epithelium. Retinal grafts
contained a dominant population of photoreceptor
cells that formed rosettes (Fig. 3A). On immunofluorescent staining, the photoreceptor cells associated
with rosettes showed S antigen-positive staining (Fig.
3B). The rosettes, which resembled the outer nuclear
layer of the retina, were surrounded by cells of the
inner nuclear layer. Ganglion cells were not usually
recognizable, probably because they were scant and
did not form a discrete layer. No inflammatory infiltration was detected in either the retinal graft or in the
host retina; however, a few macrophages occasionally
appeared around the graft margins.
VC Retinal Grafts. Retinal grafts were implanted
into the VC offiveanimals and their eyes were enucleated 12 days later and processed for histologic examination in a similar fashion to that described earlier for
SR grafts. Examination of both hematoxylin and eosin
and immunofluorescent sections revealed results resembling those found in the SR grafts (Fig. 4).
By clinical and histologic criteria, our evidence indicates that grafts implanted into either SR or VC survive and proceed to differentiate toward the morphology associated with the mature retina. Moreover, this
process occurs without evidence of immunologic rejection. In contrast, retinal grafts implanted into the
SCon soon lost their identity, and histologic examination revealed that the SCon implants contained an obvious inflammatory infiltrate, but little unidentifiable
retinal structures (results published previously).7
24 Hours After Challenge
80 _
0
24 Hours After Challenge
FIGURE 5. (A) SR graft; (B) VC graft. Capacity of allogeneic
retinal graft to induce DH in recipient mice. Allogeneic
NNR was implanted into SR or VC of recipient mice on day
0; ears were challenged on day 12 and ear swelling was assessed 24 hours later. Positive control mice received allogeneic NNR in SCon. Negative controls received ear challenge
only. Bar represents mean ±SEM. Responses of the SCon
group are significantly higher than those of the SR, VC, or
negative control groups. (P < 0.01)
Aspects of Systemic Immunity Induced by
Allogeneic Neural Retinal Grafts Implanted
into the SR and VC
FIGURE 4. Histologic appearance of an allogeneic C57BL/6
newborn neural retinal graft implanted into the VC of an
adult BALB/C mouse eye 12 days previously. The graft (G)
contains mostly photoreceptor cells which form rosettes
(Rs). Host retina (H).
Our previous studies indicated that allogeneic NNR
implanted into the AC induced a deviant form of immunity (ACATD) in recipient mice, and that ACAID
played a role in protecting these grafts from rejection.7 Because the clinical and histologic studies of SR
or VC retinal grafts yielded findings similar to those of
AC grafts, we next examined the respective fates of
SR, VC, and SCon retinal grafts and their impact on
systemic immunity.
Impaired DH Induced by Allogeneic SR or VC
NNR Grafts. Allogeneic C57BL/6 NNR were implanted into the SR and VC of adult BALB/c mice.
Five mice were used as recipients for each experimental group and each recipient received one half a donor
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933171/ on 05/06/2017
3352
Investigative Ophthalmology 8c Visual Science, November 1993, Vol. 34, No. 12
retina. Mice receiving similar NNR grafts into the
SCon served as positive controls, while negative control mice received a sham operation or no treatment at
all. Twelve days later, the ears of all mice received subcutaneous injections of irradiated (1500 R) suspended
spleen cells (106 cells/10 /x\) isolated from adult
(BALB/c X C57BL/6)F1 mice. Ear swelling was measured by micrometer 24 and 48 hours later. Figure 5
shows the results of a representative experiment. Vigorous DH reactivity directed at the alloantigens of
C57BL/6 donors was generated in BALB/c recipients
of C57BL/6 NNR grafts in SCon, whereas recipients
of allogeneic NNR grafts in the SR and VC displayed
no significant alloantigen-specific DH reactivity.
These results indicate that allogeneic NNR grafts implanted into intraocular compartments (SR or VC) as
opposed to an extraocular site (SCon) evoked different forms of systemic immunity. In contrast to conventional DH reactivity, which was induced by SCon NNR
grafts, impaired DH was generated in recipients of SR
or VC NNR grafts. The experiment was repeated twice
with similar results. In addition, two additional experiments were performed in which the donor and recipient strains were switched. The pattern of immune responses was unchanged in these experiments.
Generation of Suppressor Spleen Cells in Recipients of SR or VC Grafts. We demonstrated previously that failure to develop conventional DH in recipients of allogeneic NNR grafts in the AC is associated
with the generation of antigen-specific suppressor
splenic cells, a key feature in expression of ACAID.4'5
To determine whether C57BL/6-specific suppressor
splenic cells were induced in the BALB/c recipients by
grafts in the SR or VC, we conducted adoptive transfer
assays. Spleens were removed from five adult BALB/c
mice that received C57BL/6 NNR grafts into the SR
or VC 12 days previously. As positive controls, spleens
were harvested from five naive adult BALB/c mice.
Two groups of spleens were minced into single cell
suspensions and used for intravenous infusion into
naive BALB/c mice (5 per group) at a dose of 50 X 106
per mouse. Within 2 hours, each recipient received an
NNR graft from C57BL/6 mice in the SCon space.
Twelve days later, the ears of all mice were challenged
by injection of irradiated spleen cells of (BALB/c
X C57BL/6)F1 mice. Ear swelling was measured 24
and 48 hours after challenge. Figure 6 gives the results
of representative experiments. Naive BALB/c mice
that received spleen cells from donors bearing
C57BL/6 NNR grafts in the SR or VC failed to acquire
alloantigen-specific DH reactivity. However, BALB/c
mice of the control group, which received intravenous
infusions of naive spleen cells, developed a strong alloantigen-specific DH after receiving a C57BL/6
NNR graft in the SCon space. Thus, the DH impair-
•Control
Test
-Control
24 Hours After Challenge
80
n
+Control
Test
0
-Control
24 Hours After Challenge
FIGURE 6. (A) SR graft; (B) VC graft. Adoptive transfer of
impaired DH reactivity induced by allogeneic NNR in recipient mice. For test groups, naive mice received spleen cells
(50 X 106) intravenously from mice bearing SR or VC allogeneic NNR grafts. As positive control, a similar number of
normal spleen cells were infused intravenously into naive
syngeneic recipients. One hour later recipients received implants of allogeneic NNR grafts in SCon. Twelve days later
their ears were challenged; ear swelling responses were measured 24 hours later. Negative controls were as described in
Figure 5. Responses of positive control group are significantly greater than those of either test group or negative
control groups (p < 0.01).
ment displayed in BALB/c mice bearing C57BL/6
NNR grafts in SR or VC was associated with generation of alloantigen-specific suppressor splenic T cells,
consistent with the existence of ACAID.
DISCUSSION
During the past two decades, understanding of immune privilege in the eye and central nervous system
has advanced greatly. It has been learned that immune
privilege is a dynamic physiological process achieved
through active downregulation of systemic cell-me-
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933171/ on 05/06/2017
Immune Privilege and Retinal Allografts
diated immunity.45 For example, after antigens such
as allogeneic tumor cells,8 soluble antigens,9 or neural
retinal cells4 are injected into the AC of the eye, recipients display a deviant form of antigen- specific systemic immunity (e.g., AC AID). This phenomenon has
also been described in response to antigens injected
into the VC.8'10 However, the nature of immune privilege in the SR, an important compartment of the eye
related to retinal transplantation and retinal disease,
has not been determined previously. Our data indicate
that allogeneic NNR grafts implanted into the SR and
VC induce an antigen-specific deficit in DH associated
with suppressor splenic T cells, characteristic features
of ACAID. Moreover, NNR grafts implanted in the SR
or VC developed and differentiated into recognizable
retinal structures within 12 days of engraftment. In
contrast, NNR grafts in the SCon induced conventional DH immunity and were rejected by postimplantation day 12. These results imply that immune privilege is extended to NNR grafts implanted into the SR
and VC, and that ACAID plays a protective role in the
success of these retinal grafts.
Immune privilege in intraocular compartments
can be considered to be the consequence of an interplay between the immune system and the intraocular
microenvironment. It is known that under physiological conditions, the AC possesses an immunosuppressive microenvironment.11 This microenvironment, in
turn, plays a regulatory role in modulating systemic
immunity directed at antigens introduced into the
AC.12 On one hand, the suppressive microenvironment sustained in the AC inhibits the local expression
of preexisting systemic immunity. On the other hand,
the intraocular microenvironment participates in
modifying the primary immune responses to ocular
antigen.
The AC constitutively contains suppressive molecules such as transforming growth factor /3, and other
as-yet-unknown cytokines.7 Although it is not clear
which immunosuppressive molecules are represented
in the SR or VC, this study implies that both the SR
and VC also contain immunosuppressive microenvironments that are sustained for at least 12 days after
implantation of NNR grafts. Because both the SR and
VC are adjacent to the retina, it would be reasonable
to suggest that retinal cells may significantly contribute to the immune suppressive microenvironment. In
support of this idea are the following: production of
transforming growth factor /5 by astrocytes is upregulated under pathological conditions,12 retinal pigment
epithelium is an intraocular producer of transforming
growth factor /3,13 and Muller's cells (glia) of the retina
suppress T cell proliferation by a direct contact mechanism.14 In addition, the endothelium of retinal vessels,
Bruch's membrane, and the pigment epithelium to-
3353
gether form the so-called ocular-blood barrier. It is
interesting that the SR, although only a potential
space, has features of an immunologically privileged
site even after retinal detachment and the disruption
of the retinal-blood barrier by implantation. It would
appear that the SR's immune privilege is able to withstand passive disruption of the ocular-blood barrier,
implying that local factors may actively downregulate
systemic immunity.
Because neural retinal grafts express both transplantation antigens and retinal autoantigens,3 this
study's results do not exclude the possibility that NNR
grafts in the SR may induce ACAID directed at retinarestricted antigens. We have previously reported that
both allogeneic and syngeneic NNR grafts implanted
into the AC or VC induced ACAID directed at retinarestricted autoantigens.10 It would be interesting to
determine whether retinal autoantigen-specific ACAID
can be induced by syngeneic or allogeneic NNR grafts,
or by soluble retinal autoantigens (such as S antigen
and interphotoreceptor retinoid binding protein) injected into the SR. Such information will be vital in
determining the potential mechanisms involved in retinal graft rejection. It will also provide insight into the
immunopathogenesis of autoimmune retinitis.
Retinal autoantigens, which are expressed on retinal grafts, can evoke conventional DH reactivity when
the grafts were implanted in the immunologically nonprivileged SCon space. Such DH reactivity presents a
potential risk both to retinal grafts and to the host
retina.3 Therefore, knowing whether the SR is an immunologically privileged site for both alloantigens or
autoantigens, is essential for predicting the fate of retinal grafts in this space.
The SR is surrounded by retinal pigment epithelium and photoreceptor cells, both of which constitutively express immunogenic autoantigens.1516 It therefore is reasonable to speculate that these autoantigens
may be released into the SR under pathological (perhaps even physiological) conditions and that ACAID
may constantly exist under such conditions. Thus, the
immunologic features of the normal SR (e.g., immune
suppressive or nonsuppressive) may play a critical role
in preventing autoimmune retinitis.
Both the SR and VC are immunologically privileged sites for NNR grafts, implying that the posterior
compartments are favorable sites for retinal grafts.
However, the immune privilege induced by NNR
grafts in the SR or VC may not be absolute or permanent. It must be noted that allogeneic P815 tumor
cells implanted into the AC of C57BL/6 mice induced
only transient ACAID.17 We have preliminary results
that reveal that allogeneic retinal grafts placed in the
AC are eventually destroyed,18 suggesting that conventional immunity can emerge and overcome ACAID. A
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933171/ on 05/06/2017
3354
Investigative Ophthalmology & Visual Science, November 1993, Vol. 34, No. 12
study is underway to determine whether long-standing
retinal grafts in the SR and VC are similarly affected.
Key Words
immune privilege, ACAID, allogeneic retinal graft, subretinal space, vitreous cavity
10.
Acknowledgments
The authors thank Ms. Debra Bunch Ghosh for her editorial
assistance and Ms. Barbara French for photographic assistance.
11.
References
12.
1. LaVail MM, Li L, Turner JE, Yasumura, D. Retinal
pigment epithelial cell transplantation in RCS rats:
normal metabolism in rescued photoreceptors. Exp
Eye Res. 1992;55:555-562.
2. Rao K, Lund RD, Kunz HW, Gill TJ III. The role of
MHC and non-MHC antigens in the rejection ofintracerebral allogeneic neural grafts. Transplantation.
1989;48:1018-1021.
3. Jiang LQ, Streilein JW. Immune responses elicited by
transplantation and tissue-restricted antigens expressed on retinal tissues implanted subconjunctivally. Transplantation. 1991;529:513-513.
4. Streilein JW. Anterior chamber associated immune
deviation: the privilege of immunity in the eye. Surv
Ophthalmol. 1990; 35:67-73.
5. Niederkorn JY. Immune privilege and immune regulation in the eye. Adv Immunol. 1990;48:191-226.
6. Streilein JW, Niederkorn JY, Shadduck JA. Systemic
immune unresponsiveness induced in adult mice by
anterior chamber presentation of minor histocompatibility antigens. / Exp Med. 1980; 152:1121-1125.
7. Streilein JW, Wilbanks GA, Cousins SW. Immunoregulatory mechanisms of the eye. J Neuroimmunol. 1992;
39:185-200.
8. Jiang LQ, Streilein JW. Immune privilege extended to
allogenic tumor cells in the vitreous cavity. Invest Ophthalmol VisSci. 1991;32:224-228.
9. Hara Y, Caspi RR, Wiggert B, Chan C-C, Wilbanks
13.
14.
15.
16.
17.
18.
GA, Streilein JW. Suppression of experimental autoimmune uveitis in mice by induction of anterior
chamber associated immune deviation with interphotoreceptor retinoid binding protein. J Immunol.
1992;148:1685-1692.
Jiang LQ, Streilein JW. Immunity and immune privilege elicited by autoantigens expressed on syngeneic
neonatal neural retinal grafts. Curr Eye Res. 1992;
11:697-709.
Cousins SW, McCabe MM, Danielpour D, Streilein
JW. Identification of transforming growth factor-beta
as an immunosuppressive factor in aqueous humor.
Invest Ophthalmol VisSci. 1991;32:2201-2211.
Wahl SM, Allen JB, McCartney-Francies N, et al. Macrophage- and astrocyte-derived transforming growth
factor j8 as a mediator of central nervous system dysfunction in acquired immune deficiency syndrome. J
Exp Med. 1991; 173:981-991.
Tanihara H, Yoshida M, Matsumoto M, Yoshimura N.
Identification of transforming growth factor-/? expressed in cultured human retinal pigment epithelial
cells. Invest Ophthalmol Vis Sci. 1993; 34:413-419.
Caspi RR, Roberge FG, Nussenblatt RB. Organ-resident, non-lymphoid cells suppress proliferation of autoimmune T lymphocytes. Science. 1987; 237:10291032.
Donoso LA, Marryman CF, Edelberg KE, Naids R,
Kalsow C. S-antigen in the developing retina and pineal gland. A monoclonal antibody study. Invest Ophthalmol VisSci. 1985;26:561-567.
Broekhuyse RM, Kuhlmann ED, Winkens HJ. Experimental autoimmune anterior uveitis (EAAU). II.
Dose-dependent induction and adoptive transfer using a melanin-bound antigen of the retinal pigment
epithelium. Exp Eye Res. 1992;55:401-411.
Ksander BR, Bando Y, Acevedo J, Streilein JW. The
infiltration and accumulation of precursor cytotoxic T
cells increases with time in progressively growing ocular tumors. Cancer Res. 1991;51:3153-3158.
Jiang LQ, Streilein JW. Survival of intraocular neural
retina grafts is influenced by microenvironmental factors as well as immunogenetic disparity. Soc Neurosci
Abs. 1991;17:1137.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933171/ on 05/06/2017