Immune Privilege Is Extended, then Withdrawn, from Allogeneic Tumor Cell Grafts Placed in the Subretinal Space

Abstract

purpose. To determine whether the subretinal space can extend immune privilege to allogeneic tumor cell grafts that do not possess their own inherent immune privilege.

methods. P815 tumor cells were injected into the anterior chamber (AC), the subretinal (SR) space, or subconjunctivally in eyes of BALB/c (allogeneic), SCID (immune incompetent), normal DBA/2 (syngeneic), or DBA/2 mice presensitized with P815 cells transfected with interleukin-12 and B7.1. Tumor growth was observed clinically and histologically for up to 50 days. BALB/c recipients were tested for suppression of DBA/2-specific delayed hypersensitivity and concomitant immunity. The SR space of tumor-containing eyes was assessed for its capacity to support ovalbumin (OVA)-specific anterior chamber associated immune deviation (ACAID).

results. P815 cells injected into the SR space of presensitized and normal DBA/2 and SCID mice grew progressively, resulting eventually in recipient death. Tumor cells injected into the SR space of eyes of BALB/c mice grew progressively until day 14, followed by tumor regression resulting in phthisis bulbi (14/35) or tumor elimination (19/35) with preserved ocular anatomy by day 35. Despite elimination of tumors from the SR space, BALB/c recipients exhibited DBA/2-specific ACAID and concomitant immunity. In addition, OVA injected into the SR space of eyes from which tumor has been eliminated induced ACAID.

conclusions. Various parameters of immune privilege, originally described for the AC, are characteristic of immune privilege within the SR space. However, because P815 cells placed in the AC prove lethal for BALB/c recipients, but P815 cells placed in the SR space resolve without jeopardizing the host’s life, immune privilege in the SR space can be distinguished from immune privilege in the AC, and this may have implications for grafts of retinal tissue placed within the SR space.

The anterior chamber (AC) of the murine eye has been studied extensively for its property as an immune privileged site. ^{1 2 3 4 5 6 7} The experiments of Niederkorn et al. ⁶ first documented that minor histoincompatible tumor cells injected into the AC formed progressively growing tumors that never experienced immune rejection. Moreover, these allogeneic tumors induced a deviant systemic immune response, termed anterior chamber associated immune deviation (ACAID), which has proved to be a characteristic feature of systemic immunity induced by a wide variety of antigens injected intracamerally. ^{2 4} Because unrelenting tumor growth coincided temporally with sustained ACAID in recipients of AC tumor injections, the concept emerged that deviant immunity and immune privilege are mechanistically linked. This concept was further strengthened by the observations that major histocompatibility complex (MHC) incompatible tumor cell grafts injected into the AC experienced immune privilege only transiently, ⁸ and, correspondingly, the ACAID induced by MHC incompatible tumors also proved to be evanescent. ⁸ Immune privilege is also a feature of the vitreous cavity. ⁹ The initial pattern of growth of P815 cells injected in this compartment of eyes of BALB/c mice has been found to be essentially similar to that observed for P815 tumors placed within the AC, and P815 tumors growing within the vitreous cavity readily induced ACAID. 

The subretinal (SR) space is a third intraocular compartment with the potential to be an immune privileged site. ^{10 11 12} Our laboratory has had a long-standing interest in promoting the survival of retina-derived grafts in the SR space; to aid in this goal we have begun to examine the extent to which this intraocular compartment displays immune privilege. Several years ago Jiang et al. ^{11 13} reported that allografts of retinal pigment epithelium (RPE) as well as of neonatal neuronal retina experienced immune privilege within the SR space. Specifically, both RPE and neonatal neuronal retina allografts survived for prolonged intervals when injected into the SR space, compared with the subconjunctival space. However, both RPE and neuronal retina grafts may themselves be immune privileged tissues and, thus, capable on their own of resisting immune rejection. Therefore, we cannot be sure that their prolonged acceptance within the SR space is a property of the space or of the grafts themselves. To assess directly the immune privileged status of the SR space as a site, it is necessary to select allografts for SR space implantation that do not display the inherent property of immune privilege. P815 tumor cells represent such a graft. 

Recently, we have reported that a single injection of P815 cells into the SR space of eyes of BALB/c mice induces donor-specific ACAID. ¹² Because ACAID induction is only one manifestation of immune privilege, we elected to describe the growth pattern of P815 tumors implanted in the SR space of eyes of allogeneic recipients (BALB/c), syngeneic recipients (DBA/2), and immune incompetent recipients (SCID mice) as independent means to assess immune privilege within the SR space. Our results revealed that tumors grew progressively in the SR space of eyes of all recipients, confirming that the SR space is an immune privileged site. However, in allogeneic recipients, intraocular tumor masses began to resolve between 2 and 3 weeks post–tumor implantation, and, eventually, almost all mice were cured of their tumors. Surprisingly, P815-specific ACAID persisted even when tumors were eliminated. Additionally, the SR space of eyes that recovered from tumor growth regained the capacity to support ACAID to an unrelated antigen, implying that immune privilege persisted at the site. The possible mechanisms that might be responsible for graft rejection in a site that retains its immune privilege are discussed. 

Materials and Methods

Animals

Adult female BALB/c, DBA/2, and CB.17 (SCID) mice, 6 to 8 weeks of age, were obtained from the animal facilities at the Schepens Eye Research Institute or from Jackson Laboratories (Bar Harbor, ME). Mice were maintained in a common room of a vivarium. Inoculations, injections, clinical examinations, and enucleations were conducted under anesthesia induced by intraperitoneal injection of ketamine (Ketalar; Parke–Davis, Paramus, NJ), 0.075 mg/g body weight, and xylazine (Rompun; Pheonix Pharmaceutical, St. Joseph, MO), 0.006 mg/g body weight. All experimental procedures conformed to the ARVO Statement on the Use of Animals in Ophthalmic and Vision Research. Five mice were used for each experimental group, and experiments were repeated at least twice. 

Intraocular Injections

P815 mastocytoma cells (DBA/2 origin) were grown as previously described. ⁶ For experiments, 2 × 10⁵ P815 cells were slowly injected into different ocular sites in a volume of 2 μl using a glass micropipette connected to a 10-μl syringe. For AC, and subconjunctival injections, a 0.3-mm penetrating wound was made in the peripheral portion of the cornea, 1 mm posterior to the limbus, or in the fornix of the conjunctiva, respectively, using a 30-gauge needle. For injections into the SR space, the temporal conjunctiva was opened parallel to the limbus and the eye was rotated to expose the posterior part of the sclera. A 0.3-mm tangential sclerotomy was made with a 30-gauge needle, and a retinal bleb was created with 0.5 μl Hanks’ balanced salt solution (HBSS), and then tumor cells or antigen-containing solutions were injected through the wound into the SR space. 

Soluble Antigen

Fifty micrograms of ovalbumin (OVA; Sigma Chemicals, St. Louis, MO) in HBSS was used as soluble antigen and was injected in a volume of 2 μl into the eyes of experimental animals. 

Clinical Evaluation of Tumor Growth of P815 Cells

The anterior segment of eyes after tumor cell inoculation was examined clinically with a Topcon slit-lamp (Tokyo Optical, Tokyo, Japan), and the posterior segment was examined through a coverslip with a dissecting microscope. All eyes were examined every second day after injection of tumor cells. Subretinal tumor growth was graded using the following categories: 0, no tumor detectable; 1, minimal tumor (less than 2 quadrants); 2, medium tumor (less than 3 quadrants); 3, SR tumor in more than 3 quadrants; 4, SR space filled with tumor and infiltration of vitreous body; 5, eye filled with tumor (extension into AC); 6, same as 5 plus protrusio bulbi and extraocular tumor extension; and 7, death of animal with large orbital tumor. 

Histopathologic Evaluation of Tumor Growth of P815 Cells

For histologic evaluation, tumor-bearing and control eyes of each group were enucleated at different time points (3, 7, and 14, 21, 45 days), fixed with 10% buffered formalin, processed for routine histopathology, and stained with hematoxylin and eosin. 

Sensitization of DBA/2 Mice to P815 Tumor Antigens

Immunization of mice with P815 cells transfected with B7.1 and interleukin (IL)-12 allows recipients to reject wild-type P815 cells injected into a conventional site. ¹⁴ For this purpose P815 tumor cells had been previously transfected with vectors containing genes encoding the co-stimulatory molecules B7.1 and IL-12. ¹⁴ The transfected cells were used to sensitize groups of DBA/2 mice to the tumor-specific antigens of P815 before intraocular implantation of P815 cells. For sensitization, transfected P815 cells (1 × 10⁶) were injected into the flank of normal DBA/2 mice 14 days before experimental study. 

Assay for Delayed Hypersensitivity

Mice to be assayed for P815-specific delayed hypersensitivity (DH) received intracameral, subretinal, or subconjunctival injections of P815 cells at various times before assay. At day 45 after initial injection of P815 cells, 5 × 10⁵ x-irradiated (2000 R) P815 tumor cells in 10 μl were injected into the left ear pinnae of mice with intact ocular morphology. For mice to be assayed for OVA-specific DH, OVA (100 μg) combined with complete Freund’s adjuvant (CFA) was injected subcutaneously within 7 days of intraocular injection of OVA. Ten days later, 200 μg OVA in 10 μl was injected into the left ear pinnae of the mice. The right ear served as untreated control. Both ear pinnae were measured immediately before injection and 24 hours later with an engineer’s micrometer (Mitutoyo, Tokyo, Japan). The measurements were performed as triplicates. Results were expressed as specific ear swelling = (24-hour measurement − 0-hour measurement) experimental ear − (24 hour measurement − 0-hour measurement) negative control ear × 10^–3 mm. A two-tailed Student’s t-test was used and significance assumed if P < 0.05. 

Results

Pattern of Growth and Survival of P815 Cells in SR Space of BALB/c Eyes

Panels of normal BALB/c mice received injections of P815 cells (2× 10⁵/site) into the SR space, the AC, or subconjunctivally. The growth of these tumors was examined by direct inspection, by slit-lamp, by indirect ophthalmoscopy, and, eventually, by histology of enucleated tumor-bearing eyes. As described previously, ¹² tumors grew progressively within the AC and the SR space (Fig. 1A , Table 1 ) through day 14. By day 14 post-implantation, tumor cells had broken through the sclera and penetrated into the orbit (Fig. 1B) . In contrast, P815 cells injected subconjunctivally formed transiently growing masses that promptly receded and were no longer evident at 14 days (Table 1) . Beyond 14 days, the patterns of tumor growth in the AC and SR space began to differ. Whereas the AC tumors simply continued to grow progressively beyond day 21 post-implantation, SR space tumors began to recede when examined at this time point (Table 1) . Eyes with P815 cells in the SR space then progressed toward one of two outcomes: phthisis bulbi was observed by day 30 in 19 of 35 eyes; complete tumor elimination with preserved ocular morphology was observed in 14 of 35 eyes (Table 1) . Two eyes showed slowly progressing tumor growth that was still evident when the experiment was terminated on day 60 post-implantation. These findings indicate that immune privilege was extended equally well to allogeneic tumor cells injected into the AC and SR space for the first 2 weeks after implantation. Thereafter, privilege was apparently lost in the SR space, and as a consequence the tumors (with or without the eye) were destroyed. 

Pattern of Growth and Survival of P815 Cells in SR Space of Eyes of Syngeneic and Immune Incompetent Mice

One explanation for the different patterns of survival of P815 cells in the SR space and AC of eyes of BALB/c mice may be that unrelenting tumor cell growth is not possible in the SR space because of space and nutritional constraints, rather than because of loss of immune privilege. To examine this possibility, we injected 2 × 10⁵ P815 cells into the SR space of eyes of CB.17 (SCID) mice, which are profoundly immune incompetent, and in the eyes of DBA/2 mice, which are syngeneic with the tumor and therefore only exhibit weak immunity that is directed against tumor-specific antigens. For comparison’s sake, P815 cells were also implanted in the AC of a separate panel of normal DBA/2 mice. As revealed in Figure 2 , progressively growing tumors formed in the SR space of both groups of recipients, and all mice eventually died of the tumor. In SCID mice, tumor growth was extremely rapid and mice died between 8 and 12 days post-implantation, presumably due to systemic dissemination of the tumor. In DBA/2 mice, the rate of tumor growth in the AC and the SR space was similar, although tumor growth in the SR space tended to lag behind. Eventually, however, the same outcome was reached. Their intraocular tumors penetrated through the ocular capsule, forming large orbital tumors. These mice died of direct extension of tumor cells into the brain, a mode of death similar to that of mice with P815 cells in the AC. The findings that SCID mice died from SR space P815 cells within 12 days, whereas DBA/2 mice survived significantly longer, indicate that adaptive immunity acts to suppress intraocular tumor growth, even though that immunity may be weak and directed at tumor-specific antigens. Progressive tumor growth in the eyes of DBA/2 mice without evidence of elimination argues against the existence of either a microanatomic or a nutritional barrier to progressive tumor growth in the SR space. Thus, the elimination of tumors from the SR space of BALB/c eyes must reflect allograft rejection and, presumably, the loss of immune privilege at the site. 

Pattern of Growth and Survival of P815 Cells in SR Space of Eyes of Presensitized Syngeneic Mice

Because P815 cells injected into the flank of DBA/2 mice (a non–immune privileged site) form progressively growing tumors, progressive growth after injection into the AC or SR space may not be a reflection of immune privilege. Recently, Chen et al. ¹⁴ have reported that DBA/2 mice can be sensitized to tumor antigens by receiving a subcutaneous injection of P815 cells that had been transfected with B7.1 and IL-12. DBA/2 mice primed with B7.1 and IL-12–transfected P815 cells acquire the ability to reject wild-type P815 cells injected into the flank. ¹⁴ However, Chen et al. also reported that when wild-type P815 cells were injected into the AC of similarly primed DBA/2 mice, the tumors grew progressively, indicating that immune privilege is extended to weakly antigenic tumor placed in this intraocular site, even in the presence of preexistent immunity. We next examined whether the SR space more closely resembled a privileged site (AC) or a nonprivileged site (subconjunctival space) by using as recipients DBA/2 mice that had been immunized with B7.1 and IL-12–transfected P815 cells. In contrast to cells injected into the subconjunctival space, P815 cells were able to create progressively growing tumors in the SR space of presensitized DBA/2 recipients, and the majority of the mice (8/10) eventually died (Fig. 2) . This result confirms that the normal SR space shares with the AC the capacity to extend immune privilege to weakly antigenic tumors even when recipients have been specifically presensitized to tumor-specific antigens. The growth of P815 tumor cells in the AC or the SR space tended to be slower in presensitized DBA/2 mice when compared with naïve DBA/2 mice (Fig. 2) , indicating that persisting immunity directed against tumor antigens has a measurable effect on tumor growth in the eye. Eventually, however, this immunity fails and immune privilege is able to promote unlimited tumor growth. 

Systemic DH Response to P815 Cells in SR Space of Eyes of BALB/c Mice

We have recently reported that P815 cells placed in the SR space of eyes of BALB/c mice induce donor-specific ACAID that is detectable within 14 days. ¹² We wished to determine whether suppressed donor-specific DH (i.e., ACAID) persisted in mice whose intraocular tumors had undergone regression. To examine this point, panels of BALB/c mice received P815 cells in the SR space or subconjunctivally (positive controls). After 14, 22, or 47 days, panels of these mice received an intrapinnae injection of x-irradiated P815 cells (5 × 10⁵). Ear swelling responses as evidence of DH were evaluated 24 and 48 hours later. As the results displayed in Figure 3 reveal, impaired DH was observed among recipients of P815 cells in the SR space, whether the mice were tested at 14, 22, or 47 days. As before, tumors were eliminated in virtually all these eyes by 30 days, indicating that loss of intraocular tumor did not correlate with acquisition of donor-specific DH. Moreover, DH was equally impaired in mice with phthisis bulbi and mice with anatomically preserved eyes from which tumor had been eliminated. The finding that donor-specific DH did not emerge in mice that eliminated their intraocular tumors from the SR space means that tumor rejection was not procured by T cells that mediate donor-specific DH. 

Capacity of SR Space of Eyes with Resolved Tumors to Support ACAID Induction

The results presented above, documenting persistence of donor-specific ACAID in mice whose intraocular tumors had disappeared, do not directly demonstrate that these eyes had recovered the property of immune privilege. To assess whether SR spaces from which P815 tumors had been eliminated still possessed the property of immune privilege, we tested whether an irrelevant antigen could evoke a deviant systemic immune response if injected into the SR space of eyes from which tumors had spontaneously resolved, and which displayed intact ocular morphology. Eyes that had undergone phthisis bulbi after tumor elimination were not used in these experiments. Ovalbumin (50μ g) was injected into the SR space of recovered eyes at day 45 after P815 cell injection. By that time, SR tumors had been completely eliminated, and SR space had recovered with scar formation in some mice. Seven days later, these mice received a subcutaneous immunization with OVA and CFA. Ten days later, the ear pinnae of these mice were challenged with OVA (200 μg), and ear swelling was evaluated 24 and 48 hours thereafter. The results of a representative experiment (of three) are presented in Figure 4 . Ovalbumin-specific DH was not detected in mice that received OVA into recovered SR spaces. By contrast, mice immunized with OVA/CFA without prior intraocular injection of OVA displayed intense DH. Thus, the SR space of eyes from which tumor had been eliminated displayed, or perhaps regained, the capacity to support ACAID induction. 

Capacity of P815 Cells in SR Space to Induce Concomitant Immunity

BALB/c mice that contain progressively growing P815 tumors in one eye acquire concomitant immunity. That is, subsequent injection of P815 cells into the flank, or even into the contralateral eye, fails to result in tumor formation. ¹⁵ Although the precise mediators of concomitant immunity have not been elucidated, DH does not contribute to protection against tumors. Concomitant immunity is not only able to prevent the establishment of new tumors, but it appears to be responsible for the inability of the original tumor in the eye to seed systemic metastases. ¹⁶ We next wished to determine whether P815 cells injected into the SR space induced concomitant immunity similar to that induced by P815 cells injected into the AC of the eye. Accordingly, panels of BALB/c mice received P815 cells into the SR space. Seven days later, a second inoculum of P815 cells (2 × 10⁵) was placed into the AC of the right eye. The pattern of tumor growth within that eye was observed and recorded daily. Within a few days of the AC tumor cell injection, a haze appeared in the AC, but no tumor mass was detected. Thereafter, the haze in the AC cleared, but no visible tumor ever emerged (data not shown). This result indicates that P815 cells placed in the SR space induce concomitant immunity similar to tumors originally placed in the AC. 

Discussion

Our results provide several new pieces of evidence to support the view that the SR space is an immune privileged site. First, P815 tumor cells, which possess no inherent immune privilege, formed progressively growing tumors when placed in the SR space of eyes of allogeneic BALB/c mice. Second, P815 cells also formed progressively growing tumors when placed in the SR space of eyes of syngeneic DBA/2 mice presensitized to the weak tumor-specific antigens of this tumor cell line. Third, mice bearing P815 tumors in the SR space displayed concomitant immunity (i.e., they resisted tumor formation when P815 cells were injected into the AC of the uninjected fellow eye). Fourth, P815 cells in the SR space of BALB/c mice induced persistent DBA/2-specific ACAID, a confirmation of our recent reported finding. We conclude that the SR space is indeed an immune privileged site. 

However, the tumors that formed intraocularly after SR space injection of P815 cells into BALB/c mice did not persist and grow indefinitely. Instead, the tumors, which had filled the globe and infiltrated through the choroid and sclera, began to recede 14 to 21 days after implantation. In some cases, the tumors simply disappeared, leaving the eye anatomically intact. In the other cases, tumor-containing eyes underwent phthisis. In fact, no mice died from tumors that were implanted in the SR space. This outcome contrasts sharply with that of BALB/c mice in which P815 tumor cells were implanted in the AC; all these mice die eventually of cerebral extensions of the tumors. We considered two possible explanations for the distinctive pattern of tumor growth and elimination in the SR space of eyes of BALB/c mice. First, the SR space is incapable of supporting sustained tumor growth. Second, immune privilege in the SR space may not be permanent, as it usually is in the AC. 

We doubt that tumors are eliminated from the SR space because this site cannot sustain progressive tumor growth. This doubt is based on our finding that P815 cells readily formed tumors that continued to grow in the SR space of eyes of SCID mice and DBA/2 mice until their recipients died. In DBA/2 mice, the tempo of tumor growth in the SR space was virtually indistinguishable from that in the AC. Thus, the SR space is capable of supporting tumor growth that is progressive and eventually fatal. It is of interest that P815 cells were far more aggressive in SCID mice than in DBA/2 mice. Although SCID mice are defenseless against this tumor, T cells of DBA/2 mice are able to recognize weak tumor-specific antigens on P815 cells. We presume that the extended survival of DBA/2 mice bearing intraocular tumors, compared with SCID mice, reflects the expression of this antitumor immune response. However, this level of tumor immunity, which retarded intraocular tumor growth, failed ultimately to protect against the tumor’s lethality. The fact that tumor growth within DBA/2 eyes was less aggressive than in SCID eyes indicates that tumor immunity must be able to express itself in the SR space, even though this site is immune privileged. 

The ability of immunity to express itself in the SR space supports the second possible explanation for the elimination of tumor cells from eyes of BALB/c mice (i.e., that immune privilege for P815 cells in the SR space is transient). However, several other experimental observations cast doubt on this explanation. First, injection of OVA into the SR space of BALB/c eyes from which tumors had spontaneously resolved induced ACAID. The recent literature indicates that the existence of immune privilege in the AC is strongly correlated with the capacity of the AC to support ACAID induction. ¹⁷ Assuming that a similar correlation exists for the SR space, the ability of the recovered SR space to support OVA-specific ACAID implies that immune privilege is also present. Second, BALB/c mice from whose eyes SR tumors spontaneously resolved failed to acquire DBA/2-specific DH. In previous reports, elimination of P815 tumors from eyes of mice correlated with acquisition of donor-specific DH. In particular, P815 cells established tumors in the AC of MHC-incompatible C57BL/6 and A/J mice; and during the first 14 days while the tumors were growing progressively, the recipients displayed DBA/2-specific ACAID. However, when the ocular tumors were rejected during the subsequent 7 to 10 days, the recipients suddenly acquired DBA/2-specific DH. ⁸ Moreover, circumstantial evidence implicates T cells that mediate DH in ocular tumor rejection. ¹⁸ Thus, when grafts within immune privileged sites are eliminated because of loss of privilege, the systemic immune response shifts from a deviant one to a conventional one. No such shift was detected when P815 tumors were eliminated from the SR space of BALB/c eyes. For these reasons, we cannot be sure that the reason for tumor elimination from the SR space of BALB/c eyes is a change in the immune privileged status of the site. 

If the SR space is physically capable of supporting progressive unrelenting P815 tumor cell growth (as our evidence revealed in syngeneic DBA/2 mice), and if resolution of the tumor in eyes of BALB/c mice does not result from loss of immune privilege in the SR space, how then can the elimination of these intraocular tumors be explained? Unique anatomic features of the SR space might contribute to the tumor elimination, which is not seen after AC injection of tumor cells. Moreover, the trauma necessary to insert cells into the SR space is considerably greater than that within the AC. In the former, the choroid must be penetrated, and the retinal detachment created by the procedure may further alter the properties of the space. However, even after this surgical trauma, the SR space displays the capacity to support the induction of immune deviation and to grant immune privilege, at least for a temporary period. The difference in tumor growth begins to be apparent when the tumor penetrates and destroys the normal anatomy of the original injection site. Tumors originally placed in the AC break out and continue their unrelenting growth. However, when SR tumors break out of the space, further growth stops and the tumor begins to recede. 

A third possibility is suggested by recent experiments reported by Ksander et al. ¹⁹ These investigators have discovered that P815 cells growing in the immune privileged AC acquire (through time) unique properties that enable the eye-derived tumor cells to grow progressively when implanted anywhere, even at nonprivileged sites. The evidence suggests that tumor cells growing within the AC may assume the status of an “immune privileged tissue.” If this interpretation of the data of Ksander et al. is correct, then we must consider the possibility that mechanisms responsible for immune privilege in the AC and SR space are similar but not identical. Particularly, the SR space may differ from the AC by not possessing the ability to confer“ immune privilege” on tumor cells placed within. Experiments are currently under way to test the hypothesis that immune privilege in the SR space lacks the capacity to confer “immune privilege” on P815 cells placed within this site. 

Whatever the mechanism, our evidence indicates that immune privilege is first extended, then withdrawn, from allogeneic tumor grafts placed in the SR space. This information is likely to be relevant to the fate of allogeneic retinal tissues grafted into the SR space. As Jiang et al. reported previously, ¹¹ allogeneic neonatal retina grafts experience immune privilege when implanted in the SR space of eyes of normal mice. The recipients even develop donor-specific ACAID within 2 weeks of implantation. However, these neuronal retina allografts eventually fail in the AC as well as in the SR space, and their failure is somehow linked to the development of destructive anti-donor immunity. ²⁰ The results of Jiang et al. strongly resemble our experience with allogeneic tumor grafts in the SR space in that both types of grafts are ultimately destroyed (i.e., immune privilege is withdrawn). Working out the pathogenesis of this withdrawal may provide insights into how to extend indefinitely the survival of allogeneic retinal tissue placed in the SR space. 

 Supported by Deutsche Forschungsgemeinschaft Grant WE 1462/3-1/2; by United States Public Health Service Grant EY 09595; and by the Minda DeGunzburg Research Center for Retinal Transplantation.

 Submitted for publication December 29, 1998; revised April 16 and July 12, 1999; accepted July 14, 1999.

 Commercial relationships policy: N.

 

Figure 1.

 

Photomicrographs of sections of eyes of BALB/c mice after injection of P815 cells into the SR space. Eyes had been enucleated, fixed in 10% buffered formalin, and embedded in paraffin. After cutting 6-micrometer sections, tissue had been stained with hematoxylin-eosin. (A) Day 3 after P815 cell injection: P815 tumor cells (T) in the SR space directly underneath the retina (R). S, sclera. (B) Day 14 after P815 cell injection: Tumor cells grow underneath the retina (R), invade the choroid (C), and penetrate the sclera (S) to grow extraocularly (E).

Table 1.

 

Cell Tumor Growth in the SR Space of BALB/c Mice

		Tu. Increase	Tu. Decrease	Phthisis	No Tu.
	14 days	34/35	—	1/35	—
SR space	21 days	4/35	21/35	10/35	—
	30 days	2/35	14/35	19/35	—
	14 days	25/25	—	—	—
AC	21 days	25/25	—	—	—
	30 days	24/25	—	1/25	—
	14 days	—	—	—	15/15
Subconjunctival	21 days	—	—	—	15/15
	30 days	—	—	—	15/15

 Observation of tumor growth after injection of P815 cells into the SR space, the AC, or subconjunctivally in BALB/c mice. Tumor growth was evaluated at days 14, 21, and 30 after P815 cell injection. Tu., tumor.

Figure 2.

 

P815 tumor cell growth in SCID mice and in naïve or primed DBA/2 mice. Grading of tumor growth after injection of P815 cells subconjunctivally (Scon), or into the anterior chamber (AC) or subretinal space (SR) of naïve or primed (with P815 cells expressing IL-12 and B7.1) syngeneic DBA/2 mice or into the SR of SCID mice lacking functional T or B cells.

Figure 3.

 

P815 specific DH measurement at different time points after injection of P815 cells into the SR space of BALB/c mice. Ear swelling analysis 24 hours after ear challenge with P815 cells in animals with inoculation of P815 cells into the SR space. P815 inoculation had been performed 14 days SR(14d), 22 days SR(22d), or 47 days SR(47d) before measurement. As positive control, animals received P815 cells subconjunctivally (Scon). Naive animals with ear challenge only served as negative control (negative). * indicates mean response significantly lower than positive control, P < 0.05.

Figure 4.

 

Ovalbumin-specific DH measurement after OVA injection into the SR space after elimination of P815 cell tumors. Ear swelling analysis was performed 24 hours after ear challenge with OVA in animals with subcutaneous immunization with OVA and CFA 7 days earlier. One week before immunization, OVA had been injected into the SR space SR(naïve). One group of animals had received an injection of P815 cells in the SR space and had rejected the SR tumor before OVA injection SR(tumor). Negative control animals (negative) received no immunization. Positive control mice were only immunized (positive). * indicates mean response significantly lower than positive control, P < 0.05.