| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Published online before print June 3, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Schepens Eye Research Institute, Harvard Medical School, Boston, Massachusetts
Correspondence: J. Wayne Streilein, M.D., Schepens Eye Research Institute, 20 Staniford St., Boston, MA 02114. E-mail: waynes{at}vision.eri.harvard.edu
 |
ABSTRACT |
|---|
T cells create an antigen-presentation environment that leads to CD4+ and CD8+ 
/ß T cells, which as regulators, suppress induction and expression of T helper cell type 1 (Th1) and Th2 immune expression systems. The ways the eye influences local and systemic immune responses to ocular antigens and pathogens carry risks to and benefits for mammalian organisms. As loss of sight is a powerful, negative-selecting force, the benefits of ocular immune privilege outweigh the risks.
Key Words: anterior chamber • regulatory T cells • immune deviation • transforming growth factor-ß • thrombospondin
|
INTRODUCTION |
|---|
Immune privileged sites are defined as places within the body where foreign tissue grafts experience extended (often indefinite) survival, whereas similar grafts placed in conventional sites are promptly rejected. Immune privileged tissues differ from conventional tissues in that grafts prepared from the former experience extended (often indefinite) survival when placed at conventional sites, whereas nonprivileged tissue grafts placed in conventional sites are promptly rejected.
The last 30 years of research have also given rise to a consensus view of the meaning of ocular immune privilege. The rationale is as follows: All organs and tissues, even those with special physiologic needs and those unable to regenerate themselves, require immune protection against pathogens. As immune protection against pathogens can damage vital tissues in an innocent bystander manner, immune privilege is regarded as an evolutionary adaptation that enables local protection to be provided by immune effectors that do not disrupt specialized tissue functions or cause the loss of tissue incapable of regeneration. Immune privilege is achieved by dynamic interactions between the immune system and specialized tissues. In the case of the eye, where a precise microanatomy and clear media must be maintained for light images to fall accurately on the retina, immune privilege allows for immune protection of the eye in a manner that is largely devoid of immunogenic inflammation.
|
ANTERIOR CHAMBER OF THE EYE: FEATURES OF AN IMMUNE PRIVILEGED SITE |
|---|
|
MANIFESTATIONS OF OCULAR IMMUNE PRIVILEGE |
|---|
Intraocular microenvironment is selectively anti-inflammatory and immunosuppressive
AqH, when harvested from normal eyes and tested in vitro, has been shown to suppress activation of a wide variety of cells and molecules involved in inflammation and destructive immunity [11
, 12
]: CD4+ T cells (as revealed by inhibition of proliferation and secretion of effector cytokines), polymorphonuclear neutrophilic leukocytes (inhibiting release and destructive potential of the cellular contents), macrophages activated by bacterial lipopolysaccharide and interferon- (IFN-; reduced production of reactive oxygen intermediates and nitric oxide), natural killer (NK) cells (reduced capacity to lyse appropriate target cells), and complement components C1q and C3 NK cells. Some of these same inhibitory activities are found within the vitreous gel. It should be pointed out that AqH is not a global inhibitor of all inflammation and immunity, as fully functional cytotoxic T cells retain their capacity to lyse their specific targets in the presence of AqH, and neutralizing, noncomplement-fixing antibodies retain their properties when AqH is present.
The virtually continuous layer of pigmented epithelium that lines the posterior surface of the iris, which lies immediately beneath the secretory epithelium of the ciliary body and supports the photoreceptor cell layer of the retina, represents a kind of immunologic barrier for the eye. In part, this is because tight junctions unite retinal pigment epithelium (RPE) into an impermeable shield that prevents blood-borne cells and molecules from indiscriminately entering the subretinal space from the choriocapillaris. A similar set of tight junctions unites iris phycoerythrin (PE). Despite this physical barrier, leukocytes—especially activated T lymphocytes—have been demonstrated to be able to penetrate through pigment epithelial layers, and this appears to be an important pathway by which immunopathogenic T cells enter ocular compartments and cause inflammatory diseases. Over the past 15 years, experimental evidence has accumulated to indicate that T cells, which encounter ocular pigment epithelia, are altered by this experience [13 14 15 ]. RPE and pigmented epithelial cells cultured from iris and ciliary body have all been found to inhibit T cell activation in vitro. In the case of iris PE, inhibition has been found to be dependent on intimate cell contact between PE and T cells, whereas PE from ciliary body and retina secretes soluble factors that suppress T cells. Identification of the cell-surface molecules responsible for iris PE-induced suppression is now under way. RPE are known to secrete transforming growth factor-ß (TGF-ß), thrombospondin (TSP), prostaglandin E2, and probably other immunomodulatory molecules. One way to consider the ocular pigment epithelium as an immune barrier relates to the fact that any T cells that enter the eye must pass through this layer. Inevitably, and as a means of avoiding sight-limiting inflammation, invading T cells must be altered by this encounter. Consequently, most, if not all, of the immunopathogenic potential of T cells that approach the eye through the pigment epithelial layer are neutralized. Thus, it may well be that the "immune barrier" created by ocular pigment epithelia makes an essential contribution to ocular immune privilege.
Corneal endothelium provides a somewhat impenetrable barrier between the anterior chamber and the corneal stroma, and corneal endothelial cells have been reported to secrete molecules that suppress lymphocyte activation and to suppress inflammation [16 , 17 ]. In addition, corneal endothelial cells express cell molecules on their apical surface, which inhibit complement activation (CD46, CD55, CD59) [18 ] and promote apoptosis (CD95 ligand) among CD95+ cells that encounter them [19 ]. Circumstantial evidence links CD95L expression on corneal endothelium with enhanced survival of orthotopic corneal allografts in mice [20 , 21 ], presumably as the donor-specific, alloreactive T cells, which could mediate graft rejection, express CD95 and are thus deleted via programmed cell death.
Thus, the ability of the intraocular microenvironment to be anti-inflammatory and immunosuppressive resides among unique molecular features of the AqH and vitreous gel and the cell-surface properties of ocular parenchymal cells (pigment epithelia, corneal endothelia) that surround intraocular compartments and form a type of immune barrier to blood-borne molecules and cells that threaten the integrity of the visual axis.
Foreign tissue grafts survive for prolonged, often indefinite, intervals without immune rejection
Vast, clinical experience with penetrating keratoplasty collected over more than 50 years indicates that the eye extends immune privilege to corneal grafts [22
]. Unlike any other type of solid organ transplant, corneal allografts in humans display a very high rate of acceptance, although immunosuppression is only applied topically and in modest dosage. This salutary outcome is reserved, however, for cornea grafts placed in so-called "low-risk" eyes, i.e., eyes without evidence that includes prior corneal inflammation, neovascularization, and caustic burns. When cornea grafts are placed in "high-risk" eyes, far fewer grafts are accepted, and even extensive, systemic, immunosuppressive therapy is often unable to reverse this deleterious outcome. Experiments in rodent models have clearly indicated that immune privilege is compromised in artificially induced high-risk eyes [23
], implying that the reason for the high rate of graft failure in human high-risk eyes is a result of the absence of ocular immune privilege.
Orthotopic corneal transplants performed in rodent models have provided direct evidence for the important role of ocular immune privilege [24 , 25 ]. Allografts of cornea placed in normal murine eyes are often accepted, sometimes indefinitely, although no local or systemic immune-suppressive therapy is administered. The high rate of acceptance in this circumstance has been traced, on the one hand, to the presence of immune privilege in the recipient eye [26 ] and conversely, to inherent immune privilege of corneal tissue itself [27 ]. Experimental maneuvers that compromise the privilege of the recipient eye or of the donor graft invariably result in immune rejection of the transplant.
Experimental results obtained when other types of histoincompatible tissue grafts are placed in the anterior chamber of the eye point to the same conclusion. Allogeneic tumor cells, which are routinely destroyed before they can form detectable tumors when injected at subcutaneous sites, often form progressively growing tumors when injected into the anterior chamber [28 ]. If immune privilege is interrupted in the eye before injection of tumor cells, solid tumors usually fail to develop.
Eye-derived antigens induce systemic immune deviation (tolerance), which lacks effector cells and molecules that mediate immunogenic inflammation [anterior chamber-associated immune deviation (ACAID)]
Mice with allogeneic tumors growing progressively in the anterior chamber display evidence suggesting that the recipient’s immune system’s ability to recognize and respond to transplantation antigens on the tumor cells is unusual [29
, 30
]. First, mice bearing progressively growing allogeneic tumors in the eye are unable to reject skin grafts genetically identical to the tumor cells. As the skin grafts accepted by these mice are placed on the recipient’s flank, the alteration of the recipient’s immune response is systemic, not merely within the tumor-containing eye. Second, mice with allogeneic, intraocular tumors fail to acquire or display delayed hypersensitivity (DH) directed at the tumor’s alloantigens. Acceptance of donor-type skin grafts and failure to develop donor-specific DH do not reflect, however, a failure of the immune system of the recipient mouse to respond to tumor alloantigens. Sera of tumor-bearing mice contain antibodies directed at tumor alloantigens, and the secondary lymphoid organs of these mice contain primed, donor-specific, cytotoxic T cells. Moreover, the spleens of eye tumor-bearing mice contain regulatory T cells that can suppress induction and expression of donor-specific DH when adoptively transferred to naïve mice. This pattern of immune responses to tumor alloantigens, ACAID, has proven to be a stereotypic immune response to eye-derived antigens. Thus, if a soluble, heterologous protein antigen such as ovalbumin (OVA) is injected into the anterior chamber, the recipient mice fail to acquire OVA-specific DH when immunized with OVA and complete Freunds’ adjuvant, to acquire OVA-specific serum, noncomplement-fixing antibodies, and to acquire splenic T cells that suppress OVA-specific T helper cell type 1 (Th1) and Th2 responses when transferred into naïve mice.
|
ACAID: DEFINITION, FEATURES, AND MECHANISMS |
|---|
It is now clear that ACAID can be induced by diverse types of antigens (soluble, cell-associated, viral, tumor-specific, haptenic, autologous, allogeneic). Unlike most other forms of experimentally induced unresponsiveness, ACAID can be generated in naïve and presensitized individuals, and when it is present, it is long-lasting, dominant, and resistant to termination. It is important that ACAID has now been shown to be inducible in mice, rats, rabbits, and monkeys, and there is even circumstantial evidence to indicate that ACAID exists in humans [31 , 32 ].
ACAID: the camero-splenic axis
One of the several remarkable features of ACAID is that the phenomenon cannot be evoked in animals lacking a spleen [33
, 34
]. In fact, ACAID induction is aborted if the recipient animal’s spleen is removed surgically before and up to 4–5 days after antigen is injected into the anterior chamber. As a corollary to this time-dependent effect, ACAID is also aborted if the eye in which antigen has been injected is enucleated within 4–5 days. These results indicate that successful ACAID induction requires the antigen-bearing eye and an intact spleen to be in place for at least an initial 4- to 5-day interval, implying that a camero-splenic axis exists for the transfer of immunologically relevant information.
Direct evidence for the existence of this axis was obtained from experiments in which blood was removed from mice that had received an anterior chamber injection of antigen 48 h previously [35 ]. Normal mice that received an intravenous (i.v.) infusion of this blood subsequently developed ACAID to the original antigen, although these mice never received an anterior chamber injection of that antigen. Analysis of blood obtained at 48 h after anterior chamber antigen injection has revealed that the ACAID-inducing signal (AIS) is an F4/80+ monocyte that bears antigenic epitopes but not native antigen itself.
The source of the ACAID-inducing F4/80+ monocytes is probably the eye itself. Dendritic cells and macrophages have been demonstrated to be present within the eye: immediately adjacent to the pigment epithelium of the iris, the ciliary body, and the retina [36 , 37 ]. Similar cells are also present in the stroma of the iris and ciliary body, where ample macrophages are also found. In fact, F4/80+ cells harvested from an eye that received a direct injection of antigen into the anterior chamber 24 h previously are able to induce ACAID when i.v. injected into naïve mice [38 ]. These results suggest that eye-derived antigens are captured by indigenous APCs and are carried, presumably across the trabecular meshwork, via the blood to the spleen, where the subsequent steps in ACAID development take place.
In vitro generation of ACAID-inducing cells
Fortunately, it is possible to create an AIS in vitro by exposing conventional APCs to AqH before or during antigen pulsing. When injected into naïve mice, these AqH-treated APCs induce ACAID [39
]. The factor in AqH that is able to confer ACAID-inducing properties on conventional APCs in vitro is TGF-ß2, the isoform exclusively produced within the eye. Actually, treatment of conventional APCs with amniotic fluid and cerebrospinal fluid also converts the cells into AIS [40
], and this reflects the fact that these fluids also contain large amounts of TGF-ß. It is pertinent that AqH, amniotic fluid, and cerebrospinal fluid are all derived from tissue sites that have been characterized experimentally as immune privileged.
|
ANALYSIS OF ACAID-PROMOTING FEATURES OF TGF-ß-TREATED APCs |
|---|
APCs treated in vitro with active TGF-ß2 and pulsed with a soluble antigen, OVA, create a microenvironment (supernatant) that is rich in active TGF-ß, TSP, tumor necrosis factor (TNF-), type I IFNs, and interleukin (IL)-10. By contrast, this microenvironment is selectively deficient in IL-12. APCs treated in vitro with active TGF-ß2 and pulsed with OVA up-regulate CD40 poorly, and this failure persists even in the presence of responding T cells. Such APCs express normal levels of major histocompatibility complex (MHC) class I and class II molecules, B7-1 and B7-2, and intercellular adhesion molecule-1 and enhanced levels of surface CD1. TGF-ß2-treated, OVA-pulsed APCs readily process and present OVA-derived peptides that are loaded onto MHC class I and class II molecules expressed on the cell surface. TGF-ß2-treated, OVA-pulsed APCs accumulate in the marginal zone of the spleen (but not elsewhere) after i.v. injection into naïve mice [44
], and once they reach this site, the cells recruit NK T cells [45
], additional F4/80+ APCs, and naïve ß CD4+ and CD8+ T cells and proceed to form multicellular clusters that include marginal zone B cells and perhaps T cells [46
47
48
]. Subsequent steps in ACAID development proceed from this point.
Differential gene expression in TGF-ß-treated APCs
The preceding information provides a partial description of the changes that are induced by TGF-ß2 in conventional APCs. We have conducted differential gene expression experiments to extend our knowledge of the genes participating in the ACAID-inducing phenotype of TGF-ß2-treated cells [49
]. Among the more than 40 up-regulated genes and 50 expressed sequence tags following TGF-ß treatment are the following proteins: murine macrophage elastase, coagulation factor X, ubiquitin fusion protein, integral membrane protein 1, RW1 protein, macrophage-inflammatory protein-2 (MIP-2), CDC-10, Cu2+-transporting ATPase, TSP, rjs/Herc 2, type I IFNs (IFN-, IFN-ß), and Oryctolagus cuniculus translation-initiation factor. Our attention has been first focused on TSP, IFNs, and MIP-2 because of the capacity of these molecules to down-regulate IL-12 production by APCs (TSP, IFNs) and to recruit NK T cells (MIP-2). Faunce et al. [50
] have already demonstrated that MIP-2 secretion by blood-borne cells as well as splenic marginal zone F4/80+ cells following anterior chamber injection of antigen is required for NK T cells to be recruited to the site. Sonoda et al. [45
] have reported that NK T cells are absolutely required for ACAID induction, and others have implicated B cells and T cells in ACAID [45
46
47
48
].
|
TSP: ORGANIZER OF THE ACAID-PROMOTING PROPERTIES OF TGF-ß-TREATED APCs |
|---|
There are features of TSP-1 expression that make it an attractive candidate for participation in conferring ACAID-inducing properties on APCs. TSP is an immediate early gene found in APCs treated with TGF-ß2 [49
]. Moreover, APCs treated with TSP display a pattern of up-regulated genes that resembles that induced by TGF-ß2: TGF-ß, TNF-, p75 TNF receptor (R), type I IFNs, type I IFN R. Like TGF-ß2, TSP treatment of APCs leads to down-regulation of the genes for IL-12 and CD40 [41
, 42
, 49
].
More importantly, antigen-pulsed APCs treated with TSP in vitro, instead of TGF-ß, induce ACAID when injected i.v. into naïve mice. In a corollary experiment, conventional APCs were harvested from mice whose TSP gene had been disrupted. When these cells were treated in vitro with TGF-ß, little production of active TGF-ß was observed. In addition, TSP knockout cells, which were treated with TGF-ß2, pulsed with antigen, and then injected i.v. into normal, wild-type mice, failed to induce ACAID. Together, these results indicate that TSP plays a key role in the process by which TGF-ß2–APCs induce ACAID.
Proposed mode of action of TSP in ACAID induction
TSP, by virtue of the receptors and other molecules it is capable of binding, is particularly well suited for orchestrating the interactions between APCs and T cells that lead to ACAID [55
]. First, TSP binds CD36, a receptor well expressed on the surface of APCs, and by virtue of its ability to bind latent TGF-ß, TSP can tether latent TGF-ß to the APC surface. Second, when TSP binds latent TGF-ß, it promotes its conversion to active TGF-ß and subsequently acts to maintain the cytokine in its active state. In this manner, TSP endows the APC with a nanoenvironment highly enriched for active TGF-ß. APCs also express CD47, a receptor that binds TSP and as a consequence, sends negative signals, which inhibit IL-12 gene activation and IL-12 production, to the nucleus.
T cells also express CD47, which offers the potential that TSP could act as a trimolecular bridge (CD36–TSP–CD47) that would enhance the stability of interactions between T cells and APCs. In this context, it is relevant that TSP has been reported to bind CD47 on T cells and to cause the deviation of signal transduction proceeding from the T cell receptor away from the Th1 phenotype.
|
RESPONSES OF T CELLS EXPOSED IN VITRO TO ANTIGEN-PULSED, TGF-ß-TREATED APCs THAT INDUCE ACAID |
|---|
, and the resultant T cells have been found to suppress activation of bystander T cells in vitro and to inhibit induction and expression of DH in vivo. CD8+ T cells (OT-1) similarly exposed to ACAID-inducing APCs in vitro secrete TGF-ß, lose their capacity to lyse target cells, and suppress expression of DH in vivo. It is anticipated that eye-derived APCs that carry antigen to the spleen cause similar perturbations in T cells that they encounter in the marginal zone. |
CONCLUDING REMARKS AND PERSPECTIVE |
|---|
The perils of loss of ocular immune privilege
There are experimental situations and presumed clinical circumstances in which ocular immune privilege is lost cognately or inadvertently. The deleterious consequences of this loss may be revealed as follows: Allogeneic corneal grafts are no longer protected from immune rejection; intraocular tumors elicit immune responses of a vigor and type that destroy the tumor but also cause phthisis; irretrievable damage to the visual axis (corneal stroma, endothelium, lens, vitreous) occurs secondary to acute viral infection or pathogen-associated intraocular inflammation; and autoimmunity to strong ocular antigens is triggered, leading to anterior or posterior uveitis and glaucoma secondary to intraocular inflammation.
The perils of maintaining ocular immune privilege
The fact that ocular immune privilege exists implies that it carries a biologic cost, as immune responses to eye-derived antigens and immune responses expressed in the eye are materially blunted. Some examples of the perils that result from maintenance of immune privilege are that innate and/or adaptive-immune elimination of intraocular tumors may not be possible; acute retinal necrosis secondary to new or recurrent herpes virus infection of the anterior segment is a serious risk, as virus-specific ACAID is induced transiently, rendering the retina vulnerable to direct viral toxicity; and tumors that develop within the eye may never activate a destructive immune response and thereby cannot be eliminated.
The eye’s dim view of ocular immunity and inflammation
The dim view that the eye takes of immunity and inflammation leads to the following outcomes: Activation of effector T cells (Th1, Th2) and innate-immune effectors (natural killer cells, macrophages, polymorphonuclear neutrophils) that generate amplified, nonspecific inflammation is avoided, and a microenvironment rich in antagonists of inflammatory mediators is created; and the risk of overwhelming ocular infection is reduced by infusing the ocular microenvironment with /ß defensins, lysozyme, lactoferrins, and other molecules that can eliminate pathogens before proinflammatory innate and adaptive immunity intervenes. This spectrum of outcomes is apparently based on the evolutionary gamble that few pathogens that reach the eye require Th1/Th2 immunity and complement-fixing antibodies for elimination.
The eye’s practical view of ocular immunity and inflammation
The practical view that the eye takes of immunity and inflammation leads to the following outcomes: Activation and expression of adaptive- and innate-immune effectors that eliminate pathogens and infected cells with elegance and specificity, such as cytotoxic T cells and noncomplement-fixing antibodies, are permitted; and even as these elegant, specific effectors run the risk of elimination of critical, nonreplicating cells (such as the corneal endothelium and retinal neurons), expression of MHC class Ia and class II molecules on key ocular parenchymal cells is reduced or eliminated so that the cells are virtually invisible to effector T cells that have been primed elsewhere (extraocularly). This spectrum of outcomes is apparently based on the evolutionary hope that most pathogens can be eliminated before infection of ocular parenchymal cells by encounters outside the eye.
|
ACKNOWLEDGEMENTS |
|---|
Received November 21, 2002; accepted January 28, 2003.
|
REFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  K. Gronert Lipid Autacoids in Inflammation and Injury Responses: A Matter of Privilege Mol. Interv., February 1, 2008; 8(1): 28 - 35. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Kuffova, M. Netukova, L. Duncan, A. Porter, B. Stockinger, and J. V. Forrester Cross Presentation of Antigen on MHC Class II via the Draining Lymph Node after Corneal Transplantation in Mice J. Immunol., February 1, 2008; 180(3): 1353 - 1361. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Biteman, I. R. Hassan, E. Walker, A. J. Leedom, M. Dunn, F. Seta, M. Laniado-Schwartzman, and K. Gronert Interdependence of lipoxin A4 and heme-oxygenase in counter-regulating inflammation during corneal wound healing FASEB J, July 1, 2007; 21(9): 2257 - 2266. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. R. Smith, D. Choi, T. J. Chipps, Y. Pan, D. O. Zamora, M. H. Davies, B. Babra, M. R. Powers, S. R. Planck, and J. T. Rosenbaum Unique Gene Expression Profiles of Donor-Matched Human Retinal and Choroidal Vascular Endothelial Cells Invest. Ophthalmol. Vis. Sci., June 1, 2007; 48(6): 2676 - 2684. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Nakamura, A. Terajewicz, and J. Stein-Streilein Mechanisms of Peripheral Tolerance following Intracameral Inoculation Are Independent of IL-13 or STAT6 J. Immunol., August 15, 2005; 175(4): 2643 - 2646. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Engelbert and M. S. Gilmore Fas Ligand but Not Complement Is Critical for Control of Experimental Staphylococcus aureus Endophthalmitis Invest. Ophthalmol. Vis. Sci., July 1, 2005; 46(7): 2479 - 2486. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Gronert, N. Maheshwari, N. Khan, I. R. Hassan, M. Dunn, and M. Laniado Schwartzman A Role for the Mouse 12/15-Lipoxygenase Pathway in Promoting Epithelial Wound Healing and Host Defense J. Biol. Chem., April 15, 2005; 280(15): 15267 - 15278. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Kitaichi, K. Namba, and A. W. Taylor Inducible immune regulation following autoimmune disease in the immune-privileged eye J. Leukoc. Biol., April 1, 2005; 77(4): 496 - 502. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Vij, L. Roberts, S. Joyce, and S. Chakravarti Lumican Regulates Corneal Inflammatory Responses by Modulating Fas-Fas Ligand Signaling Invest. Ophthalmol. Vis. Sci., January 1, 2005; 46(1): 88 - 95. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Chakravarti, F. Wu, N. Vij, L. Roberts, and S. Joyce Microarray Studies Reveal Macrophage-like Function of Stromal Keratocytes in the Cornea Invest. Ophthalmol. Vis. Sci., October 1, 2004; 45(10): 3475 - 3484. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. N. Mallam, M. Y. Hurwitz, T. Mahoney, P. Chevez-Barrios, and R. L. Hurwitz Efficient Gene Transfer into Retinal Cells Using Adenoviral Vectors: Dependence on Receptor Expression Invest. Ophthalmol. Vis. Sci., June 1, 2004; 45(6): 1680 - 1687. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
INTRODUCTION