- Email: [email protected]
Immunohistochemical characterization of dendritic cells and macrophages in the aqueous outflow pathways of the rat eye
Exp. Eye Res. (1992) 55, 315-324
lmmunohistochemical Characterization Macrophages in the Aqueous Outflow P. G. McMENAMIN” Department
of Anatomy
(Received
AND
of Dendritic Cells and Pathways of the Rat Eye I. HOLTHOUSE
and Human Biology, The Centre for Human Biology, Australia, Nedlands, Perth, Western Australia 6009 Lund
75 July
7997 and accepted
in revised form
The University
73 November
of Western
7997)
Immunohistochemical studies were performed to determine the distribution, phenotype and ontogeny of macrophages and dendritic cells (DCs) in the aqueous humour oufflow pathways of the rat eye. Optimal fixation and indirect immunoperoxidase techniques were employed in conjunction with a panel of mAbs on tangential frozen sections of ocular tissues from a total of 3 7 Wistar Furth rats aged 12-13 days (n = 8), 3 weeks (n = 12), 7 weeks (n = 5) and 15 weeks (n = 12). The density ofimmunopositive cells was scored qualitatively. A moderate to low density of Ia+ cells with a dendritic morphology were observed in the trabecular meshwork. DCs were also identified in the suprachoroidal space and in the connective tissues of nerves and vessels piercing the sclera, i.e. in association with non-conventional aqueous o&low pathways. The phenotypical and morphological characteristics of these cells would indicate that they may potentially act as antigen presenting cells (APCs). Non-dendritic pleomorphic cells with a macrophage phenotype were also identified in the trabecular meshwork, and bipolar or elongated cells with a macrophage phenotype were a noticeable feature in the perivascular region of collector channels and the limbal episcleral veins. Some macrophage and DC-like cells were observed in intimate association with limbal mast cells. Theories on the mechanisms of Anterior Chamber Associated Immune Deviation (ACAID) have assumed APCs are largely absent from the tissues lining the anterior chamber. Our findings of a low but moderate density of putative APCs in the conventional and non-conventional aqueous humour oufflow pathways are discussed in relation to the various theories of ACAID. Key words: dendritic cell; macrophage; trabecular meshwork; aqueousoutflow pathway; ACAID: ocular immunology: classII; anterior chamber: rat eye.
1. Introduction Local antigen-induced immune responseshave various suppressive mechanisms that act to restrict lymphocyte activation/proliferation such as the generation of antigen-specific suppressor T-cells. Precise control mechanisms are particularly well developed in delicate organs such as the lungs, gut, brain and eye whose tissues could be functionally compromised by the non-specific ‘bystander ’ damage characteristic of
some cell-mediated
immune
responses.
eye ; and (4) a paucity or lack of Ia+ cells lining the anterior chamber (AC) (Niederkorn and Streilein, 1983 ; Abi-Hanna, Wakefield and Watkins, 198 7 ; Williamson, DiMarco and Streilein, 1987). The conventional and non-conventional aqueous outflow pathways are ideally suited to function as immunological filters of the aqueous humor (AqH). Indeed, it is well known that the trabecular meshwork acts as a self cleansing biological filter, removing cellular and extracellular debris from the AqH on its exit from the anterior chamber angle (for review see
Experimental intracameral injection of a wide variety of cell surface and soluble antigens elicits a
Rohen and Liitjen-Drecoll, 1982). The ubiquitous trabecular cells (TCs) that line the AqH filled inter-
variety of aberrant systemic immune responses, among which are antigen-specific suppression of delayed hypersensitivity reactions and normal cytotoxic and helper T-cell function (Niederkorn and Streilein, 1983). This phenomenon is termed ACAID
trabecular spaces phagocytose material which falls to pass through the lntertrabecular spaces en route to Schlemm’s canal. Whilst some studies have claimed there is a complete absence of MHC class II (Ia) antigen-bearing cells in the tissues lining the human anterior chamber (Abi-Hanna et al., 1987), there have been reports of small numbers of round Ia
(anterior
chamber-associated immune
deviation)
(Streilein and Niederkorn, 1981; for review see Tompsett, Abi-Hanna and Wakefield, 1990). A number of factors have historically been considered essential for the induction of ACAID. These include: (1) a route for antigen to reach the vascular system;
(2) an intact functional spleen (Streilein and Niederkorn, 198 1) ; (3) the anatomical integrity of the
cells (Bakker and Kijlstra, 198 5 ; Wang et al., 198 7) in the iris and ciliary body. Other studies (Williamson, Bradley and Streilein, 1989) have claimed to demonstrate significant numbers of F4/80+ macrophages in the mouse iris, of which one-third were Ia positive ; however, the morphological appearance of these cells
was not illustrated.
* Forcorrespondence. 00144835/92/080315+10
positive cells (Latina et al., 1988) or single scattered
%08.00/O
Elongated
cells, which
may
0 1992 AcademicPressLimited
P. G. McMENAMlN
316
represent macrophages (M@) or professional antigen presenting cells (APCs), have also been described in the trabecular meshwork and around Schlemm’s canal (Lynch et al., 1987; Latina et al., 1988). However, recently Tripathi et al. (1990) have claimed that all cells of the trabecular meshwork and Schlemm’s canal together with the entire cornea1 epithelium and endothelium of human eyes, are class II positive. This extent of expression is questionable in light of a considerable body of previous evidence that class II expression, for example in the epithelium, is limited to Langerhans cells (for example, Rodrigues et al., 1981; Gillette, Chandler and Greiner, 1982; Whitsett and Stulting, 1984; Williams and Coster, 1989). Ia+ cells in the trabecular meshwork could be involved in the sampling and presentation of an array of antigens that enter the AC, including both endogenous antigens (e.g. retinal S-antigen) and exogenous antigens (e.g. HSV-1 or experimentallyintroduced antigens). Previous immunohistochemical studies of the anterior segment have not provided convincing evidence that the Ia+ cells are dendritic in form (Bakker and Kijlstra, 1985; Bakker et al., 1986; Lynch et al., 1987; Latina et al., 1988). The coexistence of these two features are classically considered as prerequisites for classification as dendritic cells (DCs) or APCs (Steinman et al., 1979). Future determination of the functional capacity of DC populations in regulating immune responses in the anterior segment necessitates the acquisition of information on the distribution, phenotype and ontogeny of M+s and APCs in the AqH outflow pathways. The aim of the present study was to provide this information in the rat eye. The rat offers several advantages in terms of anatomical similarities of the AqH outflow pathways to humans (McMenamin and Al-Shakarchi, 1989) and the availability of a wide range of well characterized monoclonal antibodies (mAb). In previous studies of the respiratory tract, the detection of an intraepithelial Ia+ DC network has been dependent on novel methods of cell surface antigen preservation and appropriate sectioning plane in which planar views of the epithelium are obtained (Holt et al., 1989 ; Holt, Schon-Hegrad and McMenamin, 1990; Schon-Hegrad et al., 1991). Similar techniques of optimal fixation and sectioning were employed in the present investigation of ocular tissue.
The mAb employed in this study and their specificities are briefly summarized in Table 1. Monoclonal antibodies EDl, ED2, ED3, ED8 and ED9 were kindly supplied by Dr C. D. Dijkstra, Department of Histology, Vrije University, Amsterdam, The Netherlands. 0X-6, 0X-19, 0X-41, OX-42 and W3/25 were obtained from commercial sources (Serotec, Oxford, IJ.K.). The mAbs supplied as ascites were diluted to optimal levels in PBS (pH 7.4-7.6, 290-320 mosmol). Tissue Preparation and Immunoperoxidase Staining The tissue fixation and sectioning methods have been described previously (Holt et al., 1990). In summary, the anaesthetized animals were perfused with heparinized PBS followed by cold absolute ethanol. This served to prevent antigen diiusion and remove all blood, (a source of pseudoperoxidase activity) from the vascular bed. Eyes and lymph nodes TABLE I Monoclonal antibody specificities MoAB ED1
ED2
ED3
ED8
ED9 OX-6
and Methods
Animals
Female Wistar Furth rats (specific pathogen free) were obtained from the Animal Resource Centre, Murdoch University, Western Australia. These consisted of eight 12-13 day old rats (just before eye opening), five 7 week and 12 each of 3 and 15 week old rats (total 37).
I. HOLTHOUSE
Monoclonal Antibodies
ox-41 2. Materials
AND
OX-42 ox-19 W3/25
Specificity
___-
Cytoplasmic antigen in most rat monocytes. macrophages and dendritic cells Membrane antigen on connective tissue macrophage subpopulations in lymphoid and nonlymphoid organs Membrane antigen of macrophages mostly confined to lymphoid organs ; small DC subpopulations CDll/CD18 antigen 95 kDa protein (probably related to C3bi receptor) microglia, Langerhans cells, IDCs, some macrophages Similar (but wider) distribution to ED8 Ia molecule: DCs, some macrophages 110000-120000 MW surface protein on macrophage subpopulations and glial cells Rat C3bi receptor (CDll/CDlS) 95 kDa protein ( - RD8) CD5 T-cells CD4 T-cells some dendritic cells and macrophages
Reference Dijkstra et al. (1985) Dijkstra et al. (1985), Damoiseaux et al. (1989b) Dijkstra et al. (1985h Damoiseaux et al. (1989b) Damoiseaux et al. (1989b)
Damoiseaux et al. (1989b) McMaster and Williams (19 79) Robinson, White and Mason (1986) Robinson et al. (1986) Dallman et al. (1984) Jefferies et al. (1987)
DENDRlTlC
CELLS
AND
MACROPHAGES
IN THE
RAT
EYE
were removed and placed in absolute ethanol overnight. This was followed by rehydration and inilltration with a PBS/OCT mixture (Tissue Tek II freezing medium, Miles Laboratories, IN, U.S.A.). Tissues were embedded in OCT for cryostat sectioning (10 pm). The eyes were orientated to produce tangential sections which contained larger areas of trabecular meshwork and limbal tissue than conventional meridional sections [Figs l(A) and (B)]. A standard indirect immunoperoxidase procedure using mouse mAb (see Table I), biotinylated sheep anti-mouse and streptavidin-horseradish peroxidase (Amersham Laboratories, Buckinghamshire, U.K.) was used in this study. The horseradish peroxidase was visualised using 3,3 diaminobenzidine tetrahydrochloride (Sigma Chemical Co., St Louis, MO, U.S.A.) (12 mg per 10ml PBS) and 5~1 H,O, (30% by volume). Endogenous peroxidase activity was not blocked as it was found to compromise antigenicity. Cells displaying endogenous peroxidase activity were easily distinguishable from immunostained cells. Sections in which the primary mAb was omitted and in which an inappropriate antibody (OKT4), directed against the human CD4 marker was substituted, acted as negative controls.. Positive control tissue was stained in paraliel and consisted of lymph node (for all mAbs) and conjunctival Langerhans cells [for anti-Ia antibodies, i.e. 0X-6, see Fig. 2(A)]. Sections were lightly counterstained with haematoxylin before mounting. Qualitative Analysis
Due to the size and irregular nature of the tissue involved, quantitative analysis such as that usually performed on DC networks in epithelia was not considered appropriate. A qualitative investigation of staining patterns was made by light microscopy. Positive staining cells were classified as DCs if they possessed one or more large slender cytoplasmic processes. All others were classiiled as non-dendritic cells (NDCs). The density of these cells was scored
TABLE
317
FIG. 1. A, Conventional meridional section of the rat anterior segment. Note the small area of trabecular meshwork (TM) available for examination. The large episcleralvesselscut in transversesectionare indicatedby small arrows. B, Lower power micrograph of oblique or tangential frozen section of the anterior segmentwhich clearly demonstratesthe greater area of trabecular tissue (TM) availablefor study. In additionobliqueandlongitudinal profiles of the episcleralvessels(small arrows) are also presentin such sections.The dark cellsin the conjunctival laminapropria aroundthesevessels are mastcells(Toluidine blue stain). (A) x200; (B) x85.
qualitatively as ‘ + + + ’ (extremely high), ‘ + + ’ (moderate, approximately two to four cells per high power field), ‘ + ’ (low, about one cell per high power
II
Age-relatedchangesin the distribution andfrequency of ClassII (Ia) positive cells in the rat aqueousoutflow pathways*. Conventional pathways include trabecular meshwork (TM), Schlemm’s canal (SC) and episcleral veins (ESV); nonconventional routes include suprachoroidalspace(SCS) and scleral vessels(SV)
3 week
12-13 day __.
*Density
7 week
15 week
Tissue location
DC
NDC
DC
NDC
DC
NDC
DC
NDC
TM SC ESV scs sv
1 i c 5 +
ItI + 4 -
-t * + + .
k IL + f .
?E + + + .
+ rfr + + .
+ zk ++ f+ ++
+ + ++ +
scored as in Materials
and Methods.
P. G. McMENAMlN
318
FIG. 2. For legend see opposite.
AND
I. HOLTHOUSE
DENDRITIC
CELLS
AND MACROPHAGES
IN THE RAT EYE
319
TABLEIII
Thedistribution* of immunogositivecells in
the various regionsof the rat outflow pathways in 3 and 15 week old animals. Seelegendsto Tables1 and ZZfor abbreviations
OX-6
Tissue location
ED1
ED8
ED2
W3/25 _____ DC NDC
DC
NDC
DC
NDC
DC
NDC
DC
NDC
DC
NDC
Ik +
f +
-
+
+
++
-
+
-
+ -
-
-
k AI
f +
-
f
-
-
+
-
-
-_ -
-
+ ++
+ ++
+ -
++ -
++ +++
+ ++
+ -
-
+
++
+
-
-
TM 3 week 15 week SC 3 week 15 week ESV 3 week 15 week scs 3 week 15 week sv 15 week
OX-42
---+ + -
++ ++
++ +++
* +
f -
-
-
+ -
+_ -
+
-
-
f
+
-
5
+
++
*
.
.
.
.
* Densityscoredasdescribed in MaterialsandMethods.
field) and ‘ _+’ (rare or occasional, present but not in every section), ’ - ’ (absent). 3. Results The age-related distribution of Ia+ cells in the various components of the AqH outflow pathways of 12-13 day and 3, 7 and 15 week old animals is summarized in Table II. The distributions of the other mAb employed in this study for 3 and 15 week old animals are summarized in Table III. Conventional Outflow Pathway l’rabecuiur meshwork.Ta+ cells within the trabecular
meshwork demonstrated either true dendritic [Figs 2(B) and (C)] or elongate/bipolar morphology (i.e. DCs) or were NDCs [Fig. 2(D)]. The density of these cells increased with age (Table II) and they appeared to be associated with both the inner more robust connective tissue trabeculae (pectinate ligaments) and the outer more lamellate trabeculae. In the 15-week group, DCs also stained with ED1 and occasionally with BD2. NDCs stained positively for ED1 , ED2 [Fig. 2(D)], ED9, Ox-41 and occasionally for Ox-42 in both
age groups (ED9 and OX-41 were excluded from the 3week group because of excessive background staining). Canal of Schlemm and collector channels.Ia+ cells were located in small numbers adjacent Schlemm’s canal and the draining collector channels [Fig. 2(E)]. In longitudinal sections of the canal these round or ovoid cells were located in the connective tissue of the inner and outer walls (juxtacanalicular zone). They were observed to increase in number with age and were found to be 0X-6+, 0X-41’. 0X-42’. EDl+, ED2+ and ED8’. Episclerulveins. Ia+ cells were consistently located in association with the episcleral veins (ESV) in all age groups [Figs 2(E) and (G)]. These cells displayed both dendritic and non-dendritic morphologies in similar proportions. Longitudinal sections revealed that Ia+ cells were distributed along the length of the vessels in close proximity to the endothelial cells and pericytes [Fig. 2(G)]. An age-related increase in the density of these cells was observed. The cells around the episcleral vessels stained particularly strongly for ED2 and in varying densities using the mAbs 0X-42, EDl, ED8 and ED9. Through-
FIG. 2. A, Ia+ (0X-6) Langerhanscellswithin limbal conjunctival epithelium (sectionedparallel to the underlying basal lamina, i.e. plan view) actedasin situ positivecontrols for Ia immunostaining.B, Low power view of iridocornealanglein an obliquesection.Adjoining structures:ciliary processes (CP).sclera(S)and anterior chamber(AC).Note the presenceof Ia+ (OX6) immunopositivedendritic cellsin the trabecular tissue(TM), shown in greater detail in (C). D, ED2+non-dendritic cellsin the trabecular meshwork. Ciliary epithelium is identifiable. E, Perivascular Ia+ (0X-6) non-dendritic cells in the walls of Schlemm’scanalor outer trabecularmeshwork(TM), collector channels(smallarrows)and the episcleralveins(ESV).E-cornea1 epithelium. CP-ciliary processes.F, ED2+ bipolar DC-like cells in close associationwith mast cells (weakly stained with haematoxylin). G, NumerousIa+ elongatedbipolar perivascular macrophagesadjacent to episcleralvessels(ESV). H, Ja+ dendritic cells in the anterior suprachoroidalspace(SCS),ros-rod outer segments,arrows delineatechoroid. (A) x 600; (B) x210: (C) x 500: (D) x 600; (E) x280; (F) x 550; (G) x200: (H) x 300.
320
out the highly vascular conjunctival lamina propria, cells with a similar phenotype and morphology viz histiocytic/macrophagic were encountered. In the 3week-old animals similar cells were detected close to vessels and in the lamina propria using 0X-42, EDl, ED2. ED8 and ED9; however, the densities were generally lower. In the 15-week-old animals, 0X-6+, 0X-42’ and ED2+ bipolar DC-like cells were often located either in close proximity to or in contact with perivascular and connective tissue mast cells [Fig. 2(F)], which were particularly common in the limbal zone. Non-conventional Route Suprachoroidal space (SCS) and scleral vessels. Ia+ DCs were located in the outer choroid or SCS with dendritic processes entering both the choroid and sclera [Fig. 2(H)]. Occasional EDl+ DCs and NDCs were located in the SCS of both 3 and 15 week old animals. DCs seemed to be arranged in a two-dimensional network interposed between the choroid and sclera. The density of these cells increased with age (Tables II and III). Small numbers of Ia+ DCs which were located in the dense connective tissue of the sclera similarity showed an increase in number with age. However, close examination revealed that most of these cells lay adjacent to vessels or nerves that pierced the sclera. Small numbers of EDl+ and ED2+ DCs and NDCs were located around the large scleral vessels in some of the 15week animals. No Ia+ DCs were observed in the sclera of the 3-week animals as no intrascleral vessels or nerves were sectioned in these animals. Endogenous Peroxidase Activity Cells displaying endogenous peroxidase activity were rare within the eye and were confined mainly to the lamina propria of the limbal conjunctiva. These cells were readily identifiable from positively stained cells in other sections because of their consistent rounded shape and much more intense dark brown/ black staining. The pseudoperoxidase activity of erythrocytes was minimal as a consequence of the vascular perfusion/fixation protocol. 4. Discussion By using optimal pre-embedding fixation and appropriate sectioning techniques we have demonstrated la+ cells with both dendritic and non-dendritic morphologies related to the rat AqH outilow pathways and have further characterized these cells using a panel of mAb. Despite the current lack of a specific anti-rat DC mAb, satisfactory identification can be based on staining with an anti-Ia mAb and other M$ pan-specific mAb in conjunction with demonstration of a dendritic morphology.
P. G. McMENAMlN
AND
I. HOLTHOUSE
Distribution and Phenotype of Macrophage-likeCellsin the AqH Pathways
NDCs in the trabecular meshwork and in the walls of Schlemm’s canal displayed Ia (0X-6). ED1 , ED2. OX41 and occasionally OX-42 ; however, these cells were negative for ED3. The mAb ED1 recognizes peripheral blood monocytes and most M@, whilst ED2 recognizes resident tissue M@s of lymph node, spleen, thymus (Dijkstra et al., 1985; Damoiseaux et al., 1989a. b), skeletal muscle (Honda, Kimura and Rostami, 1990) and extraocular muscle (personal observation, P.McM). ED3 is restricted to M+ subpopulations in lymphoid organs and at sites of inflammation (Dijkstra et al., 1985). The other markers used (0X-6, 0X-41. 0X-42, ED8, ED9 and W3/25) have a highly variable frequency on M&s that may be related to microenvironmental factors or different stages of M$ maturation (Beelen et al., 1987; Damoiseaux et al., 1989a, b). On the basis of the morphological appearance and phenotype we conclude that the majority of NDCs in the trabecular meshwork and Schlemm’s canal are resident tissue macrophages. The role of the trabecular meshwork as a biological filter of the AqH places it in an optimal location for immune surveillance. It has been proposed. primarily on the basis of ultrastructural evidence, that following phagocytosis of debris TCs detach from the trabecular beams and migrate through the meshwork to Schlemm’s canal (Rohen and Van der Zypen, 1968 : Grierson and Lee, 1973). An alternative explanation is that some of the cells lining the intertrabecular spaces in the normal resting state are phagocytically active M$s, which have assumed residency in the tissue following their role in developmental remodelling of the iridocorneal angle (Rem& Urner and Aeberhard, 1983 : Richardson et al., 198 5). Our ontogeny data suggests that M&s may in fact continue to populate the anterior segment and perivascular connective tissue in the episclera until maturity (Table 111). The second population of cells which we believe to be macrophages are the NDC and elongated cells adjacent to the episcleral veins which were positive for all mAbs except 0X-41, ED3 and W3/25. Their morphology. strong reactivity with ED2 and extensive distribution throughout the conjunctival lamina propria is strong evidence that these cells are resident tissue n/r+ or histiocytes. Similar cells have been located in perivascular sites within the human adult CNS (Hickey and Kimura, 1988; Graeber, Streit and Kreutzberg, 1989) human foetal CNS (Esiri, Al Izzi and Reading, 1991), dermis (Sontheimer, 1989 ; Sontheimer, Matsubara and Seelig, 1989) and skeletal muscle (Honda et al., 1990). Some of these perivascular M$s in other tissues have also demonstrated class II staining similar to those around episcleral veins, suggesting a potential role as antigen presenting cells. Indeed some of these cells in the episcleral veins and surrounding limbal/conjunctival lamina propria
DENDRITIC
CELLS
AND
MACROPHAGES
IN THE
RAT
may represent precursors of conjunctival and cornea1 Langerhans cells (for review see Williams and Coster, 1989). Interestingly, tracer studies (Sherman, Green and Laties, 1978) have shown that proteins injected into the AC leak from episcleral limbal vessels, thus it is feasible that class II bearing cells around collector channels and episcleral veins have direct access to intracameral antigens. These cells could travel to the local lymph nodes via draining conjunctival lymphatics. thus bypassing the camero-splenic axis. Alternatively, they may be active in phagocytosis and clearance of immune complexes which, in the mouse, have been shown to accumulate around these limbal vessels (Hylkema, Broersma and Kijlstra, 1988). These cells may therefore play a role in limiting the inflammatory responses which emanate from immune complex deposition. The increase in frequency of class IT positive dendritic and non-dendritic cells around the episcleral veins and limbal/conjunctival lamina propria at 3 weeks, 1 week after eye opening (day 14) suggests that this ontogenic process may be antigen driven; however, further experimental evidence would be required to support this proposal. Perivascular and connective tissue M+s were commonly observed in intimate contact with conjunctival mast cells. An intriguing possibility is that the high numbers of mast cells in the limbus may influence DC and M+ maturation via the production of cytokines such as IL-3 and GM-CSF (Wodnor-Filipowicz, Heusser and Moroni, 1989: for review see Gordon, Burd and Galli, 1990). It has recently been shown that GM-CSF upregulates the expression of IL-2 receptors on lymphborne (veiled) dendritic cells which may potentially play a role in increasing the efficiency of T cell activation (MacPherson, Fossum and Harrison, 1989 I. Phenotype of DC-iike Cells Class II expression and dendritic morphology are classically considered as the hallmarks of professional APCs or DCs (Steinman et al., 1979; Kraal. 1989). To our knowledge, this is the first demonstration of cells in the mammalian aqueous humour outflow pathways which fulfill both these criteria and may therefore represent a population of antigen presenting cells. Small numbers of DC-like cells in these locations expressed ED1 which is known to recognize DC subpopulations (Damoiseaux et al., 1989b; Holt and Schon-Hegrad, 1987). One of the critical factors thought to play a role in ACAID induction was the lack of la+ cells lining the AC (Williamson et al., 1987). Clearly, in light of our evidence of a low to moderate density of Ia+ DC population in the trabecular meshwork this hypothesis must be reevaluated, particularly since it is known that as few as one DC is necessary to produce a mixed lymphocyte response (Langhoff and Steinman, 1989: Romani et al., 1989). The failure of previous studies to demonstrate class II antigen-bearing dendritic shaped cells 31
EYE
321
in the human and non-human anterior segment is most likely related to modes of fixation, section orientation and immunohistochemical techniques. Role of la+ Cells of the AqH Outflow Pathways In areas such as the skin and oral mucosa, la+ Langerhans cells permit sensitization to a variety of antigens. In contrast, experimental exposure of the AC to antigens results in a suppressive immune reaction (ACAID) (Streilein and Niederkorn, 1981). The induction of ACAID by resident Ia+ cells in the rat eye may occur via a number of mechanisms. These include: (1) the presence of ‘alternative ’ APCs that activate suppressor T-cells rather than helper T-cells; (2) ‘faulty ’ presentation of antigen by both bone marrow and non-bone marrow derived cells; and (3) the presence of T-cell suppressive factors and M$s. It has been suggested that an alternative population of ‘I-J’ positive APCs exist which preferentially elicit the generation of antigen specific suppressor T-cell activity (Toews et al., 1980 ; Granstein and Greene, 1985) ; however, the existance of such ce1l.sremains highly controversial. Aberrantly Ia+ non-bone marrow derived cells (e.g. keratinocytes, fibroblasts) or la+ M$s and ,4PCs that lack the ability to deliver antigen non-specific T-cell co-stimulatory signals such as IL-1 are capable of producing a state of T-cell unresponsiveness or anergy : such T-cells are effectively ’ paralysed ’ to further antigen presentation by professional APCs (for review see Geppert and Lipsky, 1989 ; Schwartz, 1990). Clearly there is potential for Ia+ cells in the eye to produce T-cell unresponsiveness, a possibility supported by recent evidence that introduction of IL-l into the AC of mice was able to abrogate ACAID (Benson and Niederkorn, 1990). A fruitful area of future study would be the investigation of the ability of anterior segment-derived DCs to activate T-cells and/or produce antigen non-specific signals such as the interleukins. Normally, APCs such as Langerhans cells require a maturational step stimulated by factors present within lymph and lymph nodes, such as IL-1 and GM-CSF, to enhance their ability to activate resting T-cells (for review see Romani and Schuler, 1989). The absence of a lymphatic drainage pathway in the eye may prevent intraocular APCs from receiving these maturation signals and they may therefore induce T-cell paralysis or the induction of suppressor T-cells in the spleen in response to antigen presentation. In organs such as the lung, resident alveolar M+s play a role in the suppression of T-cell proliferation and activation (for review see Halt, 1986 : Holt, Schon-Hegrad and Oliver, 1988). Suppression of T-cell mediated reactions, particularly delayed hypersensitivity reactions, is likely to have the physiological function of limiting inflammatory responses in organs whose function could easily be compromised by nonliEP 55
P. G. McMENAMlN
322
specific ‘bystander’ damage. In the eye, suppressive M$s could possibly function similarly to alveolar M$s, preventing T-cell activation by resident Ia+ APCs and thereby restricting destructive inflammatory reactions. Indeed, recently a population of M$s (one-third of which were Ia+) has been reported in the mouse ciliary body and iris stroma (Williamson et al., 19 8 9 ; Streilein and Cousins, 1990). Whole iris/ciliary body tissue explants were found to be incapable of activating allogeneic T-cells, and thus assumed to lack APCs. Furthermore, the explants were found to suppress Tcell proliferation in culture, and preliminary evidence suggests that transforming growth factor-beta (identified in aqueous humour) may be one suppressive factor in the tissue explant supernatant ; however, the cellular source(s) of such suppressive factors has not been identified. Non-conventional Outflow Pathways A proposed route of egress of AqH and large molecules (up to 1 pm) from the AC is posteriorly into the SCS and then escaping either via the venous system (uveo-vortex) or in the perivascular and perineural spaces in the sclera (uveo-scleral) (Inomata, Bill and Smelser, 1977). In the present study we have shown for the first time that in the rat Ia’ DC-like cells are located in the SCS and adjacent nerves and vessels that pierce the sclera. The relative importance of conventional versus non-conventional routes is highly species dependant and the requirement for a more complex network of APCs in the non-conventional routes may be more important in lower mammals such as the rat. Further studies will be necessary to reevaluate whether such cells exist in primate and human eyes. On the basis of our findings it is evident that previously undescribed populations of MI.$s and DCs exist within the normal rat anterior segment and AqH outflow pathways (conventional and non-conventional) and these cells increase with age up to maturity. This is in addition to a further population of DCs which we have observed within the ciliary body epithelium and iris stroma (unpubl. res.). Further in vitro
studies
of the capacity
of DCs from the outflow
tissues (and other anterior segment tissues) to process and present antigen, and their interactions with Tcells are required to elucidate their role in the induction of ACAID. We have also characterized the normal macrophage/histocyte population which is present
in
the
trabecular
meshwork
and
in
the
perivascular connective tissue of Schlemm’s canal, collector channels and the episcleral vessels. Some of these may also represent immature DCs and the intimate relation between those cells in the episcleral region and limbal mast cells may be a novel observation of importance.
AND
I, HOLTHOUSE
Acknowledgements We would like to acknowledge the valuable advice of Dr P. G. Holt of the Western Australian Research Institute for Child Health (WARICH) during the preparation of this manuscript.
References Abi-Hanna. D., Wakefield, D. and Watkins. S. (1987). HLA antigens in ocular tissue. I. In vivo expression in human eyes. Transplantation 45. 610-3. Bakker, M., Grumet, F. G., Feltkamp, T. E. W. and Kijlstra. A. (1986). HLA-antigens in the human uvea. Dot. Ophthalmologica 61, 271-9. Bakker, M. and Kijlstra, A. (1985). The expression of HLAantigens in the human anterior uvea. Curr. E;{e Res. 4, 599-604. Beelen. R. H. J.. Eestermans. I. L., Dopp. E. A. and Dijkstra. C. D. (1987). Monoclonal antibodies EDI, ED2 and ED3 against rat macrophages : Expression of recognised antigens in different stages of differentiation. TranspI. Proc. 19, 3166-70. Benson, J. L. and Niederkorn. J. Y. ( 1990). Interleukin-1 abrogates anterior chamber-associated immune deviation. Invesf. OphthaZmo[. Vis. Sci. 31, 2 12 3-8. Dallman, M. J., Matthew, I,. T. and Green, J, R. (1984). MKC OX-19 : a monoclonal antibody that labels rat T lymphocytes and augments in vitro proliferative responses. Eur. 1. lmmunol. 14, 260-7. Damoiseaux, J. G.. Dopp, E. A.. Beelen, R. H. J. and Dijkstra. C. D. (1989a). Rat bone marrow and monocyte cultures: influence of culture time and lymphokines on the expression of macrophage differentiation antigens. J. Leukocyte Biol. 46. 246-53. Damoiseaux. J. G., Dopp, E. A.. Neeljes, ]. 1.. Beelen, R. H. J. and Dijkstra. C. D. (1989b). Heterogeneity of macrophages in the rat evidenced by variability in determinants: two new anti-rat macrophage antibodies against a heterodimer of 160 and 95 kd (CD1 l/CD1 8). 1. Leukoc$e BioI. 46, 556-64. Dijkstra. C. D., Dopp, E. A., Joling. P. and Kraal. G. ( 198 5). The heterogeneity of mononuclear phagocytes in lymphoid organs : distinct macrophage subpopulations in the rat recognised by monoclonal antibodies EDI. ED2 and ED3. Immunology 54. 589-99. Esiri, M. M.. Al Izzi, M. S. and Reading, M. C. ( 199 1). Macrophages, microglial cells and HLA-DR antigens in fetal and infant brain. J. CJin. Pathol. 44, 102-6. Geppert, T. D. and Lipsky. P. E. (1989). Antigen presentation by cells that are not of bone marrow origin. Reg. Immunol. 2. 60-71. Gillette, T. E.. Chandler. J. W. and Greiner, J. V. ( lY82). Langerhans cells of the ocular surface. Ophthalmology 89. 700-l 1. Gordon. j. R., Burd, P. R. and Galli, S. (1990). Mast cells as a source of multifunctional cytokines. Irnmunol. Today Il. 458-64. Graeber, M. B., Streit, W. J. and Kreutzberg, G. W. (1989). Identity of EDZ-positive perivascular cells in the rat brain. I. Neurosci. Res. 22, 103-6. Granstein. R. D. and Greene, M. I. (1985). Splenic I-Jbearing antigen-presenting cells in activation of suppression: further characterisation. Cell. Immunol. 91, 12-20. Grierson, I. and Lee, W. R. (1973). Erythrocyte phagocytosis in the human trabecular meshwork. B. 1. OphthalmoL 29, 400-l 5. Hickey, W. F. and Kimura, H. (1988 ). Perivascular microglial
DENDRITIC
CELLS
AND
MACROPHAGES
IN THE
RAT
cells of the CNS are bone marrow-derived and present antigen in vivo. Science 239, 290-2. Holt, P. G. (1986). Down-regulation of immune responses in the lower respiratory tract: the role of alveolar macrophages. Clin. Exp. Immunol. 63, 261-70. Holt. P. G. and Schon-Hegrad. M. A. (1987). Localisation of T cells. macrophages and dendritic cells in rat respiratory tract tissue: implication for immune function studies. Immunology 62, 349-56. Holt. P. G.. Schon-Hegrad, M. A. and McMenamin. P. G. ( 1990). DCs in the respiratory tract. Intern. Kev. hnuno/. 6, 13949. Holt. P. G., Schon-Hegrad, M. A. and Oliver, J. (1988). MHC class II antigen-bearing dendritic cells in the pulmonary tissues of the rat. Regulation of antigen presentation activity by endogenous macrophage populations. 1. Exp. Med. 167. 262-74. Holt, P. G., Schon-Hegrad. M. A., Phillips, M. J. and McMenamin, P. G. ( 1989). Ia-positive dendritic cells form a tightly meshed network within the human airway epithelium. Clin. Exp. Allergy. 19. 597-601. Honda. H.. Kimura. H. and Rostami, A. (1990). Demonstration and phenotypic characterisation of resident macrophages in rat skeletal muscle. Immunology 70, 2 72-7. Hylkema, H. A., Broersma, L. and Kijlstra, A. (1988). In vivo and in vitro deposition of immunoglobulin aggregates in the mouse eye. Cut-r. Eye Res. 7. 593-9. Inomata, H.. Bill, A. and Smelser, G. K. (1977). Unconventional routes of aqueous humor outflow in cynomolgous monkey (Maraca irus). Am. J. Ophthalmol. 73, 893-907. Jefferies. W. A., Green, J. R. and Williams, A. F. (1987). Authentic T helper CD4 (W3/25) antigen on rat peritoneal macrophages. 1. Exp. Med. 162, 117-27. Kraal. G. (1989). Immunocytochemistry of dendritic cells. A clue to their function? Res. Immunol. 140, 891-5. Langhoff, E. and Steinman. R. M. (1989). Clonal expansion of human T lymphocytes initiated by dendritic cells. J. Exp. Med. 169. 315-20. Latina. M.. Flotte. T.. Crean. E., Sherwood, M. E. and Granstein, R. D. (1988). Immunohistochemical staining of the human anterior segment. Evidence that resident cells play a role in immunologic responses. Arrh. Ophthulmol. 106, 95-9. Lynch, M. G.. Peeler, J. S.. Brown, R. H. and Niederkorn, J. Y. ( 19 8 7 ). Expression of HLA class I and II antigens on cells of the trabecular meshwork. OphthalmoIogy 94, 851-i. MacPherson. G. G.. Fossum, S. and Harrison, B. (1989). Properties of lymph-borne cells in culture. II. Expression of the IL-2 receptor: role of GM-CSF. immunology 68, 108-l 3. McMaster, W. R. and Williams, A. F. ( 19 79). Identification of Ia glycoproteins in the rat thymus and purification from rat spleen. Eur. J. Immunol. 9. 426-33. McMenamin, P. G. and Al-Shakarchi, M. J. (1989). The effect of various levels of intraocular pressure on the rat aqueous humor outflow system. J. Anat. 162, 67-82. Niederkorn. J. Y. and Streilein. J. W. (1983). Alloantigens placed into the anterior chamber of the eye to induce specific suppression of delayed type hypersensitivity but normal cytotoxic T lymphocyte and helper T lymphocyte responses. J, Irnmunol. 131. 26704. Rem& Ch.. Urner, U. and Aeberhard, B. (1983). The occurrence of cell death during the remodelling of the anterior chamber angle recess in the developing rat eye. Graefe’s Arch. Clin. Exp. Ophthalmol. 22 1, 113-2 1. Richardson, T. M.. Marks, M. S., Ausprunk. D. H. and Miller. M. ( 19 8 5 ). A morphological and morphometric analysis of the aqueous outflow system of the developing cat eye. Exp. Eye Rrs. 41. 31-51.
EYE
323 Robinson, A. P., White, T. M. and Mason. D. W. (1986). Macrophage heterogeneity in the rat as delineated by two monoclonal antibodies MRC OX-41 and MRC OX42, the latter recognizing complement receptor type 3. Immunology 57, 23947. Rodriguez, M. M., Rowden, G., Hackett, J. and Bakos, I. (1981). Langerhans cells in the normal conjunctiva and peripheral cornea of selected species. Invest. Ophthalmol. Vis. Sci. 21, 759-65. Rohen. J. W. and Liitjen-Drecoll, E. (1982). Biology of the trabecular meshwork. In Basic Aspects of Glaucoma Research (Ed. Liitjen-Drecoll. E.). Pp. 141-66. Schattauer Verlag: Stuttgart. Rohen, J. W. and van der Zypen, E. (1968). The phagocytic activity of the trabecular meshwork endothelium. Albrecht v. Graefes Arch. Klin. Exp. Ophthalmol. 175, 143-60. Romani, N., Inaba, K.. Purse, E. Crowley, M., Witmer-Pack. M. and Steinman, R. M. (1989). A small number of anti-CD3 molecules on dendritic cells stimulate DNA synthesis in mouse T lymphocytes. J. Exp. Med. 169, 1153-68. Romani. N. and Schuler, G. (1989). Structural and functional relationships between epidermal langerhans cells and dendritic cells. Res. Immunol. 140, 895-8. Schon-Hegrad, M., Oliver, J., McMenamin. P. G. and Holt, P. G. (1991). Studies on the density, distribution and surface phenotype of intraepithelial class II MHC (Ia)bearing dendritic cells (DC) in the conducting airways. J. Exp. Med. 173, 1345-56. Schwartz, R. H. (1990). A cell culture model for Tlymphocyte clonal anergy. Science 248. 1349-56. Sherman, S. H.. Green, K. and Laties, A. M. ( 19 78). The fate of anterior chamber fluoroscein in the monkey eye 1. The anterior chamber outflow pathways. Exp. Eye Res. 27, 159-73. Sontheimer, R. D. ( 1989). Perivascular dendritic macrophages as immunological constituents of the human dermal microvascular unit. J. Invest. Dermatol. 93 (Suppl.). 96-101. Sontheimer, R. D., Matsubara, T. and Seelig. L. L. (1989). A macrophage phenotype for a constitutive, class II antigen expressing, human dermal perivascular dendritic cell. J. Invest. Dermatol. 93. 154-9. Steinman, R. M.. Kaplan, G., Witmer, M. D. and Cohn, Z. (1979). Identification of a novel cell type in peripheral lymphoid organs of mice. V. Purification of spleen dendritic cells, new surface markers, and maintenance in vitro. J. Exp. Med. 149, l-16. Streilein, J. W. and Cousins, S. W. (1990). Aqueous humor factors and their effect on the immune response in the anterior chamber. Curr. Eye Res. 9 (Suppl.). 175-82. Streilein. J. W. and Niederkorn, J. Y. (1981). Induction of anterior chamber associated immune deviation requires an intact, functional spleen. J. Exp. Med. 153. 1058-67. Toews. G. B., Bergstresser, P. R., Streilein, J. W. and Sullivan, S. (1980). Epidermal Langerhans cell density determines whether contact hypersensitivity or unresponsiveness follows skin painting with DNFB. 1. Immunol. 124, 445-52. Tompsett, E.. Abi-Hanna, D. and Wakefield. D. (1990). Immunological privilege in the eye : a review. Curr. Eye Res. 9, 1141-5. Tripathi, B. J., Tripathi. R. C., Wong, P. and Raja. S. (1990). Expression of HLA by the human trabecular meshwork and cornea1 endothelium. Exp. Eye Rrs. 51, 269-76. Wang, H.. Kaplan, H. J., Chan. W. C. and Johnson, M. (198 7). The distribution and ontogeny of MHC antigens in murine ocular tissue. Invest. Ophthalmol. Vis. Sri. 28. 1383-9.
324 Whitsett, C. F. and Stulting, R. D. (1984). The distribution of HLA antigens on human cornea1 tissue. Invest. Ophthalmol. Vis. Sci. 25, 519-24. Williams, K. A. and Coster, D. J. (1989). The role of the limbus in cornea1 allograft rejection, Eye 3, 158-66. Williamson, J. S. P., Bradley, D. and Streilein, J. W. (1989). Immunoregulatory properties of bone marrow-derived cells in the iris and ciliary body. Immunology 67. 96-102. Williamson, J. S. P., DiMarco, S. and Streilein, J. W. (1987).
P. G. McMENAMlN
AND
I. HOLTHOUSE
Immunobiology of Langerhans cells on the ocular surface. I. Langerhans cells within the central cornea interfere with the induction of anterior chamber associated immune deviation. Invest. Ophthalmol. Vis. Sci. 28, 1527-32. Wodner-Filopowicz, A., Heusser, C. H. and Moroni, C. (1989). Production of the haemopoietic growth factors GM-CSF and interleukin-3 by mast cells in response to IgE receptor mediated activation. Nature 339, 150-2.
lmmunohistochemical Characterization Macrophages in the Aqueous Outflow P. G. McMENAMIN” Department
of Anatomy
(Received
AND
of Dendritic Cells and Pathways of the Rat Eye I. HOLTHOUSE
and Human Biology, The Centre for Human Biology, Australia, Nedlands, Perth, Western Australia 6009 Lund
75 July
7997 and accepted
in revised form
The University
73 November
of Western
7997)
Immunohistochemical studies were performed to determine the distribution, phenotype and ontogeny of macrophages and dendritic cells (DCs) in the aqueous humour oufflow pathways of the rat eye. Optimal fixation and indirect immunoperoxidase techniques were employed in conjunction with a panel of mAbs on tangential frozen sections of ocular tissues from a total of 3 7 Wistar Furth rats aged 12-13 days (n = 8), 3 weeks (n = 12), 7 weeks (n = 5) and 15 weeks (n = 12). The density ofimmunopositive cells was scored qualitatively. A moderate to low density of Ia+ cells with a dendritic morphology were observed in the trabecular meshwork. DCs were also identified in the suprachoroidal space and in the connective tissues of nerves and vessels piercing the sclera, i.e. in association with non-conventional aqueous o&low pathways. The phenotypical and morphological characteristics of these cells would indicate that they may potentially act as antigen presenting cells (APCs). Non-dendritic pleomorphic cells with a macrophage phenotype were also identified in the trabecular meshwork, and bipolar or elongated cells with a macrophage phenotype were a noticeable feature in the perivascular region of collector channels and the limbal episcleral veins. Some macrophage and DC-like cells were observed in intimate association with limbal mast cells. Theories on the mechanisms of Anterior Chamber Associated Immune Deviation (ACAID) have assumed APCs are largely absent from the tissues lining the anterior chamber. Our findings of a low but moderate density of putative APCs in the conventional and non-conventional aqueous humour oufflow pathways are discussed in relation to the various theories of ACAID. Key words: dendritic cell; macrophage; trabecular meshwork; aqueousoutflow pathway; ACAID: ocular immunology: classII; anterior chamber: rat eye.
1. Introduction Local antigen-induced immune responseshave various suppressive mechanisms that act to restrict lymphocyte activation/proliferation such as the generation of antigen-specific suppressor T-cells. Precise control mechanisms are particularly well developed in delicate organs such as the lungs, gut, brain and eye whose tissues could be functionally compromised by the non-specific ‘bystander ’ damage characteristic of
some cell-mediated
immune
responses.
eye ; and (4) a paucity or lack of Ia+ cells lining the anterior chamber (AC) (Niederkorn and Streilein, 1983 ; Abi-Hanna, Wakefield and Watkins, 198 7 ; Williamson, DiMarco and Streilein, 1987). The conventional and non-conventional aqueous outflow pathways are ideally suited to function as immunological filters of the aqueous humor (AqH). Indeed, it is well known that the trabecular meshwork acts as a self cleansing biological filter, removing cellular and extracellular debris from the AqH on its exit from the anterior chamber angle (for review see
Experimental intracameral injection of a wide variety of cell surface and soluble antigens elicits a
Rohen and Liitjen-Drecoll, 1982). The ubiquitous trabecular cells (TCs) that line the AqH filled inter-
variety of aberrant systemic immune responses, among which are antigen-specific suppression of delayed hypersensitivity reactions and normal cytotoxic and helper T-cell function (Niederkorn and Streilein, 1983). This phenomenon is termed ACAID
trabecular spaces phagocytose material which falls to pass through the lntertrabecular spaces en route to Schlemm’s canal. Whilst some studies have claimed there is a complete absence of MHC class II (Ia) antigen-bearing cells in the tissues lining the human anterior chamber (Abi-Hanna et al., 1987), there have been reports of small numbers of round Ia
(anterior
chamber-associated immune
deviation)
(Streilein and Niederkorn, 1981; for review see Tompsett, Abi-Hanna and Wakefield, 1990). A number of factors have historically been considered essential for the induction of ACAID. These include: (1) a route for antigen to reach the vascular system;
(2) an intact functional spleen (Streilein and Niederkorn, 198 1) ; (3) the anatomical integrity of the
cells (Bakker and Kijlstra, 198 5 ; Wang et al., 198 7) in the iris and ciliary body. Other studies (Williamson, Bradley and Streilein, 1989) have claimed to demonstrate significant numbers of F4/80+ macrophages in the mouse iris, of which one-third were Ia positive ; however, the morphological appearance of these cells
was not illustrated.
* Forcorrespondence. 00144835/92/080315+10
positive cells (Latina et al., 1988) or single scattered
%08.00/O
Elongated
cells, which
may
0 1992 AcademicPressLimited
P. G. McMENAMlN
316
represent macrophages (M@) or professional antigen presenting cells (APCs), have also been described in the trabecular meshwork and around Schlemm’s canal (Lynch et al., 1987; Latina et al., 1988). However, recently Tripathi et al. (1990) have claimed that all cells of the trabecular meshwork and Schlemm’s canal together with the entire cornea1 epithelium and endothelium of human eyes, are class II positive. This extent of expression is questionable in light of a considerable body of previous evidence that class II expression, for example in the epithelium, is limited to Langerhans cells (for example, Rodrigues et al., 1981; Gillette, Chandler and Greiner, 1982; Whitsett and Stulting, 1984; Williams and Coster, 1989). Ia+ cells in the trabecular meshwork could be involved in the sampling and presentation of an array of antigens that enter the AC, including both endogenous antigens (e.g. retinal S-antigen) and exogenous antigens (e.g. HSV-1 or experimentallyintroduced antigens). Previous immunohistochemical studies of the anterior segment have not provided convincing evidence that the Ia+ cells are dendritic in form (Bakker and Kijlstra, 1985; Bakker et al., 1986; Lynch et al., 1987; Latina et al., 1988). The coexistence of these two features are classically considered as prerequisites for classification as dendritic cells (DCs) or APCs (Steinman et al., 1979). Future determination of the functional capacity of DC populations in regulating immune responses in the anterior segment necessitates the acquisition of information on the distribution, phenotype and ontogeny of M+s and APCs in the AqH outflow pathways. The aim of the present study was to provide this information in the rat eye. The rat offers several advantages in terms of anatomical similarities of the AqH outflow pathways to humans (McMenamin and Al-Shakarchi, 1989) and the availability of a wide range of well characterized monoclonal antibodies (mAb). In previous studies of the respiratory tract, the detection of an intraepithelial Ia+ DC network has been dependent on novel methods of cell surface antigen preservation and appropriate sectioning plane in which planar views of the epithelium are obtained (Holt et al., 1989 ; Holt, Schon-Hegrad and McMenamin, 1990; Schon-Hegrad et al., 1991). Similar techniques of optimal fixation and sectioning were employed in the present investigation of ocular tissue.
The mAb employed in this study and their specificities are briefly summarized in Table 1. Monoclonal antibodies EDl, ED2, ED3, ED8 and ED9 were kindly supplied by Dr C. D. Dijkstra, Department of Histology, Vrije University, Amsterdam, The Netherlands. 0X-6, 0X-19, 0X-41, OX-42 and W3/25 were obtained from commercial sources (Serotec, Oxford, IJ.K.). The mAbs supplied as ascites were diluted to optimal levels in PBS (pH 7.4-7.6, 290-320 mosmol). Tissue Preparation and Immunoperoxidase Staining The tissue fixation and sectioning methods have been described previously (Holt et al., 1990). In summary, the anaesthetized animals were perfused with heparinized PBS followed by cold absolute ethanol. This served to prevent antigen diiusion and remove all blood, (a source of pseudoperoxidase activity) from the vascular bed. Eyes and lymph nodes TABLE I Monoclonal antibody specificities MoAB ED1
ED2
ED3
ED8
ED9 OX-6
and Methods
Animals
Female Wistar Furth rats (specific pathogen free) were obtained from the Animal Resource Centre, Murdoch University, Western Australia. These consisted of eight 12-13 day old rats (just before eye opening), five 7 week and 12 each of 3 and 15 week old rats (total 37).
I. HOLTHOUSE
Monoclonal Antibodies
ox-41 2. Materials
AND
OX-42 ox-19 W3/25
Specificity
___-
Cytoplasmic antigen in most rat monocytes. macrophages and dendritic cells Membrane antigen on connective tissue macrophage subpopulations in lymphoid and nonlymphoid organs Membrane antigen of macrophages mostly confined to lymphoid organs ; small DC subpopulations CDll/CD18 antigen 95 kDa protein (probably related to C3bi receptor) microglia, Langerhans cells, IDCs, some macrophages Similar (but wider) distribution to ED8 Ia molecule: DCs, some macrophages 110000-120000 MW surface protein on macrophage subpopulations and glial cells Rat C3bi receptor (CDll/CDlS) 95 kDa protein ( - RD8) CD5 T-cells CD4 T-cells some dendritic cells and macrophages
Reference Dijkstra et al. (1985) Dijkstra et al. (1985), Damoiseaux et al. (1989b) Dijkstra et al. (1985h Damoiseaux et al. (1989b) Damoiseaux et al. (1989b)
Damoiseaux et al. (1989b) McMaster and Williams (19 79) Robinson, White and Mason (1986) Robinson et al. (1986) Dallman et al. (1984) Jefferies et al. (1987)
DENDRlTlC
CELLS
AND
MACROPHAGES
IN THE
RAT
EYE
were removed and placed in absolute ethanol overnight. This was followed by rehydration and inilltration with a PBS/OCT mixture (Tissue Tek II freezing medium, Miles Laboratories, IN, U.S.A.). Tissues were embedded in OCT for cryostat sectioning (10 pm). The eyes were orientated to produce tangential sections which contained larger areas of trabecular meshwork and limbal tissue than conventional meridional sections [Figs l(A) and (B)]. A standard indirect immunoperoxidase procedure using mouse mAb (see Table I), biotinylated sheep anti-mouse and streptavidin-horseradish peroxidase (Amersham Laboratories, Buckinghamshire, U.K.) was used in this study. The horseradish peroxidase was visualised using 3,3 diaminobenzidine tetrahydrochloride (Sigma Chemical Co., St Louis, MO, U.S.A.) (12 mg per 10ml PBS) and 5~1 H,O, (30% by volume). Endogenous peroxidase activity was not blocked as it was found to compromise antigenicity. Cells displaying endogenous peroxidase activity were easily distinguishable from immunostained cells. Sections in which the primary mAb was omitted and in which an inappropriate antibody (OKT4), directed against the human CD4 marker was substituted, acted as negative controls.. Positive control tissue was stained in paraliel and consisted of lymph node (for all mAbs) and conjunctival Langerhans cells [for anti-Ia antibodies, i.e. 0X-6, see Fig. 2(A)]. Sections were lightly counterstained with haematoxylin before mounting. Qualitative Analysis
Due to the size and irregular nature of the tissue involved, quantitative analysis such as that usually performed on DC networks in epithelia was not considered appropriate. A qualitative investigation of staining patterns was made by light microscopy. Positive staining cells were classified as DCs if they possessed one or more large slender cytoplasmic processes. All others were classiiled as non-dendritic cells (NDCs). The density of these cells was scored
TABLE
317
FIG. 1. A, Conventional meridional section of the rat anterior segment. Note the small area of trabecular meshwork (TM) available for examination. The large episcleralvesselscut in transversesectionare indicatedby small arrows. B, Lower power micrograph of oblique or tangential frozen section of the anterior segmentwhich clearly demonstratesthe greater area of trabecular tissue (TM) availablefor study. In additionobliqueandlongitudinal profiles of the episcleralvessels(small arrows) are also presentin such sections.The dark cellsin the conjunctival laminapropria aroundthesevessels are mastcells(Toluidine blue stain). (A) x200; (B) x85.
qualitatively as ‘ + + + ’ (extremely high), ‘ + + ’ (moderate, approximately two to four cells per high power field), ‘ + ’ (low, about one cell per high power
II
Age-relatedchangesin the distribution andfrequency of ClassII (Ia) positive cells in the rat aqueousoutflow pathways*. Conventional pathways include trabecular meshwork (TM), Schlemm’s canal (SC) and episcleral veins (ESV); nonconventional routes include suprachoroidalspace(SCS) and scleral vessels(SV)
3 week
12-13 day __.
*Density
7 week
15 week
Tissue location
DC
NDC
DC
NDC
DC
NDC
DC
NDC
TM SC ESV scs sv
1 i c 5 +
ItI + 4 -
-t * + + .
k IL + f .
?E + + + .
+ rfr + + .
+ zk ++ f+ ++
+ + ++ +
scored as in Materials
and Methods.
P. G. McMENAMlN
318
FIG. 2. For legend see opposite.
AND
I. HOLTHOUSE
DENDRITIC
CELLS
AND MACROPHAGES
IN THE RAT EYE
319
TABLEIII
Thedistribution* of immunogositivecells in
the various regionsof the rat outflow pathways in 3 and 15 week old animals. Seelegendsto Tables1 and ZZfor abbreviations
OX-6
Tissue location
ED1
ED8
ED2
W3/25 _____ DC NDC
DC
NDC
DC
NDC
DC
NDC
DC
NDC
DC
NDC
Ik +
f +
-
+
+
++
-
+
-
+ -
-
-
k AI
f +
-
f
-
-
+
-
-
-_ -
-
+ ++
+ ++
+ -
++ -
++ +++
+ ++
+ -
-
+
++
+
-
-
TM 3 week 15 week SC 3 week 15 week ESV 3 week 15 week scs 3 week 15 week sv 15 week
OX-42
---+ + -
++ ++
++ +++
* +
f -
-
-
+ -
+_ -
+
-
-
f
+
-
5
+
++
*
.
.
.
.
* Densityscoredasdescribed in MaterialsandMethods.
field) and ‘ _+’ (rare or occasional, present but not in every section), ’ - ’ (absent). 3. Results The age-related distribution of Ia+ cells in the various components of the AqH outflow pathways of 12-13 day and 3, 7 and 15 week old animals is summarized in Table II. The distributions of the other mAb employed in this study for 3 and 15 week old animals are summarized in Table III. Conventional Outflow Pathway l’rabecuiur meshwork.Ta+ cells within the trabecular
meshwork demonstrated either true dendritic [Figs 2(B) and (C)] or elongate/bipolar morphology (i.e. DCs) or were NDCs [Fig. 2(D)]. The density of these cells increased with age (Table II) and they appeared to be associated with both the inner more robust connective tissue trabeculae (pectinate ligaments) and the outer more lamellate trabeculae. In the 15-week group, DCs also stained with ED1 and occasionally with BD2. NDCs stained positively for ED1 , ED2 [Fig. 2(D)], ED9, Ox-41 and occasionally for Ox-42 in both
age groups (ED9 and OX-41 were excluded from the 3week group because of excessive background staining). Canal of Schlemm and collector channels.Ia+ cells were located in small numbers adjacent Schlemm’s canal and the draining collector channels [Fig. 2(E)]. In longitudinal sections of the canal these round or ovoid cells were located in the connective tissue of the inner and outer walls (juxtacanalicular zone). They were observed to increase in number with age and were found to be 0X-6+, 0X-41’. 0X-42’. EDl+, ED2+ and ED8’. Episclerulveins. Ia+ cells were consistently located in association with the episcleral veins (ESV) in all age groups [Figs 2(E) and (G)]. These cells displayed both dendritic and non-dendritic morphologies in similar proportions. Longitudinal sections revealed that Ia+ cells were distributed along the length of the vessels in close proximity to the endothelial cells and pericytes [Fig. 2(G)]. An age-related increase in the density of these cells was observed. The cells around the episcleral vessels stained particularly strongly for ED2 and in varying densities using the mAbs 0X-42, EDl, ED8 and ED9. Through-
FIG. 2. A, Ia+ (0X-6) Langerhanscellswithin limbal conjunctival epithelium (sectionedparallel to the underlying basal lamina, i.e. plan view) actedasin situ positivecontrols for Ia immunostaining.B, Low power view of iridocornealanglein an obliquesection.Adjoining structures:ciliary processes (CP).sclera(S)and anterior chamber(AC).Note the presenceof Ia+ (OX6) immunopositivedendritic cellsin the trabecular tissue(TM), shown in greater detail in (C). D, ED2+non-dendritic cellsin the trabecular meshwork. Ciliary epithelium is identifiable. E, Perivascular Ia+ (0X-6) non-dendritic cells in the walls of Schlemm’scanalor outer trabecularmeshwork(TM), collector channels(smallarrows)and the episcleralveins(ESV).E-cornea1 epithelium. CP-ciliary processes.F, ED2+ bipolar DC-like cells in close associationwith mast cells (weakly stained with haematoxylin). G, NumerousIa+ elongatedbipolar perivascular macrophagesadjacent to episcleralvessels(ESV). H, Ja+ dendritic cells in the anterior suprachoroidalspace(SCS),ros-rod outer segments,arrows delineatechoroid. (A) x 600; (B) x210: (C) x 500: (D) x 600; (E) x280; (F) x 550; (G) x200: (H) x 300.
320
out the highly vascular conjunctival lamina propria, cells with a similar phenotype and morphology viz histiocytic/macrophagic were encountered. In the 3week-old animals similar cells were detected close to vessels and in the lamina propria using 0X-42, EDl, ED2. ED8 and ED9; however, the densities were generally lower. In the 15-week-old animals, 0X-6+, 0X-42’ and ED2+ bipolar DC-like cells were often located either in close proximity to or in contact with perivascular and connective tissue mast cells [Fig. 2(F)], which were particularly common in the limbal zone. Non-conventional Route Suprachoroidal space (SCS) and scleral vessels. Ia+ DCs were located in the outer choroid or SCS with dendritic processes entering both the choroid and sclera [Fig. 2(H)]. Occasional EDl+ DCs and NDCs were located in the SCS of both 3 and 15 week old animals. DCs seemed to be arranged in a two-dimensional network interposed between the choroid and sclera. The density of these cells increased with age (Tables II and III). Small numbers of Ia+ DCs which were located in the dense connective tissue of the sclera similarity showed an increase in number with age. However, close examination revealed that most of these cells lay adjacent to vessels or nerves that pierced the sclera. Small numbers of EDl+ and ED2+ DCs and NDCs were located around the large scleral vessels in some of the 15week animals. No Ia+ DCs were observed in the sclera of the 3-week animals as no intrascleral vessels or nerves were sectioned in these animals. Endogenous Peroxidase Activity Cells displaying endogenous peroxidase activity were rare within the eye and were confined mainly to the lamina propria of the limbal conjunctiva. These cells were readily identifiable from positively stained cells in other sections because of their consistent rounded shape and much more intense dark brown/ black staining. The pseudoperoxidase activity of erythrocytes was minimal as a consequence of the vascular perfusion/fixation protocol. 4. Discussion By using optimal pre-embedding fixation and appropriate sectioning techniques we have demonstrated la+ cells with both dendritic and non-dendritic morphologies related to the rat AqH outilow pathways and have further characterized these cells using a panel of mAb. Despite the current lack of a specific anti-rat DC mAb, satisfactory identification can be based on staining with an anti-Ia mAb and other M$ pan-specific mAb in conjunction with demonstration of a dendritic morphology.
P. G. McMENAMlN
AND
I. HOLTHOUSE
Distribution and Phenotype of Macrophage-likeCellsin the AqH Pathways
NDCs in the trabecular meshwork and in the walls of Schlemm’s canal displayed Ia (0X-6). ED1 , ED2. OX41 and occasionally OX-42 ; however, these cells were negative for ED3. The mAb ED1 recognizes peripheral blood monocytes and most M@, whilst ED2 recognizes resident tissue M@s of lymph node, spleen, thymus (Dijkstra et al., 1985; Damoiseaux et al., 1989a. b), skeletal muscle (Honda, Kimura and Rostami, 1990) and extraocular muscle (personal observation, P.McM). ED3 is restricted to M+ subpopulations in lymphoid organs and at sites of inflammation (Dijkstra et al., 1985). The other markers used (0X-6, 0X-41. 0X-42, ED8, ED9 and W3/25) have a highly variable frequency on M&s that may be related to microenvironmental factors or different stages of M$ maturation (Beelen et al., 1987; Damoiseaux et al., 1989a, b). On the basis of the morphological appearance and phenotype we conclude that the majority of NDCs in the trabecular meshwork and Schlemm’s canal are resident tissue macrophages. The role of the trabecular meshwork as a biological filter of the AqH places it in an optimal location for immune surveillance. It has been proposed. primarily on the basis of ultrastructural evidence, that following phagocytosis of debris TCs detach from the trabecular beams and migrate through the meshwork to Schlemm’s canal (Rohen and Van der Zypen, 1968 : Grierson and Lee, 1973). An alternative explanation is that some of the cells lining the intertrabecular spaces in the normal resting state are phagocytically active M$s, which have assumed residency in the tissue following their role in developmental remodelling of the iridocorneal angle (Rem& Urner and Aeberhard, 1983 : Richardson et al., 198 5). Our ontogeny data suggests that M&s may in fact continue to populate the anterior segment and perivascular connective tissue in the episclera until maturity (Table 111). The second population of cells which we believe to be macrophages are the NDC and elongated cells adjacent to the episcleral veins which were positive for all mAbs except 0X-41, ED3 and W3/25. Their morphology. strong reactivity with ED2 and extensive distribution throughout the conjunctival lamina propria is strong evidence that these cells are resident tissue n/r+ or histiocytes. Similar cells have been located in perivascular sites within the human adult CNS (Hickey and Kimura, 1988; Graeber, Streit and Kreutzberg, 1989) human foetal CNS (Esiri, Al Izzi and Reading, 1991), dermis (Sontheimer, 1989 ; Sontheimer, Matsubara and Seelig, 1989) and skeletal muscle (Honda et al., 1990). Some of these perivascular M$s in other tissues have also demonstrated class II staining similar to those around episcleral veins, suggesting a potential role as antigen presenting cells. Indeed some of these cells in the episcleral veins and surrounding limbal/conjunctival lamina propria
DENDRITIC
CELLS
AND
MACROPHAGES
IN THE
RAT
may represent precursors of conjunctival and cornea1 Langerhans cells (for review see Williams and Coster, 1989). Interestingly, tracer studies (Sherman, Green and Laties, 1978) have shown that proteins injected into the AC leak from episcleral limbal vessels, thus it is feasible that class II bearing cells around collector channels and episcleral veins have direct access to intracameral antigens. These cells could travel to the local lymph nodes via draining conjunctival lymphatics. thus bypassing the camero-splenic axis. Alternatively, they may be active in phagocytosis and clearance of immune complexes which, in the mouse, have been shown to accumulate around these limbal vessels (Hylkema, Broersma and Kijlstra, 1988). These cells may therefore play a role in limiting the inflammatory responses which emanate from immune complex deposition. The increase in frequency of class IT positive dendritic and non-dendritic cells around the episcleral veins and limbal/conjunctival lamina propria at 3 weeks, 1 week after eye opening (day 14) suggests that this ontogenic process may be antigen driven; however, further experimental evidence would be required to support this proposal. Perivascular and connective tissue M+s were commonly observed in intimate contact with conjunctival mast cells. An intriguing possibility is that the high numbers of mast cells in the limbus may influence DC and M+ maturation via the production of cytokines such as IL-3 and GM-CSF (Wodnor-Filipowicz, Heusser and Moroni, 1989: for review see Gordon, Burd and Galli, 1990). It has recently been shown that GM-CSF upregulates the expression of IL-2 receptors on lymphborne (veiled) dendritic cells which may potentially play a role in increasing the efficiency of T cell activation (MacPherson, Fossum and Harrison, 1989 I. Phenotype of DC-iike Cells Class II expression and dendritic morphology are classically considered as the hallmarks of professional APCs or DCs (Steinman et al., 1979; Kraal. 1989). To our knowledge, this is the first demonstration of cells in the mammalian aqueous humour outflow pathways which fulfill both these criteria and may therefore represent a population of antigen presenting cells. Small numbers of DC-like cells in these locations expressed ED1 which is known to recognize DC subpopulations (Damoiseaux et al., 1989b; Holt and Schon-Hegrad, 1987). One of the critical factors thought to play a role in ACAID induction was the lack of la+ cells lining the AC (Williamson et al., 1987). Clearly, in light of our evidence of a low to moderate density of Ia+ DC population in the trabecular meshwork this hypothesis must be reevaluated, particularly since it is known that as few as one DC is necessary to produce a mixed lymphocyte response (Langhoff and Steinman, 1989: Romani et al., 1989). The failure of previous studies to demonstrate class II antigen-bearing dendritic shaped cells 31
EYE
321
in the human and non-human anterior segment is most likely related to modes of fixation, section orientation and immunohistochemical techniques. Role of la+ Cells of the AqH Outflow Pathways In areas such as the skin and oral mucosa, la+ Langerhans cells permit sensitization to a variety of antigens. In contrast, experimental exposure of the AC to antigens results in a suppressive immune reaction (ACAID) (Streilein and Niederkorn, 1981). The induction of ACAID by resident Ia+ cells in the rat eye may occur via a number of mechanisms. These include: (1) the presence of ‘alternative ’ APCs that activate suppressor T-cells rather than helper T-cells; (2) ‘faulty ’ presentation of antigen by both bone marrow and non-bone marrow derived cells; and (3) the presence of T-cell suppressive factors and M$s. It has been suggested that an alternative population of ‘I-J’ positive APCs exist which preferentially elicit the generation of antigen specific suppressor T-cell activity (Toews et al., 1980 ; Granstein and Greene, 1985) ; however, the existance of such ce1l.sremains highly controversial. Aberrantly Ia+ non-bone marrow derived cells (e.g. keratinocytes, fibroblasts) or la+ M$s and ,4PCs that lack the ability to deliver antigen non-specific T-cell co-stimulatory signals such as IL-1 are capable of producing a state of T-cell unresponsiveness or anergy : such T-cells are effectively ’ paralysed ’ to further antigen presentation by professional APCs (for review see Geppert and Lipsky, 1989 ; Schwartz, 1990). Clearly there is potential for Ia+ cells in the eye to produce T-cell unresponsiveness, a possibility supported by recent evidence that introduction of IL-l into the AC of mice was able to abrogate ACAID (Benson and Niederkorn, 1990). A fruitful area of future study would be the investigation of the ability of anterior segment-derived DCs to activate T-cells and/or produce antigen non-specific signals such as the interleukins. Normally, APCs such as Langerhans cells require a maturational step stimulated by factors present within lymph and lymph nodes, such as IL-1 and GM-CSF, to enhance their ability to activate resting T-cells (for review see Romani and Schuler, 1989). The absence of a lymphatic drainage pathway in the eye may prevent intraocular APCs from receiving these maturation signals and they may therefore induce T-cell paralysis or the induction of suppressor T-cells in the spleen in response to antigen presentation. In organs such as the lung, resident alveolar M+s play a role in the suppression of T-cell proliferation and activation (for review see Halt, 1986 : Holt, Schon-Hegrad and Oliver, 1988). Suppression of T-cell mediated reactions, particularly delayed hypersensitivity reactions, is likely to have the physiological function of limiting inflammatory responses in organs whose function could easily be compromised by nonliEP 55
P. G. McMENAMlN
322
specific ‘bystander’ damage. In the eye, suppressive M$s could possibly function similarly to alveolar M$s, preventing T-cell activation by resident Ia+ APCs and thereby restricting destructive inflammatory reactions. Indeed, recently a population of M$s (one-third of which were Ia+) has been reported in the mouse ciliary body and iris stroma (Williamson et al., 19 8 9 ; Streilein and Cousins, 1990). Whole iris/ciliary body tissue explants were found to be incapable of activating allogeneic T-cells, and thus assumed to lack APCs. Furthermore, the explants were found to suppress Tcell proliferation in culture, and preliminary evidence suggests that transforming growth factor-beta (identified in aqueous humour) may be one suppressive factor in the tissue explant supernatant ; however, the cellular source(s) of such suppressive factors has not been identified. Non-conventional Outflow Pathways A proposed route of egress of AqH and large molecules (up to 1 pm) from the AC is posteriorly into the SCS and then escaping either via the venous system (uveo-vortex) or in the perivascular and perineural spaces in the sclera (uveo-scleral) (Inomata, Bill and Smelser, 1977). In the present study we have shown for the first time that in the rat Ia’ DC-like cells are located in the SCS and adjacent nerves and vessels that pierce the sclera. The relative importance of conventional versus non-conventional routes is highly species dependant and the requirement for a more complex network of APCs in the non-conventional routes may be more important in lower mammals such as the rat. Further studies will be necessary to reevaluate whether such cells exist in primate and human eyes. On the basis of our findings it is evident that previously undescribed populations of MI.$s and DCs exist within the normal rat anterior segment and AqH outflow pathways (conventional and non-conventional) and these cells increase with age up to maturity. This is in addition to a further population of DCs which we have observed within the ciliary body epithelium and iris stroma (unpubl. res.). Further in vitro
studies
of the capacity
of DCs from the outflow
tissues (and other anterior segment tissues) to process and present antigen, and their interactions with Tcells are required to elucidate their role in the induction of ACAID. We have also characterized the normal macrophage/histocyte population which is present
in
the
trabecular
meshwork
and
in
the
perivascular connective tissue of Schlemm’s canal, collector channels and the episcleral vessels. Some of these may also represent immature DCs and the intimate relation between those cells in the episcleral region and limbal mast cells may be a novel observation of importance.
AND
I, HOLTHOUSE
Acknowledgements We would like to acknowledge the valuable advice of Dr P. G. Holt of the Western Australian Research Institute for Child Health (WARICH) during the preparation of this manuscript.
References Abi-Hanna. D., Wakefield, D. and Watkins. S. (1987). HLA antigens in ocular tissue. I. In vivo expression in human eyes. Transplantation 45. 610-3. Bakker, M., Grumet, F. G., Feltkamp, T. E. W. and Kijlstra. A. (1986). HLA-antigens in the human uvea. Dot. Ophthalmologica 61, 271-9. Bakker, M. and Kijlstra, A. (1985). The expression of HLAantigens in the human anterior uvea. Curr. E;{e Res. 4, 599-604. Beelen. R. H. J.. Eestermans. I. L., Dopp. E. A. and Dijkstra. C. D. (1987). Monoclonal antibodies EDI, ED2 and ED3 against rat macrophages : Expression of recognised antigens in different stages of differentiation. TranspI. Proc. 19, 3166-70. Benson, J. L. and Niederkorn. J. Y. ( 1990). Interleukin-1 abrogates anterior chamber-associated immune deviation. Invesf. OphthaZmo[. Vis. Sci. 31, 2 12 3-8. Dallman, M. J., Matthew, I,. T. and Green, J, R. (1984). MKC OX-19 : a monoclonal antibody that labels rat T lymphocytes and augments in vitro proliferative responses. Eur. 1. lmmunol. 14, 260-7. Damoiseaux, J. G.. Dopp, E. A.. Beelen, R. H. J. and Dijkstra. C. D. (1989a). Rat bone marrow and monocyte cultures: influence of culture time and lymphokines on the expression of macrophage differentiation antigens. J. Leukocyte Biol. 46. 246-53. Damoiseaux. J. G., Dopp, E. A.. Neeljes, ]. 1.. Beelen, R. H. J. and Dijkstra. C. D. (1989b). Heterogeneity of macrophages in the rat evidenced by variability in determinants: two new anti-rat macrophage antibodies against a heterodimer of 160 and 95 kd (CD1 l/CD1 8). 1. Leukoc$e BioI. 46, 556-64. Dijkstra. C. D., Dopp, E. A., Joling. P. and Kraal. G. ( 198 5). The heterogeneity of mononuclear phagocytes in lymphoid organs : distinct macrophage subpopulations in the rat recognised by monoclonal antibodies EDI. ED2 and ED3. Immunology 54. 589-99. Esiri, M. M.. Al Izzi, M. S. and Reading, M. C. ( 199 1). Macrophages, microglial cells and HLA-DR antigens in fetal and infant brain. J. CJin. Pathol. 44, 102-6. Geppert, T. D. and Lipsky. P. E. (1989). Antigen presentation by cells that are not of bone marrow origin. Reg. Immunol. 2. 60-71. Gillette, T. E.. Chandler. J. W. and Greiner, J. V. ( lY82). Langerhans cells of the ocular surface. Ophthalmology 89. 700-l 1. Gordon. j. R., Burd, P. R. and Galli, S. (1990). Mast cells as a source of multifunctional cytokines. Irnmunol. Today Il. 458-64. Graeber, M. B., Streit, W. J. and Kreutzberg, G. W. (1989). Identity of EDZ-positive perivascular cells in the rat brain. I. Neurosci. Res. 22, 103-6. Granstein. R. D. and Greene, M. I. (1985). Splenic I-Jbearing antigen-presenting cells in activation of suppression: further characterisation. Cell. Immunol. 91, 12-20. Grierson, I. and Lee, W. R. (1973). Erythrocyte phagocytosis in the human trabecular meshwork. B. 1. OphthalmoL 29, 400-l 5. Hickey, W. F. and Kimura, H. (1988 ). Perivascular microglial
DENDRITIC
CELLS
AND
MACROPHAGES
IN THE
RAT
cells of the CNS are bone marrow-derived and present antigen in vivo. Science 239, 290-2. Holt, P. G. (1986). Down-regulation of immune responses in the lower respiratory tract: the role of alveolar macrophages. Clin. Exp. Immunol. 63, 261-70. Holt. P. G. and Schon-Hegrad. M. A. (1987). Localisation of T cells. macrophages and dendritic cells in rat respiratory tract tissue: implication for immune function studies. Immunology 62, 349-56. Holt. P. G.. Schon-Hegrad, M. A. and McMenamin. P. G. ( 1990). DCs in the respiratory tract. Intern. Kev. hnuno/. 6, 13949. Holt. P. G., Schon-Hegrad, M. A. and Oliver, J. (1988). MHC class II antigen-bearing dendritic cells in the pulmonary tissues of the rat. Regulation of antigen presentation activity by endogenous macrophage populations. 1. Exp. Med. 167. 262-74. Holt, P. G., Schon-Hegrad. M. A., Phillips, M. J. and McMenamin, P. G. ( 1989). Ia-positive dendritic cells form a tightly meshed network within the human airway epithelium. Clin. Exp. Allergy. 19. 597-601. Honda. H.. Kimura. H. and Rostami, A. (1990). Demonstration and phenotypic characterisation of resident macrophages in rat skeletal muscle. Immunology 70, 2 72-7. Hylkema, H. A., Broersma, L. and Kijlstra, A. (1988). In vivo and in vitro deposition of immunoglobulin aggregates in the mouse eye. Cut-r. Eye Res. 7. 593-9. Inomata, H.. Bill, A. and Smelser, G. K. (1977). Unconventional routes of aqueous humor outflow in cynomolgous monkey (Maraca irus). Am. J. Ophthalmol. 73, 893-907. Jefferies. W. A., Green, J. R. and Williams, A. F. (1987). Authentic T helper CD4 (W3/25) antigen on rat peritoneal macrophages. 1. Exp. Med. 162, 117-27. Kraal. G. (1989). Immunocytochemistry of dendritic cells. A clue to their function? Res. Immunol. 140, 891-5. Langhoff, E. and Steinman. R. M. (1989). Clonal expansion of human T lymphocytes initiated by dendritic cells. J. Exp. Med. 169. 315-20. Latina. M.. Flotte. T.. Crean. E., Sherwood, M. E. and Granstein, R. D. (1988). Immunohistochemical staining of the human anterior segment. Evidence that resident cells play a role in immunologic responses. Arrh. Ophthulmol. 106, 95-9. Lynch, M. G.. Peeler, J. S.. Brown, R. H. and Niederkorn, J. Y. ( 19 8 7 ). Expression of HLA class I and II antigens on cells of the trabecular meshwork. OphthalmoIogy 94, 851-i. MacPherson. G. G.. Fossum, S. and Harrison, B. (1989). Properties of lymph-borne cells in culture. II. Expression of the IL-2 receptor: role of GM-CSF. immunology 68, 108-l 3. McMaster, W. R. and Williams, A. F. ( 19 79). Identification of Ia glycoproteins in the rat thymus and purification from rat spleen. Eur. J. Immunol. 9. 426-33. McMenamin, P. G. and Al-Shakarchi, M. J. (1989). The effect of various levels of intraocular pressure on the rat aqueous humor outflow system. J. Anat. 162, 67-82. Niederkorn. J. Y. and Streilein. J. W. (1983). Alloantigens placed into the anterior chamber of the eye to induce specific suppression of delayed type hypersensitivity but normal cytotoxic T lymphocyte and helper T lymphocyte responses. J, Irnmunol. 131. 26704. Rem& Ch.. Urner, U. and Aeberhard, B. (1983). The occurrence of cell death during the remodelling of the anterior chamber angle recess in the developing rat eye. Graefe’s Arch. Clin. Exp. Ophthalmol. 22 1, 113-2 1. Richardson, T. M.. Marks, M. S., Ausprunk. D. H. and Miller. M. ( 19 8 5 ). A morphological and morphometric analysis of the aqueous outflow system of the developing cat eye. Exp. Eye Rrs. 41. 31-51.
EYE
323 Robinson, A. P., White, T. M. and Mason. D. W. (1986). Macrophage heterogeneity in the rat as delineated by two monoclonal antibodies MRC OX-41 and MRC OX42, the latter recognizing complement receptor type 3. Immunology 57, 23947. Rodriguez, M. M., Rowden, G., Hackett, J. and Bakos, I. (1981). Langerhans cells in the normal conjunctiva and peripheral cornea of selected species. Invest. Ophthalmol. Vis. Sci. 21, 759-65. Rohen. J. W. and Liitjen-Drecoll, E. (1982). Biology of the trabecular meshwork. In Basic Aspects of Glaucoma Research (Ed. Liitjen-Drecoll. E.). Pp. 141-66. Schattauer Verlag: Stuttgart. Rohen, J. W. and van der Zypen, E. (1968). The phagocytic activity of the trabecular meshwork endothelium. Albrecht v. Graefes Arch. Klin. Exp. Ophthalmol. 175, 143-60. Romani, N., Inaba, K.. Purse, E. Crowley, M., Witmer-Pack. M. and Steinman, R. M. (1989). A small number of anti-CD3 molecules on dendritic cells stimulate DNA synthesis in mouse T lymphocytes. J. Exp. Med. 169, 1153-68. Romani. N. and Schuler, G. (1989). Structural and functional relationships between epidermal langerhans cells and dendritic cells. Res. Immunol. 140, 895-8. Schon-Hegrad, M., Oliver, J., McMenamin. P. G. and Holt, P. G. (1991). Studies on the density, distribution and surface phenotype of intraepithelial class II MHC (Ia)bearing dendritic cells (DC) in the conducting airways. J. Exp. Med. 173, 1345-56. Schwartz, R. H. (1990). A cell culture model for Tlymphocyte clonal anergy. Science 248. 1349-56. Sherman, S. H.. Green, K. and Laties, A. M. ( 19 78). The fate of anterior chamber fluoroscein in the monkey eye 1. The anterior chamber outflow pathways. Exp. Eye Res. 27, 159-73. Sontheimer, R. D. ( 1989). Perivascular dendritic macrophages as immunological constituents of the human dermal microvascular unit. J. Invest. Dermatol. 93 (Suppl.). 96-101. Sontheimer, R. D., Matsubara, T. and Seelig. L. L. (1989). A macrophage phenotype for a constitutive, class II antigen expressing, human dermal perivascular dendritic cell. J. Invest. Dermatol. 93. 154-9. Steinman, R. M.. Kaplan, G., Witmer, M. D. and Cohn, Z. (1979). Identification of a novel cell type in peripheral lymphoid organs of mice. V. Purification of spleen dendritic cells, new surface markers, and maintenance in vitro. J. Exp. Med. 149, l-16. Streilein, J. W. and Cousins, S. W. (1990). Aqueous humor factors and their effect on the immune response in the anterior chamber. Curr. Eye Res. 9 (Suppl.). 175-82. Streilein. J. W. and Niederkorn, J. Y. (1981). Induction of anterior chamber associated immune deviation requires an intact, functional spleen. J. Exp. Med. 153. 1058-67. Toews. G. B., Bergstresser, P. R., Streilein, J. W. and Sullivan, S. (1980). Epidermal Langerhans cell density determines whether contact hypersensitivity or unresponsiveness follows skin painting with DNFB. 1. Immunol. 124, 445-52. Tompsett, E.. Abi-Hanna, D. and Wakefield. D. (1990). Immunological privilege in the eye : a review. Curr. Eye Res. 9, 1141-5. Tripathi, B. J., Tripathi. R. C., Wong, P. and Raja. S. (1990). Expression of HLA by the human trabecular meshwork and cornea1 endothelium. Exp. Eye Rrs. 51, 269-76. Wang, H.. Kaplan, H. J., Chan. W. C. and Johnson, M. (198 7). The distribution and ontogeny of MHC antigens in murine ocular tissue. Invest. Ophthalmol. Vis. Sri. 28. 1383-9.
324 Whitsett, C. F. and Stulting, R. D. (1984). The distribution of HLA antigens on human cornea1 tissue. Invest. Ophthalmol. Vis. Sci. 25, 519-24. Williams, K. A. and Coster, D. J. (1989). The role of the limbus in cornea1 allograft rejection, Eye 3, 158-66. Williamson, J. S. P., Bradley, D. and Streilein, J. W. (1989). Immunoregulatory properties of bone marrow-derived cells in the iris and ciliary body. Immunology 67. 96-102. Williamson, J. S. P., DiMarco, S. and Streilein, J. W. (1987).
P. G. McMENAMlN
AND
I. HOLTHOUSE
Immunobiology of Langerhans cells on the ocular surface. I. Langerhans cells within the central cornea interfere with the induction of anterior chamber associated immune deviation. Invest. Ophthalmol. Vis. Sci. 28, 1527-32. Wodner-Filopowicz, A., Heusser, C. H. and Moroni, C. (1989). Production of the haemopoietic growth factors GM-CSF and interleukin-3 by mast cells in response to IgE receptor mediated activation. Nature 339, 150-2.