An intravital and confocal microscopic study of the distribution of intracameral antigen in the aqueous outflow pathways and limbus of the rat eye

Experimental Eye Research 79 (2004) 455–464 www.elsevier.com/locate/yexer

An intravital and confocal microscopic study of the distribution of intracameral antigen in the aqueous outflow pathways and limbus of the rat eye Serge Camelo, Adam C. Shanley, Angel S.P. Voon, Paul G. McMenamin* School of Anatomy and Human Biology, The University of Western Australia, 35 Stirling Highway, Crawley, Perth 6009, Western Australia Received 7 January 2004; accepted in revised form 13 May 2004 Available online 04 August 2004

Abstract In a previous investigation into the fate of fluorescently labelled antigen (Ag) injected into the anterior chamber (AC) of the rat eye, a large number of Agþ cells were noted in the conventional and non-conventional aqueous humour outflow pathways together with the external limbus. The aim of this study was to investigate the precise distribution and phenotype of these cells and compare their ability to capture fluorescent-labelled protein (bovine serum albumin, BSA, and ovalbumin, OVA) and polysaccharides (dextran, Dx) injected into the AC. The density of Agþ cells in the iris and limbus was investigated using in vivo video fluorescence microscopy 24 hr post-injection. The distribution and phenotype of Agþ cells in ocular tissues was analysed by confocal microscopy of frozen sections and in iris and corneoscleral/limbal wholemounts from animals sacrificed 24 hr post injection. The general distribution of labelled Ag was equivalent in OVA, BSA and Dx injected animals. Antigen-bearing cells were observed within the iris, iridocorneal angle, pre-equatorial choroid and around limbal/episcleral vessels. Localization of Agþ cells and free Ag in the anterior segment suggests that substances of these molecular weights (40 – 70 kDa) leave the eye through the conventional and non-conventional aqueous outflow pathways. The cells that internalized BSA, OVA or Dx in ocular tissues were of a similar phenotype, namely, ED1þ, ED2þ, occasionally ED3þ and predominantly MHC class II2, thus suggesting that they are of the macrophage phenotype. However, a few Agþ MHC class IIþ dendriform cells (putative DC) were also observed in the iris, trabecular meshwork, choroid and episclera. In conclusion our data reveal that the majority of intracamerally injected soluble Ag retained in the eye is taken up by resident macrophages not only in the iris but in all tissues lining the AC of the eye. q 2004 Elsevier Ltd. All rights reserved. Keywords: dendritic cells; macrophages; anterior chamber; aqueous outflow pathways; intravital microscopy; confocal microscopy

1. Introduction Experimental injection of antigen (Ag) into the anterior chamber (AC) of the eye induces a systemic state of tolerance known as Anterior Chamber Associated Immune Deviation (ACAID). This phenomenon is characterized by the suppression of delayed type hypersensitivity (DTH) responses to the same Ag injected subcutaneously (Streilein, 2003). It has been postulated that bone marrow-derived cells internalising Ag in the AC of the eye carry the ACAID-inducing signal via the blood to the spleen where populations of regulatory CD4þ and CD8þ T cells are generated (Wilbanks and Streilein, 1991; Wilbanks et al., * Corresponding author. Professor Paul G. McMenamin, School of Anatomy and Human Biology, The University of Western Australia, 35 Stirling Highway, Crawley, Perth 6009, Western Australia. E-mail address: [email protected] (P.G. McMenamin). 0014-4835/$ - see front matter q 2004 Elsevier Ltd. All rights reserved. DOI:10.1016/j.exer.2004.06.009

1991). Dense macrophage and dendritic cell (DCs) networks have been described in the uveal tract in normal experimental animals, humans and in experimental models of ocular inflammation (Knisely et al., 1991; McMenamin and Holthouse, 1992; McMenamin et al., 1992, 1994; Steptoe et al., 1997; McMenamin, 1997, 1999). In the rat, DCs in the iris, ciliary body, choroid and aqueous outflow pathways of the eye assume a variety of shapes (pleomorphic or dendriform) and express MHC class II molecules and/or OX62 (anti-a E2-integrin) (Steptoe et al., 1996). By contrast, macrophages in these tissues express the marker ED2 (anti-scavenger receptor CD163) and are generally MHC class II2 (Forrester et al., 1994; Steptoe et al., 1996). We have demonstrated that 24 hr after an intracameral injection of fluorescent labelled dextran (Dx) macrophages are the predominant cells bearing-Ag in the iris (Camelo et al., 2003). In addition that study also

456

S. Camelo et al. / Experimental Eye Research 79 (2004) 455–464

revealed cells in the ciliary body, anterior suprachoroidal space and episcleral connective tissue that were able to capture Dx in vitro and in vivo. A recent similar report in mice reveal that at 6 hr post injection, protein Ag (ovalbumin, OVA) may also be internalised by DCs in the iris (Becker et al., 2003). Determining the exact nature of these Ag-capturing cells was one of the aims of the present study. Furthermore, we wished to precisely determine whether macrophages and DCs in the anterior segment differ in their capacity to internalise protein or polysaccharides Ag injected into the AC. Observations by in vivo video-microscopy and confocal microscopy of immuno-stained sections and iris and limbal wholemounts showed that following injection of bovine serum albumin (BSA), OVA and Dx the phenotype of protein or Dx-bearing cells was equivalent. In accordance with our observations of the iris, Ag-bearing cells in the ciliary body, aqueous outflow pathway, episcleral tissues and limbus are predominantly macrophages with the exception of a few putative DCs. The data suggests that macrophages located in tissues lining the AC and throughout the course of the conventional and non-conventional aqueous outflow pathways are the major type of Ag-capturing cells in the anterior segment of the eye.

2.2. Reagents and antibodies BSA, paraformaldehyde (PFA) and sodium pentobarbitone were purchased from Rhone Merieux, (Australia, Queensland), O.C.T Compound was obtained from Sakura (USA). Alcaine was obtained from Alcon (NSW, Australia). Purified monoclonal antibody (mAb) anti-Dinitrophhenyl (DNP, clone SPE-7) was obtained from Sigma Chemical Co (St Louis, MO, USA). Purified mAbs anti-CD68 (clone ED1), CD163 (scavenger receptor type B, clone ED2), CD169 (sialoadhesin, clone ED3), I-Ab (clone MRC OX6), anti-a-E2-integrin (clone MRC OX62) and biotinylated ED1, ED2 and OX6 were purchased from Serotec Ltd. (Oxford, UK). Purified mAb antiCD11b (clone WT.5) was purchased from pharmingen (Becton Dickinson, UK). FITC conjugated-streptavidin (emission 525 nm) was obtained from Amersham-Pharmacia (Sweden). Alexa Fluor 546 conjugated goat anti-mouse IgG (emission 572 nm), Alexa Fluor 488 conjugated BSA (66 kDa, emission 525 nm), FITC conjugated OVA (45 kDa, emission 525 nm), or lysine fixable FITC labeled Dextran (40 kDa, emission 525 nm) and Cascade Blue (CB) labeled Dextran (Dx 70 kDa, emission 420 nm) were purchased from Molecular Probes (Eugene, OR USA). 2.3. In vivo video fluorescent microscopy

2. Materials and methods 2.1. Animals and anaesthesia procedures Female Lewis rats, 8 – 11-weeks-old, obtained from Animal Resources Center (Murdoch University, Western Australia) were kept under pathogen free conditions in chaff-lined cages and housed in 12 hr day/night cycles (Animal House, University of Western Australia, Western Australia) with food (Stockfeeders RM2 Autoclaved rat and mouse diet; Animal Resources Center, Western Australia) and water supplied ad libitum. Immediately preceding and during intracameral injections animal were anaesthetized by inhalation of a mix of oxygen and nitrous oxide (4:1) and 1% halothane (ICI pharmaceuticals; Melbourne, Australia). The method of intracameral Ag injections has been previously described (Camelo et al., 2003). To allow in vivo fluorescence microscopy animals were anaesthetized by intraperitoneal injection of Xylasine (0·033 mg per 100 g, Xylazil-20; Ilium, Australia) and Ketamine (0·167 mg per 100 g, Ketamil; Ilium, Australia). Before cardiac perfusion animals were deeply anaesthetized by intraperitoneal injection of sodium pentobarbitone (100 mg kg21 body weight, Rhone Merieux Australia, Queensland, Australia) and then perfused with heparinised PBS (1000 units l21) followed by paraformaldehyde (BDH Laboratories supplies, (England)) to remove the intravascular pool of cells. All procedures conform to the ARVO statement for the use of animals in ophthalmic and vision research.

In vivo video fluorescent microscopy was performed as previously described (Becker et al., 2000, 2003; Camelo et al., 2003). During examination animals placed upon the mechanical stage of the microscope were kept warm by placing their trunk inside an electrically heated plastic tube. Topical Alcaine 0·5% was administered to the right eye to eliminate reflex blinking. The head was positioned such that the right eye was resting against a drop of sterile ocular lubricating gel (Viscotears, Ciba Vision, Australia) on a glass coverslip and aligned with the microscope (Eclipse TE300, Nikon, Australia) objective. A UV filter block was used for Cascade Blue fluorescence (excitation BP 340– 380 nm, emission LP 420 nm) and blue filter block (excitation BP 450– 490, emission DM 505) for FITC and A488 fluorescence. The iris, cornea and limbus were examined using long distance £ 10 and £ 20 objective lenses (Plan fluoro, NA 0.3 and Plan fluoro, NA 0.45, respectively, Nikon, Australia). 2.4. Tissue collection, processing and immunostaining At 24 hr following intracameral injection animals were perfused and both eyes were enucleated and postfixed in 4% paraformaldehyde. Iridial sheets and corneal-limbal wholemounts were dissected as previously documented (McMenamin, 2000). Alternatively, following perfusion, both eyes were enucleated and embedded in OCT (ProSciTech, Thuringowa, Queensland, Australia) as previously described (McMenamin and Holthouse, 1992)

S. Camelo et al. / Experimental Eye Research 79 (2004) 455–464

for freezing and cryostat sectioning. Routine single immunofluorescence was performed on iris and corneallimbal wholemounts and frozen eye sections as previously described (McMenamin et al., 1994; McMenamin, 2000). To characterise FITC conjugated Dx, OVA or Alexa-Fluor 488 conjugated BSA containing cells, single staining was performed using a range of mouse anti-rat primary mAb followed by secondary goat anti-mouse IgG directly coupled to Alexa-Fluor 546. To characterise CB-Dx containing cells, we chose red (Alexa-Fluor 546 conjugated anti-mouse IgG, emission 572 nm) and green (FITC conjugated-streptavidin, emission 520 nm) fluorochromes for immunofluorescence thus allowing triple stained cells to be observed. During double staining the primary mAb (mAb 1) was revealed using Alexa-Fluor 546 conjugated secondary followed by the second biotinylated mouse anti-rat primary mAb (mAb 2) visualised using streptavidin-FITC. Negative controls (incubation for 45 min in purified mAb anti-DNP or without primary mAbs) were performed on some sections or wholemounts in each experiment. 2.5. Confocal microscopy and image analysis Wholemounts and tissue sections were examined for the presence of immunofluorescently labelled and fluorescent Agþ cells by confocal microscopy as previously described (Camelo et al., 2003). Frozen sections were imaged by confocal microscopy to detect the presence of immuno-reactivity and/or Agþ cells. The resultant files were merged and false colored to give a tri-colour image. Montage of the anterior segment was constructed by assembling the composite images using Adobe Photoshop (Version 7.0, Mountain View, Ca). Tissue wholemounts were imaged by generating a series of z-stacks from the internal aspect moving through to the external aspect in 0·5 mm increments. The regions sampled included the central cornea, peripheral cornea, limbus and the choroid, sclera and episclera at the level of commencement of the retinal pigmented epithelium. Merged images were produced using the Confocal Assistant version 4.02. The presence and relative quantity of intracellular antigen in wholemount samples was scored in 10 mm layers (2 , absence of Ag; þ , few cells bearing low levels of Ag; þ þ , moderate number of Agþ cells; þ þ þ , high number of Agþ cells bearing high levels of Ag and moderate levels of free Ag; þ þ þ þ , high number of Agþ cells with high levels of free Ag).

3. Results 3.1. Uptake of Ag by cells in the iris and limbus observed by in vivo video microscopy In vivo video fluorescent microscopic images of the right iris and limbus of PBS, Dx and BSA injected animals were

457

recorded 24 hr post injection. No fluorescent cells were observed in the iris (Fig. 1(A) and limbus (Fig. 1(B)) of control (PBS) injected animals. A dense network of Agþ cells could be observed in the right iris of 70 kDa Dx (Fig. 1(C)) and BSA (Fig. 1(E)) injected rats. Fluorescent cells tended to be more concentrated at the iris base and occasionally focal clumping of Agþ cells in the iris was noted in a few animals (data not shown). In vivo video microscopic analysis also revealed large numbers of these cells around the episcleral vessels and in the perilimbal conjunctiva following Dx (Fig. 1(D)) and BSA (Fig. 1(F)) injections. Similar results were observed in 40 kDa Dx and OVA injected animals (data not shown). 3.2. Microscopic localization of Agþ cells in the anterior segment Confirming our in vivo fluorescence analysis, 24 hr following intracameral injection of OVA, BSA and Dx, Agþ cells were observed in the iris, within the loose connective tissue of the ciliary body and in the anterior suprachoroidal space. Cells containing Ag accumulated close to Schlemm’s canal and within the loose connective tissue of the limbal conjunctiva and around limbal/episcleral vessels. Seven days following Dx injection into the AC, Agþ cells could still be observed in the iris, ciliary body (close to Schlemm’s canal) and around episcleral vessels (data not shown). As previously reported (Camelo et al., 2003), whilst most Ag appeared to have accumulated in cells, free fluorescent Ag was also visible in the interstitial space of the ciliary body suggesting that not all Ag remaining in the eye had been taken up by day 7. These observations suggest that Ag leaves the eye through both the conventional and non-conventional aqueous outflow pathways. No fluorescence could be detected in animals injected with PBS or in the uninjected eye of experimental animals at any time points (data not shown). A summary of the distribution of the different fluorescent Ags in the anterior segment determined on corneal/limbal wholemounts is presented in Table 1. 3.3. Phenotype of Agþ cells in the anterior segment The aim of this part of the study was to characterize the phenotype of the cells in the eye that had captured fluorescent Ags. To this end, tangential frozen sections of ocular tissues from injected rats (control and experimental eyes) were subjected to single and double immunofluorescent staining with a range of mAbs specific for macrophage and DC phenotypic markers and subsequently examined by confocal microscopy (Fig. 2(A)). In Dx injected animals large numbers of Agþ cells observed in the anterior segment were ED1þ, a marker present in both macrophages and DC. Some Dxþ cells also expressed MHC class II (OX6þ). Large numbers of OX6þ and OX6þ ED1þ Ag2 cells were present in the peripheral cornea and

458

S. Camelo et al. / Experimental Eye Research 79 (2004) 455–464

Fig. 1. In vivo video microscopic images of the rat iris ((A), (C) and (E)) and limbus ((B), (D)) and (F) 24 hr following injection with PBS ((A) and (B)), cascade blue Dx ((C) and (D)) or Alexa-Fluor 488 conjugated BSA ((E) and (F)). Note the absence of any fluorescence in PBS injected anterior segment while iris (A) and limbal (B) blood vessels can be visualized. Fluorescent Agþ cells can be seen in close proximity to blood vessels in the iris (C) and limbus (D) of Dx-injected animals. Even greater numbers of Agþ cells were present in the iris (E) and limbus (F) of BSA injected rats. Note that the iris of the BSA-injected animal appears ‘out of focus’ due to the interference of a significant number of Agþ cells in the cornea. Such cells are less frequent in the cornea of Dx-injected animals. Magnification: £ 20 objective.

Table 1 Intraocular Ag distribution after AC injection of OVA, BSA and Dx Depth

Tissue Peripheral corneaa

Ocular surface 60–70 mm 50–60 mm 40–50 mm 30–40 mm 20–30 mm 10–20 mm 0–10 mm Anterior chamber

Limbus

Sclera

OVA

BSA

Dx

OVA

BSA

Dx

OVA

BSA

Dx

2 NT þþ þþ þþ þ þ

NT 2 þ þþ þþ þþ þþ

2 2 þ þ 2 þ þþ

2 2 þ þ þ þ þþ

þ þ þþ þþ þþ þþ þ þþ

þþ þþ þ þþ þ þþ þ þþ þ þþ þ þþ þ

NT 2 2 2 2 2 þ

NT þ þþ þþ þþ þþ þ þþ þ þþ þ

NT þ þþ þþ þþ þþ þ þþ þþ

Distribution of BSA, OVA and Dx in the anterior segment was compared on corneal wholemounts 24 hr following an intra-cameral injection of different Ags. Confocal images were obtained at the level of the peripheral cornea, trabecular meshwork and sclera. The presence and relative quantity of antigen was scored in 10 mm layers 2, absence of Ag; þ , few cells bearing low levels of Ag; þ þ, moderate number of Agþ cells; þ þþ , high number of Agþ cells bearing high levels of Ag and moderate levels of free Ag; þþ þþ , high number of Agþ cells with high levels of free Ag; NT, Not tested. a In the peripheral cornea only cell-associated Ag has been analysed.

S. Camelo et al. / Experimental Eye Research 79 (2004) 455–464

459

Fig. 2. Confocal microscopic images of frozen sections and corneal/limbal wholemounts 24 hr following intracameral injection of Dx. (A) Montage of a frozen section showing the right iridocorneal angle of a Dx (blue)-injected eye stained with mAb ED1 (CD68, red) and OX6 (MHC class II, green). In the choroid Dxþ ED1þ double positive cells appears purple. Around episcleral vessels Dxþ OX6þ cells are cyan while triple positive cells Dxþ ED1þ OX6þ (white) are present around the canal of Schlemm. Note also ED1þ OX6þ Dx2 cells (yellow), most likely newly recruited mononuclear cells, are present in the ciliary body and OX6þ ED12 Dx2 (green) cells in the stroma of the peripheral cornea. Blue (Dxþ only) cells are numerous in the episcleral bulbar and conjunctival stroma. (B) Confocal image of corneal/limbal wholemount from FITC-conjugated Dx (green) injected rat stained with ED1 (red) at the level of the trabecular meshwork reveal many Dxþ cells expressing ED1 (yellow). ((C) and (D)), episcleral tissue section of Dx (blue) injected eyes stained with ED2 shows Dxþ ED2þ OX62 (purple) and Dxþ ED2þ OX6þ (orange/white) around episcleral vessels. (E) White triple positive cells (Dxþ ED2þ OX6þ) in the anterior choroid of Dx injected animal. Note the presence of OX6þ Ag2 (green) cells. (F) In the ciliary body, Ag has been internalized by large OX6þ ED22 cell (cyan) and a smaller OX6þ cell expressing low levels of ED2 (green, cyan and yellow). Numerous single OX6þ cells (green) and an OX6þ ED2þ (yellow) are in close proximity. (G) In the ciliary body, Dx was internalized by cells expressing low levels of OX6 (green) and CD11b (red) in close proximity to large numbers of OX6þ Ag2 cells. Bars represent 20 mm in each image.

iridocorneal angle respectively, alongside a few Agþ cells expressing both ED1 and OX6. In the anterior choroid, Agþ ED1þ OX62 macrophages were frequently observed (Fig. 2(A)). The expression of ED1 by Agþ cells at the level of the trabecular meshwork was confirmed by confocal microscopy on tissue wholemounts following injection of

40 kDa Dx (Fig. 2(B)). Similar distribution of Agþ cells was observed on sections stained with mAbs specific for ED2 and OX6 (not shown). Double positive (Agþ ED2þ, purple) and triple positive (Agþ ED2þ OX6þ) cells (white) were observed around episcleral vessels (Fig. 2(C) and (D)). Cells bearing Ag, expressing ED2 and OX6 were also present in

460

S. Camelo et al. / Experimental Eye Research 79 (2004) 455–464

Fig. 3. Confocal microscopic analysis of frozen ocular sections and tissue wholemounts 24 hr following intracameral injection of BSA. In these images, Ag appears green and staining with the mAbs red. Colocalisation is indicated in yellow. (A) Iridocorneal angle sections stained with the mAb ED1 specific for the rat equivalent of CD68 shows the presence of large BSAþ ED1þ cells (yellow) in proximity to ED1þ Ag2 cells (red). Note also large amounts of free soluble Ag (green). (B) Similar ED1þ Agþ cells were observed in the anterior choroid. (C) Iridocorneal angle section revealed the presence of ED2þ Agþ cells (yellow), ED2þ Ag2 cells (red) and ED22 Agþ cells (green). Free Ag appears as small green dots. (D) Choroidal wholemounts reveals large numbers of ED2þ Agþ elongated cells. Hexagonal shaped multinucleated retinal pigment epithelial cells are clearly identifiable. (E) Putative DC (OX6þ Agþ, yellow), OX6þ Ag2 cells (red) and free Ag (green dots) can be observed in the iridocorneal angle of BSA injected animal. (F) An ED3þ macrophage containing BSA can be seen in the anterior choroid. (G) Wholemount of the peripheral cornea and trabecular meshwork was stained with mAb specific for ED1 (red). Note the presence of numerous ED1þ cells in the trabecular meshwork (right side of image) but not in the cornea (left side of image). A few ED1þ cells have internalized BSA (yellow). (H) Iris wholemount stained with the mab OX6 (red). The majority of BSA-bearing cells are OX62. Bars represent 20 mm in each image.

the anterior choroid (Fig. 2(E)) and in the ciliary body (Fig. 2(F)). Cells bearing Ag in the ciliary body also expressed CD11b (Fig. 2(G)). Following BSA injection, ED1 þ Agþ cells were observed in the iridocorneal angle (Fig. 3(A)) and the choroid (Fig. 3(B)). The presence of large amounts of free Ag was also noted in the iridocorneal angle. Some cells bearing BSA in the iridocorneal angle also expressed ED2, a pan macrophage marker (Fig. 3(C)). Large numbers of

elongated ED2þ Agþ cells were also detected on frozen sections and in anterior segment wholemounts, on the sclerad aspect of the pigmented epithelial cells in the anterior choroid (Fig. 3(D)). Small numbers of BSAþ OX6þ potential DC were present in the ciliary body (Fig. 3(E)) reminiscent of the Dx experiment described above. The expression of sialoadhesin (ED3þ) by numerous Agþ cells in the ciliary body indicates that many of these cells were macrophages. Observation of tissue wholemounts

S. Camelo et al. / Experimental Eye Research 79 (2004) 455–464

at the level of the trabecular meshwork stained with the mAb ED1 revealed numerous ED1þ Ag2 cells, a few Agþ cells expressing ED1 and the presence of free Ag (Fig. 3(G)). However, it is highly possible that the ‘free Ag’ may have been internalized by trabecular meshwork cells (not visualized in the staining regimes used). Threedimensional reconstruction of the distribution of BSAþ cells reveal that they spanned all layers from the internal side of the anterior segment (trabecular meshwork) to the external side of the eye (limbus) (data not shown). We have previously reported that cells that internalize Dx in the iris are predominantly of the macrophage

461

phenotype expressing ED1, ED2 and ED3 with very few cells expressing the DC markers OX6 and OX62 (Camelo et al., 2003). Similarly, observation of iris wholemounts of animals injected with BSA and stained with OX6 revealed that BSAþ cells were mainly OX62 (Fig. 3(H)) indicating that iris Ag-bearing cells are more likely macrophage than DCs. In OVA injected eyes some ED1þ Agþ cells were noted in the ciliary body (Fig. 4(A) and (B)). The majority of OVAþ cells in the iridocorneal angle expressed ED2 (Fig. 4(C) and (D)) and ED3 (Fig. 4(E) and (F)). Only a few OVAþ cells were OX6þ in the ciliary body (Fig. 4(G))

Fig. 4. Confocal microscopic images of frozen section 24 hr following intracameral injection of FITC-conjugated OVA. ((A) and (B)), right iridocorneal angle stained with mAb ED1 (CD68, red), shows that the majority of OVAþ cells (green) appear not to express this marker. ((C) and (D)), in the ciliary body, majority of OVAþ cells express ED2 (yellow or red containing green) while some are ED22 (green). ((E) and (F)), in the iridocorneal angle and in close proximity to Schlemm’s canal the majority of OVAþ cells are ED3þ. (G) Views of the ciliary body and episcleral tissue (H) show that the majority of OX6þ cells do not contain OVA. All bars represent 20 mm.

462

S. Camelo et al. / Experimental Eye Research 79 (2004) 455–464

and episclera (Fig. 4(H)). Absence of background staining on control sections confirmed the specificity of the immuno-histochemistry procedure (data not shown). However, staining of tissue wholemounts with a control mAb (DNP) revealed some auto-fluorescence on the corneal endothelium that could represent fibrin aggregates.

4. Discussion Recently, our group demonstrated that bone marrowderived cells distributed throughout the anterior segment of the eye internalize dextran (Dx) injected into the AC (Camelo et al., 2003). In the present study, using in vivo video microscopy followed by confocal microscopy of frozen eye sections and corneal-limbal wholemounts, we confirm that this fluorescent labeled polysaccharide Ag is present both in a soluble and in a cell-associated form along the route of both the conventional and non-conventional aqueous outflow pathways. In previous studies of short term fluid dynamics in the eye similar distribution of fluorescent Dx in the sclera and tissues lining the AC has been reported within a few hours of intracameral injection (Bill, 1975; Cole and Monro, 1976; Kim et al., 2002; Lindsey and Weinreb, 2002). Experimental studies of the deviant ocular immune responses induced following Ag placement into the AC have relied upon protein or cellular Ags rather than polysaccharide Ags such as Dx. In this context, it is interesting that Becker and colleagues (Becker et al., 2003) have recently shown that immune cells in the iris internalize protein Ag (OVA) as early as 6 hr following AC injection. In light of these observations and our own previously published data it was thus important to determine whether protein Ags (BSA, OVA) were handled and distributed in a pattern equivalent to that observed in Dx injected eyes (Camelo et al., 2003). In concordance with our observations in Dx injected eyes, BSAþ cells were observed in the tissues lining both the non-conventional and conventional outflow pathways. By contrast, OVAþ cells were more concentrated along the latter pathway. The reason behind the different distribution of Dx and BSA vs. OVA remains unclear as it does not appear to depend on the nature of the Ag (protein vs. polysaccharides) or the quantity injected (Dx (20 mg) vs. OVA and BSA (10 mg)). The lack of OVA in the anterior choroid, when compared to BSA and Dx, is not related to the lysine-fixable property of the Ag as Dx (lysine fixable), and BSA, which was not in this form displayed a similar distribution. Possible explanations for the difference in OVA distribution may be differences in ionic charge, isoelectric points and water solubility which may influence its dynamics within the aqueous humour. On the basis of previous immunophenotypic and functional studies it is clear, there are rich networks of potential APCs, namely macrophages and DC, in the iris, ciliary body, choroid and tissues lining the outflow pathways (Knisely et al., 1991; Flugel et al., 1992;

McMenamin and Holthouse, 1992; McMenamin et al., 1992, 1994; Steptoe et al., 1997; McMenamin, 1997, 1999). Macrophages and DC, have distinct functional roles and antigen presenting capabilities which may influence the nature of immune responses (Manning and Gajewski, 2001). Therefore it was important to determine whether either or both cell types were primarily responsible for internalizing intracamerally injected Ag. Previous reports suggest that 24 hr following AC injection of OVA (Becker et al., 2003) or Dx (Camelo et al., 2003) the majority of Ag-bearing cells in the iris were of the macrophage phenotype. The present report expands previous investigations and reveals that the majority of Ag-bearing cells in the ciliary body, choroid and tissues lining the outflow pathways express ED1, ED2 or ED3, a phenotype indicative of cells of the monocytes/macrophage lineage (Dijkstra et al., 1994). Becker and colleagues (Becker et al., 2003) reported that 6 hr following injection, Ag was internalized predominantly by MHC class II þ and CD11cþ dendritic cells in the iris, however, 18 hr later these cells had mainly disappeared. In the present study, 24 hr following AC-injection we observed some OX6þ BSAþ and Dxþ cells but not OVA-bearing cells in the iridocorneal angle. This phenotype could correspond to either activated macrophages or DC. The expression of aE2-integrin (OX62þ), which identifies rat DC and gd-Tcells (Brenan and Puklavec, 1992), by a few cells in the iridocorneal angle observed in the present study (data not shown) suggests that at least some Agþ cells may be DC. Definitive identification of these cells as DC, however, would require their isolation and testing of their Agpresentation capacity, something that tissue dimensions in rodent precludes. The current tenet on the cellular basis for ACAID and ocular immune responses is that APCs (presumably DC or macrophages) carry ACAID-inducing signal from the eye to the spleen (Streilein, 2003). It has been shown that the number of iris Agþ cells (Camelo et al., 2003) and quantity of free Ag (Kim et al., 2002) decrease with time, suggesting that Ag-bearing cells or free soluble Ag leave the AC of the eye. There is some indirect evidence that AC derived Ag probably reaches the spleen (Kaplan and Streilein, 1974; Faunce et al., 2001) and draining lymph nodes of the neck (Egan et al., 1996; Hoffmann et al., 2001; Mckenna et al., 2002). More recently we have shown that fluorescent Ag injected into the AC of the eye is internalized by resident macrophage populations in the submandibular, cervical and facial lymph nodes, the spleen and surprisingly in the mesenteric lymph nodes (Camelo et al., 2004). However, the precise route by which free Ag or Agþ cells leave the eye is still unclear. The presence of Agþ cells in the rat iridocorneal angle in the present study suggests that immune cells may travel via the conventional aqueous humour drainage pathways, a concept that has been previously well accepted (Lee, 1995). This does not preclude drainage of Ag or Agþ cells via other routes. Indeed our observations,

S. Camelo et al. / Experimental Eye Research 79 (2004) 455–464

supported by previous studies of tracers dynamics in the eye (Inomata et al., 1972; Cole and Monro, 1976; Inomata and Bill, 1977; Lindsey and Weinreb, 2002), suggest that Agþ cells present in the suprachoroidal space, the sclera and the tissue spaces around episcleral vessels piercing the sclera, i.e. the non-conventional aqueous outflow pathways, may represent an important route of Ag clearance from the AC. From these tissues sites, free or cell-associated Ag would have access to conjunctival lymphatic vessels and from there to submandibular and cervical lymph nodes draining the tissues around the eye (Egan et al., 1996; Hoffmann et al., 2001; Mckenna et al., 2002; Camelo et al., 2004). In conclusion, results from this study and previous observations (Becker et al., 2003; Camelo et al., 2003) suggest that retention of Ag in the eye following an intracameral injection, is mediated by macrophages. Determining the migration capacities of these cells to peripheral secondary lymphoid organs will hopefully help understand the cellular basis of the afferent arm of the deviant ocular immune responses.

Acknowledgements This work was supported by an Australian National Health and Medical Research Council Project Grant (# 37382000). Serge Camelo is recipient of an Eric Cyril Lawrence Medical Research Fellowship. We thank Dr Paul Rigby for his helpful advice and technical help with confocal microscopy. We thank Guy Ben-Ary at the Image Acquisition and Analysis Facility (IAAF) for his help in image processing and in vivo video fluorescence microscopy. The Biomedical Confocal Microscopy Research Centre and the IAAF are supported by the Lotteries Commission of Western Australia.

References Becker, M.D., Nobiling, R., Planck, S.R., Rosenbaum, J.T., 2000. Digital video-imaging of leukocyte migration in the iris: intravital microscopy in a physiological model during the onset of endotoxin-induced uveitis. J. Immunol. Meth. 240, 23–37. Becker, M.D., Planck, S.R., Crespo, S., Garman, K., Fleischman, R.J., Dullforce, P., Seitz, G.W., Martin, T.M., Parker, D.C., Rosenbaum, J.T., 2003. Immunohistology of antigen-presenting cells in vivo—a novel method for serial observation of fluorescently labeled cells. Invest. Ophthalmol. Vis. Sci. 44, 2004–2009. Bill, A., 1975. Editorial: the drainage of aqueous humor. Invest. Ophthalmol. 14, 1–3. Brenan, M., Puklavec, M., 1992. The MRC OX-62 antigen: a useful marker in the purification of rat veiled cells with the biochemical properties of an integrin. J. Exp. Med. 175, 1457–1465. Camelo, S., Shanley, A.C., Voon, A.S., McMenamin, P.G., 2004. The distribution of antigen in lymphoid tissues following its injection into the anterior chamber of the rat eye. J. Immunol. 172, 5388–5395.

463

Camelo, S., Voon, A.S.P., Bunt, S., McMenamin, P., 2003. Local retention of soluble antigen by potential antigen presenting cells in the anterior segment of the eye. Invest. Ophthalmol. Vis. Sci. 44, 5212–5219. Cole, D.F., Monro, P.A., 1976. The use of fluorescein-labelled dextrans in investigation of aqueous humour outflow in the rabbit. Exp. Eye Res. 23, 571–585. Dijkstra, C.D., Dopp, E.A., van den Berg, T.K., Damoiseaux, J.G., 1994. Monoclonal antibodies against rat macrophages. J. Immunol. Meth. 174, 21–23. Egan, R.M., Yorkey, C., Black, R., Loh, W.K., Stevens, J.L., Woodward, J.G., 1996. Peptide-specific T cell clonal expansion in vivo following immunization in the eye, an immune-privileged site. J. Immunol. 157, 2262–2271. Faunce, D.E., Sonoda, K.H., Stein-Streilein, J., 2001. MIP-2 recruits NKT cells to the spleen during tolerance induction. J. Immunol. 166, 313– 321. Flugel, C., Kinne, R.W., Streilein, J.W., Lutjen-Drecoll, E., 1992. Distinctive distribution of HLA class II presenting and bone marrow derived cells in the anterior segment of human eyes. Curr. Eye Res. 11, 1173–1183. Forrester, J.V., McMenamin, P.G., Holthouse, I., Lumsden, L., Liversidge, J., 1994. Localization and characterization of major histocompatibility complex class II-positive cells in the posterior segment of the eye: implications for induction of autoimmune uveoretinitis. Invest. Ophthalmol. Vis. Sci. 35, 64–77. Hoffmann, F., Zhang, E.P., Mueller, A., Schulte, F., Foss, H.D., Franke, J., Coupland, S.E., 2001. Contribution of lymphatic drainage system in corneal allograft rejection in mice. Graefes Arch. Clin. Exp. Ophthalmol. 239, 850–858. Inomata, H., Bill, A., 1977. Exit sites of uveoscleral flow of aqueous humor in cynomolgus monkey eyes. Exp. Eye Res. 25, 113–118. Inomata, H., Bill, A., Smelser, G.K., 1972. Unconventional routes of aqueous humor outflow in Cynomolgus monkey (Macaca irus). Am. J. Ophthalmol. 73, 893–907. Kaplan, H.J., Streilein, J.W., 1974. Do immunologically privileged sites require a functioning spleen? Nature 251, 553–554. Kim, T.W., Lindsey, J.D., Aihara, M., Anthony, T.L., Weinreb, R.N., 2002. Intraocular distribution of 70-kDa dextran after subconjunctival injection in mice. Invest. Ophthalmol. Vis. Sci. 43, 1809–1816. Knisely, T.L., Anderson, T.M., Sherwood, M.E., Flotte, T.J., Albert, D.M., Granstein, R.D., 1991. Morphologic and ultrastructural examination of I-Aþ cells in the murine iris. Invest. Ophthalmol. Vis. Sci. 32, 2423–2431. Lee, W.R., 1995. Doyne Lecture. The pathology of the outflow system in primary and secondary glaucoma. Eye 9, 1–23. Lindsey, J.D., Weinreb, R.N., 2002. Identification of the mouse uveoscleral outflow pathway using fluorescent dextran. Invest. Ophthalmol. Vis. Sci. 43, 2201–2205. Manning, T.C., Gajewski, T.F., 2001. Cells of the monocyte/macrophage lineage. In: Lotze, M.T., Thomson, A.W. (Eds.), Dendritic Cells. Academic Press, San Diego, CA, pp. 263– 274. Mckenna, K.C., Xu, Y., Kapp, J.A., 2002. Injection of soluble antigen into the anterior chamber of the eye induces expansion and functional unresponsiveness of antigen-specific CD8þ T cells. J. Immunol. 169, 5630–5637. McMenamin, P.G., 1997. The distribution of immune cells in the uveal tract of the normal eye. Eye 11, 183–193. McMenamin, P.G., 1999. Dendritic cells and macrophages in the uveal tract of the normal mouse eye. Br. J. Ophthalmol. 83, 598–604. McMenamin, P.G., 2000. Optimal methods for preparation and immunostaining of iris, ciliary body, and choroidal wholemounts. Invest. Ophthalmol. Vis. Sci. 41, 3043–3048. McMenamin, P.G., Crewe, J., Morrison, S., Holt, P.G., 1994. Immunomorphologic studies of macrophages and MHC class II-positive dendritic cells in the iris and ciliary body of the rat,

464

S. Camelo et al. / Experimental Eye Research 79 (2004) 455–464

mouse, and human eye. Invest. Ophthalmol. Vis. Sci. 35, 3234–3250. McMenamin, P.G., Holthouse, I., 1992. Immunohistochemical characterization of dendritic cells and macrophages in the aqueous outflow pathways of the rat eye. Exp. Eye Res. 55, 315 –324. McMenamin, P.G., Holthouse, I., Holt, P.G., 1992. Class II major histocompatibility complex (Ia) antigen-bearing dendritic cells within the iris and ciliary body of the rat eye: distribution, phenotype and relation to retinal microglia. Immunology 77, 385– 393. Steptoe, R.J., Holt, P.G., McMenamin, P.G., 1996. Origin and steady-state turnover of major histocompatibility complex class II-positive dendritic cells and resident-tissue macrophages in the iris of the rat eye. J. Neuroimmunol. 68, 67–76.

Steptoe, R.J., Holt, P.G., McMenamin, P.G., 1997. Major histocompatibility complex (MHC) class II-positive dendritic cells in the rat iris. In situ development from MHC class II-negative precursors. Invest. Ophthalmol. Vis. Sci. 38, 2639–2648. Streilein, J.W., 2003. Ocular immune privilege: the eye takes a dim but practical view of immunity and inflammation. J. Leukoc. Biol. 74, 179 –185. Wilbanks, G.A., Mammolenti, M., Streilein, J.W., 1991. Studies on the induction of anterior chamber-associated immune deviation (ACAID). II. Eye-derived cells participate in generating blood-borne signals that induce ACAID. J. Immunol. 146, 3018–3024. Wilbanks, G.A., Streilein, J.W., 1991. Studies on the induction of anterior chamber-associated immune deviation (ACAID). 1. Evidence that an antigen-specific, ACAID-inducing, cell-associated signal exists in the peripheral blood. J. Immunol. 146, 2610–2617.