Perifoveal müller cell depletion in a case of macular telangiectasia type 2

Abstract

Purpose: To assess the histopathologic changes in a postmortem sample derived from an eye donor with macular telangiectasia (MacTel) type 2 to gain further insight into the cause of the disease.

Design: Clinicopathological case report.

Participants: Postmortem tissue was collected from 5 different donors: 1 MacTel type 2 patient; 1 healthy control; 2 type 2 diabetic patients, 1 with retinopathy and 1 without retinopathy; and 1 patient with unilateral Coat's disease.

Methods: Macular pigment distribution in the posterior part of freshly dissected eyes was documented by macrophotography. Paraffin sections from both the macular and peripheral regions were assessed using antigen retrieval and immunohistochemistry to study the distribution of cell-specific markers. Blood vessels were visualized with antibodies directed against collagen IV and claudin 5; glial cells with antibodies against glial fibrillary acidic protein (GFAP), vimentin, glutamine synthetase (GS), and retinaldehyde binding protein (RLBP1, also known as CRALBP); microglia with an antibody against allograft inflammatory factor 1 (also known as Iba1); and photoreceptors with antibodies against rhodopsin and opsin. Using anatomic landmarks, the sections then were matched with the macular pigment distribution and a fluorescein angiogram of the patient that was obtained before the patient's death.

Main outcome measures: Presence and distribution of macular pigment and cell-specific markers.

Results: Macular pigment was absent in the macula. Furthermore, abnormally dilated capillaries were identified in a macular region that correlated spatially with regions of fluorescein leakage in an angiogram that was obtained 12 years before death. These telangiectatic vessels displayed a marked reduction of the basement membrane component collagen IV, indicating vascular pathologic features. The presence of GFAP was limited to retinal astrocytes, and no reactive Müller cells were identified. Importantly, reduced immunoreactivity with Müller cell markers (vimentin, GS, and RLBP1) in the macula was observed. The area that lacked Müller cells corresponded with the region of depleted macular pigment.

Conclusions: These findings suggest that macular Müller cell loss or dysfunction is a critical component of MacTel type 2, which may have implications for future treatment strategies.

Conflict of interest statement

No authors have any financial/conflicting interests to disclose

Copyright © 2010 American Academy of Ophthalmology. Published by Elsevier Inc. All rights reserved.

Figures

Figure 1

Angiogram of the right (A, C, E) and left eye (B, D, F) showing fluorescein leakage in the perifovea (arrowheads). The leakage is particularly prominent temporal to the fovea, which is characteristic for MacTel type 2.

Figure 2

Photographs of dissected eye globes from three different donors before fixation showing the distribution of macular pigment. In the MacTel type 2 patient (A, B) macular pigment is absent from the center of the macula (black arrows) but faintly visible as a perifoveolar ring (white arrows). In contrast, in the two control donors (C–F) the highest concentration of macular pigment is found in the fovea (black arrows).

Figure 3

Mapping of wax sections onto the angiogram. Two sections (110, superior, and 198, inferior to the macula) stained against collagen IV show the distribution of vessels (A). Larger vessels were identified (insets in A) and plotted as red dots in the naso-temporal axis. The distribution of red dots was used to define the superior-inferior position of the two sections and to position a scale that gives the approximate position of every section in the angiogram (B). Hematoxylin & eosin stains of three sections (C, section 130, 140 and 150) shows that vascular abnormalities (visible at higher magnification in D, E) match well to fluorescein leakage visible in the angiogram (yellow arrows in B). Scale bar is in C 500µm and in D 50µm.

Figure 4

Immunohistochemistry with antibodies against collagen IV (green) and claudin-5 (red) labels retinal blood vessels (section 133) of MacTel type 2 retina. A non-specific luminal stain is also visible in the red channel and is prominent in telangiectatic, deeper plexus vessels. This is particularly obvious in the overview micrograph (A) as an orange stain. However, confocal microscopy through the entire thickness of the wax section (B, D, F) from selected regions shows that claudin-5 staining can be clearly distinguished as a junctional stain (arrowheads in B, D, F). Single confocal slices (C, E, G) reveal fine vacuoles (arrows C, E, G) in vascular basal lamina throughout the retina. In telangiectatic vessels collagen IV staining appears diffuse and interrupted (white stars in D–G). Temporal is to the left and nasal to the right. Scale bar is in A 200µm and in B 50µm.

Figure 5

Immunohistochemistry with the Iba1 antibody (red) labeling macrophages in the control (A–D) and the MacTel type 2 eye (E–H, section 153). Macrophages are distributed in both samples throughout the RGC layer, IPL and INL in the periphery (B, F) and the macula (C, D, G, H). They tend to be associated with blood vessels (visible as green/yellow autofluorescence), telangiectatic vessels indicated with arrowheads in G, H. RGC; retinal ganglion cell, IPL; inner plexiform layer, INL; inner nuclear layer, HFL; Henle’s fiber layer, OPL; outer plexiform layer, ONL; outer nuclear layer. Temporal is to the left and nasal to the right. Scale bar is in A 200µm and in D 50µm.

Figure 6

Immunohistochemistry with an antibody directed against glial fibrillary acidic protein (GFAP) (red). GFAP immunostaining labels retinal astrocytes in a control retina (A–D) and the MacTel type 2 specimen (E–H, section 134). Strong GFAP labeling in the fovea of the control eye (A, C) might indicate the “Müller cell cone”. There is also weak labeling in Henle’s fiber layer (arrow B). Retinal astrocytes are associated with blood vessels and nerve fibers in both samples and appear in similar distribution and density. Blood vessels are visible as green/yellow autofluorescence (telangiectatic vessels; arrowheads G). No GFAP labeling is visible in Henle’s fiber layer of the MacTel eye (arrow G). GCL; ganglion cell layer, IPL; inner plexiform layer, INL; inner nuclear layer, HFL; Henle’s fiber layer, OPL; outer plexiform layer, ONL; outer nuclear layer. Temporal is to the left and nasal to the right. Scale bar is in A 200µm and in D 50µm.

Figure 7

Immunohistochemistry with antibodies directed against three different Müller cell markers (A–E). The distribution of vimentin (A, B) indicates the presence of Müller cells throughout the macula and periphery in the control eye (A) but a strong reduction in the macula of the MacTel eye (B). Similarly, glutamine synthetase, GS (C) and retinaldehyde binding protein 1, RLBP1 (D) are reduced in the MacTel macula. A 48 hour postmortem delay before fixation did not affect vimentin staining in the macula of a control eye (E). In order to compare the area of macular pigment depletion with the area of Müller cell depletion, blood residues were traced in the color photograph (F, G). The image was scaled to match vessels in the angiogram, which locates the approximate position of the area of pigment depletion (green area) in the angiogram (H). The vimentin immunostaining (B) was scaled and centered on the angiogram (based on the size relationship established in Fig. 3), which demonstrates a rough correlation between macular pigment and Müller cell depletion (yellow arrows in H). The relative frequency of cell nuclei in the inner nuclear layer (based on H&E stained, perifoveal sections) is plotted in I and shows a reduction in the perifoveal region in the MacTel specimen (blue) versus control (red). Stars in I indicate statistical significance with a p-value below 0.05. Temporal is to the left and nasal to the right. Scale bar in E is 200µm

Figure 8

Immunohistochemistry showing the distribution of ML-opsin (red) and rhodopsin (green) in the control (A–D) and the MacTel eye (E–H, section 143). In both samples rods (green) are reduced and cones (red) are increased in the fovea. In the control Henle’s fiber layer is stained by the ML-opsin antibody (C) and high magnification of the ML-opsin stain (D) confirms that this particular antibody stains the entire cone cell, including cone axons. ML-opsin stain in the MacTel Henle’s fiber layer (G, H) suggests the presence of cone axons but is weaker in comparison to the control (C). Nevertheless, strong ML-opsin stain is visible in the MacTel fovea (H). IPL; inner plexiform layer, INL; inner nuclear layer, HFL; Henle’s fiber layer, OPL; outer plexiform layer, ONL; outer nuclear layer. Photoreceptor structure: P; pedicle, S; soma, IS; inner segment, OS; outer segment. Temporal is to the left and nasal to the right. Scale bar is in A 200µm and in D 50µm.