QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online 29 June 2004
doi: 10.1242/jcs.01221

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lundkvist, A. Articles by Pekny, M. Search for Related Content PubMed PubMed Citation Articles by Lundkvist, A. Articles by Pekny, M. Social Bookmarking                         What's this?

Under stress, the absence of intermediate filaments from Müller cells in the retina has structural and functional consequences

¹ Department of Medical Biochemistry, The Sahlgrenska Academy at Göteborg University, Medicinaregatan 9C, SE-41390 Göteborg, Sweden
² Paul Flechsig Institute for Brain Research, Leipzig University, Jahnallee 59, 04109 Leipzig, Germany
³ Center for Transgene Technology and Gene Therapy, Institute for Biotechnology, KU Leuven, Herestraat 49, Leuven, B-3000, Belgium
⁴ Institute for Pathology, Tübingen University, Liebermeisterstrasse 8, 72076 Tübingen, Germany

View larger version (99K): [in a new window]   Fig. 1. Light microscopy of retinas from wild-type (wt) (A) and GFAP–/– Vim–/– (g–/–v–/–) (B) mice whose eyes were dissected after postfixation, thereby eliminating mechanical stress on the retina. As shown by H&E staining, the retinal structure is normal in both cases. Abbreviation: GCL, ganglion cell layer; INL, inner nuclear layer; ONL, outer nuclear layer. Scale bar, 50 µm.

View larger version (71K): [in a new window]   Fig. 2. Light microscopy of retinas from wild-type and mutant mice whose eyes were prepared by `standard' means (i.e. imposing mechanical stress on the retina). (A-D) H&E staining reveals a normal laminar structure in wild-type (wt) and GFAP–/– (g–/–) retinas, partial separation of the ILM in a Vim–/– (v–/–) retina, and complete separation in a GFAP–/– Vim–/– (g–/–v–/–) retina. Immunohistochemical visualization of ILM by laminin antibodies (E,F), astroglial cells by S100 antibodies (G,H) and neuronal processes in the ganglion cell layer (GCL) and outer plexiform layer by antibodies against neurofilament-M (I,J, green) and the distribution of retinoschisin (Rs1) (I,J, red) in wild-type and GFAP–/– Vim–/– retinas. The separated layer in GFAP–/– Vim–/– mice consists of the ILM (F,H, arrows) and parts of Müller cells (H). Arrows indicate the ILM and asterisks indicate separation between the ILM and the rest of the retina. Abbreviations: INL, inner nuclear layer; ONL, outer nuclear layer. Scale bars, 50 µm (A-D), 20 µm (E,F), 50 µm (G-J).

View larger version (146K): [in a new window]   Fig. 3. Transmission electron microscopy of wild-type and GFAP–/– Vim–/– retinas. (A) In wild-type retinas, large Müller-cell endfeet (MC) form a continuous straight zone at the vitreal surface of the retina; the endfeet contain abundant rough endoplasmic reticulum and a few IFs (arrowheads), are closely apposed to the basal lamina (BL) and form a boundary between optic axons (ax) and the vitreous body (vit). (B) In GFAP–/– Vim–/– retinas, the basal lamina is separated from the retina by many micrometers and is not visible (arrow), and the Müller cell processes lack IFs. The Müller cell processes and endfeet are both irregularly shaped and, often, only very thin lamellae are visible (top left). (C) A lower-magnification image shows the irregularity of the inner surface of the remaining retina. Many blood vessels (bv) are not covered by Müller-cell endfeet (as they are in wild-type retinas; not shown). Instead, the basal lamina of pericytes (pc) forms the innermost surface of the remaining retina (arrowheads). (D) A large blood vessel containing many erythrocytes (ery) is incompletely covered by Müller-cell processes (MC) and is detached, along with a large axon bundle, from the rest of the retina (arrow). (E) Rarely, a Müller-cell process (asterisks) had a disrupted membrane and missing cytoplasm.

View larger version (75K): [in a new window]   Fig. 4. Vascular response to OIR at P17. (A) The OIR time scheme. 1-week-old mice are exposed to 75% oxygen for 5 days. This hyperoxia causes cessation and even some regression of the developing retinal vasculature. Thus, at P12, when the mice are put back to normal conditions, the retinal vasculature is underdeveloped, which leads to relative hypoxia and a neovascular response in the retina that becomes maximal at P17; at P28, the vasculature has remodeled into the normal appearance. (B-E) H&E staining reveals intact retinas and comparable neovascular responses in wild-type (B) and GFAP–/– mice (C). In Vim–/– mice (D), the ILM was partially separated (asterisk), as in adult Vim–/– mice (compare Fig. 2C), and the neovascular response in the vitreous body was reduced. In GFAP–/– Vim–/– mice (E), the ILM was completely separated from the rest of the retina (asterisk in E, compare with Fig. 2D,F,H) and the neovascular response in the vitreous body was limited. (F,G) Visualization of Müller cell endfeet using antibodies against carbonic anhydrase II (CAII) reveals that the separation of the ILM-containing layer of the retina occurs within Müller-cell endfeet. Arrows indicate vitreal vessels and arrowheads indicate intraretinal vessels. Asterisk indicates a separation. Abbreivations: GCL, ganglion cell layer; INL, inner nuclear layer; VB, vitreous body. Scale bar, 20 µm (B-G).

View larger version (61K): [in a new window]   Fig. 5. Vascular response to OIR at P17 and P19. (A-D) Numbers of cell nuclei and capillary tufts in the vitreous body, assessed on retinal cross sections at P17. (E) Cell nuclei in the vitreous body at P19. (F-H) Extent of the vascular-free zone as a measure of the vascular response to OIR, assessed on whole-mounted retinas at P17 in which vessels were detected by isolectin staining (F,G, red). During relative hypoxia (P12-P17; Fig. 4A), avascular zones (F) become vascularized. A massive neovascular response occurs around the veins and capillary tufts grow toward the vitreous body (G, arrows). The areas of avascularity at P17 (delineated in G) were comparable in wild-type and GFAP–/– Vim–/– retinas (H). (I) VEGF was expressed at similar levels in wild-type and GFAP–/– Vim–/– retinas from controls (wtc vs g–/–v–/–c) and from mice subjected to OIR (wtoir vs g–/–v–/–oir). The data are presented as mean±s.e.m. Asterisks indicate P-values obtained by using Student's t test: *, 0.05; **, 0.01; ***, 0.005; ****, 0.002; *****, 0.001. Scale bar, 500 µm (F,G).

View larger version (73K): [in a new window]   Fig. 6. Development of the vascular system in P0-P9 retinas. (A-D) Blood vessels identified by isolectin staining and astrocytes identified by S100 antibodies at the front of the expanding vascular plexus at P7. The network of astrocytes at the vascular front, the progression of retinal vascularization (A,B) and fibronectin production (C,D) were similar in wild-type and GFAP–/– Vim–/– mice. (E,F) The ratio of the radius of the vascular plexus to that of the retina (Lv/Lr; E) was similar in wild-type and GFAP–/– Vim–/– mice at P0-P9 (F). (E inset) The region depicted in (A-D). Scale bars, 100 µm (A-D).

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

Twitter