Glial cell-mediated spread of retinal degeneration during detachment: A hypothesis based upon studies in rabbits

Vision Research 45 (2005) 2256–2267 www.elsevier.com/locate/visres

Glial cell-mediated spread of retinal degeneration during detachment: A hypothesis based upon studies in rabbits Mike Francke a, Frank Faude b, Thomas Pannicke a, Ortrud Uckermann a,b, Michael Weick a, Hartwig Wolburg c, Peter Wiedemann b, Andreas Reichenbach a, Susann Uhlmann b, Andreas Bringmann b,* b

a Paul Flechsig Institute of Brain Research, University of Leipzig, D-04109 Leipzig, Germany Department of Ophthalmology and Eye Clinic, University of Leipzig, Liebigstrasse 1014, D-04103 Leipzig, Germany c Pathological Institute, University of Tu¨bingen, D-72076 Tu¨bingen, Germany

Received 13 July 2004; received in revised form 19 August 2004

Abstract In human subjects with peripheral retinal detachments, visual deficits are not restricted to the detached retina but are also present in the non-detached tissue. Based upon studies on a rabbit model of rhegmatogenous retinal detachment, we propose a glial cellmediated mechanism of spread of retinal degeneration into non-detached retinal areas which may also have importance for the understanding of alterations in the human retina. Both detached and attached portions of the rabbit retina display photoreceptor cell degeneration and cystic degeneration of the innermost layers. An inverse mode of photoreceptor cell degeneration in the attached tissue suggests a disturbed support of the photoreceptor cells by Mu¨ller cells which show various indications of gliosis (increased expression of intermediate filaments, cell hypertrophy, decreased plasma membrane K+ conductance, increased Ca2+ responsiveness to purinergic stimulation) in both detached and attached tissues. We propose that gliotic alterations of Mu¨ller cells contribute to the degeneration of the attached retina, via disturbance of glial homeostasis mechanisms. A down-regulation of the K+ conductance of Mu¨ller cells may prevent effective retinal K+ and water clearance, and may favor photoreceptor cell degeneration and edema development. Ó 2005 Elsevier Ltd. All rights reserved. Keywords: Retinal detachment; Photoreceptors; Edema; Mu¨llerÕs glia; P2 receptors; Potassium channels

1. Introduction Retinal detachment is a major cause of vision loss, with approximately 15,000 new cases of non-traumatic retinal detachment every year in the United States (Regillo & Bensen, 1988). Modern surgical methods in the treatment of age-related macular degeneration involves translocation of the retina after the generation of a temporary detachment (de Juan, Loewenstein, *

Corresponding author. Tel.: +49 341 97 21 557; fax: +49 341 97 21 659. E-mail address: [email protected] (A. Bringmann). 0042-6989/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.visres.2004.08.028

Bressler, & Alexander, 1998; Lewis, Kaiser, Lewis, & Estafanous, 1999b; Machemer, 1998), and proposed experimental therapies such as pigment epithelium or photoreceptor transplantation, electronic retinal implants, or injection of trophic factors or vectors into the sub-retinal space may include short- or long-term detachments. In order to limit the retinal degeneration caused by detachment, and to optimize the regenerative capacity of the tissue after reattachment surgery, it is necessary to understand the complex cellular responses to detachment. Rhegmatogeneous retinal detachment occurs after a collapse of the vitreous gel and the creation of a retinal tear by vitreoretinal adhesions through which liquefied

M. Francke et al. / Vision Research 45 (2005) 2256–2267

vitreous fluid flows into the sub-retinal space. At present, the only effective treatment of retinal detachment is surgery. However, an early and ophthalmoscopically successful retinal reattachment often fails to restore the normal visual capabilities (Chisholm, McClure, & Foulds, 1975; Isashiki & Ohba, 1986), and experimental reattachment does not completely recover the photoreceptor cell layer (Anderson, Guerin, Erickson, Stern, & Fisher, 1986; Kroll & Machemer, 1969). Even in cases of successful reattachment surgery, patients often describe permanent defects in color vision and a decline in visual acuity (Isernhagen & Wilkinson, 1988; Nork, Millecchia, Stickland, Linberg, & Chao, 1995), and the recovery of vision after surgical reattachment declines with the duration of detachment and with age (Burton, 1982). Moreover, after retinal reattachment in cases of local detachment, functional defects have also been localized to such areas of the visual field which correspond to retinal regions which had not been detached (Chisholm et al., 1975; Sasoh, Yoshida, Kuze, & Uji, 1997), and the macular function may be also depressed in cases with purely peripheral detachment (Chisholm et al., 1975). These observations in human subjects indicate that, in addition to the detached retinal areas, the surrounding non-detached retina undergoes changes. The cause for this dysfunction of attached retinal areas, which may contribute to the impaired recovery of visual acuity after reattachment surgery, is unclear. In our experiments using a rabbit model of rhegmatogenous retinal detachment, we observed distinct degeneration also of attached retinal areas surrounding a focal detachment. In the present review of our observations we propose the hypothesis that the retinal degeneration in the non-detached retina is caused, at least in part, by Mu¨ller cell gliosis. Disturbed homeostasis mechanisms—normally maintained by Mu¨ller cells—may underlie the degeneration of both the outer and inner retinal layers in the attached areas. Reactive gliosis may have cytoprotective as well as cytotoxic effects on retinal neurons. Mu¨ller glial cells are suggested to be implicated in cytokine-mediated protection of photoreceptor cells from death (Harada et al., 2000; Wen et al., 1995); they release antioxidant substances such as glutathione, and they buffer the elevated extracellular K+ and protect neuronal cells from glutamate and nitric oxide toxicity, particularly by glutamate uptake and subsequent detoxification by glutamine synthesis (Bringmann, Francke, & Reichenbach, 2004). On the other hand, activated retinal glial cells may also exert cytotoxic effects under distinct pathological conditions, e.g. by expression of proinflammatory cytokines such as tumor necrosis factor (Cotinet, Goureau, Hicks, Thillaye-Goldenberg, & de Kozak, 1997; de Kozak, Naud, Bellot, Faure, & Hicks, 1994, 1997; Drescher & Whittum-Hudson, 1996), by production of free radicals, and by failures in the normal glutamate uptake (Li &

2257

Puro, 2002) and in extracellular K+ buffering which may exacerbate neuronal hyperactivation and excitotoxicity (Bringmann, Francke, et al., 2004). In order to investigate the pathophysiological Mu¨ller cell responses upon detachment, we used a rabbit model of focal rhegmatogenous retinal detachment, with the creation of a small retinal hole and sub-retinal injection of sodium hyaluronate in order to stabilize the detachment (Faude et al., 2001). One of the main differences between human and rabbit retinas is (in addition to the absence of a cone-dominated macula) the absence of blood vessels in the sensory retina (with the exception of the highly myelinated medullary rays). This has advantages for the investigation of the involvement and responses of Mu¨ller cells to detachment since astrocytes (which surround the superficial vessels in vascularized retinas) are absent, and agonist-evoked physiological responses can be recorded selectively, e.g. from the Mu¨ller cell endfeet at the vitreal surface of retinal wholemounts (Uckermann et al., 2004).

2. Outer retinal degeneration The vision loss caused by retinal detachment is suggested to be predominantly caused by apoptotic photoreceptor cell death (Chang, Lai, Edward, & Tso, 1995; Cook, Lewis, Fisher, & Adler, 1995; Fisher & Anderson, 2001), although structural and biochemical rearrangements may occur in various retinal layers (Anderson et al., 1986; Coblentz, Radeke, Lewis, & Fisher, 2003; Erickson, Fisher, Anderson, Stern, & Borgula, 1983; Fisher, Erickson, Lewis, & Anderson, 1991; Lewis, Matsumoto, & Fisher, 1995, 1998; Marc, Murry, Fisher, Linberg, & Lewis, 1998) and contribute to vision loss (Fisher & Lewis, 2003). Photoreceptor cells begin to die during the first day of experimental detachment, with a maximum occurring around three days, and it continues to some extent as long as the retina is detached (Hisatomi et al., 2001, 2002; Rex et al., 2002). The photoreceptor cell loss can be a major cause of an irreversible change of retina even after reattachment, because of the failure to regenerate. The vulnerability of photoreceptor cells has been explained mainly by the hypoxia and nutrient deprivation caused by detachment which leads to an increased distance between the choriocapillaris and the neural retina (Linsenmeier & Padnick-Silver, 2000; Mervin et al., 1999). This may also explain the relative preservation of the inner retina which is trophically supplied by the inner retinal circulation. However, detachment of the non-vascularized rabbit retina causes a very similar degeneration pattern as described for detached vascularized retinas, e.g. of the pig and rat (Jackson et al., 2003; Zacks et al., 2003): the most marked degeneration was found in the outer retina with a markedly thinned outer nuclear layer,

2258

M. Francke et al. / Vision Research 45 (2005) 2256–2267

while the inner retina was almost unchanged in thickness (Fig. 1A) and showed only marginal loss of second and third order neurons; however, some indications of disorganization are present, such as cystic degeneration (see below) (Faude et al., 2001). The similar degeneration characteristics of vascularized and non-vascularized retinas offers the possibility to use rabbits for the investigation and understanding of (at least certain aspects of) cellular responses to detachment. Most impressively, although the outer nuclear layer was markedly thinned, there were one to three rows of photoreceptor cell nuclei which survived even six weeks of detachment (despite of a stable detachment and the absence of inner retinal blood vessels) (Fig. 1A and B). The mean number of photoreceptor nuclei per Mu¨ller cell was 2.4 after six weeks of detachment (vs. 7.2 in the surrounding non-detached areas) (Faude et al., 2001). The preservation of few rows of photoreceptor nuclei is very similar to the observations in vascularized retinas of rats and cats where many photoreceptor cells survive months of detachment (Erickson et al., 1983; Zacks et al., 2003).

Why some photoreceptor cells survive long-term detachment is unclear (but would be of great interest since they retain at least some capacity for recovery (Rex et al., 2002)). Since we observed the same phenomenon in the non-vascularized rabbit retina, the delivery of oxygen and nutrients from the inner retinal circulation as main cause (as it can be expected in vascularized retinas) seems less likely. In addition to the cell loss in the detached retina, photoreceptor cell degeneration can also be observed (albeit with smaller incidence and after longer time periods) in the surrounding non-detached retinal tissue. However, there is an apparent difference in the mode of photoreceptor cell death between both tissues. In the detached retina, the degeneration of photoreceptor cells begins in the outer segments. Detachment disrupts the contact between the outer segments and the pigment epithelium which results in collapse and disappearance of the outer segments within hours and days of detachment, whereas the inner segments and cell bodies may degenerate considerably later or may survive for longer

Fig. 1. Focal detachment of the rabbit retina causes tissue degeneration in both detached and surrounding attached retinal areas. (A) Slices of detached and non-detached regions of a retina which was focally detached for six weeks (above). Measurements of the retinal thickness (below) revealed differences between both regions especially in the ONL while the inner retinal layers were similar in thickness. (B) Ultrastructure of slices through a detached (left side) and a non-detached area (right side) of a rabbit retina which was focally detached for six weeks. In the detached tissue, the ONL is severely reduced in thickness; however, two rows of photoreceptor cell nuclei are preserved. The remaining photoreceptor cells are devoid of outer segments and, in many cases, of inner segments. The small picture (left below) shows surviving photoreceptor nuclei and abnormal inner segments at higher magnification. Beside apparently intact ganglion cells (GC) and Mu¨ller cell endfeet (MCE), the innermost retina contains cysts of various sizes (yellow arrows). In the non-detached area (right side), the ONL and the PRS are much better preserved compared to the detached area; however, several small groups of adjacent photoreceptor cells are in the process of degeneration. The small figure (right below) shows a group of swollen photoreceptor cell bodies with altered chromatin morphology (red arrows); the inner segments of these cells are swollen (blue arrow), while the outer segments are relatively well maintained (violet arrow). With permission from Faude et al. (2001). (C) Cystic degeneration of a detached human retina (blue arrows). The cysts in the proximity of the ora serrata (red arrow) reflect physiological senile cystoid degeneration. From Arruga (1936). (D) Edematous degeneration of a group of Mu¨ller cells in the attached portion of a focally detached rabbit retina. The cells and their nuclei (blue arrow) swell, the cytoplasm is vacuolized, and the cells die after disruption of their plasma membranes. The cell death is associated with disruptions of the inner limiting membrane (red arrows), and with degeneration of photoreceptor cell bodies (yellow arrow). GCL, ganglion cell layer; INL, inner nuclear layer; IPL, inner plexiform layer; NFL, nerve fiber layer; ONL, outer nuclear layer; PRS, photoreceptor segments. (For interpretation of the references in colour in this figure legend, the reader is referred to the web version of this article.)

M. Francke et al. / Vision Research 45 (2005) 2256–2267

time periods (Erickson et al., 1983; Foulds, 1963; Kroll & Machemer, 1969). The example at the left side of Fig. 1B shows a slice through a rabbit retina which was detached for six weeks; two rows of photoreceptor cell nuclei and inner segments (albeit with shorter length and abnormal morphology) survived this long duration of detachment while the outer segments were completely absent. In the attached tissue surrounding the focal detachment, there were groups of adjacent photoreceptor cells which were in the process of degeneration (Fig. 1B) (Faude et al., 2001). The degeneration of these cells begins in the cell bodies; the cells showed swollen cell bodies with disorganized chromatin and swollen mitochondria within the inner segments, while the outer segments, as well as the connection between inner and outer segments, appeared relatively well preserved (Fig. 1B). This atypical or inverse degeneration pattern, which has been found in both rods and cones, suggests that in the non-detached tissue, the support of the photoreceptor cells by the neural retina, likely by Mu¨ller glial cells, is disturbed, and that the photoreceptor degeneration is most probably not caused by a functional disturbance of pigment epithelial cells or by a disruption of oxygen and nutrient supply from the choroid. Such dying cells were rarely found after one week but regularly after three weeks, and abundantly after six weeks of detachment. The morphological alterations of the photoreceptor cell bodies and inner segments may suggest that these cells undergo apoptosis (but this remains to be proven). In the focally detached rat retina, there is a certain level of caspase activation also in the attached tissue, albeit at lower level (Zacks et al., 2003).

3. Inner retinal degeneration In the detached rabbit retina, the thickness of the innermost retinal layers (nerve fiber layer to inner plexiform layer) remained virtually unchanged over several weeks, and the thickness of the inner nuclear layer decreased rather moderately (Fig. 1A) (Faude et al., 2001). However, there was some disorganization in the inner retinal layers, as indicated by a degeneration of single ganglion cells and by an anomalous sprouting of enlarged lateral branches from the hypertrophied Mu¨ller cell bodies into the plexiform layers (Fig. 2A); both phenomena were also observed in the detached retina of the cat (Fisher & Lewis, 2003). In the rabbit retina which was detached for two days up to six weeks, ‘‘empty spaces’’ of varying sizes were present in the nerve fiber and ganglion cell layers (Fig. 1B) (Faude et al., 2001). These ‘‘empty spaces’’ in the innermost layers of detached rabbit retina resemble cystic degeneration. A cystic degeneration (in combination or not with atrophic degeneration) has been described to occur frequently in detached human retinas (Fig. 1C)

2259

(Arruga, 1936). Here, the cysts develop predominantly in the outer plexiform layer, frequently in the inner nuclear layer, and may also be present in the nerve fiber layer (Arruga, 1936). Since the location of the cysts in the detached human retina is very similar to that observed in cystoid macular edema, one may assume that they represent edematous cysts. Intra- and extracellular edema within the inner retinal layers (e.g. of the Mu¨ller cell endfeet abutting the inner limiting membrane) have been described to be an early alteration (within one day) of the experimentally detached pig retina; the intracellular edema is associated with mitochondrial swelling (Jackson et al., 2003). Similarly, experimental detachment of the primate retina caused edematous swelling and cystoid degeneration of the inner retinal layers (Machemer, 1968; Machemer & Norton, 1969). The cause of cyst formation in the inner rabbit retina is unclear. The cysts may be formed either by development of extracellular edema and/or by degeneration of ganglion cells and nerve fibers (Faude et al., 2001). There are indications that the cysts are caused by extracellular edema, as: (i) despite a degeneration of single neurons in the ganglion cell and inner nuclear layers, the overall number of neuronal cell bodies within the ganglion cell layer is not significantly decreased in the detached tissue (which is also reflected in the unaltered thickness of inner retinal layers; Fig. 1A) (Faude et al., 2001); and (ii) hypertrophied Mu¨ller cells do not fill the empty spaces as it would be expected when they were caused by degenerating neurons (and has been observed within the spaces left by dying photoreceptor cells and by retracted photoreceptor synapses in the outer plexiform layer (Erickson et al., 1983; Fisher & Lewis, 2003; Fisher et al., 1991; Lewis & Fisher, 2000)). The morphological deformation of ganglion cells which surround the cysts (Faude et al., 2001) may suggest that the death of single ganglion cells is caused, at least in part, by mechanical injury due to the cysts. Very interestingly, the presence of the cysts was not restricted to the detached retina, but was also observed in the surrounding attached tissue, several millimeters distant from the detachment (Fig. 1B) (Faude et al., 2001). The morphological appearance and the location of the cysts in the attached tissue is quite the same as in the detached tissue. The cysts in the non-detached inner retina are colocalized with groups of atypically dying photoreceptor cells (while the inner plexiform and nuclear layers do not show apparent signs of degeneration) (Fig. 1B); retinal portions without cysts do also not contain degenerating photoreceptors. The colocalization may suggest that both phenomenons are causally coupled, and it is proposed that they are caused by Mu¨ller cell dysfuction (see below). Extracellular edema was not the only degenerative change in these areas; sometimes there were single Mu¨ller cells or small groups of Mu¨ller cells in the

2260

M. Francke et al. / Vision Research 45 (2005) 2256–2267

Fig. 2. Focal detachment of the rabbit retina causes Mu¨ller cell gliosis which is not restricted to the detached tissue. (A) GFAP immunoreactivity in slices of a rabbit retina which was focally detached for one week. GFAP-immunolabeled Mu¨ller cells were also found in the neighboring nondetached retina, as well as in the retinal periphery. Arrows mark hypertrophied side branches of Mu¨ller cells within the inner plexiform layer. There was no obvious GFAP immunoreactivity in the control retina (not shown). (B) Relative plasma membrane area of Mu¨ller cells which were isolated from detached and attached portions of rabbit retinas, illustrating that cell hypertrophy is obvious after two and three days of detachment. (C) The slice through a detached retina is stained against GFAP (green), displaying a GFAP-positive sub-retinal fibrosis (arrow). All cell nuclei are blue stained. GCL, ganglion cell layer; INL, inner plexiform layer; ONL, outer plexiform layer. (D) Relative amplitude of inwardly directed K+ currents in isolated rabbit Mu¨ller cells in dependence on the duration of the detachment. Cells from detached and surrounding non-detached retinal portions were recorded. Above: Examples of current records in three cells. (E) Subcellular distribution of the K+ conductance in isolated Mu¨ller cells from control and three-days detached retinas which was recorded by focal application of a high-K+ solution onto four different membrane domains. (F) Examples of Ca2+ imaging records at the vitreal surface of isolated retinal wholemounts. The images were taken before (above) and during application of ATP (below) in a control retina, and in detached and non-detached portions of a retina which was focally detached for three days. (G) Incidence of Mu¨ller cells which responded to application of ATP with a rise of their cytosolic free Ca2+, in dependence on the duration of detachment. After three days of detachment, retinal tissues from detached and non-detached portions were investigated. With permission from Francke et al. (2001, 2003) and Uhlmann et al. (2003). (For interpretation of the references in colour in this figure legend, the reader is referred to the web version of this article.)

detached and neighboring non-detached tissues which showed features of degeneration apparently due to intracellular edema (Fig. 1D). The bodies, nuclei, and mitochondria of the cells were swollen, and photoreceptor cell somata enveloped by these cells were in the process of degeneration. The degeneration of swollen Mu¨ller cells was associated with a disruption of the inner limiting membrane (Fig. 1D), an event which may favor the development of proliferative vitreoretinopathy.

4. Mu¨ller cell gliosis Retinal detachment causes an early reactivity of Mu¨ller cells which includes morphological, biochemical, and physiological alterations. Within minutes of experimental detachment, Mu¨ller cells show increased protein phosphorylation, e.g. of the fibroblast growth factor receptor and of the mitogen-activated protein kinases,

and increased expression of transcription factors (Geller, Lewis, & Fisher, 2001). Within one day of detachment, Mu¨ller cells begin to proliferate and increase their expression of intermediate filaments, vimentin and glial fibrillary acidic protein (GFAP) (Fig. 2A) (Fisher et al., 1991; Jackson et al., 2003; Lewis, Guerin, Anderson, Matsumoto, & Fisher, 1994, 1995). The detachment-induced proliferation of Mu¨ller cells is transient and peaks at three to four days of detachment (Erickson, Guerin, & Fisher, 1990; Fisher & Lewis, 2003; Fisher et al., 1991). At this point of time, Mu¨ller cell hypertrophy is obvious in the retina (Fig. 2B) where the cells fill the spaces left by dying photoreceptors, and in the sub-retinal space (Fisher et al., 1991; Lewis & Fisher, 2000). The development of a sub-retinal fibrosis by outgrowing Mu¨ller cell processes (Fig. 2C) inhibits the regeneration of outer photoreceptor segments after successful reattachment (Anderson et al., 1986). Hypertrophied side branches of Mu¨ller cells which grow into

M. Francke et al. / Vision Research 45 (2005) 2256–2267

the plexiform layers (Fig. 2A) may inhibit the reformation of synaptic contacts which are disconnected, for example, by the retraction of synaptic terminals of photoreceptor cells (Erickson et al., 1983). The expression of Mu¨ller cell proteins which are involved in homeostatic functions of Mu¨ller cells and in glio-neuronal interactions, such as glutamine synthetase, cellular-retinaldehyde-binding protein, and carbonic anhydrase, are down-regulated in detached retinas (Fisher & Lewis, 2003; Lewis et al., 1999a; Marc et al., 1998). The Mu¨ller cell gliosis may be caused (directly or indirectly) by the detachment-induced hypoxia, since Mu¨ller cell proliferation and hypertrophy were shown to be reduced by oxygen supplementation (Lewis et al., 1999a), and may be mediated by growth factors and cytokines such as basic fibroblast growth factor released by cones (Lewis, Erickson, Guerin, Anderson, & Fisher, 1992), or by direct contacts between cones and Mu¨ller cells (Lewis & Fisher, 2000). In the focally detached retina of the rabbit, the gliotic response of Mu¨ller cells was not restricted to the detached tissue but was also observed in the surrounding non-detached tissue, as indicated by an up-regulated expression of vimentin and GFAP (Fig. 2A), as well as by the hypertrophy of cells (Fig. 2B). The appearance of elevated GFAP levels in Mu¨ller cells of non-detached retinal portions was dependent on the duration of detachment; while at two days of detachment, only faint GFAP staining was found in some Mu¨ller cells in the neighboring attached tissue, marked staining was observed after one week of detachment, and after three weeks, virtually all Mu¨ller cells throughout the entire retina expressed GFAP immunoreactivity (Francke et al., 2001). The GFAP expression by cells in the retinal periphery may be more modest than in the detached area, as indicated by the restriction of expression to the vitreal half of the cytoplasm (Fig. 2A). These data suggest a spread of the gliotic response from the focally detached into the surrounding non-detached tissue. A similar spread of Mu¨ller cell reactivity has been observed when multiple laser lesions were applied to the rabbit retina (Humphrey, Constable, Chu, & Wiffen, 1993). Mu¨ller cells of the detached rabbit retina display early alterations of physiological parameters. One of the main functions of Mu¨ller cells in the normal retina is the maintenance of the extracellular ion and water homeostasis (Newman & Reichenbach, 1996). During light-evoked neuronal activation, the extracellular K+ level increases markedly in the plexiform layers, and in order to avoid K+-induced neuronal hyperexcitation and glutamate toxicity, Mu¨ller cells take up the excess K+ via K+ channels expressed in their plasma membranes, and release it into the inner retinal blood vessels and into the vitreous. The K+ conductance of rabbit Mu¨ller cell membranes is unevenly distributed, with a high expression level of K+ channels in membranes of

2261

the endfoot which abut the vitreous body (Fig. 2E) (Newman, 1987). After detachment, the Mu¨ller cells show an early down-regulation of their expression of functional K+ channels (Fig. 2D) (Francke et al., 2001; Uhlmann et al., 2003), with decreases in all cell membrane areas (Fig. 2E). In addition to Mu¨ller cells of the detached areas, cells in the surrounding non-detached retina display also a down-regulation of their K+ channel expression, albeit after a longer time period (Fig. 2D) (Francke et al., 2001). Another characteristic feature of gliotic Mu¨ller cells in the rabbit retina is the increase of their intracellular Ca2+ responsiveness to stimulation of purinergic P2Y receptors with extracellular adenosine 5 0 -triphosphate (ATP) or with other types of nucleotides (Francke et al., 2002, 2003). In the normal retina, only few Mu¨ller cells respond to ATP with a transient rise of their cytosolic Ca2+ (Fig. 2F and G). After detachment, the incidence of responding cells increases markedly, up to a high value after three days of detachment which is comparable to that observed in experimental proliferative vitreoretinopathy (Fig. 2G) (Uckermann et al., 2003; Uhlmann et al., 2003). As observed in the cases of intermediate filament and K+ channel expression, the alteration of the Ca2+ responsiveness was not restricted to the detached retina but was also observed in the non-detached tissue, albeit with lower amplitude (Fig. 2F and G). The mechanism of elevated Ca2+ responsiveness upon purinergic receptor stimulation is unclear; however, recent observations on cultured Mu¨ller cells which show that growth factors and cytokines such as platelet-derived, epidermal, and nerve growth factor resensitize P2Y receptors which were depressed in their function by agonist application (Weick, Wiedemann, Reichenbach, & Bringmann, in press) may suggest an involvement of intraretinally released soluble factors in the induction of this gliotic alteration. In addition to this signaling from growth factor receptors to P2Y receptors, there is a signaling from P2Y receptor activation to growth factor receptors in cultured Mu¨ller cells which is mediated by P2Y receptor-stimulated autocrine/paracrine release of various factors such as heparin-binding epidermal and plateletderived growth factors (Milenkovic, Weick, Wiedemann, Reichenbach, & Bringmann, 2003). Since P2Y receptor activation stimulates the proliferation of cultured Mu¨ller cells (Milenkovic et al., 2003; Moll et al., 2002), it has been suggested that, as well as the alteration of the plasma membrane K+ currents (Bringmann et al., 2000), the enhanced responsiveness to ATP may also contribute to the proliferation of Mu¨ller cells in the detached retina (Bringmann, Pannicke, et al., 2003). A relation between enhanced P2Y receptor responsiveness and down-regulation of K+ channel expression has been suggested based upon the observation that cells in detached retinas which respond to ATP display significantly smaller K+ currents (Uhlmann et al., 2003). A depressing

2262

M. Francke et al. / Vision Research 45 (2005) 2256–2267

effect of suramin on the down-regulation of the K+ conductance and cell hypertrophy (Uhlmann et al., 2003) may be explained by the inhibiting action of the substance on both P2Y and growth factor receptor signalings. Both the decrease of K+ currents and the increase of responsiveness to ATP are signs of a de-differentiation of rabbit Mu¨ller cells in pathologicalconditions, since the postnatal differentiation of immature cells (which is associated by ceasing of precursor cell proliferation) is accompanied by a marked increase of K+ channel expression (Bringmann, Biedermann, & Reichenbach, 1999) as well as by a decrease of the responsiveness to ATP (Uckermann, Grosche, Reichenbach, & Bringmann, 2002).

5. Mu¨ller cell gliosis may contribute to retinal degeneration At present, one can only speculate about the reason for the atypical or inverse photoreceptor cell degeneration and for the cyst development in the attached tissue surrounding a focal detachment. Since the contact between the pigment epithelium and the photoreceptors is apparently not disturbed in the attached tissue (which is reflected by the well-preserved outer segments), the atypical photoreceptor cell degeneration is likely not mediated by deprivation of the oxygen and nutrient supply from the choroid (as it has been suggested to be the cause for the photoreceptor cell degeneration in the detached tissue). Since we used the non-vascularized rabbit retina in our experiments, a deprivation of nutrient supply from the inner retinal vasculature may also be ruled out as a causative factor. The inverse mode of photoreceptor cell degeneration in the attached tissue (from the cell bodies to the outer segments) suggests that the support of the photoreceptor cells by the neural retina, likely by Mu¨ller glial cells, is disturbed. The down-regulation of glial K+ channel expression may indicate that glial homeostasis mechanisms are disturbed in both the detached and non-detached retina. Is there a causative relationship between the spread of Mu¨ller cell gliosis and the spread of the retinal degeneration from the detached into the surrounding non-detached tissue? In the following, we propose a hypothetical mechanism how dysfunctional Mu¨ller cells in the non-detached tissue may contribute to the spread of retinal degeneration. A down-regulation or closure of K+ channels in gliotic Mu¨ller cells which is observed in both the detached and attached tissues (Fig. 2D) should have deleterious consequences for retinal function since the K+ channels of Mu¨ller cells are crucially involved in the maintenance of the extracellular K+ homeostasis (Newman & Reichenbach, 1996). A decreased uptake of neuronally released K+ by Mu¨ller cells in the outer retina may cause overexcitation and intracellular Ca2+ overload of photo-

receptor cells, resulting in apoptosis which is associated with an alteration of chromatin morphology, cell swelling, and swelling of mitochondria in the inner segments (Fig. 1B). On the other hand, the outer segments are preserved due to the unaltered contact to the pigment epithelial cells. The atypical photoreceptor cell degeneration in the attached tissue is associated with the presence of cysts in the inner retina likely caused by extracellular edema (Fig. 1B). The cause of edema development in the detached and non-detached tissues is unknown. Leakage of the outer blood-retinal barrier constituted by the pigment epithelium, and subsequent serum protein inflow (which osmotically drives water into the tissue) is unlikely to be the cause of cyst formation in the rabbit retina since the intraretinal serum protein distribution is restricted by the outer plexiform layer (Antcliff, Hussain, & Marshall, 2001) which would result in water accumulation and cyst formation in the outer retina (and not in the inner retina, as observed). Generally, water accumulation within the retina may be caused by vascular leakage resulting in water movement from the blood into the retinal interstitium, or by disturbed water redistribution from the retinal parenchyma into the blood or vitreous. While the sub-retinal space (and possibly the outer part of the neural retina) is normally dehydrated by the pigment epithelium (which is the basis for the proper attachment of the neural retina to the pigment epithelium: Marmor, 1999; Pederson, 1994), the inner retina is suggested to be dehydrated via the K+ clearance function of Mu¨ller cells: the currents of neuronally released excess K+ flow through the Mu¨ller cell bodies and drive osmotically water from the retinal interstitium through the Mu¨ller cells into the blood and vitreous (Bringmann, Reichenbach, & Wiedemann, 2004; Nagelhus et al., 1999). This implicates that an effective K+ clearance by Mu¨ller cells is necessary for the dehydration of the inner retina. However, when Mu¨ller cells become gliotic during retinal diseases, e.g. after transient ischemia or during detachment, they down-regulate their expression of functional K+ channels; this down-regulation should disturb the retinal K+ clearance as well as the inner retinal dehydration, and, therefore, may induce inner retinal edema development and cyst formation (Bringmann, Reichenbach, et al., 2004; Pannicke et al., 2004). During acute ischemic conditions, Mu¨ller cells may rapidly close their K+ channels, due to the ATP dependence of channel activation (Kusaka & Puro, 1997). In vascularized retinas, the predominant K+ channel subtype thought to mediate K+ currents between retinal and extraretinal spaces (the Kir4.1 channel subtype: Ishii et al., 1997; Kofuji et al., 2000), and the highest membrane conductance for K+ ions, are located in plasma membrane domains of Mu¨ller cell sheets which envelop the blood vessels in the inner nuclear layer, as well as in endfeet membranes which abut the vitreous body (Fig. 3A)

M. Francke et al. / Vision Research 45 (2005) 2256–2267

2263

Fig. 3. Proposed mechanisms of retinal cyst formation and glial cell-mediated spread of retinal degeneration from the detached into the attached tissue. (A) The sub-cellular distribution of the predominant K+ channel subtype in Mu¨ller cells (the Kir4.1 channel) correlates with the distribution of edematous cysts in detached retinas. In vascularized retinas, the Kir4.1 protein is predominantly expressed in Mu¨ller cell membranes which abut the vitreous body (arrowhead) and which surround the retinal vessels (arrows), beside a small expression at the outer limiting membrane. In nonvascularized retinas, the Kir4.1 protein is mainly expressed at the inner limiting membrane (arrowhead), beside a smaller expression at the outer limiting membrane (arrow). The down-regulation of K+ channel expression by Mu¨ller cells after detachment may cause accumulation of K+ ions in the retinal tissue at sites where Mu¨ller cells predominantly extrude K+ into extra-retinal spaces (i.e., in vascularized retinas into the retinal vessels, in non-vascularized retinas into the vitreous). The resulting higher osmotic pressure at these tissue sites may cause water influx from the blood into the INL in the case of vascularized retinas (which may underlie cyst formation within the INL and perhaps OPL), and water inflow from the vitreous in the case of non-vascularized retinas which may cause cyst formation in the nerve fiber and ganglion cell layers. (B) The spread of gliotic Mu¨ller cell responses from focally detached into non-detached portions of the rabbit retina may be mediated by soluble factors. Detachment causes intraretinal release of ATP and growth factors which diffuse into the surrounding tissue. Only those Mu¨ller cells which express P2Y or growth factor receptors respond with gliotic alterations. The autocrine release of ATP and growth factors by these stimulated cells may induce gliotic alterations in neighboring Mu¨ller cells which results in small groups of cells which cause extracellular edema formation, or which degenerate due to intracellular edema. GCL, ganglion cell layer; INL, inner nuclear layer; ONL, outer nuclear layer; OPL, outer plexiform layer.

(Kofuji et al., 2002; Nagelhus et al., 1999; Newman, 1987). However, in Mu¨ller cells from non-vascularized retinas, e.g. from the guinea pig (Fig. 3A) (Newman, 1987) and from the rabbit (Fig. 2E), the endfeet membranes display the highest K+ conductance, and the Mu¨ller cells release the excess K+ predominantly into the vitreous (some release may also occur into the subretinal space). There is an apparent correlation between the K+ secretion sites of Mu¨ller cells in the vascularized retinas (predominantly into the vessels in the inner nuclear layer) and non-vascularized retinas (into the vitreous) and the locations of cysts in both types of retinas (inner and outer plexiform layers vs. nerve fiber and ganglion cell layers) (Fig. 3A). When the release of excess K+ is disturbed, K+ accumulates within the Mu¨ller cells and in the neighboring retinal interstitium. The accumulation of K+ causes a high osmotical pressure of the retinal tissue. The only borders to extra-retinal spaces which display stable low K+ levels (and relatively low osmotical pressure) are the Mu¨ller cell membrane domains contacting the vessels and the vitreous. The great osmotical gradient at these borders favors water movement into the Mu¨ller cells (and subsequently, into the retinal interstitium) at these sites. We suggest that the down-regulation of functional K+ channels in Mu¨ller cells contributes to both the cyst formation in the inner retina and the photoreceptor cell degeneration in the outer retina; water accumulates in the retinal tissue where regularly the excess K+ is extruded into extraretinal spaces, and the disruption of the K+ clearance currents

may markedly and permanently depolarize the photoreceptor cells; this must result in intracellular Ca2+ overload and induction of apoptosis. The K+ accumulation due to the down-regulation of Mu¨ller cellÕs K+ channels may occur more extracellularly or intracellularly; in the first case, cysts develop in the retinal tissue (Fig. 1B); in the second case, Mu¨ller cells swell and die (Fig. 1D). How does the Mu¨ller cell gliosis spread from the detached into the surrounding non-detached tissue? The cause of this phenomenon is unclear; however, soluble factors which are released by the detached tissue and diffuse into the neighboring tissue are likely to be implicated (Fig. 3B). Retinal detachment is associated with mechanical stress of the tissue, and it has been shown that mechanical deformation of glial cell membranes is an important stimulus which induces autocrinous release of ATP (Newman, 2001). Additionally, retinal detachment causes the release of various growth factors, e.g. of fibroblast growth factor (Geller et al., 2001), and damaged cells at the edge of a retinal break may release various intracellular molecules (Jackson & Marshall, 2004) which may activate surrounding cells. The bidirectional stimulation loop between ATP and growth factors in Mu¨ller cells (with enhanced growth factor release upon stimulation by ATP, and an enhanced sensitivity of P2Y receptors induced by growth factors) may exacerbate the intraretinal release of soluble factors which diffuse from detached into attached portions of the retina, and which may induce gliosis in Mu¨ller cells far distant from the detachment.

2264

M. Francke et al. / Vision Research 45 (2005) 2256–2267

We observed a ‘‘patchy’’ patterns of degeneration in attached retinal areas. The photoreceptor cell degeneration in the attached tissue is not uniform; small groups of photoreceptor cells show signs of degeneration while other groups do not (Fig. 1B) (Faude et al., 2001). Similarly, the cysts (which are colocalized with degenerating photoreceptor cells) are distributed non-uniformly within the non-detached tissue, and groups of edematous Mu¨ller cells are surrounded by apparently normal Mu¨ller cells (Fig. 1D). It has been shown that the K+ current decrease in rabbit Mu¨ller cells after detachment is highly variable; there are cells with marked decrease and others with a small or no alteration (Uhlmann et al., 2003); a similar heterogeneity was observed in cells from attached retinal areas distant from detachment (Fig. 2D, above). One may assume that the groups of degenerating photoreceptor cell nuclei are enveloped by single Mu¨ller cells which show a marked down-regulation of their K+ conductance (Fig. 3B). Thus, both phenomena of retinal degeneration in the attached tissue surrounding a focal detachment (cysts in the inner layers and atypical photoreceptor cell death in the outer layers) may be caused by single Mu¨ller cells (or small groups of these cells) displaying a high degree of gliosis. Why the degree of gliosis may differ among Mu¨ller cells is unclear. We suggest that different expression levels of functional P2Y receptors may be involved in the determination of the severity of gliosis. In the healthy retina, there are only some Mu¨ller cells which express functional P2Y receptors (Fig. 2F). Similarly, in the attached tissue outside of detachment, the overall ATP-induced Ca2+ response is significantly lower than in the detached tissue (Fig. 2G), suggesting that there are Mu¨ller cells that express, and cells which do not express functional P2Y receptors. This heterogeneity of receptor expression (which may also occur in regard to other types of receptors, e.g. for growth factors) may underlie the ‘‘patchy’’ pattern of Mu¨ller cell gliosis and degeneration in attached retinal areas. Small groups of gliotic or degenerating Mu¨ller cells may be formed when one cell which shows a high degree of gliosis induces gliotic alterations in the neighboring cells (Fig. 3B). It is not known whether the gliotic alterations of Mu¨ller cells during detachment are rather cytoprotective or cytotoxic on retinal neurons (Fisher & Lewis, 2003). As previously suggested for the response of cones to detachment (Rex et al., 2002), Mu¨ller cells may enter a state which conserves metabolic energy by down-regulation of ATP-consuming molecular components (e.g. K+ channels (Kusaka & Puro, 1997), glutamine synthetase, etc.) that are not required for the survival of Mu¨ller cells but crucially involved in neuronal survival and activity. It seems as if gliotic Mu¨ller cells are functionally uncoupled from the neurons, in order to enhance their own survival. However, it remains to be elucidated whether this assumption is right, or whether the gliotic altera-

tions of Mu¨ller cells are caused by a disturbed communication between glial cells and neurons, resulting in functional failures of Mu¨ller cells despite their ‘‘intention’’ to support neuronal survival. Beside Mu¨ller cells, there are microglial cells and (in vascularized retinas) astrocytes which form a network of reactive gliosis during detachment. Microglial cells which are located in the inner retina at normal conditions, become early activated during experimental detachment (Uhlmann et al., 2003). Microglial cells have been suggested to modify photoreceptor survival by controlling the neurotrophic factor production of Mu¨ller cells (Harada et al., 2002), and were found to induce degeneration of photoreceptor cells in vitro (Roque, Rosales, Jingjing, Agarwal, & Al-Ubaidi, 1999) and the programmed neuronal cell death in the developing retina (Frade & Barde, 1998). It seems possible (but remains to be proven) that activated microglia may contribute to the detrimental effects of reactive gliosis upon detachment. The reactive gliosis may be a clinically significant limiting factor in the recovery of vision after reattachment (Anderson et al., 1986), and future attempts to reduce gliosis may also inhibit the spread of retinal degeneration into retinal portions which are not primarily destroyed.

6. Conclusions The inverse mode of photoreceptor cell degeneration, the preservation of outer segments, and the cyst development within the inner retina (together with the edematous degeneration of single groups of Mu¨ller cells) suggest that the degeneration of non-detached areas of rabbit retinas is not caused by a failure of pigment epithelial cells or by a disturbed oxygen and nutrient supply from the choroid, but rather by disturbed homeostatic mechanisms of the gliotic Mu¨ller cells. One cause of photoreceptor cell death and edema development may be the deterioration of retinal K+ clearance due to the downregulation of K+ channels in Mu¨ller cells. If these findings can be extrapolated to human patients, one may assume that disturbed glial homeostasis mechanisms contribute to the sub-optimal recovery of visual acuity after reattachment surgery and may explain (at least partially) the functional loss in attached retinal portions as well as in the macula in cases of peripheral detachments. The age-related decrease of K+ currents in human Mu¨ller cells (Bringmann, Kohen, Wolf, Wiedemann, & Reichenbach, 2003) may contribute to the decline of vision return after surgical reattachment in elderly patients (Burton, 1982). However, the proposed mechanism of Mu¨ller cell-mediated photoreceptor cell degeneration and spread of gliosis awaits confirmation by further research, and it remains to be proven whether similar alterations occur also in species with vascularized retinas.

M. Francke et al. / Vision Research 45 (2005) 2256–2267

Acknowledgements This work was supported by the Interdisziplina¨res Zentrum fu¨r Klinische Forschung (IZKF) Leipzig at the Faculty of Medicine of the University of Leipzig (Project C21), by the Sa¨chsisches Ministerium fu¨r Wissenschaft und Kunst (SMWK; HWP program), and by the Deutsche Forschungsgemeinschaft (Re 849/8-3 and Br 1249/2-1).

References Anderson, D. H., Guerin, C. J., Erickson, P. A., Stern, W. H., & Fisher, S. K. (1986). Morphological recovery in the reattached retina. Investigative Ophthalmology and Visual Sciences, 27, 168–183. Antcliff, R. J., Hussain, A. A., & Marshall, J. (2001). Hydraulic conductivity of fixed retinal tissue after sequential excimer laser ablation: Barriers limiting fluid distribution and implications for cystoid macular edema. Archives of Ophthalmology, 119, 539–544. Arruga, H. (1936). Die Netzhautablo¨sung. Barcelona. Bringmann, A., Biedermann, B., & Reichenbach, A. (1999). Expression of potassium channels during postnatal differentiation of rabbit Mu¨ller glial cells. European Journal of Neuroscience, 11, 2883–2896. Bringmann, A., Francke, M., Pannicke, T., Biedermann, B., Kodal, H., Faude, F., et al. (2000). Role of glial K+ channels in ontogeny and gliosis: A hypothesis based upon studies on Mu¨ller cells. Glia, 29, 35–44. Bringmann, A., Pannicke, T., Francke, M., Milenkovic, I., Weick, M., & Uckermann, O. (2003). Proliferation of retinal glial (Mu¨ller) cells: Role of P2 receptors and potassium channels. Biomedical Research, 14, 38–46. Bringmann, A., Kohen, L., Wolf, S., Wiedemann, P., & Reichenbach, A. (2003). Age-related decrease of potassium currents in human retinal glial (Mu¨ller) cells. Canadian Journal of Ophthalmology, 38, 464–468. Bringmann, A., Francke, M., & Reichenbach, A. (2004). Mu¨ller cells in retinopathies. In L. Hertz (Ed.), Non-neuronal cells in the nervous system: Function and dysfunction, part III: Pathological conditions (pp. 1117–1132). Amsterdam: Elsevier. Bringmann, A., Reichenbach, A., & Wiedemann, P. (2004). Pathomechanisms of cystoid macular edema. Ophthalmic Research, 36, 241–249. Burton, T. C. (1982). Recovery of visual acuity after retinal detachment involving the macula. Transactions of the American Ophthalmological Society, 80, 457–497. Chang, C. J., Lai, W. W., Edward, D. P., & Tso, M. O. (1995). Apoptotic photoreceptor cell death after traumatic retinal detachment in humans. Archives of Ophthalmology, 113, 880–886. Chisholm, I. A., McClure, E., & Foulds, W. S. (1975). Functional recovery of the retina after retinal detachment. Transactions of the Ophthalmological Society of United Kingdom, 95, 167–172. Coblentz, F. E., Radeke, M. J., Lewis, G. P., & Fisher, S. K. (2003). Evidence that ganglion cells react to retinal detachment. Experimental Eye Research, 76, 333–342. Cook, B., Lewis, G. P., Fisher, S. K., & Adler, R. (1995). Apoptotic photoreceptor degeneration in experimental retinal detachment. Investigative Ophthalmology and Visual Sciences, 36, 990–996. Cotinet, A., Goureau, O., Hicks, D., Thillaye-Goldenberg, B., & de Kozak, Y. (1997). Tumor necrosis factor and nitric oxide production by retinal Muller glial cells from rats exhibiting inherited retinal dystrophy. Glia, 20, 59–69.

2265

de Juan, E., Loewenstein, A., Bressler, N. M., & Alexander, J. (1998). Translocation of the retina for management of subfoveal choroidal neovascularization. II. a preliminary report in humans. American Journal of Ophthalmology, 125, 635–646. de Kozak, Y., Cotinet, A., Goureau, O., Hicks, D., & ThillayeGoldenberg, B. (1997). Tumor necrosis factor and nitric oxide production by resident retinal glial cells from rats presenting hereditary retinal degeneration. Ocular Immunology and Inflammation, 5, 85–94. de Kozak, Y., Naud, M. C., Bellot, J., Faure, J. P., & Hicks, D. (1994). Differential tumor necrosis factor expression by resident retinal cells from experimental uveitis-susceptible and -resistant rat strains. Journal of Neuroimmunology, 55, 1–9. Drescher, K. M., & Whittum-Hudson, J. A. (1996). Herpes simplex virus type 1 alters transcript levels of tumor necrosis factor-alpha and interleukin-6 in retinal glial cells. Investigative Ophthalmology and Visual Sciences, 37, 2302–2312. Erickson, P. A., Fisher, S. K., Anderson, D. H., Stern, W. H., & Borgula, G. A. (1983). Retinal detachment in the cat: The outer nuclear and outer plexiform layers. Investigative Ophthalmology and Visual Sciences, 24, 927–942. Erickson, P. A., Guerin, C. J., & Fisher, S. K. (1990). Tritiated uridine labeling of the retina: Changes after retinal detachment. Experimental Eye Research, 51, 153–158. Faude, F., Francke, M., Makarov, F., Schuck, J., Ga¨rtner, U., Reichelt, W., et al. (2001). Experimental retinal detachment causes widespread and multilayered degeneration in rabbit retina. Journal of Neurocytology, 30, 379–390. Fisher, S. K., & Anderson, D. H. (2001). Cellular effects of detachment on the neural retina and the retinal pigment epihtelium. In S. J. Ryan & C. P. Wilkinson (Eds.), Retina (pp. 1961–1986). St. Louis: Mosby. Fisher, S. K., Erickson, P. A., Lewis, G. P., & Anderson, D. H. (1991). Intraretinal proliferation induced by retinal detachment. Investigative Ophthalmology and Visual Sciences, 32, 1739–1748. Fisher, S. K., & Lewis, G. P. (2003). Mu¨ller cell and neuronal remodeling in retinal detachment and reattachment and their potential consequences for visual recovery: A review and reconsideration of recent data. Vision Research, 43, 887–897. Foulds, W. S. (1963). Experimental retinal detachment. Transactions of the Ophthalmological Society of United Kingdom, 83, 153–170. Frade, J. M., & Barde, Y. A. (1998). Microglia-derived nerve growth factor causes cell death in the developing retina. Neuron, 20, 35–41. Francke, M., Faude, F., Pannicke, T., Bringmann, A., Eckstein, P., & Reichelt, W. (2001). Electrophysiology of rabbit Muller (glial) cells in experimental retinal detachment and PVR. Investigative Ophthalmology and Visual Sciences, 42, 1072–1079. Francke, M., Uhlmann, S., Pannicke, T., Goczalik, I., Uckermann, O., & Weick, M. (2003). Experimental dispase-induced retinopathy causes up-regulation of P2Y receptor-mediated calcium responses in Muller glial cells. Ophthalmic Research, 35, 30–41. Francke, M., Weick, M., Pannicke, T., Uckermann, O., Grosche, J., & Goczalik, I. (2002). Upregulation of extracellular ATP-induced Muller cell responses in a dispase model of proliferative vitreoretinopathy. Investigative Ophthalmology and Visual Sciences, 43, 870–881. Geller, S. F., Lewis, G. P., & Fisher, S. K. (2001). FGFR1, signaling, and AP-1 expression after retinal detachment: Reactive Mu¨ller and RPE cells. Investigative Ophthalmology and Visual Sciences, 42, 1363–1369. Harada, T., Harada, C., Nakayama, N., Okuyama, S., Yoshida, K., & Kohsaka, S. (2000). Modification of glial-neuronal cell interactions prevents photoreceptor apoptosis during light-induced retinal degeneration. Neuron, 26, 533–541. Harada, T., Harada, C., Kohsaka, S., Wada, E., Yoshida, K., & Ohno, S. (2002). Microglia-Muller glia cell interactions control

2266

M. Francke et al. / Vision Research 45 (2005) 2256–2267

neurotrophic factor production during light-induced retinal degeneration. Journal of Neuroscience, 22, 9228–9236. Hisatomi, T., Sakamoto, T., Goto, Y., Yamanaka, I., Oshima, Y., & Hata, Y. (2002). Critical role of photoreceptor apoptosis in functional damage after retinal detachment. Current Eye Research, 24, 161–172. Hisatomi, T., Sakamoto, T., Murata, T., Yamanaka, I., Oshima, Y., & Hata, Y. (2001). Relocalization of apoptosis-inducing factor in photoreceptor apoptosis induced by retinal detachment in vivo. American Journal of Pathology, 158, 1271–1278. Humphrey, M. F., Constable, I. J., Chu, Y., & Wiffen, S. (1993). A quantitative study of the lateral spread of Mu¨ller cell responses to retinal lesions in the rabbit. Journal of Comparative Neurology, 334, 545–558. Isashiki, M., & Ohba, N. (1986). Recovery of differential light sensitivity following surgery for rhegmatogenous retinal detachment. Graefes Archives of Clinical and Experimental Ophthalmology, 224, 184–190. Isernhagen, R. D., & Wilkinson, C. P. (1988). Recovery of visual acuity following the repair of pseudophakic retinal detachment. Transactions of the American Ophthalmological Society, 86, 291–306. Ishii, M., Horio, Y., Tada, Y., Hibino, H., Inanobe, A., Ito, M., et al. (1997). Expression and clustered distribution of an inwardly rectifying potassium channel, KAB-2/Kir4.1, on mammalian retinal Mu¨ller cell membrane: Their regulation by insulin and laminin signals. Journal of Neuroscience, 17, 7725–7735. Jackson, T. L., Hillenkamp, J., Williamson, T. H., Clarke, K. W., Almubarak, A. I., & Marshall, J. (2003). An experimental model of rhegmatogenous retinal detachment: surgical results and glial cell response. Investigative Ophthalmology and Visual Sciences, 44, 4026–4034. Jackson, T. L., & Marshall, J. (2004). Fluorophore-assisted retinal break detection using antibodies to glial fibrillary acidic protein. Investigative Ophthalmology and Visual Sciences, 45, 993–1001. Kofuji, P., Biedermann, B., Siddharthan, V., Raap, M., Iandiev, I., & Milenkovic, I. (2002). Kir potassium channel subunit expression in retinal glial cells: Implications for spatial potassium buffering. Glia, 39, 292–303. Kofuji, P., Ceelen, P., Zahs, K. R., Surbeck, L. W., Lester, H. A., & Newman, E. A. (2000). Genetic inactivation of an inwardly rectifying potassium channel (Kir4.1 subunit) in mice: Phenotypic impact in retina. Journal of Neuroscience, 20, 5733–5740. Kroll, A. J., & Machemer, R. (1969). Experimental retinal detachment and reattachment in the rhesus monkey. Electron microscopic comparison of rods and cones. American Journal of Ophthalmology, 68, 58–77. Kusaka, S., & Puro, D. G. (1997). Intracellular ATP activates inwardly rectifying K+ channels in human and monkey retinal Mu¨ller (glial) cells. Journal of Physiology, 500, 593–604. Lewis, G. P., Erickson, P. A., Guerin, C. J., Anderson, D. H., & Fisher, S. K. (1992). Basic fibroblast growth factor: A potential regulator of proliferation and intermediate filament expression in the retina. Journal of Neuroscience, 12, 3968–3978. Lewis, G. P., & Fisher, S. K. (2000). Mu¨ller cell outgrowth after retinal detachment: Association with cone photoreceptors. Investigative Ophthalmology and Visual Sciences, 41, 1542–1545. Lewis, G. P., Guerin, C. J., Anderson, D. H., Matsumoto, B., & Fisher, S. K. (1994). Rapid changes in the expression of glial cell proteins caused by experimental retinal detachment. American Journal of Ophthalmology, 118, 368–376. Lewis, H., Kaiser, P. K., Lewis, S., & Estafanous, M. (1999b). Macular translocation for subfoveal choroidal neovascularization in agerelated macular degeneration: a prospective study. American Journal of Ophthalmology, 128, 135–146. Lewis, G. P., Linberg, K. A., & Fisher, S. K. (1998). Neurite outgrowth from bipolar and horizontal cells after experimental

retinal detachment. Investigative Ophthalmology and Visual Sciences, 39, 424–434. Lewis, G. P., Matsumoto, B., & Fisher, S. K. (1995). Changes in the organization and expression of cytoskeletal proteins during retinal degeneration induced by retinal detachment. Investigative Ophthalmology and Visual Sciences, 36, 2404–2416. Lewis, G. P., Mervin, K., Valter, K., Maslim, J., Kappel, P. J., & Stone, J. (1999a). Limiting the proliferation and reactivity of retinal Mu¨ller cells during experimental retinal detachment: the value of oxygen supplementation. American Journal of Ophthalmology, 128, 165–172. Li, Q., & Puro, D. G. (2002). Diabetes-induced dysfunction of the glutamate transporter in retinal Muller cells. Investigative Ophthalmology and Visual Sciences, 43, 3109–3116. Linsenmeier, R. A., & Padnick-Silver, L. (2000). Metabolic dependence of photoreceptors on the choroid in the normal and detached retina. Investigative Ophthalmology and Visual Sciences, 41, 3117–3123. Machemer, R. (1968). Experimental retinal detachment in the owl monkey. II. Histology of retina and pigment epithelium. American Journal of Ophthalmology, 66, 396–410. Machemer, R. (1998). Macular translocation. American Journal of Ophthalmology, 125, 698–700. Machemer, R., & Norton, E. W. (1969). Experimental retinal detachment and reattachment: I. Methods, clinical picture and histology. Bibliography of Ophthalmology, 79, 80–90. Marc, R. E., Murry, R. F., Fisher, S. K., Linberg, K. A., & Lewis, G. P. (1998). Amino acid signatures in the detached cat retina. Investigative Ophthalmology and Visual Sciences, 39, 1694– 1702. Marmor, M. F. (1999). Mechanisms of fluid accumulation in retinal edema. Documents of Ophthalmology, 97, 239–249. Mervin, K., Valter, K., Maslim, J., Lewis, G. P., Fisher, S. K., & Stone, J. (1999). Limiting photoreceptor death and deconstruction during experimental retinal detachment: The value of oxygen supplementation. American Journal of Ophthalmology, 128, 155–164. Milenkovic, I., Weick, M., Wiedemann, P., Reichenbach, A., & Bringmann, A. (2003). P2Y receptor-mediated stimulation of Mu¨ller glial cell DNA synthesis: Dependence on EGF and PDGF receptor transactivation. Investigative Ophthalmology and Visual Sciences, 44, 1211–1220. Moll, V., Weick, M., Milenkovic, I., Kodal, H., Reichenbach, A., & Bringmann, A. (2002). P2Y receptor-mediated stimulation of Mu¨ller glial DNA synthesis. Investigative Ophthalmology and Visual Sciences, 43, 766–773. Nagelhus, E. A., Horio, Y., Inanobe, A., Fujita, A., Haug, F. M., & Nielsen, S. (1999). Immunogold evidence suggests that coupling of K+ siphoning and water transport in rat retinal Mu¨ller cells is mediated by a coenrichment of Kir4.1 and AQP4 in specific membrane domains. Glia, 26, 47–54. Newman, E. A. (1987). Distribution of potassium conductance in mammalian Mu¨ller (glial) cells: A comparative study. Journal of Neuroscience, 7, 2423–2432. Newman, E. A. (2001). Propagation of intercellular calcium waves in retinal astrocytes and Mu¨ller cells. Journal of Neuroscience, 21, 2215–2223. Newman, E. A, & Reichenbach, A. (1996). The Mu¨ller cell: A functional element of the retina. Trends in Neurosciences, 19, 307–312. Nork, T. M., Millecchia, L. L., Stickland, B. D., Linberg, J. V., & Chao, G. (1995). Selective loss of blue cones and rods in human retinal detachment. Archives of Ophthalmology, 113, 1066–1073. Pannicke, T., Iandiev, I., Uckermann, O., Biedermann, B., Kutzera, F., Wiedemann, P., et al. (2004). A potassium channel-linked mechanism of glial cell swelling in the postischemic retina. Molecular and Cellular Neuroscience, 26, 493–502.

M. Francke et al. / Vision Research 45 (2005) 2256–2267 Pederson, J. E. (1994). Fluid physiology of the subretinal space. In C. P. Wilkinson (Ed.), Retina (pp. 1955–1968). St. Louis: Mosby. Regillo, D., & Bensen, W. E. (1988). Retinal detachment: Diagnosis and management. Philadelphia: JB Lippincott (pp. 1–30). Rex, T. S., Fariss, R. N., Lewis, G. P., Linberg, K. A., Sokal, I., & Fisher, S. K. (2002). A survey of molecular expression by photoreceptors after experimental retinal detachment. Investigative Ophthalmology and Visual Sciences, 43, 1234–1247. Roque, R. S., Rosales, A. A., Jingjing, L., Agarwal, N., & Al-Ubaidi, M. R. (1999). Retina-derived microglial cells induce photoreceptor cell death in vitro. Brain Research, 836, 110–119. Sasoh, M., Yoshida, S., Kuze, M., & Uji, Y. (1997). The multifocal electroretinogram in retinal detachment. Advances in Ophthalmology, 94, 239–252. Uckermann, O., Grosche, J., Reichenbach, A., & Bringmann, A. (2002). ATP-evoked calcium responses of radial glial (Mu¨ller) cells in the postnatal rabbit retina. Journal of Neuroscience Research, 70, 209–218. Uckermann, O., Iandiev, I., Francke, M., Franze, K., Grosche, J., & Wolf, S. (2004). Selective staining by vital dyes of Mu¨ller glial cells in retinal wholemounts. Glia, 45, 59–66.

2267

Uckermann, O., Uhlmann, S., Weick, M., Pannicke, T., Francke, M., & Reichenbach, A. (2003). Upregulation of P2Y receptormediated calcium responses in glial cells during experimental detachment of the rabbit retina. Neuroscience Letters, 338, 131–134. Uhlmann, S., Bringmann, A., Uckermann, O., Pannicke, T., Weick, M., & Ulbricht, E. (2003). Early glial cell reactivity in experimental retinal detachment: effect of suramin. Investigative Ophthalmology and Visual Sciences, 44, 4114–4122. Weick, M., Wiedemann, P., Reichenbach, A., & Bringmann, A. (in press). Resensitization of P2Y receptors by growth factor-mediated activation of the phosphatidylinositol-3 kinase in retinal glial cells. Investigative Ophthalmology and Visual Sciences, 45. Wen, R., Song, Y., Cheng, T., Matthes, M. T., Yasumura, D., & LaVail, M. M. (1995). Injury-induced upregulation of bFGF and CNTF mRNAs in the rat retina. Journal of Neuroscience, 15, 7377–7385. Zacks, D. N., Ha¨nninen, V., Pantcheva, M., Ezra, E., Grosskreutz, C., & Miller, J. W. (2003). Caspase activation in an experimental model of retinal detachment. Investigative Ophthalmology and Visual Sciences, 44, 1262–1267.