If Müller cells are to function as antigen-presenting cells, they must be able to take up and process antigens. Indeed, Müller cells have been found to phagocytize a variety of substances. The earliest documented evidence of phagocytic activity by Müller cells was reported more than six decades ago by Friedenwald and Chan (1932). After injecting a suspension of melanin granules into the vitreous of albino rabbits, they noted that the pigment accumulated in monocytic phagocytes and Müller cells. Subsequently, it was shown that a variety of materials, such as carbon and copper particles, erythrocyte debris, and subretinal hemorrhage, are phagocytized by Müller cells (Algvere and Kock, 1983; Rosenthal and Appleton, 1975; Miller et al., 1986; Koshibu, 1978; Mano and Puro, 1990). In human eyes with chalcosis, copper particles are found in retinal glial cells (Rao et al., 1976).
The phagocytic activity of Müller cells has been examined in Müller cell cultures and also in experimental animals. Mano and Puro (1990) found that Müller cell cultures from postmortem eyes were able to phagocytize retinal cell fragments as well as latex beads (Fig. 6.5). Furthermore, they reported that uptake was partially dependent on extracellular [Ca2+] and was blocked by the calcium channel blocker, nifedipine. Phagocytic activity was also reduced by the addition of 8-bromo-cAMP while vitamin D3 stimulated uptake suggesting that the phagocytic process used by Müller cells is similar to that associated with macrophages.
Freshly dissociated Müller cells, Müller cell cultures, and isolated rabbit retina have been shown to take up latex beads (Stolzenburg et al., 1992). In addition, Müller cells have been reported to phagocytize egg-lecithin- ' coated silicone particles following intraocular injection (Nishizona et al., 1993). Similarly, Müller cells in explant cultures of goldfish retina also accumulate latex beads, although no uptake was observed in vivo (Li et al., 1987; Wagner and Raymond, 1991). Phagocytosis of exogenous particles, cell debris, and hemorrhagic products may be an important scavenging func-
Figure. 6.5. Phagocytosis by Müller cell cultures. Electron micrograph of a Müller cell exposed to retinal debris. Arrows point to cellular debris inside the Müller cell. The histogram shows the effect ofvarious conditions on the percentage of cells exhibiting phagocytic activity (open bar) and the mean fluorescence per cell (hatched bar). Mean changes from the appropriate control values are shown. Glial cell cultures were preincubated in the presence of 1 mM EGTA, 10 mM nifedipine, 1 mM 8-bromo-cyclic AMP, or 10 mM vitamin D3, and exposed to microspheres and subsequently analyzed by flow cytometry (Mano and Puro, 1990). (Copyright 1990 Association for Research in Vision and Ophthalmology, reprinted with permission.)
tion of Müller cells. Indeed, it is conceivable that Müller cells phagocytose outer segment membranes in degenerative retinal diseases in which the RPE is unable to perform this function.
6.4. RETINAL CONDITIONS WITH POTENTIAL MÜLLER CELL INVOLVEMENT
There is no compelling evidence that the Müller cell is the primary site affected in any retinal disease. However, there are many retinal conditions in which Müller cells appear to play a prominent role. For instance, in nonproliferative diabetic retinopathy, Müller cells are a major site for the synthesis and secretion of VEGF, an angiogenic agent that contributes to vascular proliferation (Amin et al., 1997). Gass (1999) postulated that the Müller cell cone, a cellular component of fovea centralis, is involved in the pathogenesis of some macular diseases such as idiopathic macular holes and foveomacular schisis. Retinal disorders in which there is potential for direct Müller cell involvement are considered in the next section.
Juvenile retinoschisis (also termed congenital hereditary retinoschisis) is a rare, X-linked recessive disease that develops early in life and leads to parafoveal intraretinal cysts and marked vision loss in affected males (Yanoff et al., 1968; Harris and Yeung, 1976; Condon et al., 1986). The disease is characterized by bilateral splitting of the retina, often involving the macula and especially the inferior temporal quadrant of the peripheral retina. Retinal splitting at the level of the nerve fiber layer has been observed in histopathology studies (Condon et al., 1986) (Fig. 6.6). A major clinical feature of the disease is that the b-wave of the ERG is drastically reduced although the a-wave is normal (Harris and Yeung, 1976). Since Müller cells participate in the generation of the b-wave (Chapter 5), the disorder has been linked to Müller cell dysfunction (Harris and Yeung, 1976; Condon et al., 1986), and as expected, ultrastructural changes are seen in the Müller cells near the internal limiting membrane (Condon et al., 1986).
The Mizuo phenomenon (also known as Mizuo-Nakamura phenomenon) refers to a rapid change in the color of the fundus from red in the dark-adapted state to golden after light onset (Mizuo, 1913). A similar fundus appearance has been reported in X-linked retinoschisis (de Jong et al., 1991). Interestingly, a transient whitish-yellow sheen is also seen in "spreading depression" (see Chapter 5), a phenomenon triggered by a rise in extracellular K+ in the retina. Accordingly, deJong et al. (1991) suggested
Figure. 6.6. Histopathology of juvenile X-linked retinoschisis. A histological section showing retinal split in the nerve fiber layer (NFL). INL, inner nuclear layer, and ON1, outer nuclear layer (Eagle, 1999). (Copyright 1999 W.B. Saunders, reprinted with permission.)
that the color changes associated with retinoschisis result from spreading depression due to the inability of the damaged Müller cell endfeet to clear K+ from the extracellular space.
The gene associated with X-linked juvenile retinoschisis (XLRS1) has been cloned (Sauer et al. 1997). It encodes a 224-amino acid protein containing a "discoidin" domain at the C-terminus, which suggests that the protein is involved in cell-cell interactions such as cell adhesion or intercellular signaling. Surprisingly, the gene is expressed in photoreceptors but not in Müller cells (Reid et al., 1999). It remains to be learned whether the retinal splitting results from a direct action of the mutant protein on the Müller cell or is a secondary response to neuronal dysfunction.
Macular edema is yet another ocular disease in which Müller cells have been implicated. Cystoid macular edema (CME) is a condition in which there is an accumulation of fluid in the macula (Gass and Norton, 1966; Jampol, 1994). It is associated with a variety of ocular conditions that include retinovascular diseases, retinal degenerations, intraocular inflammation, tumors, and most frequently, as a complication of postcataract surgery (Gass and Norton, 1966; Jampol, 1994). A common feature of the disorder is the presence of clear cysts in Henle's layer in the foveal region and occasionally in other retinal layers (Fig. 6.7). CME involves breakdown of the blood-retina barrier with accumulation of serous transudates in the retina, but no single etiologic factor is likely to be responsible for the occurrence of CME in such a wide variety of conditions. The question as to whether the fluid
Figure. 6.7. Histology of cystoid macular edema. Retina from a patient with pseudophakic cystoid macular edema. Note the presence of large cystic spaces in the outer (asterisk) and inner retinas. Courtesy of Dr. Deepak Edward, University of Illinois Medical School.
accumulates in extracellular spaces or within Müller cells remains controversial owing to limited access to clinical specimens at early stages of CME development.
Early histopathological studies showing swelling and degeneration of Müller cells led to the suggestion that the cystoid spaces represented swollen Müller cell processes (Fine and Brucker, 1981;Yanoffet al., 1984). Either the edema, subsequent necrosis of Müller cells, or both, may cause further degenerative changes and the formation of larger cystic spaces. At the ultrastructural level, Müller cell enlargement occurs without much accumulation of extracellular fluid (Fine and Brucker, 1981). This suggests that the primary site of edema is the Müller cell itself, and that the buildup of extracellular fluid is a secondary event (Yanoff et al., 1984). A similar conclusion was reached based on a histopathologic study of dominantly inherited cystoid macular edema, a rare clinical entity (Loeffler et al., 1992).
The results of other studies, however, suggest that the fluid is accumulated primarily in the extracellular space, and that neuronal loss and Müller cell changes are secondary events (Gass et al., 1985; Tso, 1982 Wolter, 1981). In an ultrastructural study of an eye with CME, Gass et al. (1985) reported finding no cellular components in the enlarged spaces of the INL and the OPL. Surprisingly, they found no evidence for Müller cell degeneration or necrosis. In an extensive histopathological study of eyes with CME, Tso (1982) reported finding cystic spaces in the outer and inner retina, but was unable to resolve the question of whether the spaces were due to extracellular fluid accumulation or Müller cell swelling.
Although results from the various CME studies appear to be at odds, the observed differences might also be due to a number of complicating factors: stage of disease; processing of the tissue; prior systemic disease or treatment; and previous surgical or medical history. In this context, development of a representative animal model for CME would be of great value (Tso, 1982; Bellhorn, 1984), but at present no suitable model exists.
6.4.3. Müller Cell Sheen Dystrophy
In addition to retinoschisis and CME, Müller cell dysfunction has been implicated in an autosomal dominant retinal dystrophy termed "Müller cell sheen dystrophy" (Polk et al., 1997; Kellner et al., 1998). In this condition, the fundus shows multiple folds at the level of the internal limiting membrane at the posterior pole. In some cases, this is accompanied by macular edema. Histological studies indicated a diffuse thickening and undulation of the internal limiting membrane (Polk et al., 1997), and the ERG shows a reduction in the amplitude of the b-wave and flicker response (Kellner et al., 1998). It has been proposed that membrane thickening and subsequent
folding result from accumulation of a protein synthesized and secreted by Müller cells (Kellner et al., 1998).
Retinal detachment results not only in physical separation of the neural retina from the RPE, but also leads to rapid cellular changes in the retina and RPE (Fisher and Anderson, 1994). The condition can have a potentially devastating effect on visual function. Experimental studies of retinal detachment have been made possible by the development of a superb animal model in cats where the cellular changes accompanying retinal detachment have been extensively documented (Fisher and Anderson, 1994). In cats with surgically detached retinas, detachment evokes striking cytological changes in photoreceptors and Müller cells (Fig. 6.8; Fisher and Anderson, 1994; Okada et al., 1990). In the early stages of detachment, Müller cells undergo hypertrophy and proliferation. At more advanced stages, Müller cell processes extend into subretinal space to cover the entire area of detachment, resulting in a scar-like structure in the distal retina (Fisher and Anderson, 1994; Erickson et al., 1987; Fisher et al., 1991; Lewis et al., 1994). These morphological changes are accompanied by increases in RNA and protein synthesis in Müller cells (Erickson et al., 1990; Fisher and Anderson, 1994).
The consequences of the Müller cell changes can be profound. Formation of abnormal membranes in the subretinal space results in subretinal fibrosis which can prevent regeneration of photoreceptor outer segments after reattachment surgery and thus can lead to subsequent photoreceptor degeneration (Fisher and Anderson, 1994; Korte et al., 1992).
6.4.5. Müller Cells in Retinal Membranes
In many ocular conditions, such as proliferative diabetic retinopathy, a series of pathological processes result in cells invading the vitreous, adhering to the ILM, proliferating, and secreting extracellular matrix molecules. The net effect of these changes is the formation of epiretinal membranes
Figure 6.8. Cytological changes in Müller cells following retinal detachment. A. The electron micrograph shows a Müller cell nucleus (MN) that has migrated to the outer plexiform layer after detachment of cat retina. PN, photoreceptor nucleus. B. Micrograph showing numerous 10 nm filaments in the Müller cell cytoplasm; some of the filaments show immunolabeling for GFAP (gold particles). C. Micrograph showing extensive Müller cell (MC) hypertrophy in the subretinal space (SRS) following detachment (Fisher and Anderson, 1994). (Copyright 1994 W. B. Saunders Co./Mosby, reprinted with permission.)
and development of traction forces on the retina (McDonald et al., 1994). The presence of epiretinal membranes leads to adverse effects on vision and is frequently associated with ocular morbidity (McDonald et al., 1994). Epiretinal membranes appear to contain multiple cell types, and it was generally assumed that most are derived from astrocytes, endothelial cells, fibroblasts, pericytes, macrophages, and RPE cells. Recently, however, it was shown that Müller cells participate in the formation of epiretinal membranes (Guerin et al., 1990b; Kono et al., 1995).
The availability of Müller cell-specific antibodies has been crucial in this regard. From previous work it was known that GFAP is expressed by fibrous astrocytes and reactive Müller cells but not by RPE, fibroblasts, or macrophages (Sarthy, 1991). CRALBP, on the other hand, is expressed by Müller cells and RPE but not by astrocytes (Bunt-Milam and Saari, 1983). Based on these antigenic differences, Guerin et al. (1990) used double labeling methods to distinguish Müller cells from fibrous astrocytes and RPE in epiretinal membranes. They found cells that were GFAP+/CRALBP- (astrocytes), GFAP/CRALBP+ (RPE cells), and GFAP+/CRALBP+ (Müller cells). These data suggest that in addition to astrocytes and RPE cells, Müller cells are an important component of epiretinal membranes (Fig. 6.9).
Was this article helpful?