Constitutive and Inducible Heat Shock Protein 70 Immunoreactivity in the Normal Rat Eye

Abstract

purpose. Distributions for the constitutive and inducible 70-kDa heat shock proteins, Hsc70 and Hsp70, in different parts of the rat eye are likely to be related to the metabolic demands required for absorption and detection of light. This study was conducted to better understand the functions of Hsc70 and Hsp70 in these tissues and to provide a basis for elucidating their contributions to the maintenance and repair of ocular structures subsequent to tissue injury or cellular degeneration.

methods. Eyes from male Sprague–Dawley rats (200–300 g) were fixed in methacarn and embedded in paraffin. Sagittal sections (10 μm) through the optic nerve were stained with hematoxylin and eosin or incubated with heat shock protein antibody or control IgG. Bound antibody was visualized using an avidin–biotin–horseradish peroxidase detection system.

results. Hsc70 immunoreactivity was detected in all layers of the retina, except the outer segments. In the retinal pigment epithelium, staining was restricted to cells near the optic nerve–retina junction. Intense staining was also observed in glial nuclei of the optic nerve, whereas weaker staining was observed in the basal and wing cells of the limbal and corneal epithelia. In contrast, Hsp70 immunoreactivity was restricted to the outer nuclear layer and inner segments of the retina. Hsp70 staining was also prominent in basal and wing cells of the limbal cornea and to a lesser extent in the central corneal epithelium. The optic nerve was Hsp70 negative.

conclusions. Hsc70 and Hsp70 have distinct distributions in the normal rat eye, which imply regional and cell-specific functions.

When living cells are exposed to metabolic stress, they respond by increasing transiently the expression of a small subset of proteins, known as the heat shock proteins (Hsps). These proteins, so named because temperature elevation (heat shock) was the first described inducer of the response, have been found in virtually all organisms and are highly conserved phylogenetically (for review, see ^{Refs. 1 2 3 4 5} ). In mammals, Hsps are known to be induced by a wide range of noxious stimuli, such as hypoxia and ischemia, as well as by damaging chemical agents, such as metabolic poisons and toxic heavy metals. 

Hsps are divided into families based on their molecular weights. The 70-kDa family is composed of several related proteins that are found commonly in distinct intracellular localizations. ^{6 7} In mammalian cells, they are Hsc70 (cytoplasm/nucleus) ^{8 9} ; Hsp70 (cytoplasm/nucleus) ^{10 11 12 13 14} ; Grp78/Bip (ER lumen) ^{15 16} ; and Grp75/mtp70 (mitochondrial matrix) ^{17 18} and Ly-hsp70 (lysosomes). ¹⁹ Hsc70 and Hsp70 are interesting because not only are they found in the same cellular compartments but they are highly homologous (∼95%) and show similar biological properties. ²⁰ Moreover, Hsc70 is synthesized constitutively in mammalian cells and is only moderately stress-inducible, whereas Hsp70 is usually undetectable, under normal conditions but is highly induced in cells experiencing stress. The proteins are often referred to as the constitutive (cognate) and inducible forms of the 70-kDa Hsp. 

The first evidence that Hsps play a crucial role in the eye came from the work of Barbe et al. ²¹ They showed that induction of Hsp synthesis by hyperthermia correlated with the time when photoreceptors were protected from light-induced damage. ²¹ They also documented that after hyperthermia, retinal levels of several Hsps were maximally elevated with the same time course as the protection from light damage. ²¹ More recently, it was discovered, using an antibody against both forms of the 70-kDa Hsp (anti-Hsc70/Hsp70), that unstressed animals have high levels of Hsc70, Hsp70, or both in many ocular tissues, including the retina and cornea, and that these levels can be increased by the onset of retinal degeneration in a rat model of retinal dystrophy. ²² Interestingly, additional studies, using an antibody specific for Hsp70 (anti-Hsp70), indicated that there were only low levels of Hsp70 immunoreactivity in the normal rat retina and cornea. ^{23 24} These observations might lead one to hypothesize that the intense staining observed using anti-Hsc70/Hsp70 was primarily due to reactivity for Hsc70. Although this may, or may not, be true, the story is clearly more complicated. For example, another study, which compared immunoreactivity for anti-Hsc70/Hsp70 and anti-Hsp70 in fresh rat corneas, suggests that Hsp70 is the primary form of 70-kDa Hsp in this tissue. ²⁵  

In the study reported here, we examined the distribution of Hsc70 and Hsp70 immunoreactivity in normal ocular structures to gain a better understanding of the function of these proteins in the normal eye. Furthermore, a comprehensive study of the distribution of each isoform in the major structures of the eye has not yet been done, and that information will be needed to evaluate Hsc70 and Hsp70 changes after light-induced ocular damage, a project currently in progress (Dean DO and Tytell M, unpublished). For the present study, we made use of two fairly recent technical improvements that have permitted a more accurate assessment of tissue immunoreactivity for these proteins. The first was the development of an antibody specific for Hsc70, ²⁶ which allowed the two forms to be distinguished unequivocally. The second was the discovery that methacarn fixative, ^{27 28} a modification of Carnoy’s solution, markedly improved retention of Hsc70 and Hsp70 immunoreactivity (in rat tissue, Kent CR and Tytell M, unpublished observations; and in human tissue ²⁹ ). We report here that the patterns of staining with the constitutive and inducible 70-kDa Hsp antibodies were distinct in most structures of the eye. Hsc70 reactivity was strongly expressed in the optic nerve, throughout much of the retina, in the epithelium of the cornea, including the limbal region, and in the ciliary body. On the other hand, Hsp70 reactivity was restricted to the inner segments and nuclei of the photoreceptors, as well as the outer limiting membrane of the retina and the corneal and limbal epithelia. These results imply that there are regional and cell-type–specific differences in the expression of Hsc70 and Hsp70 in the normal rat. They also suggest that Hsp70 is not just a stress-activated protein. 

Materials and Methods

Animal Handling and Euthanasia

Animals were handled in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Male Sprague–Dawley rats (200–300 g) were maintained between 21°C and 24°C on a 12-hour light/12-hour dark cycle with food and water ad libitum. The light intensity in the housing cages was less than 2 foot candles (ft-c). For euthanatization, the rats were transferred to portable cages covered with a blanket and transported to the dissection room. The animals were allowed to settle for approximately 30 minutes, still with the blanket in place, before they were killed by CO₂ inhalation. The eyes or appropriate ocular tissues were removed, as quickly as possible. The dissection room was maintained between 22°C and 24°C with a light level of not more than 6 foot-candle (fc) during the euthanatization procedure. All animals were killed between 1 and 3 PM. 

Western Blot Analysis

Retinal tissue was dissected, homogenized in sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) buffer (2% SDS, 20% glycerol, 5% β-mercaptoethanol, 0.01% bromophenol blue, 10 mM Tris–HCl [pH 8.0]), 1 mM phenylmethylsulfonyl fluoride and 1 mM EDTA) and boiled for 3 minutes. Aliquots (2 μg) were then separated on an 8% SDS–polyacrylamide minigel (Bio-Rad). Samples of purified bovine Hsc70 (∼90% Hsc70 and ∼10% Hsp70; SPP-750; StressGen Biotechnologies) and recombinant human Hsp70 (>90% Hsp70; SPP-755; StressGen Biotechnologies) were included as controls. The proteins in the gel were transferred electrophoretically to nitrocellulose (Hoefer Mighty Small Transphor System) and the filter split into sections. After blocking for 1 hour with 5% blotto (5% nonfat milk in 10 mM phosphate-buffered saline [pH 7.4], 0.1% Tween-20 [PBS–T]), the membranes were incubated for 1 hour in 5% blotto containing either the Hsc70 (0.25 μg/ml; SPA-810; StressGen), or Hsp70 (1 μg/ml; SPA-815; StressGen) primary antibody, followed by incubation for 45 minutes in 5% blotto containing either the peroxidase-conjugated goat anti-rat (1 μg/ml; Kirkegaard & Perry Laboratories, KPL) or goat anti-mouse (1 μg/ml; KPL) secondary antibody, as appropriate. Immune complexes were visualized using the ECL detection system (Amersham). 

Preparation of Tissue for Paraffin Embedding

Eyes were excised and immersed overnight at room temperature, in methacarn solution (60% methanol, 30% 1,1,1-trichloroethane (TCE), 10% acetic acid). The next day the lens was extracted from each eye and discarded, as described previously. ²⁴ Eyes were then immersed in a series of graded isopropanol solutions, passed through several changes of TCE (at room temperature) and paraffin (infiltration medium; Surgipath Medical Industries), maintained at 60°C, and embedded in paraffin (embedding medium, Surgipath), using standard techniques. Ten-micrometer-thick sagittal sections were collected from each eye and mounted onto ProbeOnPlus microscope slides (Fisher Scientific). Only sections including the optic nerve were used for histology and immunohistochemistry. 

Histology

Slide-mounted sections were deparaffinized and rehydrated according to normal histologic procedures. Tissue sections were stained with hematoxylin and eosin. Bluing reagent was used to neutralize the hematoxylin. Slides were dehydrated, cleared in xylene, and placed on a coverslip with Shandon-Mount. Hematoxylin, eosin, bluing reagent, and Shandon-Mount were purchased from Shandon, Inc. 

Immunohistochemistry

Slide-mounted sections were deparaffinized, rehydrated, and blocked with 10% normal goat serum (NGS; Vector Laboratories) in PBS–T, followed by incubation overnight with one of the primary antibodies in PBS–T, plus 2% NGS. The primary antibodies were anti-Hsc70 (0.25 μg/ml) and anti-Hsp70 (5 μg/ml) or (instead of those antibodies) rat IgG (0.25 μg/ml; Sigma) and mouse ascites fluid (5 μg/ml; Sigma). The latter were used to monitor nonspecific immunoreaction product. The next day slides were incubated for 1 hour in PBS–T, containing biotinylated secondary antibody, followed by 1 hour in PBS–T containing peroxidase-conjugated streptavidin (1μ g/ml; KPL). The secondary antibodies were goat anti-rat (1 μg/ml; KPL), in the case of rat IgG and anti-Hsc70, and goat anti-mouse (1μ g/ml; KPL), in the case of mouse ascites fluid and anti-Hsp70. Bound antibody was visualized with the DAB-nickel detection system (Vector). 

Results

Specificity of Anti-Hsc70 and Anti-Hsp70

Comparison of the patterns of ocular staining for Hsc70 and Hsp70 requires that the antibodies be selective for these two closely related proteins. To document that specificity, western blot analysis was performed using purified samples of bovine Hsc70 and recombinant human hsp70, as well as homogenates of whole retinas. The results are shown in Figure 1 . Results indicated that anti-Hsc70 recognized the constitutive (Fig. 1A , lane C) but not the inducible (Fig. 1A , lane I) form of the protein. Anti-Hsc70 also recognized a band of the expected size in the retinal homogenate (Fig. 1A , lane R). In contrast, anti-Hsp70 recognized the inducible but not the constitutive form of the protein (Fig. 1B , lanes C and I). The weak signal with bovine Hsc70 was due to the small amount (∼10%) of Hsp70 in this preparation. It was also noted that the faint Hsp70 band migrated slightly slower than expected (Fig. 1B , lane C). This probably reflects some steric hindrance in the mobility of the Hsp70 protein by the slightly larger, and more prevalent, Hsc70 protein. Anti-Hsp70 recognized a band of the expected size in the retinal homogenate (Fig. 1B , lane R), which suggested that detectable levels of inducible Hsp70 were present in the unstressed retina. Because the intensity of this signal was lower than the corresponding signal with anti-Hsc70 (compare Figs. 1A and 1B , lane R), this blot also suggested that Hsc70 was the predominant 70-kDa Hsp isoform in the unstressed retina. The observations noted here regarding the specificity of the individual antibodies are consistent with previous findings. ^{26 30}  

Retina

Hsc70 reactivity was observed in all layers of the normal retina, from the retinal ganglion cell (RGC) layer to the inner segments (IS) of the photoreceptors (Fig. 2) . The most prominent staining was observed in the IS and outer limiting membrane (OLM). Staining was also observed in RGC axons. In this case, reactivity was most visible at the optic disc (Fig. 3) . Only in two layers of the retina was there was no Hsc70 immunoreactivity, the outer segments of the photoreceptors, and the retinal pigment epithelium (RPE), distal to the optic nerve (Figs. 2 and 3) . Intriguingly, RPE cells, which were located less than 200 to 250 μm from the optic disc, showed relatively high levels of Hsc70 immunoreactivity, particularly within the cell nuclei (Fig. 3) . Background immunoreaction product was absent from all regions of the retina and optic nerve, as indicated by the lack of binding with nonimmune IgG (Figs. 2 and 3) . 

Hsp70 immunostaining was observed exclusively in the IS, the OLM, and the outer nuclear layer (ONL) of the retina (Fig. 2) . Interestingly, the reactivity along the length of the OLM appeared punctate in comparison to the relatively continuous immunostaining seen with anti-Hsc70 (insets, Fig. 2 ). No reactivity was observed in any other region of the retina, even near the optic nerve (Fig. 3) . There was no nonspecific staining with mouse ascites fluid (not shown), which was used as a negative control. 

Optic Nerve

Hsc70 immunoreactivity was detected in the cytoplasm of many glial cells (solid black arrows, Fig. 4 ), although closer inspection revealed that most of the staining was perinuclear (see inset, Fig. 4 ). Reactivity was also detected in the nerve fibers of this structure, in which the immunostaining appeared as wavy lines running in a predominantly horizontal direction across the section (compare Hsc70 panel to that of rat IgG and Hsp70, Fig. 4 ). An example of one of these threads is indicated by the black arrow with a white center (Fig. 4) . Whether that staining corresponded to axons or to the myelin sheath was impossible to resolve in these sections. No staining was observed with anti-Hsp70 (Fig. 4) or with antibody controls consisting of nonimmune rat IgG (Fig. 4) and mouse ascites fluid (not shown). 

Central and Peripheral Cornea and Limbus

We examined the abundance of Hsc70 and Hsp70 proteins in the central (Fig. 5) and peripheral regions of the cornea, as well as in the limbal epithelium (Fig. 6) . Relatively low levels of Hsc70 immunoreactivity were detected in the central cornea. Staining was localized to the cytoplasm of the epithelial basal cells (solid black arrow, Fig. 5 ) and the nucleus of an occasional wing cell (black arrows with white center, Fig. 6 ). No reactivity was observed in the superficial layer of the epithelium. In some of the stromal cells, or collagen fibers and endothelial cells, there were traces of patchy staining (Fig. 5) . However, the specificity was equivocal, especially in the case of endothelium, because that reactivity was seen also in tissue treated with nonimmune rat IgG. 

Specific Hsc70 immunoreactivity was observed also in the epithelial layer of the peripheral cornea and limbus (Fig. 6) . Most of the staining here was localized in the cytoplasm of the basal cells, as was described for the central cornea (compare Figs. 5 and 6 ). Intriguingly, the intensity of Hsc70 reactivity was higher in the basal cells of this region than in the same cells of the central cornea. There was also reactivity in numerous wing cells within this region (e.g., see solid arrow in Fig. 6 ), which was in contrast to the mostly negative central corneal wing cells (Fig. 5) . This staining pattern was also observed, although to a lesser extent, when peripheral corneal epithelium was compared with that of the central cornea. Taken together these results indicated that there were variations in Hsc70 levels across the entire length of the epithelial layer and that the highest levels were found in the peripheral cornea and limbus. This finding was interesting because wing cell layers of the peripheral cornea and limbus are thinner than the corresponding layer in the central cornea (compare Figs. 5 and 6 ). 

Of particular significance was the finding that Hsp70 immunostaining was much more prominent than that for Hsc70 in the central corneal epithelium (Fig. 5) . This result contrasted with the pattern of staining in all other ocular structures. Hsp70 reactivity was detected in the nuclei (see inset, Fig. 5 ) of most wing cells, including those bordering the superficial layer (black arrows with white center, Fig. 5 ). There was also considerable Hsp70 immunoreactivity in the cytoplasm of numerous basal cells (solid black arrow, Fig. 5 ) and in general the intensity of staining within basal cells seemed to be greater for Hsp70 than for Hsc70. The cytoplasm of basal cells of the limbus also contained considerable Hsp70 immunoreactivity (Fig. 6) . Like Hsc70, the intensity of staining was greater in the basal cells of this region than in the corresponding cells of the central cornea (compare Fig. 6 to Fig. 5 ). More surprisingly, the reactivity showed that the nuclei of only a few isolated wing cells stained for Hsp70. It appeared that there was a transition, from the majority of the wing cells being Hsp70 immunopositive to the majority of them being Hsc70 immunopositive, as one scans from the center of the cornea to the more peripheral regions. In sum, these results suggest that Hsp70 is more abundant than Hsc70 in the central region of the cornea but that the two forms are distributed equivalently in peripheral corneal and limbal epithelial cells. This location-dependent shift in the relative proportions of Hsc70 to Hsp70 appears to be the consequence of higher levels of Hsc70 in the peripheral cornea and limbus, rather than a decrease in overall Hsp70 levels. 

Ciliary Body and Iris

Hsc70 reactivity was observed exclusively in the cytoplasm of the ciliary body epithelial cells (Fig. 7) . The most intense staining was juxtanuclear, as indicated by the dark line surrounding the nuclei that can be seen on the lower left corner of the inset. No Hsc70 immunoreactivity was observed in epithelial cell nuclei or in other cell types of the ciliary body. Staining for Hsp70 was not detected in the ciliary body. 

Inspection of the iris where it joins the ciliary body suggested that it contained some staining for both 70-kDa Hsps (Fig. 7) . Hence, the entire length of this structure was examined at higher magnification. Low levels of Hsc70 reactivity were observed at the edges of the tissue in the posterior epithelium and the cells of the anterior limiting layer (not shown). Some patchy staining for Hsp70 was also present in the limiting layer. However, the staining for both isoforms was extremely low when compared with that observed for other heat shock proteins, such as Hsp25 and Hsp90 (Dean DO and Tytell M, unpublished observations). 

Discussion

Overview

We have shown that both 70-kDa Hsps can be detected readily in normal ocular tissues. Furthermore, the expression of these isoforms is regulated in a spatial manner. Previously, studies of ocular Hsc70 and Hsp70 from this laboratory made use of only an antibody specific for Hsp70 or an antibody that did not discriminate between the two 70-kDa Hsp isoforms (anti-Hsc70/Hsp70). ^{22 23 24} Thus, the interpretation of those results was constrained. In the former case, the methodology detected the high levels of Hsp70 in the retina and cornea after hyperthermia but did not reveal the distinct pattern of Hsp70 staining present in the unstressed animal. In the latter report, the use of an antibody that reacted with Hsc70 and Hsp70 meant that the drawing of conclusions about the protein levels of each isoform was inferential. By studying the separate expression of Hsc70 and Hsp70, we have been able to show that some ocular cells appear to require Hsc70 alone, under normal metabolic conditions, whereas others use Hsp70 alone or both Hsc70 and Hsp70. Our results suggest that referring to Hsc70 as “the constitutive, or cognate, protein” and Hsp70 as“ the stress-inducible protein” is misleading, at least for the eye. We suggest that the definition of Hsp70 be modified so that less emphasis is placed on the stress-inducible nature of that protein. Growing evidence from our work and that of others indicates that both isoforms can be present in unstressed, as well as stressed, cells. 

Hsp70 Expression Is Not Due to Whole Body Stress

The presence of detectable levels of Hsp70 in any given tissue is typically associated with cell stress. Because any stress response would alter the levels and distribution of both Hsc70 and Hsp70, we thought it prudent to point out corroborating evidence from the literature that our findings were not the result of unintended stress produced when the animals were removed from their housing cages, euthanatized, and exsanguinated. There are four pieces of evidence that mitigate against this interpretation. First, peak accumulation of Hsp70 was shown to occur at approximately 18 hours after hyperthermic stress in the cornea and retina. ^{23 24} In the present study, animals were removed from their housing cages and killed in about half an hour, well before that point. Second, Hsp70 immunoreactivity was limited to a subset of retinal cells known to upregulate the protein early in the hyperthermic stress response. ²⁴ Specifically, there was staining in the ONL and IS, but not in the IPL, OPL, or RGCs. Third, Hsp70 mRNA has been detected previously in the retinas of unstressed rats. ³¹ Last, Hsp70 was found in unstressed HeLa cells ³² and in normal rat spinal cord motor neurons. ³³ Therefore, there is ample precedent for our observation of Hsp70 staining in normal ocular cells, and it is clear that we are not witnessing a reaction to some form of whole body stress. 

Retina

If the presence of Hsp70 in the retina is not due to whole body stress, then it is possible that the retina is normally in a state of stress. One explanation for this observation is that the retina is subject to light-induced stress or damage in the course of normal functioning. Certainly the absorption of light energy by the OS may produce reactive species, such as lipid peroxides, which can damage the retina. ^{34 35 36} We used a low level of light illumination (<6 fc), and we killed the rats during the second half of the light cycle, when it has been shown that the retinas of animals exposed to bright light are less susceptible to light-induced damage. ³⁷ We also examined the retinal tissue for light-induced degenerative changes. The retinas of our animals contained none of the characteristic signs of light-induced cellular damage, such as degeneration of the photoreceptor segments and thinning of the ONL. ^{38 39 40} We also saw no change in Hsp70 expression across the retina, even though it has been documented that the superior retina is more sensitive to light-induced damage than the inferior retina. ^{38 41} Thus, the expression of Hsp70 in the retinas of these animals probably is not due to light-induced stress but rather is a consequence of normal function. It should be added that hyperthermia, which is the most studied inducer of Hsps in the retina, causes elevation of Hsp70 levels in inner retinal layers, where no staining was observed under conditions used in this report. ²⁴ Therefore, unless the stress response to light is dramatically different from that for hyperthermia, this would further support our suggestion that light-induced stress seems unlikely as an explanation for our findings. 

The outer layers of the rat retina, from the OPL to the OS, are avascular and, consequently, the photoreceptors must rely on diffusion from capillaries on the vitreal surface, in the inner retina, and on the choriocapillary system for their continuous supply of oxygen and other nutrients. ⁴² Intraretinal oxygen profiles indicate that there is a decline in oxygen tension from the pigment epithelium–choriocapillaris region to the ONL and that this decline is steeper in the retinas of dark-adapted animals than light-adapted ones. ^{42 43} Because oxygen tension levels in the IS are close to zero in dark-adapted animals, indicating the rate of oxygen replenishment and the rate of oxygen consumption is about equal, it has been postulated that the retina may be at risk of hypoxia after an increase in oxygen demand or a reduction in oxygen supply. ^{43 44} The potential consequences of hypoxia may be one reason for the presence of Hsps in the normal retina. 

Based on the comments above, it is not totally unexpected that the 70-kDa Hsps are present in the normal retina. What was surprising to us was the observation that in IS and the ONL, the distribution of Hsc70 was coextensive with that of Hsp70. Apparently both isoforms are needed in the same layers (and it would seem, in the same cells). This observation suggests that Hsc70 and Hsp70 play different roles in photoreceptor metabolism and survival. However, a clear functional difference between the two forms has not been established. ²⁰ Further work is necessary to determine what distinctive functions Hsc70 and Hsp70 serve in the normal photoreceptor. 

Another region of the retina that showed an interesting distinction in Hsc70 and Hsp70 distribution was the OLM, which contains the zone of junctional specializations between Müller cells and photoreceptors. In the Hsc70-stained sections, the intensity of immunoreactivity along the entire length of the OLM was uniform and slightly greater than that in the rest of the IS, making the OLM visible (Fig. 1) . This observation suggests that Hsc70 is focally associated with the gap and zonulae adherens junctions between Müller cell terminal processes and/or between Müller cells and photoreceptors. In the Hsp70-stained sections, however, the pattern of immunoreactivity in this same location appeared punctate (Fig. 1) , suggesting that Hsp70 was even more focally concentrated in the intercellular junctional complex. These observations suggest a special role for Hsc70 and Hsp70 in the structural organization of those junctions. 

Retinal Pigment Epithelium

The RPE, which is composed of a monolayer of cuboidal cells sandwiched between the choroidal and photoreceptor layers of the retina, is responsible for a wide variety of processes that are indispensable to normal retinal function. These include the daily phagocytosis of effete photoreceptor outer segment fragments; the absorption of stray light; the uptake, processing, transport, and release of vitamin A and some of its visual cycle intermediates; and the maintenance of proper retinal levels of dehydration and optical clarity by means of net movement of ions and water from that structure. ⁴⁵ By virtue of its location, the RPE is also involved in the maintenance of the retinal blood-brain barrier. We observed no staining in the RPE for Hsc70, or Hsp70, except in those cells located less than 200 to 250 μm from the optic nerve. There the staining was specific for Hsc70 and appeared to be strongest in the nucleus (Fig. 3) . Because Hsp70 is known to be stimulated in macrophages that are actively engaged in phagocytosis, ⁴⁶ it is possible that the Hsc70 immunoreactivity may be similarly induced and that this observation is related to differences in RPE phagocytic activity across the retina. The difficulty with this premise is that we removed the eyes during the middle of the day, when the level of photoreceptor disc shedding is normally low. Also, we did not observe any signs of phagosomes, or phagolysosomes, hallmark features of ongoing phagocytosis, although this may in part be a consequence of the limitations of light microscopy. Another possible explanation for our findings is that the RPE around the optic nerve may be more stressed than the RPE that lies at the distal end of the retina. This could be a result of the fact that there is a defect in the blood-retinal barrier around the optic nerve head that could allow entry of stress-inducing factors (Bok D, personal communication). Clearly, the functional significance of this apparent distinction between RPE cells near the optic nerve and those in more peripheral parts of the retina needs to be explored further. 

Optic Nerve

This laboratory previously described Hsc70/Hsp70 staining in the juxtanuclear cytoplasm of glial cells and in the optic nerve fibers of rats comparable in age to those used in this study. ²² However, that report did not discriminate between the two forms. The present work showed that only Hsc70 immunoreactivity was detectable in the normal rat optic nerve, not Hsp70. We are unable to say at this time whether the prominent immunoreactivity visualized by Yamaguchi et al. ²² in the optic nerve of the Royal College of Surgeons rat included Hsp70, in addition to Hsc70. As suggested in that report, the ongoing degeneration of the dystrophic rat retina could induce a stress response, which might lead to induction of Hsp70. 

Cornea

Yamaguchi et al., ²³ from this laboratory, reported that the central corneal epithelium contained low levels of Hsp70 immunoreactivity under normal conditions and that those levels were elevated in basal and wing cells after hyperthermia. Taken together with another report, which showed that epithelial Hsp70 levels were elevated in experimental alkali burn in the rabbit and in the human corneal diseases keratoconus and bullous keratopathy, ⁴⁷ this observation suggested that the corneal epithelial cells, like other cells, respond to metabolic stress by increasing Hsp70. Other work from this laboratory has confirmed that the basal cells of the corneal epithelium contained Hsc/Hsp70 in rats as young as postnatal day 15, but did not distinguish between the two forms. ²² Interestingly, those previous studies failed to reveal one unusual aspect of the normal central corneal epithelium shown here. That is, the distribution and prominence of immunoreactive Hsp70 over that of Hsc70 (Fig. 4) . This observation is consistent with a preliminary finding that Hsp70 accounts for most of the Hsp immunoreactivity detected in the corneal epithelium. ²⁵ Thus, in contrast to other tissues, in which Hsc70 is the more abundant form and Hsp70 levels are low, the normal rat corneal epithelium appears unique, having a greater requirement for Hsp70. It is possible that this distinction may be related to the robust healing capacity of the corneal epithelial cells, which can repair an abraded area in just a few hours, ⁴⁸ but further study will be needed to address this hypothesis. 

Another unexpected observation in the cornea was that the patterns and relative levels of immunoreactivity for Hsc70 and Hsp70 were different in the limbal epithelium compared with the central region. Both forms of the protein had nearly equivalent distributions in the limbus, with the minor exception that the wing cells contained mainly Hsc70 immunoreactivity and were negative for Hsp70 (Fig. 5) . Because this region of the cornea is the site of the proposed epithelial stem cells, ^{49 50} it would suggest that the proliferative capacity of these undifferentiated cells may require greater amounts of Hsc70 than is found in the more differentiated cells of the central cornea. 

Ciliary Body

In this structure, Hsc70 reactivity was detected in the cytoplasm of the epithelial cells. Because many epithelia are active in secretion, like those of the ciliary body, the prominence of Hsc70 may be a consequence of its role as the clathrin uncoating ATPase involved in vesicle cycling. ^{51 52} The significance of the greater juxtanuclear concentration of Hsc70 in these cells is not known. 

Conclusions

We have described the pattern of Hsc70 and Hsp70 immunostaining in different ocular structures in the absence of overt whole body stress and have shown that each structure has a distinctive complement and distribution of these two proteins. We have also shown that some cell types can vary in their Hsc70 and/or Hsp70 content depending on their location in the structure, as seen in the RPE and corneal epithelia cells. It should be noted, however, that, although these results suggest differences in the absolute amounts of Hsc70 relative to Hsp70, they must be interpreted with caution, because factors other than absolute protein content can affect relative levels of tissue immunoreactivity. Nonetheless, our observations show clearly that the generalization commonly stated in the Hsp literature that one isoform, Hsc70, functions as a constitutive housekeeping protein required for the folding of other cellular proteins as they are synthesized and the other, Hsp70, functions as an inducible “stress/damage repair” protein is an oversimplification. Furthermore, our results provide a foundation from which it will be possible to explore the relationships between the distributions of Hsc70 and Hsp70 and the unique metabolic attributes of the various ocular cells, ultimately leading to a clearer general understanding of the functions of these two proteins and their roles in the maintenance of the normal structure and function of the retina. 

 Supported by grant EY07616 (MT) from the National Institutes of Health, Bethesda, Maryland.

 Submitted for publication January 21, 1999; revised May 18, 1999; accepted June 21, 1999.

 Commercial relationships policy: N.

 Presented in part at the 1997 annual meeting of the Society for Neuroscience in New Orleans, LA, and the 1998 Cold Spring Harbor Symposium on Molecular Chaperones and the Heat Shock Response at Cold Spring Harbor, NY.

 

Figure 1.

 

Western blot analysis using the antibodies for constitutive Hsc70 and inducible Hsp70. Samples of bovine Hsc70 (100 ng), recombinant human Hsp70 (100 ng), and retinal homogenate (2 μg) were prepared and loaded on an SDS–polyacrylamide gel. After electrophoresis and transfer to nitrocellulose, the membrane was cut in half, and one side of the blot was incubated with anti-Hsc70 (A), whereas the other side was incubated with anti-Hsp70 (B). The letters above each lane indicate the contents of that lane: C, bovine Hsc70; I, recombinant human Hsp70; and R, retinal homogenate.

Figure 2.

 

Distribution of Hsc70 and Hsp70 immunoreactivity in normal rat retina. Tissue sections were stained with hematoxylin and eosin (H&E; upper left panel), rat IgG (upper right panel), anti-Hsc70 (lower left panel), or anti-Hsp70 (lower right panel). Images of the retina were photographed at a position about halfway between the optic nerve and the peripheral end of the retina. The pattern of staining at this location was typical of all areas of the retina, except the region adjacent to the optic nerve. The layers of the retina are (bottom to top): retinal ganglion cell layer (rgc), inner plexiform layer (ipl), inner nuclear layer (inl), outer plexiform layer (opl), outer nuclear layer (onl), photoreceptor inner segment layer (is), photoreceptor outer segment (os), and retinal pigment epithelium cell layer (rpe). Arrows indicate the location of the OLM. Magnified images of the OLM are shown in the insets.

Figure 3.

 

Distribution of Hsc70 and Hsp70 immunoreactivity at the optic nerve–retina junction. Tissue sections were processed as described in Figure 1 . The retinal pigment epithelium (rpe) and the fibers (axons) of the retinal ganglion cells (rgc fibers), which converge at the optic nerve head, are indicated by arrows.

Figure 4.

 

Distribution of Hsc70 and Hsp70 immunoreactivity in the optic nerve of normal rats. This photograph was taken at the center of the optic nerve approximately 50 to 100 μm away from the optic nerve head. Note that the staining in the glial cells (solid arrows) and optic nerve fiber (open arrow). The inset shows a magnified view of two immunopositive glial cells.

Figure 5.

 

Distribution of Hsc70 and Hsp70 immunoreactivity in the central region of the cornea. Tissue sections were processed and photographed as described earlier. Images were captured from the center of the cornea. The following three layers in the corneal epithelium were identifiable in these micrographs: the basal cell layer (adjacent to the stroma), the wing cell layer (middle 3–4 layers of flattened nuclei), and the superficial cell layer (outermost 1 or 2 layers of nuclei). The most darkly stained structures within the H&E-stained epithelium are hematoxylin-stained nuclei. Comparison of their size, shape, and location in this panel with Hsp immunoreactivity in the Hsc70 and Hsp70 panels indicated that staining in the wing cells (open arrows) was largely nuclear. The lighter gray areas surrounding the cell nucleus reflect eosin-stained cytoplasm. Comparison of this cytoplasmic staining with the distribution of Hsc70 and Hsp70 reactivity in the basal cells (solid arrows) indicated that it was localized to the cytoplasm around the nucleus. The insets show images of the corneal epithelium enlarged to make it easier to see the cellular distribution for Hsc70 and Hsp70.

Figure 6.

 

Distribution of Hsc70 and Hsp70 immunoreactivity in the peripheral cornea and limbus. The limbal region (LR) differs from the peripheral cornea (PC) in that the epithelium overlies a rather less compact and highly vascularized loose connective tissue (LC), rather than an avascular compact stroma. The blood vessels (BV) and epithelial cell layer (E) are indicated. An example of a wing cell stained for Hsc70 (solid arrow) is indicated. The inset shows a magnified view of the wing cell.

Figure 7.

 

Distribution of Hsc70 and Hsp70 immunoreactivity in the ciliary body. These low-power magnification images show that only Hsc70 immunostaining is present in the ciliary body. They also show that staining is found mainly in the epithelial cells. The inset shows Hsc70 staining in the epithelium in more detail.