immunostaining of heat shock proteins in the retina and

Immunostaining of Heat Shock Proteinsin the Retina and Optic Nerve Headof Normal and Glaucomatous EyesGulgun Tezel, MD; M. Rosario Hernandez, DDS; Martin B. Wax, MD

Purpose: To examine immunostaining of 60-kd and27-kd heat shock proteins (HSP 60 and HSP 27), whichare known to increase cell survival in response to stress,in glaucomatous retina and optic nerve head.

Methods: Six postmortem eyes from patients with pri-mary open-angle glaucoma, 6 eyes from patients with nor-mal-pressure glaucoma, and 6 eyes from age-matched nor-mal subjects were studied by immunohistochemistry. Thesections of the retina and optic nerve head were exam-ined after immunostaining with antibodies to HSP 60 andHSP 27.

Results: The intensity of the immunostaining and thenumber of labeled cells for heat shock proteins (HSPs)were greater in retina sections from glaucomatous eyesthan in sections from normal eyes from age-matcheddonors. Retinal immunostaining of HSP 60 was promi-nent in the retinal ganglion cells and photoreceptors,

whereas immunostaining of HSP 27 was prominent inthe nerve fiber layer and ganglion cells as well as in theretinal vessels. In addition, retinal immunostaining ofthese HSPs exhibited regional and cellular differences.Optic nerve heads of glaucomatous eyes exhibited in-creased immunostaining of HSP 27, but not HSP 60,which was mostly associated with astroglial cells in thelamina cribrosa.

Conclusion: The increased immunostaining of HSP 60and HSP 27 in the glaucomatous eyes may reflect a roleof these proteins as a cellular defense mechanism in re-sponse to stress or injury in glaucoma.

Clinical Relevance: These findings suggest that im-munoregulation is an important component of glauco-matous optic neuropathy.

Arch Ophthalmol. 2000;118:511-518

H E A T S H O C K prote ins(HSPs), also called stressproteins, are a group ofhighly conserved pro-teins that are constitu-

tively expressed in most cells under nor-mal physiological conditions.1,2 Heat shockproteins are classified into families on thebasis of their molecular weight, includ-ing 90-kd (HSP 90), 70-kd (HSP 70), 60-kd(HSP 60), and small (25- to 30-kd) HSPfamilies. They are thought to play a vitalrole in normal cellular function. One of themain roles that HSPs play is that of mo-lecular chaperones. Specifically, HSPs havebeen shown to function in protein matu-ration events such as protein folding, un-folding, and translocation across mem-branes. In response to environmentalstresses such as heat, anoxia, and expo-sure to cytokines, cells newly synthesizelarge quantities of HSPs. Because of theirprotective capacity, the increased expres-sion of HSPs helps the cells to survive

stressful conditions and also promotes re-covery from stress.3-5

Constitutive expression of HSPsoccurs in the central nervous system in avariety of cell types, including oligoden-drocytes, astrocytes, and neurons.6,7 Theexpression of HSPs in neuronal cells sug-gests that, since neurons are structurallyand functionally very complex cells andexhibit nonmitotic characteristics, theymay require constitutive levels of theseproteins for protective purpose againstvarious stresses, such as hypoxia, anoxia,and excessive excitatory stimulation.

In addition to their constitutive ex-pression, the accumulation of HSPs invarious cells of the nervous system dur-ing acute toxic-metabolic states and in avariety of degenerative, inflammatory,and neoplastic neurological diseases fur-ther suggests their role for neuronal sur-vival.8-15 Up-regulation of HSP 27 hasbeen shown after damage to peripheralnerves. For example, vagus nerve lesions

LABORATORY SCIENCES

From the Department ofOphthalmology and VisualSciences, WashingtonUniversity School of Medicine,St Louis, Mo.

ARCH OPHTHALMOL / VOL 118, APR 2000 WWW.ARCHOPHTHALMOL.COM511

©2000 American Medical Association. All rights reserved.Downloaded From: https://jamanetwork.com/ on 10/01/2021

result in a time-dependent up-regulation of HSP 27 invagal motor and ganglion sensory neurons.16 Transec-tion of the sciatic nerve similarly results in a 9-fold up-regulation of HSP 27 messenger RNA and protein inganglion cells.17

In the retina, the expression of HSPs is develop-mentally regulated through ocular organogenesis.18,19

The most commonly studied retinal HSPs are the mem-bers of HSP 70 family, which are rapidly induced byhyperthermic, light, or ischemic injury in rat and rabbitretinas.20-25 However, studies of retinal expression ofother HSPs are limited. For example, heat-induced reti-nal expression of small HSPs, including HSP 23 and

HSP 27, has been shown in Drosophila melanogaster.26

In addition, retinal synthesis of a small HSP, HSP 30,increases after optic nerve crush in goldfish.27

Since elevated titers of serum autoantibodies to HSP60 and HSP 27 were found in many patients with glau-coma,28,29 we studied HSP 60 and HSP 27 immunoreac-tivity in postmortem glaucomatous eyes in comparisonwith normal donor eyes. Our observations disclosed in-creased immunostaining of these HSPs in the retina andoptic nerve head of the glaucomatous eyes, which sug-gests that HSP 60 and HSP 27 may be components of anatural defense mechanism that are activated in glauco-matous optic neuropathy.

MATERIALS AND METHODS

EYES

Six postmortem human eyes with a diagnosis of primaryopen-angle glaucoma (ages, 56-93 years), 6 eyes with a di-agnosis of normal-pressure glaucoma (ages, 68-84 years),and 6 eyes from age-matched normal donors (ages, 61-91years) were obtained from the Glaucoma Research Foun-dation (San Francisco, Calif), the National Disease Re-search Interchange (Philadelphia, Pa), the Mid-America EyeBank (St Louis, Mo), and one of us (M.B.W.). Clinical find-ings of the patients were well documented and includedintraocular pressure readings, optic disc assessments, andvisual field tests (Table). Normal donors had no historyof eye disease or diabetes. There was no infection or sepsisin any of the donors. The cause of death for all of the do-nors used in this study was myocardial infarction or car-diopulmonary failure.

The eyes were enucleated within 2 to 4 hours afterdeath and processed and fixed within the following 6 to12 hours in either 10% buffered formaldehyde or 4%paraformaldehyde. The posterior poles were dissectedfree of surrounding tissues, washed extensively in 0.2%glycine in phosphate-buffered saline at pH 7.4, embeddedin paraffin, and oriented sagittally for 6-mm sections. Im-munoperoxidase staining and double immunofluores-cence labeling were used for localization of different epit-opes of HSP 60 and HSP 27 to the retina and optic nervehead. Serial sections from the glaucomatous eyes and nor-mal donor eyes were stained simultaneously to controlvariations in the immunostaining.

IMMUNOHISTOCHEMICAL ANALYSIS

For immunostaining of HSP 60, mouse monoclonal anti-bodies to human HSP 60, which were either a gift fromRadhey S. Gupta, PhD (McMaster University, Hamilton,Ontario)30 or purchased (StressGen, Victoria, British Co-lumbia) were used at 1:200 dilution and 1:1000 dilution,respectively. A rabbit polyclonal anti–human HSP 27, agift from Kanefusa Kato, PhD (Institute for Developmen-tal Research, Aichi, Japan)11 was used at 1:200 dilution,and a mouse monoclonal anti–human HSP 27 (Sigma-Aldrich Corp, St Louis, Mo) was used at 1:1000 dilutionfor immunostaining for HSP 27. Additional monoclonaland polyclonal antibodies against human HSP 27 (Stress-Gen) were used at 1:1000 and 1:400, respectively. Non-

immune rabbit and mouse serum samples (Sigma-AldrichCorp) were used to replace the primary antibodies toserve as negative controls.

For immunoperoxidase staining, sections from nor-mal and glaucomatous eyes were deparaffinized, rehy-drated, and pretreated with 0.3% hydrogen peroxide inphosphate-buffered saline to decrease endogenous per-oxidase activity. Primary antibodies were localized by im-munoperoxidase using a commercial kit (Vector Labora-tories, Burlingame, Calif). The biotinylated secondaryantibody was incubated on the sections for 30 minutes,washed with phosphate-buffered saline containing 0.1%bovine serum albumin, and reacted with streptavidin-horseradish peroxidase conjugated for 30 minutes. Afterseveral washes, color was developed by incubation with3,39-diaminobenzidine tetrahydrochloride (Sigma-Al-drich Corp) as cosubstrate for 5 to 7 minutes. Sectionswere counterstained with hematoxylin and coverslippedwith a mounting medium (Permount; Fischer, Pitts-burgh, Pa).

To study colocalization of HSPs, we performed a doubleimmunofluorescence procedure. Sections were incubatedwith a mixture of mouse antibody against HSP 60 and rab-bit antibody against HSP 27 at 1:100 dilution for 30 min-utes. The sections were then incubated with a mixture ofrhodamine red– and Oregon green–labeled secondary an-tibodies (Molecular Probes, Eugene, Ore) for another 30minutes. Negative controls were performed by replacingthe primary antibody with nonimmune serum or by incu-bating sections with each primary antibody followed by theinappropriate secondary antibody to determine that eachsecondary antibody was specific to the species against whichit was made.

Slides were examined and documented with a fluo-rescence microscope (Olympus, Tokyo, Japan) equippedwith bright-field illumination and epifluorescence light. Im-ages were recorded on 400 ASA color print film or 400 ASAblack-and-white print film (Eastman Kodak Company,Rochester, NY). Images were also recorded by means of adigital camera (Diagnostic Instruments, Sterling Heights,Mich) attached to the microscope.

The stained sections were examined by 2 observers(G.T. and M.R.H.) in a masked fashion regarding theidentity or diagnosis of the donors. Qualitative evaluationof the immunostaining (negative, faint, or increased im-munostaining) in the optic nerve head and different lay-ers of the retina was independently recorded by these ob-servers. Judgment of increased immunostaining wasmade only when both observers independently agreed.



RESULTS

The examination of retina sections by immunohisto-chemical analysis demonstrated low intensity of immu-nostaining of HSP 60 and HSP 27 in retina from normaldonor eyes. The immunostaining was noticeably in-creased for both HSPs in glaucomatous eyes with eitherprimary open-angle glaucoma or normal-pressure glau-coma. Immunostaining with different antibodies to ei-ther HSP 60 or HSP 27 from 2 different sources yieldedidentical results. Control sections stained with nonim-mune serum to replace the primary antibodies did notexhibit immunostaining.

In normal eyes, faint immunostaining of HSP 60 wasobserved in a few retinal ganglion cells and photorecep-tors. However, in glaucomatous eyes, the intensity of theimmunostaining and the number of stained cells for HSP60 were greater than in normal eyes. In the eyes with pri-mary open-angle glaucoma, some retinal ganglion cells,but not all, strongly stained for HSP 60. In the eyes withnormal-pressure glaucoma, in addition to increased im-munostaining in the retinal ganglion cells and photo-receptors, immunostaining was observed in some cellslocated in the inner nuclear layer of the retina (Figure1).No prominent staining for HSP 60 was associated with reti-nal vasculature or nerve fiber layer.

Although clearly positive immunostaining of HSP27 was observed in the vascular wall in normal retinas,there was faint immunostaining of HSP 27 in retinal tis-sue (Figure 2). As shown in Figure 2, B, few retinalganglion cells stained for HSP 27 in normal retinas, andno staining was observed in the nerve fiber layer. How-ever, the intensity of the immunostaining and the num-ber of stained cells for HSP 27 were greater in glauco-matous eyes than in normal eyes. The increasedimmunostaining of HSP 27 in glaucomatous eyes wasprominent in the nerve fiber layer as well as in retinalganglion cells (Figure 2).

Immunostaining of HSP 60 and HSP 27 with the useof adjacent retina sections from an eye with normal-pressure glaucoma is shown in Figure 3. Immuno-staining of HSP 60 was prominent in the retinal photo-receptors but not around the retinal vasculature, whereasimmunostaining of HSP 27 was prominent around theblood vessels but not in the retinal photoreceptors. Doubleimmunofluorescence labeling of HSP 60 and HSP 27in a glaucomatous eye as well as negative controls areshown in Figure 4. Similar to the immunoperoxidasestaining, no labeling for HSP 60 was associated with reti-nal vasculature, but there was prominent immuno-staining for HSP 27 around the retinal blood vessels. Mostof the retinal ganglion cells were heavily stained forboth HSP 60 and HSP 27. However, some of the retinalganglion cells exhibited marked staining for only oneof the HSPs.

The immunostaining of HSPs in retinal ganglion cellsexhibited regional differences that seemed to be relatedto their localization with respect to vascular structures.For example, immunostaining of HSP 27 in retinal gan-glion cells close to blood vessels (Figure 2, D) exhibiteda lower intensity than that in the cells far from blood ves-sels (Figure 2, B, E, and F). In addition to the regional

differences, there was an apparent difference between theimmunostaining of individual retinal ganglion cells (Fig-ure 1, D, and Figure 4, A, B, D, and E).

Immunostaining of HSP 60 in the optic nerve headwas similarly faint in normal and glaucomatous eyes(Figure 5). However, immunostaining of HSP 27 in-creased markedly in the optic nerve head of glaucoma-tous eyes with either primary open-angle glaucoma ornormal-pressure glaucoma compared with age-matched controls. The increased HSP 27 immunostain-ing was mostly located in the laminar region and asso-ciated with astroglial cells as assessed by morphologicalexamination (Figure 6).

COMMENT

Immunohistochemical analysis in postmortem eyesdemonstrated that retinal immunostaining of HSP 60and HSP 27 was greater in glaucomatous eyes than ineyes from age-matched normal donors. In addition,optic nerve heads of glaucomatous eyes exhibitedincreased immunostaining of HSP 27, but not HSP 60,which was associated mostly with astroglial cells.

We did not observe marked differences in the HSP60 and HSP 27 immunostaining between eyes withnormal-pressure glaucoma and those with primaryopen-angle glaucoma. This may suggest that increasedexpression of these HSPs may be, at least in part, inde-pendent of the level of intraocular pressure, and there-fore related to tissue stress or damage in all eyes withglaucoma. Induction of HSPs in the central nervous sys-tem in response to several environmental stresses,including ischemia, has been suggested to be an earlyresponse against stress and restoration of damaged areasin the brain after injury. Increased expression of HSPsin astrocytes within the affected area has been similarlyimplicated so as to increase neuronal survival.31 Theincreased immunostaining of HSP 60 and HSP 27 in theretina and optic nerve head of the glaucomatous eyestherefore may suggest that these proteins play a role asa defense mechanism of stressed or injured neurons inglaucoma. In addition, there were regional differencesin the immunostaining of retinal HSPs as well as differ-

Clinical Data of Postmortem Glaucomatous Eyes*

Patient No./Sex/Age, y Diagnosis C/D VF Damage

1/M/58 POAG 0.9 Advanced2/F/74 POAG 0.9 Advanced3/F/67 POAG 0.5 Moderate4/F/93 POAG 0.7 Moderate5/F/56 POAG 0.9 Advanced6/F/67 POAG 0.4 Moderate7/F/84 NPG 0.95 Advanced8/F/84 NPG 0.95 Advanced9/F/68 NPG 0.8 Moderate

10/F/68 NPG 0.6 Moderate11/F/74 NPG 0.8 Moderate12/F/74 NPG 0.9 Advanced

*C/D indicates vertical cup-to-disc ratio; VF, visual field; POAG, primaryopen-angle glaucoma; and NPG, normal-pressure glaucoma.



ences between HSP immunostaining of individual reti-nal ganglion cells. This may correspond to local andindividual differences in the susceptibility of neuronalcells to damage in various regions of the retina. Such apreferential response to stress appears consistent with

our limited understanding of the patterns of glaucoma-tous defects. However, the precise assessment of therelationship between HSP immunostaining and corre-sponding functional or anatomical damage in glauco-matous eyes needs further study.

A B

C D

E F

nfl

v

p

p

p

v

nfl

nfl

Figure 1. Immunoperoxidase staining of heat shock protein (HSP) 60 in human retina. A and B, Retina sections from normal donor eyes. C and D, Retina sectionsfrom eyes with primary open-angle glaucoma. E and F, Retina sections from eyes with normal-pressure glaucoma. C and E are from eyes with moderateglaucomatous damage; D and F, from eyes with advanced glaucomatous damage. There was faint immunostaining of HSP 60 in some retinal cells including retinalganglion cells (arrows) in normal retina. However, glaucomatous eyes with either primary open-angle glaucoma or normal-pressure glaucoma exhibited increasedimmunostaining. In addition to the immunostaining of retinal ganglion cells and photoreceptors in the eyes from patients with normal-pressure glaucoma, somecells located in the inner nuclear layer (black arrowheads) exhibited immunostaining. Note in D that one retinal ganglion cell was immunostained (black arrow) butanother adjacent cell was not (white arrowhead) (v indicates vessel; nfl, nerve fiber layer; and p, photoreceptors) (chromagen, 3,39-diaminobenzidinetetrahydrochloride; nuclear counterstain with hematoxylin; magnification bars represent 150 µm).



The ability of neuronal and glial cells to increase HSPexpression in response to stress is key to their survival.After transection, up-regulated HSP 27 in dorsal root gan-glion cells of the sciatic nerve is transported to injured axons

and may promote survival and contribute to alterationsin the cytoskeleton associated with axonal growth.17 Simi-lar to peripheral nerves,16,17 up-regulation of HSPs occursafter optic nerve crush.27 Additionally, it has been shown

A B

C D

E F

nfl

vv

p

p

v

v

v

v

v

nfl

nfl

Figure 2. Immunoperoxidase staining of heat shock protein (HSP) 27 in human retina. A and B, Retina sections from normal donor eyes. C and D, Retina sectionsfrom eyes with primary open-angle glaucoma. E and F, Retina sections from eyes with normal-pressure glaucoma. C and E are from eyes with moderateglaucomatous damage; D and F, from eyes with advanced glaucomatous damage. Note in A that, despite clearly positive immunostaining of HSP 27 in retinalvessels, there was faint immunostaining in retinal tissue including a few retinal ganglion cells (arrows) in normal retina. The arrow in B points to a stained retinalganglion cell in normal retina. In comparison with normal eyes, glaucomatous eyes with either primary open-angle glaucoma (C and D) or normal-pressureglaucoma (E and F) exhibited increased immunostaining, which was most prominent in the retinal nerve fiber layer. Immunostaining of retinal ganglion cells forHSP 27 exhibited differences between different cells. Notice in D that a retinal ganglion cell close to a vessel was not as heavily immunostained as other retinalganglion cells shown in B, E, and F (v indicates vessel; nfl, nerve fiber layer; and p, photoreceptors) (chromagen, 3,39-diaminobenzidine tetrahydrochloride;nuclear counterstain with hematoxylin; magnification bars represent 150 µm).



A

B

vit

vit

p

p

v

v

Figure 3. Immunoperoxidase staining of heat shock protein (HSP) 60 and HSP 27in adjacent retina sections through the same blood vessel from an eye withadvanced normal-pressure glaucoma. A, Retina section exhibiting immunostainingof HSP 60. B, Retina section exhibiting immunostaining of HSP 27.Immunostaining of HSP 60 was positive in the retinal photoreceptors but negativearound the retinal blood vessel. Immunostaining of HSP 27 was positive aroundthe same blood vessel as shown in A but negative in the retinal photoreceptors.There was significant immunostaining for HSP 60 and HSP 27 in the retinalganglion cells (arrows) (v indicates vessel; vit, vitreous; and p, photoreceptors)(chromagen, 3,39-diaminobenzidine tetrahydrochloride; nuclear counterstain withhematoxylin; magnification bar represents 150 µm).

A B C

D E

G H

F

vit

nfl

vit

nfl

vit

nfl

nfl nfl

nfl nfl

nfl

vit vit

v

v

v

v

v

v

v

v

v

Figure 4. Double immunofluorescence labeling for heat shock protein (HSP)60 (red) and HSP 27 (green) in retina sections from an eye with primaryopen-angle glaucoma and moderate glaucomatous damage. A and D,Immunostaining of HSP 60. B and E, Immunostaining of HSP 27. C and F,Colocalization of HSP 60 and HSP 27. G and H, Negative controls of the HSP60 and HSP 27 immunostaining, respectively. Immunostaining of HSP 60was negative, but immunostaining of HSP 27 was positive around the bloodvessels. Glial cells lining the internal limiting membrane were also positivefor HSP 27. Most of the retinal ganglion cells were stained for both HSP 60and HSP 27 (arrows). However, notice a retinal ganglion cell in B and C(arrowheads) that was positive for HSP 60 but negative for HSP 27. Alsonotice a retinal ganglion cell in D and F (arrowheads) that was positive forHSP 27 but negative for HSP 60 (v indicates vessel; vit, vitreous; and nfl,nerve fiber layer) (magnification bar represents 150 µm).



A

B

C

vit

v

v

vit

v

lc

vit

v

lc

lc

Figure 5. Immunoperoxidase staining of heat shock protein (HSP) 60 inhuman optic nerve head. A, Optic nerve head section from a normal donoreye. B, Optic nerve head section from an eye with primary open-angleglaucoma and advanced glaucomatous damage. C, Optic nerve head sectionfrom an eye with normal-pressure glaucoma and advanced glaucomatousdamage. There was no prominent immunostaining of HSP 60 in the opticnerve head of normal donor eyes. Glaucomatous eyes did not exhibit aprominent difference in the immunostaining of HSP 60 compared withnormal eyes (v indicates vessel; vit, vitreous; and lc, lamina cribrosa)(chromagen, 3,39-diaminobenzidine tetrahydrochloride; nuclear counterstainwith hematoxylin; magnification bar represents 1 mm).

A

B

C

vit

Bm

Bm

v

v

vit

v

lc

vit

Bmv

cs

lc

lc

Figure 6. Immunoperoxidase staining of heat shock protein (HSP) 27 inhuman optic nerve head. A, Optic nerve head section from a normal donoreye. B, Optic nerve head section from an eye with primary open-angleglaucoma and advanced glaucomatous damage. C, Optic nerve head sectionfrom an eye with normal-pressure glaucoma and advanced glaucomatousdamage. There was immunostaining of HSP 27 around the vessels in theoptic nerve head of normal donor eyes. Glaucomatous eyes exhibited aprominent difference in the immunostaining of HSP 27 compared withnormal eyes. The increased immunostaining was mostly located in thelaminar area (v indicates vessel; vit, vitreous; lc, lamina cribrosa; Bm, Bruchmembrane; and cs, cavernous spaces of Schnabel degeneration)(chromagen, 3,39-diaminobenzidine tetrahydrochloride; nuclear counterstainwith hematoxylin; magnification bar represents 1 mm).



in rabbits that HSPs can be transported in axons betweenretinal ganglion cells and optic nerve.32 In addition to in-creased immunostaining of HSP 27 in the optic nerve headand cell body of retinal ganglion cells in glaucomatous eyes,increased immunostaining in the retinal nerve fiber layermay indicate the feasibility of HSP transport between op-tic nerve head and retinal ganglion cells to increase sur-vival of the injured axons in glaucomatous eyes.

Heat shock proteins are highly antigenic, and im-mune responses to HSPs are implicated in the develop-ment of a number of human autoimmune diseases.33 Al-though increased expression of HSPs in glaucomatous eyesmay serve initially to protect cells from further destruc-tion and facilitate repair, they subsequently may recruitimmune responses that contribute to the progression ofdisease.33-35 Glial cells of the retina and optic nerve headare antigen-presenting cells.36-38 Since these glial cells be-come activated in glaucomatous eyes,39,40 enhanced ex-pression of HSPs in these eyes may be an immunostimu-latory signal, which leads to a break in immune tolerancethat occurs under normal conditions.41 Increased titers ofautoantibodies to HSP 60 and/or HSP 27 in many pa-tients with glaucoma28,29 therefore may represent a gen-eralized response to tissue stress and/or damage, which sub-sequently may contribute to disease progression bydiminishing the protective abilities of native HSPs.

Accepted for publication November 10, 1999.This study was supported in part by grants EY12314

(Dr Wax), EY06416 (Dr Hernandez), and EY02687(Dr Hernandez) from the National Institutes of Health,Bethesda, Md; grants from the Glaucoma Foundation, NewYork, NY (Dr Tezel), and the Glaucoma Research Founda-tion, San Francisco, Calif (Dr Wax); and an unrestrictedgrant to the Department of Ophthalmology and Visual Sci-ences, Washington University School of Medicine, St Louis,Mo, from Research to Prevent Blindness Inc, New York, NY.

Belinda McMahan, HT (ASCP), provided assistanceduring immunohistochemistry.

Reprints: Martin B. Wax, MD, Department of Oph-thalmology and Visual Sciences, Washington UniversitySchool of Medicine, Box 8096, 660 S Euclid Ave, St Louis,MO 63110 (e-mail: [email protected]).

REFERENCES

1. Subjeck JR, Shyy TT. Stress protein systems of mammalian cells. Am J Physiol.1986;250:C1-C17.

2. Welch WJ. Mammalian stress response: cell physiology structure/function ofstress proteins and implications for medicine and disease. Physiol Rev. 1992;72:1063-1081.

3. Lindquist S, Craig EA. The heat shock proteins. Annu Rev Genet. 1988;22:631-677.4. Ellis RJ, van der Vies SM. Molecular chaperones. Annu Rev Biochem. 1991;60:

321-347.5. Parsell DA, Lindquist S. The function of heat shock proteins in stress tolerance: deg-

radation and reactivation of damaged proteins. Annu Rev Genet. 1993;27:437-496.6. D’Souza SM, Brown IR. Constitutive expression of heat shock proteins Hsp90,

Hsp70 and Hsp60 in neural and non-neural tissues of the rat during postnataldevelopment. Cell Stress Chaperones. 1998;3:188-199.

7. Plumier JC, Hopkins DA, Robertson HA, Currie RW. Constitutive expression ofthe 27-kDa heat shock protein (Hsp27) in sensory and motor neurons of the ratnervous system. J Comp Neurol. 1997;384:409-428.

8. Head MW, Corbin E, Goldman JE. Overexpression and abnormal modification ofthe stress proteins aB-crystallin and HSP27 in Alexander disease. Am J Pathol.1993;143:1743-1753.

9. Iwaki T, Iwaki A, Tateishi J, Sakaki Y, Goldman JE. AB-crystallin and 27 kilodal-ton heat shock protein are regulated by stress conditions in the central nervous

system and accumulate in Rosenthal fibers. Am J Pathol. 1993;143:487-495.10. Renkawek K, Bosman GJ, de Jong WW. Expression of small heat-shock protein

hsp27 in reactive gliosis in Alzheimer disease and other types of dementia. ActaNeuropathol (Berl). 1994;87:511-519.

11. Kato H, Liu Y, Kogure K, Kato K. Induction of 27-kda heat shock protein followingcerebral ischemia ina ratmodelof ischemic tolerance. BrainRes.1994;634:235-244.

12. Higashi T, Takechi H, Uemura Y, Kikuchi H, Nagata K. Differential induction ofmRNA species encoding several classes of stress proteins following focal cere-bral ischemia in rats. Brain Res. 1994;650:239-248.

13. Kato H, Araki T, Itoyama Y, Kogure K, Kato K. An immunohistochemical study ofheat shock protein-27 in the hippocampus in a gerbil model of cerebral ische-mia and ischemic tolerance. Neuroscience. 1995;68:65-71.

14. Kato H, Kogure K, Liu XH, Araki T, Kato K, Itoyama Y. Immunohistochemical lo-calization of the low molecular weight stress protein HSP27 following focal ce-rebral ischemia in the rat. Brain Res. 1995;679:1-7.

15. Aquino DA, Capello E, Weisstein J, et al. Multiple sclerosis: altered expressionof 70- and 27-kDa heat shock proteins in lesions and myelin. J Neuropathol ExpNeurol. 1997;56:664-672.

16. Hopkins DA, Plumier JC, Currie RW. Induction of the 27-kDa heat shock protein(Hsp27) in the rat medulla oblongata after vagus nerve injury. Exp Neurol. 1998;153:173-183.

17. Costigan M, Mannion RJ, Kendall G, et al. Heat shock protein 27: developmentalregulation and expression after peripheral nerve injury. J Neurosci. 1998;18:5891-5900.

18. Tanaka Y, Kobayashi K, Kita M, Kinoshita S, Imanishi J. Messenger RNA expres-sion of heat shock proteins (HSPs) during ocular development. Curr Eye Res.1995;14:1125-1133.

19. Kojima M, Hoshimaru M, Aoki T, et al. Expression of heat shock proteins in thedeveloping rat retina. Neurosci Lett. 1996;205:215-217.

20. Lindquist S. The heat-shock response. Annu Rev Biochem. 1986;55:1151-1191.21. Masing TE, Rush SJ, Brown IR. Induction of a heat shock gene (hsp70 ) in rab-

bit retinal ganglion cells detected by in situ hybridization with plastic-embeddedtissue. Neurochem Res. 1990;15:1229-1235.

22. Tytell M, Barbe MF, Brown IR. Induction of heat shock (stress) protein 70 andits mRNA in the normal and light-damaged rat retina after whole body hyper-thermia. J Neurosci Res. 1994;38:19-31.

23. Caprioli J, Kitano S, Morgan JE. Hyperthermia and hypoxia increase tolerance ofretinal ganglion cells to anoxia and excitotoxicity. Invest Ophthalmol Vis Sci. 1996;37:2376-2381.

24. Beasley TC, Tytell M, Sweatt AJ. Heat shock protein 70 in the retina of Xenopuslaevis, in vivo and in vitro: effect of metabolic stress. Cell Tissue Res. 1997;290:525-538.

25. Lewden O, Garcher C, Assem M, Morales C, Rochette L, Bron AM. Changes ofthe inducible heat shock protein 70 mRNA level in rat retina after ischemia andreperfusion. Ophthalmic Res. 1998;30:291-294.

26. Marin R, Demers M, Tanguay RM. Cell-specific heat-shock induction of Hsp23in the eye of Drosophila melanogaster. Cell Stress Chaperones. 1996;1:40-46.

27. Cauley KA, Sherman TG, Ford-Holevinski T, Agranoff BW. Rapid expression ofnovel proteins in goldfish retina following optic nerve crush. Biochem BiophysRes Commun. 1986;138:1177-1183.

28. Wax MB, Tezel G, Saito I, et al. Anti-Ro/SS-A positivity and heat shock proteinantibodies in patients with normal-pressure glaucoma. Am J Ophthalmol. 1998;125:145-157.

29. Tezel G, Seigel GM, Wax MB. Autoantibodies to small heat shock proteins in glau-coma. Invest Ophthalmol Vis Sci. 1998;39:2277-2287.

30. Singh B, Gupta RS. Expression of human 60-kd heat shock protein (HSP60 orP1) in Escherichia coli and the development and characterization of correspond-ing monoclonal antibodies. DNA Cell Biol. 1992;11:489-496.

31. Higashi T, Takechi H, Uemura Y, Kikuchi H, Nagata K. Differential induction ofmRNA species encoding several classes of stress proteins following focal cere-bral ischemia in rats. Brain Res. 1994;650:239-248.

32. Clark BD, Brown IR. Axonal transport of a heat shock protein in the rabbit visualsystem. Proc Natl Acad Sci U S A. 1985;82:1281-1285.

33. Young DB. Heat-shock proteins: immunity and autoimmunity. Curr Opin Immu-nol. 1992;4:396-400.

34. Aquino DA, Capello E, Weisstein J, et al. Multiple sclerosis: altered expressionof 70- and 27-kDa heat shock proteins in lesions and myelin. J Neuropathol ExpNeurol. 1997;56:664-672.

35. Birnbaum G, Kotilinek L. Immunity to heat shock proteins and neurological dis-orders of women. Infect Dis Obstet Gynecol. 1999;7:39-48.

36. Roberge FG, Caspi RR, Nussenblatt RB. Glial retinal Muller cells produce IL-1activity and have a dual effect on autoimmune T helper lymphocytes: antigen pre-sentation manifested after removal of suppressive activity. J Immunol. 1988;140:2193-2196.

37. Penfold PL, Provis JM, Liew SC. Human retinal microglia express phenotypiccharacteristics in common with dendritic antigen-presenting cells. J Neuroim-munol. 1993;45:183-191.

38. Ma N, Streilein JW. T cell immunity induced by allogenic microglia in relation toneuronal retina transplantation. J Immunol. 1999;162:4482-4489.

39. Rao K, Lund RD. Optic nerve degeneration induces the expression of MHC an-tigens in the rat visual system. J Comp Neurol. 1993;336:613-627.

40. Hernandez MR, Pena JD. The optic nerve head in glaucomatous optic neuropa-thy. Arch Ophthalmol. 1997;115:389-395.

41. Gregerson DS. Immune privilege in the retina. Ocul Immunol Inflamm. 1998;6:257-267.



immunostaining of heat shock proteins in the retina and

Documents