Graefe's Archive

Graefe's Arch Clin Exp Ophthalmol(1992)230:421-427

for Clinical and Experimental

Ophthalmology © Springer-Verlag1992

Extracellular matrix changes of the optic nerve lamina cribrosa in monkey eyes with experimentally chronic glaucoma Takeo Fuknchi, Shoichi Sawaguchi, Hiroaki Hara, Motohiro Shirakashi, and Kazuo Iwata

Department of Ophthalmology,Niigata University, School of Medicine, l-Asahi Machi, Niigata City 951, Japan Received June 25, 1991 / Accepted September 30, 1991

Abstract. Using light microscopic immunohistochemistry, we studied the immunolocalization and immunoreactivity of the extracellular matrix, including collagen types III, IV, VI, laminin, and alpha elastin in the lamina cribrosa of monkey eyes with normal and experimentally chronic glaucoma. Our results showed: (1) abnormal linearlike immunodeposits of both collagen type IV and laminin in the margin of the lamina cribrosa with significant density in the glaucomatous eyes; (2) the immunoreactivity of collagen type III resembled that of the normal eye, but was slightly stronger at the laminar surface; (3) findings with collagen type VI resembled those of type III with an enhanced linearlike staining surrounding the nerve-fiber bundles. Furthermore, staining of alpha elastin demonstrated dramatic changes in both reactivity and localization. The lamina cribrosa of glaucomatous eyes showed a markedly reduced immunoreactivity as well as an irregular, interrupted pattern. These obervations suggest that the changes might be a secondary to the long-standing elevation of intraocular pressure. The alteration of these macromolecules may modify the course of glaucomatous optic damage.

Introduction

The extracellular matrix consists of a variety of many macromolecules that fill a large part of the extracellular space [5, 6, 12] and has important functions in maintaining the architecture and mechanical stability of the tissue. Besides this structural role, the extracellular matrix and its components also interact with the cells and influence a variety of critical cell functions such as migration, differentiation, polarity, and tissue-specific gene expression. In the lamina cribrosa of the optic nerve, the extracellular matrix has been postulated to provide both mechanical and functional support for the optic nerve fiber Correspondence to: T. Fukuchi

bundles [1, 10]; thus an abnormality or an alteration of the extracellular matrix could affect the strength and elasticity of the laminar beams as a result of glaucomatous changes in the optic nerve head [11, 16, 17]. Previous studies on the lamina cribrosa in the eyes of normal humans [4, 7-9, 14, 19] and normal monkeys [3, 13] have demonstrated the existence of structural components and several kinds of glycoprotein as a portion of the normal extracellular matrix. The interstitial laminar beams were composed of mainly collagen types I, III, V, VI, and fibronectin. Collagen type IV and laminin were localized at the margin of the laminar beams and basement membrane of the blood vessels. Alpha elastin and fibrillin, which make up the elastic fibers, were abundant. The glial basement membrane covering the laminar beams showed a fine meshlike structure traversing the scleral canal and may affect tissue elasticity via an interaction with the elastic fibers [8]. In human eyes with glaucoma, biochemical analysis has demonstrated the presence of altered extracellular matrix components in the lamina cribrosa [20]. Also, few studies [2, 9, 15] have dealt with changes in the macromolecules that comprise the extracellular matrix of the lamina cribrosa in human eyes with glaucoma or monkey eyes with experimental glaucoma using immunohistochemical procedures. The biotin-streptoavidin (B-SA) method and light microscopy were used in this study. Considered in conjunction with clinical findings, we have demonstrated the alteration of the macromolecules that comprise the extracellular matrix of the optic nerve heads in monkeys with laser-induced chronic glaucoma as compared to findings in the normal contralateral eye. The differences and similarities of these macromolecules with those reported in previous studies are discussed. Materials and methods Chronic glaucoma induced by laser administration

Three normal adult cynomolgusmonkeys(Macaca irus) were used in this study. In each animal, glaucoma was induced in the right

422 lOP mmHg

I

• • rt. eye A ~ It, eye

50 g Q

40

ot

•

30

I f

•t

20

Z~

A

Z~

z~

A

A

10 months

Case

1

Case

2

2

3

Case

3

4

Fig. 1. Course in IOP in three monkeys with experimentally induced chronic glaucoma (arrows:the time of laser application)

5

Fig. 2. Stereofundus photograph of case 1. At 2.5 months after the initial elevation of IOP, the right eye shows glaucomatous large

cupping of the optic nerve head and temporal peripapillary choroidal atrophy (B)

Fig. 3. A Comparison of histological sections of the optic nerve heads of a normal eye and B an eye with experimental glaucoma (stained with hematoxylin, x 60). Cryo-section of the optic nerve head of the right eye of case 1 (B) showed the reduced volume

of prelaminar tissue, a marked increase in the number of nuclear stainings for glial cells, and an irregular arrangement of glial cells in the laminar and the retrolaminar portions (arrows)

eye while the left eye served as the control. All procedures were performed under deep surgical anesthesia induced by administering intramuscular ketamine hydrochloride 9 mg/kg and intravenous sodium pentobarbital 11 mg/kg. Procedures conformed with those of the Research in Vision and Ophthalmology resolution on the use of animals in research. After the eye of each monkey was treated with oxybuprocaine-HC1 0.4%, 201~400 circumferential ar-

gon laser burns were made at one time and then repeated with a gonioscopic lens, aiming at the middle of the trabecular meshwork. We used a 100 gm beam diameter, 0.2 s duration and 600800 MW power. IOP measurement by Alcon applanation pneumotonograph and fundus photography using a Topcon stereo-fundus camera was performed every 1 2 weeks. The monkeys were then

423

Fig. 4A-F, Comparison of immunostaining for collagen type IV (A, B, C, D) and laminin (E, F) in normal eyes (A, C, E) and eyes with experimental glaucoma (B, D, F) (A, B, x 80; C, D, x 240; E, F, x 160). In the glaucomatous eyes, the staining pattern of the optic nerve head for collagen type IV (B, D) and laminin (F) was similar to that of the normal counterpart (type IV, A, C; laminin, E). The edges of the laminar beams and pial septa showed linear staining, staining along the glial columns and vessel

killed with an overdose of intravenous pentobarbital and intramuscular ketamine-HC1 at 2.5, 3, and 5 months after the initial rise in IOP.

Immunohistochemical studies Soon after enucleation, the bilateral optic nerve heads were carefully dissected and rinsed in 0.01 M phosphate-buffered saline (PBS; pH 7.2). Small pieces of the optic nerve heads were embedded in O.C.T compound and a plastic embedding capsule (both by TissueTek II from Miles Laboratories, Naperville, Ill.) and were then flash-frozen in thiopentane cooled in liquid nitrogen. Frozen sec-

walls in the prelaminar portion was well preserved. Also the anterior part of the lamina cribrosa showed immunostaining around the blood vessels (B, F, arrows). Under high power (D, F) the laminar beams in the surface of the lamina cribrosa of the glaucomatous eyes appeared compressed. In the deeper part, thickened laminar beams were evident. The linear immunoreactivity seen along the margin of the laminar beams was denser and more irregular than normal (D, arrowheads)

tions 8 gm thick were cut sagittally at - 2 0 ° C using a microtomecryostat (International Equipment Co, Needham, Mass.), then placed on slides precoated with poly-L-lysine (Sigma Chemical Co, St Louis, Mo.). For immunohistochemical study, we used a biotinstreptoavidin method. The slides were fixed in absolute acetone at - 2 0 ° C for 5 min, air dried, and washed several times with ice-cold 0.01 M PBS. The specimens were then treated with 3% H202 for 10 min to block endogenous peroxidase. Following repeated washing with ice-cold 0.01 M PBS, the sections were soaked in 5% normal goat serum for 20 min at room temperature to reduce non-specific binding.

424 After the blocking serum had been drained off, primary antibodies were applied to the sections, diluted in 0.01 M PBS and incubated for 90 min at room temperature. We used the following polyclonal antibodies: rabbit anti-bovine skin collagen type III (1:600), rabbit anti-bovine renal collagen type 1V (1:1000), rabbit anti-mouse EHS tumor laminin (1:1500) (Advance Co, Tokyo), rabbit anti-human placental collagen type VI (1:500) (Chemicon International, Inc, E1 Segundo, Calif.), rabbit anti-bovine skin alpha elastin (1:500) (Elastin Products Co, Owensvill, Mich.). The specificity of these polyclonal antibodies was evaluated by Ouchterlony's method and by Western blot immunoelectrophoresis. In addition, their specificity was evaluated by incubating the control sections with antibodies to collagen types III, IV and VI, laminin and alpha-elastin, which had been preabsorbed with excess amounts of the following: calf skin collagen type III (Sigma Chemical, St Louis, Mo.); human placental collagen type IV (Cosmo Bio, Tokyo); collagen type VI (Sigma and Fuji Yakuhin, Tokyo); mouse EHS tumor laminin (Collaborative Research, Bedford, Mass.) and human aortic alpha elastin (Elastin Products). No cross-reactivity was shown among these antibodies. As negative controls, sections were also incubated with non-immune normal rabbit serum instead of the primary antibodies in appropriate dilutions. After being rinsed three times with 0.01 M PBS for 5 min each, the sections were reacted with biotinylated secondary antibody (anti-rabbit immunoglobulin)(Biogenex Laboratories, Dublin, Calif.) for 40 min at room temperature, followed by reaction with streptavidin-labeled peroxidase for 40 min at room temperature.

The color reaction was developed with 0.006% H202 in 0.02% 3,3'-diaminobenzidine tetrahydrochloride (Wako Pure Chemical Product, Tokyo) in TRIS-HC1 buffer (pH 7.6). After a final washing in distilled water, all sections were mounted with a mounting media, Crystal-Mount (Biomeda, Foster City, Calif.), examined, and photographed under the light microscope.

Fig. 5A-D. Comparison of immunostaining for collagen type III (A, B) and type VI (C, D) in normal eyes (A, C) and eyes with experimental glaucoma (B, D) (A, B, x80); C, D, x120). For collagen type III (A, B), the thickened laminar beams in glaucomatous eyes (B) showed diffuse stainings similar to the normal control eyes (A). The immunoreactivity of collagen type III at the laminar surface of some glaucomatous eyes appeared slightly stronger in some areas (B, arrows). The areas of the pial septa in the retrola-

minar portion were larger and stained more densely than normal eye. The immunostaining for collagen type VI (C, D) was virtually identical to findings for collagen type III (A, B). An enhanced linearlike immunostaining around the nerve fiber bundles was also detected. In the glaucomatous eyes (D), the laminar beams were stained irregularly and roughly. Thickened laminar beams and irregular laminar channels were also noted

Results T h e course o f I O P changes in the n o r m a l a n d the experim e n t a l eyes are s u m m a r i z e d in Fig. 1. All three right e x p e r i m e n t a l eyes showed deep g l a u c o m a t o u s large cupTable 1. Intraocular pressure and optic nerve damage estimates Monkey Diagnosis case no.

Mean IOP Duration Neural area (mmHg) (weeks) remaining (%)

1 R L 2 R L 3 R L

27.1 17.8 32.6 17.5 31.8 18.6

Mild glaucoma Normal control Moderate glaucoma Normal control Severe glaucoma Normal control

12 10 19 -

60 ~ 70 50~60 < 20 -

425

ping at the time of enucleation (the stero-fundus photographs of case 1 are shown in Fig. 2). The estimation of the optic nerve damage in each of the eyes is demonstrated in Table 1. The normal monkey optic nerve head showed linearlike staining in the surface of the laminar beams and the pial septa for collagen type IV and laminin (Fig. 4). Along the glial columns and the vascular walls in the prelaminar portion, the staining pattern and intensity were obvious. In glaucomatous eyes, although the staining pattern for collagen type IV and laminin appeared similar, the immunoreactivity just anterior to the lamina cribrosa appeared more intense. The anterior portion of the laminar beams appeared compressed. Thickened laminar beams were evident in the deeper part of the lamina cribrosa in glaucomatous eyes. The linearlike immunoreactivity seen along the margin of the laminar beams was more intense and more irregular compared to the normal control. In addition, dotlike staining for collagen type IV and laminin was identified in areas of the prelaminar portion. Collagen type III (Fig. 5A, B) was demonstrated in the scleral portion of the lamina cribrosa diffusely with an immunoreactivity identical to that of the circumferential adjacent sclera and posterior pial septa of the normal eye. Scattered

staining was also seen along the glial columns in the normal eye for type III collagen. The immunoreactivity and the pattern of the glaucomatous laminar beams appeared the same as their normal counterparts for collagen type III. Only a slightly stronger iummunoreactivity with this type III collagen was observed in the laminar surface of one glaucomatous eyes, the pattern and immunoreactivity for type III collagen were similar in these eyes. The staining of collagen type VI (Fig. 5C, D) showed virtually identical findings to those for collagen type III in normal and glaucomatous eyes. The sclera, laminar beams and pial septa were all diffusely stained. An enhanced linearlike staining along the edge of the laminar beams, the pial septa, and the vascular walls was demonstrated in the normal eye. In the normal eye, numerous fine bandlike immunodeposits of alpha elastin (Fig. 6) were detected within the glial columns and laminar beams crossing the nerve fiber bundles. Faint, bandlike staining was observed in the pial septa along the optic nerve fiber bundles. A strong immunoreactivity for alpha elastin was demonstrated in the arachnoid membrane of the normal optic nerve heads. Immunoreactivity was markedly reduced along the glial columns. In the lamina cribrosa, a mild-to-moderate reduction in immunoreactivity was demonstrated in glaucomatous

Fig. 6A-D. Comparison of immunostaining for alpha elastin in normal eyes (A, C) versus eyes with experimental glaucoma (B, D). (A-D, x 120). Whereas immunostaining for alpha elastin was always prominent in normal eyes (A, C), in glaucoma eyes (B, D) immunoreactivity along the glial columns was markedly reduced; immunostaining for alpha elastin was also reduced moderately in the laminar portion. The pattern and degree of immunore-

activity for alpha elastin varied among the glaucomatous eyes. In case 1 (A, left normal eye; B, right glaucomatous eye), the immunoreactivity was reduced to the greatest extent and abnormal immunodeposits were detected in the surface of the lamina cribrosa. In case 3 (C, left normal eye; D, right glaucomatous eye), immunoreactivity was also reduced along the glial columns, but still persisted

426 eyes. Interestingly, the severity of this reduction varied from mild to moderate among the three monkeys. In case 1 (Fig. 6A, B) with mild nerve-fiber damage, a marked reduction in immunoreactivity was seen in the lamin cribrosa, and abnormal immunodeposits were observed on the surface of the lamina cribrosa of the glaucomatous eye. In case 3 (Fig. 6C, D) with the most severe nerve fiber damage, only a mild reduction of immunoreactivity was seen in the lamina cribrosa with a significant reduction of immunoreactivity in the glial columns in the glaucomatous eye as compared with that of the contralateral control eye. Furthermore, in histological cryo-sections stained with hematoxylin (Fig. 3), glaucomatous eyes showed a reduced amount of both prelaminar tissue volume and retinal nerve fiber layer thickness. Nuclear staining showed an increased number of glial cells in the lamina cribrosa and retrolaminar portion.

Discussion

A persistent elevation of IOP damages the optic nerve and the lamina cribrosa, resulting in glaucomatous optic disc cupping [11, 16, 17]. Besides the level of IOP and its duration, the extracellular matrix itself may play a role in the pathogenesis of optic disc cupping since the susceptibility of the optic disc to the elevated IOP varies with the individual. To clarify the pathogenesis of glaucomatous optic disc cupping, few studies have reported changes in the macromolecules constituting the extracellular matrix of the lamina cribrosa. This study of laser-induced glaucoma in monkeys has examined changes in the following structural proteins: collagen type III, IV, VI, alpha elastin, and laminin, a glycoprotein. Some authors [2, 9, 20] have reported extracellular matrix changes in the lamina cribrosa in human glaucoma. However, it is uncertain whether these findings are a primary abnormality or are secondary to glaucoma. Because we compared the experimental eye with its normal counterpart, we conclude that the findings observed a secondary to a long-standing elevation of IOP and follow the progression of optic disc cupping. Hernandez et al. [9] and Floyd et al. [2] have reported changes of the extracellular matrix in human glaucoma by an immunohistochemical method. Hernandez [9] has shown a marked loss of elastin, an increase in collagen type VI, and the extension and accumulation of collagen type IV and other basement membrane molecules in 17 human eyes with primary open-angle glaucoma. An increased laminin staining in the prelaminar portion of eyes with severe end-stage glaucomatous cupping has been described by Floyd et al. [2]. In a study of experimental glaucoma in the monkey, Morrison et al. [15] have demonstrated compressed and thickened laminar beams without qualitative differences between normal and glaucomatous eyes for collagen types I, III, elastin, and a thicker astrocyte-associated immunolabeling of basement membrane. Quigley et al. [18] have described structural changes in alpha elastin, but the number of fibers in eyes with human glaucoma and monkey experi-

mental glaucoma is as high as in normal eyes, as observed by light and electron microscopy. The abovementioned authors have also demonstrated a loss of small-diameter collagen fibers and a sparse distribution of the same in the lamina cribrosa. Although a decrease in the density of collagen fibers has been shown in glaucomatous lamina cribrosa by electron microscopy, the laminar beams are generally thickened, and the interstitial collagen such as types I and III show no significant immunohistochemical changes. During the development of glaucomatous optic disc cupping, cellular components such as astrocyte and fibroblast may change their cellular activities, resulting in an altered extracellular matrix composition in addition to structural changes. Interestingly, the findings reported for alpha elastin by Morrison [15] differ both from those of Hernandez [9] and the present study. Quigley [18] has reported that the number of the elastic fibers in glaucoma is as same as in the normal eye, but that most of the fibers appear to be degenerated. In contrast, the study by Hernandez and that reported here each show a marked loss of alpha elastin in both the lamina cribrosa and the glial columns in the prelaminar portion. The differences may reflect the severity of optic disc cupping or stage of glaucoma. Because of the differing sources of antibodies against alpha elastin, it is possible that our antibody did not react with degenerated alpha elastin. A marked loss of immunoreactivity or an extremely altered structure for alpha elastin should respond to the decrease of elasticity in the optic nerve heads seen in glaucoma. There are several similarities between glaucomatous optic nerve heads in humans and monkeys regarding the immunoreactivity of extracellular matrix components. These similarities suggest that there is an advantage to using an experimental model of glaucoma to investigate the extracellular matrix of optic nerve heads during this disease process. The similarities suggest that findings observed in human glaucoma may, at least in part, be secondary to a long-standing elevation of IOP. The results of our study as well as those of others indicate that immunoreactivity for extracellular matrix components, especially for alpha elastin, are not closely correlated with the level of IOP, the duration of elevation, or the severity of glaucomatous cupping. This suggests that some extracellular matrix components, at least alpha elastin, have their own critical strength against each level of IOP. Once changes have occurred either in the structure or the components of the lamina cribrosa that maintain the mechanical and functional support of the optic nerve head, the process of glaucoma may be accelerated.

References

1. AndersonDR, Francisco S (I 969) Ultrastructure of human and monkey lamina cribrosa and optic nerve head. Arch Ophthalmol 82:800-814 2. Floyd BB, Cleveland PH, Worthen DM (1988) Extracellular matrix of the optic nerve head in normal and glaucomatous eyes. Invest OphthalmolVis Sci 29 [ARVO Suppl] :353

427 3. Fukuchi T (1990) Extracellular matrix of the normal monkey optic nerve lamina cribrosa. Acta Soc Ophthalmol Jpn 94:1024-1030 4. Goldbaum MH, Jeng S, Longemann R, Weinreb RN (1989) The extracellular matrix of the human optic nerve head. Arch Ophthalmol 107:1225-1231 5. Goldberg BD, Rabinovitch M (1988) Connective tissue. In: Weiss L (ed) Cell and tissue biology, 6th edn. Urban and Schwarzenberg, Baltimore, pp 155 173 6. Hay ED (1981) Extracellular matrix. J Cell Biol 91:205 223 7. Hernandez MR, Igoe F, Neufeld AH (1986) Extracellular matrix of the human optic nerve head. Am J Ophthalmol 102 : 139148 8. Hernandez MR, Luo XX, Igoe F, Neufeld AH (1987) Extracellutar matrix of the human lamina cribrosa. Am J Ophthalmol 104:567 576 9. Hernandez MR, Andrzejewska WM, Neufeld AH (1990) Changes in the extracellular matrix of the human optic nerve head in primary open angle glaucoma. Am J Ophthalmol 109:180-188 10. Hogan MJ, Alvarade JA, Weddell JE (1971) Optic nerve. In: Histology of the human eye. Saunders, Philadelphia, pp 523606 11. Iwata K (1985) Pathophysiology in the early stage of primary open-angle glaucoma. Jpn J Clin Ophthalmol 39:407-424 12. Kajikawa K (1984) Connective tissue. Kinbara, Tokyo, pp I 8, 31 223, 279-382

13. Morrison JC, Jerden JA, L'Hernault NL, Quigley HA (1988) The extracellular matrix composition of the monkey optic nerve head. Invest Ophthalmol Vis Sci 29 : 1141-1150 14. Morrison JC, Jerden JA, Dorman ME, Quigley HA (1989) U1trastructural localization of extracellular matrix components in the optic nerve head. Arch Ophthalmol 107:123-129 15. Morrison JC, Dorman Pease ME, Dunkelberger MS, etal. (1990) Optic nerve head extracellular matrix in primary optic atrophy and experimental glaucoma. Arch Ophthalmol 108:1020-1024 16. Quigley HA, Addicks W, Green R, et al. (1981) Optic nerve damage in human glaucoma. II. The site of injury and susceptibility of damage. Arch Ophthalmol 99:635-649 17. Quigley HA, Hohman RM, Addicks W, et al. (1983) Changes in the lamina cribrosa correlated with neural loss in open-angle glaucoma. Am J Ophthalmol 95:673-691 18. Quigley HA, Dorman-Pease ME, Dunkelberger G, et al. (1990) Changes in collagen and elastin in the optic nerve head in chronic human and experimental monkey glaucoma. Invest Ophthalmol Vis Sci 31 [ARVO suppl] :564 19. Rehenberg M, Ammitzboll T, Tengroth B (1987) Collagen distribution in the lamin cribrosa and the trabecular meshwork of the human eye. Br J Ophthalmol 71 : 886-892 20. Tengroth B, Ammitzboll T (1984) Change in the content and composition of collagen in the glaucomatous eye. A preliminary report. Acta Ophthalmol (Copenh) 62:999 1008