Quantitative Studies of Elastin in the Optic Nerve Heads of Persons with Primary Open-angle Glaucoma

Quantitative Studies of Elastin in the Optic Nerve Heads of Persons with Primary Open-angle Glaucoma

Quantitative Studies of Elastin in the Optic Nerve Heads of Persons with Primary Open--angle Glaucoma Erica N. Quigley, Harry A. Quigley, MD, Mary E. ...

2MB Sizes 0 Downloads 11 Views

Quantitative Studies of Elastin in the Optic Nerve Heads of Persons with Primary Open--angle Glaucoma Erica N. Quigley, Harry A. Quigley, MD, Mary E. Pease, BS, Lisa A. Kerrigan, BS Purpose: To compare quantitative and qualitative differences in elastin content in the optic nerve heads of glaucomatous and control human eyes. Methods: Transmission electron microscopy and quantitative histomorphometry on ten control and ten glaucomatous eyes. Results: Elastin fiber complexes in the control lamina cribrosa were smaller and more numerous than in the insertion zone of the sclera immediately surrounding the lamina. Although the density of elastin fibers in the normal lamina was twice that of the insertion zone (P = 0.004), the percent area of the connective tissue matrix occupied by elastin was the same for both zones (P > 0.4). There was no difference between control and glaucomatous eyes in the quantified parameters of elastin content or in the ultrastructure of elastin between control and glaucomatous eyes. Conclusions: The authors demonstrated for the first time that elastin in the normal lamina consists of fibers of smaller diameter than in the adjacent sclera, although the total amount of elastin is similar in both locations. This may provide maximum viscoelasticity within the limited connective tissue beam area of the lamina. Despite using a large number of specimens, the authors again found no differences between normal and glaucomatous eyes in the number or ultrastructural appearance of elastin fibers. Ophthalmology 1996; 103: 1680-1685

Primary open-angle glaucoma is an optic neuropathy that consists of a characteristic alteration in the physical appearance of the optic nerve head and loss of visual function beginning in the midperipheral field. It is the excavation of the nerve head tissues that most easily distinguish glaucoma from other optic neuropathies that produce similar field defects. 1,2 Although loss of ganglion cell axons causes a loss of nerve head tissue volume, the deep and undermined appearance of the optic disc in glaucoma Originally received: March 6, 1996. Revision accepted: May 30, 1996. From the Glaucoma Service and the Daria Center for Preventive Ophthalmology, Wilmer Institute, Johns Hopkins University School of Medicine, Baltimore, Supported in part by PHS research grants EY 02120 (Dr. H. Quigley) and EY 01765 (core facility grant, Wilmer Institute), both awarded by the National Eye Institute, National Institutes of Health, Bethesda, Maryland. Reprint requests to Erica N. Quigley, Wilmer 120, Johns Hopkins Hospital, 600 N. Wolfe St,Baltimore, MD 21287-9019.

1680

results from an alteration in the connective tissues of the lamina cribrosa of the nerve head. 3 - 5 These consist of stretching of the lamina cribrosa beams and their backward and lateral movement. In many human eyes with primary and secondary glaucoma, as well as in animal models, intraocular pressure above the range considered normal is clearly associated with the production of nerve head excavation. In some persons with glaucoma, the excavation and field loss occur at normal intraocular pressure. In these persons, the change in connective tissue occurs under physical forces that are tolerated by many eyes. This may indicate that they have a fundamental weakness in extracellular matrix structure in the nerve head. To understand the pathogenesis of glaucoma, it is important to determine the factors that lead to alterations in the extracellular matrix of the nerve head. Anderson 6 first detected the presence of elastin in the nerve head by transmission electron microscopy. Hernandez? showed the abundance of elastin both within the lamina cribrosa

Quigley et al . Elastin in the Optic Nerve Heads in POAG and in the insertion zone immediately surrounding the nerve head by immunofluorescence and immunoelectron microscopy. There have been several examinations of elastin in the nerve heads of human glaucoma eyes and in monkey eyes with experimental glaucoma. These reports indicate that there is a change in elastin in glaucoma, but they do not agree on the type or amount of alteration. Hernandez and colleagues8 initially reported a loss of elastin fibers within the cores of the cribriform plates and suggested that tubular elastic fibers were no longer identifiable. 7 More elastin rnRNA was detected in glaucomatous nerve heads than in controls,9 although the significance of this finding is as yet unclear. In another report, Netland and co-workers 1o described the findings in glaucomatous nerve heads as showing "elastosis," defined as proliferation and degeneration of elastic elements. It was suggested that there are quantitatively more elastin fibrils present in glaucoma nerve heads than in normal nerve heads. Previous studies from our laboratory indicate that abnormalities in elastin do occur in glaucoma eyes. Results of examinations that demonstrate the disposition of elastin fibrils show that in normal eyes the fibrils are straight, as if under tension. In glaucomatous eyes, on the other hand, the fibrils are often curled. I 1.12 In addition, results of electron microscopic examination of elastin profiles show that the fibrils are separated from the surrounding collagenous matrix.13 Our quantitative evaluations showed no change in the number of elastin profiles detected in glaucoma nerve heads per unit area in six glaucomatous eyes. To evaluate further the state of elastin in the glaucomatous optic nerve head, we examined the largest number of human glaucomatous eyes yet studied by quantitative histomorphometry. We measured the number, size, and shape of elastin profiles and described their disposition in the extracellular matrix in masked evaluations by electron microscopy.

Methods This research was approved by the Joint Committee for Clinical Investigation of the Johns Hopkins University School of Medicine. Eyes were obtained after donation to eye banks or from the Ophthalmic Pathology Laboratory of the Johns Hopkins Hospital. Normal eyes were from persons whose hospital or eye bank record showed no history of ocular disease (other than cataract extraction in some cases). All eyes were examined at a dissecting microscope and were excluded if any pathologic findings other than glaucoma were present. Both control and glaucoma globes were embedded in paraffin and sectioned in the horizontal plane through the pupil and posterior pole. A variety of histochemical stains, including hematoxylineosin and periodic acid-Schiff, were used for these paraffin sections to identify any abnormalities, with special attention to deposits of exfoliative material in the anterior ocular structures. No eye had any evidence of exfoliation by history or on postmortem or histologic examination. Ten control and ten glaucomatous eyes (from 20 individuals) were fixed in glutaraldehyde and paraformalde-

hyde buffered in phosphate. The time to enucleation was a mean of 5.3 ± 4.0 hours in control eyes and 3.2 ± 1.1 hours in glaucomatous eyes (P = 0.21, two-tailed Student's t test). The additional time to fixation was 12.8 ± 6.8 hours in control eyes and 10.4 ± 14.8 in glaucomatous eyes (P = 0.8, two-tailed Student's t test). Control eyes were selected to match the age range of the donors with glaucoma. The mean age for control persons was 79.3 ± 10.3 years, and for those with glaucoma, it was 79.9 ± 8.7 years (P = 0.9, two-tailed Student's t test). Of the ten persons in each group, five of the control subjects and six of the glaucoma subjects were male. All subjects were white. The optic nerve was removed 1 to 3 mm posterior to the globe and marked with razor cuts at the superior and nasal borders. The optic nerve head was removed with a surrounding portion of sclera that was marked to indicate the superior and temporal positions. The tissues were post-fixed in 1% osmium tetroxide in phosphate buffer and embedded in Epoxy resin. The optic nerve heads were divided into quadrants, and their inferior aspect was selected for electron microscopic examination. The thin sections were stained with uranyl acetate and lead citrate. Optic nerve cross-sections were sectioned at 1 J.lm and examined by light microscopy in masked fashion to determine the degree of damage. This was performed by one of us (HAQ), who measured the neural area of the whole nerve and determined whether there were localized areas of atrophy. In one control and one glaucoma eye, the optic nerve was too poorly preserved for evaluation; the presence and extent of glaucoma damage in this subject with glaucoma was confirmed by severely abnormal results of a Humphrey visual field test, which was provided as part of the premortem clinical record. Glaucoma was defined by a clinical history of treatment by an ophthalmologist or optometrist for glaucoma and the presence of optic nerve damage on the crosssection. To maximize the difference between control subjects and those with glaucoma, we selected only glaucomatous optic nerve heads from eyes with more than 50% loss of neural area. Four eyes had essentially no remaining fibers. In the one example that had no available optic nerve cross-section, results from a Humphrey visual field test from premortem records showed severe damage. All controls had neural areas within 25% of the normal mean value established in previous data from this laboratory 14 and had no localized atrophy in their nerves. Sections from the inferior quadrant of the optic nerve head were examined using a transmission electron microscope (JEOL 100CX, Tokyo) by an observer masked as to whether the nerve was a control or a glaucoma specimen. The lamina cribrosa and insertion zone adjacent to the lamina were identified. The lamina consists of the portion of the optic nerve head at the level of the sclera, containing both collagenous connective tissue and nerve fiber bundles. The insertion zone is the portion of the sclera immediately adjacent to the nerve head and extending up to 500 microns away from it. From each zone (lamina and insertion zone), 25 photographs were taken overlying randomly selected areas of the extracellular matrix. Contact prints of 3 X 4-inch negatives were evaluated

1681

Ophthalmology

Volume 103, Number 10, October 1996

Table 1. Reproducibility of Image Analysis Method Analysis No.

2

Total fiber area (J-lm ) Average fiber area (J-lm 2 ) Mean:!:: SO No. of elastin fibers % Fiber area Mean fiber diameter (J-lm)

so =

2

3

22.03

22.59

22.40

0.46 :!:: 0.46

0.47 :!:: 0.45 48 6.2 1.01

0.47 :!:: 0.44 48 6.1 0.99

48 6.0 0.99

standard deviation.

by another masked observer using a Micro-Plan II image analysis tablet (Laboratory Computer Systems, Inc, Cambridge, MA). Elastin complexes were identified by their characteristic configuration, including a peripheral zone of fine fibrillar material and a central amorphous component. The observer carrying out this portion of the experiment is experienced in the ultrastructural appearance of elastin as identified by immunoelectron microscopic studies with antibodies against alpha elastin in previously published experiments. 15 .1 6 Within each photograph, the area of extracellular matrix (not including axonal tissue) was outlined and measured. Then, each elastin profile was outlined individually, with calculations producing values for maximum diameter of a fiber, form factor (deviation from circular shape), percent of matrix occupied by elastin, number of elastin fibers per unit area (density), and mean area of elastin. One set of 25 electron-microscopic negatives from a control insertion zone specimen was analyzed three times for elastin content to estimate the reproducibility of the method. The negatives were printed three times and analyzed in three separate sessions with-

out reference to the other sessions. The values obtained indicate a high level of reproducibility (Table 1). Another observer examined the photographs in masked fashion to determine whether any qualitative differences could be reproducibly discerned in the appearance of the elastin complexes.

Results The total area of connective tissue matrix that was measured in each set of 25 photographs from an individual specimen varied between 300 and 400 j.lm 2 • In control and glaucoma specimens, the form factor (a measure of the shape of each elastin fiber complex) had an identical mean for the lamina and the insertion zone. The mean number of elastin fiber complexes detected in the control lamina cribrosa was 64% greater than in the control insertion zone (Table 2). The density of elastin fibers, calculated as the number per square micron of matrix area, was almost exactly twice as great in the lamina as in the

Table 2. Image Analysis Data: Glaucoma and Control Specimens Mean Area (p.m 2 )

Fiber Number

Density*

% Fiber Areat

Maximum Diameter (p.m)

Glaucoma Lamina Mean:!:: SO Insertion Mean:!:: SO

0.19 :!:: 0.07

154 :!:: 97

0.50 :!:: 0.31

7.85 :!:: 2.65

0.59 :!:: 0.12

0.37 :!:: 0.14 0.47

106 :!:: 59 0.20

0.29 :!:: 0.16 0.08

8.96 :!:: 3.16 0.40

0.83 :!:: 0.20 0.004

Control Lamina Mean:!:: SO Insertion Mean:!:: SO

0.17 :!:: 0.13

170:!:: 73

0.61 :!:: 0.26

7.65 :!:: 2.20

0.54 :!:: 0.19

0.30 :!:: 0.12 0.03

109 :!:: 49 0.04

0.30 :!:: 0.14 0.004

7.87 :!:: 1.94 0.82

0.70 :!:: 0.18 0.06

N

p:j:

SO = standard deviation. * Number of elastin fiber complexes per square micron of extracellular matrix. t Area of elastin complexes/area of extracellular matrix X 100. Differences between glaucoma and control in each unit category are not significant (two-tailed Student's t tests, lowest P = 0.16). See Table 3 for power calculations.

* 1682

Quigley et al . Elastin in the Optic Nerve Heads in POAG 25

en

IX: 20

w

CD

u:: 15

11.

0

IX: 10

w

CD ~

:::J 5 Z

0

A

0

0.1

FIBER DIAMETER (microns)

-e- glaucoma LC ...... glaucoma IZ ....... control LC

0.2

........ controllZ

16.-------------------------------.

en

IX:

14

W 12

CD

u:: 10 11. o 8 In

6 4

CD ~

:::J Z 2

O~"._,,~_r,,~~~~""_r~~~

B

o

0.1

0.2

FIBER DIAMETER (microns) ....... control LC ....... control IZ

16

en

14

W CD

12

IX:

~---------------------------------.

u:: 10 o 8

insertion zone. The average area per fiber was 56% smaller in the lamina zone compared with the insertion zone. Likewise, the maximum diameter of each fiber was 33% smaller in the lamina compared with the insertion zone. Thus, when one analyzes the percent of each zone that is comprised of elastin, the fibers in the lamina constituted nearly the same percent area of the matrix as the fibers in the insertion zone. However, laminar fibers were smaller and more numerous, whereas insertion zone fibers were larger and fewer. These differences are seen graphically when the number of elastin profiles is plotted against elastin mean diameter (Fig 1). The glaucoma specimens were not different from agematched control specimens in any parameter (Table 2). As in the control specimens, the glaucomatous eyes had smaller, more numerous elastin fibers in the lamina cribrosa, but an equal proportion of their lamina consisted of elastin compared with their insertion zone. We calculated the statistical power to determine the difference between control and glaucoma specimens for all of the major variables (Table 3). For percent elastin area, the most important variable, we had an 89% to 91 % power to determine a difference of 50% or greater. In the masked, qualitative evaluation, no difference was found between glaucoma and controls in the appearance of the elastin. Specifically, none of the glaucoma specimens differed from the control subjects in the appearance of the amorphous central component of the elastin complex (Fig 2). It was clear that the elastin fibers were more numerous and smaller in the lamina compared with the insertion zone, just as the quantitation demonstrated. One glaucoma specimen appeared to have dramatically more elastin, both qualitatively and quantitatively, than any of the others.

Discussion

11.

IX:

W 6

CD ~

4

Z

2

:::J

C

o L,,,-.,,-."".-,,.-~._,,._~~~~

o

0.1

0.2

FIBER DIAMETER (microns)

We found that the elastin content in normal human lamina cribrosa is different from that of the insertion zone. In the lamina, the elastin fiber complexes are smaller and more numerous than in the insertion zone. Even though the elastin fibers are smaller in the lamina, the percent of the connective tissue occupied by elastin is the same for

-e- glaucoma LC ...... glaucoma IZ

Figure 1. The graphs show the distribution of elastin diameter in normal and glaucomatous eyes, both in the lamina cribrosa and in the peripapillary insertion zone. A, the distributions of both control and glaucoma lamina (open square and filled triangle) show more smaller fibrils (2 upper lines) than the control and glaucoma insertion zones (cross and filled circle, the 2 lower lines) . Notice that there isno substantive difference between control and glaucoma. B, to show the comparative shape of the distribution in control lamina (filled triangle) compared with control insertion zone (cross), the total number of elastin fibrils was equated in the two distributions (the lamina number was made equal to the insertion zone number). This s hows that the lamina has a greater number of the smallest elastin fibrils. C, an identical normalization was performed to compare the lamina and the insertion zone in the glaucoma speci mens. As in controls, the lamina has a greater proportion of the smallest fibrils.

Table 3. Power to Determine Differences in Elastin Parameters between Glaucoma and Control Groups Potential % Difference

Lamina Density % Fiber area Maximum diameter Insertion zone Density % Fiber area Maximum diameter

100%

80%

50%

0.99 1.00 1.00

0.95 1.00 1.00

0.61 0.9 1 0.95

0.99 1.00 1.00

0.92 1.00 1.00

0.56 0.89 0.98

1683

Ophthalmology

Volume 103, Number 10, October 1996

Figure 2. Elastin fibrils are shown by transmission electron microscopy. A, in the insertion zone, the elastin complexes (arrows) are typically larger than in the lamina (shown in 2B). Notice that the collagen fibrils (in areas indicated by asterisks) of the insertion zone (A) are also generally larger in diameter than in the lamina (B). In one glaucoma specimen that was unique in our series, there were substantially more profiles of elastin in the lamina than in either control specimens or among other glaucoma specimens (C) (A = glaucoma specimen, insertion zone; B = control specimen, lamina cribrosa; C = glaucoma specimen, lamina cribrosa; original magnification, x30,OOO).

both regions. Although the insertion zone is made up almost entirely of extracellular matrix, the lamina is approximately one-half extracellular matrix and one-half nerve fibers. Hence, there is a regional specialization in the fiber structure of the extracellular matrix that may maximize the structural support of the lamina. The finding of small, uniform-sized elastin fibers is analogous to our previous finding that collagen is smaller and more uni-

1684

form in diameter in the lamina than in the insertion zone or in the remainder of the sclera. II For both elastin and collagen, these smaller fibers may permit the narrow beams of the lamina to contain the highest possible density of fibrous connective tissue elements. This would maximize viscoelastic support to protect the axons that pass through the lamina. The size and disposition of connective tissue fibers is determined by post-translational modification of tropoelastin and tropocollagen. It is reasonable to assume that the susceptibility to abnormal stretching and failure of connective tissue support in the optic nerve head is a result of differences in the detailed disposition of these fibers at the optic nerve head. This leads to the conclusion that there may be no difference in basic genetic sequences for collagen and elastin among persons with widely varying end-organ connective tissue properties (and sensitivities to glaucomatous damage). There was no difference between glaucomatous and control eyes in elastin morphology, either quantitatively or qualitatively. We did not find that there were either more or fewer elastin fibers; nor was there any detectable difference in the ultrastructural appearance of the amorphous elastin component in masked examinations of a large number of glaucoma and control specimens. If there is no difference between severely damaged glaucomatous eyes and control eyes, then glaucomatous eyes are unlikely to differ at any stage from normal eyes. Did we miss some difference in elastin as suggested by other researchers?lO This is not likely, because our study was based on a larger sample of both glaucoma and control eyes. In addition, the glaucoma eyes that we used had well-characterized glaucoma both in diagnosis and stage of disease. Our calculations of statistical power suggest that if the threefold difference between control and glaucoma as reported by Netland and colleagues had been present in our specimens, we would have been almost certain to have detected it. In fact, we had a high probability of detecting a difference as small as 50% in the major quantitative properties of elastin fiber complexes. One major difference between our study and that of Netland et al is that none of our subjects had evidence of exfoliation glaucoma. It may be that exfoliation eyes are different in both number and appearance of elastin fibers from both normal and primary open-angle glaucoma eyes. However, in this and in a previous studyll we found that primary open-angle glaucoma eyes have no differences from control eyes in the number and size of elastin fiber complexes. We have shown l2 ,13 that the elastin of glaucomatous eyes takes on a curled appearance when viewed with the light microscopic, Luna staining method. This method shows the three-dimensional configuration of the tissue, which is not feasible with thin sections of transmission electron microscopy. On the other hand, electron microscopy shows the detailed internal structure of the elastin fiber complex. We found no measurable change in the internal structure or number of elastin fibers. This is consistent with the fact that elastin is a hardy molecule that is quite resistant to degradation by a variety of strong acids and bases.17 It is our hypothesis that elastin takes

Quigley et al . Elastin in the Optic Nerve Heads in POAG up a curled appearance because it is disinserted from its connections to other elements of the extracellular matriX.12 Other disorders that affect elastin, such as aortic aneurysm, lead to a similar, curled appearance of elastin. 18 A potential consequence of this disinsertion is the loss of the normal viscoelastic properties of the lamina. It has been shown previously that there is a change in the elasticity of the lamina in enucleated human and canine glaucomatous eyes. 19.20 Likewise, Burgoyne and co-workers 21 demonstrated alterations in the compliance of the optic nerve head in monkey eyes with experimental glaucoma. This change in compliance could occur without detectable changes in the elastin itself, if its connection to the other elements of the matrix were altered by glaucoma. We suspect that this is the case in the majority of those with primary open-angle glaucoma. Acknowledgments. The authors thank W. Richard Green, MD, who provided control tissues and examined eyes from this study by light microscopy. They also thank Yulan Ding, MS, for biostatistical consultation.

References 1. Emery 1M, Landis D, Paton D, et al. The lamina cribrosa in normal and glaucomatous human eyes. Trans Am Acad Ophthalmol Otolaryngol 1974;78:290-7. 2. Radius RL, Maumenee AB, Green WR. Pit-like changes of the optic nerve head in open-angle glaucoma. Br I Ophthalmol 1978;62:389-93. 3. Quigley HA, Green WR. The histology of human glaucoma cupping and optic nerve damage: clinicopathologic correlation in 21 eyes. Ophthalmology 1979;86:1803-27. 4. Quigley HA, Addicks EM, Green WR, Maumenee AE. Optic nerve damage in human glaucoma. II. The site of injury and susceptibility to damage. Arch Ophthalmol 1981;99:635-49. 5. Quigley HA, Hohman RM, Addicks EM, et al. Morphologic changes in the lamina cribrosa correlated with neural loss in open-angle glaucoma. Am I Ophthalmol 1983;95:67391. 6. Anderson DR. Ultrastructure of human and monkey lamina cribrosa and optic nerve head. Arch Ophthalmol 1969; 82:800-14. 7. Hernandez MR. Ultrastructural immunocytochemical anal-

8.

9. 10. 11.

12.

13. 14. 15. 16.

17. 18. 19. 20.

21.

ysis of elastin in the human lamina cribrosa. Changes in elastic fibers in primary open-angle glaucoma. Invest Ophthalmol Vis Sci 1992;33:2891-903. Hernandez MR, Andrzejewska WM, Neufeld AH. Changes in the extracellular matrix of the human optic nerve head in primary open-angle glaucoma. Am I Ophthalmol 1990; 109:180-8. Hernandez MR, Yang I, Ye H. Activation of elastin mRNA expression in human optic nerve heads with primary openangle glaucoma. I Glaucoma 1994;3:214-25. Netland PA, Ye H, Streeten BW, Hernandez MR. Elastosis of the lamina cribrosa in pseudoexfoliation syndrome with glaucoma. Ophthalmology 1995; 102:878-86. Quigley HA, Dorman-Pease ME, Brown AE. Quantitative study of collagen and elastin of the optic nerve head and sclera in human and experimental monkey glaucoma. Curr Eye Res 1991;10:877-88. Quigley H, Pease ME, Thibault D. Change in the appearance of elastin in the lamina cribrosa of glaucomatous optic nerve heads. Graefes Arch Clin Exp Ophthalmol 1994; 232:257-61. Quigley HA, Brown A, Dorman-Pease ME. Alterations in elastin of the optic nerve head in human and experimental glaucoma. Br I OphthalmolI991;75:552-7. Repka MX, Quigley HA. The effect of age on normal human optic nerve fiber number and diameter. Ophthalmology 1989;96:26-32. Morrison IC, Ierdan lA, Dorman ME, Quigley HA. Structural proteins of the neonatal and adult lamina cribrosa. Arch Ophthalmol 1989; 107:1220-4. Morrison IC, Dorman-Pease ME, Dunkelberger GR, Quigley HA. Optic nerve head extracellular matrix in primary optic atrophy and experimental glaucoma. Arch Ophthalmol 1990; 108:1020-4. Rosenbloom I. Elastin: an overview. Methods Enzymol 1987; 144:172-96. White IV, Haas K, Phillips S, Comerota AI. Adventitial elastolysis is a primary event in aneurysm formation. I Vasc Surg 1993; 17:371-81. Zeimer RC, Ogura Y. The relation between glaucomatous damage and optic nerve head mechanical compliance. Arch Ophthalmol 1989; 107:1232-4. Zeimer RC, Slocum I, Wilensky IT, Shapiro A. Noninvasive measurement of optic nervehead mechanical compliance in normal and ocular hypertensive beagles. Curr Eye Res 1984;3:699-709. Burgoyne CF, Quigley HA, Thompson HW, et al. Early changes in optic disc compliance and surface position in experimental glaucoma. Ophthalmology 1995; 102:1800-9.

1685