Pathophysiologic Changes in the Optic Nerves of Eyes with Primary Open Angle and Pseudoexfoliation Glaucoma

Abstract

purpose. To determine whether differences in the optic nerve occur in eyes with primary versus secondary open-angle glaucoma.

methods. Optic nerves obtained at autopsy from 36 eyes with primary open-angle glaucoma (POAG) and 15 with pseudoexfoliation glaucoma (PEXG) were studied quantitatively and qualitatively. Axon counts, fibrosis, capillary number and density, and arteriosclerotic changes were assessed in the postlaminar optic nerve and compared to normal age-matched autopsy eyes. Changes in composition of extracellular matrix components were evaluated by immunohistochemistry and electron microscopy.

results. Marked differences were found between POAG and PEXG. Axon loss in eyes with POAG but not in PEXG was associated with increasing connective tissue in the septa and surrounding the central retinal vessels, including increased amounts of type IV and VI collagen. The total number of capillaries decreased with the loss of axons in both POAG and PEXG. POAG nerves, however, had a decrease in the density of capillaries, whereas in PEXG the capillary density did not change with axon loss. Arteriosclerotic changes were more common in glaucomatous eyes than in age-matched control eyes.

conclusions. The difference in morphology of the optic nerves between POAG and PEXG indicates that in eyes with POAG, elevated IOP cannot be the only pathogenetic factor in glaucomatous optic neuropathy. Additional factors, inducing fibrosis and loss of capillaries, seem to be involved. Such additional factors may also contribute to the clinical finding in POAG that nerves can become damaged without elevation of intraocular pressure.

Optic nerve damage in glaucoma is usually considered to be a problem caused by increased intraocular pressure (IOP). Optic discs differ in sensitivity to IOP, with some discs withstanding pressure that causes others to become damaged. In normal-pressure glaucoma, cupping also develops, yet with IOPs that are not elevated. Optic disc damage in both primary open-angle glaucoma (POAG) and secondary open-angle glaucoma, such as pseudoexfoliation glaucoma (PEXG), has similar clinical characteristics, with diffuse loss of axons, increased central cupping, and thinning of the disc rim at the superior and inferior poles. 

The cause of the different susceptibilities of optic discs to IOP is not understood. Size and shape of the optic disc, characteristics of the lamina cribrosa, blood supply to the nerve and disc, and inherent axonal characteristics have all been postulated to play a role in susceptibility to damage. ^{1 2 3} Because nearly one half of patients with POAG do not have consistently elevated IOP, this condition has been termed an “optic neuropathy,” with IOP being a “risk factor.” ^{4 5} The multiple potential damaging factors may represent a spectrum, ranging from elevated IOP on one end to ischemic and other factors on the other end. ⁴ Eyes with POAG may lie in the middle of the spectrum, with optic nerve damage caused by a combination of factors. 

Recent histologic studies lend weight to this idea. In eyes with POAG an increase in the amount of elastic tendon sheath material is found in the trabecular meshwork. ^{6 7} This material, also known as sheath-derived (SD)-plaques, is increased in eyes with axon loss in the optic nerve, but does not correlate with IOP. ⁸ This suggests that SD-plaques alone are not responsible for the problem in the meshwork that causes the increase in IOP, but rather accompany the disease process. ⁸ In eyes with PEXG there was a correlation between the PEX material in the meshwork and IOP levels, ⁹ as well as axon loss in the optic nerve. ¹⁰ We hypothesized from these findings that those factors involved in the pathogenesis of the meshwork changes in POAG leading to the accumulation of SD-plaques may also induce changes in the optic nerve, making the nerve more susceptible to elevated IOP than normal eyes. SD-plaques are not increased in eyes with PEXG. ^{7 10} In addition, PEX material differs profoundly from SD-plaques material in its morphology and distribution within the eye. We therefore assume that factors responsible for PEX and SD-plaque formation are different. 

If the factors inducing optic nerve changes in patients with POAG differ from those in patients with secondary glaucomas such as PEXG, or are not present in the secondary glaucomas, the histologic findings in the optic nerve may also differ. To evaluate this hypothesis, we studied the optic nerves from donor eyes with POAG or PEXG. 

Methods

Donor Eyes

Eyes were obtained at autopsy through the Glaucoma Foundation (San Francisco, CA), and the Mayo Clinic Eyebank (Rochester, MN) in accordance with the provisions of the Declaration of Helsinki for research involving human tissue. Informed consent was obtained from every donor. Eyes with all stages of severity of disease were included, ranging from elevated pressure without damage, to eyes that had undergone laser trabeculoplasty or filtration surgery. Eyes were obtained from a variety of sources around the country, each donor had a different ophthalmologist, and the clinical records were of variable completeness. The clinical assessment of cupping in this collection of records ranged from brief descriptions to detailed drawings or photographs; similarly, visual field examinations were assessed with a variety of techniques. Changes in the anterior segments of these eyes have been reported. ^{8 10}  

POAG Eyes.

Thirty-six eyes from 23 donors (mean age, 75 ± 9 years; range, 58–86) were studied. Clinical details are in Table 1 . 

PEX Eyes.

Fifteen eyes from 10 donors (mean age, 88 ± 8 years; range, 71–100) were studied. Eleven eyes were under glaucoma therapy, and four eyes had PEX syndrome but no glaucoma. Clinical details are in Table 2 . 

Normal Eyes.

Thirty-seven eyes from 22 normal donors were studied. For age-matched comparison of glaucomatous eyes, a subgroup of 20 of these normal eyes from 10 donors (mean age, 76 ± 12 years; range, 50–90) were used. The remaining 17 eyes were from 12 donors younger than 50 years (range, 16–48 years) and were studied to determine changes with age. 

Histologic Analysis

Because eyes were received from various eye banks around the United States, they had been fixed in a variety of fixatives, from paraformaldehyde-glutaraldehyde mixtures to 10% formalin. Portions from the postbulbar region of the optic nerve, 1 to 2 mm behind the globe, were embedded in Epon for electron microscopy, or in paraffin for immunohistochemistry. In this region, the central retinal artery and vein are still located in the center of the nerve. Semithin cross sections were stained with toluidine blue and fuchsin. Quantitative analysis of sections was performed similar to our previous reports, ^{8 10} based on the method of Quigley et al. ¹¹ The area of the entire nerve, excluding the meninges, was measured. The area of the nerve fiber bundles and the area of connective tissue between the bundles and surrounding the central retinal vessels were measured separately. For determination of the number of axons, the cross section of the optic nerve was divided into eight sectors. In each sector, axons in five sample areas of 1000 μm² were counted. These sample areas extended from the central to the peripheral nerve at equal distances from each other. The mean of all measurements was multiplied with the nerve fiber area, to yield the total axon counts. 

The postbulbar retrolaminar portion of the optic nerve was chosen for study because of the regularity of the tissue and ease of making quantitative measurements when compared with the bowing and cupping of the laminar region. This retrolaminar portion of the optic nerve receives most of its blood supply from the pial arterioles, which penetrate the connective tissue septa and branch into capillaries; only a few capillaries originate at the central retinal artery (CRA). ¹² Evaluation of both the capillaries and arterioles was performed. Within the connective tissue septa, both the total number and the density of capillaries were determined. The total number of capillaries was determined by dividing the cross section of the nerve into quadrants. In the middle of each quadrant, three regions with an area of 0.155 mm² each were analyzed at 40× magnification. These regions were spaced at equal distances from the central to the peripheral portions of the nerve. The mean of these 12 regions was multiplied with the entire cross-sectional area of the nerve, to produce the capillary number per nerve. The potential tortuosity of capillaries did not allow the determination of their absolute number, and so a simple count of the number of visible capillary lumens (or capillary profiles) present in a cross-section was made (Fig. 1) . The measurements were performed by two of the authors (JG, AK) independently, and the average of the data was taken. The density of capillaries was defined as the total number of capillaries per square millimeter of total nerve area. 

Arterioles within the pial sheath of the nerve and the CRA were evaluated for arteriosclerosis, and a grading system was devised. In stage 1, there was no identifiable arteriosclerosis. Stage 2: The basement membrane of the endothelium was thickened and at places showed small infiltrations of light-microscopic homogeneous material. The lumen of the arteriole was not narrowed; in the CRA, in addition, the internal elastic membrane was partly fragmented, and an ingrowth of muscle cells in the intimal layer was possible. Stage 3: Arterioles showed a concentric narrowing of the lumen due to a lift-off of the endothelium by additional infiltration of amorphous material; in the CRA arteriosclerotic plaques occurred (Fig. 2) . Grading was performed independently by two pathologists (Amann K, Institute of Pathology, University of Erlangen-Nürnberg, Erlangen, Germany; and Blaser B, Institute of Pathology, City Hospital, Bamberg, Germany). Limited information on general medical diseases, such as systemic hypertension and myocardial infarction, was available, and we were unable to make an assessment of general arteriosclerosis. 

For electron microscopic study, 11 of the best-preserved POAG nerves spanning the entire range of axon loss, from 790,000 to 42,000, were selected. Similarly, eight PEXG specimens (axon range, 170,000–872,000), and six age-matched normal specimens were selected. Ultrathin sections were viewed with an electron microscope (model EM902; Zeiss, Oberkochen, Germany). 

Immunohistochemistry

Depending on the fixation protocols, in only nine eyes with POAG (mean age, 76 ± 12.8 years; range, 58–92) and five eyes with PEXG (mean age, 83 ± 7.8 years; range, 83–93) was immunostaining possible. These results were compared to five control eyes (mean age, 82 ± 6.7 years; range, 74–90). 

Immunohistochemistry was performed with affinity-purified rabbit and mouse polyclonal antibodies, according to the manufacturers’ instructions. Immunoreactivity was visualized on paraffin-embedded sections. Ten- to 20-μm cross sections through the postbulbar optic nerve were placed on poly-l-lysine-coated slides and initially incubated with dry milk solution (Blotto; Santa Cruz Biotechnology, Heidelberg, Germany) at room temperature for 30 minutes, to reduce nonspecific background staining. Incubation with the primary antibody (Table 3)was performed in a moist chamber for 12 to 36 hours at room temperature. The sections were rinsed in Tris-buffered saline (TBS) three times for 10 minutes each and then incubated for 1 hour with the appropriate secondary antibodies (Table 3) . After they were rinsed in PBS, the sections were mounted in Kaiser’s glycerin jelly (Merck KGaA; Darmstadt, Germany). The sections were viewed either with a fluorescence microscope (Aristoplan; Ernst Leitz, Wetzlar, Germany) or with a confocal laser-scanning microscope (LSM 5 Pascal; Carl Zeiss, Jena, Germany). 

Statistics

For comparison of the different groups a two-sided Student’s t-test for independent samples was performed, and P < 0.05 was considered significant. For data that did not have a normal distribution, a Wilcoxon rank-sum test was used. The generalized estimating equation (GEE) method was used to account for the use of both eyes from some donors. ¹³ In addition, in each group, the coefficient of correlation was determined, and α < 0.05 was considered significant. 

Results

Control Eyes

Age-Related Changes.

In control eyes, the cross-sectional area of the optic nerve and the thickness of the connective tissue septa surrounding the bundles of myelinated nerve fibers did not change with increasing age (Table 4) . The number of capillaries located within the connective tissue septa remained nearly constant with age (7% decrease between eyes aged <50 and >50 years, P = 0.23). Arteriosclerosis was not present in the pial vessels in eyes of donors less than 50 years of age. In eyes of donors older than 50 years, 6 of 20 cases showed some pial arterioles with mild to moderate arteriosclerotic changes (grade 2). In the CRA, 9 of 17 eyes of older donors and 3 eyes of younger donors showed grade 2 changes. Advanced arteriosclerosis (grade 3 changes) was not seen in any of the normal control eyes. 

Quantitative Changes.

In the 24 control eyes (19 donors) in which the axons were counted, the number of axons varied between 897,998 and 1,274,260, with a mean of 1,026,668 ± 96,891 (Tables 4 5) . Axon counts decreased with age (r = −0.660, P < 0.001; Fig. 3a ). There was no correlation between axon counts and total cross-sectional area of the optic nerve. The mean area was 7.5 ± 1.4 mm² (range, 4.2–10.1 mm²). 

Connective tissue septa occupied 20.0% ± 1.2% of the area of the optic nerve. The area of connective tissue surrounding the central retinal artery and vein was 0.091 ± 0.012 mm² (Tables 4 5) . No correlation was found between the amount of connective tissue within the optic nerve and age or axon counts. 

The number of capillaries within the connective tissue septa was 450 ± 78 (range, 277–564). The number of capillaries per square millimeter—the capillary density—was 61.1 ± 5.9. There was no correlation between number of capillaries or capillary density and age or axon counts. The number of pial arterioles in the pial connective tissue surrounding the optic nerve ranged from 9 to 24 (mean, 13.7 ± 3.4). 

Glaucomatous Eyes

In both POAG and PEXG eyes, loss of axons was accompanied by a decrease in cross-sectional area of the optic nerve and an increase in the relative proportion of the amount of connective tissue. When the absolute amount of connective tissue was quantitated, however, PEXG eyes had only a minimal increase in the thickness of the septa, and the proportional increase in connective tissue was due to the decreased size of the axon bundles (see quantitation below). In contrast, one half of the POAG eyes with significant axon loss had not only a proportional increase in connective tissue but also an actual increase in the amount of connective tissue. 

In both POAG and PEXG the total number of capillaries was decreased when compared to age-matched normal eyes (quantitation described later). The density of capillaries (total number of capillaries per square millimeter of total nerve area), however, differed between the diseases. In PEXG the density of capillaries remained constant. In contrast, the density was less in POAG. As in the normal eyes, in both PEXG and POAG, capillaries were seen only within the connective tissue septa, usually in the middle. In POAG with increased thickness of the septa, the “diffusion distance” between the capillary wall and nerve fiber bundles was therefore enlarged. 

Arteriosclerosis was more common and more advanced in both the CRA and the pial arterioles in POAG eyes than in age-matched control eyes. 

Quantitative Findings

Connective Tissue.

In POAG eyes, a decrease in overall nerve cross-sectional area was found with axon loss (r = 0.7144, P < 0.001; Fig. 3b ). As axons decreased, an increase in the total amount of connective tissue was found in 16 of 36 eyes, resulting in an overall 26% increase in the mean amount (1.96 ± 0.71 mm², n = 36 POAG eyes, compared with 1.55 ± 0.27 mm² in age-matched normal eyes; P = 0.04). This increase occurred in both the connective tissue in the region surrounding the central retinal vessels and also in the septa. The amounts ranged as high as 3.83 mm², compared with the highest value of 2.04 mm² in normal eyes and 2.52 mm² in PEXG (Fig. 3c) . The proportion of the nerve composition changed with increasing axon loss: the connective tissue in the septa comprised a proportionately larger fraction of the nerve (r = −0.7608, P < 0.001; Fig. 3d ).Comparison of the entire group of POAG eyes (including all stages of axon loss) with age-matched normal eyes showed a 65% increase in thickness of the connective tissue septa (33.30% ± 11.9% vs. 20.2% ± 1.5% of total nerve cross-sectional area; P = 0.0002; Tables 5, 6). 

The connective tissue surrounding the central retinal vessels also increased with axonal loss (r = 0.4117, P < 0.05; Fig. 4a ). Overall this increase was 17% when comparing all POAG eyes with age-matched normal eyes (0.106 ± 0.019 mm² vs. 0.090 ± 0.014 mm²; P = 0.02; Tables 5 6 ). 

In PEXG, a decrease in cross-sectional area of the nerve with axon loss was also found (r = 0.4727, P < 0.05; Fig. 3b ). In contrast to POAG eyes, as axon bundles decreased, there was no significant change in the absolute amount of connective tissue in PEXG. There was also no change in the connective tissue surrounding the central retinal artery. In contrast, as in POAG eyes, the connective tissue of the septa comprised a larger proportion of the overall nerve area (r = −0.7217, P < 0.001; Fig. 3d ). Comparison of the entire group of PEXG eyes (including all stages of axon loss) with the age-matched normal eyes, there was a 42% increase (28.8% ± 6.7% vs. 20.2% ± 1.5% of total nerve cross-sectional area; P = 0.0001; Tables 5 7 ). 

Vasculature.

In POAG, capillaries were still easily distinguishable within the thickened connective tissue septa by their endothelial lining (Fig. 1) . There was a decrease in both the absolute number of capillaries (r = −0.6877, P < 0.001; Fig. 3e ) and the capillary density (r = −0.578, P ≤ 0.001; Fig. 3f ) with increasing axon loss. Comparison of all POAG eyes (all stages of axon loss combined) with age-matched normal eyes showed a 30% decrease in the number of capillary profiles in the septa (304 ± 205 vs. 435 ± 80; P = 0.01). The corresponding decrease in capillary density was 23% (48.2 ± 27.1 vs. 62.4 ± 5.7 capillaries/mm²; P = 0.01; Tables 5 6 ). In keeping with this finding, nerves with more connective tissue septa had fewer capillaries (r = −0.36; P < 0.05) and a reduced capillary density (r = −0.33; P < 0.05; Fig. 3g ). 

Fewer arterioles in the pia were present in POAG when compared with age-matched normal control eyes (25% decrease: 10.2 ± 2.4 vs. 13.6 ± 3.6; P < 0.0001; Tables 5 6 ). The decrease in number of pial arterioles correlated with the decrease in density of capillaries in the septa (r = 0.35; P < 0.05). 

In POAG eyes there were more arteriosclerotic changes than in the age-matched normal eyes for both the CRA and pial vessels. The arteriosclerosis score for the CRA was 1.9 ± 0.6 vs. 1.6 ± 0.5 in age-matched control eyes (P = 0.03; scores range from 1 to 3: see the Methods section). The mean arteriolosclerosis score for the pial arterioles was 1.9 ± 0.6 vs. 1.3 ± 0.5 (P = 0.001; Tables 5 6 ). In the CRA, 5 of the POAG eyes had advanced arteriosclerosis (grade 3), and 20 eyes had mild to moderate changes (grade 2). Similarly, in the pial vessels five eyes showed advanced arteriolosclerosis (grade 3), but these were different eyes from different donors than those with advanced arteriosclerosis of the CRA. In the pial vessels, 21 eyes had mild to moderate changes (grade 2). Only four eyes had no evidence of arteriolosclerosis in any of the vessels. 

In PEXG, the number of capillaries decreased with axon loss, but in contrast to the POAG eyes, the capillary density did not change with axon loss (Figs. 3e 3f) . No correlation was found between the number of capillaries or capillary density and the area of septa (Fig. 3g) . The number of pial arterioles was comparable to that in the age-matched control (Tables 5, 7). 

In eyes with PEXG, the arteriosclerotic changes were somewhat less prominent than in eyes with POAG, but the differences were not statistically significant (Tables 5 7) . Only 1 of 15 PEXG eyes had advanced arteriolosclerosis in the pial vessels (grade 3), whereas seven had mild to moderate changes (grade 2). In the CRA the arteriosclerotic changes were somewhat more pronounced: 2 of 15 eyes had advanced changes (grade 3) and 6 had mild to moderate changes. Almost one half of the donors with PEXG had normal-appearing vessels within and surrounding the nerve. 

Immunohistochemistry and Electron Microscopy

Fixation of the material gave reliable results in POAG eyes only for type IV and VI collagen staining and in PEXG eyes only for type VI collagen. Staining for type I, II, and III collagen was not possible. 

In normal eyes, a thin layer of type IV collagen–immunoreactive (IR) material surrounded the blood vessels and the nerve fiber bundles (Fig. 4a) . This finding was the same in all areas of the nerve and in all five eyes investigated. In contrast, in all POAG eyes, the collagen IV layers were markedly thickened at the rim of the nerve fiber bundles, and stained fibers extended into the periphery of the nerve bundles, forming an irregular or spike-like pattern (Fig. 4b) . The pattern was similar in both the central and peripheral regions of the nerves. Ultrastructurally, the basement membranes surrounding the nerve fiber bundles were thin in the normal eyes, even in the advanced age groups investigated (Fig. 4c) . In most eyes with PEXG the morphology of the basement membranes resembled that in the normal control eyes. In contrast, in eyes with POAG, there was a thickening of the basement membrane and protrusions of this thickened basement-membrane–like material extended into the nerve fiber bundles (Fig. 4d) . 

In control eyes, a thin layer of type VI collagen was located between the collagen IV layer around both the blood vessels and nerve fiber bundles and the adjacent connective tissue of the septa. In addition, thin strands of collagen VI were also present within the connective tissue septa (Fig. 5a) . PEXG nerves showed a pattern and thickness of type VI collagen staining similar to that in the control eyes. In contrast, in POAG eyes collagen VI layers appeared thickened around the nerve fiber bundles, and within the septa (Fig. 5b) . 

At the electron microscopic level within the septa, thick collagen fibers, presumably type I and III collagen, were easily distinguishable from fine fibrillar material. This fine fibrillar material surrounded the elastic fibers within the septa, and connected them with the basement membrane of the astrocytes surrounding the nerve fiber bundles and with the basement membrane of the capillaries. The location of these fibrils was consistent with the type VI collagen seen on light microscopic immunohistochemical examination. 

In normal and PEXG, only a thin sheath of fine fibrillar material surrounded the elastic fibers in the septa, regardless of age or amount of axon loss (Fig. 5c) . In contrast, in POAG eyes with increased thickness of the septa, an increased amount of this fine fibrillar material was present surrounding the elastic fibers and also adjacent to basement membranes (Fig. 5d) . 

Discussion

Our quantitative data show for the first time that there are significant differences in the retrolaminar optic nerves of POAG eyes compared with those of age-matched PEXG as an example of secondary glaucomas. Both groups had an increase in the relative amount of connective tissue within the septa as axon loss increased. POAG eyes, however, also had an actual increase in the total amount of connective tissue within the nerve. The variability of the amount of connective tissue in POAG was high, and eyes appeared to fall into two groups. Rather than appearing as a continuous spectrum with most values near the mean, approximately half of the POAG eyes had a marked increase of connective tissue, whereas the other half had relatively normal levels despite similar levels of axon loss. In the eyes with marked increase of connective tissue, an increase in fine fibrillar material and a thickening of the basement membranes in the nerve were observed. The immunohistochemistry results suggest that this fine fibrillar material is type VI collagen. 

This increase in connective tissue in POAG eyes seems to be more than a simple reaction to axon loss, especially when compared with PEXG eyes with similar amounts of axon loss. A significant correlation between axon loss and age was even found in normal eyes, but without concomitant increase in connective tissue. It is possible that a growth factor or other stimulus caused this increase in connective tissue. Increased levels of TGF-β2 have been reported in the optic nerves of eyes with POAG. ¹⁴ TGF-β2 can induce increased expression of ECM components in monolayer cultures of human optic nerve astrocytes. ¹⁵ TGF-β2 levels are increased in the aqueous humor and vitreous of 50% patients with POAG. ^{16 17} In anterior chamber perfusion experiments, treatment with TGF-β2 leads to an increase in extracellular material under the inner wall of Schlemm’s canal and a simultaneous decrease in outflow facility. ¹⁸ It is therefore tempting to speculate that TGF-β2 is one possible common factor inducing trabecular meshwork and optic nerve changes in eyes with POAG. 

A decrease in the total number of capillaries was found with axon loss in both POAG and PEXG. A difference in capillary density—the number of capillaries per area of optic nerve—was present between POAG and PEXG. In PEXG, the capillary density remained constant despite axon loss. This is similar to changes in monkey eyes with laser-induced ocular hypertension. ^{19 20} Both Quigley et al. ¹⁹ and our group ²⁰ studied monkeys with experimentally increased IOP and found that the percentage of capillary area (capillary density) in the prelaminar portion of the optic nerves was the same as in control nerves. These findings support the idea that PEXG behaves as a secondary glaucoma, with pressure-induced optic nerve damage rather than damage caused by primarily ischemic or other factors. 

In contrast to PEXG, the optic nerves of POAG eyes had a decrease in capillary density with axon loss. Thus, the loss of capillary density cannot be caused solely by axon loss. The decrease in capillary density was also correlated with increased thickness of connective tissue in the septa of eyes with POAG but not PEXG. Common factors could be responsible for the increase in connective tissue and the decrease in capillary density in POAG. It is tempting to speculate that these factors also induce the changes in the trabecular meshwork in POAG eyes. 

A decrease in the number of pial arteries was found in POAG. We are not certain why. It could exist before the development of POAG and thus make the nerves more susceptible to damage. In keeping with this, atherosclerosis appeared more prominent in POAG eyes than in normal and PEXG eyes. Second, the decrease in pial arterioles could be caused by the same factors responsible for the pathogenesis of the nerve damage, and thus change in parallel with the nerve damage. Finally, the decrease in arterioles could be a secondary reaction to atrophy of the nerve. 

In summary, this study suggests that, in addition to IOP, other factors are involved in the degeneration of the optic nerve in POAG eyes. This could contribute to the development of optic neuropathy in eyes without elevated IOP. The elevated levels of TGF-β2 reported in the aqueous and vitreous of some eyes with POAG and the known effects of TGF-β2 on the trabecular meshwork and optic nerve astrocytes may play a role in this development. 

 Supported by the Deutsche Forschungsgemeinschaft Grant SFB 539 (EL-D).

 Submitted for publication March 8, 2005; revised June 20 and July 29, 2005; accepted September 20, 2005.

 Disclosure: J. Gottanka, None; A. Kuhlmann, None; M. Scholz, None; D. H. Johnson, None; E. Lütjen-Drecoll, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

Clinical Characteristics of Eyes with POAG

Donor	Age/Sex	Duration of Disease (y)	Maximum IOP	IOP under Therapy	Cup/Disc Ratio	Visual Field	Medication	Comments
88/90
R	76/M	14	?	16	0.2	Normal	T	Cataract surg. OU; lung cancer
L			?	16	0.2	Normal	T
154/90
R	65/M	8	35	22	0.4	Normal	T E D	Cataract surg.; cupping increased
L			35	22	0.9	Sup arc	T E D
155/90
R	72/F	14	?	?	?	?	T E	?
L			?	?	?	?	T E	?
156/90
R	63/F	13	34	20	0.1	Normal	T P	—
L			34	22	0.1	Normal	T P	—
159/90
L	80/M	23	25	20	0.4	Normal	P E	LTP; cataract surg.
161/90
R	89/F	10	28	15	?	?	T P	?
L			29	15	?	?	T P	?
153/91
R	69/M	2	26	?	0.2	Normal?	—	—
L			26	?	0.2	Normal?	—	—
154/91
R	81/M	5	26	14	0.8	Normal	T	Cataract surg.
L			26	16	0.2	Sup arc	T	—
156/91
R	86/F	7	32	17	0.7	Sup arc	T P E	Cataract surg. OU; Cupping increased
L			32	17	0.6	Normal	T P E
158/91
R	84/F	13	25	12	0.7	?	P E T D	LTP × 3; trabeculectomy OU; cataract surg. OU
L			22	13	0.6	?	—
163/91
R	74/F	5	30	22	0.6	Normal	P T E	LTP
165/91
R	58/M	?	?	?	?	?	T E
L			?	?	?	?	T E
166/91
R	60/M	5	?	22	0.3	Sup arc	E D	Cataract surg. OU
L			?	16	0.3	Constricted	E D
151/01
L	63/M	?	26	19			T E	Myocardial infarction
153/01
R	80/F	5	28	18	0.6	Normal	T	Cataract surg. × 2; aortic valve sclerosis; myocardial infarction
L				14	0.6	Normal
154/01
R	83/F	5	?	?	?		T P	Myocardial infarction; macular degeneration OU
L							T
156/01
R	83/M	13	?	?	?		T P	Cataract surg. OU
L							T P	Macular degeneration OU
157/01
L	78/M	?	?	?	?		T	Diabetes
159/01
R	74/M	?	?	21	Advanced			Myocardial infarction
160/01
R	85/F	1	26	20	?		T
161/01
L	79/F	4	?	?	?		T	Cataract surg.; myocardial infarction

 sup arc, superior arcuate; T, timolol; E, epinephrine; D, diamox; P, pilocarpine; LTP, laser trabeculoplasty.

Table 2.

 

Clinical Characteristics of Eyes with PEXG

Donor	Age/sex	Duration of the Disease (y)	Maximum IOP	IOP under Therapy	Cup/disc Ratio	Visual Field	Medication	Comments
89/90
L	93/F	7	30	23	0.7	Normal	T	Cupping increased
152/90
R	83/F	1	16	16	0.4	Normal?	None	PEX syndrome
153/90
R	87/F	2	20	14	0.5	Normal?	None	PEX syndrome
161/91
R	79/F	4	28	19	0.3	Normal	T
L			31	21	0.3	Normal	T
116/93
L	93/F	7	45	25	0.2	Normal	T P	Trabeculectomy; cataract surg.; LTP
149/93
R	100/M	12	29	15	0.9	Sup arc	P T E	Cataract surg. OU
L			33	15	0.9	Dbl arc	P T E		Trabeculectomy OS, LTP OU
82/95
R	86/M	10	21	17	0.3	Normal?	None	Diabetes
152/95
R	89/M	5	46	20+	0.8	Blind; central scotoma	P T	Macular degeneration OU
L			26	16	?				P T
181/95
R	98/M	7	22	13	0.2	Constricted	E
L			35	17	0.2	Dbl arc	E
81/96
R	71/F	20	22	17	0.3	Normal	None
L			45	30	1.0	Dbl arc	T P	LTP and trabeculectomy × 2; cataract

 Sup arc, superior arcuate; Dbl arc, double arcuate. See Table 1for all other abbreviations.

Figure 1.

 

Cross section through the postlaminar region of an optic nerve of an eye with POAG (58 years). In the connective tissue septa, capillaries (arrows) are recognizable by their shape and their endothelial lining. Semithin section, toluidine blue/fuchsin.

Figure 2.

 

Semithin sections through pial arterioles (left) and central retinal arteries (right), showing different stages of arteriolosclerosis and arteriosclerosis, respectively. Stage 1: (a) No arteriolosclerosis, the intimal layer (arrowheads) is not thickened. (b) In the CRA, in addition, the inner elastic membrane (arrowheads) is not split. Stage 2: (c) In the pial arterioles the basement membrane of the endothelium is thickened (arrows), and at places there are inclusions of light-microscopic homogeneous material (arrowheads). (d) In the CRA, in addition to a thickening of the basement membrane (arrows), at some places, an ingrowth of muscle cells in the intimal layer occurs (arrowheads) and the inner elastic membrane is partly fragmented. Stage 3: (e) The lumen of the arterioles (arrow) is narrowed concentrically due to a lift-off of the endothelium by deposits of amorphous material (arrowheads). (f) In the CRA in the intimal layer there are arteriosclerotic plaques (arrows).

Table 3.

 

Antibodies Used for Immunohistochemistry

	Host	Dilution	Manufacturer	Location
Primary antibody
Anti-GFAP	Rabbit	1:200	Sigma-Aldrich	Deisenhofen, Germany
Anti-collagen I	Rabbit	1:250	Rockland	Gilbertsville, PA
Anti-collagen III	Rabbit	1:300	Rockland
Anti-collagen IV	Mouse	1:250	Dako	Hamburg, Germany
Anti-collagen VI	Rabbit	1:300	Rockland
Secondary antibody
Anti-rabbit Alexa 488	Goat	1:2000	MobiTec	Göttingen, Germany
Anti-mouse Cy3	Goat	1:2000	Dianova	Hamburg, Germany

Table 4.

 

Measurements of Normal Eyes

Donor	Age (y)	Axon Counts (n)	Total Cross-Sectional Area (mm2)	Pial Arterioles (n)	Capillary Count	Capillary Density	Connective Tissue Septa (%)	Connective Tissue area (A. centralis, mm2)	Total Connective Tissue Area (mm2)	Arteriosclerosis			Comments
																					A. Centralis Retinae	Pial Arterioles
90/90 R	90	897,998	5.78	9	319	56	18.9	0.081	1.17	1	1
90/90 L			5.92	12			20.0				1
181/96 L	88	1,021,356	7.40	10	448	61	19.7	0.083	1.54	1	1
160/93 R	87	970,313	8.05	14	515	66	21.3	0.096	1.81	1	1
160/93 L		963,605	8.19	21	551	69	20.1	0.080	1.65	2	2
121/95 R	83	877,520	6.43	10	419	66	20.6		1.41		1
121/95 L			6.99	10			19.3		1.35		1
13/95 R	80	898,516	4.16	12	277	68		0.064		2	2
13/95 L			4.90	12	356	75	24.4		1.20	2	2
25/04 A*	77	920,560	8.16	15	459	59		0.090		2	2	Diabetes
25/04 B*			7.76	13	422	57				2	2
78/96 R	75	1,050,672	7.78	22	477	62	19.7	0.111	1.64	2	1
78/96 L			5.26	17	342	65				2	1
123/95 R	74	1,036,820	6.80	14	462	68	21.3	0.082	1.53	1	1
123/95 L		962,667	6.96	14	438	63				1	1
182/93	71	1,116,975	8.78	16	518	60	18.8	0.090	1.74
21/04 A	66	1,006,300	8.67	18	461	55	20.1	0.100	1.84	2	1	Alcoholism
21/04 B			7.50	13	489	67				2	1
24/04 A	57	968,073	7.68	10	426	57	18.8	0.110	1.56	1	1
24/04 B			6.36	11	329	53				1	2
23/04 A	50	1,096,980	9.72	12	559	59	20.1	0.090	2.04	1	1
38/94	48	981,541	8.36	24	517	63	19.6	0.079	1.72	1	1
20/04 A	48	928,638	7.32	11	406	57	18.3	0.083	1.42	1	1
2/94	43	1,042,965	9.64		548	58	21.1	0.099	2.13	1	1
28/04 A	42	1,139,158	7.79	17	493	65	19.6	0.090	1.62	1	1	Myocardial infarction
28/04 B			6.98	17	505	74				1	1
19/04 A	36	1,107,220	8.34	13	471	58	20.1	0.099	1.78	1	1	Asthma
19/04 B			8.04	14	490	63				1	1
22/04 A	36	1,038,465	6.14	12	307	50	18.6	0.113	1.26	—	1
90/03 A	28		8.75	10	490	56	19.3	0.098	1.79	1	1	Hypertension; diabetes; thrombosis
90/03 B		1,274,260	9.65	12	511	53				1	1
27/04 A	26	1,030,260	5.66	14	312	56	20.0	0.096	1.23	—	1	Epileptic
27/04 B			10.09	15	564	57				2	1
29/04 A	19	1,107,048	7.88	11	489	64	20.3	0.081	1.68	2	1
29/04 B			8.43	13	467	57				2	1
89/03 A	16	1,202,600	7.37	12			21.0	0.083	1.63	1	1
89/03 B			8.77	13						1	1
Mean ± SD	58 ± 23.6	1,026,688 ± 96,891	7.5 ± 1.4	13.7 ± 3.4	450 ± 78	61.1 ± 5.9	20.0 ± 1.2	0.091 ± 0.012	1.60 ± 0.25	1.4 ± 0.49	1.2 ± 0.37

 A and B were used when the position of the donor eyes was unknown.

Table 5.

 

Summary of the Quantitative Data of the Optic Nerves

Donor	Age (y)	Eyes (donors) n	Axon Counts	Total Cross-Sectional Area (mm2)	Pial Arterioles (n)	Capillary Count	Capillary Density	Connective Tissue Septa (%)	Connective Tissue Area (A. centralis, mm2)	Total Connective Tissue Area (mm2)	Arteriosclerosis
																						A. Centralis Retinae	Pial Arterioles
Normal (all eyes)	58 ± 23.6	37 (22)	1,026,688 ± 96,891	7.5 ± 1.4	13.7 ± 3.4	450 ± 78	61 ± 5.9	20.0 ± 1.2	0.091 ± 0.012	1.60 ± 0.25	1.4 ± 0.5	1.2 ± 0.4
Normal (Age-matched)	76 ± 12.3	20 (10)	984,883 ± 73,872	7.1 ± 1.4	13.6 ± 3.6	435 ± 80	62 ± 5.7	20.2 ± 1.5	0.090 ± 0.014	1.55 ± 0.27	1.5 ± 0.5	1.3 ± 0.5
POAG	75 ± 8.8	36 (23)	545,603 ± 271,651	5.7 ± 1.5	10.2 ± 2.4	304 ± 205	48 ± 27.1	33.3 ± 11.9	0.106 ± 0.019	1.95 ± 0.69	1.9 ± 0.6	1.9 ± 0.6
PEXG	88 ± 8.3	15 (10)	441,349 ± 304,596	5.5 ± 1.7	13.9 ± 3.7	322 ± 167	58 ± 22.9	28.8 ± 6.7	0.101 ± 0.022	1.65 ± 0.46	1.6 ± 0.7	1.6 ± 0.6

Figure 3.

 

(a–f) Data in each panel show correlation of axon counts with the parameter on the y-axis in optic nerves with POAG (♦) and PEXG (▴). (○) Mean of control eyes ± SD. (a) Correlation of axon counts and age in optic nerves of control eyes (r = −0.6598; P < 0.001). (b) Correlation of axon counts and cross-sectional area. In both groups the correlation is significant (r _POAG = 0.7144, P < 0.001; r _PEXG = 0.4727, P < 0.05). (c) Correlation of axon counts and the total area of connective tissue. Whereas in POAG eyes, the total area increased with axon loss, in PEXG eyes, the area did not increase (r _POAG = 0.1593, not significant; r _PEXG = 0.0245, not significant). (d) Correlation of axon counts and proportion of connective tissue septa. In both groups, the correlation is significant (r _POAG = −0.7608, P < 0.001; r _PEXG = −0.7217, P < 0.001). (e) Correlation of axon counts and number of capillaries. In both groups, the correlation is significant (r _POAG = 0.6877, P < 0.001; r _PEXG = 0.5078, not significant). (f) Correlation of axon counts and capillary density. Whereas in POAG the capillary density decreased significantly with axon loss, in PEXG it remained nearly constant (r _POAG = −0.5780, P < 0.001; r _PEXG = −0.1497, not significant). (g) Correlation of the amount of connective tissue and capillary density in optic nerves with POAG (♦) and PEXG (▴). Whereas in POAG the capillary density decreased significantly with an increase of connective tissue, no such correlation existed in optic nerves with PEXG (r _POAG = −0.3854, P < 0.05; r _PEXG = −0.1507 (not significant). (○) Mean of control eyes ± SD.

Table 6.

 

Measurements of the Eyes with POAG

Donor	Age (y)	Axon Counts	Total Cross-Sectional Area (mm2)	Pial Arterioles (n)	Capillary Count	Capillary Density	Connective Tissue Area Septa (%)	Connective Tissue Area (A. Centralis, mm2)	Total Connective Tissue Area (mm2)	Arteriosclerosis
																				A. Centralis	Pial Arterioles
151/01 L	63	1,118,630	6.95	9	534	77	21.3	0.081	1,56	2	2
88/90 L	76	938,730	9.25	9	210	23	24.6	0.102	2,34	2	2
88/90 R	76	867,665	5.80	8	340	60	27.2	0.076	2,59	2	2
161/90 L	89	852,495	5	12	480	96	23.0	0.069	1,22	2	2
165/91 R	58	790,281	7.62	13			39.6	0.113	3,13	1	1
153/91 R	69	784,000	8.25	11	748	91	29.3	0.086	2,50	1	1
159/90 L	80	775,233	5.83	10	392	67	23.3	0.081	1,44	1	1
161/01 L	79	773,146	5.94	15	531	89	23.5	0.118	1,51	3	2
161/90 R	89	768,180		12			19.9	0.095	1,34	1	1
156/90 R	63	764,085	4.91	9	602	59	18.7	0.104	1,02	2	1
156/01 R	83	758,250	5.87	11	149	25	22.9	0.089	1,43	2	1
165/91 L	58	725,200	6.95	13	290	42	19.4				2
154/90 R	65	716,869	7.84	9	778	56	32.0	0.115	2,62	1	1
153/01 L	80	698,828	5.11	13	395	77	20.4	0.091	1,13	2	3
156/90 L	63	692,310	4.89	11			18.6	0.109	1,02	2	1
166/91 L	60	677,163	8.05	10	443	55	36.5	0.126	3,06	2	3
163/91 R	74	667,038	5.31	6	151	28	29.8	0.075	1,66	1	2
154/01 L	83	637,018	5.79	10	413	71	22.3	0.133	1,42	2	1
153/91 L	69	608,300	6.89	12			34.6	0.126	2,51	2	2
154/01 R	83	579,200	5.54	11	340	61	26.8	0.134	1,62	2	2
155/90 L	72	501,394	6.57	10	343	52	31.5			/	2
154/90 L	65	451,990	6.16	9			45.8	0.130	2,95	2	1
153/01 R	80	434,160	3.81	10	381	100	27.4	0.097	1,14	2	3
157/01 L	78	414,557	4.15	9	12	3	30.2	0.107	1,36	1	2
155/90 R	72	352,843	4.41	9	218	49	46.5	0.096	2,15	1	2
156/91 R	86	342,375	5.65	11	325	58	47.1	0.132	2,79	2	2
166/91 R	60	321,480	4.94	10	44	8	36.6	0.090	1,90	1	2
154/91 R	81	317,858	5.05	10	110	22	38.3	0.117	2,05	3	2
154/91 L	81	267,879	6.05	9	363	60	61.5	0.105	3,83	2	2
159/01 R	74	265,967	4.38	12	146	33	34.6	0.145	1,66	3	2
156/91 L	86	263,288	4.2	15	71	17	37.9	0.092	1,68	3	2
167/91 L	72	220,163	4.48	10	139	31	59.3	0.101	2,76	2	2
156/01 L	83	146,304	3.77	9	20	5	22.9	0.107	1,45	2	2
160/01 R	85	71,462	3.55	4	19	5	39.7	0.112	1,54	2	3
158/91 L	84	41,693	3.44	11	87	25	58.3				2
158/91 R	84	35,658	4.1	4	112	27	55.2	0.131	2,39	2	3
Mean ± SD	75.2 ± 8.8	545,603 ± 271,651	5.7 ± 1.54	10.2 ± 2.38	304 ± 205	48.2 ± 27.1	33.3 ± 11.9	0.106 ± 0.019	1.96 ± 0.71	1.9 ± 0.61	1.9 ± 0.63

Figure 4.

 

(a, b) Cross sections of control (Co) and POAG eyes through the retrobulbar region of optic nerves immunolabeled for type IV collagen. (a) In control eyes, a thin layer of type IV collagen surrounded the blood vessels within the septa (S) and the nerve fiber (NF) bundles (age, 89 years). (b) In POAG eyes, the type IV collagen layer was thickened and formed spikelike protrusions into the periphery of the nerve fiber bundles (age, 86 years; 342,375 axons). (c, d) Electron micrographs of the connective tissue septa adjacent to the nerve fiber bundles. (c) In control eyes, the connective tissue septa were separated from the nerve fiber bundles by a thin basement membrane (arrows). The elastic fibers were surrounded by fine fibrillar material (arrowheads) (age, 89 years). (d) In eyes with POAG, basement membranes were thickened and formed protrusions into the nerve fiber bundles (arrows) (age, 72 years; 352,843 axons).

Table 7.

 

Measurements of the Eyes with PEXG

Donor	Age (y)	Axon Counts	Total Cross-Sectional Area (mm2)	Pial Arterioles (n)	Capillary Count	Capillary Density	Connective Tissue Area Septa (%)	Connective Tissue Area (A. Centralis, mm2)	Total Connective Tissue Area (mm2)	Arteriosclerosis
																				A. Centralis	Pial Arterioles
82/95 R	86	1,033,505	10.42	9	534	51	20.1	0.107	2,20	2	1
153/90 R	87	871,763	5.25	7	405	77	19.7	0.112	1,15	2	2
81/96 R	71	858,404	6.4	17	465	73	24.4	0.082	1,64	2	1
152/90 R	83	775,762	7.41	17	582	79	21.3			1	1
181/95 R	98	610,305	5.24	13	201	38	27.5	0.075	1,53	2	1
161/91 R	79	490,104	5.86	13	347	59	53.5	0.126	2,21	1	2
89/90 L	93	412,854	3.98	17	223	56	18.1	0.074	0,79	1	1
149/93 R	100	344,338	5.06	16	283	56	27.5	0.125	1,52	2	1
181/95 L	98	277,760	4.51	22	269	61	34.7	0.137	1,70	2	2
116/93 L	93	229,241	6.43	10	82	13	38.0	0.080	2,52	1	2
149/93 L	100	227,850	4.09	17	164	40	35.5	0.101	1,55	3	3
152/95 L	89	172,420	4.78	13	321	67	56.9	0.123	1,41	1	2
161/91 L	79	122,200	4.57	14	146	32	31.8	0.068	1,52	1	2
152/95 R	89	98,175	3.1	11	159	51	32.2	0.081	1,08	1	2
81/96 L	71	95,546	5.65	12	651	115	38.7	0.115	2,30	3	1
Mean ± SD	88 ± 8.3	441,348 ± 304,596	5.4 ± 1.66	13.9 ± 3.7	322 ± 167	57.9 ± 23.0	28.8 ± 6.7	0.101 ± 0.0217	1.65 ± 0.48	1.6 ± 0.71	1.6 ± 0.61

Figure 5.

 

(a, b) Cross-sections through the retrobulbar region of control (Co) and POAG optic nerves immunolabeled (a) for type IV collagen and (b) double-labeled for type IV and type VI collagen. (a) In control eyes, only a thin layer of type VI collagen surrounded the blood vessels within the septa (S) and the nerve fiber (NF) bundles. In addition, there were stained fiber bundles within the septa (age, 89 years). (b) The type VI collagen labeled fibers (green) surrounding the blood vessels within the septa and the nerve fiber bundles are always located directly adjacent to the type IV-labeled basement membranes (red). In POAG eyes with thickened septa, there appeared to be more type VI–labeled fibers within the septa (age, 86 years; 342,357 axons). (c, d) Electron micrographs of the connective tissue septa adjacent to the nerve fiber bundles. (c) In control eyes, the elastic fibers (E) are surrounded by a small layer of fine fibrils (arrows) (age, 89 years). (d) In POAG eyes with thickened septa, there was a marked thickening of the fine fibrillar sheath surrounding the elastic fibers (arrows) (age, 72 years; 352,843 axons).

The authors thank Kerstin U. Amann and Barbara Blaser for their expert help with pathologic evaluation, Anke Fischer, Katja Gedova, Elke Kretzschmar, and Hong Nguyen for excellent assistance with immunohistochemistry and electron microscopy, and Marco Gösswein for the preparation of the micrographs.