Trabecular Meshwork in Human and Cynomolgus Monkey Eye Ted S. Acorr,*t Mary Wesrcorr,* Michael 5. Posso,* and E. Michael Van Duskirk*

The glycosaminoglycans (GAGs) extractable from the trabecular meshworks (TM) of human and non- human primate eye have been analyzed by sequential enzymatic degradation and cellulose acetate elec- trophoresis. For comparison, similar extracts of the cornea, sclera, iris, and ciliary body have also been analyzed. The distribution of glycosaminoglycans in human and in cynomolgus monkey TMs are similar, although not identical. The human TM contains (HA), - 4-sulfate and/or 6- sulfate (CS), (DS), keratan sulfate (KS), (HS), and an unidentified band of Alcian Blue staining material, which is resistant to the enzymes that we used. Based upon quantitation of the Alcian Blue staining intensities of extracted GAGs, which have been corrected by a relative dye-binding factor, the GAGs of the human TM include: 29.0% HA, 14.1% CS, 21.5% DS, 20.3% KS, and 15.0% HS. The cynomolgus monkey trabecular GAGs include: 12.8% HA, 14.3% CS, 15.2% DS, 42.1% KS, and 15.6% HS. Invest Ophthalmol Vis Sci 26:1320-1329, 1985

The preponderance of evidence suggests that the We have examined the GAGs primary site of aqueous outflow resistance resides of the human and nonhuman primate trabecular within the trabecular meshwork and possibly within meshwork as the initial step in studies intended to clar- the deep portion of the corneoscleral meshwork and/ ify the role that GAGs play in the regulation of aqueous or the amorphous juxtacanalicular basement mem- outflow. In essentially all of the tissues studied to date, brane near the canal of Schlemm.1"3 Extracellular ma- the GAGs are found as components of larger proteo- trix appears to comprise a significant portion of this and aggregates; the GAGs appear tissue and probably of the aqueous outflow barrier in to function only as a part of these larger molecules.14 primates.3"13 Quantitative aqueous perfusion5-6 and The GAG distribution profile of the trabecular mesh- histochemical7"13 studies with degradative enzymes work of enucleated primate eyes was determined as a have lead to the suggestion that glycosaminoglycans prelude to studying these more complex (GAGs) may be a key physiologic component of the and to serve as reference points for studies of the reg- resistance to the outflow of aqueous humor. The GAGs ulation of the biosynthetic and degradative activities may reduce the functional diameter of the flow chan- of the human trabecular meshwork in explant organ nels through the deep corneoscleral intertrabecular culture. spaces and/or regulate flow through the juxtacanali- cular basement membrane. Recently, significant biochemical advances in the Materials and Methods techniques for characterizing GAGs, and the proteo- Materials glycans of which they are usually a component, have 14 been reported. Micro techniques have also been de- Human eye bank eyes were obtained from 6 to 72 veloped, which facilitate the analysis of GAGs from 15 20 hr after enucleation and used immediately or stored small tissues. " These advances make the analysis of frozen at -20°C. Cynomolgus monkey (Macacafas- the unlabeled GAGs from the primate trabecular 420 cicularis) and a few Rhesus monkey (Macaca mulatto) meshwork feasible. eyes were enucleated at autopsy and used immediately or stored frozen at -20°C. Hyaluronidase (Hyaluro- nate lyase, EC 4.2.2.1, from Streptomyces hyaluroly- From the Departments of Ophthalmology* and Biochemistry,! School of Medicine, Oregon Health Sciences University, Portland, ticus), Chondroitinase AC ( AC Oregon. lyase, EC 4.2.2.5, from Arthrobacler aurescens), Chon- Supported in part by grants from the USPHS (EY-03279), the droitinase ABC (Chondroitin sulfate ABC lyase, EC American Health Assistance Foundation: National Glaucoma Re- 4.2.2.4, from Proteus vulgaris), Keratanase (Keratan search Program, and Research to Prevent Blindness. sulfate endo-B-galactopyranosyl-glycanohydrolase, Submitted for publication: March 30, 1984. from Pseudomonas species), Heparitinase (heparitin Reprint requests: Ted S. Acott, PhD, Department of Ophthal- mology, LI05, 3181 SW Sam Jackson Park Road, Portland, OR sulfate lyase, EC 4.2.2.8, from Flavobacterium hepar- 97201. inum), keratan sulfate, chondroitin-4-sulfate, chon-

1320

Downloaded from iovs.arvojournals.org on 09/25/2021 No. 10 PRIMATE TRADEOJLAR MESH WORK GLYCOSAMINOGLYCANS / Acorr er ol. 1321

droitin-6-sulfate, dermatan sulfate, heparan sulfate, and 10% ethanol16; and scanned without clearing on an hyaluronic acid were purchased from Miles Labora- Helena Laboratories Rapid-Scan (Beaumont, TX) or tories, Inc; Elkhart, IN (prepared by Seikagaku Kogyo; a Beckman model CDS-200 densitometer. Areas under Tokyo, Japan); papain, cysteine hydrochloride, hyal- the curves were determined on a Zeiss Digiplan MOP- uronidase (from bovine testes), , deoxyribo- 3 image analyzer (Carl Zeiss, Inc., Oberkochen, West nuclease I (EC 3.1.21.1) and ribonuclease A (EC Germany). Uniform quantities of standards, usually 3.1.27.5) were purchased from Sigma Chemical Co (St. hyaluronic acid and chondroitin-4-sulfate, were in- Louis, MO); bovine testicular hyaluronidase was also cluded on all runs for quantitative comparisons be- purchased from Worthington (Freehold, NJ); Bio-Gel tween runs. In separate experiments, a "relative dye P-6DG (desalting gel) was purchased from Bio-Rad binding factor" for Alcian Blue staining and interaction Laboratories, Richmond, CA; Alcian Blue 8GX was was determined for each type of commercially available purchased from Aldrich Chemicals (Milwaukee, WI) GAG applied to CAE strips. Several concentrations of and n-butyl nitrite was purchased from American Sci- each GAG were scanned densitometrically and the entific (McGraw Park, IL). All other chemicals were areas were used to evaluate linearity within the ranges the best Reagent Grade available. of interest and to obtain estimates of the dye binding factors of each type of GAG relative to HA for com- parison with unknowns. Methods Identification of GAGs by sequential enzymatic digestions: Measurement of the GAG profiles of the Extraction of GAGs from tissues: The trabecular respective eye structures was accomplished by a mod- meshwork, cornea, iris, ciliary body and a 5-mm ring ified version of methods previously described.151719 of the sclera were removed from enucleated eye bank Enough sample or standard to produce good visible or cynomolgus monkey eyes and separated under a bands when stained after CAE using 0.25-/zl applica- Wild M5D (Heerbrugg, Switzerland) dissecting scope. tions, was dissolved in 30 /x\ of water and a 5-n\ aliquot The tissues were blotted and weighed. Lipids were ex- was removed for electrophoresis on CAE. The remain- tracted for 24 hr with one change, 10 volumes each, ing 25 /xl were diluted to 100 /A final volume with the of chloroform:methanol (2:1). The tissues were rinsed first enzyme buffer to give the final concentrations given in acetone, dried, weighed, and digested with papain below, the enzyme digestion was completed and the for 24 to 48 hr at 65 °C. The papain incubation buffer reaction was terminated by the addition of 3 volumes was 5 mM ethylenediaminetetraacetic acid (EDTA), 2 of cold 5 mM potassium acetate in ethanol. The GAGs mM cysteine hydrochloride, 200 mM sodium acetate were precipitated over night at -20°C, centrifuged at at pH 5.5, and contained 50 units of papain per mil- 12,000 X g for 90 min at 4°C, the supernatant was liliter and 1 ml of incubation mixture per 10 eyes. The removed and the pellet was dried and resuspended in papain and other protein remaining were precipitated 25 n\ of water. A 5-/il aliquot was removed for CAE by incubation with 10% (final volume) trichloroacetic and the remainder was diluted to 100 /nl with the next acid (TCA) for 2 hr at 4°C and were centrifuged for enzyme solution; the process was repeated removing 15 min at 4°C and 12,000 X g; the pellet was rinsed a 5-/A aliquot and reducing the sample volume until once and the pooled supernatants were applied to Bio- all the steps were completed. We eliminated the TCA gel P-6DG (desalting-gel) columns to separate TCA and precipitation step after each treatment161719 because other small compounds from the GAGs. The high mo- we found that our dermatan sulfate standards were in- lecular weight fractions were pooled, the GAGs were completely precipitated from 10% TCA solutions by precipitated overnight at -20°C with three volumes of ethanol (data not shown). absolute ethanol containing 5% potassium acetate and The sequence of treatment buffers (final concentra- centrifuged for 90 min at 4°C and 12,000 X g. This tions), enzymes, incubation conditions (including our partially purified fraction was then redissolved in water modifications of reported methods), and components for cellulose acetate electrophoresis (CAE) or sequential degraded were: enzymatic digestion. Cellulose acetate electrophoresis: The various GAGs 1. Streptomyces hyaluronidase (5 TR units) in 20 and standards were applied to cellulose acetate strips mM sodium acetate, 2 mM EDTA, 150 mM sodium (Microzone Plus, Beckman Instruments, Fullerton, chloride, pH 5.0, was incubated with the sample for 5 CA) with a 0.25-/xl applicator and electrophoresed in hr at 60°C to degrade hyaluronic acid.2122 a Beckman Microzone model R-101 electrophoresis 2. Chondroitinase AC (1 unit) in 20 mM Tris (tris cell at room temperature for 2 hr at 4 mA using 0.3 (hydroxymethyl) aminomethane), 50 mM sodium ac- M cadmium acetate buffer (pH 4.1).1618 Strips were etate, 100 mM sodium chloride, pH 8.0, was incubated stained for 15 min in 1% Alcian Blue, 5 mM sodium with the sample for 5 hr at 37°C to degrade chon- acetate in 50% ethanol; destained in 5% acetic acid in droitin-4 and 6-sulfates.2324

Downloaded from iovs.arvojournals.org on 09/25/2021 1022 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / Ocrober 1985 Vol. 26

IE Ml of IN HC1 and 2.5 n\ of 20% n-butyl nitrite in II I ethanol; this was incubated for 5 hr at room temper- ature; neutralized with 2.5 ;ul of IN NaOH; diluted to 100 fx\ final volume with water; and ethanol precipi- tated as above. This process is reported to degrade hep- arin and heparan sulfate.27'28 6. In some cases, pools of material from step 5 and pools of the original starting material were subjected to degradation by heparitinase before electrophoresis. Heparitinase (10 units) in 1 mM calcium chloride, 0.1 iT • • • i' M sodium acetate, pH 7.0, was incubated with the sample for 6 hrs at 37°C to degrade heparan sulfate.29 Testicular hyaluronidase was used to check the HA chondroitinase AC degradation for inhibition by der- matan sulfate. None was observed under the conditions detailed above. In pilot experiments with standards, the amount of each enzyme used was shown to be in excess, thereby insuring complete degradation of each GAG. The data presented in this article represent two or more separate degradation experiments for each GAG from each species and tissue and two or more CAE analyses of each fraction using a total of 150 hu- man and 180 cynomolgus monkey eyes. The standard error of the mean of the individual analysis values were never in excess of 10% of the individual values. The nondegradable material from the sequential enzymatic degradation process was also incubated with deoxyri- bonuclease I, ribonuclease A, and alpha amylase with no detectable changes (data not shown).

Results CAE Profiles of the GAGs Extracted from Human and Cynomolgus Monkey Trabecular Meshworks and Surrounding Tissues

i • Figure 1 shows the CAE migration pattern of GAGs extracted from human trabecular meshwork as com- Fig. 1. Cellulose acetate electrophoretic profiles of glycosaminog- pared to the migration of the various standards. Three lycans extractable from human trabecular meshwork compared with peaks were observed, one coincident with hyaluronic standards. Extracted GAGs from human trabecular meshwork (panel acid, one intermediate, and one broader peak covering I) were electrophoresed with standards to establish the electrophoretic the range of the other standards. These standards are migration profile. Panel 2 shows the migration of keratan sulfate and separable, but due to variations in the degree of sul- hyaluronic acid. Panel 3 shows the migration of chondroitin sulfate and dermatan sulfate. Panel 4 shows the migration of heparin sulfate. fation or other characteristics in the various tissues, The electrophoresis was run in the direction of the arrow, ie, from they could not be used for absolute identification of negative to positive. These are densitometric traces of the CAE strips the respective GAGs. after Alcian Blue staining without clearing. Figure 2 shows a comparison of similar electropho- retic profiles of the GAGs extractable from human and cynomolgus monkey trabecular meshwork, sclera, 3. Chondroitinase ABC (1 unit) in the same buffer cornea, ciliary body, and iris. Hyaluronic acid, peak I, as in step 2 (above) was incubated with the sample for is detectable in all the tissues studied except cornea; 5 hr at 37°C to degrade dermatan sulfate.23-24 band II is present in relatively high amounts in all of 4. Keratanase (1 unit) in 50 mM Tris, 50 mM so- the tissues except sclera, although it may not be the dium acetate, pH 7.4, was incubated with the sample same material in each tissue. The remainder of the for 5 hr at 37°C to degrade keratan sulfate.2526 GAGs run as a broader peak (III) and comprise a con- 5. To the remaining 5 n\ of sample was added 2.5 glomerate of several individual GAGs, which require

Downloaded from iovs.arvojournals.org on 09/25/2021 No. 10 PRIMATE TRADECULAR MESHWORK GLYCOSAMINOGLYCAN5 / Acorr er ol. 1323

HUMAN CYNOMOLGUS

T M

i • • • • i • • • • i i • • • • i

Sclera

Fig. 2. Cellulose acetate electrophoretic profiles of glycosaminoglycans from human and cynomolgus monkey eye structures. Extracted GAGs from human and cyno- molgus monkey trabecular meshwork, sclera, cornea, ciliary body, and iris were electrophoresed, stained with alcian blue, and scanned densitometrically. Electro- Cornea phoretic direction (see arrow) is from right (-) to left (+) and the scale is in centimeters from the origin. 1 ' I ' ' ' ' I

CB

I ' ' ' ' I ' • ' ' I

Iris

enzymatic treatments to identify specifically. The rel- and in the samples after the degradation of HA, CS, ative amounts of the GAGs and their electrophoretic DS, KS, and HS, respectively. The HS and remaining mobilities are distinctly different in each tissue; simi- peak were both resistant to the n-butyl nitrite treatment, larities can be seen, however, when the pattern of each although the HS was degraded by heparitinase. Hatched tissue in human is compared to that of the monkey. regions represent the amount removed by each treat- ment; the stippled area, which appears as a new peak Identification of Trabecular Meshwork GAGs Via area not present in the previous scan, actually reflects Sequential Enzymatic Degradation a change in the migration rate of the remaining GAGs. The electrophoretic profiles of the trabecular GAGs The final remaining peak is resistant to all of the treat- in aliquots removed after each step in the sequential ments. enzymatic degradation are shown in Figure 3. These The TM of both species contain HA, although the profiles show a comparison of the human and cyno- human has relatively more than the monkey. Both molgus monkey GAGs in the initial, untreated sample contain significant amounts of CS and of DS, although

Downloaded from iovs.arvojournals.org on 09/25/2021 1324 INVESTIGATIVE OPHTHALMOLOGY G VISUAL SCIENCE / October 1985 Vol. 26

HUMAN CYNOMOLGUS Identification of GAGs from Surrounding Eye Tissues by Sequential Enzymatic Degradation The four adjacent tissues were also analyzed, both to provide a comparison of the tissue GAGs and to aid in the establishment of the uniqueness of the trabecular meshwork GAGs. Figure 4 shows the results of the enzymatic degradation method as applied to extracts of the human and cynomolgus monkey sclera. This tissue is relatively low (<10%) in its total content of hyaluronic acid, heparan sulfate, and undegraded material; although the cynomolgus monkey has a sig- nificant amount of heparan sulfate, as determined by n-butyl nitrite treatment, and considerable KS. The human has much more CS and the migration is con- siderably different from the CS of the monkey, sug- gesting different degrees of sulfation. Both have an in- termediate quantity of DS. Figure 5 shows the results of enzymatic analysis of the human and cynomolgus monkey cornea. Hyal- uronic acid is not detectable on these strips, although it can be seen when they are overloaded. Both species have an intermediate amount of CS and less DS. Treatment with keratanase removes considerable KS in each species and causes a major increase in the elec- trophoretic mobility of the material remaining after treatment (stippled area). HS is not detected by n-butyl nitrite treatment and a large peak of undegraded ma- terial with increased mobility remains. Figure 6 shows the same analysis of human and cy- nomolgus monkey ciliary body. HA is detectable in small amounts; slightly more CS is present, and in the monkey its migration is increased by the treatment (stippled area). DS is the predominant GAG in both species and KS is present in small amounts in the monkey and not detectable in the human. Treatment with n-butyl nitrite did not reveal any HS in either species and a large peak of undegraded material resists all treatments. Fig. 3. Electrophoretic profiles after sequential enzymatic degra- Figure 7 shows the same analysis of the GAGs of dation of glycosaminoglycans from the trabecular meshwork. Human the human iris. Cynomolgus monkey iris was not and cynomolgus monkey trabecular meshwork GAGs were subjected to sequential enzymatic digestion, electrophoresed on cellulose acetate, available in sufficient amounts for complete rigorous stained with Alcian Blue, and scanned densitometrically. Traces are analysis, but the basic profile was similar to that for for equal aliquots of extracted GAGs before treatment, and after the human. HA is present, as are significant amounts treatment with streptomyces hyaluronidase, chondroitinase AC, of CS and DS. No KS or HS degradation is detectable chondroitinase ABC, keratanase, and heparitinase. The hatched areas represent the Alcian Blue staining material removed by each enzyme and an intermediate amount of material remains after treatment. The stippled areas, which appear as new peaks, are not all of the treatments. actually new material but are due to changes in migrational position of the remaining material resulting from the digestion. No degradation Percentage of the GAGs from Human and was observed after n-butyl nitrite treatment (data not shown). Cynomolgus Monkey Eye Tissues The hatched areas in Figures 3-7 were measured to the relative mobilities are somewhat different. Both obtain an estimate of the percentages of each type of contain KS and HS in appreciable amounts and there GAG that was degraded by each of the enzymes. These is a small amount of material that is not degraded by data are presented in Table 1. The "Total" column is any of the enzymes. Relative percentages will be pre- the sum of all the other columns and its deviation from sented later (Tables 1, 2). 100 is an indicator of the cumulative error in the mea-

Downloaded from iovs.arvojournals.org on 09/25/2021 No. 10 PRIMATE TRADECULAR MESHWORK GLYCOSAMINOGLYCANS / Acorr er ol. 1325

Table 1. Comparison of the percentages of each tissue GAG based upon direct quantitation of alcian blue staining*

Type of GAG HA CS DS KS H/HS Other Total

Human TM ISA 13.6 20.9 15.3 17.8 9.5 102.5 Monkey TM 9.9 12.2 13.1 28.1 16.4 19.8 99.5 Human sclera 5.6 58.6 19.6 14.9 ND 5.9 104.6 Monkey sclera 6.0 18.8 10.9 46.4 10.4 5.4 97.9 Human cornea ND 27.2 14.8 32.6 ND 29.3 103.9 Monkey cornea ND 17.9 12.5 29.8 ND 49.9 110.1 Human CB 10.3 15.7 26.9 ND ND 40.4 93.3 Monkey CB 7.2 2.0 22.5 7.8 ND 54.1 93.6 Human iris 13.7 31.7 28.0 ND ND 25.0 98.4

* Values are for the percentages of total alcian blue staining material removed The "Other" column is simply the amount of stain remaining after the treatment by each enzyme treatment. The last column is the sum of the previous columns sequence. ND: none detectable. The standard error of the mean of individual and its deviation from 100 reflects the relative cumulative error in these values. determinations was never in excess of 10% of the individual values.

surements. The standard error in the mean of the in- relative dye-binding factor. The "other" column is the dividual values was never over 10% of the individual amount of material, which resisted all of the degra- means. These values represent the percentages of Alcian dation treatments, and with the exception of the sclera Blue staining lost after each enzyme treatment. and cornea ran at exactly the same relative migration Since the different GAGs may be expected to bind rate. In the sclera, there was only a small peak re- different amounts of dye per dimeric unit, based upon maining and in the cornea, the remaining material mi- the number and type of charged groups present, an grates faster than it does before the keratanase treat- alternative basis for the evaluations was also used. Sev- ment. The area of this peak is included in the"Other" eral dilutions of known amounts of commercially column although it may have been partially degraded. available GAGs of each type were applied to CAE strips, stained, and scanned; the areas of these profiles Discussion were used to evaluate linearity and to obtain a "dye- The trabecular mesh works of both human and non- binding factor" for each type of GAG. These factors human primates contain significant amounts of the were then normalized relative to the factor for HA. five major GAGs found in the extracellular matrix of These relative dye-binding factors and the "corrected" other tissues, as well as an unidentified component(s) values for the percentage each type of GAG contributed that is resistant to all the degradative treatments. This to the total trabecular GAGs are shown in Table 2. "Other" component may include the nonGAG oli- The validity of each of these two estimation methods gosaccharide chains from proteoglycans and glycopro- is evaluated in the Discussion section. They are in teins and possibly even some other unidentified cellular moderate agreement and the relative dye-binding fac- or extracellular material which binds Alcian Blue; it tors do not vary greatly from 1.00. does not appear to be undegraded GAGs and we could Hyaluronic acid is present in relatively high amounts only speculate upon its nature. No strong preponder- in the human TM, in somewhat lower amounts in the ance of any GAG was observed. The resistance of tra- monkey TM, in both scleras, ciliary bodies, and the becular HS to n-butyl nitrite and its degradation by human iris. We could not detect any hyaluronic acid heparitinase has been observed by other investiga- in the cornea at these levels of sample application. Chondroitin sulfate is present in intermediate amounts in all of the tissues studied, although there is less in the Table 2. Relative dye binding factors and monkey ciliary body and considerably more in the hu- comparison of the "corrected" percentages of total man iris and sclera. The percentages of dermatan and trabecular GAGs* keratan sulfate are also intermediate, although KS is low in the iris and ciliary body and higher in the sclera, Type of GAG HA CS DS KS HS cornea and monkey TM. Factors 1.00 1.10 1.11 0.86 1.36 Heparan sulfate, as detected by chemical degrada- Human TM 29.0 14.1 21.5 20.3 15.0 tion, is in low to undetectable amounts in all the tissues Monkey TM 12.8 14.3 15.2 42.1 15.6 except the monkey sclera. Heparitinase digestion of the trabecular GAGs of both species demonstrates that * Values are for the percentage of the total GAGs that are removed by each enzymatic degradation. The relative dye binding factors for each GAG were HS is actually present at much higher levels than was determined for commercially available GAGs of each type as described in the Methods Section. The factor for HA was arbitrarily set at 1.00 and the others indicated by chemical degradation. The relatively high were normalized relative to it. The "Other" peak was not included in these charge density of this GAG is also apparent from its calculations.

Downloaded from iovs.arvojournals.org on 09/25/2021 1326 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / October 1985 Vol. 26

HUMAN CYNOMOLGUS are in good qualitative agreement with the reports of others.19'20'32 Incorporation of radiolabel into HA, CS, DS, KS, and HS by human trabecular meshwork in organ or cell culture has been observed by various in- vestigators.3'4'32"41 The quantitative values vary with the culture conditions and the assay method. As a rule, the relative production of HA is higher in dividing cell culture than the level of total HA in eye bank mesh- works. This disparity may indicate that HA synthesis occurs at a relatively higher rate in dividing cells in

HUMAN CYNOMOLGUS

Fig. 4. Electrophoretic profiles after sequential enzymatic degra- dation of glycosaminoglycans from the sclera. Human and cyno- molgus monkey scleral GAGs were subjected to sequential enzymatic digestions, and treated identically to those in Figure 3, except that n-butyl nitrite was used in the last step instead of heparitinase (see caption to Fig. 3).

tors.3031 This may be due to the degree of sulfation of this GAG. Based upon the quantitation of Alcian Blue staining with or without "correction," the human TM contains relatively high percentage of HA, DS, and HS, although Fig. 5. Electrophoretic profiles after sequential enzymatic degra- there are also significant percentages of CS, KS, and dation of glycosaminoglycans from the cornea. Human and cyno- molgus monkey corneal GAGs were subjected to sequential enzymatic undegradable material. The cynomolgus monkey TM digestions and treated identically to those in Figure 4. The stippled contains relatively less HA but more KS and unde- areas again represents a change in the migrational position rather gradable material than the human TM. These results than new material.

Downloaded from iovs.arvojournals.org on 09/25/2021 No. 10 PRIMATE TRADECULAR MESHWORK GLYCOSAMINOGLYCANS / Acorr er ol. 1327

culture than its concentration "in situ" would indi- HUMAN cate.3'38"41 The relative distributions of the GAGs in the adja- cent tissues of the anterior chamber are similar to find- ings of other investigators.20'3042 The heparitinase digestion used on the TM as a final step degraded much of the "Other" peak, which was resistant to n-butyl Untreated nitrite degradation. We assume that some of this peak in the other tissues is also sensitive to this enzyme, but have not verified this experimentally. The relatively high amounts of some of the GAGs make it difficult

HUMAN CYNOM0LGUS

HyolurooidOM

Chondroitinase AC

Chondroitinose ABC

Kerotonose

n-Butyl Nitrite

Fig. 7. Electrophoretic profiles after sequential enzymatic degra- dation of glycosaminoglycans from the iris. Human iris GAGs were subjected to sequential enzymatic digestions, and treated identically to those in Figure 4. Cynomolgus monkey iris tissue was not available in sufficient quantity to complete this treatment sequence.

to detect the presence of relatively small amounts of other GAGs without heavy overloading of the CAE strips, eg, corneal HA. Therefore, our inability to detect a GAG should not be construed as proof of its absence, but merely as evidence that it is not present in large amounts. We should point out, also that our tissue does include the trabecular wall of Schlemm's canal and this may contribute one or more of the GAGs. Fig. 6. Electrophoretic profiles after sequential enzymatic degra- dation of glycosaminoglycans from the ciliary body. Human and We should caution that the quantitations of GAGs cynomolgus monkey ciliary body GAGs were subjected to sequential by the two methods that we have used (percentage of enzymatic digestions, and treated identically to those in Figure 4. Alcian Blue staining material and "corrected" per-

Downloaded from iovs.arvojournals.org on 09/25/2021 1328 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / Ocrober 1985 Vol. 26

centages based upon "relative dye-binding factors" for least in part, of different but significant amounts of commercially available GAGs) have certain limita- each of the five common GAGS; HA, CS, DS, KS, and tions. Clearly, the extent of Alcian Blue staining will HS. One or more of these may contribute directly to vary for each type of GAG, therefore, the amount of the trabecular barrier of aqueous humor outflow. Alcian Blue staining material is not simply convertible Key words: outflow pathway glycosaminoglycans, sequential to some number of mg of GAG. It is also unlikely that enzymatic degradation, primate trabecular meshwork the sulfate and carboxyl groups of each GAG bind this dye with identical affinities. The extent of sulfation also varies with the type of GAG and for any specific type Acknowledgments of GAG with the tissue source and possibly also in The authors thank the Oregon Health Sciences University response to unknown physiologic conditions. GAGs and Lions' Eye Banks for providing human eyes and the also exhibit considerable microheterogeneity. There- Oregon Regional Primate Research Center for providing cy- nomolgus monkey eyes. fore, the comparisons based upon the "relative dye- A brief account of this work has been presented previously binding factors" of commercially available GAGs (see Acott et al4). should also be viewed with some caution. Without having standards of the specific GAGs from the same References tissue and identical physiologic conditions, neither method can be used to obtain exact molar or milligram 1. Grant WM: Facility of flow through the trabecular meshwork. values for the individual GAGs. The similarity between AMA Arch Ophthalmol 54:245, 1956. 2. McEwen WK: Application of Poiseuille's law to aqueous outflow. the results based upon the two quantitative estimation AMA Arch Ophthalmol 60:290, 1958. methods, the agreement with radiolabel incorporation 3. Rohen JW and Liitjen-Drecoll E: Biology of trabecular mesh- studies39 and the agreement of the relative dye-binding work: In Basic Aspects of Glaucoma Research, Liitjen-Drecoll factors with predictions from the structures of the E, editor. Stuttgart, FS Schattauer Verlag, 1982, pp. 141-66. GAGs from other tissues do add credibility to these 4. Acott TS, Passo MS, Westcott M, and Van Buskirk EM: Tra- becular meshwork glycosaminoglycans in excised primate eyes values. and in human organ cultures. ARVO Abstracts. Invest Ophthal- Since there is detectable Alcian Blue and ruthenium mol Vis Sci 24(Suppl):86, 1983. red staining in several regions of the aqueous outflow 5. Barany EH: In vitro studies of the resistance to flow through the angle of the anterior chamber. Acta Soc Med Uppsala 54:260, pathway, we can only speculate upon the distribution 1953. of the individual GAGs within the TM. It seems likely, 6. Barany EH and Scotchbrook S: Influence of testicular hyaluron- however, that some of these GAGs, as components of idase on the resistance to flow through the angle of the anterior proteoglycans, are intercalated between the collagen chamber. Acta Physiol Scand 30:240, 1954. molecules within the trabecular beams; some may be 7. Berggren L and Vrabec F: Demonstration of a coating substance components of the trabecular wall of the canal of in the trabecular meshwork of the eye. Am J Ophthalmol 44: 200, 1957. Schlemm; others may be located on the surfaces of the 8. Grierson I, Lee WR, and Abraham S: A light microscopic study trabecular cells extending out into the trabecular spaces of the effects of testicular hyaluronidase on the outflow system and some may be within the juxtacanalicular basement of a baboon (Papio cynocephalus). Invest Ophthalmol Vis Sci membrane-like region. 18:356, 1979. 9. Grierson I and Lee WR: Acid mucopolysaccharides in the outflow GAGs may contribute to the outflow resistance by apparatus. Exp Eye Res 21:417, 1975. reducing the functional diameters of the intertrabecular 10. Grierson I, Lee WR, and Abraham S: The appearance of the channels in the deep corneoscleral meshwork. Po- outflow apparatus of the eye after staining with ruthenium red. iseuille's law, which is thought to describe the outflow Acta Ophthalmol 55:827, 1977. resistance through the meshwork,2 contains a term for 11. Richardson TM: Distribution of glycosaminoglycans in the aqueous outflow system of the cat. Invest Ophthalmol Vis Sci the functional pore diameter raised to the fourth power. 22:319, 1982. This is conceptually appealing and recent photomicro- 12. Vrabec F: The amorphous substance in the trabecular meshwork. graphs by Richardson" are suggestive of this type of BrJ Ophthalmol 41:20, 1957. GAG distribution. The amorphous juxtacanalicular 13. Zimmerman LE: Demonstration of hyaluronidase-sensitive acid basement membrane may also contribute to the mucopolysaccharide. Am J Ophthalmol 44:1, 1957. 14. Hascall VC and Kimura JC: Proteoglycans: isolation and char- aqueous outflow resistance. If this is a true basement acterization. Methods Enzymol 82:769, 1982. membrane, heparan sulfate may be found in this re- 15. Breen M, Weinstein HG, Andersen M, and Veis A: Microanalysis gion.43 By analogy with the kidney, we might expect and characterization of acidic glycosaminoglycans in human tis- that the HS proteoglycan would serve as the primary sues. Anal Biochem 35:146, 1970. flow barrier in the TM.44'45 Further experiments, how- 16. Stack MT and Golichowski AM: Cellulose acetate electrophoresis of glycosaminoglycans: determination of concentration and spe- ever, will be necessary to test this hypothesis. cific activity by solid-phase spectrophotometry and fluorography. We can conclude then, that the extracellular matrix Electrophoresis 2:307, 1981. of the primate trabecular meshwork is composed, at 17. Breen M, Knepper PA, Weinstein HG, Blacik LJ, Lewandowski

Downloaded from iovs.arvojournals.org on 09/25/2021 No. 10 PRIMATE TRABECULAR MESHWORK GLYCOSAMINOGLYCANS / Acorr er ol. 1329

DG, and Baltrus BM: Microanalysis of glycosaminoglycans. Anal glycosaminoglycans by cell cultures of the trabecular meshwork Biochem 113:416, 1981. of the primate eye. Exp Eye Res 24:71, 1977. 18. Curwen KD and Smith SC: Quantitative microanalysis of aortic 34. Rohen JW, Lutjen-Drecoll E, and Rohrbach M: Morphological glycosaminoglycans. Anal Biochem 19:291, 1977. and biochemical studies on cell and tissue cultures of human 19. Knepper PA, Farbman AI, and Telser AG: Aqueous outflow trabecular meshwork. In Glaucoma Update II, Krieglstein GK pathway glycosaminoglycans. Exp Eye Res 32:265, 1981. and Leydhecker S, editors. Berlin, Springer-Verlag, 1983, pp. 20. Knepper PA, Collins J, Zaparackas Z, Weinstein H, and Breen 39-43. M: Human trabecular meshwork glycosaminoglycans. ARVO 35. Hassel JR, Newsome DA, and Ballintine EJ: Synthesis of extra- Abstracts. Invest Ophthalmol Vis Sci 22(Suppl):93, 1982. cellular matrix components by trabecular meshwork in organ 21. Ohya T and Kaneko Y: Novel hyaluronidase from Streptomyces. culture. ARVO Abstracts. Invest Ophthalmol Vis Sci 19(Suppl): Biochem Biophys Acta 198:607, 1969. 273, 1980. 22. Hatae Y and Makita A: Colorimetric determination of hyaluro- 36. Schachtschabel DO, Rohen JW, Wever J, and Sames K: Synthesis nate degradation by Streptomyces hyaluronidase. Anal Biochem and composition of glycosaminoglycans by cultured human tra- 64:30, 1975. becular meshwork cells. Graefes Arch Clin Exp Ophthalmol 218: 23. Yamagata T, Saito H, Habuchi O, and Suzuki S: Purification 113, 1982. and properties of bacterial chondroitinases and chondrosulfatases. 37. Grierson I, Marshall J, and Robins E: Human trabecular mesh- J BiolChem 243:1523, 1968. work in primary culture: A morphological and autoradiographic 24. Saito H, Yamagata T, and Suzuki S: Enzymatic methods for the study. Exp Eye Res 37:349, 1983. determination of small quantities of isomeric chondroitin sulfates. 38. Polansky JR, Wood IS, Maglio MT, and Alvarado JA: Trabecular J BiolChem 243:1536, 1968. meshwork cell culture in glaucoma research: evaluation of bio- 25. Nakazawa K, Suzuki N, and Suzuki S: Sequential degradation logical activity and structural properties of human trabecular of keratan sulfate by bacterial enzymes and purification of a cells in vitro. Ophthalmology 91:580, 1984. in the enzymatic system. J Biol Chem 250:905, 1975. 39. Acott TS, Westcott-Hodson M, Samples JR, and Van Buskirk 26. Nakazawa K and Suzuki S: Purification of keratan sulfate-en- EM: Human trabecular meshwork produc- dogalactosidase and its action on keratan sulfates of different tion in explant organ culture. ARVO Abstracts. Invest Ophthal- origin. J Biol Chem 250:912, 1975. mol Vis Sci 25(Suppl):100, 1984. 27. Conrad HE and Hart GW: Heparan sulfate biosynthesis by em- 40. Polansky JR, Weinreb RN, Baxter JD, and Alvarado J: Human bryonic tissues and primary fibroblast populations. Dev Biol 44: trabecular cells. I. Establishment in tissue culture and growth 253, 1975. characteristics. Invest Ophthalmol Vis Sci 18:1043, 1979. 28. Cifonelli J A and King J: The distribution of 2-acetamido-2-deoxy- 41. Crean EV, Casey R, and Richardson TM: Glycosaminoglycan D-glucose residues in mammalian . Carbohydr Res 21: synthesis by calf trabecular meshwork cell cultures. ARVO Ab- 173, 1972. stracts. Invest Ophthalmol Vis Sci 25(Suppl):317, 1984. 29. Oosta GM, Favreau LV, Beeler DL, and Rosenberg RD: Puri- 42. Hassell JR, Newsome DA, and Hascall VC: Characterization fication and properties of human platelet heparitinase. J Biol Chem 257:11249, 1982. and biosynthesis of proteogl yeans of corneal stroma from rhesus 30. Funderburgh JL, Stenzel-Johnson PR, and Chandler JW: Corneal monkey. J Biol Chem 254:12346, 1979. glycosaminoglycan synthesis in long-term organ culture. Invest 43. Martinez-Hernandez A and Amenta PS: The basement mem- Ophthalmol Vis Sci 24:208, 1983. brane in pathology. Lab Invest 48:656, 1983. 31. Kraemer PM: Heparan sulfates of cultured cells. I. Membrane- 44. Kanwar YS, Linker A, and Farquhar MG: Increased permeability associated and cell-sap species in Chinese hamster cells. Bio- of the glomerular basement membrane to ferritin after removal chemistry 10:1437, 1971. of glycosaminoglycans (heparan sulfate) by enzyme digestion. J 32. Knepper PA, Breen M, Weinstein HG, and Blacik LJ: Intraocular Cell Biol 86:688, 1980. pressure and glycosaminoglycan distribution in the rabbit eye: 45. Kanwar YS, Hascall VC, and Farquhar MG: Partial character- effete of age and dexamethasone. Exp Eye Res 27:567, 1978. ization of newly synthesized proteoglycans isolated from the glo- 33. Schachtschabel DO, Bigalke B, and Rohen JW: Production of merular basement membrane. J Cell Biol 90:527, 1981.

Downloaded from iovs.arvojournals.org on 09/25/2021