Proc. Natl. Acad. Sci. USA Vol. 77, No. 6, pp. 3705-3709, June 1980 Medical Sciences

Macular corneal dystrophy: Failure to synthesize a mature keratan sulfate (human genetic disease/ processing/blindness//corneal stroma) JOHN R. HASSELL*, DAVID A. NEWSOME*, JAY H. KRACHMERt, AND MERLYN M. RODRIGUESt *Section on Retinal and Ocular Connective Tissue Diseases, Clinical Branch, National Eye Institute, National Institutes of Health, Bethesda, Maryland 20205; tDepartment of Ophthalmology, University of Iowa, Iowa City, Iowa 52242; and tSection on Ocular Pathology, Clinical Branch, National Eye Institute, National Institutes of Health, Bethesda, Maryland 20205 Communicated by Elizabeth F. Neufeld, March 20,1980

ABSTRACT Corneal specimens obtained during surgery dystrophy is first detected before age 10 yr. Visual acuity pro- from patients with macular corneal dystrophy and obtained at gressively deteriorates until corneal transplantation is required, autopsy from control eyes were incubated in a medium con- usually by age 25-30 yr. The small amount of tissue produced taining radioactive precursors of and proteogly- cans. Biosyntheticallyradiolabeled material was extracted and by transplantation is insufficient for direct chemical analysis. characterized by using molecular sieve chromatography and Furthermore, stromacytes derived from normal corneas and specific enzymes. Cells in control corneas synthesized both a grown in cell culture do not synthesize the sulfate proteoglycan and a keratan sulfate proteo- characteristic of the stroma (10). Therefore, the organ culture similar to those present in monkey and bovine corneas. system appeared to be a promising technique for studying the Cells in macular corneas synthesized a normal chondroitin proteoglycans of macular corneas obtained from keratoplasty. sulfate proteoglycan but did not synthesize either keratan sul- fate or a mature keratan sulfate proteoglycan. Instead, macular The data presented here suggest that the cloudiness in corneas corneas synthesized a glycoprotein with unusually large oligo- from patients with macular corneal dystrophy results from the saccharide side chains. This glycoprotein was not detected in failure to synthesize a mature keratan sulfate proteoglycan. normal corneas and is slightly smaller than normal keratan sulfate prot can. The failure to synthesize a mature keratan MATERIALS AND METHODS sulfate proteoglycan may produce corneal opacity and result in blindness. Because of evidence indicating that the corneal Organ Culture Procedures. Corneas excised with a 7- to keratan sulfate proteoglycan is normally synthesized through 8-mm trephine blade were obtained from human eyes at ker- a glycoprotein intermediate [Hart, G. W. & Lennarz, W. (1978) atoplasty or from control donor eyes at autopsy. Corneas ob- J. Biol. Chem. 253,5795-5801, macular corneal dystrophy may tained at surgery included three from two patients with macular be a defect in glycoprotein processing. dystrophy (here termed "macular corneas"), one from a patient Macular, lattice, and granular corneal dystrophies are charac- with lattice dystrophy, two from two patients with granular terized by specific abnormalities of the corneal stroma and dystrophy, one from a patient with keratoconus, and two from result in loss of vision (1). Macular corneal dystrophy is char- normal donor eyes (here termed "control corneas"). Specimens acterized clinically by recessive inheritance, a general diffuse were placed immediately in Eagle's minimal essential medium haze throughout the corneal stroma, and the accumulation of supplemented with 5% fetal calf serum, 2 mM glutamine, and irregularly shaped deposits which tend to be anterior in the 100 units of penicillin and 50 ,ug of streptomycin per ml. The central region of the stroma. Histologically, the deposits are medium also contained 200-400 ,Ci of [6-3H]glucosamine per primarily extracellular and stain positively with Alcian blue and ml and 1.0-1.5 mCi of Na235SO4 per ml (New England Nuclear; the periodic acid-Schiff reagent (2, 3). These observations 1 Ci = 3.7 X 1010 becquerels). The corneas were incubated at suggest that the deposits may be and oli- 370C for 18-20 hr in 5% C02/95% air at 100% humidity. At gosaccharides accumulating in the stromal extracellular ma- the end of the incubation period the corneas were rinsed briefly trix. in phosphate-buffered saline and stored frozen at -700C. The extracellular matrix of the normal corneal stroma consists Biochemical Procedures. Each cornea was individually primarily of type I collagen and two proteoglycans: a chon- thawed, the epithelium and endothelium were removed rapidly droitin- proteoglycan and a keratan sulfate by scraping, and the stroma was placed in 1.0 ml of extraction proteoglycan (4-9). Because most of the sulfated carbohydrate solvent containing 4 M guanidine-HCL 0.1 M 6-aminohexanoic in the cornea is contained in these two proteoglycans, macular acid, 0.01 M Na2 EDTA, 5 mM benzamidine hydrochloride, corneal dystrophy may involve alterations in one or both of and 0.5 M sodium acetate (pH 5.8-6.0) for 24 hr at 4°C (11). these proteoglycans. Recently, an organ culture system utilizing The extracting solvent was removed, and the stroma was monkey corneas has been developed that allows the biosynthesis reextracted in 0.5 ml of the same solvent for 4-6 hr at 40C. The of radiolabeled corneal stromal proteoglycans in vItro (9). This extracts were combined, and unincorporated radioactivity was system involves incubating intact corneas in a tissue culture removed by chromatography on a Sephadex G-25 PD-10 col- medium containing radioactive precursors. Analyses of the umn (Pharmacia) that had been equilibrated with and eluted radiolabeled proteoglycans synthesized during organ culture with 4 M guanidine-HCI/0.02 M Tris-HCI, pH 7.0. The residual indicated that they were representative of the proteoglycans tissue was digested with 0.5 mg papain (Sigma) in 1.0 ml of 1 that accumulate in the stromal extracellular matrix. M sodium acetate, pH 6.5/5 mM cysteine/5 mM EDTA at The corneal cloudiness in individuals with macular corneal 55-i60C for 6 hr. Any unincorporated radioactivity in the di- gest was removed by chromatography on PD-10 columns as The publication costs of this article were defrayed in part by page described above. charge payment. This article must therefore be hereby marked "ad- The in the 4 M guanidine.HCI extract were verttsement" in accordance with 18 U. S. C. §1734 solely to indicate fractionated on a Sepharose CL-4B column (200 X 1.6 cm; 4.2 this fact. ml collected per tube), equilibrated, and eluted with 4 M gua- 3705 Downloaded by guest on September 25, 2021 3706 Medical Sciences: Hassell et al. Proc. Natl. Acad. Sci. USA 77 (1980) nidine-HCI/0.02 M Tris, pH 7.0. The radioactively labeled peak fractions were pooled, dialyzed against distilled water, and 250C lyophilized. These peak fractions were then dissolved in 4 M guanidine-HCI/0.02 M Tris, pH 7.0, and divided into two parts. 200C 400 One part was directly applied to a Sepharose CL-6B column (90 X 1.5 cm, 2.0 ml collected per tube), equilibrated, and 150C 300 eluted with 4 M guanidine.HCl/0.02 M Tris.HCl, pH 7.0. The 200 second part was digested with papain and fractionated on 6-00 Sepharose CL-6B as described to isolate the glycosaminoglycans 50C and . The tubes containing the peak fractions 1- F. UE were pooled, dialyzed against distilled water, and lyophi- 8000 0 lized. :r: 10,000 1250 >6 The glycosaminoglycans and glycopeptides isolated from Sepharose CL-6B were sequentially digested with chondroiti- 8000 1000 nase ABC (Miles), keratanase (keratan sulfate-f3-endogalac- tosidase, a gift from Sakaru Suzuki), and nitrous acid as de- 6000 750 scribed (9, 12). In brief, the samples were digested with chon- 4000 500 droitinase ABC and chromatographed on Sephadex G-50 (0.6 X 30cm, 1.0 ml collected per tube). The amount digested was 2000 250 determined by the shift in elution position from a higher to a lower molecular weight. The undigested higher molecular 10 20 30 40 50 60 70 80 weight material was then digested with the keratanase and the 9V amount digested was determined again. Undigested samples Vo eluted from the same column served as "blanks." Samples also Tube number were digested for 18 hr at 370C with 2 units of mixed exogly- FIG. 1. Sepharose CL-4B chromatography of 4 M guanidine.HCl cosidases per ml (Miles) in 0.4 M sodium acetate buffer, pH extracts of (A) control and (B) macular corneal stromas. Corneas were 4.4/0.1 M NaCI. radiolabeled in organ culture with [3H]glucosamine and Na235SO4 Radioactivities in fractions were measured by liquid scin- before extraction. Brackets denote pooled tubes. tillation spectrophotometry with 10 ml of ACS (Amersham). Aliquots (0.01-0.2 ml) from each tube were assayed for radio- of this material. Similar results were obtained by chromatog- activity. An equivalent amount of 70% ethanol was added when raphy of intact (Fig. 2C) and degraded (Fig. 2D) peak 2 ma- measuring samples in 4 M guanidine.HG. An external standard terial from macular corneas on Sepharose CL-6B. was used to correct for spillover of M5S into the 3H channel. Peak 3 material, isolated by Sepharose CL-4B chromatog- raphy, was chromatographed on Sepharose CL-6B. Intact peak RESULTS 3 material from control corneas (Fig. 3A) eluted as a single peak There were two major differences between labeled macro- at tube 30. After papain digestion, two radiolabeled peaks were molecules obtained from control and macular corneas incu- observed (Fig. 3B): a peak at tube 41 bated with [3H]glucoaaniine and [35Sisulfate. First, substantially (designated peak 3b) with a preceding shoulder (designated less of the total incorporated 3H and [35S]sulfate in macular peak 3a) containing both 3H and 35S and another peak at tube corneas was extracted by 4 M guanidine.HCl. Extracts of con- 50 containing only 3H. By comparing its elution position to trol corneas contained 81-88% (8148% for 3H and 8588% for positions of chains of known molecular MS) of the total incorporated radioactivity, whereas extracts of weight, the molecular weight of the major glycosaminoglycan macular corneas contained only 55-60% (55-57% for 3H and peak was estimated to be 15,000 (13). Intact peak 3 material 59-60% for35S) of the total incorporated radioactivity. from macular corneas was eluted as a single peak at tube 34 Fractionation of the extract on Sepharose CL-4B revealed (Fig. 3C), which indicated a slightly lower molecular weight the second major difference between control and macular than that of intact peak 3 material from control corneas (Fig. corneas (Fig. 1). The extract of control corneas was fractionated 3A). Papain digestion prior to chromatography produced three into two peaks containing 3H and two peaks (labeled 2 and 3) radiolabeled peaks (Fig. 3D) in elution positions similar to those containing both 3H and 35S (Fig. 1A). The extract of the ma- obtained by digestion of control peak 3 material. There were, cular corneas also eluted as four peaks of radioactivity in ap- however, some major differences between the oligosaccharides proximately the same positions as the controls. However, peak of control and macular peak 3 materials. First, although the 3 from macular corneas contained low levels of 35S and eluted major 3H-labeled peak (peak 3b) also contained a5S in the later than peak 3 from control corneas (Fig. 1B). control (Fig. 3B), it contained only the 3H in the macular ma- Peak 2 material isolated by elution from the Sepharose CL-4B terial (Fig. 3D). Furthermore, peak 3b from macular corneas column was chromatographed on Sepharose CL-6B (Fig. 2). eluted at tube 44, which indicates that the oligosaccharides in Intact peak 2 material from control corneas (Fig. 2A) eluted peak 3b of macular corneas are lower in molecular weight (t as a single peak at tube 28. After papain digestion, peak 2 ma- 10,000) than the glycosaminoglycan in peak 3b of control cor- terial produced two radiolabeled peaks on CL-6B (Fig. 2B): a neas. The peak 3a material in both control and macular corneas major glycosaminoglycan peak at tube 32 (designated peak 2a) are glycosoaminoglycans derived from peak 2 proteoglycans containing both 3H and 35S and a minor peak at tube 53 con- that contaminate peak 3. taining only 3H. By comparing its elution position to positions The oligosaccharides in peaks 2a and 3b were subjected to of chondroitin sulfate chains of known molecular weight, the treatments that degrade certain types of glycosaminoglycans molecular weight of the glycosaminoglycan peak was estimated (Table 1). Peak 2a material from control corneas was almost to be 55,000 (13). The shift in the elution position of the 35S in entirely digested by chondroitinase ABC, indicating that peak the intact proteoglycan (Fig. 2A) to a lower molecular weight 2 contained the chondroitin sulfate proteoglycan. Similar results after papain digestion (Fig. 2B) is evidence that protein is part were obtained by digestion of peak 2a glycosaminoglycans of Downloaded by guest on September 25, 2021 Medical Sciences: Hassell et al. Proc. Natl. Acad. Sci. USA 77 (1980) 3707

IL,

- c6 vmAe cE vb

11

10 20U 30 40 50

VO VT Tube number FIG. 3. Sepharose CL-6B chromatography of peak 3 material in Fig. 1. (A) Intact peak 3 material from control corneas; (B) papain- treated peak 3 material from control corneas; (C) intact peak 3 ma- terial from macular corneas; (D) papain-treated peak 3 material from macular corneas. Brackets denote pooled tubes. f it appears that the oligosaccharide in peak 3b from macular VO VT corneas is derived from a glycoprotein. Tube number The material not extracted from the stroma by 4 M guani- FIG. 2. Sepharose CL-6B chromatography ofpeak 2 material in dine-HCI (residue) was digested with papain and fractionated Fig. 1. (A) Intact peak 2 material from control corieas; (B) papain- on Sepharose CL-6B (Fig. 5). Four peaks of radiolabeled ma- treated peak 2 material from control corneas; (C) intact peak 2 ma- terial were obtained from the residue of control corneas, with terial from macular corneas; (D) papain-treated peak 2 material from the majority of the 3H eluted as a low molecular weight peak macular corneas. Brackets denote pooled tubes. at tube 50 (Fig. 5A). Peaks R-b and R-c correspond in elution position to the chondroitin sulfate glycosaminoglycan side chain macular corneas. The majority of the peak Sb material from and the keratan sulfate side chain, respectively (see Fig. 2, Fig. control corneas was digested by keratanase, indicating that peak 3, and Table 1). Only three peaks of radiolabeled material were 3 material contained the keratan sulfate proteoglycan. How- obtained from the residue of macular corneas, with the majority ever, peak 3 material from macular corneas was resistant to keratanase as well as chondroitinase and nitrous acid. Kerata- Table 1. Sensitivity of 3H-labeled peak 2a and 3b nase degrades an unsulfated keratan sulfate (Sakaru Suzuki, personal communication). oligosaccharides to digestion Peak 2a Peak 3b Peak 3b materials from control and macular corneas were Control, also digested with a mixture of Macular, Control, Macular, exoglycosidases, and the digests Treatment N % % % were chromatographed on Sephadex G-50 (Fig. 4). The results showed that peak Sb material from macular corneas was sen- Chondroitinase ABC 96 94 12 8 sitive to this mixture of enzymes (Fig. 4B) but peak Sb material Keratan sulfate- from control corneas was resistant (Fig. 4A). Because this fl-endogalactosidase 1 1 73 5 mixture contains enzymes that attack specific glycosidic link- Nitrous acid - 1 ages found only in the oligosaccharide portion of glycoproteins, Resistant 3 5 15 86 Downloaded by guest on September 25, 2021 3708 Medical Sciences: Hassell et al. Proc. Natl. Acad. Sci. USA 77 (1980)

R-b B M 10,000o 1000 C6 In' 7500 750

5000 R-a 500

2500X 250

10 20 30 40 50 ft VO VT Tube number FIG. 5. Sepharose CL-6B chromatography of stromal residue (4 M guanidine-HCl-inextractable material) after papain digestion. (A) Residue from control corneas; (B) residue from macular corneas. 10 20 30 1 Brackets denote pooled tubes. VO VT Tube number droitin sulfate proteoglycan contains one or two chondroitin FIG. 4. Sephadex G-50 chromatography of peak 3a material in sulfate side chains with a molecular weight of --55,000 attached Fig. 3. Samples were incubated with a "mixed" glycosidase prepara- to a protein core to produce a macromolecule with a molecular tion (- - -) or without enzyme (-) before chromatography. (A) Peak weight of t150,000. The keratan sulfate proteoglycan consists 3a material from control corneas; (B) peak 3a material from macular of 2 to 3 keratan sulfate side chains with a molecular weight of corneas. -10,000 attached to a protein core, forming an intact proteo- glycan with a molecular weight of -75,000. The results of the of the 3H eluted in peak R-b (Fig. 5B). Peak R-c was absent. present study indicate that stromacytes in normal human cor- Greater than 95% of the 3H in peak R-b of both control and neas synthesize proteoglycans in organ culture similar to those macular samples was sensitive to chondroitinase ABC. These previously found in monkey and bovine corneas. observations indicate that the proteoglycans not extracted by Corneas from patients with macular corneal dystrophy 4 M guanidine.HCl in both the control and the macular samples synthesize an apparently normal chondroitin sulfate proteo- contain glycosaminoglycans that resemble the glycosamino- glycan, although it is not as extractable as the chondroitin sulfate of the proteoglycans extracted from those tissues. proteoglycan from control corneas. Macular corneas, however, Furthermore, the reduced extractability of incorporated ra- do not synthesize a normal keratan sulfate proteoglycan or diolabel in macular corneas noted previously can be attributed keratan sulfate. Instead, they synthesize a glycoprotein with to radioactivity in chondroitin sulfate remaining in the residue oligosaccharide side chains that are larger than other oligo- and not keratan sulfate. Thus, our analyses confirmed previous saccharides but smaller than normal keratan sulfate side chains. work that also could not detect labeled keratan sulfate in ma- This oligosaccharide is not digested by keratanase, chondroi- cular corneas radiolabeled in an organ culture system different tinase ABC, or nitrous acid but is degraded by a mixture of from that used in this study (14). exoglycosidases. The glycoprotein is not detected in normal The proteoglycan data for granular, lattice, and keratoconus corneas and is slightly smaller than the keratan sulfate proteo- corneas were indistinguishable from that found for control glycan synthesized by normal corneas. We suggest that this corneas (not shown). The lattice, granular, and keratoconus unusual glycoprotein synthesized by stromacytes in macular corneas are more appropriate controls than normal corneas corneas may be an incompletely or abnormally processed because the patients were subjected to the same preoperative precursor to the mature keratan sulfate proteoglycan. procedures as the patients with macular dystrophy. Moreover, Corneal keratan sulfate is chemically different from carti- it indicates that not all corneal dystrophies that involve stromal laginous keratan sulfate (15-18). Keratan sulfate is a repeating extracellular matrix abnormalities also involve changes in sulfated of glucosamine- in both tissues. stromal proteoglycans. The disaccharide-protein linkage in corneal keratan sulfate is, however, different from the linkage in cartilaginous keratan DISCUSSION sulfate. Corneal keratan sulfate contains , a sugar found Previous work with monkey and bovine corneas indicates that in the oligosaccharide portion of glycoproteins, and is linked the corneal stroma contains two proteoglycans: a chondroitin to the protein core by an N-glycosidic linkage at an sulfate and a keratan sulfate proteoglycan (5, 8, 9). The chon- residue. Cartilaginous keratan sulfate, on the other hand, is Downloaded by guest on September 25, 2021 Medical Sciences: Hassell et al. Proc. Natl. Acad. Sci. USA 77 (1980) 3709

O-linked and recent evidence suggests that mannose may be either each other or the collagen fibrils present in the stroma absent (19). Corneal chondroitin sulfate, like cartilaginous to produce the crystalline lattice necessary for the optical clarity chondroitin sulfate, contains no mannose and is linked to the of the cornea. Macular corneal dystrophy, then, provides the protein core by an 0-glycosidic linkage at a serine. first direct evidence that the corneal keratan sulfate proteo- Corneal keratan sulfate proteoglycan is probably synthesized glycan plays an important functional role in the transparency from a glycoprotein intermediate. The oligosaccharide portions of this tissue. are also N-linked of many mannose-containing glycoproteins The authors wish to thank Dr. Randal Olson for providing one of to the protein core at an asparagine residue (see ref. 20 for a the macular corneas used in this study and Dr. Vincent Hascall for his review). These particular oligosaccharides are derived from the many helpful suggestions. dolichol-oligosaccharide synthetic pathway. Inhibition of dolichol-mediated glycoprotein synthesis also blocks the syn- 1. Francois, J. (1966) Trans. Ophthalmol. Soc. 86,367-416. thesis of keratan sulfate but not the synthesis of chondroitin 2. Klintworth, G. L. & Vogel, F. S. (1964) Am. J. Pathol. 45, sulfate in chick corneas in organ culture (21). This strongly 565-576. suggests that the synthesis of keratan sulfate chains in the ker- 3. Garner, A. (1969) Invest. Ophthalmol. 8,475-483. D. & Gardell, S. atan is dependent upon or involves the 4. Antonopoulous, C. A., Axelsson, I., Heinegird, sulfate proteoglycan (1974) Biochim. Blophys. Acta 338, 108-119. dolichol pathway. 5. Axelsson, I. & Heinegird, D. (1975) Biochem. J. 145, 491- Studies on dolichol-mediated glycoprotein synthesis indicate 500. that the oligosaccharide part of glycoproteins are, in some cases, 6. Muthiah, P., Stuhlsatz, H. W. & Greiling, H. (1974) Hoppe- considerably larger in their immature form than that found in Seyler's Z. Physiol. Chem. 355, 924-934. the finished glycoprotein (22-25). The oligosaccharide portion 7. Speziale, P., Speziale, M., Galligani, L. & Balduini, C. (1978) is then reduced in size by removal of some sugars by a proce- Biochem. J. 173,935-939. dure termed "processing." This is followed by the addition of 8. Axelsson, I. & Heinegird, D. (1978) Biochem. J. 169, 517- many sugars, in some cases more sugars than originally present 530. in the immature glycoprotein, onto the stub to 9. Hassell, J. R., Newsome, D. A. & Hascall, V. C. (1979) J. Biol. produce the finished glycoprotein. In the synthesis of the cor- Chem. 254, 12346-12354. 10. G. K. & Smith, C. F. (1976) Lab. Invest. 35, 258- a inter- Klintworth, neal keratan sulfate proteoglycan from glycoprotein 263. mediate, "processing" would presumably expose the recogni- 11. Oegema, T. R., Hascall, V. C. & Dziewiatkowski, Z. (1975)J. Biol. tion site, which would only then allow the addition of keratan Chem. 250, 6151-6159. sulfate to the oligosaccharide stub and would produce a mature 12. Hart, G. W. (1976) J. Biol. Chem. 251, 6513-6521. corneal keratan sulfate proteoglycan. It is possible that macular 13. Wasteson, A. (1971) J. Chromatogr. 5,87-97. corneal dystrophy represents a failure to process the oligosac- 14. Klintworth, G. K. & Smith, C. F. (1977) Am. J. Pathol. 89, charide portion of the glycoprotein intermediate into keratan 167-181. sulfate. Thus, the glycoprotein detected in macular corneas 15. Roden, L. & Schwartz, N. B. (1975) in Biochemistry of Carbo- could be the unprocessed, immature glycoprotein precursor to hydrates, ed. Whelan, W. J. (Univ. Park Press, Baltimore, MD), the keratan sulfate proteoglycan Alternatively, the glycoprotein pp. 95-152. 16. Mathews, M. B. & Cifonelli, J. A. (1965) J. Biol. Chem. 240, precursor may have been only partially processed and, perhaps, 4140-4145. as a result, sugars were added to the oligosaccharide stub instead 17. Baker, J., Cifonelli, J. & Roden, L. (1975) Connect. Tissue Res. of keratan sulfate to produce a mature but unique glycoprotein 3, 149-156. instead of a keratan sulfate proteoglycan. 18. Bray, B. A., Lieberman, R. & Meyer, K. (1967) J. Biol. Chem. Corneal macular dystrophy is distinctly different from the 242,3373-3380. diseases termed "mucopolysaccharidoses" (26-28). For ex- 19. De Luca, S., Lohmander, S., Nillson, B., Hascall, V. & Caplan, ample, light and electron microscopic studies have showed that, A. (1980) J. Biol. Chem., in press. in addition to the accumulations of deposits in the stromal ex- 20. Lennarz, W. (1976) Annu. Rev. Biochem. 45,95-112. tracellular matrix, the endoplasmic reticulum of the corneal 21. Hart, G. W. & Lennarz, W. (1978) J. Biol. Chem. 253, 5795- (28, 5801. stromacytes also contain accumulations of similar material 22. Robbins, P. W., Hubbard, S. C., Turco, S. J. & Wirth, D. F. (1977) 29). Since the smooth endoplasmic reticulum is the site for Cell 12,893-900. glycoprotein processing, these morphological observations 23. Kornfeld, S., Li, E. & Tabas, I. (1978) J. Biol. Chem. 253, support our hypothesis that corneal macular dystrophy is caused 7771-7778. by a failure to process a glycoprotein precursor into a mature 24. Tabas, I., Schesinger, S. & Kornfeld, S. (1978) J. Biol. Chem. 253, proteoglycan. In contrast, the mucopolysaccharidoses are 716-722. caused by the failure to degrade the glycosaminoglycan portion 25. Chapman, A., Trowbridge, I. S., Hyman, R. & Kornfeld, S. (1979) of certain proteoglycans, which results in accumulations of these Cell 17,509-515. substances in lysosomal bodies. Furthermore, abnormalities in 26. Goldberg, M. F., Maumenee, E. & McKusick, V. A. (1965) Arch. patients with macular corneal dystrophy are limited to the Ophthalmol. 74,516-520. 27. McKusick, V. A., Neufeld, E. F. & Kelly, T. E. (1978) in The cornea. This would suggest that the keratan sulfate proteoglycan Metabolic Basis of Inherited Disease, eds. Stanbury, J. B., of the cornea may be the only proteoglycan synthesized through Wyngaarden, J. B. & Fredreckson, D. S. (McGraw-Hill, New a processed glycoprotein intermediate. Patients with muco- York), pp. 1282-1307. polysaccharidoses on the other hand, exhibit many medical 28. Snip, R. C., Kenyon, K. R. & Green, W. R. (1973) Invest. problems due to the larger number of tissues affected. Ophthalmol. 12, 88-97. The corneal proteoglycans probably interact specifically with 29. Teng, C. C. (1960) Am. J. Ophthalmol. 62,436-454. Downloaded by guest on September 25, 2021