Molecular Vision 2003; 9:355-359 <>
Received 10 June 2003 | Accepted 19 August 2003 | Published 20 August 2003

Transforming growth factor beta induced protein accumulation in granular corneal dystrophy type III (Reis-Bücklers dystrophy). Identification by mass spectrometry in 15 year old two-dimensional protein gels

Chris J. Hedegaard,1 Ida B. Thøgersen,1 Jan J. Enghild,1 Gordon K. Klintworth,2 Torben Møller-Pedersen3

1Cornea Proteomics Center, Department of Molecular Biology, University of Aarhus, Aarhus, Denmark; 2Departments of Pathology and Ophthalmology, Duke University Medical Center, Durham, North Carolina; 3Department of Ophthalmology, Aarhus University Hospital, Aarhus, Denmark

Correspondence to: Jan J. Enghild, Ph.D., Cornea Proteomics Center, Department of Molecular Biology, University of Aarhus, Gustav Wieds Vej 10C, DK-8000 Aarhus C, Denmark; email:


Purpose: To investigate the accumulation of TGFBIp in GCDIII and to demonstrate the ability to perform mass spectrometry on old two-dimensional protein gels.

Methods: Proteins were extracted from one cornea with GCDIII from a person with an Arg124Leu mutation in the TGFBI (BIGH3) gene and from one normal human cornea in 1987 and subjected to two-dimensional (2D) gel electrophoresis. After keeping the gels at room temperature for 15 years, protein spots of interest were excised and digested with trypsin. The tryptic-derived peptides were analyzed using mass spectrometry.

Results: Four areas of interest were examined and four different proteins (TGFBIp, aldehyde dehydrogenase class 3, actin, and albumin) were identified in the 15-year-old gels. Using image analysis, the amount of TGFBIp was found to be 77 fold higher in the GCDIII affected cornea relative to the normal tissue. In both situations, TGFBIp migrated on the 2D gels as a 63 kDa protein. Mass spectrometry revealed the same nine peptides in TGFBIp from both the normal and the GCDIII affected corneas, including one peptide situated at the amino terminus. Moreover, the cornea with GCDIII contained abundant 40 kDa TGFBIp fragments that were lacking sequences in both the amino and carboxy termini.

Conclusions: Mass spectrometry can be performed on old 2D polyacrylamide gels. In both normal and GCDIII affected corneas, the majority of TGFBIp migrated on 2D gels as a 63 kDa protein with an intact amino terminus. However, the amount of the 63 kDa TGFBIp was 77 fold higher in the GCDIII affected cornea. Furthermore, the GCDIII affected cornea contained abundant 40 kDa fragments that were trunctated in both the amino and carboxy termini.


Loss of corneal transparency is related to several factors including a disturbance in the spatial arrangement of collagen fibers [1-3], increased cellular reflectivity of stromal keratocytes [4-6], and abnormal deposition of extracellular matrix proteins, as in certain corneal dystrophies. A group of the latter conditions (including various types of lattice and granular corneal dystrophy and Thiel-Benhke dystrophy) result from mutations in the TGFBI (BIGH3) gene [7,8]. This gene codes for the extracellular transforming growth factor-beta induced protein (TGFBIp), which has a theoretical mass of 72 kDa. The carboxy terminus of TGFBIp contains an Arg-Gly-Asp (RGD) sequence that acts as an important recognition site for many integrins providing a system for cell adhesion [9].

Prior to the mapping of granular corneal dystrophy (GCD) to human chromosome 5 (5q31) [10,11] and the identification of mutations in the TGFBI (BIGH3) gene [7], we had hoped to identify the protein that was known to accumulate within the corneal stroma in GCD [12]. By doing so, we planned to identify the responsible gene by first obtaining a limited peptide sequence of the isolated protein and then deducing the corresponding DNA sequence. Allowing for the degeneracy of the genetic code, this approach would have enabled oligonucleotides to be synthesized and used as probes to screen genomic libraries to isolate and identify the responsible gene. As part of this study, 2D gel electrophoresis was performed on extracts of corneal proteins with several corneal dystrophies and from corneas free from disease. One cornea evaluated in this manner was from an individual with GCD type III (GCDIII, Reis-Bücklers dystrophy). In the present study, we reexamined the gel in the light of available evidence indicating that the protein accumulations involve the mutant protein product of the TGFBI gene [13,14].

Several mutations in codon 124 of the TGFBI gene (Arg124Cys, Arg124His, Arg124Ser, and Arg124Leu) are known to cause distinct corneal dystrophies [8]. In some of these conditions, a protein that presumably represents the mutant TGFBIp has been identified in the abnormal accumulations immunohistochemically [13]. The nature of these deposits, however, remains to be characterized biochemically to understand the pathobiology of these inherited disorders, and to formulate rational strategies for their treatment and the prevention of the corneal deposits.

In the present study, we analyzed TGFBIp from a 15-year-old polyacrylamide gel containing proteins from an extract of a cornea with GCDIII using the direct method of mass spectrometry. We show that TGFBIp migrates as a 63 kDa protein and has an intact amino terminus in both normal and GCDIII affected corneas. However, the level of 63 kDa TGFBIp is 77 fold higher in the affected cornea. Additionally, the GCDIII affected cornea contained abundant 40 kDa fragments of TGFBIp that were lacking parts of both the amino and carboxy termini.


Patients and tissue handling

Proteins were extracted from one half of a corneal button (7 mm in diameter) removed from a 29-year-old man who had undergone a penetrating keratoplasty for recurrent GCDIII in his left eye in 1986. This individual was a member of a family with GCDIII and was first examined at 8 years of age during a genetic evaluation of the family when his mother and maternal aunt were found to have a bilateral corneal dystrophy. At that time he was found to have a milder version of the same condition and a visual acuity of 20/50 (right eye) and 20/25 (left eye). The patient died at 35 years of age from a heart attack. An autopsy was not performed. Different ophthalmologists examined the eyes of many family members and four individuals had a gray-white corneal opacification associated with frequent episodes of ocular pain, lacrimation, photophobia, and conjunctival hyperemia starting at 6 months to 2 years of age. The disorder was rapidly progressive and was accompanied by poor vision early in life. Corneal sensation was decreased in affected persons, but normal in other family members who were examined. Slit-lamp biomicroscopy of an affected 3-year-old child revealed that the earliest corneal opacification seemed to involve primarily the subepithelial tissue and Bowman's layer in the peripheral cornea. Older affected individuals contained similar opacities as well as numerous fine gray-white dots within the central stroma and sometimes in the posterior corneal layers. They also had evidence of recurrent epithelial erosions.

Light and transmission electron microscopy

Following penetrating keratoplasty, corneal tissue was fixed in 10% neutral buffered formaldehyde at room temperature and processed for light microscopy. Part of the affected cornea was placed in glutaraldehyde and processed for transmission electron microscopy. Paraplast sections of the corneal tissue were stained with hematoxylin and eosin, Hale's colloidal iron stain, Wilder's stain, and Masson trichrome.


DNA from three affected members of the family (mother, maternal aunt, and brother) and one unaffected brother was isolated from blood using the Dneasy tissue kit (Qiagen, Valencia, CA). Exon 4 and exon 12 of the TGFBI gene were amplified by PCR using the primers described by Munier, et al. [7]. The PCR products were separated on an agarose gel and then extracted with a Qiagen gel extraction kit (Qiagen). Nucleotide sequencing was performed in a forward and reverse direction using an ABI 377XL DNA Sequencer and dRhodamine/BigDye terminator cycle sequencing reagents (PE Applied Biosystems, Foster City, CA).

2D gel electrophoresis

In 1987, 2D polyacrylamid gel electrophoresis was performed on extracted proteins from samples of a normal and a GCDIII affected cornea by Protein Databases Inc. (PDI, Huntington Station, NY) [7,15-17]. The proteins were subjected to isoelectric focusing (pH 3.5 to 10) in the first dimension and separated according to mass (12.5% w/v acrylamide gel, 300x260x1 mm) in the second dimension. The gels were stained with silver [18] and scanned, and protein spots were quantified using densitometry (PDQUEST, PDI). The gels were dried and kept at room temperature for 15 years.

Mass spectrometry

In 2002, two areas of the gels were chosen for analysis by mass spectrometry, because of an abundance of protein in those regions (Figure 1, areas 1 and 2). Three additional areas (Figure 1, areas 3 to 5) were selected to identify markers appropriate for estimating the molecular weight and density of the specific spots of interest. Small gel plugs within the chosen areas of the dried 2D gels, containing the regions of interest, were excised, washed three times in 0.5 ml sterile H2O, and incubated twice in 50% acetonitrile (HPLC-purified) for 15 min and once in 100% acetonitrile for 15 min. After incubation for 5 min in 0.1 M NH4HCO3, an equal amount of 100% acetonitrile was added. After 15 min, the gel plugs were dried in a vacuum desiccator and incubated for 16 h at 37 °C in 50 mM NH4HCO3 containing 0.3 ng/μl trypsin (HPLC purified, Promega). The samples were applied to the mass spectrometer target using a ZipTip (P10, Millipore, Glostrup, Denmark) containing C18 column material (equilibrated in 100% acetonitrile and washed in 0.1% trifluoroacetic acid before sample handling). The samples were eluted from the ZipTip onto the target using α-cyano-4-hydroxy-cinnamic-acid (Sigma, Copenhagen, Denmark) and analyzed using mass spectrometry (MALDI-TOF, Perceptive Biosystems Voyager DE-PRO, Applied Biosystem). Mass spectra were calibrated using trypsin auto-digestion ion peaks as internal standards. The Internet based search program ProFound was used for protein identification by peptide mapping.


Microscopy and genotyping

The deposits within all corneal specimens examined from different affected members of the family had the typical light and electron microscopic features of GCD (images not shown). The deposits were located predominantly in the subepithelial region. The three affected individuals in the family all had the Arg124Leu mutation in the TGFBI gene.

2D gel electrophoresis and mass spectrometry

Despite showing some variation in the isoelectric focusing of the proteins (Figure 1), the 2D gels were of sufficient quality for protein extraction and analysis. The silver stained 2D gel from the individual with GCDIII had 197 identifiable spots (Figure 1A) as compared to 140 spots on the control gel (Figure 1B). We only examined areas with obvious accumulated material (and adjacent areas as markers for molecular weight and densitometry). We identified four proteins in five selected areas (Figure 1). In both the normal cornea and the cornea with GCDIII, TGFBIp was identified in area 1 as a 63 kDa protein based on the detection by mass spectrometry of nine peptides, spanning amino acid residues from 28 to 588 (Figure 2). This identification was based on several gel plugs taken across the encircled area (Figure 1, area 1). Using densitometry, the amount of the 63 kDa TGFBIp was found to be 77 fold higher in the GCDIII affected cornea relative to the normal tissue (compare the encircled area 1 in Figure 1A to Figure 1B). We were unable to identify the outermost amino terminal peptide (amino acids residues 24-27) because this peptide was too small for identification by the mass spectrometer [19]. Moreover, the peptide containing the Arg124/Leu124 residue was too large to be extracted from the polyacrylamide gel [20]. In two instances, the same peptide was represented by two peaks (Figure 2A, peak B and G) due to methionine oxidation. It is noteworthy that the same nine peptides were identified in both the normal and the GCDIII affected cornea including one peptide situated close to the amino terminus (amino acids 28-38, Figure 2B, peptide D). The GCDIII affected cornea contained abundant 40 kDa TGFBIp fragments (Figure 1A, area 2) that were lacking peptide D close to the amino terminal end and peptide F close to the carboxy terminus (Figure 2B). In both gels, aldehyde dehydrogenase class 3 (50 kDa) was found in area 3 (Figure 1) with a comparable intensity by densitometry. During the characterization of the 40 kDa TGFBIp fragments, actin (42 kDa, Figure 1A, area 4) was recognized. In the normal cornea, albumin (66 kDa; Figure 1B, area 5) was identified, whereas the high amount of TGFBIp obscured the detection of albumin in the GCDIII affected cornea (Figure 1A).


Several techniques have been employed to characterize the nature and biochemical composition of the corneal accumulations in the dystrophies caused by mutations in the TGFBI gene. In tissue sections, special stains, histochemical procedures, and immunohistochemical methods have been used. These methods have led to the characterization of certain corneal deposits as amyloid or to another non-specific protein. Antibody-based techniques, such as immuno-histochemistry and western blotting have been employed to characterize TGFBIp isoforms in both normal and diseased corneas [13,14,21-28]. Although antibody-based techniques are superior to the usual histochemical methods, their indirect nature has inherent weaknesses. The antibodies may not be specific and cross-reactions with proteins having similar epitopes may, for instance, give rise to false positive reactions [29]. Direct methods including amino terminal protein sequencing [14,30] and mass spectrometry are therefore preferable because they provide information that is more precise on the amino acid sequence of the protein in question [29].

Although the theoretical molecular weight of TGFBIp is 72 kDa, all previous studies have observed a lower molecular weight [14,30,31]. Also in the current study of normal and GCDIII affected corneas, TGFBIp migrated as a 63 kDa protein (Figure 1) and not as a 72 kDa protein. This unexpected difference in size of TGFBIp remains to be fully elucidated but is clearly not an artifact from analyzing the protein. It can be explained by the observation that extracellular TGFBIp lacks the encoded carboxy terminal [32].

It is noteworthy that the same nine peptides were identified in both the normal and the GCDIII cornea including one peptide (peptide D in Figure 2) situated close to the amino terminus. The GCDIII cornea contained abundant 40 kDa TGFBIp fragments that were lacking peptide D at the amino terminus and peptide F at the carboxy terminus. We speculate that the R124L mutation may affect the resistance of TGFBIp to degradation by proteolysis leading to an accumulation of 40 kDa fragments in addition to the 63 kDa fragment.

Our identification within the gels of three proteins of known molecular weight, aldehyde dehydrogenase class 3 (50 kDa), albumin (66 kDa) and actin (42 kDa), enabled us to provide a good prediction of the actual size of the TGFBIp and its fragments within the GCDIII cornea. Our finding of a 77 fold higher amount of the 63 kDa TGFBIp in an extract of the GCDIII cornea relative to the normal tissue is considerably higher than the 2.5 fold excess of TGFBIp reported by Korvatska, et al. based on western blot analysis of 1D gels [14]. This discrepancy probably reflects different amounts of mutated TGFBIp present in the corneal samples of the affected individuals in addition to differences in the applied methodology.

A noteworthy finding is the demonstration that it is possible to obtain valuable data from 15-year old 2D gels by excising and analyzing the protein-spots in question using mass spectrometry. The significance of this is that old (one or two-dimensional) polyacrylamide gels saved for many years holding valuable information can be analyzed with the advanced methodology of today. The data obtained from the 15-year old 2D gels show that the accumulated corneal protein in GCDIII is almost exclusively amino terminal intact TGFBIp. However, it probably lacks a short segment of the carboxy terminal, since TGFBIp migrates as a 63 kDa protein, and not as the expected 72 kDa protein. The same situation seems likely for other types of GCDs, as illustrated by the apparent molecular weight of TGFBIp in a one dimensional SDS gel [30].


Supported in part by Research Grant R01 EY 12712 from the National Eye Institute. DNA sequencing was performed by the Duke University Comprehensive Cancer DNA Analysis Facility.


1. Maurice DM. The structure and transparency of the cornea. J Physiol 1957; 136:263-86.

2. Benedek GB. Theory of transparency of the eye. Applied Optics 1971; 10:459-73.

3. Vaezy S, Clark JI. A quantitative analysis of transparency in the human sclera and cornea using Fourier methods. J Microsc 1991; 163 (Pt 1):85-94.

4. Møller-Pedersen T, Vogel M, Li HF, Petroll WM, Cavanagh HD, Jester JV. Quantification of stromal thinning, epithelial thickness, and corneal haze after photorefractive keratectomy using in vivo confocal microscopy. Ophthalmology 1997; 104:360-8.

5. Møller-Pedersen T, Li HF, Petroll WM, Cavanagh HD, Jester JV. Confocal microscopic characterization of wound repair after photorefractive keratectomy. Invest Ophthalmol Vis Sci 1998; 39:487-501.

6. Jester JV, Møller-Pedersen T, Huang J, Sax CM, Kays WT, Cavanagh HD, Petroll WM, Piatigorsky J. The cellular basis of corneal transparency: evidence for 'corneal crystallins'. J Cell Sci 1999; 112 (Pt 5):613-22.

7. Munier FL, Korvatska E, Djemai A, Le Paslier D, Zografos L, Pescia G, Schorderet DF. Kerato-epithelin mutations in four 5q31-linked corneal dystrophies. Nat Genet 1997; 15:247-51.

8. Klintworth GK. The molecular genetics of the corneal dystrophies--current status. Front Biosci 2003; 8:d687-713.

9. Ruoslahti E. RGD and other recognition sequences for integrins. Annu Rev Cell Dev Biol 1996; 12:697-715.

10. Eiberg H, MøllerHU, Berendt I, Mohr J. Assignment of granular corneal dystrophy Groenouw type I (CDGG1) to chromosome 5q. Eur J Hum Genet 1994; 2:132-8.

11. Stone EM, Mathers WD, Rosenwasser GO, Holland EJ, Folberg R, Krachmer JH, Nichols BE, Gorevic PD, Taylor CM, Streb LM, et al. Three autosomal dominant corneal dystrophies map to chromosome 5q. Nat Genet 1994; 6:47-51.

12. Garner A. Histochemistry of corneal granular dystrophy. Br J Ophthalmol 1969; 53:799-807.

13. Streeten BW, Qi Y, Klintworth GK, Eagle RC Jr, Strauss JA, Bennett K. Immunolocalization of beta ig-h3 protein in 5q31-linked corneal dystrophies and normal corneas. Arch Ophthalmol 1999; 117:67-75.

14. Korvatska E, Henry H, Mashima Y, Yamada M, Bachmann C, Munier FL, Schorderet DF. Amyloid and non-amyloid forms of 5q31-linked corneal dystrophy resulting from kerato-epithelin mutations at Arg-124 are associated with abnormal turnover of the protein. J Biol Chem 2000; 275:11465-9.

15. Garrels JI, Franza BR Jr. The REF52 protein database. Methods of database construction and analysis using the QUEST system and characterizations of protein patterns from proliferating and quiescent REF52 cells. J Biol Chem 1989; 264:5283-98.

16. Garrels JI. The QUEST system for quantitative analysis of two-dimensional gels. J Biol Chem 1989; 264:5269-82.

17. Garrels JI, Franza BR Jr. Transformation-sensitive and growth-related changes of protein synthesis in REF52 cells. A two-dimensional gel analysis of SV40-, adenovirus-, and Kirsten murine sarcoma virus-transformed rat cells using the REF52 protein database. J Biol Chem 1989; 264:5299-312.

18. Switzer RC 3rd, Merril CR, Shifrin S. A highly sensitive silver stain for detecting proteins and peptides in polyacrylamide gels. Anal Biochem 1979; 98:231-7.

19. Mann M, Hendrickson RC, Pandey A. Analysis of proteins and proteomes by mass spectrometry. Annu Rev Biochem 2001; 70:437-73.

20. Speicher KD, Kolbas O, Harper S, Speicher DW. Systematic Analysis of Peptide Recoveries from In-Gel Digestions for Protein Identifications in Proteome Studies. J Biomol Tech 2000; 11:74-86.

21. Akhtar S, Bron AJ, Meek KM, Bennett K. Congenital hereditary endothelial dystrophy and band keratopathy in an infant with corpus callosum agenesis. Cornea 2001; 20:547-52.

22. Kawasaki S, Nishida K, Quantock AJ, Dota A, Bennett K, Kinoshita S. Amyloid and Pro501 Thr-mutated (beta)ig-h3 gene product colocalize in lattice corneal dystrophy type IIIA. Am J Ophthalmol 1999; 127:456-8.

23. Konishi M, Yamada M, Nakamura Y, Mashima Y. Immunohistology of kerato-epithelin in corneal stromal dystrophies associated with R124 mutations of the BIGH3 gene. Curr Eye Res 2000; 21:891-6.

24. Korvatska E, Munier FL, Chaubert P, Wang MX, Mashima Y, Yamada M, Uffer S, Zografos L, Schorderet DF. On the role of kerato-epithelin in the pathogenesis of 5q31-linked corneal dystrophies. Invest Ophthalmol Vis Sci 1999; 40:2213-9.

25. Lee EH, Seomun Y, Hwang KH, Kim JE, Kim IS, Kim JH, Joo CK. Overexpression of the transforming growth factor-beta-inducible gene betaig-h3 in anterior polar cataracts. Invest Ophthalmol Vis Sci 2000; 41:1840-5.

26. Schmitt-Bernard CF, Chavanieu A, Derancourt J, Arnaud B, Demaille JG, Calas B, Argiles A. In vitro creation of amyloid fibrils from native and Arg124Cys mutated betaIGH3((110-131)) peptides, and its relevance for lattice corneal amyloid dystrophy type I. Biochem Biophys Res Commun 2000; 273:649-53.

27. Takacs L, Boross P, Tozser J, Modis L Jr, Toth G, Berta A. Transforming growth factor-beta induced protein, betaIG-H3, is present in degraded form and altered localization in lattice corneal dystrophy type I. Exp Eye Res 1998; 66:739-45.

28. Takacs L, Csutak A, Balazs E, Berta A. Immunohistochemical detection of betaIG-H3 in scarring human corneas. Graefes Arch Clin Exp Ophthalmol 1999; 237:529-34.

29. Rappsilber J, Mann M. What does it mean to identify a protein in proteomics? Trends Biochem Sci 2002; 27:74-8.

30. Klintworth GK, Valnickova Z, Enghild JJ. Accumulation of beta ig-h3 gene product in corneas with granular dystrophy. Am J Pathol 1998; 152:743-8.

31. Skonier J, Neubauer M, Madisen L, Bennett K, Plowman GD, Purchio AF. cDNA cloning and sequence analysis of beta ig-h3, a novel gene induced in a human adenocarcinoma cell line after treatment with transforming growth factor-beta. DNA Cell Biol 1992; 11:511-22.

32. Klintworth GK, Hedegaard C, Valnickova Z, Thogersen IB, Enghild JJ. Purification and characterization of transforming growth factor beta induced protein from porcine cornea. ARVO Annual Meeting; 2002 May 5-10; Fort Lauderdale, FL.

Hedegaard, Mol Vis 2003; 9:355-359 <>
©2003 Molecular Vision <>
ISSN 1090-0535