EXPERIMENTAL
AND
MOLECULAR
PATHOLOGY
55,
190-19.5 (191)
Elastase Digestion of Normal and Pseudoxanthoma Lesional Skin Elastins ELAINE
SCHWARTZ,
MARK
THIEBERG, FREDERICK MARK LEBWOHL
Elasticum
A. CRUICKSHANK,
AND
Deparrment of Dermatology, Mount Sinai School of Medicine, New York, New York 10029 Received February 5, 1991; and in revised form April 11, 1991 Pseudoxanthoma elasticum (PXE) is a heritable disorder of connective tissue that is characterized by redundant folds of skin in flexural areas. There is considerable evidence that suggests that the elastic fiber is the main site of the abnormality although the primary molecular defect has not been identified. The aim of this study was to identify differences between PXE and normal skin elastins. Elastins from normal, nonsolar-exposed skin, and pseudoxanthoma elasticum lesional skin were purified and their solubilization by pancreatic elastase was compared. Results demonstrated that elastin derived from normal skin was more susceptible to proteolytic cleavage than elastin purified from either pseudoxanthoma elasticum lesional skin or ligamentum nuchae. Pretreatment of the lesional elastin with testicular hyaluronidase increased its solubilization two-fold and generated a unique 15,000 Da molecular weight fragment. Elastin prepared from PXE skin may contain bound glycosaminoglycans which interfere with elastase activity. The susceptibility of normal skin elastin to proteolytic degradation may have implications in the study of aging skin. 0 1991 Academic
Press, Inc.
INTRODUCTION Pseudoxanthoma elasticum (PXE) is a heritable disorder of connective tissue whose lesions involve the skin, eyes, and cardiovascular system (McKusick, 1972). Skin lesions are characterized by redundant folds in flexural areas and by the presence of yellow xanthoma-like papules and plaques. Vascular changes include calcification of arteries and the sequelae include hypertension, coronary artery occlusion, recurrent gastrointestinal bleeding, and cerebral hemorrhage. The eye changes are described as angioid streaks in the ocular fundus. The primary etiology of PXE has been a matter of controversy for many years. Histological studies revealed that PXE skin had increased quantities of glycosaminoglycans and calcified elastic fibers (Smith et al., 1962; Martinez-Hernandez and Huffer, 1974; Danielsen, 1982; Walker et at., 1989) whereas biochemical studies demonstrated increased amounts ofhexosamine, calcium, and elastin in PXE skin (Smith et al., 1962; Pasquali-Ronchette et al., 1981). The major glycosaminoglycan is known to be hyaluronic acid (Smith et al., 1964) although increased amounts of chondroitin sulfate are also present (Longas et al., 1986). The earliest detectable PXE abnormality is believed to be the accumulation of polyanions in the PXE dermis. Polyanions were demonstrated within mineralized elastic fibers in clinically affected dermis and within nonmineralized elastic fibers in unaffected dermis (Martinez-Hernandez and Huffer, 1974). Koreen et al. (1987) found increased amounts of hyaluronic acid and dermatan sulfate in lesional and nonlesional skins of five patients with PXE. These polyanions are believed to serve as initiating factors for the mineralization. The elastic fibers in PXE skin are increased in quantity and are abnormal in appearance (clumped and fragmented). However, the amino acid composition and 190 0014-4800/91 $3.00 Copyright Q 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
ELASTASE
DIGESTION
OF
SKIN
ELASTIN
191
desmosine content of PXE skin elastin are similar to that of normal skin elastin (Smith et al., 1%3; Schwartz et al., 1990). It was our aim to identify differences between PXE and normal skin elastins that might explain the altered appearance of the elastic fibers in PXE. This paper describes studies on the digestion by pancreatic elastase of elastins prepared from normal and PXE skins. MATERIALS
AND METHODS
Preparation of e&tin. Nonsolar-exposed skins (n = 4) were obtained from post-surgical specimens (reduction mammoplasty) of normal individuals (34-39 years of age). Lesional axillary skins from unequivocal cases of PXE were obtained from postsurgical specimens. The PXE patients ranged in age from 38 to 74 and all had extensive axillary involvement and angioid streaks. Normal-appearing skin was obtained from one PXE patient. Skin was homogenized using a Virtis in phosphate-buffered saline (pH 7.4) containing proteolytic inhibitors, defatted, and extracted overnight with a solution containing 1% sodium dodecyl sulfate (SDS), 0.05 M Tris-HCl (pH 7.5), 0.33 M mercaptoethanol, and proteolytic inhibitors. The suspension was then centrifuged and the pellet was washed extensively with water (at least 10 times) prior to lyophilization. The residue was then treated with cyanogen bromide (Schwartz et al., 1990). The insoluble fraction (elastin) was washed with water and lyophilized. The purity of the elastin was confirmed by amino acid analysis. The elastins prepared from PXE and normal skins had amino acid compositions typical of elastin (Smith et al., 1963). Elustuse digestion. Elastin samples (5 mg) were preincubated overnight (37°C with mixing) in 3 ml of 0.2 M NH,HCO, (pH 8) centrifuged, and resuspended in fresh buffer containing porcine pancreatic elastase at an enzyme:substrate ratio of 1:250. Samples were incubated at 37°C with mixing. Aliquots were removed at defined time points and phenylmethylsulfonyl fluoride was added to stop the incubation. The aliquots were centrifuged and the amount of solubilized protein was quantified (Bradford, 1976). In some experiments, testicular hyaluronidase (2 mg/ml) was included in the preincubation buffer. SDS polyucrylumide gel electrophoresis. Samples were electrophoresed on polyacrylamide gels as described by Laemmli (1970). Precast (l&20%) gels and molecular weight standards were purchased from Integrated Separation Systems (ISS) (Hyde Park, MA). Equivalent amounts of protein (50 kg) were applied to the lanes. Gels were stained with the ISS Pro-Blue Staining System. Binding of elustuse to elustin. The binding of pancreatic elastase to normal human skin and bovine ligamentum nuchae elastins was determined as described by Kagan et al. (1972). Insoluble elastins (10 mg) from normal skin and from bovine ligamentum nuchae were incubated with 20 kg elastase for 2 min in 2 ml of 0.05 M ammonium acetate buffer (pH 8.45). Following centrifugation the pellets were resuspended in fresh buffer and incubated at 37°C. Ten milligrams of elastin was added to the supernatants and these suspensions were incubated at 37°C. The rates of solubilization of the elastins reflect the proportion of elastase present in the pellets and supernatants.
RESULTS Digestion
The normal skin elastin was soluelastase to a greater extent than was PXE skin elastin. Time is shown in Fig. 1. Results (Table I) expressed as total p,g
of PXE and normal skin elustins.
bilized,by pancreatic course of elastolysis
192
SCHWARTZ
0
2
ET AL.
4
6
8
10
Time (hours) FIG. 1. Time course of elastolysis of elastin by porcine pancreatic elastase. Values (mean of duplicate determinations) represent the amount of solubilized elastin (as determined by Bradford protein assay) present in the supematants following incubation with porcine pancreatic elastase. (Cl) PXE; (+) bovine ligamentum nuchae; (m) normal shin.
protein solubilized at 4 hr (means + SEM) were 98 + 10 for PXE skin elastin (n = 3) and 336 + 9 for normal skin elastin (n = 4). The PXE elastin therefore exhibited significantly decreased susceptibility to elastase. Interestingly, the vaiues (means of duplicate determinations) for bovine ligamentum nuchae and human aortic elastins (obtained from Elastin Products Co., Pacific, MO) were 145 and 141 p,g protein solubilized, respectively. Furthermore, elastin prepared from nonlesional skin obtained from a PXE patient also exhibited decreased elastolysis (125 pg protein solubilized in 4 hr). The assay was linear with respect to enzyme and elastin concentrations. These results demonstrated that normal skin elastin was more susceptible to proteolytic cleavage than either bovine ligamentum nuchae or human aortic elastin. Furthermore, elastin prepared from PXE skin was more resistant to elastolysis than was normal skin elastin. Effect of preincubation with hyaluronidase. When PXE elastin was preincubated with testicular hyaluronidase, the amount of elastin solubilized by elastase increased two-fold (2.03 ? 0.16, mean + SEM, n = 3). The amount of normal skin elastin that was solubilized only slightly increased following hyaluronidase preincubation (1.11 + 0.02, mean f SEM, n = 3). Solubilization Source
PXE lesional
TABLE I of Dermal Elastin by Pancreatic Elastase
of elastin
Total pg solubilized in 4 hr (1)
(2) (3) Normal
112 78 103 mean (?SEM) = 98 (210)
(1)
311
(2)
349 345 340
(3) (4)
mean (*SEM)
= 336 (29)
ELASTASE
DIGESTION
OF
SKIN
ELASTIN
193
The peptides solubilized from PXE and normal skin elastins by elastase (with and without hyaluronidase preincubation) were resolved by SDS-PAGE. Results (Fig. 2) demonstrated a distinct peptide (molecular weight of 15,000 Da) in the PXE with hyaluronidase sample (lane C) that was absent in the normal skin samples (lanes D and E). This peptide was found in the three elastins tested from the skins of PXE patients. The protein band at 25,000 Da represents elastase. Binding ofelastase to elastin. Elastin is known to bind to elastase to form a precipitable complex. Sodium dodecyl sulfate can facilitate this binding leading to an increase in elastase activity (Kagan et al., 1972). We compared the binding of elastase to normal skin elastin and to ligamenturn nuchae elastin in order to determine if the normal skin elastin could bind increased amounts of elastase. Approximately 35% of the elastase originally added was bound to the pelleted ligamenturn nuchae elastin whereas 65% was found in the supernatant. These results are in agreement with those obtained by Kagan et al. (1972). Approximately 28% of the elastase was bound to the pellet of normal skin elastin whereas 72% was found in the supernatant. Therefore, the increased elastolysis of normal skin elastin is not due to increased binding of elastase. DISCUSSION There is considerable histological and ultrastructural evidence that the elastic fiber is the main site of the abnormality in PXE skin. Histologically PXE skin samples have been shown to contain increased amounts of elastic fibers in the middle and lower dermis. These fibers are clumped, fragmented, and calcified. Studies on elastin purified from PXE skin have failed to reveal any alteration in
FIG. 2. SDS-polyacrylamide gel electrophoretic pattern of digestion supematants. Lane A contains protein standards (phosphorylase B, 95,000; glutaraldehyde dehydrogenase, 55,000; ovalbumin, 43,000; lactate dehydrogenase, 36,000; carbonic anhydrase, 29,000; lactoglobulin, 18,400; and cytochrome c, 12,400). Lane B contains the supematant from the digestion of PXE elastin. Lane C contains the supematant from the digestion of PXE elastin with hyaluronidase preincubation. Lane D contains the supematant from the digestion of normal skin elastin. Lane E contains the supematant from the digestion of normal skin elastin with hyaluronidase preincubation.
194
SCHWARTZ
ET AL.
amino acid composition or desmosine content. Histological and electron microscopic studies have identified the binding of polyanions to the elastic fibers (Martinez-Hernandez and Huffer, 1974) as an important abnormality in PXE. In this investigation we studied the digestion of normal and PXE skin elastins by pancreatic elastase. Results demonstrated first that normal skin elastin was more susceptible to proteolytic cleavage by elastase than was either ligamenturn nuchae or human aortic elastins. This property of skin elastin may play a role in the degeneration of elastic fibers seen in cutaneous aging. The increased activity was not due to increased binding of elastase to skin elastin. Since the amino acid composition and desmosine content of the elastins are similar (Schwartz et al., 1990), human skin elastin may have a different conformation which favors proteolytic degradation. Variations in the structure of elastin mRNA resulting from alternative splicing have been reported. Therefore a unique splicing pattern for dermal elastin could result in an altered conformation (Indik et al., 1989) perhaps by deleting a cross-linking site. Interestingly, PXE skin elastin was more resistant than was normal skin elastin to elastase digestion. The resistance of PXE elastin to proteolysis may be due to the glycosaminoglycans blocking the enzyme activity. Preincubating with testicular hyaluronidase increased the amount of elastin solubilized two-fold but did not equal values obtained for normal skin elastin. It is possible that hyaluronidase resistant glycosaminoglycans (such as dermatan sulfate) were bound to the elastin. Martinez-Hemandez and Huffer (1974) reported that polyanions remained associated with elastin in PXE skin following hyaluronidase digestion. Furthermore, Koreen et al. (1987) found increased amounts of both hyaluronic acid and dermatan sulfate in lesional skin from PXE patients. Elastase digestion of PXE elastin with hyaluronidase pretreatment generated a unique 15,000-Da molecular weight fragment. This fragment either lacks an essential site for enzyme activity or this site is inaccessible. Work is in progress to characterize this peptide. These studies suggest that elastin purified from PXE skin is associated with bound glycosaminoglycans. Pasquali-Ronchetti et al. (1984) found that lathyritic elastin bound mainly dermatan and heparan sulfates whereas collagen bound mainly hyaluronic acid and chondroitin sulfate. The glycosaminoglycans may bind to the positively charged amino groups on elastin (Fornieri et al., 1987). It is clear from studies on nonlesional PXE skin that the increase in glycosaminoglycans is an early event and may precede the increased synthesis of elastin. The glycosaminoglycans may then bind to tropoelastin and interfere with normal elastic fiber formation to yield the clumped, thickened elastic fibers characteristic of the disorder. REFERENCES 1. BRADFORD, M. M. (1976). A rapid and sensitive method for quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248-254. 2. DANIELSEN, L. L. (1982). Morphologic changes in pseudoxanthoma elasticum and senile skin. Int. J. Dermatol.
21, 449-454.
3. FORNIERI, C., BACCARANI-CONTRI, M., QUAGLINO, D., and PASQUALI-RONCHETTI, I. (1987). Lysyl oxidase activity and elastin/glycosaminoglycan interactions in growing chick and rat aortas. J. Cell Biol. 105, 1463-1469. 4. INDIK,
Z., YEH,
H., ORNSTEIN-GOLDSTEIN,
N., KUCICH,
U., ABRAMS,
W., ROSENBLOOM,
J. C.,
and ROSENBLOOM, J. (1989). Structure of the elastin gene and alternative splicing of elastin mRNA: Implications for human disease. Am. J. Med. Gen. 34, 81-90.
ELASTASE
DIGESTION
OF
SKIN
ELASTIN
195
5. KAGAN, H. M., CROMBIE, G. D., JORDAN, R. E., LEWIS, W., and FRANZBLAU, C. (1972). Proteolysis of elastin-ligand complexes. Stimulation of elastase digestion of insoluble elastin’by sodium dodecyl sulfate. Biochemisrry 11, 3412-3418. 6. KOREEN, R., WEINFLASH-KONSTADT, J., GRABOWSKI, G., LONGAS, M. O., TOOME, B., LEMLICH, G., and LEBWOHL, M. G. (1987). Quantitative evidence for increased hyaluronic acid and dermatan sulfate in lesional and nonlesional pseudoxanthoma elasticum skin. J. Invest. Dermatol. 88, 500. 7. LAEMMLI, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685. 8. LONGAS, M. O., WISCH, P., LEBWOHL, M. G., and FLEISCHMAJER, R. (1986). Glycosaminoglycans of skin and urine in pseudoxanthoma elasticum. Evidence for chondroitin 6 sulfate alteration. Clin. Chim. Acta 155, 227-236. 9. MARTINEZ-HERNANDEZ, A., and HUFFER, W. E. (1974). Pseudoxanthoma elasticurn: Dermal polyanions and the mineralization of elastic fibers. Lab. Invest. 31, 181-186. 10. MCKUSICK, V. A. (1972). Pseudoxanthoma elasticurn. In “Heritable Disorders of Connective Tissue,” pp. 475-520. Mosby, St. Louis. 11. PASQUALI-RONCHETTI, I., BRESSAN, G. M., FORNIERI, C., BACCARANI-CONTRI, M., CASTELLANI, I., and VOLPIN, D. (1984). Elastin fiber-associated glycosaminoglycans in B-aminopropionitrile induced lathyrism. Exp. Mol. Pathol. 40, 235-245. 12. PASQUALI-RONCHETTI, I., VOLPIN, D., BACCARANI-CONTRI, M., CASTELLANI, I., and PESERICO, A. (1981). Pseudoxanthoma elasticum. Biochemical and Ultrastructural studies. Dermatologica 163, 307-325. 13. SCHWARTZ, E., CRUICKSHANK, F. A., and LEBWOHL, M. (1990). Determination of desmosines in elastin-related skin disorders by isocratic high performance liquid chromatography. Exp. Mol.
Pathol. 52, 63-68. 14. SMITH, J. G., DAVIDSON, E. A., and HILL, R. L. (1963). Composition of normal and pathological cutaneous elastin. Nature 197, 1108-l 109. 15. SMITH, J. G., DAVIDSON, E. A., and TAYLOR, R. W. (1964). Cutaneous acid mucopolysaccharides in pseudoxanthoma elasticurn. J. Invest. Dermatol. 43, 42W30. 16. SMITH, J. G., SAMS, W. M., DAVIDSON, E. A., and CLARK, R. D. (1962). Pseudoxanthoma elasticum: Histochemical alterations. Arch. Dermatol. 86, 741-744. 17. WALKER, E. R., FREDERICKSON, R. G., and MAYES, M. D. (1989). The mineralization of elastic fibers and alterations of extracellular matrix in pseudoxanthoma elasticurn. Arch. Dermatol. 125, 70-76.
AND
MOLECULAR
PATHOLOGY
55,
190-19.5 (191)
Elastase Digestion of Normal and Pseudoxanthoma Lesional Skin Elastins ELAINE
SCHWARTZ,
MARK
THIEBERG, FREDERICK MARK LEBWOHL
Elasticum
A. CRUICKSHANK,
AND
Deparrment of Dermatology, Mount Sinai School of Medicine, New York, New York 10029 Received February 5, 1991; and in revised form April 11, 1991 Pseudoxanthoma elasticum (PXE) is a heritable disorder of connective tissue that is characterized by redundant folds of skin in flexural areas. There is considerable evidence that suggests that the elastic fiber is the main site of the abnormality although the primary molecular defect has not been identified. The aim of this study was to identify differences between PXE and normal skin elastins. Elastins from normal, nonsolar-exposed skin, and pseudoxanthoma elasticum lesional skin were purified and their solubilization by pancreatic elastase was compared. Results demonstrated that elastin derived from normal skin was more susceptible to proteolytic cleavage than elastin purified from either pseudoxanthoma elasticum lesional skin or ligamentum nuchae. Pretreatment of the lesional elastin with testicular hyaluronidase increased its solubilization two-fold and generated a unique 15,000 Da molecular weight fragment. Elastin prepared from PXE skin may contain bound glycosaminoglycans which interfere with elastase activity. The susceptibility of normal skin elastin to proteolytic degradation may have implications in the study of aging skin. 0 1991 Academic
Press, Inc.
INTRODUCTION Pseudoxanthoma elasticum (PXE) is a heritable disorder of connective tissue whose lesions involve the skin, eyes, and cardiovascular system (McKusick, 1972). Skin lesions are characterized by redundant folds in flexural areas and by the presence of yellow xanthoma-like papules and plaques. Vascular changes include calcification of arteries and the sequelae include hypertension, coronary artery occlusion, recurrent gastrointestinal bleeding, and cerebral hemorrhage. The eye changes are described as angioid streaks in the ocular fundus. The primary etiology of PXE has been a matter of controversy for many years. Histological studies revealed that PXE skin had increased quantities of glycosaminoglycans and calcified elastic fibers (Smith et al., 1962; Martinez-Hernandez and Huffer, 1974; Danielsen, 1982; Walker et at., 1989) whereas biochemical studies demonstrated increased amounts ofhexosamine, calcium, and elastin in PXE skin (Smith et al., 1962; Pasquali-Ronchette et al., 1981). The major glycosaminoglycan is known to be hyaluronic acid (Smith et al., 1964) although increased amounts of chondroitin sulfate are also present (Longas et al., 1986). The earliest detectable PXE abnormality is believed to be the accumulation of polyanions in the PXE dermis. Polyanions were demonstrated within mineralized elastic fibers in clinically affected dermis and within nonmineralized elastic fibers in unaffected dermis (Martinez-Hernandez and Huffer, 1974). Koreen et al. (1987) found increased amounts of hyaluronic acid and dermatan sulfate in lesional and nonlesional skins of five patients with PXE. These polyanions are believed to serve as initiating factors for the mineralization. The elastic fibers in PXE skin are increased in quantity and are abnormal in appearance (clumped and fragmented). However, the amino acid composition and 190 0014-4800/91 $3.00 Copyright Q 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
ELASTASE
DIGESTION
OF
SKIN
ELASTIN
191
desmosine content of PXE skin elastin are similar to that of normal skin elastin (Smith et al., 1%3; Schwartz et al., 1990). It was our aim to identify differences between PXE and normal skin elastins that might explain the altered appearance of the elastic fibers in PXE. This paper describes studies on the digestion by pancreatic elastase of elastins prepared from normal and PXE skins. MATERIALS
AND METHODS
Preparation of e&tin. Nonsolar-exposed skins (n = 4) were obtained from post-surgical specimens (reduction mammoplasty) of normal individuals (34-39 years of age). Lesional axillary skins from unequivocal cases of PXE were obtained from postsurgical specimens. The PXE patients ranged in age from 38 to 74 and all had extensive axillary involvement and angioid streaks. Normal-appearing skin was obtained from one PXE patient. Skin was homogenized using a Virtis in phosphate-buffered saline (pH 7.4) containing proteolytic inhibitors, defatted, and extracted overnight with a solution containing 1% sodium dodecyl sulfate (SDS), 0.05 M Tris-HCl (pH 7.5), 0.33 M mercaptoethanol, and proteolytic inhibitors. The suspension was then centrifuged and the pellet was washed extensively with water (at least 10 times) prior to lyophilization. The residue was then treated with cyanogen bromide (Schwartz et al., 1990). The insoluble fraction (elastin) was washed with water and lyophilized. The purity of the elastin was confirmed by amino acid analysis. The elastins prepared from PXE and normal skins had amino acid compositions typical of elastin (Smith et al., 1963). Elustuse digestion. Elastin samples (5 mg) were preincubated overnight (37°C with mixing) in 3 ml of 0.2 M NH,HCO, (pH 8) centrifuged, and resuspended in fresh buffer containing porcine pancreatic elastase at an enzyme:substrate ratio of 1:250. Samples were incubated at 37°C with mixing. Aliquots were removed at defined time points and phenylmethylsulfonyl fluoride was added to stop the incubation. The aliquots were centrifuged and the amount of solubilized protein was quantified (Bradford, 1976). In some experiments, testicular hyaluronidase (2 mg/ml) was included in the preincubation buffer. SDS polyucrylumide gel electrophoresis. Samples were electrophoresed on polyacrylamide gels as described by Laemmli (1970). Precast (l&20%) gels and molecular weight standards were purchased from Integrated Separation Systems (ISS) (Hyde Park, MA). Equivalent amounts of protein (50 kg) were applied to the lanes. Gels were stained with the ISS Pro-Blue Staining System. Binding of elustuse to elustin. The binding of pancreatic elastase to normal human skin and bovine ligamentum nuchae elastins was determined as described by Kagan et al. (1972). Insoluble elastins (10 mg) from normal skin and from bovine ligamentum nuchae were incubated with 20 kg elastase for 2 min in 2 ml of 0.05 M ammonium acetate buffer (pH 8.45). Following centrifugation the pellets were resuspended in fresh buffer and incubated at 37°C. Ten milligrams of elastin was added to the supernatants and these suspensions were incubated at 37°C. The rates of solubilization of the elastins reflect the proportion of elastase present in the pellets and supernatants.
RESULTS Digestion
The normal skin elastin was soluelastase to a greater extent than was PXE skin elastin. Time is shown in Fig. 1. Results (Table I) expressed as total p,g
of PXE and normal skin elustins.
bilized,by pancreatic course of elastolysis
192
SCHWARTZ
0
2
ET AL.
4
6
8
10
Time (hours) FIG. 1. Time course of elastolysis of elastin by porcine pancreatic elastase. Values (mean of duplicate determinations) represent the amount of solubilized elastin (as determined by Bradford protein assay) present in the supematants following incubation with porcine pancreatic elastase. (Cl) PXE; (+) bovine ligamentum nuchae; (m) normal shin.
protein solubilized at 4 hr (means + SEM) were 98 + 10 for PXE skin elastin (n = 3) and 336 + 9 for normal skin elastin (n = 4). The PXE elastin therefore exhibited significantly decreased susceptibility to elastase. Interestingly, the vaiues (means of duplicate determinations) for bovine ligamentum nuchae and human aortic elastins (obtained from Elastin Products Co., Pacific, MO) were 145 and 141 p,g protein solubilized, respectively. Furthermore, elastin prepared from nonlesional skin obtained from a PXE patient also exhibited decreased elastolysis (125 pg protein solubilized in 4 hr). The assay was linear with respect to enzyme and elastin concentrations. These results demonstrated that normal skin elastin was more susceptible to proteolytic cleavage than either bovine ligamentum nuchae or human aortic elastin. Furthermore, elastin prepared from PXE skin was more resistant to elastolysis than was normal skin elastin. Effect of preincubation with hyaluronidase. When PXE elastin was preincubated with testicular hyaluronidase, the amount of elastin solubilized by elastase increased two-fold (2.03 ? 0.16, mean + SEM, n = 3). The amount of normal skin elastin that was solubilized only slightly increased following hyaluronidase preincubation (1.11 + 0.02, mean f SEM, n = 3). Solubilization Source
PXE lesional
TABLE I of Dermal Elastin by Pancreatic Elastase
of elastin
Total pg solubilized in 4 hr (1)
(2) (3) Normal
112 78 103 mean (?SEM) = 98 (210)
(1)
311
(2)
349 345 340
(3) (4)
mean (*SEM)
= 336 (29)
ELASTASE
DIGESTION
OF
SKIN
ELASTIN
193
The peptides solubilized from PXE and normal skin elastins by elastase (with and without hyaluronidase preincubation) were resolved by SDS-PAGE. Results (Fig. 2) demonstrated a distinct peptide (molecular weight of 15,000 Da) in the PXE with hyaluronidase sample (lane C) that was absent in the normal skin samples (lanes D and E). This peptide was found in the three elastins tested from the skins of PXE patients. The protein band at 25,000 Da represents elastase. Binding ofelastase to elastin. Elastin is known to bind to elastase to form a precipitable complex. Sodium dodecyl sulfate can facilitate this binding leading to an increase in elastase activity (Kagan et al., 1972). We compared the binding of elastase to normal skin elastin and to ligamenturn nuchae elastin in order to determine if the normal skin elastin could bind increased amounts of elastase. Approximately 35% of the elastase originally added was bound to the pelleted ligamenturn nuchae elastin whereas 65% was found in the supernatant. These results are in agreement with those obtained by Kagan et al. (1972). Approximately 28% of the elastase was bound to the pellet of normal skin elastin whereas 72% was found in the supernatant. Therefore, the increased elastolysis of normal skin elastin is not due to increased binding of elastase. DISCUSSION There is considerable histological and ultrastructural evidence that the elastic fiber is the main site of the abnormality in PXE skin. Histologically PXE skin samples have been shown to contain increased amounts of elastic fibers in the middle and lower dermis. These fibers are clumped, fragmented, and calcified. Studies on elastin purified from PXE skin have failed to reveal any alteration in
FIG. 2. SDS-polyacrylamide gel electrophoretic pattern of digestion supematants. Lane A contains protein standards (phosphorylase B, 95,000; glutaraldehyde dehydrogenase, 55,000; ovalbumin, 43,000; lactate dehydrogenase, 36,000; carbonic anhydrase, 29,000; lactoglobulin, 18,400; and cytochrome c, 12,400). Lane B contains the supematant from the digestion of PXE elastin. Lane C contains the supematant from the digestion of PXE elastin with hyaluronidase preincubation. Lane D contains the supematant from the digestion of normal skin elastin. Lane E contains the supematant from the digestion of normal skin elastin with hyaluronidase preincubation.
194
SCHWARTZ
ET AL.
amino acid composition or desmosine content. Histological and electron microscopic studies have identified the binding of polyanions to the elastic fibers (Martinez-Hernandez and Huffer, 1974) as an important abnormality in PXE. In this investigation we studied the digestion of normal and PXE skin elastins by pancreatic elastase. Results demonstrated first that normal skin elastin was more susceptible to proteolytic cleavage by elastase than was either ligamenturn nuchae or human aortic elastins. This property of skin elastin may play a role in the degeneration of elastic fibers seen in cutaneous aging. The increased activity was not due to increased binding of elastase to skin elastin. Since the amino acid composition and desmosine content of the elastins are similar (Schwartz et al., 1990), human skin elastin may have a different conformation which favors proteolytic degradation. Variations in the structure of elastin mRNA resulting from alternative splicing have been reported. Therefore a unique splicing pattern for dermal elastin could result in an altered conformation (Indik et al., 1989) perhaps by deleting a cross-linking site. Interestingly, PXE skin elastin was more resistant than was normal skin elastin to elastase digestion. The resistance of PXE elastin to proteolysis may be due to the glycosaminoglycans blocking the enzyme activity. Preincubating with testicular hyaluronidase increased the amount of elastin solubilized two-fold but did not equal values obtained for normal skin elastin. It is possible that hyaluronidase resistant glycosaminoglycans (such as dermatan sulfate) were bound to the elastin. Martinez-Hemandez and Huffer (1974) reported that polyanions remained associated with elastin in PXE skin following hyaluronidase digestion. Furthermore, Koreen et al. (1987) found increased amounts of both hyaluronic acid and dermatan sulfate in lesional skin from PXE patients. Elastase digestion of PXE elastin with hyaluronidase pretreatment generated a unique 15,000-Da molecular weight fragment. This fragment either lacks an essential site for enzyme activity or this site is inaccessible. Work is in progress to characterize this peptide. These studies suggest that elastin purified from PXE skin is associated with bound glycosaminoglycans. Pasquali-Ronchetti et al. (1984) found that lathyritic elastin bound mainly dermatan and heparan sulfates whereas collagen bound mainly hyaluronic acid and chondroitin sulfate. The glycosaminoglycans may bind to the positively charged amino groups on elastin (Fornieri et al., 1987). It is clear from studies on nonlesional PXE skin that the increase in glycosaminoglycans is an early event and may precede the increased synthesis of elastin. The glycosaminoglycans may then bind to tropoelastin and interfere with normal elastic fiber formation to yield the clumped, thickened elastic fibers characteristic of the disorder. REFERENCES 1. BRADFORD, M. M. (1976). A rapid and sensitive method for quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248-254. 2. DANIELSEN, L. L. (1982). Morphologic changes in pseudoxanthoma elasticum and senile skin. Int. J. Dermatol.
21, 449-454.
3. FORNIERI, C., BACCARANI-CONTRI, M., QUAGLINO, D., and PASQUALI-RONCHETTI, I. (1987). Lysyl oxidase activity and elastin/glycosaminoglycan interactions in growing chick and rat aortas. J. Cell Biol. 105, 1463-1469. 4. INDIK,
Z., YEH,
H., ORNSTEIN-GOLDSTEIN,
N., KUCICH,
U., ABRAMS,
W., ROSENBLOOM,
J. C.,
and ROSENBLOOM, J. (1989). Structure of the elastin gene and alternative splicing of elastin mRNA: Implications for human disease. Am. J. Med. Gen. 34, 81-90.
ELASTASE
DIGESTION
OF
SKIN
ELASTIN
195
5. KAGAN, H. M., CROMBIE, G. D., JORDAN, R. E., LEWIS, W., and FRANZBLAU, C. (1972). Proteolysis of elastin-ligand complexes. Stimulation of elastase digestion of insoluble elastin’by sodium dodecyl sulfate. Biochemisrry 11, 3412-3418. 6. KOREEN, R., WEINFLASH-KONSTADT, J., GRABOWSKI, G., LONGAS, M. O., TOOME, B., LEMLICH, G., and LEBWOHL, M. G. (1987). Quantitative evidence for increased hyaluronic acid and dermatan sulfate in lesional and nonlesional pseudoxanthoma elasticum skin. J. Invest. Dermatol. 88, 500. 7. LAEMMLI, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685. 8. LONGAS, M. O., WISCH, P., LEBWOHL, M. G., and FLEISCHMAJER, R. (1986). Glycosaminoglycans of skin and urine in pseudoxanthoma elasticum. Evidence for chondroitin 6 sulfate alteration. Clin. Chim. Acta 155, 227-236. 9. MARTINEZ-HERNANDEZ, A., and HUFFER, W. E. (1974). Pseudoxanthoma elasticurn: Dermal polyanions and the mineralization of elastic fibers. Lab. Invest. 31, 181-186. 10. MCKUSICK, V. A. (1972). Pseudoxanthoma elasticurn. In “Heritable Disorders of Connective Tissue,” pp. 475-520. Mosby, St. Louis. 11. PASQUALI-RONCHETTI, I., BRESSAN, G. M., FORNIERI, C., BACCARANI-CONTRI, M., CASTELLANI, I., and VOLPIN, D. (1984). Elastin fiber-associated glycosaminoglycans in B-aminopropionitrile induced lathyrism. Exp. Mol. Pathol. 40, 235-245. 12. PASQUALI-RONCHETTI, I., VOLPIN, D., BACCARANI-CONTRI, M., CASTELLANI, I., and PESERICO, A. (1981). Pseudoxanthoma elasticum. Biochemical and Ultrastructural studies. Dermatologica 163, 307-325. 13. SCHWARTZ, E., CRUICKSHANK, F. A., and LEBWOHL, M. (1990). Determination of desmosines in elastin-related skin disorders by isocratic high performance liquid chromatography. Exp. Mol.
Pathol. 52, 63-68. 14. SMITH, J. G., DAVIDSON, E. A., and HILL, R. L. (1963). Composition of normal and pathological cutaneous elastin. Nature 197, 1108-l 109. 15. SMITH, J. G., DAVIDSON, E. A., and TAYLOR, R. W. (1964). Cutaneous acid mucopolysaccharides in pseudoxanthoma elasticurn. J. Invest. Dermatol. 43, 42W30. 16. SMITH, J. G., SAMS, W. M., DAVIDSON, E. A., and CLARK, R. D. (1962). Pseudoxanthoma elasticum: Histochemical alterations. Arch. Dermatol. 86, 741-744. 17. WALKER, E. R., FREDERICKSON, R. G., and MAYES, M. D. (1989). The mineralization of elastic fibers and alterations of extracellular matrix in pseudoxanthoma elasticurn. Arch. Dermatol. 125, 70-76.
