Biochimica et Biophysica Acta, 439 (1976) 107-115
© Elsevier Scientific Publishing Company, Amsterdam -- Printed in The Netherlands BBA 37388 UREA-EXTRACTABLE PROTEIN FROM H U M A N EPIDERMIS
J. BAYNES and M. LEVINE Department of Biochemistry, University of Manchester, Manchester MI3 9PL ( U.K.)
(Received December 3rd, 1975)
SUMMARY The major protein fraction extracted from human epidermis by urea has the following characteristics: (1) A molecular weight of approximately 600 000. (2) A copper content of 0.03 ~ giving a minimum molecular weight of 212 000. (3) A non-protein moiety indicated by (a) protein/dry weight ratio (b) anomalous ultraviolet absorption spectrum (c) high-voltage electrophoresis after mild alkaline or acid treatment (d) an excess of imidazole groups by Pauly reaction compared with the number of histidine residues by amino acid analysis. The non-protein component is not urocanic acid or RNA as has been reported (Bernstein, I. A. (1970), J. Soc. Cosmet. Chem. 21, 583-594 and Sibrack, L. A., Gray, R. H. and Bernstein, I. A. (1974), J. Invest. Dermatol. 62, 394-405). (4) The presence of interchain disulphide linkages indicated by polyacrylamide gel electrophoresis in sodium dodecyl sulphate with and without fl-mercaptoethanol. (5) Polyacrylamide gel electrophoresis in presence and absence of 6 M urea suggests that aggregates are broken down in urea.
INTRODUCTION The study of keratinisation in epidermis has been hampered by the difficulties in isolating the insoluble epidermal proteins involved in the process. The traditional method for extraction of the fibrous protein has been by concentrated urea. Rudall [1] in 1952 gave the name "epidermin" to the urea-soluble protein obtained from cow snout epidermis. Epidermin gave an a-type X-ray diffraction pattern, was classified with the a-keratin group and identified as a fibrous protein. Similar extractions have been made on other mammalian species [2, 3]. Citric acid/sodium citrate buffer, pH 2.6, has also proved successful for the isolation of fibrous protein from bovine nose epidermis and the multichain molecule obtained was termed "prekeratin" [4, 5]. There is now evidence indicating that proAbbreviation: SDS, sodium dodecyl sulphate.
108 teins extracted either by urea or by citrate buffer are related by common identical components [6] but their nature and chemical structure are still poorly understood. In a recent model postulated by Bernstein [3], the urea-extractable protein is depicted "in situ" as covalently bonded to a urocanic acid molecule at one end and copper is shown as associated through the imidazole residue. Our observations reported here have been made on the main protein fraction isolated from human epidermis by urea. We have confirmed Bernstein's finding that the large protein aggregate isolated is associated with copper. In addition we have isolated free urocanic acid from skin and identified it. However, our evidence indicates that there is no bound urocanic acid associated with the protein. Another unknown component has been separated from the protein by either mild acid or alkaline treatment. It is not urocanic acid or any of the common nucleic acid components. MATERIALS AND METHODS Hanks' balanced salt solution and Eagle's minimal essential medium were obtained from Wellcome Reagents Ltd. The Radiochemical Centre, Amersham supplied L-[2,5-3H2]histidine, 58 Ci/mmol.
Preparation of tissue Samples of human skin were obtained from the lower thigh region after leg amputations using a dermatome set at a thickness of 0.38 mm. The tissue was placed on gauze moistened with Hanks' balanced salt solution and stored at --20 °C or at 4 °C when required for incorporation studies. Labelled histidine was incorporated into skin using a method adapted from Brotherton [7] and Marks et al. [8]. The skin was floated, epidermis upwards in Eagle's minimal essential medium containing antibiotics, 1 ~ (w/v) glutamine and L-[2,5-3H2]histidine (1 ffCi/ml). Incubation was at 37 °C and 75 ~ humidity for 24 h in a gas mixture of 9 5 ~ air and 5 ~ carbon dioxide. The epidermis was separated from the dermis in early experiments by heat treatment, 60 °C for 2 rain. In later experiments the skin was frozen over dry ice and tile epidermis removed by scraping with a scalpel (Dr. Sprott, Unilever Research. Colwerth House, Bedford, personal communication).
Preparation of urea-extractable proteins The epidermal mince or scrapings were homogenized in 8 M urea, 0.2 M Tris-HC1 buffer (pH 8.5). Debris was removed by low speed centrifugation and the supernatant dialyzed against 0.1 M ammonium hydroxide at 4 °C overnight. Dialysis was continued for 48 h against 0.01 M ammonium hydroxide after which any aggregates formed were removed by centrifugation at 40 000 x g for 20 min at 0 °C and the supernatant was lyophilized. In certain experiments the urea extract was treated with 0.1 M perchloric acid. The effect of perchloric acid on the protein extracted was investigated.
Isolation and characterization of urea-extractable protein The dried urea extract was dissolved in 0.02 M sodium carbonate, applied to a sephadex G-200 column (bed dimensions 2 × 30 cm) and eluted with 0.01 M am-
109
monium hydroxide. Each fraction was tested for absorbance at 280 nm, protein by the method of Lowry [9], RNA by the orcinol reaction and radioactivity after incorporation of L-[2,5-3H2]histidine. The main protein fractions were pooled and lyophilized for further tests.
High voltage paper electrophoresis High voltage paper electrophoresis was carried out following the method of Markham and Smith [10]. The apparatus consisted of two compartments with platinum electrodes, and a central tank filled with carbon tetrachloride for cooling purposes. Samples were applied to the centre of a strip of paper 10 x 80 cm (Whatman No. 1) which was then moistened with the electrophoretic buffer: pyridine (10 ml)/acetic acid (92 ml)/water (1.6 litres) (pH 3.5). A voltage of 1500 V was applied for 1 h. After drying, pyridine was washed from the electrophoretogram using 95 ethanol (100 ml)/ether (900 ml), containing a few drops of "880" ammonia. Ultraviolet-absorbing materials were localized by ultraviolet photography.
Polyacrylamide gel electrophoresis The method used for polyacrylamide gel electrophoresis in SDS was based on Dunker and Rueckert [1 I] and Weber and Osborn [12]. Standard protein samples were prepared at a concentration of 1-2 mg/ml by incubation for 1-2 h at 37 °C in 0.01 M sodium phosphate buffer (pH 7.0), containing 1 ~ (w/v) SDS and 1 ~ (v/v) fl-mercaptoethanol. Samples of 100 ffl containing bromophenol blue as a marker, were applied to 7.5 ~ polyacrylamide gels, 3 ~ cross-linking, and electrophoresis was carried out in a continuous system of 0.1 M sodium phosphate (pH 7.0) containing 0.1 ~ (w/v) SDS. An initial current of 5 mA/tube was applied until the samples entered the gels and was then increased to 8 mA/tube for 3-4 h. Gels were cut through the bromophenol blue region and stained for protein with 1 ~ (w/v) amido black in 7 ~ (v/v) acetic acid. Destaining was achieved by continuous elution with 7 ~ (v/v) acetic acid. Relative mobilities were calculated as the ratio of the distance travelled by the protein to the length of the gel. For molecular weight determinations ureaextractable protein incubated in SDS with and without fl-mercaptoethanol and standard proteins prepared as above were run at the same time. In addition, polyacrylamide gel electrophoresis was carried out in the absence of SDS using 5 ~ polyacrylamide gels, 3 ~ cross-linking and a discontinuous buffer system. The cathode compartment contained 25 mM imidazole/acetate buffer (pH 7.4) whereas the gel and anode contained 50 mM sodium chloride/25 mM imidazole-HC1 buffer (pH 7.4). Samples of urea-extractable protein were prepared in 0.02 M sodium carbonate or 6 M urea. A current of 5 mA/tube was applied and the gels run for 1-2 h. The staining procedure was as previously described. Gels were scanned on a Joyce Loebl gel scanner at a wavelength of 265 nm.
Amino acid analysis 'Urea-extractable protein was hydrolyzed in 6 M HCI (re-distilled) in sealed tubes under nitrogen overnight at 100 °C. Analyses were performed on a Jeol JLC6AH amino acid analyzer.
t10
The Pauly reaction tests Pauly reagent was freshly prepared for each experiment according to Reaven and Cox [13]. It was used as a spray for qualitative purposes. Estimation ofimidazole content was carried out by adding 3.6 ml Pauly reagent to 0.4 ml sample. The colour produced was measured against a reagent blank at 520 nm. Histidine was used as a standard.
Paper chromatography Paper chromatography of amino acids was carried out in a single dimension using 2-propanol/acetone/1 M HCI (60:15:25, v/v) as a developing solvent. Histidine, tyrosine and urocanic acid were localized using Pauly reagent spray. RESULTS The elution profile of urea-extracted protein after Sephadex G-200 chromatography (Fig. 1) revealed one main protein peak (fraction A), with a molecular weight of about 600 000. Approximately 75 % of the total Lowry protein in the urea extract appears in fraction A and its position was indicated by absorbance at 280 nm, radioactivity after labelled histidine incorporation, estimation of Lowry protein, and by orcinol reaction. Fractions in this region were pooled and lyophilized. A shoulder
30
OOO Cl:)m / rnl
i
i
o 20
I
o51-
I; 7
,
\
,,
"A
OOO
dlO o o o
-~.
0 10
30
50 O.O1 iv1 N H 4 O H
70
90
( rrtl )
Fig. 1. Gel filtration of the urea extract. After incorporation of [aH]histidine,'.the~epidermis was homogenized in 8 M urea. The urea extract was dialyzed, lyophilized and chromatographed on a sephadex G-200 column (2 x 30 cm), equilibrated and eluted with 0.01 M NH4OH. O---O cpm/ml, 0 - - 0 absorbance at 280 nm, A---A absorbance at 700 nm after the Lowry assay on each fraction, J,--A absorbance at 660 nm after the orcinol reaction on each fraction.
111 of lower molecular weight protein was also detectable. It had a similar anomalous ultraviolet spectrum, A2ao/Lowry protein ratio and the same specific activity after histidine incorporation. One other significant peak was eluted in the low molecular weight region. It was detected by absorbance at 280 nm and reaction with orcinol but was largely non-protein in nature by the Lowry method. Treatment of the urea extract with 0.1 M perchloric acid appeared to split fraction A into smaller units since the main protein in perchloric acid-insoluble material had a molecular weight of 160 000 and in perchloric acid-soluble material 50 000. In addition, copper and ultraviolet absorbing material was released both of which became dialyzable. Following in vitro incorporation of tritiated histidine into whole skin for 24 h it was found that the specific activities of the urea extract and perchloric acid-soluble material were similar (Table I). However, the specific activity of perchloric acidinsoluble material was 67 7oo higher. These results are not consistent with those obtained by Bernstein and co-workers who found that a histidine-rich protein was solubilized by treatment of the urea extract with 0.1 M perchloric acid [14].
TABLE I INCORPORATION OF [3H]HISTIDINE INTO EPIDERMAL EXTRACTS [aH]Histidine (5.27 Ci/mol) was incorporated into whole skin for 24 h. The epidermis was extracted with 8 M urea and the dried urea extract treated with 0.1 M perchloric acid. Specific activities of the crude urea extract, perchloric acid-insoluble material and perchloric acid-soluble material were measured. Fractionation of the crude extracts by Sephadex chromatography revealed one main protein fraction in each case. The estimated molecular weights and specific activities of the fractionated proteins are also given for comparison. PCA, perchloric acid. Extract
Specific activity of crude extract (a) dpm/mg protein
Main protein peak after fractionation
(b) Molecular nmol weight histidine/mg ( × 10-a) protein/24 h
Specificactivity (a) (b) dpm/mg nmol protein histidine/mg ( × l0 -a) protein/24h
9.23 15.4 9.23
106 140 90
( × 10-3) Urea extract PCA-insolublematerial PCA-solublematerial
108 180 108
600 160 50
9.06 12.0 7.69
Measurement of the ultraviolet absorption spectrum of fraction A in 0.01 M ammonium hydroxide showed a maximum at 266 nm. This is uncharacteristic of protein indicating the presence of an additional ultraviolet-absorbing moiety in fraction A. Protein estimations carried out by several methods were compared with dry weight measurements. The ratio of protein to dry weight by A,4ns_225 nm, Lowry and A2a0 nm/A260nm measurements were 0.47, 0.52 and 0.59, respectively. Bovine serum albumin was used as a reference standard in the Lowry protein estimations. The
112 a m i n o acid c o m p o s i t i o n o f bovine s e r u m a l b u m i n a n d f r a c t i o n A m a y n o t be similar e n o u g h to p r o d u c e a c c u r a t e results. The o t h e r m e t h o d s used involved the m e a s u r e m e n t o f a b s o r b a n c e in the u l t r a v i o l e t region. Since f r a c t i o n A has been s h o w n to have a n u n c h a r a c t e r i s t i c u l t r a v i o l e t s p e c t r u m o f p r o t e i n these also c o u l d be subject to error. H o w e v e r , the three m e t h o d s are in fairly close a g r e e m e n t suggesting that f r a c t i o n A c o n t a i n s n o n - p r o t e i n m a t e r i a l which w o u l d be i n c l u d e d in the d r y weight b u t n o t in the p r o t e i n e s t i m a t i o n s . It was, therefore, necessary t h a t all calculations involving the c o n c e n t r a t i o n o f p r o t e i n s h o u l d specify the m e t h o d used. C o p p e r analysis o f l y o p h i l i z e d f r a c t i o n A by a t o m i c emission s p e c t r o s c o p y gave a c o p p e r c o n t e n t o f 0.0370 (weight c o p p e r / d r y weight). This value gives a m i n i m u m m o l e c u l a r weight o f the m o l e c u l e as 212 000. T h e i m i d a z o l e c o n t e n t o f f r a c t i o n A , using the P a u l y r e a c t i o n with a histidine s t a n d a r d , was f o u n d to be 1.85 70 (weight i m i d a z o l e / w e i g h t L o w r y protein) which is e q u i v a l e n t to 6.7 i m i d a z o l e g r o u p s p e r 100 a m i n o acid residues. The a m i n o acid analysis o f f r a c t i o n A is s h o w n in T a b l e I1. O t h e r results obt a i n e d for v a r i o u s e p i d e r m a l p r o t e i n s are given for c o m p a r i s o n . T h e r e a p p e a r s to be a similarity between f r a c t i o n A , the u r e a - e x t r a c t e d p r o t e i n s isolated by C a r r u t h e r s a n d " p r e k e r a t i n " i s o l a t e d by M a t o l t s y [2, 5]. T h e histidine c o n t e n t o f fraction A was f o u n d to be 2.1 p e r 100 a m i n o a c i d residues, less t h a n o n e - t h i r d the n u m b e r o f
TABLE II AMINO ACID ANALYSIS OF EPIDERMAL PROTEINS Urea-extracted protein fraction A was subjected to amino acid analysis as described in the text. Results are expressed as residues/100 residues and represent the average of duplicate analyses. Results for other epidermal proteins are given for comparison. Amino acid
Aspartic acid Threonine Serine Glutamic acid Proline Glycine Alanine Half-cystine Valine Methionine Isoleucine Leueine Tyrosine Phenylalanine Lysine Histidine Arginine
Human urea extract protein A 9.5 4.9 8.6 13.6 4.9 12.3 7.7 1.2 4.7 2.4 2.7 8.6 2.4 3.5 6.1 2.1 4.9
Mouse urea extract protein (Carruthers [21)
Bovine prekeratin (Skerrow [15])
10.2 5.9 6.6 15.0 4.6 11.2 6°8 1.7 5.5 1.2 3.6 7.9 2.7 3.4 6.2 2.3 5.5
9.0 3.5 9.4 14.4 1.5 15.7 6.8 0.9 5.1 2.0 4.1 8.8 2.8 3.8 5.2 0.9 5.9
Bovine prekeratin (Baden [16])
Rat urea extract (Bernstein [3])
Human histidinerich protein (Bernstein [3])
9.1 4.0 11.1 14.1 1.4 16.4 6.7 0.6 4.0 1.3 3.5 9.2 2.8 3.6 5.1 1.0 6.1
8.3 4.8 11.2 14.8 -15.8 7.6 -3.9 -3.1 6.2 2.7 2.6 4.6 2.5 6.8
15.4 8.3 6.0 12.o 16.8
10.1 0 1.8 0 1.2 1.6 4.3 0.6 1.3 8.3 10.0
113 estimated imidazole groups by the Pauly reaction. The difference indicates the presence of a non-histidine Pauly-reactive material. In addition to the standard amino acids a ninhydrin-positive material from the acid hydrolyzate of fraction A was eluted immediately before lysine on the short column. It therefore carried a positive charge under the conditions used but has yet to be identified. When the acid hydrolyzate of fraction A was separated by paper chromatography and sprayed with Pauly reagent no urocanic acid was detectable. Since fraction A gave a positive orcinol reaction, the possibility of the presence of RNA or oligonucleotides was investigated by mild alkaline hydrolysis of fraction A and separation by high voltage paper electrophoresis. Ultraviolet photography of the electrophoretogram did not reveal any of the nucleotides of the four common RNA bases. However, one ultraviolet absorbing spot was detectable. After treatment of fraction A with 0.1 M HC1 a similar spot was obtained. Simply dissolving fraction A in 0.02 M sodium carbonate did not release this material. In each case samples had to be freshly prepared since the material appeared to be very labile. It travelled towards the anode at a faster rate than nucleotides or nucleoside triphosphates. Under similar conditions urocanic acid travelled towards the cathode. Therefore neither urocanic acid nor oligonucleotides are present in fraction A. The ultraviolet absorbing material could also be localized as a yellow spot following spraying of the electrophoretogram with Pauly reagent. It has not yet been identified. SDS-polyacrylamide gel electrophoresis of fraction A, incubated with and without fl-mercaptoethanol revealed similar complex patterns of numerous sub-units with a range of molecular weights from 160 000 to 14 000. The molecular weight of the two main sub-units of fraction A in the presence of fl-mercaptoethanol were found, from the average of duplicate experiments, to be 77 000 and 63 000, whereas in the absence of fl-mercaptoethanol the bands were less resolved with molecular weights of 76 000 and 69 000. Polyacrylamide gel electrophoresis was also carried out in the absence of SDS. The samples of fraction A were dissolved in 0.02 M sodium carbonate or 6 M urea. The patterns obtained after scanning the stained gels at 265 nm are shown in Figs. 2a and 2b, respectively. The results indicate that fraction A consists of aggregates since the high molecular weight band obtained from a solution in 0.02 M sodium carbonate is broken down in the presence of 6 M urea. DISCUSSION Although fibrous protein can be extracted from human epidermal tissue, the nature of the protein molecule "in situ" is still uncertain and the mechanism by which it is finally stabilised remains unknown. The protein in the present study formed more than 75 70 of the total protein extracted from human whole epidermis by urea. The amino acid composition, the aggregate molecular weight value of 600 000 and the solubility at extremes of pH are characteristics of fraction A which ale similar to prekeratin as isolated by Skerrow [15] and Matoltsy [5] from bovine snout epidermis by citric acid/sodium citrate buffer, pH 2.6. This is in agreement with work by Carruthers [6] who presented immuno-
114 (a)
(b)
I
0
2
4
O 2 Distonce migrated (cm)
I
4
Fig. 2. Polyacrylamide gel electropboresis of fraction A. The main protein fraction after gel filtration of the urea extract was lyophilized. Electrophoresis was carried out on 5 ~ gels, 3 % cross-linking, in a discontinuous buffer system. Gels were stained with amido black and scanned at 265 nm. (a) Fraction A in 0.02 M sodium carbonate, (b) Fraction A in 6 M urea.
logical evidence that bovine snout epidermal proteins, whether extracted by urea or citrate buffer p H 2.6, are related by c o m m o n identical components. Polyacrylamide gel electrophoresis of fraction A in the presence of SDS and /3-mercaptoethanol revealed two main bands of molecular weights 77 000 and 63 000. This is comparable with work by Skerrow [17] who found that bovine prekeratin consisted of two chains with molecular weights of 72 000 and 60 000. However, unlike Skerrow, numerous less prominent sub-units were also detectable some of which may be due to contaminants of other proteins removed. Therefore, although there are many similarities between fraction A and bovine prekeratin, extraction with urea results in a heterogeneous preparation. The difference in the pattern obtained in the absence of fl-mercaptoethanol indicates the presence of interchain disulphide linkages, in contrast to Steinert [18] who found that urea-extractable protein from bovine epidermis whether from living cell layers or stratum corneum, gave similar patterns with and without fl-mercaptoethanol. He concluded that cysteine residues in the protein from living cell layers of bovine epidermis were in their thiol form whereas in the stratum corneum they formed intrachain disulphide bonds rather than interchain linkages. The protein fraction we have isolated from human epidermis has several unusual features which have not been reported for bovine prekeratin, namely associated copper and an ultraviolet-absorbing, Pauly-reactive component, both of which are released from the large molecule by treatment with dilute acid.
115 On the basis of studies of urea-extractable proteins from both human and rat epidermis, a model has been postulated by Bernstein [3] for the structure of the large protein obtained and an important role in the keratinisation process assigned to it. In the model the protein is shown "in situ" with a urocanic acid molecule covalently linked at one end. Copper is associated through imidazole groups linking two parts of the large molecule, a histidine-rich portion and another section of the molecule reported to be insoluble in 0.1 M perchloric acid. It has also been suggested that the same protein molecule is present in larger quantities in keratohyalin granules but the granules are not extracted from the whole epidermis by urea [19]. However, this view would appear to be in contradiction to the idea, commonly held, that keratohyalin is largely non-fibrous protein. We have isolated and identified unbound urocanic acid in an aqueous extract of epidermis and found that the free urocanic acid becomes radioactive after [3H]histidine incorporation (unpublished observations). However, paper chromatography of the acid hydrolyzate of radioactive fraction A, showed that the radioactivity was localised entirely in the histidine spot. In addition, although an ultraviolet-absorbing, Pauly-reactive component has been found in the present study under conditions reported by Bernstein to release urocanic acid, our high voltage paper electrophoresis study has shown that it is not urocanic acid. Alkaline hydrolysis of fraction A did not reveal any of the common nucleotides. The orcinol reaction obtained therefore tends to indicate the presence of carbohydrate rather than R N A or nucleotides. ACKNOWLEDGMENT We thank Unilever Ltd., Port Sunlight, Cheshire for a grant to J. Baynes.
REFERENCES 1 2 3 4 5
Rudall, K. M. (1952) Adv. Protein Chem. 7, 253-290 Carruthers, C. (1974) Oncoiogy 30, 125-133 Bernstein, 1. A. (1970) J. Soc. Cosmet. Chem. 21,583-594 Matoltsy, A. G. (1964) Nature 201, 1130-1131 Matoltsy, A. G. (1965) in Biology of the Skin and Hair Growth (Lyne, A. G. and Short, B. F., eds.), pp. 291-305, Angus and Robertson, Sydney 6 Carruthers, C. (1973) Br. J. Dermatol. 89, 477-485 7 Brotherton, J. (1969) J. Invest. Dermatol. 52, 78-88 8 Marks, R., Fukui, F. and Halprin, K. (1971) Br. J. Dermatol. 84, 453-460 9 Lowry, O. H., Rosebrough, N. J., Farr, A. L. and Randall, R. J. (1951) J. Biol. Chem. 193, 265275 10 Markham, R. and Smith, J. D. (1952) Biochem. J. 52, 552-557 11 Dunker, A. K. and Rueckert, R. R. (1969) J. Biol. Chem. 224, 5074-5080 12 Weber, K. and Osborn, M. (1969) J. Biol. Chem. 244, 4406-4412 13 Reaven, E. P. and Cox, A. J. (1965) J. Invest. Dermatol. 15, 422-431 14 Sugawara, K. and Bernstein, I. A. (1971) Biochim. Biophys. Acta 238, 129-138 15 Skerrow, D. (1972) Bioehim. Biophys. Acta 257, 398-403 16 Baden, H. P., Gifford, A. M. and Goldsmith, L. A. (1971) J. Invest. Derrnatol. 56, 446 A.A.9 17 Skerrow, D. (1974) Biochem. Biophys. Res. Commun. 59, 1311-1316 18 Steinert, P. M. (1975) Biochem. J. 149, 39-48 19 Sibrack, L. A., Gray, R. H. and l~ernstein, I. A. (1974) J. Invest. Dermatol. 62, 394--405
© Elsevier Scientific Publishing Company, Amsterdam -- Printed in The Netherlands BBA 37388 UREA-EXTRACTABLE PROTEIN FROM H U M A N EPIDERMIS
J. BAYNES and M. LEVINE Department of Biochemistry, University of Manchester, Manchester MI3 9PL ( U.K.)
(Received December 3rd, 1975)
SUMMARY The major protein fraction extracted from human epidermis by urea has the following characteristics: (1) A molecular weight of approximately 600 000. (2) A copper content of 0.03 ~ giving a minimum molecular weight of 212 000. (3) A non-protein moiety indicated by (a) protein/dry weight ratio (b) anomalous ultraviolet absorption spectrum (c) high-voltage electrophoresis after mild alkaline or acid treatment (d) an excess of imidazole groups by Pauly reaction compared with the number of histidine residues by amino acid analysis. The non-protein component is not urocanic acid or RNA as has been reported (Bernstein, I. A. (1970), J. Soc. Cosmet. Chem. 21, 583-594 and Sibrack, L. A., Gray, R. H. and Bernstein, I. A. (1974), J. Invest. Dermatol. 62, 394-405). (4) The presence of interchain disulphide linkages indicated by polyacrylamide gel electrophoresis in sodium dodecyl sulphate with and without fl-mercaptoethanol. (5) Polyacrylamide gel electrophoresis in presence and absence of 6 M urea suggests that aggregates are broken down in urea.
INTRODUCTION The study of keratinisation in epidermis has been hampered by the difficulties in isolating the insoluble epidermal proteins involved in the process. The traditional method for extraction of the fibrous protein has been by concentrated urea. Rudall [1] in 1952 gave the name "epidermin" to the urea-soluble protein obtained from cow snout epidermis. Epidermin gave an a-type X-ray diffraction pattern, was classified with the a-keratin group and identified as a fibrous protein. Similar extractions have been made on other mammalian species [2, 3]. Citric acid/sodium citrate buffer, pH 2.6, has also proved successful for the isolation of fibrous protein from bovine nose epidermis and the multichain molecule obtained was termed "prekeratin" [4, 5]. There is now evidence indicating that proAbbreviation: SDS, sodium dodecyl sulphate.
108 teins extracted either by urea or by citrate buffer are related by common identical components [6] but their nature and chemical structure are still poorly understood. In a recent model postulated by Bernstein [3], the urea-extractable protein is depicted "in situ" as covalently bonded to a urocanic acid molecule at one end and copper is shown as associated through the imidazole residue. Our observations reported here have been made on the main protein fraction isolated from human epidermis by urea. We have confirmed Bernstein's finding that the large protein aggregate isolated is associated with copper. In addition we have isolated free urocanic acid from skin and identified it. However, our evidence indicates that there is no bound urocanic acid associated with the protein. Another unknown component has been separated from the protein by either mild acid or alkaline treatment. It is not urocanic acid or any of the common nucleic acid components. MATERIALS AND METHODS Hanks' balanced salt solution and Eagle's minimal essential medium were obtained from Wellcome Reagents Ltd. The Radiochemical Centre, Amersham supplied L-[2,5-3H2]histidine, 58 Ci/mmol.
Preparation of tissue Samples of human skin were obtained from the lower thigh region after leg amputations using a dermatome set at a thickness of 0.38 mm. The tissue was placed on gauze moistened with Hanks' balanced salt solution and stored at --20 °C or at 4 °C when required for incorporation studies. Labelled histidine was incorporated into skin using a method adapted from Brotherton [7] and Marks et al. [8]. The skin was floated, epidermis upwards in Eagle's minimal essential medium containing antibiotics, 1 ~ (w/v) glutamine and L-[2,5-3H2]histidine (1 ffCi/ml). Incubation was at 37 °C and 75 ~ humidity for 24 h in a gas mixture of 9 5 ~ air and 5 ~ carbon dioxide. The epidermis was separated from the dermis in early experiments by heat treatment, 60 °C for 2 rain. In later experiments the skin was frozen over dry ice and tile epidermis removed by scraping with a scalpel (Dr. Sprott, Unilever Research. Colwerth House, Bedford, personal communication).
Preparation of urea-extractable proteins The epidermal mince or scrapings were homogenized in 8 M urea, 0.2 M Tris-HC1 buffer (pH 8.5). Debris was removed by low speed centrifugation and the supernatant dialyzed against 0.1 M ammonium hydroxide at 4 °C overnight. Dialysis was continued for 48 h against 0.01 M ammonium hydroxide after which any aggregates formed were removed by centrifugation at 40 000 x g for 20 min at 0 °C and the supernatant was lyophilized. In certain experiments the urea extract was treated with 0.1 M perchloric acid. The effect of perchloric acid on the protein extracted was investigated.
Isolation and characterization of urea-extractable protein The dried urea extract was dissolved in 0.02 M sodium carbonate, applied to a sephadex G-200 column (bed dimensions 2 × 30 cm) and eluted with 0.01 M am-
109
monium hydroxide. Each fraction was tested for absorbance at 280 nm, protein by the method of Lowry [9], RNA by the orcinol reaction and radioactivity after incorporation of L-[2,5-3H2]histidine. The main protein fractions were pooled and lyophilized for further tests.
High voltage paper electrophoresis High voltage paper electrophoresis was carried out following the method of Markham and Smith [10]. The apparatus consisted of two compartments with platinum electrodes, and a central tank filled with carbon tetrachloride for cooling purposes. Samples were applied to the centre of a strip of paper 10 x 80 cm (Whatman No. 1) which was then moistened with the electrophoretic buffer: pyridine (10 ml)/acetic acid (92 ml)/water (1.6 litres) (pH 3.5). A voltage of 1500 V was applied for 1 h. After drying, pyridine was washed from the electrophoretogram using 95 ethanol (100 ml)/ether (900 ml), containing a few drops of "880" ammonia. Ultraviolet-absorbing materials were localized by ultraviolet photography.
Polyacrylamide gel electrophoresis The method used for polyacrylamide gel electrophoresis in SDS was based on Dunker and Rueckert [1 I] and Weber and Osborn [12]. Standard protein samples were prepared at a concentration of 1-2 mg/ml by incubation for 1-2 h at 37 °C in 0.01 M sodium phosphate buffer (pH 7.0), containing 1 ~ (w/v) SDS and 1 ~ (v/v) fl-mercaptoethanol. Samples of 100 ffl containing bromophenol blue as a marker, were applied to 7.5 ~ polyacrylamide gels, 3 ~ cross-linking, and electrophoresis was carried out in a continuous system of 0.1 M sodium phosphate (pH 7.0) containing 0.1 ~ (w/v) SDS. An initial current of 5 mA/tube was applied until the samples entered the gels and was then increased to 8 mA/tube for 3-4 h. Gels were cut through the bromophenol blue region and stained for protein with 1 ~ (w/v) amido black in 7 ~ (v/v) acetic acid. Destaining was achieved by continuous elution with 7 ~ (v/v) acetic acid. Relative mobilities were calculated as the ratio of the distance travelled by the protein to the length of the gel. For molecular weight determinations ureaextractable protein incubated in SDS with and without fl-mercaptoethanol and standard proteins prepared as above were run at the same time. In addition, polyacrylamide gel electrophoresis was carried out in the absence of SDS using 5 ~ polyacrylamide gels, 3 ~ cross-linking and a discontinuous buffer system. The cathode compartment contained 25 mM imidazole/acetate buffer (pH 7.4) whereas the gel and anode contained 50 mM sodium chloride/25 mM imidazole-HC1 buffer (pH 7.4). Samples of urea-extractable protein were prepared in 0.02 M sodium carbonate or 6 M urea. A current of 5 mA/tube was applied and the gels run for 1-2 h. The staining procedure was as previously described. Gels were scanned on a Joyce Loebl gel scanner at a wavelength of 265 nm.
Amino acid analysis 'Urea-extractable protein was hydrolyzed in 6 M HCI (re-distilled) in sealed tubes under nitrogen overnight at 100 °C. Analyses were performed on a Jeol JLC6AH amino acid analyzer.
t10
The Pauly reaction tests Pauly reagent was freshly prepared for each experiment according to Reaven and Cox [13]. It was used as a spray for qualitative purposes. Estimation ofimidazole content was carried out by adding 3.6 ml Pauly reagent to 0.4 ml sample. The colour produced was measured against a reagent blank at 520 nm. Histidine was used as a standard.
Paper chromatography Paper chromatography of amino acids was carried out in a single dimension using 2-propanol/acetone/1 M HCI (60:15:25, v/v) as a developing solvent. Histidine, tyrosine and urocanic acid were localized using Pauly reagent spray. RESULTS The elution profile of urea-extracted protein after Sephadex G-200 chromatography (Fig. 1) revealed one main protein peak (fraction A), with a molecular weight of about 600 000. Approximately 75 % of the total Lowry protein in the urea extract appears in fraction A and its position was indicated by absorbance at 280 nm, radioactivity after labelled histidine incorporation, estimation of Lowry protein, and by orcinol reaction. Fractions in this region were pooled and lyophilized. A shoulder
30
OOO Cl:)m / rnl
i
i
o 20
I
o51-
I; 7
,
\
,,
"A
OOO
dlO o o o
-~.
0 10
30
50 O.O1 iv1 N H 4 O H
70
90
( rrtl )
Fig. 1. Gel filtration of the urea extract. After incorporation of [aH]histidine,'.the~epidermis was homogenized in 8 M urea. The urea extract was dialyzed, lyophilized and chromatographed on a sephadex G-200 column (2 x 30 cm), equilibrated and eluted with 0.01 M NH4OH. O---O cpm/ml, 0 - - 0 absorbance at 280 nm, A---A absorbance at 700 nm after the Lowry assay on each fraction, J,--A absorbance at 660 nm after the orcinol reaction on each fraction.
111 of lower molecular weight protein was also detectable. It had a similar anomalous ultraviolet spectrum, A2ao/Lowry protein ratio and the same specific activity after histidine incorporation. One other significant peak was eluted in the low molecular weight region. It was detected by absorbance at 280 nm and reaction with orcinol but was largely non-protein in nature by the Lowry method. Treatment of the urea extract with 0.1 M perchloric acid appeared to split fraction A into smaller units since the main protein in perchloric acid-insoluble material had a molecular weight of 160 000 and in perchloric acid-soluble material 50 000. In addition, copper and ultraviolet absorbing material was released both of which became dialyzable. Following in vitro incorporation of tritiated histidine into whole skin for 24 h it was found that the specific activities of the urea extract and perchloric acid-soluble material were similar (Table I). However, the specific activity of perchloric acidinsoluble material was 67 7oo higher. These results are not consistent with those obtained by Bernstein and co-workers who found that a histidine-rich protein was solubilized by treatment of the urea extract with 0.1 M perchloric acid [14].
TABLE I INCORPORATION OF [3H]HISTIDINE INTO EPIDERMAL EXTRACTS [aH]Histidine (5.27 Ci/mol) was incorporated into whole skin for 24 h. The epidermis was extracted with 8 M urea and the dried urea extract treated with 0.1 M perchloric acid. Specific activities of the crude urea extract, perchloric acid-insoluble material and perchloric acid-soluble material were measured. Fractionation of the crude extracts by Sephadex chromatography revealed one main protein fraction in each case. The estimated molecular weights and specific activities of the fractionated proteins are also given for comparison. PCA, perchloric acid. Extract
Specific activity of crude extract (a) dpm/mg protein
Main protein peak after fractionation
(b) Molecular nmol weight histidine/mg ( × 10-a) protein/24 h
Specificactivity (a) (b) dpm/mg nmol protein histidine/mg ( × l0 -a) protein/24h
9.23 15.4 9.23
106 140 90
( × 10-3) Urea extract PCA-insolublematerial PCA-solublematerial
108 180 108
600 160 50
9.06 12.0 7.69
Measurement of the ultraviolet absorption spectrum of fraction A in 0.01 M ammonium hydroxide showed a maximum at 266 nm. This is uncharacteristic of protein indicating the presence of an additional ultraviolet-absorbing moiety in fraction A. Protein estimations carried out by several methods were compared with dry weight measurements. The ratio of protein to dry weight by A,4ns_225 nm, Lowry and A2a0 nm/A260nm measurements were 0.47, 0.52 and 0.59, respectively. Bovine serum albumin was used as a reference standard in the Lowry protein estimations. The
112 a m i n o acid c o m p o s i t i o n o f bovine s e r u m a l b u m i n a n d f r a c t i o n A m a y n o t be similar e n o u g h to p r o d u c e a c c u r a t e results. The o t h e r m e t h o d s used involved the m e a s u r e m e n t o f a b s o r b a n c e in the u l t r a v i o l e t region. Since f r a c t i o n A has been s h o w n to have a n u n c h a r a c t e r i s t i c u l t r a v i o l e t s p e c t r u m o f p r o t e i n these also c o u l d be subject to error. H o w e v e r , the three m e t h o d s are in fairly close a g r e e m e n t suggesting that f r a c t i o n A c o n t a i n s n o n - p r o t e i n m a t e r i a l which w o u l d be i n c l u d e d in the d r y weight b u t n o t in the p r o t e i n e s t i m a t i o n s . It was, therefore, necessary t h a t all calculations involving the c o n c e n t r a t i o n o f p r o t e i n s h o u l d specify the m e t h o d used. C o p p e r analysis o f l y o p h i l i z e d f r a c t i o n A by a t o m i c emission s p e c t r o s c o p y gave a c o p p e r c o n t e n t o f 0.0370 (weight c o p p e r / d r y weight). This value gives a m i n i m u m m o l e c u l a r weight o f the m o l e c u l e as 212 000. T h e i m i d a z o l e c o n t e n t o f f r a c t i o n A , using the P a u l y r e a c t i o n with a histidine s t a n d a r d , was f o u n d to be 1.85 70 (weight i m i d a z o l e / w e i g h t L o w r y protein) which is e q u i v a l e n t to 6.7 i m i d a z o l e g r o u p s p e r 100 a m i n o acid residues. The a m i n o acid analysis o f f r a c t i o n A is s h o w n in T a b l e I1. O t h e r results obt a i n e d for v a r i o u s e p i d e r m a l p r o t e i n s are given for c o m p a r i s o n . T h e r e a p p e a r s to be a similarity between f r a c t i o n A , the u r e a - e x t r a c t e d p r o t e i n s isolated by C a r r u t h e r s a n d " p r e k e r a t i n " i s o l a t e d by M a t o l t s y [2, 5]. T h e histidine c o n t e n t o f fraction A was f o u n d to be 2.1 p e r 100 a m i n o a c i d residues, less t h a n o n e - t h i r d the n u m b e r o f
TABLE II AMINO ACID ANALYSIS OF EPIDERMAL PROTEINS Urea-extracted protein fraction A was subjected to amino acid analysis as described in the text. Results are expressed as residues/100 residues and represent the average of duplicate analyses. Results for other epidermal proteins are given for comparison. Amino acid
Aspartic acid Threonine Serine Glutamic acid Proline Glycine Alanine Half-cystine Valine Methionine Isoleucine Leueine Tyrosine Phenylalanine Lysine Histidine Arginine
Human urea extract protein A 9.5 4.9 8.6 13.6 4.9 12.3 7.7 1.2 4.7 2.4 2.7 8.6 2.4 3.5 6.1 2.1 4.9
Mouse urea extract protein (Carruthers [21)
Bovine prekeratin (Skerrow [15])
10.2 5.9 6.6 15.0 4.6 11.2 6°8 1.7 5.5 1.2 3.6 7.9 2.7 3.4 6.2 2.3 5.5
9.0 3.5 9.4 14.4 1.5 15.7 6.8 0.9 5.1 2.0 4.1 8.8 2.8 3.8 5.2 0.9 5.9
Bovine prekeratin (Baden [16])
Rat urea extract (Bernstein [3])
Human histidinerich protein (Bernstein [3])
9.1 4.0 11.1 14.1 1.4 16.4 6.7 0.6 4.0 1.3 3.5 9.2 2.8 3.6 5.1 1.0 6.1
8.3 4.8 11.2 14.8 -15.8 7.6 -3.9 -3.1 6.2 2.7 2.6 4.6 2.5 6.8
15.4 8.3 6.0 12.o 16.8
10.1 0 1.8 0 1.2 1.6 4.3 0.6 1.3 8.3 10.0
113 estimated imidazole groups by the Pauly reaction. The difference indicates the presence of a non-histidine Pauly-reactive material. In addition to the standard amino acids a ninhydrin-positive material from the acid hydrolyzate of fraction A was eluted immediately before lysine on the short column. It therefore carried a positive charge under the conditions used but has yet to be identified. When the acid hydrolyzate of fraction A was separated by paper chromatography and sprayed with Pauly reagent no urocanic acid was detectable. Since fraction A gave a positive orcinol reaction, the possibility of the presence of RNA or oligonucleotides was investigated by mild alkaline hydrolysis of fraction A and separation by high voltage paper electrophoresis. Ultraviolet photography of the electrophoretogram did not reveal any of the nucleotides of the four common RNA bases. However, one ultraviolet absorbing spot was detectable. After treatment of fraction A with 0.1 M HC1 a similar spot was obtained. Simply dissolving fraction A in 0.02 M sodium carbonate did not release this material. In each case samples had to be freshly prepared since the material appeared to be very labile. It travelled towards the anode at a faster rate than nucleotides or nucleoside triphosphates. Under similar conditions urocanic acid travelled towards the cathode. Therefore neither urocanic acid nor oligonucleotides are present in fraction A. The ultraviolet absorbing material could also be localized as a yellow spot following spraying of the electrophoretogram with Pauly reagent. It has not yet been identified. SDS-polyacrylamide gel electrophoresis of fraction A, incubated with and without fl-mercaptoethanol revealed similar complex patterns of numerous sub-units with a range of molecular weights from 160 000 to 14 000. The molecular weight of the two main sub-units of fraction A in the presence of fl-mercaptoethanol were found, from the average of duplicate experiments, to be 77 000 and 63 000, whereas in the absence of fl-mercaptoethanol the bands were less resolved with molecular weights of 76 000 and 69 000. Polyacrylamide gel electrophoresis was also carried out in the absence of SDS. The samples of fraction A were dissolved in 0.02 M sodium carbonate or 6 M urea. The patterns obtained after scanning the stained gels at 265 nm are shown in Figs. 2a and 2b, respectively. The results indicate that fraction A consists of aggregates since the high molecular weight band obtained from a solution in 0.02 M sodium carbonate is broken down in the presence of 6 M urea. DISCUSSION Although fibrous protein can be extracted from human epidermal tissue, the nature of the protein molecule "in situ" is still uncertain and the mechanism by which it is finally stabilised remains unknown. The protein in the present study formed more than 75 70 of the total protein extracted from human whole epidermis by urea. The amino acid composition, the aggregate molecular weight value of 600 000 and the solubility at extremes of pH are characteristics of fraction A which ale similar to prekeratin as isolated by Skerrow [15] and Matoltsy [5] from bovine snout epidermis by citric acid/sodium citrate buffer, pH 2.6. This is in agreement with work by Carruthers [6] who presented immuno-
114 (a)
(b)
I
0
2
4
O 2 Distonce migrated (cm)
I
4
Fig. 2. Polyacrylamide gel electropboresis of fraction A. The main protein fraction after gel filtration of the urea extract was lyophilized. Electrophoresis was carried out on 5 ~ gels, 3 % cross-linking, in a discontinuous buffer system. Gels were stained with amido black and scanned at 265 nm. (a) Fraction A in 0.02 M sodium carbonate, (b) Fraction A in 6 M urea.
logical evidence that bovine snout epidermal proteins, whether extracted by urea or citrate buffer p H 2.6, are related by c o m m o n identical components. Polyacrylamide gel electrophoresis of fraction A in the presence of SDS and /3-mercaptoethanol revealed two main bands of molecular weights 77 000 and 63 000. This is comparable with work by Skerrow [17] who found that bovine prekeratin consisted of two chains with molecular weights of 72 000 and 60 000. However, unlike Skerrow, numerous less prominent sub-units were also detectable some of which may be due to contaminants of other proteins removed. Therefore, although there are many similarities between fraction A and bovine prekeratin, extraction with urea results in a heterogeneous preparation. The difference in the pattern obtained in the absence of fl-mercaptoethanol indicates the presence of interchain disulphide linkages, in contrast to Steinert [18] who found that urea-extractable protein from bovine epidermis whether from living cell layers or stratum corneum, gave similar patterns with and without fl-mercaptoethanol. He concluded that cysteine residues in the protein from living cell layers of bovine epidermis were in their thiol form whereas in the stratum corneum they formed intrachain disulphide bonds rather than interchain linkages. The protein fraction we have isolated from human epidermis has several unusual features which have not been reported for bovine prekeratin, namely associated copper and an ultraviolet-absorbing, Pauly-reactive component, both of which are released from the large molecule by treatment with dilute acid.
115 On the basis of studies of urea-extractable proteins from both human and rat epidermis, a model has been postulated by Bernstein [3] for the structure of the large protein obtained and an important role in the keratinisation process assigned to it. In the model the protein is shown "in situ" with a urocanic acid molecule covalently linked at one end. Copper is associated through imidazole groups linking two parts of the large molecule, a histidine-rich portion and another section of the molecule reported to be insoluble in 0.1 M perchloric acid. It has also been suggested that the same protein molecule is present in larger quantities in keratohyalin granules but the granules are not extracted from the whole epidermis by urea [19]. However, this view would appear to be in contradiction to the idea, commonly held, that keratohyalin is largely non-fibrous protein. We have isolated and identified unbound urocanic acid in an aqueous extract of epidermis and found that the free urocanic acid becomes radioactive after [3H]histidine incorporation (unpublished observations). However, paper chromatography of the acid hydrolyzate of radioactive fraction A, showed that the radioactivity was localised entirely in the histidine spot. In addition, although an ultraviolet-absorbing, Pauly-reactive component has been found in the present study under conditions reported by Bernstein to release urocanic acid, our high voltage paper electrophoresis study has shown that it is not urocanic acid. Alkaline hydrolysis of fraction A did not reveal any of the common nucleotides. The orcinol reaction obtained therefore tends to indicate the presence of carbohydrate rather than R N A or nucleotides. ACKNOWLEDGMENT We thank Unilever Ltd., Port Sunlight, Cheshire for a grant to J. Baynes.
REFERENCES 1 2 3 4 5
Rudall, K. M. (1952) Adv. Protein Chem. 7, 253-290 Carruthers, C. (1974) Oncoiogy 30, 125-133 Bernstein, 1. A. (1970) J. Soc. Cosmet. Chem. 21,583-594 Matoltsy, A. G. (1964) Nature 201, 1130-1131 Matoltsy, A. G. (1965) in Biology of the Skin and Hair Growth (Lyne, A. G. and Short, B. F., eds.), pp. 291-305, Angus and Robertson, Sydney 6 Carruthers, C. (1973) Br. J. Dermatol. 89, 477-485 7 Brotherton, J. (1969) J. Invest. Dermatol. 52, 78-88 8 Marks, R., Fukui, F. and Halprin, K. (1971) Br. J. Dermatol. 84, 453-460 9 Lowry, O. H., Rosebrough, N. J., Farr, A. L. and Randall, R. J. (1951) J. Biol. Chem. 193, 265275 10 Markham, R. and Smith, J. D. (1952) Biochem. J. 52, 552-557 11 Dunker, A. K. and Rueckert, R. R. (1969) J. Biol. Chem. 224, 5074-5080 12 Weber, K. and Osborn, M. (1969) J. Biol. Chem. 244, 4406-4412 13 Reaven, E. P. and Cox, A. J. (1965) J. Invest. Dermatol. 15, 422-431 14 Sugawara, K. and Bernstein, I. A. (1971) Biochim. Biophys. Acta 238, 129-138 15 Skerrow, D. (1972) Bioehim. Biophys. Acta 257, 398-403 16 Baden, H. P., Gifford, A. M. and Goldsmith, L. A. (1971) J. Invest. Derrnatol. 56, 446 A.A.9 17 Skerrow, D. (1974) Biochem. Biophys. Res. Commun. 59, 1311-1316 18 Steinert, P. M. (1975) Biochem. J. 149, 39-48 19 Sibrack, L. A., Gray, R. H. and l~ernstein, I. A. (1974) J. Invest. Dermatol. 62, 394--405
