Skin Proteasomes (High - Molecular-Weight Protease): Purification, Enzymologic Properties, Gross Structure, and Tissue Distribution

Yasushi Suga, Kenji Takamori, and Hideoki Ogawa Department of Dermatology, School of Medicine, Juntendo University, Tokyo, Japan

Proteasomes (high - molecular-weight protease) were puri­fied from rat skin, and their enzymologic properties, gross structure, and tissue distribution were investigated. Skin pro­teasomes were purified by successive (NH4)2S04 fractiona­tion and by phenyl Sepharose CL-4B and HPLC gel filtration chromatography. The molecular weights of the proteasomes were estimated from gel filtration to be 750 kD. On sodium dodecylsu lfate - ~olya~ryla~ide gel electrophoresis, th~ ~u­rified enzymes dissoCiated mto several bands, the maJonty falling into the range of 36 - 20 kD . Two-dimensional elec­trophoretric analysis demonstrated approximately 10-15 separate protein spots with pi values varying between 3 and 10. As analyzed by electron microscopy, the gross structure of the enzymes showed an almost symmetrical ring-shaped par­ticle with a small hole in the center. Succynyl-leucyl-Ieucyl­valyl-tyrosine-4-methylcoumaryl-7 -amide, a fluorogenic

P roteasomes are nonlysosomal enzymes with high mo­lecular weights (Mr = 750 kD). They have been found as multicatalytic proteinase complexes [1) and have been shown to play an essential role in an energy-dependent non lysosomal proteolytic pathway responsible for the

selective removal of unnecessary proteins generated in cell s [2 - 6). Proteasomes are widely distributed in various eukaryotic cells and tissues ranging from humans to yeast [7,8). Moreover, recent studies have reported their localization in the nucleus and cytoplasm [9,10). This wide intracellular distribution may be of significance, because the physiologic functions of proteasomes are still unknown. Previ­ous ly, Tanaka et al (7) showed the existence of proteasomes in rat skin by quantitative enzyme immunoassay. In this study, we report the purification of rat skill proteasomes and describe some of their properties and their gross structure from electron microscopy, as wel l as their tissue distribution as measured by enzymologic and immunologic techniques.

MATERIALS AND METHODS

Materials Ten male Wistar rats (7 weeks old, 200-250 g) were obtained from Takasugi Institutc (Tokyo). Phenyl Sepharose CL-4B and molccular weight marker proteins were purchased from Pharmacia Co. High-per­formance liquid chromatography (HPLC) was carried out using a TSKG3000SWXL or TSK4000SW gel filtration column (Toyo Soda,

Manuscript received July 14, 1992; accepted for publication May 3, 1993. Reprint rcqucsts to: Yasushi Suga, Department of Dermatology, School

of Medicinc, Juntendo University, 2-1-1 Hongo, Bunkyo-ku, Tokyo 11 3, Japan.

Abbreviations: AMC, 7-amino-4-mcthylcoumarin; SLLVY-MCA, suc­cyn y l-leucy l-lcucyl-val y I-tyrosine-4-methylcoumary 1-7 -amide.

substrate for serine proteinases, demonstrated the highest ac­tivity in terms of substrate specificity. Sodium dodecylsulfate, Ca++, and some free fatty acids activated enzyme activity. Activity was inhibited by diisopropylfluorophosphate, leu­peptin, N-ethylmaleimide, iodoacetamide, and chymostatin. These results show that both serine and cysteine residues are related to the enzyme activity of proteasomes. Total and specific enzyme activities in the epidermis were, respectively, 10 and 20 times higher than in the dermis. Immunohisto­chemical studies utilizing the avidin-biotin complex method with monoclonal antibody revealed that the enzyme is dis­tributed throughout the epidermis. These findings indicate the epidermal localization of proteasomes. Key words: non­lysosomal enzyme/rat skin/multicatalytic proteinase com­plex.] Invest DermatoI101 :346-351, 1993

Tokyo) connected to an HPLC system (J asco, Tokyo). The fluo rogenic peptide, succyny l-leucyl-leucy I-val y I-tyrosine-4-methy lcoumaryl-7 -amide (SLL VY -MCA), and other synthetic peptide substrates , soybean trypsin in­hibitor, leupeptin, chymostatin, E-64, and pepstatin A, were obtained from the Peptide Research Institute (Osaka). Ethylenediaminetetraacetic acid, tetrasodium salt, and ethylene glycol bis-N ,N ,N ',N '-tetraacetic acid were from Wako-Junyaku Co. (Osaka). Diisopropylfluorophosphate, phenyl­methylsulfonylfluoride, N-ethylmaleimide, iodoacetamide, and mono­iodoacetic acid were obtained from Sigma. The pi marker proteins were from Oriental Yeast Co. (Tokyo). The avidin-biotin complex (ABC) kit was obtained from Vector Laboratories Inc. All other reagents were ofthe high­est quality commercially available.

Enzyme assay Proteasomes activity were assayed by measuring the fluo­rescence intensity of the 7-amino-4-methylcoumarin (AMC) liberated from the fluoro genic peptide, SLLVY-MCA, using a fluorescence spectropho­tometer (Hitachi model 204) at an excitation wavelength of 345 nm and an em mission wavelength of 440 nm. Ten microliters of enzyme solution was incubated at 37"C for 1 h in a medium containing 20 mM Tris-HCl buffer (pH 8.5), 0.1 mM SLL VY -MCA, and 0.03% sodium dodecylsulfate (SDS). The reaction was terminated by the addition of 100 ,ull0% SDS and 1.3 ml of 0.2 M Tris-HCl (pH 8.5). The amount of AMC liberated during the enzyme reaction was calculatcd using a standard curve for AMC. One unit of activity was defined as the amount of enzyme degrading 1 nmol of substrate per hour.

Purification

Preparatioll oj Crude Extracts: Crude extract was prepared from rat skin according to the method of Takamori el al [9] . Wister rat skin was minced finely with scissors after the removal of hair and subcutaneous tissue. The minced skin was homogenized in a Polytron Homogenizer (Kinematica, Switzerland) with 5 volumes of 5 mM KH 2PO./0.1 M KCI (PH 7.0). The

0022-202X/93/S06.00 Copyright © 1993 by The Society for Investigative Dermatology, Inc.

346

VOL. 101 , NO. 3 SEPTEMBER 1993

4oo..,.-......... _------------<;l·O:-----ip------r1.0

'c 2->-

2 200 (3 ell

'" E >­N

iii

......• ..............

-"-..

\

O .~··jll l " 10 20 30 ~O

frac tion number

2.0

0.8 0-0) N o Q.

0.6.Q "§ c Q) o

.4 § o C 'w <5

.2 0..

0-0"_ .1..0

70

Figure 1. Elution profile of proteasomes from Phenyl Sepharose CL-4B. Ammonium-sulfate-fractionated enzyme was applied to a Phenyl Sephar­ose CL-4B colunU1 and eluted with a KCl gradient (.). Each fraction (5.0 ml) was assayed for protein concentration (e) and enzyme activity (0). Aliquots of the fractIons lI1dlcated by the arrow were subjected to gel fi ltra­tion HPLC chromatography.

homogenate was centri fuged at 8000 X g for 30 min at 4 "C, and the result­ing supernatant was used for the next step.

Atnmonium Sulfate Fractionation Solid ammonium sul fate was added to the supernatant to 30% saturation (176 g/l) , and the suspension was cen trifugated at 12000 X g for 10 min. The ammonium sulfate concentra­tion of the supernatant was then increased to 60% sa turation (1 98 g/I), and the suspension was centrifuged at 27000 X g for 30 min. The pell.et from the second centrifugation was dissolved in equilibration buffer [5 mM KH

2PO./0.l M KCl (pH 7.0)] and dialyzed aga inst several changes of

equilibration buffer. A small a~,ount of precipitate was removed by centnfu­gati.on at 27000 X g for 30 m111 .

Phenyl Sepharose CL-4B Chromatography: Solid KCl was added to the en­zyme solution to a final concentration of2.5 M and the solutIOn was applIed

, to a Phenyl Sepharose CL-4D column (2.5 X 10 cm) prev iously equilibrated with 5 mM KH2PO./2.5 M KCl (pH 7.0) (Fig 1). The column was eluted with 800 ml of a 2.5 M - O.l M linear gradient ofKCl and then washed with equilibration buffer. The effiucnt was collected in 5.0-ml fractions. Active frac tions were collected and concentrated with 40 g/lOO 1111 solid ammo­nium sulfate. The mixtures were centrifuged for 30 min at 27000 X g and the supernatants were discarded. The pellets were then dissolved and ex-

I haus rively dialyzed aga inst equilibration buffer and utilized for the next step.

HPLC Gel Filtratioll Chromatography: The enzyme solution was subjected to gel filtration HPLC (TSKG3000SWXL) after small amounts of precIpI­tate were removed with Steradisc 0.45 11m (Kurabou, Japan) . The effiuent was collected in 0.5-ml fract ions. Active fractions were collected and used as the purified enzyme.

Other Procedures Polyacrylamide gel electrophoresis (PAGE) was per­formed in 6.0% gels according to the method ofOavis [12]. SDS-PAGE was carried out using 4 - 20% or 10-20% linear gradient gels containing 0.1 % SDS according to the method of Laemmli [13j. Two-dimensional PAGE was carried out according to the method of O'Farrell [14] . Prote111 was stained with Coomassie Brilliant Blue R-250. Protein was determined by measuring the absorbance at 280 nm or according to the method of Lowry et al [1 5] using bovine serum albumin (Sigma) as a standard. Electron rrucro­scopic analysis was performed according to the method of Tanaka and col­leagues [8,9]. The purifie~ enzyme was applied to a TSKG4000SW gel filtration column, eluted With 100 mM ammOl1lum acetate buffer (pH 7.4), and negatively stained on a glow-discharge carbon-coated Formvar mem­brane with 3% uranyl acetate (pH 4.5). The negatively stained sample was examined with a Hitachi H7000 electron microscope.

Biochemical Properties

Substrate Specificity: T en microliters of purified enzyme solution (25 flg/ rnJ) was incubated at 37"C for 1 h in a medium containing 20 mM Tris-HCl buffer (pH 8.5) and 0.1 mM of various f1uorogenic pertides. T he reactions were terminated and the increase in the fluorescence 0 AMC was measured by a fluoresce nce spectrometer.

SKIN PROTEASOMES 347

Table I. Effect of Various Protease Inhibitors on Proteasomes Activityn

Residual Activity! (%)

Reagent Concentration 0.03% SOS SDS (-)

N one 100.0 100.0 OFP' 1 mM 52.2! 41.3!

10mM 5.9! 35.5! PMSF J lOOllg/ml 96.1 92.0 SBTI' 100 flg/m l 98.3 73.4 NEMJ 100 p g/ml 13.4! 30.2! 1M lOOllg/ml 28.2! 46.0! IAN lOOllg/ ml 90.8 101.5 E-64 lOOllg/ ml 92.3 121.6 Leupeptin lOOllg/ml 29.4! 68.2! C hymostatin 100 p g/ml 13.1! 20.t! Pepstatin A 100 flg/ ml 93.9 91.8 EDTA' lOOllg/ml 118.1 132.5 EGTN 100,ug/ml 102.4 90.0

, Enzyme (25 Jlg/ml) was preincubated with various protease inhibitors for 5 min at 37 °C, and enzyme activity was measured in the presence o r absence 0(0.030/0 SDS, as described ill Materials alld Methods.

I Activity relative to that in the absence of inhibitors. , Diisopropylfluorophosphate. J Phenylmethylsul fony lfluoride. , Soybean trypsin inhibitor. f N-ethylmaleimide. g Iodoacetamide. , MOlloiodoacetic acid. 1 Ethylcnediamincrctraacetic acid retrasodium sa lt. j Ethylene glycol bis-N,N,N ',N'-tetraacetic acid.

Effect of Free Fally Acids and Divalellt Caliolls: Ten microliters of purified enzyme solution (25,ug/ml) was incubated at 37"C for I h in medium containing 0.1 mM SLLVY-MCA, 20 mM Tris-HCI buffer (pH 8.5), and various free fatty acids and divalent cations. The reactions were terminated and enzyme activity was measured f1uorometrically.

Effect of Protease [/I!t ibilors: Purified enzyme (25 Ilg/ml) was preincubated with various protease inhibitors at the indicated concentrations (Table I) for 5 min at 37°C. Enzyme activity was then assayed in the presence and absence of 0.03% SDS.

Distribution of Skin Protcasomcs in Rat Skin The hair and subcuta­neous tissues were removed from rat skin with a scalpel and surgical scissors, and the skin was cut into rectangular pieces (1. 5 X 5.0 cm). Each piece was soaked in 1.0 M KCl at 4°C overnight. This treatment made it easy to separate the epidermis from the dermis with a scalpel. Crude proteasomes extracts were isolated independently from the epidermis and dermis accord­ing to the method described above.

Immunologic Methods A monoclonal antibody (MoAb2 _ t 7) against pu­rified human liver proteasomes, a kind gift from D r. Keij i T anaka (Insritlltc for Enzyme Research, The University ofTokushima), was used as the first antibody. This antibody reacts specifically with the largest component (C2

subunit) of rat proteasomes as wel l as human proteasomes [1 6,17]. Immunoelectrophoretic blot analysis was carried out by the method of

Towbin et al [18]. The crude extract and purified enzyme separated by SDS-P AGE in 10 - 20% gradient gels were transferred electrophoretically to Hybond-C (Amersham). The membranes were pretrea ted with Block Ace (Yukijirushi Co., Sapporo, Japan) and then rreated wirh the first antibody and the peroll.;dase-conjuga tcd second antibody (anti-mouse rabbit IgG). 3,3' -Oiaminobenzidine was used as a peroxjdase substrate for the detection of proteasomes.

For immunohistochemical staining, paraffin-embedded rat skin tissue was stained by the avidin-bjotin-peroxidase complex method of Hsu et al [1 9]. After incubation with 0.01 % H20 2 in methanol to reduce the endogenous peroxidase activity, the tissue was trea ted with normal goat serum and the first antibody. The method was carried out using biotinized antimouse IgG rabbit serum and avidin-biotin complex with horseradish peroxidase. Di­aminobenzidine was used as the peroxidase substrate for the detection of proteasomes. Sections were counters tained with Meth yl Green solution, dehydrated, cleared, and mounted.

348 SUGA ET AL

1000

9 8

a

t ?

<;;

b t thyroglobulin

7SJ lerritin K

catalase

~Idolase

]11

1 l~-,-----,----,-56789

elution volume (ml)

4 5 6 7 8 9 10 11 12 13 14

elution volume (ml)

0.25

0.20

0.15

0.10

r

t o co N o Q. c .Q "§ C Ql () c o ()

c .0;

o n.

Figure 2. Elution profile of proteasomes from gel filtration HPLC (TSKG3000SWXL). (a) The enzyme from the Phenyl Sepharose CL-4B column was applied to a gel filtration HPLC column . . Each fraction (0.5 ml) was assayed for protein concentration (e) and enzyme activity (0). Aliquots of the fract ion indicated by the arrow were used as the purified enzyme. (b) The molecular weight was eStImated as approximately 750 kD by gel filtra­tion. Molecular markers were thyroglobulin (669 kD), ferritin (440 kD), catalase (232 kD), and aldolase (158 kD).

RESULTS

Purification of Pro tea somes Figure 1 shows the elution profile of ammonium-su lfate-fractionated proteasomes from the Phenyl Sepharose CL-4B column. Over 90% of the enzyme activity was eluted with 1.25 M KCI, with the remaining small amount eluted with equilibration buffer. The active fractions (indicated by the arrow) were pooled, concentrated, and subjected to gel filtration HPLC (TSKG3000SWXL). Proteasomes were eluted at 5.5-6.5 ml (indicated by the thick arrow, Fig 2a) as a single peak of enzyme activity. The molecular weight was estimated as 750 kO from gel filtration HPLC, as shown in Fig 2b. The final preparation gave a single protein band on PAGE (Fig 3a, Lane 1) and a single peak on repeat HPLC (data not shown). The results of purification from rat skin are summarized in Table II. Recovery of enzyme activity through ammonium sulfate fractionation, Phenyl Sephar­ose CL-4B chromatography, and HPLC chromatography was ap­proximately 30% with 156-times purification.

Electrophoresis The electrophoretic patterns of the purified en­zyme in PAGE and SOS-PAGE are shown in Fig 3a. As shown in lane 1, the enzyme migrated as a single protein band in PAGE. However, in SOS-PAGE, the purified enzyme dissociated into at least four to five bands, of which the majority fell into a Mr range of 36-20 kO. Two-dimensional PAGE of the purified proteasomes showed 10- 15 components with different pi values in the range of 3-10 (Fig 3b) . .

Electron Microscopic Analysis Figure 4 shows electron mi­crographs of purified skin proteasomes. At low magnification (Fig 4a), .the en~yme n~olecules appeared as sy.mmetr~cal, ring-shaped particles with a ul1lform diameter of 150 A. At higher magnifica­tion (Fig. 4b), a hole, usually with a diameter of 30-50 A, was seen centrally located in the molecule. We detected semi globular units around the central hole (thick arrows) and one or two slits connect­ing the central hole to the outside (thin arrows).

THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

a 1

Top- ~

2

Mr

-36.5Kd

"" -20.1 Kd

b p i

10.6 9.9 8.3 6.4 I I I I

4.9 4.1 I I Mr

36.5Kd

I ~-;; 20.1 Kd --

Figure 3. Electrophoretic analysis of the purified enzyme. (a) Lalles 1 alld 2, results of PAGE and SDS-PAGE, respectively, on the purified enzyme. Migration was from the cathode at the top. T.D. , position of the tracking dye. (b) Purified enzyme (50 J1g) was subjected to two-dimensional PAGE. The purified enzyme was applied to an isoelectric focusing gel and SDS­PAGE was carried out. Molecular weight marker proteins were lactic dehy­drogenase (36.5 kD) and trypsin inhibitor (20.1 kD). Cytochrome c and cytochrome c acetate from horse heart were used as pi markers.

Properties of the Purified Enzyme

Substrate Specificity: The purified enzyme specifi.cally hydrolyzed SLL VY -MCA, a good substrate for chymotrypsin, with an enzyme activity of approximately 5.65 unit/flg protein. The enzyme was also somewhat active toward Suc-Arg-Pro-Phe-His-Leu-Leu-Val­Tyr-MCA (approximately 1.52 unit/!Lg protein) and Z-Phe-Arg­MCA (approximately 0.41 unit/flg protein), suitable substrates for renin and cathepsin B, respectively.

Effect oj Fatty Acids: The effects of saturated and unsaturated fatty acids with various even-numbered carbon chain lengths on protea­some activities were studied. All the fatty acids tested had great stimulatory effects. Among the saturated fatty acids, 5.0 roM pal­mitic and stearic acids increased the relative activity by 5.7 and 4.7 times, respectively. Among the unsaturated fatty acids, 5.0 111M palmitoleic and arachidonic acids increased the relative activity by 5.0 and 5.6 times, respectively. However, the methyl esters of pal­mitic acid, stearic acid, and arachidonic acid showed no activating effects.

Effect oj Divalent Cations: The effect of divalent cations on the activity of proteasomes was investigated. The relative activity was inhibited by 1.0 m.M of C u++ and Zn++ to 9% and 47% of control respectively. The activity was inhibited especially by Hg++, ~ whose presence the activity was reduced to 1 % of control. How­ever, enzyme activity increased by approximately 1.5 times (143%) III the presence of 1.0 mM Ca++.

Effect oj Protease Inhibitors: Table I shows the effects of various protease inhibitors on enzyme activity. Some serine protease inhibi­tors, such as diisopropylfluorophosphate, inhibited enzyme activity; however.' phenyl methyl sulfonyl fluoride and soybean trypsin inhib­Itor, which are also serine protease inhibitors, exhibited no such effect. Sulhydryl-blocking agents, such as N-ethylmaleimide, iodo­acetamide, and leupeptin, inhibited enzyme activity. On the other hand, ~onoiodoacetic acid and E-64 had no effect on the activity. Interestl~l~ly, chymostatin, a serine and cysteine protease inhibitor, also IIllllblted enzyme activity. Other inhibitors, such as metallo protease and carboxyl protease, had no effect on enzyme activity. The effect of protease inhibitors on enzyme activity was very similar in both the presence and absence of 0.03% SOS.

Immtmoblot Arzalysis: A monoclonal antibody that cross-reacts only with a .31 kO compOl~ent of human liver proteasomes [16,1 7] reacted speCifically With a slIlgle component of similar size in crudr extracts and purified proteasomes from rat skin.

VOL. 101, NO.3 SEPTEMBER 1993 SKIN PROTEASOMES 349

Table II. Purification of Proteasomes from Rat Skin"

Purifica tion Step

Crude extract (NH.hS04 frac tionation Phenyl sepharose CL-4B HPLC (TSKG3000SWXL)

Total Volume

(ml)

1200 190 150 210

Total Activityb·'

(unit X 10' )

240.8 11 2.7 199.7 65.7

Total Protein

(mg)

2760 874 127.5

4.83

Specific Activityb., Purification Yield

(unit/ mg X 10') (fold) (%)

0.087 1.0 100 0.129 1.5 46.8 1.566 18.0 82.9

13.60 156.3 27.3

• All purification procedures we re performed at 4'C except for HPLC ge l fi lt ration column chromatograph y. At each step, enzyme activity and protein concentration were measured as described in Material, alld Method,. .

• Protease activity was assayed using SLLVY -MCA as substrate. , One unit of activity was defined as the amount of enzyme de gradin g 1 nmol of substrate per hour.

Distribution or Proteasomes; The l oca l~zation of prote~somes was inves tigated by measurin g enzyme actIvItIes 111 the eplderlms and dermis. As shown in Table III, the total and specific enzyme activi­ties in the epidermis were approx imately 10 and 20 times higher, respectively, than th e activity in the dermis. .

An immunohIstochemIca l study by the ABC method uSll1g monoclonal antibody revealed that the enzyme was diffusely dis­tributed , mainly in the cytoplasm of the epidermis (Fig 5b). The dermis did not stain at all except for infiltratin g cells, follicular epidermal cell s, and sebaceous glands. N o positive reactions were seen in control sectIOns of the skll1 (data not shown).

DISCUSSION

The presence of proteasomes has been reported in various tissues incl uding liver, spleen, stomach, lung, and small intestine [7]. Tan­aka et al [7] examined the proteasome content in a variety of rat tissues using sandwich enzyme immunoassays and presented evi­dence for the ex istence of proteasomes in skin tissue. However, purification and characterization of the enzyme from skin tissue were not performed. Therefore, we purified and investigated the properties, gross structure, ~nd distribution of proteasomes in skin. Skin proteasomes were punfied from rat skill by a procedure that i.ncluded ammonium sulfate fractionation, column chromatogra­phy on Phenyl Sepharose CL-4B, and gel filtration HPLC on TSKG3000SWXL. This purification procedure is relatively simple compared to other purification procedures but yielded purified en­zyme that was chromatographically and electrophoretically ho­mogenous, as indicated in Fi~s 2 and 3 . A 156-times purification with a yield of 27.3% was ach Ieved (Table II) . DUring the pUrifica­tion process , the total activity and yield of proteasomes increased

a b

Figure 4. Electron microscopy of rat skin proteasomes. The proteasomes were negatively stained with 3% uranyl acetate at pH 4.5 after the rcmova l of residual oligomers by TSKG4000SW gel chromatography. T lrick, tlrill arrows, semi globular units and slits, respectively. Bar, 400 A (a), 200 A (b).

after Phenyl Sepharose CL-4B chromatography, i11dicating protea­some activation during this step. There have been reports of similar activatio n during diethylaminoethyl cell ulose chromatography [37]. These findings may ass ist in identifying the mechanism of activation of proteasomes.

The molecul ar weight of the purified proteasomes was estimated to be approximately 750 kO by gel filtration chromatography (Fig 2b). On SOS-PAGE, the purified enzyme dissociated into several bands, of w hich the majority fe ll into the range of 36-20 kD (Fig 3a, la,.,e 2). O n two-dimensional PAGE, 10 - 15 components were recognized (Fig 3b) . In other rat t issues, including liver, bra in , mus­cle, etc., the enzyme consists of 15 -20 subunits with varying mo­lecular weights ranging from 25 - 35 kD. These results indicate that proteasomes from rat skin are very similar to those from other rat tissues. Electron microscopy showed the enzyme as symmetrical ring-shaped particles approximately 150 A in diameter with a small hole in the center. W e were able to detect semiglobu lar units and slits that are reported to resu lt from proteasome fragility or fl ex ibil­ity [1 ,8,20,21 ].

Several characterization studies were performed on the purified proteasomes. First, the substrate specifi city of skin proteasomes was studied using various syntheti c substrates. Among those tested, pro­teasomes showed th e hi ghest activity toward SLLVY -MCA, a sub­strate suitable for chymottypsin-like proteases. Proteasomes were also active toward Suc-Arg-Pro-Phe-His-Leu-Leu-Val-Tyr-MCA and Z-Phe-Arg-MCA, substrates for renin and cathepsin B-like en­zymes, respectively. T hese results show that serine and cysteine residues, at least, are involved in the activity of skin proteasomes. These findings support other reports that proteasomes have multi­ple catalytic sites within each enzyme molecule [8 ,22,23]. In gen­eral, an enzy me has only one active site; however, the broad sub­strate specificity of proteasomes may be advantageous for protein hydrolysis. Inhibition studies indicated that diisopropylfluorophos­phate, N-ethylmaleimide, leupeptin, and chymostatin preferen­tially inhibit the cleavage of SLL VY -MCA by proteasomes (Table I) . This also indicates that proteasomes have properties similar to those of both seri ne and cysteine proteases. Specific activation by SOS has been previously reported for proteasomes from various tissues [7 ,23 - 27]. In th e case of rat skin proteasomes, the highest ac tivation (approximately 20 times) was observed in the presence of 0 .03% SDS (data not shown) . To define the biologic activator under physiologic conditions, the effects of various fatty acids and diva lent

Table III. Proteasomes Activities of Rat Skin"

Epidermis Dermis

Total Enzyme Activity (unit X 10'/100g wet tissue)

1.01 ± 0.50b

0.10±0.10

Specific Activity (unit/ mg)

785.48' 36.08

• Rat skin was separated into two layers (epidermis and derm is) by incubation in 1.0 M KC I. The enzyme activi ties of each layer were then measured as described in Materials alld Melhods.

, Mean ± standard deviation (n = 5). ' T he mean of total enzyme activity/ protein concentration (mg/ 100 g wet tissue).

350 SUGA ET AL

a 1 2 34

36. 5kd-

20.1kd -

b 1"/ -

Figure 5. Immunoelectrophoretic blot analysis and immunohistochemical distribution of proteasomes in rat skin. (a) Latles 1 atld 3, staining patterns of the crude extrac ts and purified enzyme, respectively, w ith Coomassie bril­liant blue. Latles 2 and 4, immullostaining patterns of the crude extracts and purified enzyme, respectively, with monoclonal antibody. (b) A section of normal rat skin stained with monoclonal antibody by the avidin-biotin-per­oxidase complex method. Note that proteasomes are predominantly local­ized in the epidermis. Bar, 100/1m.

cations were subsequently studied. Among saturated and unsatu­rated fatty acids, palmitic and stearic acids, and palmitoleic and arachidonic acids activated proteasomes. These findings suggest that proteasomes might be activated by these fatty acids under phys­iologic conditions. The mechanism of activation by fatty acids re­mains unclear; however, the carboxylic termini seem to be indis­pensable for the activation process because esterification of the carboxy lic termini eliminated the activating effect. Proteasomes were inhibited by a majority of divalent cations. Only 1.0 mM calcium ion produced a 1.5 times activation, whereas copper, zinc, and mercury ions were potent inhibitors. The strong inhibition by the mercury ion indicates that skin proteasomes are cysteine pro­teases . These findin gs suggest that fatty acids and calcium ions playa role in regulating the activity of proteasomes under physiologic conditions.

Enzyme localization in rat skin was investigated by measuring the activity of proteasomes in the epidermis and dermis separated with 1.0 M KCI. The total and specific activities in the epidermis were higher by 10 and 20 times, respectively, than those in the dermis (Table III) . Immunohistochemical analysis with monoclonal anti­body indicated that proteasomes are distributed throughout the epi­dermis. In contrast, the dermis tested mostly negative for protea­somes except for the infiltrating cells, follicular epidermal cells, and sebaceous glands (Fig 5b). These findings also strongly suggest that proteasomes are localized predominantly in the epidermis.

Nonlysosomal proteases, calpain [28-30]' and proteasomes [7,8] have been detected in several mammalian tissues, where they are conceivably responsible for the limited proteolysis of proteins and peptides in the cytosol. The wide distribution of these enzymes suggests their general importance in cellular function; however, their physiologic roles have not yet been clarified. In the case of cal pain, a close relationship with the differentiation and prolifera­tion of epidermal cells is suspected [31,32]. Recently, several inves-

THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

tigators have reported independently that proteasomes are responsi­ble for the differentiation and proliferation of animal cells [17,33 - 38]. The prominent presence of proteasomes in the epider­mis also suggests that proteasomes may be related to the differentia­tion and proliferation of epidermal cells. Further studies are now in progress to define the physiologic function of proteasomes in skin.

We express ollr sincere thatl ks to Prof A. Ikai (Departmetlt oj Scift/Ce, Tokyo It,stitllte oJ Tecllllology) Jor assistance itl carrying Ollt tire electron microscopic analy­sis, atld Dr. K. Tanaka (InstituteJor Etlzyllle Research, The Utl iversity oJ Tokll­shima) Jor the lise oj monoc/onal an tibody to proteasollles.

This IVork IVas supported, in pari, by grants from Ihe Ministry oj Edllca tioll , Scietlce and Cuitllre, j apatl (03670531).

REFERENCES

1. Rivett AJ: The multicatalytic proteinase of mammalian cells. Are/Jill Bioe/IeI" Biop/oys 268:1 - 8. 1989

2. McGu.ire M), Croll DE, DeMartino GN : ATP-stimulated proteolysis in soluble extracts ofBHK 21/c13 cells. A rc/o BiDe/if'" Biop/oys 262:273-285,1988

3. T anaka K. !chi hara A: Involvement of proteasomes (multicatalytic proteinase) in ATP-dcpendent proteolys is in rat reticulocyte extracts. FEBS Lett 236:159-162. 1988

4. Matthews W . T anaka K. Driscoll) , !chihara A. Goldberg A: Involvement of the proteasome in various degradative processes in mammalian cells. Pro, Nat! A,ad Sci USA 86:2597 -2601, 1989

5. Eytan E, Ganoth D. Armon T, H ershko A: ATP-dependent incorporation of20S protease into the 26S complex that degrades proteins conjugated to ubiquitin. Proc Nat! Acad Sci USA 86:7751-7755. 1989

6. Driscoll) , Goldberg AL: The proteasome (multicatalytic protease) is a component of the 1500-kDa proteolytic complex which degrades ubiquitin-conjugated proteins. ] Bioi Cil, ,,, 265:4789-4792. 1990

7. Tanaka K, Li K, !chihara A, Waxmann L, Goldberg AL: A high molecular weight protease in the cystosol of rat liver (purification, enzymological properties and tissue distribution).] Bioi Cil, ,,, 261:15197 -15203, 1986

8. Tanaka K. Yoshimura T. Kumatori A. !chihara A. Ikai A. Nishigai M. Kameyama K, Takagi T : Proteasome (multi-protease complex) as 20S ring-shaped particles in a variery of eukaryotic cells. ] Bioi C/o , ,,, 263: 16209- 16217, 1988

9. Tanaka K. Yoshimura T . !chihara A, Kameyama K. Takagi T: A high molecular weight protease in the cystosol of rat liver (properties of the purified enzyme). ] Bioi C/o , ,,, 261:15197-15203, 1986

10. T anaka K, Yoshimura T, Tamura T, Fujiwara T, Kumatori A, !chihara A: Pos­sible mechanism of nuclear translocation of proteasomes. FEBS Lett 271:41-46,1990

11. Takamori K, Thorpe ) , Goldsmith LA: Skin sulfhydryl oxidase purification and some properties. BiDe/ii", Biop/oys Acta 6 15:309 - 323,1980

12. Davis BJ: Disc electrophoresis. ll . Method and application to human serum pro­teins. A IIII NY Acad Sci 121 :404 - 427, 1964

13. Laemmli UK: C leavage of strucrural proteins during the assembly of the head of bacteriophage T •. Nalllre 227, 680 - 685. 1970

14. O'Farrell PH: High resolution two-dimensional electrophoresis of proteins.] Bioi c l"", 250:4007 - 4021 , 1975

15. Lowry OH, Rosebrouth N), Farr AL, Randall RJ: Protein measurement with the folin phenol reagents. ] Bioi CIoe", 193:265 - 275, 1951

16. Kanayama H. Tanaka K, Aki M. Kagawa S, Miyaji H. Satoh M, Okada F, Sato S, Shimbara N , Akira 1: Changes in expressions of proteasomes and ubiquitin genes in human renal cancer cells. Callcer Res 51 :6677 -6685,1991

17. Shimbara N, Orino E, Sone S, Ogura T , Takashiro M, Shono M. Tamura T . Yasuda H, Tanaka K, !chihara A: Regulation of gene expression of proteasomes (multi-protease complex) during growth and differentiation of human hema­topoietic cells. ] Bioi CIo e", 267, 18100-18109, 1992

18. Towbin H, Staehelin T, Gordon J: Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proe Na tl A cad Sci USA 76:4350 - 4354.1979

19. Hsu SH, Raine L, Fanger H: Usc of avidin biotin-peroxidase complex (ABC) in immunoperoxidase tech niques. ] Histocloe", Cytoe/oe", 29:577 - 580, 1981

20. Arrigo AP, Tanaka K, Goldberg AL, Welch WJ: Identiry of the 19S' prosome' particle with large multifunctional protease complex of mammalian cells (tbe proteasome). Natllre 331 :192- 194. 1988

22. Dahlmann B. Kuehn L, Rutschmann M. Reinauer H: Purification and character­ization of a mlllticataIytic high-molecular-mass proteinase from rat skeletal muscle. Bioe/lem] 228, 161- 170, 1985

23. Wilk S, Orlowski M: Evidence that piruitary cation-sensitive neutral endopepti­dase is a multicatalytic protease complex.] NCO/roe/oe", 40:842-849, 1983

24. Rivett AJ: Purification of a liver alkaline protease which degrades ox idatively modified glutam ine synthetase.] Bioi C/o .", 260: 12600-12606.

25. Dahlmann B. Rutschmann M, Kuehn L, Reinaller H : Activation of the multicata­lytic proteinase from rat skeletal muscle by fatry acids or sodium dodecyl sul fate. Bioe/oe",] 228:171-177.1985

26. T anaka K, Yoshimura T, !chih.ra A: Role of substrate in reversible activation of

VOL. 101, NO.3 SEPTEMBER 1993

27.

28. 29.

30.

31.

32.

proteasomes (multi-protease complexes) by sodium dodecyl sulfate. Biochem] 106:495 -500,1989

Saitoh Y, Yokosawa H, Ishii S: Sodium dodecyl sulfate induced conformational and enzymatic changes of multicatalytic proteinase. Bioclrem Biophys Res Com­m,,11 162:334-339, 1989

Murachi T: Calpain and calpastatin. Trwds BioclwlI Sci 8:167 - 169, 1983 Murachi T: Intracellular Ca2+ protease and its inhibitor proteins; calpain and

calpastatin, in calcium and cell functions, Vo!' 4. Cheung WY (ed.). Academic Press, New York, 1983, pp 377-410

Murachi T: Ca lcium-dependent proteinases and specific inhibitors: cal pain and calpastatin. Bioclwn Soc Symp 49:149-167,1984 .

Miyachi Y, YoshImura . N , Suzukt S, Hamakubo T, K..l1.l11agl R, Imamura S, Murachi T: Blochenllcal demonstration and l1nmunolllstochenllcal locahza­cion of calpain in human skin .] Invest Dermd(o/80:346-349, 1986

Ando Y, Miyachi Y, Yoshimura N, ImamuraS, Kannagi R, Murachi T: Purifica­tion and characterization of cal pains from pig epidermis and their action on epidermal keratin.] ltwest Dermato/80:346-349, 1986

SKIN PROTEASOMES 351

33. Haass C, K10etzel PM: The Drosophila proteasome undergoes changes in its subunit pattern during development. Exp Cell Res 180:243-252, 1989

34. Haass C, Pesold-HurtB, Multhaup G, Beyreuther K, K10etzel PM: The PROS-35 gene encodes the 35 protein subunit of Drosophia melanogaster proteasome. EMBO] 8:2373-2379, 1989

35. Pal JK, Gounon MF, Grossi SA, Scherrer K: Presence and distribution of specific prosomc antigens change as a function of embryonic development and tissue­rype differentiation in Pleurodeles walt!.] Cell Sci 90:555-567, 1988

36. Akhayat 0, Grossi de Sa F, Infante AA: Sea urchin prosome: characterization and changes during development. Proc Nail Acad Sci USA 84:1595-1599,1987

37. Gautier J, Pal JK, Grossi MF, Beetschen J C, Scherrer K: Differential cytolocaliza­tion of prosomes in axolotl during oogenesis and meiotic maturation.] Cell Sci 90:543-553,1988

38. Fujiwara T , Tanaka K, Orino E, Yoshimura T, Kumatori A, Tamura T , Chung CH, Nakai T, Yamaguchi K, Shin S, Kakizuka A, Nakanishi S, Jchihara A: Proteasomes are essential for yeast proliferation.] Bioi Chern 265 :1 6604-16613,1990