Voh~me 21t(l, namb~r ;L 21 I =21 ~

FEllS 09491

M~reh 1991

Inactivation ofhuman eystatin C and kininogen by human cat,hepsin D Brigita Lenar¢:i~,~. Marta Kra.~ovee ~, Anka R,itonja t , [sleifur Ohlfsson = and Vito Turk ~ DD f i ~ r t m ¢ , t a f Btuehemt~rry. Jo~=¢fSt¢fim h~ttt.l¢, lamovo 39. 6100f7 LJubljtma. Shwen/a, Yu~loM¢tvla and = Drpmime, t t~f Clinieal

Ch~mtsl~y. Unlr¢~:~lt~~f L~ml, Univcrslt,v th~sptml, S.221 85 l.und, 3"wcdcn Received i 6 J a n u l t r y 1991 A p~p~ln tnltibitor of 22 k D a was isolated from human pt=te=nta and shown to ~ identical to residues Cy~146.keu:t?.~ of the third domain of human klninol~en. This kinino~en dam;tin anti reeon~binan~ hun~an cystatin C were tmtettvated by pcptlde bond eleavalles at hydrophobic amino acid residues due to Ihe etClion of cathepsin D, Tlie~e results ftmher support 11~cprop=seal role ofcalhepsin D in the regulation orcysteinc proteinus~ activity. Cystatin C: KininoJen; Cathepsin D; Prot~inase inhihitor inaetlvation

2.2 Isnlatlan of thv third don~ain a f kinhlogc, n

1, iNTRODUCTION Cathepsin D (EC 3.4.Z3.5) is an aspartic proteinase widely distributed in all mammalian tissues (reviewed in [1]). The enzyme is located intracellularly, in Iysosomes, but has also been detected extracellularly under a variety of pathological conditions [2-5]. Cathepsin D plays an important role in the degradation o f proteins through limited proteolysis. Thus, cathepsin D is responsible for the specific cleavage of the myelin basic protein [6] and other brain proteins [7], for the conversion of proc~llagen to collagen [8], parathyroid hormone degradation [9], the release of T-kinin from rat T-kininogen [I0,11] and inactivation of human stefin A and stefin B, intracellular protein inhibitors of cysteine proteinases [12]. In this report we show that cathepsin D is able to degrade human cystatin C and the third domain of kininogen, both extracell.ular protein inhibitors of cysteine proteinases. This finding might be important for elucidation of additional e×tracellular actions of cathepsin D. 2. MATERIALS AND METHODS 2,1 Materials Recombinant human cystatin C was prepared as described [13]. Human cathepsin D was isolated from spleen [14], Papain (type 111) (EC 3,4,22,2) and sttbstrate BANA were from Sigma, Bovine haemoglobin was prepared as described in [15]. Pepstatin was from Peptide Research Foundation (Osaka, Japan). Acetonitrile and t rifluoroacetic acid were from Merck and BDH Chemicals, respectively. All other chemicals were of analytical grade,

Correspondence address: B. Lenar~:i~:, Department of Biochemistry, Jo~2ef Stefan Institute, J a m o w Yugoslavia

39, 61000 Ljubljana, Slovenia,

Published by Elsevier Science Publishers B. V,

The isolation was carried out followinj tile procedure as described in [161 with some modirietttions. Briefly, human placenta (1000 g) was homogenized in 0.9~, NaCI (I :1.5, w/v) sohnlon and centrifuged for 60 rain a= 14 000 x /~. The supernatant war adjusted to pet 10,$ with 5 M NaOH, and after standing I h at roan1 temperature, to pH ?.S with 3 M HCI, After ¢entrifugation the supernatant was applied to an affinity column (7.5 × $.0 era) of Cm-papain-Sepharose 4B prepared as in [1"7], et|uilibrated with 10 mM Tris-HCI buffer (pH 7.5). The coluntn was washed with the same t~uffer and the bomtd proteins eluted tl~en with 10 mM NaOH (pH I 1.01, Papain.inhibitlng fractions were pooled, adjusted to pH g.O with 3 M HCI and concentrated by ultrafiltration on an YM-5 membrane (Amicon). The concentrated sample war dialyzed against 10 mM Tris-HCI buffer (pH 7,5) containing 0,1 M NaCI and applied to a column (120 x 5 cm) of Sephadex G-75 equilibrated in the same buffer, Papain.inhihiting fractions were resolved as three peaks corresponding to M, of about 12 25 attd 6S kDa, The 25 kDa.M, protein peak was dialyzed against 10 mM TrisHCI buffer (pH 8,5) at~d subsequently cllromatographed on DEAESephacel eq uilibrated with the same buffer, The inhibitor was eluted in a single symmetrical peak with linear gradient of NaCI (0,05-0,15 M) in the same buffer, The sample was concentrated, assayed for its purity by SDS-PAGE and amino acid sequence analysis, and stored at - 25"C until use,

2,3. Assays of enzymes and inhibitors The inhibitory activity of cystatin C and the third domain of kininogen was determined by measuring the inhibition of papain using BANA as substrate [18], The cathepsin D activity was measured using haemoglobin as substrate [14] and checked by active-site titration with pepstatin [19]. Protein concentrations were determined by the method of Lowry [20]. One unit of the inhibitor activity was defined as the amount which inhibited I /~g of papain completely, 2.4, The interaction of CPIs with cathepsin D The experiment was performed essentially as in [12]. Solutions of cystatin C or the third domain of kininogen (2/LM) were mixed with an equal volume of cathepsin D (0,5 t~M) in 100 #M sodium acetate buffer (pH 3.51 and incubated at 37°0. At various times 100 .1 samples were removed and the cathepsin D activity was stopped by acidifying the samples to pH 2.0 with 3 M HCI, As a control, pepstatin (i ~M) was added tO ~athepSi~ D ~tior to incubation. The cleavage products were separated by RP-HPLC on Vydac C=~-column (250 × 4.6 mm) (10 #m) in a LDC Milton.Roy HPLC using a linear gradient of 5-80°70 acetonitrile in O. I °70 trifluoracetic acid. Flow rate

21 1

V o l u m e 2tt0, number 2

FEI~$ LETT~:R$ L]L •

A280[

.

[NaCt:l

0,3.

A520 • 0,15 M ,

3Cl .., SBatm,

0,2''

'tI

.I

!

I

'

'

,,,,

till Ill]/'"'

.,"

'I = . , " '~ '

i

It..,~,'*"

I a

i i

I

,

i~

! I i I1 !

i

,,,#

•

,,.",

,," ,.!

.

..

.,,,.-t1 I I fill

! ~

.

_

~ ~! \l

s I

!l\ !i! ~ .....~

.,/"/'

A

100

II I J

!

,

.

I !,

/~

-O,07s

-O,s

I II ,. I

200 fraction

,

,

l~ " -

300

number

Fill, I. DEAE.SePhaeeldlron;atography of the 25 kDa.M, protein peak ¢:luted from Septtad¢,~ G.75, (--) protein ,4.,,.; (.... ) inhibil0ry tt¢livily A~=,; (. . . . . ) gradient NaCl; SB dim = dimer form or .~tefin B; 3¢1 ,- third don~ah'mof kininoaen, Fraetiom of 6.4 ml were collected at a flo~v rate of IB ml/min,

was 1,0 ml/min. Peptides weredetected by UV.absorption at 220 nm, collected and dried on a Speed Vac (Savant). 2.5, Amino acM sequence umtlysis Seqt,enc¢ analysis was performed by atttomatic gas-phase Edman degradation on an Applied Biosystems sequenalor 470 A [211, tvith on.line detection of the released phenylthi0hydantoin derivatives of amino acids by all Applied Biosystemsamino acid analyser 120A [22], 2,6. Electrophoresis SDS gradient PAGE (8-25%) was carried out according to Lacmmli [231, 3. RESULTS AND DISCUSSION

3.1. 1solution of the third domain o f kininogen Ion exchange chromatography o f the 25 kDa protein fractions, obtained from gel c h r o m a t o g r a p h y on Sephadex G-75, resulted in two m a j o r inhibitory peaks eluting at 0.04 M NaCI and O. 13 M NaCI, respectively (Fig. I). The purity o f both inhibitory peaks was checked by SDS-PAGE (not shown). T h e first peak showed an apparent Mr o f 25 kDa without prior reduction by 2-mercaptoethanol whereas an apparent Mr of 13 kDa was obtained after reduction (not shown). This behaviour corresponds to h u m a n stefin B which forms an inactive dimer in the absence o f reducing agents [16]. The second peak (Fig. I, 3d) was cluted as a homogenous protein with Mr o f a b o u t 22 kDa. The Nterminal amino acid sequence analysis of the first 15 residues revealed that the isolated inhibitor is pure. This protein corresponds to the third domain o f the 212

kininogen molecule [24], T h i s fragment starting with Cy.~-246 (kininoger~ numbering), was g e n e r a t e d b y cleavage at lle-245-Cys.246 due to an unknown pro. teinase. Similarly, the isolated and sequenced human kinmogen fragment of 20 kDa from human synovial fluid [25] resulted from the cleavage at position Thr-243-Lys-244 o f tile kininogen. These two fragments derived frona the junction o f the second and the third cystatin-like domain, constituting the kininogen heavy chain, This junction represents one of the proteinase-sensitive regions in the kininogen molecule [24,26], 3,2, Fragmentation of human cystatin C and the third

domain of kininogen by cathepsin D Incubation of cystatin C and the kininogen domain with cathepsin D resulted in rapid fragmentation of both protein inhibitors. Prolonged incubation up to 45 rain resulted in additional fragments. These fragments were separated by R P . H P L C (Fig. 2) and subjected to NH2-terminal sequence analysis. Points o f cleavages for human cystatin C and the third domain of kininogen are presented in Fig. 3. Cathepsin D preferentially cleaved h y d r o p h o b i c residues, one often being aromatic. When digestion was performed at pH 5.0, otherwise under identical e×perimental conditions the same fragments were obtained (data not shown), although the rate was significantly slower. This pH corresponds to that of lysosomes [27]. Degradation o f both protein inhibitors was inhibited completely when

A220 0.500

MarCh 1~)91

FEB~ Ll~TTER$

V~lL=m¢ 2a0, .umber 2

A220 0,500

-100

A

//ljill

I

"7! / I !

1I 1 /

0.25~

,100

B

'50

LU

u. u,.

O,250

I

/ /

t~ a, +~

I

I

I

CO

1

so

i

m

I

3"

t

I

/

t t~

GD

l

/ / 1 1

Ill 1

J

/ I

0

10

I

10

20

time (m/n)

time (rain)

A220 0.500

d tJ~d

C

A220 0.500

100

;-100

0

I

//

/

1

/

// / p / / !

/

I

I

,

!

/

/

o.=so

50

Ill

I

1

)

, /

/ i

!

¢

o.2so

/

/

en /

I.~ Ill

1

/

I

I

yt__

/ t'.

/ /

/

.

0

10 time

(rnin)

n~

20

I

.,50

i

uJ IJ. u.

/

/

I

I

/

I 1

I /

"7

/'

"7 !

I

I

/

I

I

eP tlJ I,L

3~

I

"1 i

I

I:

I

/

g

O

1Q time

20

(m/n)

Fig, 2, R P - H P L C o f peptides released f r o m cystatin C a n d t h e third d o m a i n o f klninogen after incubation with cathepsin D. (A) cystatin C (c¢) after I rain o f incubation; (B) cystatin C (co) after 40 rain o f incubation; (C) third d o m a i n o f kininogen (3d) after 30 s o f incubation; (D) third d o m a i n of kininogen (3d} after 40 rain of incubation; ( - - ) absorbancy A220; (- - -) acetonitrile, V o l u m e of 100 ~I of reaction mixture w a s injected into the reverse-phase column. Major peaks are marked and their positions in the sequences are indicated in Fig, 3.

213

Volume 280) .umb¢¢ 2

I0 (a)

Mar~:h I99I

FIeldS h~vr'rER5

20

3~0

40

$ S POK"PPRLVG0~HDASVEEE~VRRALDFAVG .......

....

~

,,,i,,,, ;

,

g ...........

(b)

LWGC~RDI

E~

d

KAS H DM~HSRA/~

i .... ~.... =

~

"

,

' .....

i

PTHS PELEETLTHTITKI.~AE~INATFY~KI

D)~V

II

I

III

I

I

S0 60 # 70 80 90 VRARKQ ~'¢AG'J~YFLDVE LGRTT CTKTQ PH L -- DNC P FHDQ PH ~KA2

' ''

"

7'

-

~

a

, KIq~RV Q V ' V A G I-~'YF I D F V A R E T T C S

--

KESNEELT-V-S CE--- TKKLG¢~S LD

,,,,, ,, , , , . . . . . - -

(a)

,,,,-

rn

~ '-~

0 ,,,

I~

I'I •

,,,,,,,-,,,,I~II

"

,' ,,,

cs FQ'r~AV'PWQ~ ~,T~2LS K S T C Q r~A

-=-,=,=----~m, ~

_

~

_

I)

b=.,~.,.,.-=e

.... ( ~ )

C~-'V~VVP~TEI%~:~:Y PTV'~'- ¢ Q P.T.~M

Fig, 3. Action of human cathepsin D On cystatln C (a) and the third domain of kininogen (b). (,I.) Major sites of cleavage; ( v ) additional cleavage sites. C0ntinuous lines indicates regions of' peptide sequenced in cystatin C (a, b, c, d, e, f, g, h, i) and in the tlfird domain of kininogen (j, k, I, m, n, o, p, r),

pepstatin [2 ~M] was added, prior to the inhibitors, tO . . . . the incubation rni×ture (results not shown). This indicaies•that the cleavages were generated by cathepsin D. The proteolytic fragments obtained after interaction with cathepsin D did not show inhibitory activity on papain. The exception were both peptides containing residues 55-59 in cystatin C and the corresponding region in the kininogen domain. This region has been identified as important in the binding to proteinases [28,29]. gi values were above 10- 6 M indicating a greatly decreased binding. However, a similar inhibitory peptide containing this regions with similar Kls was ob214

tah led recently after CNBr cleavage of chicken cystatin [30]. Cleavage by cathepsin D inactivates most of the in. hibitory activity of the cysteine proteinase inhibitors. Consequently, limited proteolysis of cystatins by this action indicates a possible physiological role for cathepsin D in the regulation of cysteine proteinase activities. Inactivation of CPIs by the aspartic proteinase cathepsin D is important since we would not expect the cysteine proteinases of lysosomes to inactivate them. Specifically, it is known that the interaction of CPIs with ttieir target enzymes does not lead to peptide bond cleavage [29,31].

V o l . m e 2g0. number 2

F~BS L f f T T E R S

Uniwrttty ~ledteal C~mr¢, I~nrham, NC~, LlSAI rot hh ,;rilleal rtadtn~ Bribe maflu~¢¢tpl, Tht~ ~*~ork wa~, ~upporltd ill parl by Iit¢ R¢~,'ar~:tt Co.n¢ll of Sio~erli= t11td Federal S'ler, flwlin [l?l I.tarr¢u, ^.J. tl*Jl~l}M¢lhJoth En,~ymol #0, "I'/I=~?R Jill[ Barr~ll, A.J~ (1972} Anal, I~iod~m. 47, ]~10=19], [191 Kni~hh C.G. and Barrcll, A.J. (197611~t~htm, J, I$~. I 1%125. []0} Lowry, O,H., R~cbr~ul~h. N~J,. Farr. A,L, ~nd R~ndall, R.J. tlg'~|) .1, [tiOL Ch~n~, |93. 265=275. [211 H~wi*~k. K,M,. Hunknptll~r, M,W.. Hood, L,E. and Drtytr. W,J, (19~11 J. Biol. Chem, 2:;6, 7991~-7997. 1221 Hnnk,pitler, M.W. nnd Hood, L.E, (191t~) Mclhods En~ymol. 91.4~6-493. !-~:~1 I.aem.~li, U,K. (19711) NaB,ire 227, 61¢O.685, 1241 Barr¢tl. A.J,, Rawlillt~S, N,I)., Davies, M,E., Maehl~tdl, W,. Salvtscn, CI, ~nd Turk, V, (19~6) in: Prolelmtse Inllibilors(Bar. roll, A,J, and Salvcscn, G. eds) pp. :,1S=~69, Els©vier, Amsttrdt~m, 12:~1 kcm~rP.i~. B,, Gnbrijcl,'.i~:, D,. Rozm'm0 B,. Kolorok-DrobniP., M. anti Turk. V. (198~} biol. Chtm, Hopp¢.Sc~'ler 369, Suppl,, 2.~9-26 |, 126} Vo~el, R,, As~falLl-Maehldd|, I,. Esterl, A., Machl¢tdt, W. and Muller-Eslerl, W. (1988) J. Biol, Chem, 263, 1266|-1266B, [2"/] Tapper, H. and Sundltr, R. (1990) Biochem, J. 272, 40%414, [281 Bode. W.. Enl¢ll, R,, Musil, Dj,, T1~i¢1¢, U,, Huber, R,, Karsilikov. A., Brzio, J,, Ko~. J, and Turk, V, (1988) EMBO J. 7, 2593-2599, [29} Stubbs, M,T,, Laber. B,, Bode, W,. Hubcrr, R., Jtrala,r R., [=.¢nar~i~., B,, nnd Turk, V, (1990) EMBO J. 9, 1939-I947. [30] Machleidl, W., Thld¢. U,. Laber, B,, A~sfalg-Machleidt !,, Esterl, A., Wie~and, G., Kos, J., Turk, V, attd Bode, W, (1989) FEBS Loll. 243. 234..238, [311 Bode, W., Et~gh, R,, Musil. Dj.. Laber, B,, Stubbs. M,, Hubcr, R, anti Turk, V, 11990) Biol, Chem. Hoppe.Seylcr 371, Suppl,, 111-118.

215