Biochimica et Biophysica Acta, 11}94( 1991}249-256
249
© 1991 Elsevier Science Publishers B.V. All rights reserved 016%4889/91/$03.50 ADONIS 016748899nlo240M
Activation of intracellular calcium-activated neutral proteinase in erythrocytes and its inhibition by exogenously added inhibitors Masami Hayashi
t, M i t s u s h i I n o m a t a ~, Y u m i k o S a i t o ~, H i s a s h i I t o 2 and Seiichi Kawashima 1
1 Department of Biochemistry. Tokyo Metropolitan hlstitute of Gerontoh;g); Tokyo flapan) and 2 Department of Chemistry. Aoyama-Gakuin Unicersity Tokyo (Japotl)
(Received 7 May 1991)
Key words: Calcium-activated;Proteinase: Auto!vileactivation;Proteinase inhibitor; Erythrocyte !ntracellnlar calcium-activated neutral proteinase (CANP) in rabbit erythrocytes was activated by an influx of Ca 2 + into the cells. The catalytic large subunit changed from the original 79 kDa form to the 77 kDa and 76 kDa forms on activation just in the same manner as occurs in the autolytic activation of purified CANP in vitro. The activation required both extracellular Ca 2 + and A23187, and was accompanied by the degradation of some membrane proteins and morphological changes in erythrocyte shape from discocytes to echinodisl~, echinocytes, and spherocytes. Exogenously added Cbz-Leu-Leu-Leu-aldehyde inhibited the activation of intracellular CANP as well as the degradation of membrane proteins and the morphological changes indicating that the latter two processes are due to the action of CANP. Leupeptin and E64d were without effect on intracellular CANP.
Introduction Calcium-activated neutral proteinase (CANP) is an intracellular proteinase that requires Ca 2 + for its activity [1]. From many experiments in vitro, the mechanism underlying the activation of C A N P has been visualized. The activation process appears to include C A N P autolysis in the presence of Ca 2+ during which the N-terminal portion of the large subunit, which inhibits proteinase activity, is removed [2-4]. In the case of the low-Ca2÷-requiring form of CANP (/.tCANP), the
Abbreviations:CANP, calcium.activatedneutral pmteinase; mCANP and/zCANP, CANPs which are active in the presence of Ca2. ions of mM and p.M order, respectively; SDS. sodium dodecyl sulfate; SDS-PAGE, polyacrylamidegel electrophoresis in the presence of SDS; ZLLLal, benzyloxycarbonyl-leucyl-leucyl-leucinealdehyde: calpeptin, benzyloxycarbonyl-leucyl-norlgucine aldehyde; E64d, (2 S,3S )-3-( S )-3-methyl-l-(3-methylbulylcarbamoyl)butylcarbamoyl oxirane-2-carboxylate; E64c, N-[N-(L.3.trans.carboxyoxirane-2-
earbonyl)-L-leu~l]-isoamylamine;kDa, kilo-daltons. Correspondence: S, Kawashima,Department of Biochemistry,Tokyo Metropolitan Institute of Gerontology, 35-2 Sakae-cho, ltabashi-ku, Tokyo 173,Japan.
molecular mass of the large subunit decreases from the original 79 kDa to 76 kDa via a 77 kDa intermediate [3]. Furthermore, CANP binds to cytoplasmic membranes in a Ca2+-dependent manner and its autolytic activation is facilitated greatly by this binding [5-9]. This activation scheme, however, is drawn from results obtained in vitro and is not necessarily applicable to the activation t;f CANP in vivo. Evidence is needed to show that intracellular C A N P behaves similarly upon Ca 2+ influx into cells. Synthetic proteinase inhibitors are powerful tools for elucidating the physiological functions of proteinases. Leupeptin and E64c are well-known inhibitors of CANP, although their actions are not restricted to CANP [10]. Leupeptin and E64c inhibit CANP quite well in vitro, but their effects on intracellular CANP when they are added extracellularly is not known. To answer the questions of whether intracellular CANP is activated in a manner similar to that expected from in vitro experiments, whether synthetic CANP inhibitors are effective on intracellu!ar CANP, and of what happens to cells when intracetlular CANP is activated, we examined erythrocytes as a simple cell model. Erythrocytes are especially suited to these stud-
250 ies because they contain only one type of CANP, ~CANP [11], whose activation can be followed easily by observing the changes in molecular weight of the catalytic large subunits [3], and because they lack a lysosomal system, the presence of which makes the analysis of intracellular proteolysis and proteinase inhibition by exogeneonsly added inhibitors complicated.
79~. f" 77K
Materials and Methods
Materials. Monoclonal antibodies specific for p.CANP were produced by immunizing female BALB/C mice with purified p~CANP from rabbit skeletal muscle and screening hybridoma cells which secrete #CANP-specific antibodies, as described before for the isolation of mCANP-specific antibodies [12]. The monoclonal anti-/~CANP (1AsA 2) used in this experiment recognized only the large subunit of /LCANP and did not cross-react with the large subunit of mCANP or with the small subunit. The ionophore A23187 and peroxidase-conjugated anti-mouse IgG (Fab') were purchased from Sigma and Medical & Biological Labs. (Nagoya, Japan), respectively. E64d was a kind gift from Taisho Pharmaceutical (Omiya, Japan). Benzyloxycarbonyl-leucyl-norleucine aldehyde (ZLnLal) and benzyloxycarbonyl-leueyl-leueyl-leucine aldehyde (ZLLLal) were synthesized according to the method of Ito et al. [13]. Methods. Erythrocytes were isolated from the peripheral blood of New Zealand white rabbit and purified by successive washing with PBS (phosphatebuffered saline). The washed erythrocytes were diluted to a concentration of 5 • 10S/ml and preincnbated with A23187 at 25 °C for 5 rain. The reaction was started by the addition of CaCI 2 solution and stopped after 15 rain by the addition of an equal volume of sodium dodeeyl sulfate (SDS) sample buffer. The samples were heated to 100 ° C for 5 rain and subjected to SDS-polyacrylamide gel electrophoresis (PAGE) [14] in 12.5% polyacrylamide for Western blot analysis of CANP and in 7.5% polyacrylamide for protein staining. CANP on nitrocellulose sheets was reacted with monoclonal 1AsA 2 antibody and then with peroxidase-conjugated anti-mouse IgG (Fab') and visualized by reaction with diaminobenzidine [15]. For analysis of CANP and proteins in the cytosol and membrane fractions, erythrocytes were lysed hypotonically after reaction and separated into eytosol and membrane fractions by centrifugation at 22000 Xg for 20 rain. The membrane fraction was washed three times with 5 mM phosphate buffer (pH 8.0) before subjection to SDS-PAGE. Morphological changes in erythrocytes fixed in 1% glutaraldehyde were observed under a differential interference microscope.
7SK
Fig. 1. Autolysisof intracellular/.~CANP in rabbit erythrocytesby treatment with A23187 and/or Ca2+. Erythrocyteswere preincubated in the presence(lanes 2 and 4) or the absence(lanes I and 3) of 20 p.M A23187 at 25°C for 5 min. The reaction was started with (lanes 3 and 4) or without (lanes 1 and 2) the addition of CaCI2 solutionto a finalconcentrationof 400 ~M and stoppedafter t5 min by the addition of an equal volumeof SDS-samplebuffer. Proteins were electrophoresedon SDS-polyacrylamidegelsand transferredto nitrocellulose membranes.The blotted proteinswere incubated with anti-CANP monoclonal antibody and visualized using peroxldaseconjugated anti-mouseIgG and diaminobenzidine.
Results
Activation of intracellular CANP in erythrocytes by treatment with A23187 and Ca e+ Rabbit erythrocytes were incubated in the presence of A23187 a n d / o r Ca 2+ and intracellular CANP was analyzed by the immunoblot method (Fig. l). N o changes in the large subunit of/~CANP were observed in the absence of A23187 and Ca 2+ or in the presence of either alone. However, when both compounds were present and Ca 2+ influx occurred, the molecular mass of the CANP large subunit decreased from the original 79 kDa form to the 77 kDa and 76 kDa forms. This conversion is identical to that observed during in vitro activation of/zCANP by Ca 2+ [3] and, therefore, it was suggested that the 76 kDa form is the active form in vivo as in vitro. The autolysis of intracellular IzCANP was dependent on the extracellular Ca 2+ concentration and was observed at concentrations as low as 6.25/xM (Fig. 2). It was reported that the Ca 2+ uptake into erythrocytes in the presence of A23187 is more than 80% [16]. This means that the autolysis of intracellular p~CANP may occur at about 5/~M Ca 2+ concentration. Since membranes have been shown to be involved in the activation of CANP [6-9], CANP binding to membranes and autolytic changes in intraceUular/.LCANP were followed (Fig. 3). The time-dependent changes of /J,CANP in the cytoplasmic fraction were almost the
251
A23187 ( - ) 1
2
3
A 23187 (+) 4
5
1
2
3
4
5
79K
~
~
~
~
~
~
77K 76K
Fig. 2. Effect of Ca~'+ con(:entration on the autolysis of gCANP in rabbit erythrocytes. Erythrocyteswere preincubated in the presence or the ahsence of 20 #M A23187 at 25 oC for 5 min. The reaction ~,.asstarted with the addition of CaCI_, solution to final concentrationsof O, 6.25, 25. 10t)and 400 #M (lanes I, 2, 3, 4 and 5, respectively).The other procedures were the same as in Fig. 1.
same as those in whole cells; the amount of activated p, C A N P increased with time. This is not surprising because more than 98% of the t o t a l / . t C A N P is recovered in the cytoplasmic fraction. A part of the p, C A N P was found bound to m e m b r a n e s mainly in the activated form. T h e s e results are consistent with the activation scheme obtained in vitro in which /zCANP binds to m e m b r a n e s in the presence of Ca 2+, autolyzes on the m e m b r a n e s , and is then released to the soluble fraction [9]. T o confirm further the involvement of membranes in the autolytic activation o f / . t C A N P , erythrocytes were lysed and half were incubated in the presence of Ca 2+. T h e other half were centrifuged to remove membranes and the membrane-free cytoplas-
mic fraction was incubated similarly in the presence of Ca 2+. Autolysis occurred more efficiently in the whole hemolysates at low Ca2. concentrations than in the cytoplasmic fractions (Fig. 4), showing that membranes, possibly acting as the activation loci, also facilitate btCANP autolysis in a crude system like this.
time
3o
0
30
0
A23187 C82+
-
+
+
+ + -F -I-
1
(A)
2
5 10 15 30 111111
+ -I-
+ + + +4--t-
+ .4-
,, =,,, ~ m . = , = l , ~ ,
(a)
(c)
i
i
!
Substrates of activated/.¢C~hVPin erythrocytes T o identify the protein species which are degraded upon activation of intracellular p.CANP, the proteins recovered in the cytoplasmic and m e m b r a n e fractions after treatment of erythrocytes with Ca 2+ and A23187 were analyzed by S D S - P A G E (Fig. 5). No changes were detected in the banding patterns of the cytoplasmic proteins. O n the other hand, in the m e m b r a n e fractions, spectrin, Band 3 and Band 4.1 decreased with time, and a new band a p p e a r e d just below spectrin, Thus, the main substrate proteins degraded by activated p.CANP are m e m b r a n e proteins.
Fig. 3. Time-course of p,CANP autolysis in •rabbit erythrocytes. Erythroc,jtes were preincubated in the presence or absence of 2O/zM A2_'H87at ~ ° C for 5 min. The reaction was started with or without the addition of CaCI2 to a final concentration of 400 .u.M and stopped after various intervalsby the addition of an equal volume of SDS-sample buffer for whole cell analysis(A). For analysesof CANP in the cytosol (B) and membrane (C) fractions, erythrocyteswere lysed hypotonicallyafter the reaction and separated into cytosol and membrane fractions by centrifugation. The other procedures were the same as in Fig. I. Numbersof er,,thr,'~..,,tesu~cdv,cic 2.3.10' for (A), 2.4"10" for (B) and 1.3.10s for (C).
252
(B)
(A) A23187
+
+
Ca 2 + (pM)
0
400
0
6
~
,m
mm
~
25
(C) 100 4 0 0
,row ~
~
0
mm
6
m
25
100
400
ew-
,ram
mm
Fig. 4. Effect of Ca z* concentration on the autolysis of/zCANP in whole cells, hemolysates and cytosol of rabbit ewthrocytes. (A) Whole cells. The incubation conditions were the same as in Fig. 1. (B) Hemolysates. Erythrocytes were lysed hypotonically and treated with various concentrations of CaC[ 2 as indicated. (C) Cytosol. Erythrocytes were lysed and separated into cytosol and membrane fractions. The membrane-free cytosol fraction was treated with various conccntralions of CaCI 2 as indicated. The other procedures were the same as in Fig. 1.
Effect of synthetic CANP inhibitors on activation of intracellular I.tCANP T o p r o v i d e definite e v i d e n c e that t h e d e g r a d a t i o n o f m e m b r a n e p r o t e i n s is really d u e to t h e activity o f activated p . C A N P , inhibition o f intracellular ~ C A N P activation was a t t e m p t e d (Fig. 6). Extracellularly a d d e d
l e u p e p t i n a n d E64d, b o t h k n o w n to b e inhibitors o f C A N P in vitro, h a d little effect on t h e autolysis o f intraceIlular p C A N P e v e n at c o n c e n t r a t i o n s o f 1 m M . C a l p e p t i n ( Z L n L a l ) , a c o m p o u n d d e s c r i b e d by Tsujin a k a et al. [17], w a s a p p a r e n t l y effective at a c o n c e n tration o f 100 p.M. T h e m o s t effective inhibitor w a s
(A) 300300 A23187 C a 2+
1
2
(B) 5101530300300
1 2
5101530
min
+ + 4 - + + + + + 4- - - + 4 - 4 - + + 4 - 4 - 4 - 4 - - - 4 - + 4 - + + + + - - - + 4 - + + + + +
200 K
spectrin
116K "" g7.4 K "
•,,, band 3 ..,,,, 4.1 "1 4.2
66.2 K I""
/
i
Fig. 5. Ca2+-induced degradation patterns of cytosol and membrane proteins in erythrocytes. Cytosol (A) and membrane (B) fractions were prepared as described in Fig. 3. The proteins were electrophoresed on an SDS-polyacrylamide gel and stained with Coomassie brilliant blue. Numbers of erythrocytes used were 1.2' 107 for (A) and 2.6.10 7 for (B).
253
(A) A23187 Ca2+
+
+ _
(a) + +
e"m ~"" ~
(e)
(D)
(E)
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
i
liii ~
-ni
WL4K
0
0
0
10 100 1 0 0 0
110100
0.1
10
i
~
spectrin
band 3
10 100 1 0 0 0
Inhibitor C o n c e n t r a t i o n ( pM ) Fig. 6. Effect of exogeneously added proteinase inhibitors on intracellular autolysis of ,~CANP and degradation of membrane proteins. Erythrocytes were preincubated at 25°C for 15 rain in the presence or absence of 20 ,~M A23187 and/or proteinase inhibitors at the concentrations indicated. The re,teflon was started by the addition of CaCI, to a final concentration of 400 pM and stopped after 5 rain. The procedures for CANP immunoblotting(upper panel) and protein staining (lower panel) were the same as in Fig. 1 and Fig. 5, respectively.(A), no proteinase inhibitor; (B), leupeptin; (C), ZLnLal; (D), ZLLLaI: (E). E64d. In (E) with 1 mM E64d, some part of erythrocyteslysed due to the presence of 10% dimethylsulfoxidewhich was used to dissolvethe inhibitor and the amount of proteins applied to SDS-PAGE was less than the others.
ZLLLal, developed in this laboratory. It inhibited the autolytie activation of intracellular IxCANP at concentrations as low as 10 I~M. The appearance of membrane protein degradation products was also suppressed by this inhibitor, showing that activated CANP is responsible for the degradation.
echinodisks remained. This result shows that activation of p.CANP is the primary event leading to the deformation of erythrocyte shape, possibly through degradation of membrane proteins.
Morphological change in the shape of erythrocytes accompanying the activation of I~CANP
The results presented here clearly demonstrate that intracellular p C A N P is activated upon increases in intracellular Ca 2÷ levels by the same mechanism as that established in in vitro experiments; i.e., t~CANP autolyzes in the presence of Ca 2+ and the molecular mass of the catalytic large subonit changes from 79 kDa to 76 kDa via a 77 kDa intermediate [3]. Moreover, membranes are involved in the activation of /zCANP in vivo as well as in vitro. That is, upon an increase in cytoplasmic Ca 2÷ levels, /~CANP is first bound to membranes where it is autalyzed by low Ca 2+ concentrations; activated/zCANP is then released into the cytoplasm. Autolysis and the effect of membranes were also observed in ruptured cells. This was rather
The shape of normal erythrocyte is discoid. Incubation with either Ca 2+ or A23187 alone produced no changes in erythrocyte shape (Fig. 7). On the other hand, when erythrocytes were incubated in the presence of both A23187 and Ca 2+, a condition in which i n t r a c e i l u l a r / x C A N P is activated and membrane proteins are degraded, erythrocytes changed their shape from the original discocytes to echinodisks and echinocytes; finally, after 5 min, all cells were converted to spherocytes. Inclusion of ZLLLal, the potent inhibitor of intracellular p C A N P mentioned above suppressed this final change and many discocytes and
Discussion
254
(A)
(B)
.... ~
¢,,.~f"
.f.+ ~
4, f,
r7~ ; S ; ~
r
rf~.~
r
r-r
'i¢>]
...-,
r
249
© 1991 Elsevier Science Publishers B.V. All rights reserved 016%4889/91/$03.50 ADONIS 016748899nlo240M
Activation of intracellular calcium-activated neutral proteinase in erythrocytes and its inhibition by exogenously added inhibitors Masami Hayashi
t, M i t s u s h i I n o m a t a ~, Y u m i k o S a i t o ~, H i s a s h i I t o 2 and Seiichi Kawashima 1
1 Department of Biochemistry. Tokyo Metropolitan hlstitute of Gerontoh;g); Tokyo flapan) and 2 Department of Chemistry. Aoyama-Gakuin Unicersity Tokyo (Japotl)
(Received 7 May 1991)
Key words: Calcium-activated;Proteinase: Auto!vileactivation;Proteinase inhibitor; Erythrocyte !ntracellnlar calcium-activated neutral proteinase (CANP) in rabbit erythrocytes was activated by an influx of Ca 2 + into the cells. The catalytic large subunit changed from the original 79 kDa form to the 77 kDa and 76 kDa forms on activation just in the same manner as occurs in the autolytic activation of purified CANP in vitro. The activation required both extracellular Ca 2 + and A23187, and was accompanied by the degradation of some membrane proteins and morphological changes in erythrocyte shape from discocytes to echinodisl~, echinocytes, and spherocytes. Exogenously added Cbz-Leu-Leu-Leu-aldehyde inhibited the activation of intracellular CANP as well as the degradation of membrane proteins and the morphological changes indicating that the latter two processes are due to the action of CANP. Leupeptin and E64d were without effect on intracellular CANP.
Introduction Calcium-activated neutral proteinase (CANP) is an intracellular proteinase that requires Ca 2 + for its activity [1]. From many experiments in vitro, the mechanism underlying the activation of C A N P has been visualized. The activation process appears to include C A N P autolysis in the presence of Ca 2+ during which the N-terminal portion of the large subunit, which inhibits proteinase activity, is removed [2-4]. In the case of the low-Ca2÷-requiring form of CANP (/.tCANP), the
Abbreviations:CANP, calcium.activatedneutral pmteinase; mCANP and/zCANP, CANPs which are active in the presence of Ca2. ions of mM and p.M order, respectively; SDS. sodium dodecyl sulfate; SDS-PAGE, polyacrylamidegel electrophoresis in the presence of SDS; ZLLLal, benzyloxycarbonyl-leucyl-leucyl-leucinealdehyde: calpeptin, benzyloxycarbonyl-leucyl-norlgucine aldehyde; E64d, (2 S,3S )-3-( S )-3-methyl-l-(3-methylbulylcarbamoyl)butylcarbamoyl oxirane-2-carboxylate; E64c, N-[N-(L.3.trans.carboxyoxirane-2-
earbonyl)-L-leu~l]-isoamylamine;kDa, kilo-daltons. Correspondence: S, Kawashima,Department of Biochemistry,Tokyo Metropolitan Institute of Gerontology, 35-2 Sakae-cho, ltabashi-ku, Tokyo 173,Japan.
molecular mass of the large subunit decreases from the original 79 kDa to 76 kDa via a 77 kDa intermediate [3]. Furthermore, CANP binds to cytoplasmic membranes in a Ca2+-dependent manner and its autolytic activation is facilitated greatly by this binding [5-9]. This activation scheme, however, is drawn from results obtained in vitro and is not necessarily applicable to the activation t;f CANP in vivo. Evidence is needed to show that intracellular C A N P behaves similarly upon Ca 2+ influx into cells. Synthetic proteinase inhibitors are powerful tools for elucidating the physiological functions of proteinases. Leupeptin and E64c are well-known inhibitors of CANP, although their actions are not restricted to CANP [10]. Leupeptin and E64c inhibit CANP quite well in vitro, but their effects on intracellular CANP when they are added extracellularly is not known. To answer the questions of whether intracellular CANP is activated in a manner similar to that expected from in vitro experiments, whether synthetic CANP inhibitors are effective on intracellu!ar CANP, and of what happens to cells when intracetlular CANP is activated, we examined erythrocytes as a simple cell model. Erythrocytes are especially suited to these stud-
250 ies because they contain only one type of CANP, ~CANP [11], whose activation can be followed easily by observing the changes in molecular weight of the catalytic large subunits [3], and because they lack a lysosomal system, the presence of which makes the analysis of intracellular proteolysis and proteinase inhibition by exogeneonsly added inhibitors complicated.
79~. f" 77K
Materials and Methods
Materials. Monoclonal antibodies specific for p.CANP were produced by immunizing female BALB/C mice with purified p~CANP from rabbit skeletal muscle and screening hybridoma cells which secrete #CANP-specific antibodies, as described before for the isolation of mCANP-specific antibodies [12]. The monoclonal anti-/~CANP (1AsA 2) used in this experiment recognized only the large subunit of /LCANP and did not cross-react with the large subunit of mCANP or with the small subunit. The ionophore A23187 and peroxidase-conjugated anti-mouse IgG (Fab') were purchased from Sigma and Medical & Biological Labs. (Nagoya, Japan), respectively. E64d was a kind gift from Taisho Pharmaceutical (Omiya, Japan). Benzyloxycarbonyl-leucyl-norleucine aldehyde (ZLnLal) and benzyloxycarbonyl-leueyl-leueyl-leucine aldehyde (ZLLLal) were synthesized according to the method of Ito et al. [13]. Methods. Erythrocytes were isolated from the peripheral blood of New Zealand white rabbit and purified by successive washing with PBS (phosphatebuffered saline). The washed erythrocytes were diluted to a concentration of 5 • 10S/ml and preincnbated with A23187 at 25 °C for 5 rain. The reaction was started by the addition of CaCI 2 solution and stopped after 15 rain by the addition of an equal volume of sodium dodeeyl sulfate (SDS) sample buffer. The samples were heated to 100 ° C for 5 rain and subjected to SDS-polyacrylamide gel electrophoresis (PAGE) [14] in 12.5% polyacrylamide for Western blot analysis of CANP and in 7.5% polyacrylamide for protein staining. CANP on nitrocellulose sheets was reacted with monoclonal 1AsA 2 antibody and then with peroxidase-conjugated anti-mouse IgG (Fab') and visualized by reaction with diaminobenzidine [15]. For analysis of CANP and proteins in the cytosol and membrane fractions, erythrocytes were lysed hypotonically after reaction and separated into eytosol and membrane fractions by centrifugation at 22000 Xg for 20 rain. The membrane fraction was washed three times with 5 mM phosphate buffer (pH 8.0) before subjection to SDS-PAGE. Morphological changes in erythrocytes fixed in 1% glutaraldehyde were observed under a differential interference microscope.
7SK
Fig. 1. Autolysisof intracellular/.~CANP in rabbit erythrocytesby treatment with A23187 and/or Ca2+. Erythrocyteswere preincubated in the presence(lanes 2 and 4) or the absence(lanes I and 3) of 20 p.M A23187 at 25°C for 5 min. The reaction was started with (lanes 3 and 4) or without (lanes 1 and 2) the addition of CaCI2 solutionto a finalconcentrationof 400 ~M and stoppedafter t5 min by the addition of an equal volumeof SDS-samplebuffer. Proteins were electrophoresedon SDS-polyacrylamidegelsand transferredto nitrocellulose membranes.The blotted proteinswere incubated with anti-CANP monoclonal antibody and visualized using peroxldaseconjugated anti-mouseIgG and diaminobenzidine.
Results
Activation of intracellular CANP in erythrocytes by treatment with A23187 and Ca e+ Rabbit erythrocytes were incubated in the presence of A23187 a n d / o r Ca 2+ and intracellular CANP was analyzed by the immunoblot method (Fig. l). N o changes in the large subunit of/~CANP were observed in the absence of A23187 and Ca 2+ or in the presence of either alone. However, when both compounds were present and Ca 2+ influx occurred, the molecular mass of the CANP large subunit decreased from the original 79 kDa form to the 77 kDa and 76 kDa forms. This conversion is identical to that observed during in vitro activation of/zCANP by Ca 2+ [3] and, therefore, it was suggested that the 76 kDa form is the active form in vivo as in vitro. The autolysis of intracellular IzCANP was dependent on the extracellular Ca 2+ concentration and was observed at concentrations as low as 6.25/xM (Fig. 2). It was reported that the Ca 2+ uptake into erythrocytes in the presence of A23187 is more than 80% [16]. This means that the autolysis of intracellular p~CANP may occur at about 5/~M Ca 2+ concentration. Since membranes have been shown to be involved in the activation of CANP [6-9], CANP binding to membranes and autolytic changes in intraceUular/.LCANP were followed (Fig. 3). The time-dependent changes of /J,CANP in the cytoplasmic fraction were almost the
251
A23187 ( - ) 1
2
3
A 23187 (+) 4
5
1
2
3
4
5
79K
~
~
~
~
~
~
77K 76K
Fig. 2. Effect of Ca~'+ con(:entration on the autolysis of gCANP in rabbit erythrocytes. Erythrocyteswere preincubated in the presence or the ahsence of 20 #M A23187 at 25 oC for 5 min. The reaction ~,.asstarted with the addition of CaCI_, solution to final concentrationsof O, 6.25, 25. 10t)and 400 #M (lanes I, 2, 3, 4 and 5, respectively).The other procedures were the same as in Fig. 1.
same as those in whole cells; the amount of activated p, C A N P increased with time. This is not surprising because more than 98% of the t o t a l / . t C A N P is recovered in the cytoplasmic fraction. A part of the p, C A N P was found bound to m e m b r a n e s mainly in the activated form. T h e s e results are consistent with the activation scheme obtained in vitro in which /zCANP binds to m e m b r a n e s in the presence of Ca 2+, autolyzes on the m e m b r a n e s , and is then released to the soluble fraction [9]. T o confirm further the involvement of membranes in the autolytic activation o f / . t C A N P , erythrocytes were lysed and half were incubated in the presence of Ca 2+. T h e other half were centrifuged to remove membranes and the membrane-free cytoplas-
mic fraction was incubated similarly in the presence of Ca 2+. Autolysis occurred more efficiently in the whole hemolysates at low Ca2. concentrations than in the cytoplasmic fractions (Fig. 4), showing that membranes, possibly acting as the activation loci, also facilitate btCANP autolysis in a crude system like this.
time
3o
0
30
0
A23187 C82+
-
+
+
+ + -F -I-
1
(A)
2
5 10 15 30 111111
+ -I-
+ + + +4--t-
+ .4-
,, =,,, ~ m . = , = l , ~ ,
(a)
(c)
i
i
!
Substrates of activated/.¢C~hVPin erythrocytes T o identify the protein species which are degraded upon activation of intracellular p.CANP, the proteins recovered in the cytoplasmic and m e m b r a n e fractions after treatment of erythrocytes with Ca 2+ and A23187 were analyzed by S D S - P A G E (Fig. 5). No changes were detected in the banding patterns of the cytoplasmic proteins. O n the other hand, in the m e m b r a n e fractions, spectrin, Band 3 and Band 4.1 decreased with time, and a new band a p p e a r e d just below spectrin, Thus, the main substrate proteins degraded by activated p.CANP are m e m b r a n e proteins.
Fig. 3. Time-course of p,CANP autolysis in •rabbit erythrocytes. Erythroc,jtes were preincubated in the presence or absence of 2O/zM A2_'H87at ~ ° C for 5 min. The reaction was started with or without the addition of CaCI2 to a final concentration of 400 .u.M and stopped after various intervalsby the addition of an equal volume of SDS-sample buffer for whole cell analysis(A). For analysesof CANP in the cytosol (B) and membrane (C) fractions, erythrocyteswere lysed hypotonicallyafter the reaction and separated into cytosol and membrane fractions by centrifugation. The other procedures were the same as in Fig. I. Numbersof er,,thr,'~..,,tesu~cdv,cic 2.3.10' for (A), 2.4"10" for (B) and 1.3.10s for (C).
252
(B)
(A) A23187
+
+
Ca 2 + (pM)
0
400
0
6
~
,m
mm
~
25
(C) 100 4 0 0
,row ~
~
0
mm
6
m
25
100
400
ew-
,ram
mm
Fig. 4. Effect of Ca z* concentration on the autolysis of/zCANP in whole cells, hemolysates and cytosol of rabbit ewthrocytes. (A) Whole cells. The incubation conditions were the same as in Fig. 1. (B) Hemolysates. Erythrocytes were lysed hypotonically and treated with various concentrations of CaC[ 2 as indicated. (C) Cytosol. Erythrocytes were lysed and separated into cytosol and membrane fractions. The membrane-free cytosol fraction was treated with various conccntralions of CaCI 2 as indicated. The other procedures were the same as in Fig. 1.
Effect of synthetic CANP inhibitors on activation of intracellular I.tCANP T o p r o v i d e definite e v i d e n c e that t h e d e g r a d a t i o n o f m e m b r a n e p r o t e i n s is really d u e to t h e activity o f activated p . C A N P , inhibition o f intracellular ~ C A N P activation was a t t e m p t e d (Fig. 6). Extracellularly a d d e d
l e u p e p t i n a n d E64d, b o t h k n o w n to b e inhibitors o f C A N P in vitro, h a d little effect on t h e autolysis o f intraceIlular p C A N P e v e n at c o n c e n t r a t i o n s o f 1 m M . C a l p e p t i n ( Z L n L a l ) , a c o m p o u n d d e s c r i b e d by Tsujin a k a et al. [17], w a s a p p a r e n t l y effective at a c o n c e n tration o f 100 p.M. T h e m o s t effective inhibitor w a s
(A) 300300 A23187 C a 2+
1
2
(B) 5101530300300
1 2
5101530
min
+ + 4 - + + + + + 4- - - + 4 - 4 - + + 4 - 4 - 4 - 4 - - - 4 - + 4 - + + + + - - - + 4 - + + + + +
200 K
spectrin
116K "" g7.4 K "
•,,, band 3 ..,,,, 4.1 "1 4.2
66.2 K I""
/
i
Fig. 5. Ca2+-induced degradation patterns of cytosol and membrane proteins in erythrocytes. Cytosol (A) and membrane (B) fractions were prepared as described in Fig. 3. The proteins were electrophoresed on an SDS-polyacrylamide gel and stained with Coomassie brilliant blue. Numbers of erythrocytes used were 1.2' 107 for (A) and 2.6.10 7 for (B).
253
(A) A23187 Ca2+
+
+ _
(a) + +
e"m ~"" ~
(e)
(D)
(E)
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
i
liii ~
-ni
WL4K
0
0
0
10 100 1 0 0 0
110100
0.1
10
i
~
spectrin
band 3
10 100 1 0 0 0
Inhibitor C o n c e n t r a t i o n ( pM ) Fig. 6. Effect of exogeneously added proteinase inhibitors on intracellular autolysis of ,~CANP and degradation of membrane proteins. Erythrocytes were preincubated at 25°C for 15 rain in the presence or absence of 20 ,~M A23187 and/or proteinase inhibitors at the concentrations indicated. The re,teflon was started by the addition of CaCI, to a final concentration of 400 pM and stopped after 5 rain. The procedures for CANP immunoblotting(upper panel) and protein staining (lower panel) were the same as in Fig. 1 and Fig. 5, respectively.(A), no proteinase inhibitor; (B), leupeptin; (C), ZLnLal; (D), ZLLLaI: (E). E64d. In (E) with 1 mM E64d, some part of erythrocyteslysed due to the presence of 10% dimethylsulfoxidewhich was used to dissolvethe inhibitor and the amount of proteins applied to SDS-PAGE was less than the others.
ZLLLal, developed in this laboratory. It inhibited the autolytie activation of intracellular IxCANP at concentrations as low as 10 I~M. The appearance of membrane protein degradation products was also suppressed by this inhibitor, showing that activated CANP is responsible for the degradation.
echinodisks remained. This result shows that activation of p.CANP is the primary event leading to the deformation of erythrocyte shape, possibly through degradation of membrane proteins.
Morphological change in the shape of erythrocytes accompanying the activation of I~CANP
The results presented here clearly demonstrate that intracellular p C A N P is activated upon increases in intracellular Ca 2÷ levels by the same mechanism as that established in in vitro experiments; i.e., t~CANP autolyzes in the presence of Ca 2+ and the molecular mass of the catalytic large subonit changes from 79 kDa to 76 kDa via a 77 kDa intermediate [3]. Moreover, membranes are involved in the activation of /zCANP in vivo as well as in vitro. That is, upon an increase in cytoplasmic Ca 2÷ levels, /~CANP is first bound to membranes where it is autalyzed by low Ca 2+ concentrations; activated/zCANP is then released into the cytoplasm. Autolysis and the effect of membranes were also observed in ruptured cells. This was rather
The shape of normal erythrocyte is discoid. Incubation with either Ca 2+ or A23187 alone produced no changes in erythrocyte shape (Fig. 7). On the other hand, when erythrocytes were incubated in the presence of both A23187 and Ca 2+, a condition in which i n t r a c e i l u l a r / x C A N P is activated and membrane proteins are degraded, erythrocytes changed their shape from the original discocytes to echinodisks and echinocytes; finally, after 5 min, all cells were converted to spherocytes. Inclusion of ZLLLal, the potent inhibitor of intracellular p C A N P mentioned above suppressed this final change and many discocytes and
Discussion
254
(A)
(B)
.... ~
¢,,.~f"
.f.+ ~
4, f,
r7~ ; S ; ~
r
rf~.~
r
r-r
'i¢>]
...-,
r
