etudes fonctionnelles et structurales de l'atpase-ca2+ du réticulum
TRANSCRIPT
HAL Id: tel-00430785https://tel.archives-ouvertes.fr/tel-00430785v2
Submitted on 12 Nov 2009
HAL is a multi-disciplinary open accessarchive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come fromteaching and research institutions in France orabroad, or from public or private research centers.
Larchive ouverte pluridisciplinaire HAL, estdestine au dpt et la diffusion de documentsscientifiques de niveau recherche, publis ou non,manant des tablissements denseignement et derecherche franais ou trangers, des laboratoirespublics ou privs.
Etudes fonctionnelles et structurales de lATPase-Ca2+du rticulum sarcoplasmique de lapin et de la protinerecombinante exprime chez Saccharomyces cerevisiae
C. Montigny
To cite this version:C. Montigny. Etudes fonctionnelles et structurales de lATPase-Ca2+ du rticulum sarcoplasmique delapin et de la protine recombinante exprime chez Saccharomyces cerevisiae. Biochimie [q-bio.BM].Universit Pierre et Marie Curie - Paris VI, 2009. Franais.
https://tel.archives-ouvertes.fr/tel-00430785v2https://hal.archives-ouvertes.fr
!"#$%"%&"'
(#)*
#+
$,"""$$&&$#$-./#&(&&0 $%$01%"
#%("02""3%0(45
!"!#$%& '!()*+,) '!(-#./% "*!(+%'0+$)$ !(.,)!"#1 -
0 2!.3456"+6758783)'#."-"94
%(:;:4! ":;:4&:;:4
!!"#$%!&!'!(!)&'(!!''!!*'$!!"+++,-.#+///0'"1!0-
!" ! !#"$
% &
'( )*+,
-.$-,-,/0
$!,$$
% $ / / 1 )*+
23/
".
$!4"$$
5
%6!73
..%$.."8
7/8$!$$
. $9.3.$
:39);
./=,
,7
&$/0$?0:
@:"$., ,
% A /B C % 3 %
4D5$ -
"7
EFG
E"G
,%%H0:
..
%/F3
% $ . . . $
.498I5,.
7$$,
% ='444!-$0
0
$ ! !"+ $ ## $5 + # 5 & 0 $ 5 $ &(***+02$0##+ 0 # 2$ & & &+.// ( ***"(>50+ **1+
&
-$$ 0 # # $ #+ # #(
;+5$&&02## # + 5 0 $ %2 $ @A$ $ & &+ ##& # $ ##$**132&+(**1"(&
-$5$#&
2#+#$#+$$2(
#5##-2&
-0 $2 0# +45/ ( !676"(B&22& + + #- 2#-+55 # $%$$++$$2&$32&+(**7"(
222&+55#-0$$$+0&%2&&0#!$2+$ #2 # #& 32&+ ( **7"+ 5#2#/&&3+:(**7"( )!(3;94"'"-"-( (3(.(9 32.(M9'!U>!"!N$$&M%2()9'">-!N&$()? 9 ' ">-! K 4"!"!' 4+ 2
&&)00"( !9 +$!(9&5&( 9(9(3)9.+.+.O+.O&&(9PC&%&!+!#&%&&"&DQ$(2922(9>&%&&"(((9#(
0
0
+0"
0
R0-'!4!'!# )!$--4-$+/!--4!?
)!( +M08'!%*4!!5-,YJ +VZR08 +V
(2# !T6
%
#(2##L !T6(2 !T6( !I*
%
%
!
)'
#
( # # ,
2
$$
($+ 2 # #+ 20#
00
+0
,22(
#
, #+ # 0#$+%2$# + 0
34567)*8979:68;793?57@A68B76>=7>C47>)93??!6E"(B9)0)
!6TE"(93@a$A2.2.!61"(
]0%#+
#(
#+00& ( ) &2+ 2 #( ) & ##0(
( . $+
@A&!**,E***)"$ &%() + 0 ( ) ##2+2 0+++2&+2= "+ 0 + 00 #
& N & ( # & 2&9 2!2 T ( 2 # 0 2 + + + = "( 40+$V+20(2 2+# 0 0 $292 ,+ 2&+ 2 2+ #(% 2&0%($##+2 #0$ 0&a
%+ & "( %+ ' $2
T
34567 )* D79 E7 =3 7;A> E8;> =7> 979:68;7> :3G=G43?57>) 0 ">44!"98F):9 2&4#"#!
T!7#0#6!79!c6""(04'"924(" # 0 $
0 $a !79*+ "(0($0 0
5# 09WW555(0("(
#( S# +$2 +
& + = "+ #202
I
29++
2&&,$&2& )?"
E
34567)*,A65JA567.@,LE76F>G=5A3G;/E56FJ7 E7 E3II68JA3G;.;N3;**/)$,#&2 4 '!'!! 9* !:0 < 2 hFGi" 0 + %
% F+ % 2# h h
( "+ %
b
U
$#0+&&+%***"(
? ( $
$$ 0 + $
$ 0 2+ 9 '#
e(
e + 0 0 ?$2;2+>&(!67!"(
#
2,%e#e$-5/+ L/ ( ***"( $2
1
$ + 0 #+0$%0
$$+0 $%%0-&(
%$%0 $$+#2 (!00$$%
L
&+2
(!661"( -02%#&+ + 0 % ( + %$$'02 &()+ 0
%&(
7
( 2 $2
%'0,#2
& %
+ 2 $$ ( 2+ $ $ +#(#!61Ub?3+ ( !67Ib >2 !67U"" 2 3+ < !67*""+ & 0( # 2 2 0 $ $$9 $ $
&B0k
6
% % #3( $
#$
( 20 #% #
(+
00$22( !66+(20
$
@%A,+ +
( !66"( 220#2$
+0$V#$+0"(+$+HTH!)!''-!4!'+##
22&3+L/(!667b.5#&+4O
( **7"( 4 # $ + # $$ & M *** 3+ 3/ ( ***"9 020:!!(!4!'""($2 E7 ?57=?57> 8?58 G:A7;57> A8==G47;M>7 7A E3II68JA3G; 85D 68HG;> %) +@$+:2(**U"(22 0 2 ( B+ @HEA >$+.(**7"0(+@$H!A+2&)!'' -!4!' .5#&+ 4O
( **7"
2&( )+ @&!A+ 0)!4! ;+L//(**6"( $!+!Ed 0 &(
$
0 %
*!" #0 #*"(+4!!%*"(
,2+02,),(24"-$ + & HE+ & H\ 0 ]2+)(**U""(420$$200(
!*
2 $
# $ $$ $ $
% $ $ 0 $ $( ($
0$
($
%% 029 0 2 #( %+ 0 % % $%0%($
$%
2 % % 0 + k
!!
$$ $ , &+ 0 $ # , 0
( #+&0#+($2$+##20$& $$ 2 % 0(
!
34567 )*+ A65JA567A63E397;>3G;;7==7 E7=S A68;>=GJ8>7 E5JP56 E7:P5I.7:8H!7H6G5=8$ 8KG5A!G;Q8=7Q7A8=) **/) W) #,U2#(
# $+ #%&& + 2 "+0$%(+0
#(B+0#(
4 0 2 2 2 .W U 5!" #+2
4+](**1bL+4(**6"(22C.D%
lC.
D
"9%
+(2#00kE(("(
o >!4 $&"0
.> ( I" #"#!*"+X"
$ 2 ( $
#%
$&B0 + k
!T
$
e
(
!"
&,@B2AB"2$
#+ 0 2 00 0+ 0 >22 !66b + ] ( **6""( $
O2 O&& O& O O%
0 # O 0 # #+ 'O%('+
&B-&+&B%00( B $ #$ 2( < 0 0 + 029%;E*7 ; + , 2 $ , 0( #- & - &+ 0 2 %+ %,%+ , ( - O+ % $ 00 # 2 0 "+ H)>"4J2"@&!!!)>+A3 !"+0 @$A
% &% 3) "+ % ' ?#/**!"(
( &B
+ # , B ( 4 , ' !* $$+ 1 & T & %
+ X ( **6b +](**6"( #2)(( #**+0)L(+0B!#0+(**"(
]**U+#,?$$&20#**!
#+3#2+**!"(+ 0%$ #' **U2+**U"+ 2@A
$$ &M % 2+**! b 2+ **T b
+ **E" & ( +
2+**1b:+**1"+00L((
!I
34567)*A65JA567A63E397;>3G;;7==7E7>A68;>E7AH
!E
# +##&%23) +32 b0"(
#( SB%% %@A#+>2+= "+0 %
00]**T""(%+$$'"0-9#0#-&
$%
0$
@2-A"(+
+%&0&($$%22>2!66!"( @&BA+X!+
00 #_ .2&+32& ( **U"'0 ,
3+#,%%++02(
o 6!!!5!!'' )0&$V &!777b3!677"+2?>92&&&2&+$2#!$$&&(
#% S #
+ 2 2%
# , %+ $ $$% %
+ % 0- 0 "(% + + +#
#!0%2 0 0 3 !677b
/% **""9 2 -&?>$$
#,%
+ 0+ 0 + )& ( **E""( $
% 20 % 2( % #+3 !(
!U
o 85!!''4!!4&)+ 3 ! % #
#
& ( %
+
$$% $ #% &%+
$$ ' '2 2 ?>" + ]/& (!66Ib3
+3'(!66I"((!66U""+X$!+
+%,#
00+$,
--&/+:(!66I"+2m,,$O%02+ --&/ ( !66Ub + ( !661"( % 3 ! 0 X$! , 0 + 0 ( !66Ub + ;/5 ( ***"( ) + ##23 !-0+X!+2#2+##
$ %(?" .2&+32& ( **U"(. 0 #
3 !$%0?>($
3 !
2
#+3?2( .+%$"2##+,020 (
# 03 ?2+ 0 , $
% # 2 0+&0$%2#
?>+%%
2( R
!1
>+L"+&h &"+F 0 0 -&%+ 2 , # 2 &-&?2**!"+$ G2
20-&?2**E"(]/,$# $
!6E1+ .+L / !6E1"+ # , &
#%.#!661( & & + @2A $ 0 & + )L?(2!E6&+2,
0 U7 , **/)" E2 2 % 2 !667b 2 % !667"( &2
, 2 0 $ "
+$_0 "(4#2$
&($
&
!7
+ 2 ,
00,00+0 -**1""($+20 +#0#0 #
2(S&2,(
( S& 9%&W%
,0
++ 0 22e0$
+',0$**I"(@A+
2
$#+!**,**3+
# !**** $ #
" %
+ ?2+ 0 - 0B2+(***"(
&2$$S$$$( $$ ' & W% $$ $ % & $
(40&29+#0b+
2 2b + # @0A + 2(]0&W%( 82#
0" 0& & 0% ',
00%022%(0
!6
0 &+ 0 % 020 "#&00%%$&"($%0$0&0#&"
%+%&22&$&(&0$%,&"#&"
9+"#!0#0%02 #
R4R24
+ 2
, &&"( %
"+ R4 0 0 2 22 , (#
$$$ 2 02+@ #A+ 2 $#0
@ A $#+,%02# 0 + #+ +R40%,& & "# 9 0 R4 0 %,&(0 22$@A"+_
+0&+_ ' , # 0 0+
+ - B2 !661b B2 **"( &( " 2 0 &0(#$2&"+22&0 0&
($9%(
)!
&'()"!
$$$& 2 $$ - + )++ ) )V+
2$,!*,'j"$$+0 !!#+ #+ T,T$B#+B%(**Eb3&L& **1"( $ ' % 0 + $V 0# $$ &
b $ !$'+ '
+ $ # #+ $ T 2 # T %
2+ + 0+0B#+B%(**E""( 2 )+ $ !67E 3
3
+ B ( !67E"+
&
*
3
+B(!67E"(82 )+$%,2+B&(&S2 % %&+2
B&!6U6"($$ , %4+(!66Ub4+(!661"+,
176 4+B ( ***" 0 , % # 3U+ S K 0 "9 + %$
(+$,2+
$
# 2 )+ $0%(
&**()"!+',
-***:
! #+_
1* , 7* f # 0 (
!
34567)HJ=7J8A8=HA3?57E7=8>7!8$8E88;E38;;8++/)$$$@!A"$%%&!""( $ , )TE!+
, $ &"+@2AT"(@2A00,SI"(2$+$$$( E"($0&U"',$1"+00,
!(
0!7"($$$$
22&2+-&$'0+
$%0 (
' # #% 0% , ! && + &+0 2#,-2J&($$'20+00
#3R
!616"+ # 2 + 3 ( !67U"+ # R!66E"(((0+03o
+](!66Ub3- ?
!661""( #2 + 0 #
+ 0 %( # 20 $0#0
(
%*',
.'
*, -!$%*
( #3
2! #&9 2 #2 66U 0 !!*/) 3
+ B ( !67E"( # + 2 + 22 !* 2 #+
2 & $ &"
+ % #
2 #3
+B(!67E"0$2
34567 ) GEM=7 >A65JA568= E7=8>7!8$ >+ I** 0( ]((b+!67*"()
#+ # 2$ 22+$,%0%+!67bB
+**!"()/#&+,$++$V
+$,0$$+ !4!!!!(@!"A/#&") L#&+**IbB+**6"(
T
e+0#;+(**!"+0$$#L'+42#(!66!b&+:(!66!b2
#2(8+,.WL#,22+ ?0 # #
+ (!66T" 0 # 3T 3T
2 0 .WL .2+R(!66I""(+0&;EU+
$%?X+\2(!667"( 0 0?+ % #@(?(R4IA 2$V 2$%#
,%$0 # 22 2+: ( !66b/!66I"(!667+
022&?"##, % #
, 7 d \2+&(!667""+0$$0%#0$2
I
& ' * -!$%/"0
,
***+ / & # & +
!+
,2+ @(!A+ &+ .// ( ***""( ) + $ 9 #%#
0(& 0 + $% ."+ & " 0 & )TE!"+ @A2"#,&0$2
E
&+0E*d&0+&+.(**Tb&**7""(% $ 3T+ 3E 31+
0 2 22+$ &;EU0&?0$2
U
#,2 # , $$ &&-&($#& 02
2
# , && ( 2 0 2 0+ +% $+ #= + 0 $2
1
A8AE7>=375DE7>E3IIF67;A7>IG697>J63>A8==3>F7>E7=8>7!88.T53==7A**+$E8.GHG>K398**O//
A8A>55A3=3>F>FA8=7;A85EG983;7
348;E>7;A>E8;>=7EG983;7979:68;3867
F>G=5A3G;.8JA57==7:/$L
GE7E7;A6F77A6FIF67;J7>
(! .3" +I+T" !8I&+.//(
***"+63+4(**U"
(!( 3 +U+E" !E+3o
(
**I"+!R;&3-**I"
(!q() 32)+;I
32
+I \B)&3-**I"+!E+3o
(**I"
(!g() 3. +7 TBU4+(**1"
B;T
32 32 +UE TB6B4+(**1" T+7T+*" \B&+.(
**1" ? +I \B;&+.(
**1"q ;I
32 ? +EE \B?&+.(**1"+!ME4+(**I"
q( ;I+3 32 T+* TB6 4+(**1"
( 32;I+32)" 32 ? +T !:?&3-
**I" 32;I
32 +UE 46]3+#(**1" 3(3 3 (3 +E T;?4+B#-(**6" ? T+!+!" !H +I ?R4#+3&(**E"
T+I 4*3+#(**1"+/+L(**1"
? +6 /+L(**1"
" +7 8/+L(**1"
( 32(3 ? +7 7L]+(**U" 32(3 32 ? +E )H(":>+%R>(@"$9Y2:
34567)OA65JA567JK393?57E7?57=?57>3;K3:3A756>E7=8>7!8)
B>H+, 2%& #t#"+ 2#& :+32(!66"(#$%#+$$?L+:(!66E"+V+0+#3T3I+#! &2+0 $#&0$2
T!
0
%#+# 0&+ 2 ( !66*b X+
( !66Tb > B **1bL'+ > ( **1b + > ( **7""( @(?(B>HA+ 2 $ % &"()% #+ e + ,% $% B>H(
& ' $ %+ + , + 0 $,#09B/+X/(!66!" 22 $$0 $
0 2 0 %
$
+
0 -& 0 0 3+3
(!61U""(
34567 )+ 3A7 E7 I3D8A3G; E7 =8 $!E3.A76A!:5AH=/!$,!:7;QGKHE6G?53;G;7 .'/ 7A E7 =8J3E7JHJ=GHA
+ )B p?R+ /+ L (**1""@((3(32;IA
B+)BpT;?4++B#-(**6""# 0 % # 31 , 3!*"( # #0 &+%, #&"(#3!,360#
!,6+!3!
,#$3!30+3T3I2+3E,3!*#"( 2+ 2 %&2+# -# #(#+?+B>H+ 2 0 #'(922++$3T+3E31+0'0&;EU0$2
T
o ! ( )! 4 '' 4 !!
TT
( $ # $ $ + &
2
],'+'
229&&)4"&2&&!7"(4+'
+2#0$V$+3(!67Tb+
3(!67U"b0(H+k
TI
22B>H#+$ @(B;TA $ & @A -&+ $
0# 220(
TE
TU
T1
' 2 + % 2 0 ( ) + ' $ 0& '
+'
+
$ # ##2 # , 0 #00( @33A
# ' , 9 ' 0
2$2
) './$!./$
'',+#
++#9 !1*f $2 , ( $2 # + ?
( !67E"+ 0
&3>#!61!"( )%,'I7+$2&2& -& # # &&2&22!" %# ( $ + m#+
,$# @#A+ $# $$ # $ - 5***""(#&2L*(!3+h&!r2(!" && 2&2 & ( #& $2 # , # 0 #
+#&T*I**B/]!*,6(E**+((!E***2%"+2$2@#A!U*1*B/IE,I*(***+((!*****2%"(#0 0
, $ 0 % , $ #"(#$ *+E3 + # #
9 $#$ $0 ## 22 0 # 02 #+0 "
+
+ 02 , &3!61E" !"(8$2,#006E**+B/]!*+&T***"
#m 22(4 L 2 , *+U3 0
% # # & 22+ $2 0 , 0
T7
%gTI***2%+ B/IE+!E1*"$$+ !( !*3.4>>1(I*+T3+0$$$m% $ 2 (
# , 7* # " !f )+ 202# #!0!,!2( $2#$ $0+
02U*,**"+,$
,$0$2 E76FA3J5=59 >86JG93?57) !+ #0 + &$+ < !61U""( + )? & $ #+ 0
# 2+ 20"(
)% ! $
(****
( 0% 0%X)U*0,2# ) ) + _ % & IE*+ ( !66U"( 0 2 ,
#+ % 0 $ $ 9 & , % %+32&(**"+#&,
% ]/+ ( **U" 0 , % .+ 0 % 0
$
#+32&(**6""( 0 , $ 0 0 00 #!+
% $2
T6
r"+ 0 $+0# 0
+E,*
(&%'%9X)U*2V
%0:T*T(!#0"(?m,2V+#
0 $09 0$0#(+8+ 9 & ( < 0 , 0+ % + @A+ 0
+ ( !66U"9 ,
00X)U*+00
0&
1(
34567)*7JA756E7D>3G;5A3=3>F3G;KFAF6G=G457E7=8>7!88E8;>=8=7B567(*)0+S$$(!"'4-!!!(!0 9(:9 2 #
4 B(" #+ 2 $ ,
" # ("&!'4!#!!#9!:920u"000 ,$#0"+%V",0$0%+ & 8"( T"54 +!9#:A 2 ! S
$?!*X!7+#22(2 !$
T002#"#&B)+B)"(@A%2&/?L"(
U-0#_+2&02%9V"+
2$!V"+2 &2'@A%!V"( 2
'@A8T"(1 2 8 0 #%& #&- #%&+ " #&(-&%&-#< "#%&-#< "()0!>+< 0
( #+ 0 %&
2 2 ,
20
(0+ + 0 @# A #
+ !661b%+**I"(7?!*X!$+$0?!?!*-+!671" 00 0:T*T(!#?I"&!>+!67E"(
I*
#( :T*T(!#W,R 0!!'"#_:T*T(!#2!%%V%4+%!V%!>%!%?"62
)?3+?$X0"+ ( !66U"( $ :T*T(!#W,R" $V, % ?I'2+ (!66U"(
$$+?I$0$$
,+"()2+:T*T(!#+?I%++ # 2 2"+ ?7*" 2 ?I + # S 0 ( ) 2
+2,R?I'+
S ,+9 ,R ?I" 0 #!,+" + & ?I"$( $ S 2 2+2$%?7*#?I+00
0 &,+ S &?I 2:T*T(!#W,R( 3 0 :T*T(!#W,R 0 $ 0 0 %+X)U*+ % !
S
&,+0?I"(#22"+0+'?7*"2
?I ( 0+ 2+ ?I 0 0 0 2 #! & ,+" % !0X)U*(
( %-&02:" 0$002:"?-!0%,).+). +,$0+0&
# 0 &&2& ?" ?-+(!66E"(0$
@UA*+!f&2#+*+1f#+!f2!+Ef2"# I7 , 1 , 7v( + % # # 0 +32&(**b**I"(0% 0 , 02+ 00 0 *+!2 #20gI2(!"(
B)$
!#!66*b+(!66T"0 ##00#!( $+ % - 0 #&
9
%&&0#%&+%-+)3(**Tb+B&(**7"+
62%+0:T*T(!#!%%V%4+%!V%!>%!%?"
$ & + + & (
-&28(,02!+2+ 0"(@?A0%r"+@>A#0+$ 2(@!A#2&% 0 @3 A @3 hA"9 0a 0 & %"+ $ 0 0 %$$+_
a(
I!
!#& .+2$(!66U"(#& 0
#+ 2
$b %#(
A!4#!'!
I
34567)*;8=H>7EF;8A568;A7>.!$K76GDHE8>7$E3IIF67;A7>FA8E7795AF7.3J3$*+'/)08>),A 2& , 7f $$ $ $(
+k" /) , 2( @!B)A , ! 0 #& $
3:N!!6/)"+ @!A , ! 0+ #&3:N!!*/)"(5+# 0 02b5+ 2 02 "b)3+ @A# 0+ #))3b)3+ # # , # !
@A+ +0
$2+2+0b. +
b>+02!3."#!**3L"b *+ !+ (++0 *"#+$#b2$0#"+# !+ "T*U*#0 #+ # ("b !T+ $ b + $$b*+!*+Er2# 0 $(08*! ! #>Y)0 + $ #R);( 0 , %&> 0 $% % #&
& 0 , &?>"(T+2#+ #& 2 0+ #9 ( #(+ %& 3:NE* /)"b&( #(+ &0 #%& 3:N!T! /)"b ! 3:NI/)"(8$ 0#!B):"(02+,#,$2
0+ :" ' ( # & #+ + + #0!B) 2$$()3 + 0+ #0+#))3b)3+0+ # 2 # 0
$29
#& ' $( B %9 7* 2 0,
+$#+1E*2!E2(B$9*+U72$, !E2 %+ IEf(. E*f , # E*f $% , 0 b + +##
+$%,(
IT
)' 0$ 2 %$+#w0,+,&&(
#,0%&.)>.)+%&,&!&-&(&@-&A$!6U*
+$/&(!6U*",$$ &-&&&(0#3o
+(!67*b+45/(!676"(
&',1*%56&78
& $2
II
#E*3+!**3L+!332+!**r3E*r3?+E332(+,>1+E*v(22 &-&+!3+*+*E*(!2(! L )>+ *+!E , *+T3 .)> ##
!,"()
+0IUr&&(2! "(!(
",9
0$ 0
( + 0 0
%$$+$00,0>+#L0%
+= ),+$+# & -& $$ @#A #0
%( )2_00
+)>$&
"(0+ L!*-&0 # % %+ 0 0 9$2
IE
)+ @2A % .)> , && ( + ,
L
wj"+),
r3#
"(
34567 )* F359./$7;=3AK359./$7;J8=J359./7A7;984;F>359./)L,*+**E2(!E*33+*+T3.)>*+! 2(! )>+ ! 3 + !** 3 L % L $ #+4!! " *332 % 32_$#+4!!8"+,>U*v($$',0,/9T3L+T3+!3332"( 0 % " , 0 .)> !E*r3 )00,!,0$E,!*r)&(2!L(!(
)
&***-:$!
#+02+2'%?+# "(? > ' #" 3
!61E"9 $ + , > 1+E !** 3 L , *v+ L(?yI%!*
7 3 # "()
?+# ,
2 + % @3%A $5 # , 9WW%($(b B+ ( !66I""+ 0
$%+0(?$$2#09
?Z(?+
IU
009 CDNCD#C(?D C?DNC?D#C(?D09 PCD#%C?D#QWC(?DNL
(?+
+29 CD#NL
(?%PC(?DWC?D#Q
3 # %+ $ % CD#22#0CD()+C(?DyCD+C?D#yC?DCD$%09
X8Y=3:67\"E8)D]X8YAGA8=S.XYAGA8=!X8YAGA8=/
% #(
0# ? $$$$+ %> $$ ? $$$(.C?DN(CD+CD#\L
(?(
!****
+%220+# 2 0 0( + & + 0
$% (>1+E+ $$ %$$# 2L(g!U*r3+L
32(g7*r3"9 0
E3+U3320E332"!**r3+#%0I*r39$00( ]00R
I1
2!E3,>U+7+*+!f)Ef &W#& 6W!+ / , T*f"( 2 + 2
*+!f$*+!f3)( ) + $2 &+!3'0&%2"(
+20*+!,E*r20*r+$E*3>7+*+*+13K+ *+f)+ +E3)+I3 *+Ef##+3o
(!66U"($&0+
$$!,6E!**v$0
+2)+2)( 2 *3+ E*3 2& !f )5W5"+,>7+U>2+'>"($2 g6%1"+ % !E*RE2+$ $$ 2 2 "( 2 + + 0 *+!f # E* 2 I*f !*f ( !*+2,,!*f+!*f!f2&(
%+*-!$%/"0
5
,#0$&$$ +$ 2 $%$( $ + # &0$$%# 0(&@2A % ( )% -& 2$
9&+L(
34567 )*0 3A7> E7 JG57!8
I7
( & & % #+ V2 $ ( & 0 & 0 & 2() & 2+
0 0 % '+ ! E*E" !67"+ 2 B+
! &5?!611"()% %TT#2 $2#
I6
2 # $ $2 & ( % 0 #
2 $ #&3o
+(**"( + !!6L!*(-&&+9&$V+$$()%$2 %$2 + ##2m%$%$2# 2"( , + %
' #
& +(**I"( 02$ # 0 # 0 2(
E*
)!%
!**
+
%+ 2 V $ 9 ( 0 % "+
$+ $(+
20%$%+ %2##"2"#9'2$# % ##+ 2 0 0$2
E!
34567)*+JKF98E776B8;A@97>5676=3;A7;>3AFE5I83>J7853;J3E7;A.EFA7JA756E76FIF67;J7$>G5B7;A5;7E3GE7/7A=85A67=8I=5G67>J7;J7F93>77;>3:=75;8J8;EG57>>8J***//)
( **'-!$%"0
2 $ 0 & #2(.+ $8R9 & &+ & & ### 8R7*"+ 2#2 $$( % 7*6* +
2 $ % 0 T*TI* + ,
&2#(
!T &+ ! #+
,$#(2$,S0$$&+2$
,0
$)!61U"92$%+$% + & + $ % $= "+ 0 $# "$#0#( ],+ 2%+ $ % 3+
( **!b ]/+ ( **Ub+ ]% ( **U"+ % 0+ # # 2( ] 0%0#02%2
#+$0(
E
& , #% 2 2
$ 0 ( 56 =8 I=5G67>J7;J7 E7 =8>77II7AE7=8JKF=8A3G;E58$>53B37E7=8I3D8A3G;E547AE54)7;8:>7;J7E7J8=J359) + *r2(! E* 3 3+ !**3L!**r3,>1*v+'# 3 ? (4 ' E332+ 32($ $2+r3"( # $ , T** r39 % 0 $$ $
$ #0 , $##"( 2 $# 0 % ,
2( 0 !** $%+#+ 0 #0 ,
+ , 0+00?(
$2
ET
& @A( 2 :EE+
+',#%0$ , 2 + 2+ "+ ]% ( **U"( 22 & 0 # $$+ + , $ (%+00',2$V0+#
+202$-&+,2+$$&&(
;** ' -!$% , ;*
*2;$"4
)
7*+ ;
EI
2+ ,,02()%
00$ # $ ( $+ 0 2 $% , + $ 2
+ 0 % , ,
B/T7*!E7*!"(
)_$%+#w
2 2 # -&" && %9%? 0
&&( 0+$,#,##V2$9 $$ 2 2 @
A$,"(0
2$20@A,()+$#+0% @
A( #+ CT>D2
( 2 2 0 ##(0+0$V2+0$%2$($0002$0@0
A" 2 $ # CT>D2( CT>D2 0 0 $0
($00$
# # $ 0 2 2 0
$ 0 "( ) %+
(!677b+&(**U"() 0 + ' % &0$% $%###+?+B>H0$2U32&+(**1""(
)
+' 2 , # +% & 2 0 (
EE
EU
E1
'+#$
0 0 ' & ( H+2$%#
2 $ 0 2" ( + 0 $V&2$+$+
0;
E6
$
+(4 , $ % $+ ' 0 $ #&
9 $ $ % @(B;TA # 2$,
#0$,2@0A$@A)/+X/(**Ib+&(**U"+0$2)32&+(**1""( %
$2 A6H @ =8I3D8A3G;E5JG9HT*r3'"!r3'"'0L;B+
2,2$
2"(0%$2%$$'(
U*
&*
8 $$ # #+ + % +$%2$$$%$&&$2
U!
34567 )* 8 IG697 Z)7)[ 7>A6F>3>A8;A7@=8JG51*v(A , !4
U
" *
UT
#( 0& + , @(?A & . **"+2 $$#$$$, L( ) @(B;TA+ % 02 +!!6L!*#.$2
UI
) % %
$0 % 0 + 0 +
,$$##+$$+#$%32,$#(
%+
?0
-!$%
&&0,02$
#
+ # && 32(3/!61"(+$$#
% % " '" $% 0,%&)TE!+)1*T)1*1+,%
& 4+ ( **Ib +3o
(**Ib&3-**Ib]+(**Ub&+.((4(""+&$%(
422220
+ %% ( 2 $$
$% % 9 #0 $$+ $$
2+2##@32(!A0+2#$%32 # 2% $ #0 ,32,#R
!61Eb?
+?2(!67b+?2(!67Tb;2+3-(!66T#"($+$ T*6H + $%
UE
.(B;T+(!667b5+L/(**E"(
34567 )* 8;> Z)7D[$ =7 984;F>359 JG;A63:57 @ 5; 6F>785 E7 =383>G;> 3;A76!KF=3J7> E8;> =7EG983;7979:68;8367)@(B;%A)BpTB6B+4+(**1""9$%#+ & 2 @A '( # 93!s30+3Ts3I2 3E,3!*2( 2 #0 3! 3+
, # -L+ 2+ &
(
$ 0 & $$% +#,2(+
+!61U"(
UU
32+ $%@(B;TA# E32+ # % @&2A
"( 4+ B 5 2 20 2 $%%4'B5!677"+,>00 # 0 0 "( $%# 32 0 (B;T #
# 32W S + # 32+ @/A+0,,S(.$+220$%#3I+3E+3U37"+
026*3(
?*@?0 "%!
$ $$ , % % $$ # # 9 # , 2(+ ',0$2
U1
34567)*083;K3:7JG9A6H359b79:=7cA67E85497;A76=3;A768JA3G;E58B7J=81*v($$$$94!!9!**r3+!3?+*332%!3 (
B9 #?
329#!r2(!?B"+#*+Tr2(!"(2%$%,6*
,T*+$ &,
2?
+?2(!67"(2$#%$$($2$2E32&+(**1"(
) % + 0 32 0#%#()%+,$%#2+##+ 0 #0 $$ # 32$2
U7
34567)*87A=73;K3:7;AJG9A6H79:=7 cA67 E85497;A76=3;A768JA3G; E5 8B7J =8 1*v($$$$94!!9 !**r3+ !3?+E332 0Tr3+T*r3$T**r3.(%$$$0(
U6
0 2% $ 0 #0$2
1*
34567)*+G37E8JJM>>53G;>J8=J35985D>3A7>E7A68;>G=)#
@(!A , 2+ &+ .// ( ***"" @(!()(;IA+ 2 $ @(!()A , + + 3o
(**I""+0$(3!30m2+ # 2(
1!
!!"#
#$%#
#
32&+3(+?(+?(+&(
B&**1+IU9!E!U!E!1I(
1
Subscriber access provided by CEA
Biochemistry is published by the American Chemical Society. 1155 Sixteenth StreetN.W., Washington, DC 20036
Article
Inhibitors Bound to Ca2+
-Free Sarcoplasmic Reticulum Ca2+
ATPaseLock Its Transmembrane Region but Not Necessarily ItsCytosolic Region, Revealing the Flexibility of the Loops
Connecting Transmembrane and Cytosolic DomainsCdric Montigny, Martin Picard, Guillaume Lenoir, Carole Gauron, Chikashi Toyoshima, and Philippe Champeil
Biochemistry, 2007, 46 (51), 15162-15174 DOI: 10.1021/bi701855r
Downloaded from http://pubs.acs.org on January 23, 2009
More About This Article
Additional resources and features associated with this article are available within the HTML version:
Supporting Information Links to the 1 articles that cite this article, as of the time of this article download Access to high resolution figures Links to articles and content related to this article Copyright permission to reproduce figures and/or text from this article
http://pubs.acs.org/doi/full/10.1021/bi701855r
Inhibitors Bound to Ca2+-Free Sarcoplasmic Reticulum Ca2+-ATPase Lock ItsTransmembrane Region but Not Necessarily Its Cytosolic Region, Revealing the
Flexibility of the Loops Connecting Transmembrane and Cytosolic Domains
Cedric Montigny,, Martin Picard,,,| Guillaume Lenoir,,, Carole Gauron,, Chikashi Toyoshima, andPhilippe Champeil*,,
CNRS, URA 2096 (Proteines Membranaires Transductrices dE nergie), F-91191 Gif-sur-YVette, France, CEA, iBiTecS(Institut de Biologie et Technologies de Saclay), F-91191 Gif-sur-YVette, France, and Institute of Molecular and
Cellular Biosciences, UniVersity of Tokyo, Bunkyo-ku, Tokyo 113-0032, Japan
ReceiVed September 11, 2007; ReVised Manuscript ReceiVed October 16, 2007
ABSTRACT: Ca2+-free crystals of sarcoplasmic reticulum Ca2+-ATPase have, up until now, been obtainedin the presence of inhibitors such as thapsigargin (TG), bound to the transmembrane region of this protein.Here, we examined the consequences of such binding for the protein. We found that, after TG binding,an active site ligand such as beryllium fluoride can still bind to the ATPase and change the conformationor dynamics of the cytosolic domains (as revealed by the protection afforded against proteolysis), but itbecomes unable to induce any change in the transmembrane domain (as revealed by the intrinsicfluorescence of the membranous tryptophan residues). TG also obliterates the Trp fluorescence changesnormally induced by binding of MgATP or metal-free ATP, as well as those induced by binding of Mg2+
alone. In the nucleotide binding domain, the environment of Lys515 (as revealed by fluorescein isothiocyanatefluorescence after specific labeling of this residue) is significantly different in the ATPase complex withaluminum fluoride and in the ATPase complex with beryllium fluoride, and in the latter case it is modifiedby TG. All these facts document the flexibility of the loops connecting the transmembrane and cytosolicdomains in the ATPase. In the absence of active site ligands, TG protects the ATPase from cleavage byproteinase K at Thr242-Glu243, suggesting TG-induced reduction in the mobility of these loops. 2,5-Di-tert-butyl-1,4-dihydroxybenzene or cyclopiazonic acid, inhibitors which also bind in or near thetransmembrane region, also produce similar overall effects on Ca2+-free ATPase.
Sarcoplasmic reticulum (SR)1 Ca2+-ATPase (SERCA1a)is an ion pump belonging to the family of P-type ATPases,responsible in muscle cells for active transport of Ca2+ fromthe cytosol into the sarcoplasmic reticulum lumen (see, e.g.,refs1-6 for review): the resulting reduction of the cytosolicfree Ca2+ concentration triggers muscle relaxation. This
active transport requires ATP hydrolysis as the energy source,and its catalytic cycle has been studied at length. At somestep during ATP hydrolysis, the pumps affinity for Ca2+
changes from high to low, coupled with topological reori-entation of the binding sites from one side of the membraneto the other (with intermediate occluded species). Thecatalytic cycle comprises transitions between various inter-mediates, some of which are phosphorylated (hence the nameof the P-type family) at a critical aspartate residue in thecatalytic site (e.g., the so-called Ca2E1P or E2P forms). In2000, the detergent-solubilized purified ATPase was crystal-lized in the presence of Ca2+, and in subsequent years otherforms of the ATPase, thought to be fair analogues of variouscatalytic intermediates, were also obtained, allowing elucida-tion of the main structural features required for ATP-drivenion pumping (7-18).
However, the traditional question of the extent to whichcrystalline structures reflect the more dynamic structures ofan enzyme during its catalytic cycle can of course be raised(see, e.g., refs19 and20). This is especially the case for thestructures obtained in the absence of Ca2+, as the crystalscorresponding to these forms have been obtained, until now,by compensating the instability of detergent-solubilizedATPase in the absence of Ca2+ (see, e.g., refs21-24) bythe addition of inhibitors (such as those mentioned in refs
We thank the Human Frontier Science Program Organization forsupport, Grant RGP 0060/2001-M.
* To whom correspondence should be addressed. Phone: 33-1-6908-3731. Fax: 33-1-6908-8139. E-mail: [email protected].
CNRS. CEA.| Present address: Laboratoire de Cristallographie & RMN Bio-
logiques, CNRS UMR 8015 and Universite Paris Descartes, 4 avenuede lObservatoire, 75270 Paris Cedex 06, France.
Present address: Department of Membrane Enzymology, BijvoetCenter and Institute of Biomembranes, Utrecht University, Padualaan8, 3584 CH Utrecht, The Netherlands.
University of Tokyo.1 Abbreviations: SR, sarcoplasmic reticulum; ATPase, adenosine
triphosphatase; E2 and E1, nonphosphorylated forms of Ca2+-ATPase;E2P and E1P, phosphorylated forms of Ca2+-ATPase, respectivelyADP-insensitive or ADP-sensitive; TG, thapsigargin; BHQ, 2,5-di-tert-butyl-1,4-dihydroxybenzene; CPA, cyclopiazonic acid; prK, proteinaseK; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electro-phoresis; EGTA, [ethylenebis(oxyethylenenitrilo)]tetraacetic acid; Mops,4-morpholinepropanesulfonic acid; Tris, tris(hydroxymethyl)ami-nomethane; Mes, 4-morpholineethanesulfonic acid; FITC, fluorescein5-isothiocyanate.
15162 Biochemistry2007,46, 15162-15174
10.1021/bi701855r CCC: $37.00 2007 American Chemical SocietyPublished on Web 12/01/2007
25-27). Note that crystals of a given form of Ca2+-freeATPase obtained in the presence of different inhibitors(thought to act on the protein similarly) reveal slightlydifferent structures (15-17). In a previous work, we haveshown that adding inhibitors such as thapsigargin (TG), 2,5-di-tert-butyl-1,4-dihydroxybenzene (BHQ), or cyclopiazonicacid (CPA) to Ca2+-free ATPase complexed with BeFx (atransition-state analogue of E2P) indeed prevents theATPase Ca2+ binding pocket from opening toward the lumen,in contrast with what is considered to happen with thenormal E2P form, and with the unfortunate consequencethat, in the presence of such inhibitors, we may never beable to obtain crystals showing the ATPase structure withanopenCa2+ release pathway from the transport sites towardthe lumen (28).
In the present work, we further document the effects ofTG, BHQ, and CPA on Ca2+-free ATPase. By monitoringthe fluorescence changes occurring either in one of thecytosolic domains or within the transmembrane domain(thanks to a fluorescein label attached to Lys515 in thenucleotide binding domain or to the ATPase intrinsictryptophan residues, respectively) upon addition of specificligands such as fluoride, vanadate, Mg2+, or ATP, we nowshow that the presence of TG or related inhibitorsfixesthestructure of the ATPasetransmembrane domainand preventsit from experiencing the changes which would normally haveoccurred in the absence of the inhibitor,without necessarilypreventing binding events in the cytosolic domains fromoccurring. This is presumably a consequence of the likelyflexibility of the loops connecting the A-domain and thetransmembrane part of the ATPase and probably explainswhy the various structures previously described for crystallineCa2+-free ATPase forms have all been found to be verysimilar in their transmembrane domains, irrespective of thebound ligands and conformations in their cytosolic domains.By monitoring the proteolytic susceptibility of a loopintervening between the ATPase transmembrane and cyto-solic domains, we also document the effect of the inhibitorson the protein average structure or dynamics.
EXPERIMENTAL PROCEDURES
Chemicals.TG (the stock solution was 1 mg/mL inDMSO, i.e., about 1.5 mM) was from VWR International(no. 586005, Fontenay-sous-bois, France), BHQ (the stocksolution was 22 mg/mL in DMSO, i.e., 100 mM) was fromAldrich (no. 11,297-6, Saint Quentin Fallavier, France), andCPA (the stock solution was 10 mg/mL in DMSO, i.e., about30 mM) was from Sigma (no. C1530, Saint QuentinFallavier, France). Proteinase K (the stock solutions were 6or 1.5 mg/mL in 10 mM Tris-Cl, pH 7.4) was from RocheDiagnostics (no. 745723, Meylan, France).
Proteolysis by Proteinase K.For controlled proteolysis,SR membrane vesicles (prepared as in refs28 and29) weretreated with proteinase K (prK) under conditions describedin the corresponding figure captions. As previously (30, 31),proteolysis was stopped by adding 0.5-1 mM PMSF to thesamples and transferring them on ice for at least 10 min.After proteolysis arrest, the samples were diluted 10-fold intoa 4 M urea-containing denaturation buffer (otherwise asdescribed in ref32) and boiled for 60 s, and aliquots wereloaded onto an SDS-PAGE Laemmli gel (33) prepared in
the presence of 1 mM Ca2+ (8-12% acrylamide, as needed).Depending on the experiment and on the desired level ofresponse linearity, 4 or 2g of protein was loaded per lane;after Coomassie Blue staining, the intensity of the individualpeptide bands was deduced from gel scanning. The electro-phoresis miniprotean 2 supply, the GS700 densitometer, andMolecular Analyst and Quantity One software were providedby Bio-Rad SA (Marnes-la-Coquette, France); low molecularweight (LMW) markers were from GE Healthcare (Velizy,France).
Fluorescence Measurements.Fluorescence measurementswere carried out with a SPEX Fluorolog spectrofluorometerprovided by Horiba/Jobin-Yvon (Longjumeau, France). Theintrinsic fluorescence of the ATPase was recorded withexcitation and emission wavelengths set at 290 nm (or298 nm; see the figure captions) and 330 nm (or 320 nm),respectively (bandwidths were 2 and 10 nm, respectively),with SR membranes at 0.1 mg/mL protein in a thermostatedand continuously stirred cuvette (see, e.g., ref19). Extrinsicfluorescence of FITC-labeled SR membranes (suspended at0.02 mg/mL protein) was also recorded (ATPase labelingwith FITC had been performed as described (34), except thatFITC had been initially diluted in dimethyl sulfoxide(DMSO) instead of dimethylformamide); excitation andemission wavelengths were then set at 495 and 520 nm(bandwidths were 2 and 10 nm, respectively).
Filtration Measurements.SR membranes were first sus-pended at 0.1 mg/mL protein, with or without inhibitors (TGat 1 g/mL (about 1.5M), CPA at 0.3 g/mL (about1.35 M), or BHQ at 30M), in a Ca2+-free buffer in thepresence or absence of Mg2+ at pH 7 and 20C (see thefigure captions). Then, at time zero, 30M [-32P]ATPtogether with 300M [3H]glucose was added from aconcentrated solution (already containing the two isotopes).After various periods, aliquots (2.5 mL, i.e., about 0.25 mgof protein) were loaded onto two HA filters (Millipore, no.HAWP02500, Saint Quentin en Yvelines, France) on top ofeach other, under vacuum. The two filters, without rinsing,were counted individually in a scintillation counter, with [3H]-glucose in each filter serving as a marker of the backgroundunbound ATP (trapped in this filter together with the wettingvolume). In the upper filter, unbound ATP was subtractedfrom the total ATP, to give the amount of ATP boundspecifically to the ATPase. The lower filter served to checkthat SR membranes were adsorbed onto the upper filter only(see, e.g., ref28 or 34 for related filtration experiments).
RESULTS
Effects of TG or Other Inhibitors on E2P-Related Formsof Ca2+-Free ATPase: Proteolysis and Intrinsic Fluores-cence Experiments.In a remarkable series of papers, Dankoand colleagues previously demonstrated that Ca2+-freeATPase becomes fully resistant to proteolysis by proteinaseK after formation of E2P-related species from either ortho-vanadate or fluoride complexes (or, to some extent, fromphosphate itself) (35, 36). These authors also mentioned thatprotection by such phosphate analogues was still observedin the presence of TG. As a preliminary check, we repeatedthose experiments under our own favorite conditions (Supple-mental Figure S3 in ref28) for forming E2P-like species,i.e., at pH 7 in the presence of KCl and the additional
Effects of TG, BHQ, or CPA on SERCA1a Biochemistry, Vol. 46, No. 51, 200715163
presence of 20% DMSO, known to stabilize formation ofE2P from Pi (37).
The results illustrated in Figure 1A extend the previousresults of Danko et al. to these experimental conditions andshow that the ATPase indeed becomes fully resistant toproteinase K after reacting with aluminum or berylliumfluoride (a few other incubation conditions, e.g., withmagnesium fluoride or vanadate, are illustrated in Figure S1of the Supporting Information) and remains so if TG is addedsubsequently. TG binding under these conditions is provenby its inhibitory effect on Ca2+ binding to the luminal sideof the ATPase (28) and also by its effects on fluorescencelevels, as shown below. We also confirmed under the sameconditions the more modest protective effect of Pi itself
(despite the fact that, thanks to the presence of DMSO, Piaddition here allows almost full phosphorylation), as wellas the fact that this effect of Pi was disrupted by TG (35,36). This more modest protective effect and its subsequentdisruption by TG are obviously due to the dynamic natureof the E2 to E2P equilibrium, combined with the fact thatTG reduces the extent of the reaction of E2 with Pi (25, 27,38). Most importantly for our purpose, we then found thatthe blockade of ATPase proteolysis was independent ofwhether TG had been added after orbeforethe fluoride salt(as shown by the identical results in panels A and B of Figure1). This implies that the prior presence of TG doesnotprevent the cytoplasmic domains of ATPase from reactingwith aluminum or beryllium fluoride in a subsequent step.
FIGURE 1: Fluoride-containing E2P-like forms (but not E2P itself) are fully protected from cleavage in the presence of TGirrespectiVe ofwhether TG has been added after or before the fluoride salt, implying that the prior presence of TG doesnot prevent the ATPase fromreacting with the fluoride compound. (A) SR membranes (1 mg of protein/mL) in 100 mM KCl, 20% DMSO, 150 mM Mops-Tris, 0.5mM EGTA, and 5 mM Mg2+ at pH 7 and 20C were incubated in the absence (lanes 2 and 3) or presence of various additions: 0.05 mMAlCl3 and 1 mM KF (lanes 4 and 5), 0.05 mM BeCl2 and 1 mM KF (lanes 6 and 7), or 2 mM Pi (lanes 8 and 9). Incubation lasted 5-10min. Similar incubations in the presence of KF and Mg2+ alone, KF in the absence of free Mg2+, BeCl2 without KF, or orthovanadate areshown in Figure S1 of the Supporting Information. Then, TG (0.01 mg/mL) was added to some of these samples (lanes 3, 5, 7, and 9) for5-10 min of incubation again. Last, proteinase K was added to all samples, at 0.3 mg/mL, and proteolysis took place for 60 min. (B)Similar experiment, except that TG was added first to the Ca2+-deprived SR membranes, followed after 5-10 min by fluoride (togetherwith Al or Be) or Pi and only then, 5-10 min later, by proteinase K. In both cases, after proteolysis arrest and sample denaturation, 2gof protein per lane was loaded for SDS-PAGE, here on a 9% acrylamide gel. Lanes 10 contain molecular mass markers (LMW Pharmaciakit). Lanes 1 contain intact SR (2g of protein per lane, again) in the absence of proteinase K.
15164 Biochemistry, Vol. 46, No. 51, 2007 Montigny et al.
From the results in ref28 we know that adding TG (orother inhibitors) to Ca2+ -free ATPase after the formationof ATPase complexes with metal fluoride prevents thenormal opening of the Ca2+ binding sites toward the luminalmedium. To further characterize this influence of inhibitorson the ATPase transmembrane domain, we studied theconsequences of TG addition on the ATPase intrinsicfluorescence, a fluorescence due to Trp residues mainlylocated (12 out of 13) in the transmembrane section of theATPase. Related experiments performed in the absence ofDMSO have previously been reported (36, 28), but thanksto the presence of DMSO we were here able to directlycompare E2P-like species, formed from fluoride salts, withthe true E2P species itself, formed from Pi. Formation ofthe transition-state analogue E2AlF4 (from KF and AlCl3)only led to a slight relatively slow fluorescence decrease(Figure 2B), as previously found in the absence of DMSO(28). In contrast, the intrinsic fluorescence level of theATPase rose,both when the E2P phosphoenzyme wasformed from Pi (Figure 2D) and when the E2BeFx analoguewas formed from KF and BeCl2 (Figure 2C). This confirmsthe previous conclusion that, from the Trp fluorescence pointof view, the E2BeFx species is a much closer relative to
the true E2P state than the E2AlF4 transition-state analogue(28, 36).
Figure 2 then also shows the effect of TG addition to thesevarious species: TG addition reduced the ATPase intrinsicfluorescence level much more for E2BeFx or E2P than forE2AlF4 or E2, so that the final levels were not very different.Because of our proteolysis results in Figure 1, we caninterpret the rather similar final fluorescence levels in theabsence or presence of Pi (Figure 2A,D) as simply reflectingthe quasi-disappearance of E2P in the presence of TG;however, the again similar final fluorescence level in Figure2C implies that TG addition to the ATPase complex withBeFx has modified the transmembrane region of this complex,to make it closely resemble that of the fluoride-free complex(consistent with closure of the Ca2+ release channel),despitethe fact that BeFx is not displaced from the cytoplasmicdomains by TG.
Conversely, and to make this even more clear-cut, Figure3 shows the result of a complementary experiment whereTG was added from the start and BeFx was added at a laterstep only: after the prior addition of TG, subsequent additionof fluoride and beryllium no longer inducedany change inTrp fluorescence (Figure 3B; compare with Figure 3A),although from Figure 1B we know that susceptibility toproteolysis by proteinase K is fully blocked and hence
FIGURE 2: In contrast with what happens upon formation of E2AlF4 (or E2MgF4), formation of E2BeFx or E2P raises Trpfluorescence. Further addition of TG reduces it, much more forE2BeFx or E2P than for E2AlF4 or E2, so that the final levels aresimilar. The fluorometer cuvette contained 0.1 mg/mL SR in100 mM KCl, 20% DMSO, 5 mM Mg2+, and 50 mM Mops-Trisat pH 7 and 20C. Sequential additions were made, when indicatedby arrows: Ca2+ (100M) and EGTA (2 mM) (in all panels), KF(1 mM) and AlCl3 (B) or BeCl2 (C) (both at 50M) or Pi (2 mM)(D), and last TG (1g/mL each addition, in all panels). Theexcitation wavelength was 290 nm, and the emission wavelengthwas 330 nm (bandwidths were 2 and 10 nm, respectively). In thevarious panels, the first addition of TG saturates the ATPase(1 g/mL TG corresponds to about 1.5M TG, whereas 0.1 mg/mL SR corresponds to 0.5-0.7 M ATPase), while each furtheraddition induces an additional small quenching of fluorescence,mainly due to an inner filter effect (see also refs25 and36). Thetraces have been corrected for the (very small) artifacts inducedby dilution.
FIGURE 3: ATPase preincubation with TG, CPA, or BHQ (withpoor affinity for the latter inhibitor)preVentsthe membranous Trpresidues from experiencing fluorescence changes following forma-tion of E2BeFx. The fluorometer cuvette contained 0.1 mg/mL SRin 100 mM KCl, 20% DMSO, 5 mM Mg2+, and 50 mM Mops-Tris at pH 7 and 20C, as for the experiment illustrated in Figure2. Sequential additions were made, when indicated by arrows: Ca2+(100M) and EGTA (2 mM) (in all panels) and then KF (1 mM)and BeCl2 (50 M) directly (A), TG (1g/mL, i.e., about 1.5M,twice) followed by KF and BeCl2 (B), BHQ (30M) followed byKF and BeCl2 (C), or CPA (0.3g/mL, i.e., about 1M) followedby KF and BeCl2 (D). As for the experiment illustrated in Figure2, the excitation wavelength was 290 nm and the emissionwavelength was 330 nm (bandwidths were 2 and 10 nm, respec-tively). The traces have been corrected for the (very small) artifactsinduced by dilution.
Effects of TG, BHQ, or CPA on SERCA1a Biochemistry, Vol. 46, No. 51, 200715165
reaction with the fluoride compoundhasoccurred. In otherwords, upon reaction of BeFx with the ATPase catalytic site,the transmembrane region of the ATPase complex with TG,at least as evidenced by the Trp residues, hasnotexperiencedany change, although one of the loops connecting thetransmembrane and the cytosolic domainshasexperienceda drastic change. Similar results were obtained after reactionof ATPase with aluminum fluoride, with the same conclusion(data not shown). These results make it possible to under-stand that previous crystals of Ca2+-free ATPase complexedwith TG and fluoride salts revealed transmembrane domainspretty similar to those in the absence of fluoride,despitethefact that the cytosolic domains were different (8, 12, 13):TG fixes the ATPase transmembrane domain irrespective ofwhat happens in the cytosolic domains following binding ofthe fluoride compound (AlF4 or MgF4 in previously publishedcrystalline forms) and thereby may mask part of theinformation that would have been derived from inhibitor-free crystals.
Almost similar results were obtained using other well-characterized inhibitors of Ca2+-ATPase, instead of TG,either BHQ or CPA (Figure 3C,D). In the case of BHQ,however, concentrations of BHQ much larger than stoichio-metric (30 M in Figure 3C) were required to preventfluorescence changes upon addition of BeFx almost com-pletely (2 M BHQ hardly changed anything, data notshown). This fits with the relatively poor affinity with whichunder similar conditions BHQ was found to prevent luminalopening of the Ca2+ binding sites (28). In contrast, and againconsistent with the45Ca2+ binding measurements in (28),CPA was effective at 1M; in this case, the fluorescencerecording suggests that the binding of CPA (or the resultingblockade of transmembrane segment movements) is notinstantaneous, but requires a few minutes of incubation (seealso below). For both BHQ and CPA, addition of theinhibitor resulted in an apparent drop in Trp fluorescence,while as for TG, subsequent additions induced smaller drops(not shown). This is because all these inhibitors have a dualeffect on ATPase intrinsic fluorescence: they perturb it bybinding to the ATPase and simultaneously induce inner filtereffects due to their own absorbance in the UV region (datanot shown). BHQ also has some weak fluorescence itself,which counterbalances in part the inner filter effect andcomplicates more detailed analysis (data not shown).
Effects of TG or Other Inhibitors on a Reporter Group(Fluorescein) at Lys515 in the Nucleotide Binding Domain.From the above results, we concluded that in the presentlyavailable crystals where Ca2+-free ATPase is complexed withAlF4 or MgF4 andTG, TG has probably driven the ATPasetransmembrane domain into a fixed state irrespective of whathad happened in the cytosolic domains following bindingof the fluoride compound (see, e.g., refs8, 12, and13). Onequestion, then, is to which extent, in these crystals, TG couldhave altered the organization of the cytosolic domains too,despite the fact (Figure 1) that ATPase reaction with thefluoride compound had remained possible. To address thisissue, we explored the response to the addition of TG of anextrinsic label located in the cytosolic region, namely, FITCbound at Lys515 in the nucleotide binding domain. TG hasalready been reported to block the FITC fluorescence changesinduced by Ca2+ addition or removal, but reports of the effectof TG per se on FITC fluorescence are scarce in the literature
(see, e.g., refs25 and66).As a preliminary result, under the exact same conditions
as those documented above, i.e., in the presence of DMSOand KCl at pH 7, Figure 4 first shows the previouslyundocumented fact that the response of FITC fluorescenceto ATPase reaction with BeFx is a drop in FITC fluorescence(Figure 4C), similar in sign to its response to ATPase reactionwith Pi (Figure 4D, a previously undocumented drop too),whereas ATPase reaction with AlF4 triggers a response ofopposite sign (Figure 4B, a rise in FITC fluorescence, rathersimilar to that classically observed (39) upon ATPase reactionwith another putative transition state, orthovanadate; seeSupporting Information Figure S2A). Thus, the E2BeFxspecies again appears to be more similar to the true E2Pspecies than to the E2AlF4 or E2VO4 species. When TGwas added to the FITC-labeled E2BeFx species, FITCfluorescence rose quite significantly (Figure 4C); in contrast,there was only a slight change upon addition of TG to E2AlF4 (Figure 4B) (while TG addition to E2P (Figure 4D)increased FITC fluorescence back to the level found in theabsence of Pi (Figure 4A), as expected from the dynamicnature of phosphorylation from Pi).
From these results, we may infer that the ATPase structurearound Lys515 in the previously published TGE2AlF4crystals (13) probably didnotsuffer much from the presenceof TG, but the ATPase structure in a putative TGE2BeFx
FIGURE 4: In E2BeFx (or E2P), the fluorescence of FITC boundto Lys515differsfrom that in E2AlF4 (or E2VO4, not shown). FITCfluorescence hardly responds to TG addition to E2AlF4 (or E2VO4) but responds quite significantly to TG addition to E2BeFx.The fluorometer cuvette contained 0.02 mg/mL FITC-SR in100 mM KCl, 20% DMSO, 5 mM Mg2+, and 150 mM Mops-Tris at pH 7 and 20C. Sequential additions were made, whenindicated by arrows: Ca2+ (20 M) and EGTA (0.2 mM) (in allpanels), KF (1 mM) and AlCl3 (B) or BeCl2 (C) (both at 50M)or Pi (2 mM, followed by another identical addition of 2 mM) (D),and TG (1g/mL each addition, in all panels). The excitationwavelength was 495 nm, and the emission wavelength was520 nm (bandwidths were 2 and 10 nm, respectively). In the variouspanels, the first addition of TG saturates the ATPase. The traceshave been corrected for the (small) artifacts induced by dilution.Effects of TG addition to the E2VO4 species are shown inSupporting Information Figure S2.
15166 Biochemistry, Vol. 46, No. 51, 2007 Montigny et al.
crystal, if such a structure were to become available, mightdeviate more significantly, even in the cytosolic region, fromits TG-free counterpart, again assuming the latter could alsobecome available. It is however fair to recall that in previousexperiments where TNP-AMP bound in the cytosolicdomain, instead of FITC, was used as the reporter group,addition of TG to the E2BeFx species left hydrophobicityaround the TNP group essentially unaltered (36); thus, thoseresults with TNP-AMP point to the possibility that the effectof TG on the cytosolic domains suggested by our FITC datamight remain rather limited. Nevertheless, consistent with ageneral effect of TG at the nucleotide binding site (wherethe fluorescein is located), it may be recalled that TG isknown to reduce the ATPase affinity for MgATP (14, 40-42; see also below).
Focusing on the E2VO4 and E2AlF4 species, whoseformation is most clearly revealed by the fluorescein reportergroup, formation of these species was not prevented by thepreliminary addition of TG (Supporting Information FigureS2C,D), consistent with the proteolysis results. This was truealso in the absence of DMSO (Supporting Information FigureS3), i.e., under conditions where the fluorescence riseassociated with VO4 binding has been extensively described(39). Note that TG-dependent and VO4-dependent FITCsignals are significantly larger in the absence than in thepresence of DMSO (as shown by comparing Figure S3A withFigure S2A and Figure S3C with Figure S2C), which makesthe former conditions more favorable to study the effects ofTG or VO4 on FITC fluorescence. Other inhibitors influencedFITC fluorescence in a manner qualitatively similar to thatof TG: they all increased the basic fluorescence level ofFITC but did not prevent formation of the transition-stateanalogue (Supporting Information Figure S4). BHQ wasagain less effective than TG or CPA, as at a concentrationof a few micromolar it blocked the effect of subsequentaddition of Ca2+ only partially.
Effects of TG or Other Inhibitors on Other Forms of Ca2+-Free ATPase, As Deduced from Intrinsic FluorescenceResponses to Mg2+ or ATP.We then explored the effect ofTG, BHQ, and CPA on two other events: binding of Mg2+
and ATP to Ca2+-free ATPase. These two events are knownto give rise to distinct Trp fluorescence changes, which wereobserved here in the absence of DMSO and KCl. The Trpfluorescence changes accompanying Mg2+ addition at neutralor alkaline pH (43) were completely obliterated in thepresence of TG (part A versus part B of Figure 5). This wasalso essentially true in the presence of BHQ at 10M (Figure5C), but again BHQ appeared to bind with relatively pooraffinity, as the rate of its dissociation upon subsequentaddition of excess Ca2+ was much faster than for TG (atonly 3 M BHQ, Trp fluorescence remained sensitive toMg2+ and the rate of the subsequent Ca2+-induced rise wasfaster than at 10M BHQ, while at 30M BHQ, this ratewas slowest, data not shown). In the presence of CPA, itturned out that addition of Mg2+ after CPA induced a furtherdrop in Trp fluorescence (Figure 5D): this peculiar featureprobably reveals that CPA itself binds tighter (or to a largerextent) to the ATPase at millimolar concentrations of Mg2+
than at the micromolar concentration found in nominallyMg2+-free buffer and hence is more efficient in quenchingTrp fluorescence (see the Discussion).
We then studied the effects of inhibitors on the Trpfluorescence changes that normally accompany ATP additionto Ca2+-free ATPase (44, 45, 34, 19). In the presence ofMg2+, where the affinity for ATP (MgATP, in fact) is highin the absence of TG, TG completely inhibited thesefluorescence changes (Figure 6A,C), a fact which waspreviously noted (46) but only briefly commented upon asreflecting the known TG-induced drop in the affinity of theATPase for MgATP (46, 41, 67). Remarkably, however, thiswas also the case in the total absence of free Mg2+ (Figure6B,D), i.e., under conditions where the binding of ATP (withpoorer affinity) has been suggested to be almost unalteredby TG (47). In relation with that suggestion, and because ofthe existence of seemingly contradictory results obtainedunder different ionic conditions (14), we performed [-32P]-ATP binding experiments under our ionic conditions (usinga filtration assay), and we were able to confirm the suggestionin ref 47: at the relatively high concentration of [-32P]ATPwe used (30M), i.e., with the ATPase nucleotide bindingsite essentially fully saturated in the presence of Mg2+ andonly partially in the absence of Mg2+, TG hardly affectedthe amount of ATP bound in the absence of Mg2+ (and theamount of ATP bound to the ATPase-TG complex was infact higher in the absence of Mg2+ than in its presence;
FIGURE 5: TG (as well as BHQ) completely obliterates the responseof ATPase Trp residues to the addition of Mg2+ to Ca2+-freeATPase. The response to Mg2+ in the presence of CPA is maskedby enhanced CPA interaction with the ATPase following additionof Mg2+. The fluorometer cuvette contained 0.1 mg/mL SR in150 mM Mops-Tris at pH 7 and 20C. Sequential additions weremade as follows, when indicated by arrows: Ca2+ (100M), EGTA(1 mM), inhibitor if any [in panel B TG (1g/mL), in panel CBHQ (10 M), and in panel D CPA (0.3g/mL)], then Mg2+(20 mM), and last (double arrow) Ca2+ again (at 1 mM, addedtwice in panel A, which is the control). The excitation wavelengthwas 290 nm, and the emission wavelength was changed to 320 nmto optimize the Mg2+-dependent signal (43); bandwidths were2 and 10 nm, respectively. This experiment was performed at pH7 in the absence of KCl, but similar results were also obtained atpH 7.5 (which further optimizes the Mg2+-dependent signal; seeref 43) or at pH 7 but in the additional presence of 100 mM KCl(which slightly depresses it, unpublished results). Traces have beencorrected for the (small) artifacts induced by dilution.
Effects of TG, BHQ, or CPA on SERCA1a Biochemistry, Vol. 46, No. 51, 200715167
see the squares in part F versus part E of Figure 6). Both inthe presence and in the absence of Mg2+, the amount ofbound ATP remained significant in the presence of TG. Thus,the total absence of ATP-induced fluorescence changes underany of these two conditions implies that TG completelyprevented the Trp residues in the membrane from experienc-ing any distinct change in environment upon the residualbinding of ATP to the ATPase cytosolic domain: ATP stillbinds to the ATPase (with modest affinity) despite thepresence of TG, but (as previously observed after formation
of an E2BeFx complex) no longer induces any conforma-tional change in the transmembrane domain, where most ofthe Trp residues are located.
Similar experiments were repeated using BHQ or CPAinstead of TG. BHQ proved to act very similarly to TG,although again with poorer affinity. In the presence of Mg2+,30 M BHQ nearly completely prevented (SupportingInformation Figure S5A) the normal response of ATPase Trpresidues to MgATP addition previously illustrated in Figure6A, although about half of the [-32P]ATP remained boundin the presence of BHQ (Figure S5E); 10M BHQ was notsufficient in the presence of Mg2+ (data not shown). In theabsence of Mg2+, 10 M BHQ was sufficient to eliminatethe fluorescence response to ATP addition (compare FigureS5B with its control in Figure 6B), although [-32P]ATPbinding was only minimally altered by BHQ (Figure S5F).CPA revealed specific features, to be considered in parallelwith its previously noted specific behavior in the experimentsillustrated in Figures 3D and 5D. In the absence of free Mg2+,addition of CPA no longer induced any observable time-dependent fluorescence response, and the rapid fluorescencedrop observed now matched exactly the drop expected foran inner filter effect alone (Supporting Information FigureS5D); the binding of [-32P]ATP to ATPase was againminimally altered by CPA (Figure S5F), but in this case theaccompanying fluorescence changes, too, were minimallyaltered (compare Figure S5D with its control in Figure 6B):thus, CPA seems to hardly interact with the ATPase in theabsence of Mg2+. In the presence of Mg2+, the time-dependent fluorescence drop, which we believe to be amanifestation of CPA interaction with the ATPase, wascompleted on a time scale of minutes (Figure S5C). Afterthis equilibration, addition of ATP (i.e., MgATP) no longerinduced any fast rise in Trp fluorescence, as for TG or BHQ,despite the reduced but again nonzero amount of [-32P]ATPbound after short periods (Figure S5E; see also ref67). Atlater times, Trp fluorescence rose slowly (compare FigureS5C with its control in Figure 6A), possibly revealing slowdisplacement of CPA from its binding site(s) (as also foundin Figure 5D) concomitant with a slow increase in the amountof bound [-32P]ATP (Figure S5E).
Effects of TG or Other Inhibitors on Ca2+-Free ATPasein the Absence of Any Other Ligand, As Deduced from MildProteolysis Experiments with Proteinase K.Last, we studiedthe effects of inhibitors on Ca2+-free ATPase in the absenceof any other ligand, by using the ATPase susceptibility toproteinase K as an indicator of possible changes in itsconformation or dynamics. To be consistent with theexperiments in Figures 5 and 6, we chose here to keepproteolysis media devoid of DMSO and KCl; compared withthe previous proteolysis experiments illustrated in Figure 1,we had to use a lower proteinase K to SR membrane ratioin the new experiments, because both DMSO and KClappear to slow ATPase proteolysis (Figure S6 in theSupporting Information). As an introduction to these experi-ments, let us first recall (30, 48) that, during mild treatmentof SR vesicles with proteinase K in the presence of Ca2+, anumber of well-characterized initial products are formed,including membranous peptides derived from either theC-terminal (p83C, p27/28C, and p19C) or the N-terminal(p28N and a small amount of p81N) part of the ATPase, aswell as water-soluble p29/30 fragments derived from the
FIGURE 6: TG completely obliterates the response of ATPase Trpresidues to the residual binding of either MgATP or ATP. (A-D)The fluorometer cuvette contained 0.1 mg/mL SR in 150 mMMops-Tris at pH 7 and 20C. Sequential additions were made asfollows, when indicated by arrows: Ca2+ (100M), EGTA (1 mM),Mg2+ (5 mM; A, C) or EDTA (2 mM; B, D), and TG if any(1 g/mL; C, D); then, increasing concentrations of ATP wereadded, 3M first, then 30M more, then 300M more, and lasttwo or three additions of 300M again (arrowheads). The excitationwavelength was set at 298 nm to reduce the inner filter effect dueto ATP absorbance, while the emission wavelength was kept at320 nm; bandwidths were 2 and 10 nm, respectively. Traces havebeen corrected for the (very small) artifacts induced by dilution.(E, F) [-32P]ATP binding to the ATPase (in the presence of30M ATP) under almost exactly the same conditions, as deducedfrom filtration experiments. At time zero, 30M [-32P]ATP wasadded to Ca2+-free ATPase (0.1 mg/mL), suspended in 150 mMMops-Tris (pH 7 and 20C), 0.5 mM EGTA, and either 5 mMMg2+ (E) or 0.5 mM EDTA (F), and preincubated in the absence(circles) or presence (squares) of 1g/mL TG. After variousperiods, 2.5 mL aliquots were filtered on Millipore HA filters anddouble-counted without rinsing. In panel E, the tendency for boundATP to become slightly lower after long periods most probablyreveals Mg2+-dependent ATP hydrolysis by contaminant enzymes.
15168 Biochemistry, Vol. 46, No. 51, 2007 Montigny et al.
central and cytosolic nucleotide binding domain (see lanes2-5 in gel A of Figure 7, corresponding to an experimentperformed at pH 6.5). The C-terminal peptide p83C resultsfrom ATPase cleavage at T242-E243, and the T242-E243bond is indeed peripheral (and therefore accessible toproteolysis) in the structure of the Ca2+-bound form of Ca2+-ATPase (7). In the absence of Ca2+, as seen by lanes 6-9in gel A, a new high molecular mass peptide is also formed(compare lanes 6-9 with lanes 2-5; see also Figure 1 inref 49): this p95C peptide, a C-terminal one, is formedtogether with its N-terminal counterpart, p14N, after ATPasecleavage at L119-K120 (30).
Similar experiments were repeated in the absence of Ca2+
but now in the presence of TG. TG was previously describedto only moderately slow overall proteolysis of the initial
ATPase (35). We confirmed this, but in addition we foundthat the presence of TG has a clear qualitative effect on thepattern of proteolysis: p83C becomes much weaker, whilep95C is now predominant (lanes 2-5 in gel B versus lanes6-9 in gel A of Figure 7). In Danko et al.s previous paper(35), the same qualitative trend (although less pronouncedunder the conditions of that paper) is also visible (see Figure2 in ref 35), and a similar TG-induced inversion of the p83to p95 ratio has been reported in Western blots (50, 67).We also found that the presence of CPA, instead of TG,during proteolysis in the presence of EGTA, exerts effectson formation of p95C and p83C that are qualitatively similarto those exerted by TG, although slightly less pronounced(Figure 7B, lanes 6-9): the overall loss of intact ATPase isagain slowed, and the formation of p95C is again favored at
FIGURE 7: The presence of TG (or CPA) during proteinase K treatment of Ca2+-free ATPase slows cleavage at T242-E243, resulting inthe formation of large amounts of p95C. (A) SR membranes (2 mg of protein/mL) were treated with proteinase K (0.03 mg/mL) at 20Cfor various periods of time, in proteolysis buffer (100 mM Mops-NaOH and 5 mM Mg2+ at pH 6.5) containing either 0.3 mM Ca2+ (lanes2-5) or 0.5 mM EGTA (lanes 6-9). After proteolysis arrest and sample denaturation, 4g of protein was loaded per lane, on a 12%acrylamide gel. Lanes 1 and 10 contain molecular mass markers (LMW kit, GE Healthcare). This is a repeat of the experiment illustratedin Figure 1A of ref49, but performed exactly in parallel with the one illustrated in panel B. (B) SR membranes were similarly treated inthe presence EGTA, except for the additional presence of either TG (lanes 2-5) or CPA (lanes 6-9), both at 20g/mL (i.e., about 30Mfor TG and 60M for CPA; 2 mg/mL SR corresponds to about 10-14 M ATPase, assuming 5-7 nmol of ATPase/mg of protein).
Effects of TG, BHQ, or CPA on SERCA1a Biochemistry, Vol. 46, No. 51, 200715169
the expense of p83C. The same result, including the factthat the effect was weaker, was found when BHQ was used(Figure S7 in the Supporting Information). The moreprominent p95C band and the reduced p83C band, in thepresence of TG or the other inhibitors, must be due to aslowing of the cut at T242-E243 rather than to stimulationof the cut at L119-K120, as disappearance of intact ATPaseis slower in the presence of inhibitors than in their absence(i.e., in the absence of inhibitors, formation of the p95Cpeptide in the absence of Ca2+ is underestimated becausecleavage at T242-E243 also occurs). Thus, we conclude thatTG, BHQ, and CPA all reduce the rate of cleavage at T242-E243, with the effect of BHQ or CPA being slightly lesspronounced.
A conceivable interpretation for this inhibitor-inducedalteration in the proteolysis pattern might be that Ca2+-freeATPase explores various conformations, only a few of whichcorrespond to the true E2 conformation, with this trueconformation, defined as being the conformation reactive toPi, being the only one able to bind TG (25, 26, 40, 51). Ithas indeed already been suggested that Ca2+-free ATPase isin dynamic equilibrium between a true E2 state and a Ca2+-free E1 state (partly resembling the Ca2+-saturated ATPaseconformation) (39). However, the E2/E1 equilibrium isgenerally thought to depend on the pH (and temperature),being about 50/50 under neutral conditions at 20C, butbeing almost fully shifted toward the E2 form under moreacidic conditions (39), partly because the E2 species corre-sponds to a form in which protons (to be countertransportedfor Ca2+) are bound and occluded (52); it was recentlysuggested explicitly that, at pH 6, TG could stabilize theATPase in the protonated HnE2 state (14).
To test this interpretation, we repeated our experimentsunder different pH conditions. We predicted that if TG weremerely stabilizing a pre-existing protonated E2 form, theeffect of TG on susceptibility to proteolysis would beminimal at pH 6 (where the E2/E1 equilibrium is alreadyshifted toward the E2 state in the absence of TG) butprominent at pH 7 (where half of the Ca2+-free ATPase isin the Ca2+-free E1 state in the absence of TG). However,the result of the experiment did not fit with this prediction:the effect of TG on ATPase susceptibility to proteolysis inthe absence of Ca2+ was prominentboth at pH 6 (Figure8A) and at pH 7 (Figure 8B), and the p95C to p83C ratio inthe absence of TG was even lower at pH 6 than at pH 7, sothat the relative effect of TG was even stronger at pH 6 thanat pH 7. Whether this result fits better or not with thealternative and previously proposed possibility (53-55) that,irrespective of pH, TG (as well as other inhibitors such asBHQ or CPA) drives the ATPase toward a state thatsomehowdiffers from that of inhibitor-free ATPase, will bediscussed below.
DISCUSSION
At present, due to the known instability of detergent-solubilized Ca2+-free forms of ATPase, the diffractingcrystals we have for some of these forms of ATPase wereall obtained in the presence of inhibitors (mostly TG, butalso BHQ or CPA). These forms (with or without fluoride,for instance) do differ in their cytosolic regions, but for agiven type of inhibitor (TG for instance) they have rather
similar organizations in their transmembrane domain. Thelikely reason for the latter fact is now made clear by ourresults: TG does not prevent a number of events fromoccurring in the ATPase cytosolic domain, e.g., reaction withfluoride compounds (Figure 1B) or binding of ATP withreduced affinity (Figure 6E,F), but it apparently prevents thetransmembrane helices from responding to these cytosolicevents (Figures 3 and 6A-D). Thus, the present workemphasizes that, to better understand the various conforma-tions of sarcoplasmic reticulum ATPase during its catalyticcycle, especially its Ca2+-free forms, we do need crystalsprepared in theabsenceof inhibitor. Indeed, from thebeginning, TG was considered as a dead-end complex (25),and as was explicitly stated a few years later after directedmutagenesis and chimeric exchange experiments, perturba-tions produced by binding of either inhibitor within the stalksegment interfere with the long-range functional linkagebetween ATP utilization in the ATPase cytosolic region andCa2+ binding in the membrane-bound region (56). Here,we have demonstrated this directly with native sarcoplasmicreticulum ATPase. Although we are not able to exactlypinpoint the nature of the perturbations exerted by theinhibitors, we show that, beyond the transmembrane region,they extend to the loop(s) connecting this region and thecytosolic domains (Figures 7 and 8) and even higher abovethe membrane, up to fluorescein bound to Lys515 in thenucleotide binding domain (Figure 4); the effects of TG onthe headpiece of Ca2+-free ATPase were in fact predicted(67). All this should be kept in mind when the structures ofthe forms obtained in the presence of TG or other inhibitorsare considered. ThiscaVeatholds especially for the ATPasecrystalline forms obtained with fluoride compounds (TGE2MgF4, or TGE2AlF4), which at present are the onlyanalogues available for the true E2P form but still differ fromit or from its closest relative, the TG-free E2BeFx form(Figures 2 and 4), both in the transmembrane domain (Figure3 and ref28) and probably in the nucleotide domain too(Figure 4C). However, it possibly holds also for the ATPaseforms obtained in the presence of nucleotide and inhibitors,as TG apparently blocks long-distance reorganization bynucleotides of the ATPase transmembrane domain, eventhough nucleotide binding still takes place (Figure 6 andSupporting Information Figure S5).
Our results about the effect of Mg2+ binding to Ca2+-ATPasesat least as concerns the particular type of Mg2+
binding which results in Trp fluorescence changes observableunder neutral or alkaline conditionssmay also be discussed.The exact reason for this effect of Mg2+ has been disputedfrom the start. One of the possibilities is that this Mg2+ ioncould bind to one of the two Ca2+ sites, resulting in a shiftof the ATPase conformation toward a Ca2+-free but E1-like new conformation (43, 57). In this case, the blockadeof the effect of Mg2+ by TG would indeed be easilyunderstandable. However, Trp fluorescence experiments withthe E309Q mutant (mutated at Ca2+ binding site II) showedthat Mg2+ was able to bind to this mutant and to inducenormal fluorescence changes, although Ca2+ was no longerable to induce such changes (58); thus, in that work, it wassuggested that Mg2+ wasnotbinding at any of the Ca2+ sitesin the transmembrane region. If this is correct, the bindingsite for Mg2+ (at least for this Mg2+ ion) could then bebetween Asp703 and Asp707 in the catalytic site, as observed
15170 Biochemistry, Vol. 46, No. 51, 2007 Montigny et al.
in a recently obtained E2(TG)AMPPCP crystal (PDBstructure deposited as 2DQS, to be commented uponelsewhere), and the Trp response to Mg2+ would revealMg2+-induced long-distance reorganization (in the absenceof inhibitor) of the transmembrane region of the ATPase,and again this long-distance reorganization would turn outto be blocked by TG. However, at variance with thesuggestion in ref58, it is fair to recognize that site II ofE309Q might have retained the ability to bind Mg2+ eventhough it has lost the ability to bind Ca2+. Thus, a firmconclusion about the exact site to which Mg2+ has to bindto induce Trp fluorescence changes must probably be delayeduntil experiments with other mutants have been performed.
The results obtained with ATP and MgATP may also beshortly discussed here. Although, as well-known, TG affectsthe binding of MgATP very significantly, we confirmed that,
as recently described (47), TG affects the binding of ATPin the absence of Mg2+ to a very slight extent only; it waspreviously known, too, that TG hardly affects the bindingof ATP analogues such as TNP-ATP (46) or TNP-8N3-ATP (41). MgATP and ATP might in fact bind in slightlydifferent conformations (59), and a complex of Ca2+-freeATPase with TG might conceivably bind nucleotides(e.g., AMP-PCP) in a still somewhat different conformation.
Our results also add a few details to the characterizationof the effect of BHQ and CPA. As concerns BHQ, our resultsconfirm that this inhibitor binds with relatively poor affinity(in the range of a few micromolar in most of our experi-ments), but otherwise acts rather similarly to TG. In likelyrelation with this, the structure of the ATPase in BHQATPase crystals was not found to be very different fromthat in the presence of TG (17). As concerns CPA, we deduce
FIGURE 8: At both pH 6 and pH 7, the presence of TG in the absence of Ca2+ blocks ATPase cleavage by proteinase K at T242-E243almost fully, resulting in overwhelming formation of the p95C peptide. SR membranes were treated as described in Figure 7 in the presenceof 5 mM Mg2+ and 0.5 mM EGTA and the absence (lanes 2-5) and presence (lanes 6-9) of 0.02 mg/mL TG, but now in a mediumconsisting of either 50 mM Mes-Tris at pH 6 (A) or 50 mM Mops-Tris at pH 7 (B). After proteolysis arrest and sample denaturation,4 g of protein was loaded per lane, on a 12% acrylamide gel.
Effects of TG, BHQ, or CPA on SERCA1a Biochemistry, Vol. 46, No. 51, 200715171
from the Trp fluorescence results (Figures 3D, 5D, and S5C)that it reorganizes the transmembrane domain with kineticswhich is not very fast (on the minute time scale in thepresence of Mg2+), compared with the much faster kineticswith which TG and BHQ act (54, 60). However, we alsosuspect that the mere binding of CPA to ATPase mightdepend on Mg2+ availability (part C versus part D ofSupporting Information Figure S5). Similarly, partition ofthe well-known ionophore calcimycin (A23187) into sarco-plasmic reticulum membranes (see, e.g., ref61) is favoredby millimolar concentrations of Mg2+ (unpublished data).
As concerns the effects of inhibitors on the proteolyticsusceptibility of ligand-free ATPase (at least using proteinaseK as the protease), our similar results at different pH values(Figure 8) might at first sight be taken as rendering less likelythe possibility that inhibitors could stabilize the true E2conformation of the pump, the true E2 conformation beingdefined as the dynamic ensemble of conformations makingthe ATPase reactive to Pi, as opposed to the Ca2+-free E1conformation which, in Ca2+-free ATPase, is in a pH-dependent equilibrium with E2 (39): they might seem tosupport the previous alternative suggestion (53-55) thatthese inhibitors drive the ATPase toward states which, atleast in some respect, differ from the average E2 stateprevailing at pH 6 in Ca2+-free ATPase. In fact, in availablecrystals, the exact conformations of ATPase complexed withTG and CPA seem to be slightly different from each other(16, 17). However, old discussions in terms of conforma-tions of enzymes during their catalytic cycle might be partlyoutdated by the more modern views about the intrinsicdynamics of protein structures, and our proteolysis data mightbe worth a comment in this respect. As an alternative to (orin addition to) the idea of TG shifting the averageconformation of the protein into a slightly different one,we suggest that TG and the other inhibitors probably slowthe cut at T242-E243 mainly because their freezing effecton the movements of the transmembrane segments of theATPase propagates up to the links between the A-domainand the membrane domain: i.e., reduced susceptibility toproteolysis of the peripheral T242-E243 bond is probablynot mainly due to steric hindrance in a new conformation,but to slowing of the protein dynamics in the region of theselinks, a slowing which turns out to be more or less similarfor hydrophobic inhibitors bound at different sites near thetransmembrane region. A similar slowing of the chaindynamics probably explains the protection of the cleavagesites around V734 and V747 in the M5 helix which is alsoobserved upon TG binding (67). The ATPase might beinhibited as the result of such binding because proteinbreathing (i.e., transient increases in the protein trans-membrane cross-sectional area) is probably involved in thevarious transitions in the catalytic cycle (24, 62). This couldwell be what has been dubbed by Logan-Smith et al. aglobal conformational change (64), except that we nowmight have to change our traditional loose way of talkingabout conformational changes into a more realistic one,taking into account the intrinsic flexibility and dynamic stateof the protein.
The suggestion that the changes in proteolytic susceptibil-ity of the T242-E243 bond observed in the present workafter TG addition are due to changes in peptide chaindynamics, rather than to domain rotation and changes in bond
accessibility, makes even more remarkable the fact that thealternative explanation was correctly offered by Danko etal. (35) to explain the earlier and somewhat similar findingsthat the same T242-E243 bond was protected (but nowfully) upon formation of E2P analogues (as confirmed inFigure 1): in that case, this alternative explanation did provetrue, as rotation of the A-domain during the transition to E2P-related structures was subsequently deduced from crystalstructures (10, 12).
To conclude, our main finding in this work is that,especially in the case of the formation of Ca2+-free ATPase-fluoride complexes in the presence of beryllium, but alsoafter the mere binding of ATP or MgATP, the membranehelices no longer respond to what is happening in the headdomains after binding of TG or other inhibitors, as if thevarious regions of the protein were now uncoupled. Thisis obviously due to the fact that the loops connecting theATPase transmembrane and cytosolic domains are flexible.The role of these flexible links to accommodate thermalfluctuations and simultaneously permit coupling of Ca2+
transport to ATP hydrolysis has been discussed recently (17).Such flexibility might also be the basis for the long-disputedpossible slipping of the pump under special circumstances(see, e.g., refs63-65).
ACKNOWLEDGMENT
We are grateful to David McIntosh for suggestions duringthis work and for his help in improving an initial version ofthe manuscript and to Hiroshi Suzuki for critical reading.
SUPPORTING INFORMATION AVAILABLE
Figure S1 showing the protection of E2MgF4 and E2VO4 from proteolysis, in the absence or presence of TG;Figures S2-S4 showing FITC fluorescence changes uponaddition of VO4 (or AlF4) and the effects of TG, BHQ, orCPA; Figure S5 showing Trp fluorescence responses to ATP(and [-32P]ATP binding) in the presence of BHQ or CPA;Figure S6 showing the effect of DMSO and KCl on ATPaseproteolysis; and Figure S7 showing that BHQ, too, favorsp95 formation after proteolysis. This material is availablefree of charge via the Internet at http://pubs.acs.org.
REFERENCES
1. Makinose, M. (1972) Phosphoprotein formation during osmo-chemical energy conversion in the membrane of the sarcoplasmicreticulum,FEBS Lett. 25, 113-115.
2. Hasselbach, W. (1978) The reversibility of the sarcoplasmiccalcium pump,Biochim. Biophys. Acta 515, 23-53.
3. de Meis, L., and Vianna, A. L. (1979) Energy interconversion bythe Ca2+-dependent ATPase of the sarcoplasmic reticulum,Annu.ReV. Biochem. 48, 275-292.
4. Toyoshima, C., and Inesi, G. (2004) Structural basis of ionpumping by Ca2+-ATPase of the sarcoplasmic reticulum,Annu.ReV. Biochem. 73, 269-292.
5. Kuhlbrandt, W. (2004) Biology, structure and mechanism of P-typeATPases,Nat. ReV. Mol. Cell. Biol. 5, 282-295.
6. Mller, J. V., Nissen, P., Srensen, T. L., and le Maire, M. (2005)Transport mechanism of the sarcoplasmic reticulum Ca2+-ATPasepump,Curr. Opin. Struct. Biol. 15, 387-393.
7. Toyoshima, C., Nakasako, M., Nomura, H., and Ogawa, H. (2000)Crystal structure of the calcium pump of sarcoplasmic reticulumat 2.6 resolution,Nature 405, 633-634.
8. Toyoshima, C., and Nomura, H. (2002) Structural changes in thecalcium pump accompanying the dissociation of calcium,Nature418, 605-611.
15172 Biochemistry, Vol. 46, No. 51, 2007 Montigny et al.
9. Toyoshima, C., Nomura, H., and Sugita, Y. (2003) Structural basisof ion pumping by Ca2+-ATPase of sarcoplasmic reticulum,FEBSLett. 555, 106-110.
10. Srensen, T. L.-M., Mller, J. V., and Nissen, P. (2004) Phosphoryltransfer and calcium ion occlusion in the calcium pump,Science304, 1672-1675.
11. Toyoshima, C., and Mizutani, T. (2004) Crystal structure of thecalcium pump with a bound ATP analogue,Nature 430, 529-535.
12. Toyoshima, C., Nomura, H., and Tsuda, T. (2004) Lumenal gatingmechanism revealed in calcium pump crystal structures withphosphate analogues,Nature 442, 361-368.
13. Olesen, C., Srensen, T. L.-M., Nielsen, R. C., Mller, J. V., andNissen, P. (2004) Dephosphorylation of the calcium pump coupledto counterion occlusion,Science 306, 2251-2255.
14. Jensen, A.-M. L., Srensen, T. L., Olesen, C., Mller, J. V., andNissen, P. (2006) Modulatory and catalytic modes of ATP bindingby the calcium pump,EMBO J. 25, 2305-2314.
15. Obara, K, Miyashita, N., Xu, C., Toyoshima, I., Sugita, Y., Inesi,G., and Toyoshima, C. (2005) Structural role of countertransportrevealed in Ca2+ pump crystal structure in the absence of Ca2+,Proc. Natl. Acad. Sci. U.S.A. 102, 14489-14496.
16. Moncoq, K., Trieber, C. A., and Young, H. S. (2007) Themolecular basis for cyclopiazonic acid inhibition of the sarco-plasmic reticulum calcium pump,J. Biol. Chem. 282, 9748-9757.
17. Takahashi, M., Kondou, Y., and Toyoshima, C. (2007) Interdomaincommunication in calcium pump as revealed in the crystalstructures with transmembrane inhibitors,Proc. Natl. Acad. Sci.U.S.A. 104, 5800-5805.
18. Toyoshima, C. (2007) Ion pumping by calcium ATPase ofsarcoplasmic reticulum,AdV. Exp. Med. Biol. 592, 295-303.
19. Picard, M., Toyoshima, C., and Champeil, P. (2005) The averageconformation at micromolar [Ca2+] of Ca2+-ATPase with boundnucleotide differs from that adopted with the transition state analogADPAlFx or with AMPPCP under crystallization conditions atmillimolar [Ca2+], J. Biol. Chem. 280, 18745-18754.
20. Picard, M., Jensen, A.-M., Srensen, T. L., Champeil, P., Mller,J. V., and Nissen, P. (2007) Ca2+ versus Mg2+ coordination atthe nucleotide-binding site of the sarcoplasmic reticulum Ca2+-ATPase,J. Mol. Biol. 368, 1-7.
21. Andersen, J. P., Mller, J. V., and Jrgensen, P. L. (1982) Thefunctional unit of sarcoplasmic reticulum Ca2+-ATPase. Activesite titration and fluorescence measurements,J. Biol. Chem. 257,8300-8307.
22. McIntosh, D. B., and Ross, D. C. (1985) Role of phospholipidand protein-protein associations in activation and stabilization ofsoluble Ca2+-ATPase of sarcoplasmic reticulum,Biochemistry 24,1244-1251.
23. Lund, S., Orlowski, S., de Foresta, B., Champeil, P., le Maire,M., and Mller, J. V. (1989) Detergent structure and associatedlipid as determinants in the stabilization of solubilized Ca2+-ATPase from sarcoplasmic reticulum,J. Biol. Chem. 264, 4907-4915.
24. Picard, M., Dahmane, T., Garrigos, M., Gauron, C., Giusti, F., leMaire, M., Popot, J.-L., and Champeil, P (2006) Protective andinhibitory effects of various types of amphipols on the Ca2+-ATPase from sarcoplasmic reticulum: a comparative study,Biochemistry 45, 1861-1869.
25. Sagara, Y., Wade, J. B., and Inesi, G. (1992) A conformationalmechanism for formation of a dead-end complex by the sarco-plasmic reticulum ATPase with thapsigargin,J. Biol. Chem. 267,1286-1292.
26. Wictome, M., Michelangeli, F., Lee, A. G., and East, J. M. (1992)The inhibitors thapsigargin and 2,5-di(tert-butyl)-1,4-benzohyd-roquinone favour the E2 form of the Ca2+, Mg2+-ATPase,FEBSLett. 304, 109-113.
27. Seekoe, T., Peall, S., and McIntosh, D. B. (2001) Thapsigarginand dimethyl sulfoxide activate Medium PiT HOH oxygenexchange catalyzed by sarcoplasmic reticulum Ca2+-ATPase,J.Biol. Chem. 276, 46737-46744.
28. Picard, M., Toyoshima, C., and Champeil, P. (2006) Effects ofinhibitors on luminal opening of Ca2+ binding sites in an E2P-like complex of sarcoplasmic reticulum Ca2+-ATPase with Be2+-fluoride, J. Biol. Chem. 281, 3360-3369.
29. Champeil, P., Guillain, F., Venien, C., and Gingold, M. P. (1985)Interaction of magnesium and inorganic phosphate with calcium-deprived sarcoplasmic reticulum adenosinetriphosphatase as re-flected by organic solvent induced perturbation,Biochemistry 24,69-81.
30. Juul, B., Turc, H., Durand, M. L., Gomez de Gracia, A., Denoroy,L., Mller, J. V., Champeil, P., and le Maire, M. (1995) Dotransmembrane segments in proteolyzed sarcoplasmic reticulumCa2+-ATPase retain their functional Ca2+ binding properties afterremoval of cytoplasmic fragments by proteinase K?,J. Biol. Chem.270, 20123-20134.
31. Mller, J. V., Lenoir, G., Marchand, C., Montigny, C., le Maire,M., Toyoshima, C., Juul, B. S., and Champeil, P. (2002) Calciumtransport by sarcoplasmic reticulum Ca2+-ATPase. Role of theA-domain and its C-terminal link with the transmembrane region,J. Biol. Chem. 277, 38647-38659.
32. Soulie, S., Mller, J. V., Falson, P., and le Maire, M. (1996) Ureareduces the aggregation of membrane proteins on Sodium DodecylSulfate-Polyacrylamide Gel Electrophoresis,Anal. Biochem 236,363-364.
33. Laemmli, U. K. (1970) Cleavage of structural proteins during theassembly of the head of bacteriophage T4,Nature 227, 680-685.
34. Champeil, P., Riollet, S., Orlowski, S., Guillain, F., Seebregts, C.J., and McIntosh, D. B. (1988) ATP regulation of sarcoplasmicreticulum Ca2+-ATPase. Metal-free ATP and 8-bromo-ATP bindwith high affinity to the catalytic site of phosphorylated ATPaseand accelerate dephosphorylation,J. Biol. Chem 263, 12288-12294.
35. Danko, S., Daiho, T., Yamasaki, K., Kamidochi, M., Suzuki, H.,and Toyoshima, C. (2001) ADP-insensitive phosphoenzymeintermediate of sarcoplasmic reticulum Ca2+-ATPase has acompact conformation resistant to proteinase K, V8 protease andtrypsin,FEBS Lett. 489, 277-282.
36. Danko, S., Yamasaki, K., Daiho, T., and Suzuki, H. (2004) Distinctnatures of Beryllium Fluoride-bound, Aluminum Fluoride-bound,and Magnesium Fluoride-bound stable analogues of an ADP-insensitive Phosphoenzyme intermediate of sarcoplasmic reticulumCa2+-ATPase: changes in catalytic and transport sites duringphosphoenzyme hydrolysis,J. Biol. Chem. 279, 14991-14998.
37. de Meis, L., Martins, O. B., and Alves, E. W. (1980) Role ofwater, Hydrogen ion, and temperature on the synthesis of ATPby the sarcoplasmic reticulu