Rôle des ARN hélicases Ddx5 et Ddx17 dans la progression tumorale Etienne Dardenne

To cite this version:

Etienne Dardenne. Rôle des ARN hélicases Ddx5 et Ddx17 dans la progression tumorale. Biologie moléculaire. Université Claude Bernard - Lyon I, 2014. Français. ￿NNT : 2014LYO10048￿. ￿tel- 01198928￿

HAL Id: tel-01198928 https://tel.archives-ouvertes.fr/tel-01198928 Submitted on 14 Sep 2015

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés.     ?   

  v@%@&`JN?HFGJ 77(&`|z{~   d   (5.I@(&>@ d  {        m@@)D(%E9GD|zz€n  9ED&7E&>E!5.?E&6&7D5&}{6@A|z{~ @ 0$)).))   ` ;5&%&A-(5."A&A%K&D%K{%7A5>@9,@&AA.97DE69@5&

 `

@ E5.7&7!5&A  >>9@D&E@   @E@&7D9@"9A  >>9@D&E@   @&.7.E"9  K6.7D@."&   @ &@6.7 .55&D  K6.7D&E@   @.%.&@E!9&E+  .@&"D&E@%&D-'A&        ?  G   . /$)0'=)$2./$0  <.)*$/? * '  ."&j>@(A.%&7D%E97A&.5%d%6.7.AD@D.97 a5&@9+&AA&E@ 6%   ."&j>@(A.%&7D%E97A&.5%&ADE%&A&D%&5.&[email protected].@& a5&@9+&AA&E@-.5.>>& ."&j>@(A.%&7D%E97A&.5".&7D.+.?E& a5&@9+&AA&E@ &@6.7  .@&"D&E@ (7(@5%&A&@I."&A a5.7          "E5D(%&(%&".7&L97ADk5E%&&@7@% .@&"D&E@`a5&@9+&AA&E@ a  "E5D( %& (%&".7& &D %& 1&ED.?E& L97 E% k -@5&A .@&"D&E@`6&5@9+&AA&E@&a  (@.&EK .@&"D&E@`a5&@9+&AA&E@a   "E5D(%d%97D959,.& .@&"D&E@`6&5@9+&AA&E@&a   7AD.DED%&A".&7"&A-@6"&ED.?E&A&D.959,.?E&A .@&"D&E@`a5&@9+&AA&E@a  7AD.DED%&A".&7"&A&D&"-7.?E&A%&5(%>DD.97 .@&"D&E@`a5&@9+&AA&E@a  (>@D&6&7D%&+9@6D.97&D&7D@&%&&"-&@"-&&7.959,.& E6.7&           

"E5D(%&A".&7"&A&D&"-7959,.&A .@&"D&E@`a5&@9+&AA&E@a  (>@D&6&7D.959,.& .@&"D&E@`a5&@9+&AA&E@a (>@D&6&7D-.6.&.9"-.6.& .@&"D&E@`6&5&@9+&AA&E@ a (>@D&6&7D  .@&"D&E@`aa  (>@D&6&7D 7+9@6D.?E& .@&"D&E@`a5&@9+&AA&E@a   (>@D&6&7DD-(6D.?E&A .@&"D&E@`a5&@9+&AA&E@a  (>@D&6&7D("7.?E& .@&"D&E@`a5&@9+&AA&E@ a   (>@D&6&7D-LA.?E& .@&"D&E@`6&a  (>@D&6&7D".&7"&A%&5&@@& .@&"D&E@`6&5@9+&AA&E@& a  ".&7"&A&D&"-7.?E&A%&A"D.I.D(A-LA.?E&A&D>[email protected]&A .@&"D&E@`aa  !A&@ID9.@&%&A".&7"&A%&5d7.I&@A%&L97 .@&"D&E@`aa    95LD&"-L97 .@&"D&E@`aa  "95&E>(@.&E@&%&-.6.&-LA.?E&5&"D@97.?E& .@&"D&E@`a a  7AD.DED[email protected].@&%&&"-7959,.&%&L97{ .@&"D&E@`aa  7AD.DED[email protected].@&%&9@6D.97%&A0D@&A .@&"D&E@`aa   7AD.DED%&".&7"&.77".'@&&D%rAAE@7"&A %6.7.AD@D&E@>@9I.A9.@&`aa   

 |  %  '&  & D.&7A D9ED %d!9@%  @&6&@".&@ 5&A 6&6!@&A ?E. "96>9A&7D 697 3E@L %dI9.@ ""&>D(%d(I5E&@697D@I.5%&D-'A&`5&@a E5.7&7!5&A&D5&@E@&7D9@"9A?E.^ &7 5&E@ ?E5.D( %& @>>9@D&E@A^ 97D %977( %& 5&E@ D&6>A >9E@ 5.@& &D >9@D&@ E7 @&,@% "@.D.?E&AE@69767EA"@.Da &@&6&@".&(,5&6&7D5&@&.7.E"9>9E@5d&K6&7%&697 D@I.5.7A.?E&5&@a &@6.7 .55&D?E.""&>D(%&>@(A.%&@"&3E@L%&D-'A&a  ,@D.DE%&I(,5&6&7DED-.694-&DD(>-7&,7&@^?E.D9EDE597,%& "&A77(&A^D@I&@A5&"96.D(%&AE.I.%&D-'A&^97DAE.I.697D@I.5a  &A @&6&@".&6&7DA 5&A >5EA "-5&E@&EK Ad%@&AA&7D (,5&6&7D  697 %.@&"D&E@ %& D-'A&^.%.&@E!9&E+^?E.%.@.,("&DD&D-'A&%7A5"97D.7E.D(%&6&AD@IEK%&6AD&@a %.A>97.!.5.D(^5"97+.7"&&DA&A&7"9E@,&6&7DA%E@7D"&AA.K77(&Am6AD&@&DD-'A&n 97D (D( &AA&7D.&5A %7A 5d!9ED.AA&6&7D %& 6&A @&"-&@"-&A &D 5d"?E.A.D.97 %dE7& "&@D.7& 6[email protected](a   @&"977.AA7"& Ad%@&AA& EAA.  [email protected] >9E@ 5d.%& ?Ed.5 6d D9E39E@A ""9@%(& %7A 6&A @&"-&@"-&A^ >9E@ A&A @&6@?E&A &D A&A "97A&.5A &D A @&5&"DE@& DD&7D.I& %E 67EA"@.Da  &(D(E75.&ED@'A&7@."-.AA7D>9E@@(5.A&@"&DD&D-'A&&D3&>&7A&D9ED&A 5&A .7D&@"D.97A &D 5&A %.A"EAA.97A ?E& 3d. &E&A I&" D9EA 5&A 6&6!@&A %&A (?E.>&A %E %(>@D&6&7Da7&>&7A(&++&"DE&EA&D9EA5&A6&6!@&A%&5d(?E.>&^"&EK?E.79EA97D @&39.7DA@("&66&7Dm EAA&.7^.&@@&^6.5.&^(9^D.6^5@^^67%.7&^@.&j.&@@&^ D-.&E^ &5&7^ 6.55&^ 797n &D 5&A g7".&7Ah .697^ 5(979@^ 9>-.&^ @D.7^ .A&^ ."&5^ &7j-.5.>>&^I&"?E.3d.>@D,(5d(>9?E&>@.A.&77&>E.A5dI&7DE@&5L977.A&a &DD&D-'A&%9.D!&E"9E>EK("-7,&A^+@E"DE&EK&D>@+9.AI.+A^?E&79EAI97A&EAE "9E@A%&D9ED&A"&A77(&Aa d%@&AA&%&A@&6&@".&6&7DAA>(".EK.697&D."&5?E. 97D(D(6&A"96>,797A%&D-'A&&D?E.97DD9E39E@A&EE7&9@&.55&DD&7D.I&\   &A@&6&@".&6&7DA7dE@.&7D>A%&A&7AA.3&7&6&7D.977.A"&EK?E.97D "97D@.!E(+.@&%&"&A?ED@&77(&AE7696&7D.79E!5.!5&a &>&7A&>@D."E5.'@&6&7D -.!ED^ !^-59(^E@&7Da^D-.&E^77&jE@&^E@&7D a^@.97^E5^5(6&7D^&D D9EA5&AED@&Aaaa  &A @&6&@".&6&7DA Ad%@&AA&7D &7+.7  6 +6.55& &D  E5.& ?E. 97D D9E39E@A 67.+&AD(5&E@&7D-9EA.A6&>9E@697D@I.5&D6d97D>>9@D(E7A9ED.&7.7%(+&"D.!5&a  &D@I.57dE@.D>E)D@&6&7(A7A5d.%&%&5.,E&D.975&"97D@&5&7"&@?E. 6d>>9@D(5d.%&6D(@.&55&>9E@6d&7,,&@%7A5D-'A&a      } &(   >@9,@&AA.97DE69@5&^?E."97%E.D5+9@6D.97%&6(DADA&A^&AD5&@(AE5DD%& >@9+97%&A 69%.+."D.97A %&A %.++(@&7DA 7.I&EK %& @(,E5D.97 %& 5d&K>@&AA.97 %&A ,'7&A "966&5D@7A"@.>D.979E5d(>.AA,&5D&@7D.+a  E"9E@A%&6D-'A&^3r.(DE%.(5&@;5&%&&D{^%&EK-(5."A&A?E.^ E"9E@A%&5>@9,@&AA.97DE69@5&^A97D.6>5.?E(&A%7A5@(,E5D.97D@7A"@.>D.977&55& 5c(>.AA,&5D&@7D.+^&D5!.9,(7'A&%&A6."@9Aa9E@"&5^3r.ED.5.A(%&EK69%'5&A%& >@9,@&AA.97 DE69@5& ` 5& 69%'5& [email protected] ~{^ "96>9A( %& "&55E5&A "7"(@&EA&A ?E. >@(A&7D&7D%&A>@9>@.(D(A6(DADD.?E&A%.++(@&7D&A^&D5&A"&55E5&A-E6.7&A{z?E.^ >@'A D@.D&6&7D E  jX^ A97D ">!5&A %& @(5.A&@ 5 D@7A.D.97 (>.D-(5.9j 6(A&7"-L6D&EA&^ E7 >@9"&AAEA %& D@7Aj%.++(@&7".D.97 ?E. "97D@.!E&  5 +9@6D.97 %& 6(DADA&Aa  7A 5& 69%'5& ~{^ 3r. 697D@( ?E& %K{ &D %K "97D@.!E&7D  5d.7IA.I.D( %&A "&55E5&ADE69@5&A&7"97D@;57D%&A>@9,@66&AD@7A"@.>D.977&5A&D%d(>.AA,&5D&@7D.+a 5EA >@(".A(6&7D^ 3r. %(697D@( ?E& %K &D %K{ +[email protected]&7D 5d,@&AA.I.D( %&A "&55E5&A "7"(@&EA&A &7 @(,E57D 5d(>.AA,& %&A I@.7DA %&  5d-.AD97& 6"@9 |{ ?E.^  5&E@ D9E@^ "97D@;5&7D5d&K>@&AA.97%&,'7&A.6>5.?E(A%7A5>@9,@&AA.97DE69@5&a  7A 5& 69%'5& {z 9F 5 D@7A.D.97 (>.D-(5.9j6(A&7"-L6D&EA& >&ED )D@& .7%E.D& A9EA  jX^ 3r. 697D@( ?E&  &D { 9@"-&AD@&7D %L76.?E&6&7D %&A >@9,@66&A D@7A"@.>D.977&5A &D %r(>.AA,&a  &D { "97D@;5&7D E7 >@9,@66& D@7A"@.>D.977&5 A>(".+.?E& %& 5 %.++(@&7".D.97  &7 ,.AA7D "966& "9@(,E5D&E@A D@7A"@.>D.977&5A%&A+"D&E@A%&D@7A"@.>D.97.6>5.?E(A%7A5I9.&%&A.,75.AD.97  jXa&A>@9D(.7&A"97D@;5&7D(,5&6&7D5d&K>@&AA.97%&6."@9A%&5%.++(@&7".D.97a &A6."@9A"97D@;5&7D5dd&K>@&AA.97%&&D{%7AE7&!9E"5&%&@(D@9"97D@;5& 7(,D.I&a%.6.7ED.97%&"&A>@9D(.7&AE"9E@A%&5%.++(@&7".D.97>&@6&D56.A&&7 >5"& %dE7 >@9,@66& %d(>.AA,& ?E. >&ED "97D@.!E&@ EK "-7,&6&7DA 69@>-959,.?E&A 9!A&@I(AE"9E@A%&5D@7A.D.97(>.D-(5.9j6(A&7"-L6D&EA&a  &D@I.5&++&"DE(>&7%7D6D-'A&6&D&7(I.%&7"&5r.6>9@D7"&%&A-(5."A&A &D{"966&@(,E5D&E@A"5(A%&5>@9,@&AA.97DE69@5&^&DA9E5.,7&5&@;5&%& 5r(>.AA,& 5D&@7D.+ 59@A %& 5 >@9,@&AA.97 DE69@5&a & >5EA^ "& D@I.5 6&D 5r""&7D AE@ 5r.6>9@D7"& %r.7D(,@&@ 5&A %.++(@&7DA 7.I&EK %& @(,E5D.97 %& 5d&K>@&AA.97 %&A ,'7&A mD@7A"@.>D.97^ (>.AA,&^ 6."@9n >9E@ E7& "96>@(-&7A.97 ,59!5& %& 5 >@9,@&AA.97 DE69@5&a 

 ~ &'%'  E69@ >@9,@&AA.97 5&%.7, D9 D-& +9@6D.97 9+ 6&DADA&A @&AE5DA +@96 %&&> 69%.+."D.97A 9+ ,&7& &K>@&AA.97 >@9,@6A D A&I&@5 5&I&5A^ .7"5E%.7, D@7A"@.>D.97 7% A>5.".7,a[email protected],6L-^ .7I&AD.,D&%D-&@95&.7DE69@>@9,@&AA.979+7%{^ DJ9-.,-5L@&5D&%6E5D.+97"D.975!9K-&5."A&AD-D@&.7I95I&%.7D@7A"@.>D.97 7%A>5.".7,AJ&55A.76."@9!.9,&7&A.Aa9@D-.A>E@>9A&^ EA&%DJ9!@&AD"7"&@ 69%&5A9+DE69@>@9,@&AA.97`D-&~{69EA&69%&5"96>9A&%9+"7"&@"&55AD-D&K-.!.D %.++&@&7D 6&DADD." >@9>&@D.&A 7% {z -E67 "&55A D-D E7%&@,9 &>.D-&5.5jD9j 6&A&7"-L65D@7A.D.97E>97,+jXD@&D6&7D^D@7Aj%.++&@&7D.D.97>@9"&AA"97D@.!ED&AD9 6&DADA.A+9@6D.97a  7 D-& ~{ 69EA& 69%&5^  A-9J&% D-D %K{ 7% %K "97D@.!ED& D9 DE69@j"&55 .7IA.I&7&AA !L "[email protected], !9D- D@7A"@.>D.975 7% A>5.".7, >@9,@6Aa 9@& A>&".+."55L^  %&697AD@D&%D-D%K7%%K{>@969D&"7"&@"&55A,,@&AA.I&7&AA!L@&,E5D.7,D-& A>5.".7, 9+ D-& 6"@9 |{ -.AD97& J-."- .7 DE@7 .6>"DA 97 D-& &K>@&AA.97 9+ ,&7&A .6>5."D&%.7DE69@"&55.7IA.I&7&AAa  7D-&,+jX.7%E"&%&>.D-&5.5jD9j6&A&7"-L65D@7Aj%.++&@&7D.D.9769%&5^ A-9J&% D-D  7% { %L76."55L 9@"-&AD@D& D@7A"@.>D.97^ 6."@9 7% A>5.".7, >@9,@6Aa  7% { 6&%.D&  %.++&@&7D.D.97jA>&".+." D@7A"@.>D.97 >@9,@6 !L "D.7,AD@7A"@.>D.975"9@&,E5D9@A9+D-&D@7A"@.>D.97+"D9@A.7I95I&%.7D-&,+jX A.,75.7,>D-JLa 7D&@&AD.7,5L^7%{5A9%.@&"D5L"97D@95D-&&K>@&AA.979+4&L %.++&@&7D.D.976."@9Aa&6@4!5L^D-&A&6.A%9J7@&,E5D&7%{.7 7&,D.I& +&&%!"4 599> %[email protected], D-& &7AE.7, %.++&@&7D.D.97 >@9"&AAa .755L^  A-9J&% D-D 7%{%9J7@&,E5D.97%.@&"D5L&5.".DA%.++&@&7D.D.97jA>&".+."A>5.".7,>@9,@6 D-D 6L "97D@.!ED& D9 D-& 69%.+."D.97 9+ 69@>-959,."5 >-&79DL>& %[email protected], &>.D-&5.5jD9j 6&A&7"-L65 D@7A.D.97a -&@&+9@&^  "-@"[email protected]&% D-&A&  -&5."A&A A %L76." 9@"-&AD@D9@A9+D@7A"@.>D.97^6.7%A>5.".7,>@9,@6Aa  -&J9@4>&@+9@6&%%[email protected],6L--.,-5.,-DAD-&.6>9@D7"&9+7%{ -&5."A&AA4&L@&,E5D9@A9+DE69@>@9,@&AA.97.7!@&AD"7"&@^7%5A9E7%[email protected]&A D-&@95&9+5D&@7D.I&A>5.".7,%[email protected],DE69@>@9,@&AA.97aE@D-&@69@&^D-.AJ9@4&6>-A.M&A D-& .6>9@D7"& 9+ .7D&,@D.7, D-& %.++&@&7D 5L&@A 9+ D-& ,&7& &K>@&AA.97 >@9"&AA mD@7A"@.>D.97^A>5.".7,^6."@9n+9@"96>@&-&7A.I&E7%&@AD7%.7,9+DE69@>@9,@&AA.97a

  $/0/. 2$0$*)/

 }d`}d7D@7A5D&%&,.97 95 `95L6(@A&  6`6&AA,&@ `%(79A.7&@.>-9A>-D& - `-@96D.7 66E79>@(".>.DD.97  `@9AAj.74.7, 66E79@&".>.DD.97 `E!&""9dA69%.+."D.979+,5&dA6&%.E6 `K97.">5.".7,7-7"&@ `K97.">5.".7,.5&7"&@   `5E9@&A"&7D 7.DE L!@.%.MD.97 ` @&&75E9@&A"&7D@9D&.7 ` .AD97&&"&DL5A& -7`-&D&@9,&7&9EA7E"5&@.!97E"5&9>@9D&.7 ` 7D@97.">5.".7,7-7"&@ ` 7D@97.">5.".7,.5&7"&@ ` 79"4jED 6.`6."@9 `97jA&7A&6&"L 7`7&E@97595L>[email protected].%.7&@"D.7%.7,@9D&.7 `E"5&9D.%&@.-9A>-D& `>&7&%.7,@6&  `>5D&5&Dj%[email protected]&%,@9JD-+"D9@  `@9D&.7 .7A&  `@9D&.7 .7A& `95L>[email protected].%.7&@"D.7%.7,@9D&.7  `.7%E"&%A.5&7".7,"96>5&K  ` 66E79>@&".>.DD.97 `&"9,7.D.979D.+ j`&D@9jD@7A"@.>DA&95L6&@A&-.7&"D.97 `LAD&6D."I95ED.979+.,7%A!L>97&7D.5&7@."-6&7D A7`A6557E"5&@ A7`A6557E"5&@.!97E"5&9>@9D&.7 `AD&@9.%@&"&>D9@"D.ID9@ ~z`.6.7.@EA~z D`%&@7A+&@D |`|EK.55.@L"D9@ `5D@I.95&DA     

 € $/0//!$"1./0/0'/

 '%41!J? !213#1'23'04!2 A4,3'22434+-1*@A/1"2 ,&,H!',!1%KIJJ@DDDDDDDDDDDDDDDDDD JL '%41!K? ++-13*'23'-, !$'1-*23!2&4+',2/13'1 A4,/!3'3,-+1! A*3#13'-,2%#,#3'04!2@ DDDD JN '%41!L?&#,-38/!2 !234+!412+++'1!2OP>JOQ>MIP!3MJ@DDDDDDDDDDDDDDDDDDDDDDDDD JQ '%41!M?4'5'&'+'*4+',!2!,3 !*$-1+3'-, !34+!412!3 !+#3232!2/1"2',(!3'-, !!**4*!2MJ &!9*2-41'2@DDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDD JR '%41!N? 2 !+#3233'04!@ /3#! A/1"2&!!*H!',!1%KIJJDDDDDDDDDDDDDDDDD KJ '%41!O?-'! !2'%,*'23'-, !2/1-3#',!2+ 2!,1#/-,2!4=@DDDDDDDDDDDDDDDDDDDDDDDDDDDDD KN *!J? '23!2 !2/1','/47%",!2'*!2 !2$3!412 !31,21'/3'-, !*@DDDDDDDDDDDDDDDDDDDDDKP *!K?- 4*3'-, !*A3'5'3# !2$3!412 !31,21'/3'-, !*5'*!41*''2-, !2/1-3#',!2 !* &1-+3',!> !2 -C1#%4*3!412 31,21'/3'-,,!*2> !2 $3!412 ! 31,21'/3'-, -4  !2 +- '$'3'-,2 /-23C 31 43'-,,!**!2@DDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDD LI '%41!P?+'1-2'+/*'04#2 ,2*@ DDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDD LO '%41!Q?'3! A#/'22%!-,2!,242!31#3'-, A#/'22%!@DDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDD LR '%41!R?#04!,!'2C1#%4*31'! !*A#/'22%!!3$3!412 A#/'22%!@DDDDDDDDDDDDDDDDDDDDDDDDDDDDDD MJ '%41!JI? ,$*4!,! !*31,21'/3'-,241*A#/'22%!*3!1,3'$@DDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDD ML '%41!JJ?7!+/*! A',$*4!,! !+- '$'3'-,2 !*&1-+3',!241*A#/'22%!*3!1,3'$@DDDDDDDDDDDDDD MM '%41!JK?7!+/*! !2314341!2!-, '1! !*A',$*4!, ,3*!2 #'2'-,2 A#/'22%!@DDDDDDDDDDDDDDD MN '%41! JL? 7!+/*!2 ! '$$#1!,3!2 '2-$-1+!2 /1- 4'3!2 /1 #/'22%! *3!1,3'$ 1!*!5,3!2 4 -412 ! * /1-%1!22'-,34+-1*!@DDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDD MR '%41! JM? EF -,3'-, !3 *-*'23'-, !2 '*!2 ! @ EF- ! A#/'22%!1!*3'$47/1-3#',!2!3 $-7K@ DDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDD NK '%41!JN?+-3'$2 !2&#*'2!2@DDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDD NN '%41!JO?!/1#2!,33'-, !2/1-!2242!**4*'1!2'+/*'04,3*!2&#*'2!2+-3'$2@DDDDDDDDDDNO '%41!JP?314341!!3&-+-*-%'! !2&#*'2!2 7N!3 7JP@DDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDD NP *! M? 3!412 ! 31,21'/3'-, !3 /1-3#',!2 1#%4*31'!2 ! * 31,21'/3'-, ',3!1%'22,3 5! N !3 JP@ DDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDD NR '%41!JQ?&#+ !2 '$$#1!,321.*!2 !N!3JP ,2*+3413'-, ! '$$#1!,322@DDDDDDDDD OL '%41!JR?- '$'3'-,/-23C31 43'-,,!**! !N!3JP?'+/*'3'-,241*!412$-,3'-,2@ DDDDDDDDD OM '%41!KI?-,3'-, !N ,2*@DDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDD OQ *!4//*#+!,3'1!J? '23! !2/1','/47$3!412 A#/'22%! !2$+'**!2-4&,',2'04!*!2$3!412 A#/'22%!!7/1'+#2 !+,'"1!3'22422/#'$'04!2@DDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDD JMM *!4//*#+!,3'1!K? '23! !2/1','/4731,21'321#%4*#2/1#/'22%!*3!1,3'$4-412 !*1,2'3'-, /'3&#*'-C#2!,&8+3!42!@DDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDDD JMN  

  '/(0$./



$/0/. 2$0$*)/<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<L

$/0//!$"1./0/0'/<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<M

'/(0$./<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<N

  <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<GF .0$ @ ,.*".//$*)01(*.'<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<GG j(+.7.D.97&D6("7.A6&Aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa{{ {j@"D(@.AD.?E&A%&A"&55E5&A"7"(@&EA&Aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa{| |j+"D&E@A+[email protected]7D5>@9,@&AA.97DE69@5&` 7AD!.5.D(,(7(D.?E&^6."@9&7I.@97&6&7D&D>5AD.".D( "&55E5.@&aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa{} }j9%'5&%&>@9,@&AA.97DE69@5&`&K&6>5&%E69%'5&~{aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa{ ~j"A"%&6(DADD.?E&aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa|z j@7A.D.97>.D-(5.9j(A&7"-L6D&EA&`"97D&KD&>-LA.9j>D-959,.?E&aaaaaaaaaaaaaaaaaaaaaaaaaaa|| {j(+.7.D.97&D"@"D(@.AD.?E&Aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa|| |j;5&>D-959,.?E&%&5aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa|} }j 7%E"D.97%&5@7A.D.97>.D-&5.9j(A&7"-L6D&EA&aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa|~ j@7A.D.97>.D-(5.9j(A&7"-L6D&EA&`6("7.A6&A695("E5.@&Aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa|€ {j(,E5D.97D@7A"@.>D.977&55&aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa| {a{j&A+"D&E@A7.5Aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa|ƒ {a|j&A+"D&E@AAaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa|ƒ {a}jJ.AD&D5&AED@&A+"D&E@A%&D@7A"@.>D.97%&5aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa}{ {a~j(,E5D.97"@9.A(&%&A+"D&E@A%&D@7A"@.>D.97aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa} {  |j;5&%&A6."@9Aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa}| |a{j6."@9A&D"7"&@aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa}| |a|j6."@9A&Daaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa}} .a(,E5D.97%&A+"D&E@A%&D@7A"@.>D.97aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa}~ ..j(,E5D.97%&A&++&"D&E@Aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa} .0$ ;,$//"'0.)0$!0,.*".//$*)01(*.'<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<IN j@("D.97%d(>.AA,&aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa}‚ j("7.A6&%&@(,E5D.97%&5d(>.AA,&5D&@7D.+aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa~z {j(,E5D.97>@5&A+"D&E@A%d(>.AA,&aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa~z |j 7+5E&7"&%&5D@7A"@.>D.97&D%&5"-@96D.7&aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa~{ }j 7+5E&7"&%&AAD@E"DE@&AA&"97%.@&Aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa~~ j>.AA,&5D&@7D.+&D>@9,@&AA.97DE69@5&aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa~€  ‚ {j 5D(@D.97%E>@9,@66&%d(>.AA,&E"9E@A%&5>@9,@&AA.97DE69@5&aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa~€ |j97"D.97A%&AI@.7DA%d(>.AA,&E"9E@A%&5>@9,@&AA.97DE69@5&aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa~‚ j(,E5D.97%&5d(>.AA,&%7A5D@7A.D.97(>.D-(5.9j6(A&7"-L6D&EA&aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaz .0$ ; / # '$//KA,LNB0GMA,MHB<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<KJ j&A-(5."A&A69D.+aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa~ j97"D.97A695("E5.@&A%&A-(5."A&A&D{aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa {j;5&%7A5D@7A"@.>D.97aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa |j;5&%7A5d(>.AA,&aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaƒ }j;5&%7A56DE@D.97%&A6."@9Aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa€z ~j97"D.97A%&&D{%7A5&AED@&A>@9"&AAEA%E6(D!95.A6&%&5daaaaaaaaaaaaaaaaaaaaaaaaaaa€{ j9%.+."D.97>9ADD@%E"D.977&55&Aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa€} j;5&A%&A-(5."A&A&D{%7A5>@9,@&AA.97DE69@5& &D 5 D@7A.D.97 (>.D-(5.9j6(A&7"-L6D&EA&aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa€ {j 6>5."D.97%&&D{%7A5>@9,@&AA.97DE69@5&aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa€ |j;5&%&%7A5@7A.D.97>.D-(5.9j(A&7"-L6D&EA&aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa€‚

%0$!/<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<MF

 /1'00/<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<MG .0$' CG<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<MG ,'$$)"/3$0#*!),$")0$."1'0*.5 #'$//,.*(*0/01(*.?''$)2/$2)//<MG .0$' CH<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<OG   #'$// 5)($''5 *.#/0.0 0.)/.$,0$*): ($.*  ) /,'$$)" ,.*".(/ $) '' $!!.)0$0$*)<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<OG

$/1//$*)<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<GIJ ;5&%&A-(5."A&A&D{%7A5>@9,@&AA.97DE69@5&aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa{}~ ("7.A6&%&@(,E5D.97%&5d(>.AA,&>@5&A-(5."A&A&D{aaaaaaaaaaaaaaaaaaaaaaaaaaaa{}ƒ @9,@66&A%d(>.AA,&@(,E5(A>@&D{aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa{~{

'/1,,' ()0$./<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<GJI

))4;+'/2.$)0/'=#$/0*)(.* HG)/',.*".//$*)01(*.'<<<<<<<<<<GJM

,,)$<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<GKG

$'$*".,#$<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<GMM 

 ƒ   

              d9!3&"D.+%&6D-'A&(D(%&"@"D(@.A&@5&@;5&%&A-(5."A&A&D{ E "9E@A %& 5 >@9,@&AA.97 DE69@5&a &A >@9D(.7&A 97D %& D@'A 796!@&EA&A +97"D.97A 695("E5.@&Aa97D@I.5Ad&AD+9"5.A(^D9ED%d!9@%^AE@5d.6>9@D7"&%&5@(,E5D.97%& 5d(>.AA,&5D&@7D.+E"9E@A%&5>@9,@&AA.97DE69@5&&7ED.5.A7D5&69%'5&[email protected]~{a @5AE.D&^3&6&AE.A.7D&@&AA(5"99@%.7D.97%&5D@7A"@.>D.97^5d(>.AA,&5D&@7D.+&D 5 @(,E5D.97 %&A 6."@9 >@  &D { E "9E@A %& 5 D@7A.D.97 (>.D-(5.9j 6(A&7"-L5D&EA&^ E7 >@9"&AAEA %& %.++(@&7D.D.97 @(">.DE57D "&@D.7A A>&"DA %& 5 >@9,@&AA.97DE69@5&a  &DD&.7D@9%E"D.97&AD%.I.A(&&7[email protected]>@D.&Aa7A5>@&6.'@&>@D.&^3&%(D.55&@. 5&A 6("7.A6&A &D 5&A +"D&E@A ?E. +[email protected]&7D 5 >@9,@&AA.97 DE69@5&a & %("@.@. &7 >@D."E5.&@5&69%'5&~{&D5%.++(@&7".D.97(>.D-(5.9j6(A&7"-L6D&EA&^5&A%&EK69%'5&A %& 697 D@I.5a 7A 5 A&"97%& >@D.&^ 3& "966&7"&@. >@ %("@.@& 5&A 6("7.A6&A %& 5d(>.AA,& &D %& 5d(>.AA,& 5D&@7D.+ >E.A (DE%.&@. 5 +$97 %97D "&A 6("7.A6&A A97D .6>5.?E(A&D>@D.".>&7D5>@9,@&AA.97DE69@5&a%&@7.'@&>@D.&^&AD"97A"@(&EK -(5."A&A  &D {^  5 %&A"@.>D.97 %& 5&E@A +97"D.97A 695("E5.@&A &D  5&E@A .6>5."D.97A%7A5>@9,@&AA.97DE69@5&a 

 {z .0$ @ ,.*".//$*)01(*.'     &>@&6.@&"->.D@&&AD"97A"@(5%&A"@.>D.97%&A6("7.A6&A?E."97D@.!E&7D 5>@9,@&AA.97DE69@5&a7A"&"->.D@&^3&@(AE6&5&A%.++(@&7DAA>&"DA%&5>@9,@&AA.97 DE69@5& &D 5&A 69%'5&A ED.5.A(A >9E@ 5d(DE%.&@a >@'A I9.@ %("@.D 5&A "@"D(@.AD.?E&A %&A "&55E5&A"7"(@&EA&A>E.A5&A+"D&E@A?E.+[email protected]&7D5>@9,@&AA.97DE69@5&^3&%("@.@.5& 69%'5& ~{a 7A 5 A&"97%& >@D.& %& "& "->.D@&^ 3& 6& +9"5.A&@. AE@ 5 D@7A.D.97 (>.D-(5.9j6(A&7"-L6D&EA&a



 ? !$)$0$*)0( )$/(/    & "7"&@ %(A.,7& E7 &7A&6!5& %& >D-959,.&A %d9@.,.7&A D.AAE5.@&A I@.(&A &D ?E. A97D"@"D(@.A(&A>@E7&>@95.+(@D.97"&55E5.@&79@65&6&7D.6>9@D7D&a>@9,@&AA.97 DE69@5& @&>@(A&7D& 5d&7A&6!5& %&A (I(7&6&7DA ?E. "97%E.A&7D E %(I&59>>&6&7D %dE7 "7"&@  >@D.@ %dE7& "&55E5& A.7&a & >@9"&AAEA %& >@9,@&AA.97 DE69@5& "97%E.D  5 +9@6D.97%&5DE6&E@&D5"A"%&6(DADD.?E&aE"9E@A%&5>@9,@&AA.97DE69@5&^ %&EK ,@9E>&A %& ,'7&A A97D ++&"D(A` {j 5&A 97"9,'7&A &D | 5&A ,'7&A AE>>@&AA&E@A %& DE6&E@A ?E. +[email protected]&7D 9E .7-.!&7D 5& %(I&59>>&6&7D %& 5 DE6&E@a & 796!@&EA&A 5D(@D.97A++&"D7D5d"D.I.D(%&A97"9,'7&A&D%&A,'7&AAE>>@&AA&E@A%&DE6&E@A97D(D( .%&7D.+.(&Aa 7 %.AD.7,E& 79D66&7D 5&A 6EDD.97A &D 5&A 5D(@D.97A "-@969A96.?E&A ?E. ++&"D&7D %.@&"D&6&7D 5 A(?E&7"& %E ,'7&a @ .55&E@A^  "&A 5D(@D.97A >&EI&7D ++&"D&@ 5d&K>@&AA.97%&A,'7&A&7D9E"-7D%&A6("7.A6&A.6>5.?E(A%7A5D@7A"@.>D.97^%7A5 6DE@D.97&D5AD!.5.D(%&5d6&AA,&@m(>.AA,&5D&@7D.+&D6."@9jn9E%7A5 D@%E"D.97&D5&A@(,E5D.97A>9ADjD@%E"D.977&55&Aa   

 {{   &$             .K"@"D(@.AD.?E&A97D(D(>@9>9A(&A>@a 7-7&Da&.7!&@,>9E@%(+.7.@ 5&A>@9>@.(D(A%dE7D.AAEDE69@5`{j6.7D.&7%&AA.,7EK>@95.+(@D.+A^|j("->>&6&7DEK A.,7EK 7D.>@95.+(@D.+A^ }j @(A.AD7"&  5 69@D "&55E5.@&^ ~j "?E.A.D.97 %& ">".D( @(>5."D.I& .55.6.D(^ j %(@',5&6&7D %E 6(D!95.A6&^ €j .7%E"D.97 %& 5d7,.9,(7'A&^ j "D.ID.97%&5d.7IA.97&D+9@6D.97%&6(DADA&A^‚j("->>&6&7D5%&AD@E"D.97>@5& ALAD'6& .66E7.D.@&  ƒj.7+566D.97 %E D.AAE "7"(@&EK  &D {zj .7AD!.5.D( ,(7(D.?E& m 7-7^ &.7!&@,^ |zzz_ 7-7 s &.7!&@,^ |z{{na & D.AAE DE69@5 &AD E7 D.AAE -(D(@9,'7&"96>9A(%&"&55E5&ADE69@5&A6.AEAA.%&"&55E5&Ag79@65&Ah?E.39E&7DE7 @;5&"D.+%7A5DE69@.,(7'A&a&A"&55E5&A79@65&A?E.+9@6&7D5&AD@96DE69@5A97D %&A"&55E5&A&7%9D-(5.5&A^%&A>(@."LD&A^%&A"&55E5&A%EALAD'6&.66E7.D.@&&D%&A"&55E5&A AD@965&Aa&A%.K"@"D(@.AD.?E&A%dE7D.AAEDE69@5A97D&AA&7D.&55&6&7DA>(".+.?E&AEK "&55E5&A DE69@5&A 6.A "&@D.7&A >@9I.&77&7D (,5&6&7D %& 5 "955!9@D.97 %& "&55&Aj". I&" 5&A "&55E5&A %E AD@96 DE69@5a d&7A&6!5& %& "&A "@"D(@.AD.?E&A @&,@9E>&7D 5 ">".D(%&A "&55E5&A "7"(@&EA&A  >@95.+(@&@ A7A "97D@;5& m6.7D.&7 %&A A.,7EK >@95.+(@D.+A^ ("->>&6&7D EK A.,7EK 7D.>@95.+(@D.+A^ @(A.AD7"&  5 69@D "&55E5.@& &D "?E.A.D.97 %& ">".D( @(>5."D.I& .55.6.D(&  %(I&59>>&@ %&A ">".D(A >9E@ 5.6&7D&@ 5& D.AAEDE69@5&7[email protected]&7DAm%(@',5&6&7D%E6(D!95.A6&^.7%E"D.97%&5d7,.9,(7'A&nD9ED &7("->>7DEALAD'6&.66E7.D.@&m$'%41!{na  

 {|  '%41!J? !213#1'23'04!2 A4,3'22434+-1*@A/1"2 ,&,H!',!1%KIJJ@    d"?E.A.D.97 %& "&A >@9>@.(D(A >@ 5&A "&55E5&A "7"(@&EA&A @(AE5D& %& 6EDD.97A A96D.?E&A %7A 5&A ,'7&A @&?E.A >9E@ 5 >@95.+(@D.97 &D 5 %.++(@&7".D.97 >E.A %& 5 A(5&"D.97 %&A "&55E5&A 6ED(&Aa & >-(796'7& &AD 6>5.+.( >@ 5d.7AD!.5.D( ,(796.?E& %&A "&55E5&A "7"(@&EA&Aa &A "&55E5&A g79@65&Ah (I95E&7D >@9,@&AA.I&6&7D I&@A E7 ADDED "7"(@&EK&7"?E(@7D>5EA.&E@A%&"&A"@"D(@.AD.?E&Aa  

  '$        "  !           #   &A(?E&7$,&%&"9-9@D&A%&DE6&E@A697D@&?E&%&D@'A796!@&EA&A5D(@D.97A ,(7(D.?E&A A97D >@(A&7D&A %7A %&A DE6&E@A &7 "96>@.A97 %E D.AAE A.7 %3"&7Da  %.++."E5D( &AD %d.%&7D.+.&@ 5&A 6EDD.97A ?E. @(AE5D&7D %& 5d.7AD!.5.D( ,(7(D.?E& >@9>@& EK DE6&E@A I.AjjI.A %&A ?E&5?E&A 6EDD.97A ?E. 97D "97%E.D  5d(D!5.AA&6&7D %& 5 DE6&E@ m&! ,!1 !,-+! 3*2 !36-1)^ |z{|na & >5EA^ "&@D.7&A 5D(@D.97A ,(796.?E&A A&6!5&7D A(5&"D.977(&A EK "9E@A %& 5 >@9,@&AA.97 DE69@5&a &5 AE,,'@& ?E& "&A 5D(@D.97A39E&7DE7@;5&+97"D.977&5a@&K&6>5&^%7A5&"A%E"7"&@%EA&.7^ 

 {}      >@(+(@&7D.&55&6&7D  6EDD.97A9E%&A69%.+."D.97A%E 796!@&%&"9>.&A%&,'7&A.7"5E7D      ^^&D{{?E.A97D"97A.%(@(A"966&>9EI7D>9D&7D.&55&6&7D+.@&5d9!3&D %&D-(@>.&m-.7&D5a^|zz€na7A5&"%@&%E"7"&@%EA&.7-(@(%.D.@&^E7&6EDD.97 %(5(D'@& %7A 5& ,'7& {^ E,6&7D& +9@D&6&7D 5& @.A?E& %& %(I&59>>&@ E7 "7"&@ ma -&7s@6.,.7.^|zzna   d(D!5.AA&6&7D%&%.I&@A69%'5&A%&>@9,@&AA.97DE69@5&>[email protected]%&697D@&@?E& {z5D(@D.97A,(7(D.?E&A%.++(@&7D&AA97D7("&AA.@&A5D@7A+9@6D.97DE69@5&a@ &K&6>5&^ 5& ,@9E>& %& a &.7!&@,  D@7A+9@6&@ %&A +.!@9!5AD&A -E6.7A &D  5&E@A "97+(@&@ E7 >9EI9.@ DE69@.,'7& ', 5'5- &7 .7D@9%E.A7D  j                     655jDh&D g @,&j hn &D E7&                j  m5&7!A^ |zz{_ -7&D5a^{ƒƒƒnm$'%41!|na&A>@9D(.7&AI.@5&A97DE7&"D.97>5&.9D@9>.?E&&DA97D5 +9.A">!5&A%&@(>@.6&@>}&D&D">!5&A%d"D.I&@"jL"^E797"9,'7&a&A@(AE5DDA 97D(D(9!D&7EA%7A%&A[email protected],4 !&D7&@&+5'D&7D7.5&@;5&%E6."@9&7I.@977&6&7D^7. "&5E. %& 5 >5AD.".D( %&A "&55E5&A "7"(@&EA&Aa .7A.^ %7A "& 69%'5&^  5D(@D.97A A97D AE++.A7D&A>9E@D@7A+9@6&@%&A"&55E5&A-E6.7&Aa&A@(AE5DDAA.6.5.@&A97D(D(9!D&7EA I&" %dED@&A DL>&A %& "&55E5&A >@.6.@&A -E6.7&A       ` "&55E5&A >E5697.@&A^>@9ADD.?E&A^9I@.&77&AnmM@9I&D5a^|zz|_E7%!&@,&D5a^|zz|_."-&D 5a^|zz{na

 {~ 

'%41!K? ++-13*'23'-, !$'1-*23!2&4+',2/13'1 A4,/!3'3,-+1! A*3#13'-,2 %#,#3'04!2@ A/1"2!'39+, @@!3,'5@?!4'* ',%3&!1- 3-,!1@341!JRRR@ S,3'%",! %1,  45'142MI@   7>5EA%&5d.7AD!.5.D(,(7(D.?E&^%dED@&A+"D&E@A^"966&5&6."@9&7I.@977&6&7D 9E5>5AD.".D("&55E5.@&^>@D.".>&7D5>@9,@&AA.97DE69@5&a  7&++&D^&A"&55E5&ADE69@5&AA97D&7"97D"D%.@&"DI&"5&6."@9&7I.@977&6&7Da E "9E@A %& 5 >@9,@&AA.97 DE69@5&^ 5& AD@96 &AD 69%.+.(^  5 +9.A %7A 5 7DE@& %&A "&55E5&A?E.5&"96>9A&7D6.AEAA.%7A5AD@E"DE@&%&56D@."&&KD@"&55E5.@&m 7-7 s9EAA&7A^|z{|na&6."@9&7I.@977&6&7DDE69@5A&"96>9A&%&+.!@9!5AD&A"D.I(A^?E. >@9%E.A&7D%&A&7ML6&A%&%(,@%D.97%&56D@."&&KD@"&55E5.@&m|^}^ƒn.7A.?E&

 { %&A+"D&E@A%&"@9.AA7"&m ^ ^ ^7D{^ |^ nmaa-9J6."4^&.5A97^s 9A&A^ |zz~_ a  -9J6."4 &D 5a^ |zz~n &D %&A "&55E5&A %E ALAD'6& .66E7.D.@&a  A("@(D.97 %& "&A +"D&E@A %& "@9.AA7"& >&ED >&@6&DD@& 5 >@95.+(@D.97 %&A "&55E5&A (>.D-(5.5&A%3"&7D&Aa.7A.5&A69%.+."D.97A%E6."@9&7I.@977&6&7DDE69@57&A97D>A E7.?E&6&7DE7&"97A(?E&7"&_&55&A>@D.".>&7DEAA.5>@9,@&AA.97DE69@5&m 7-7s 9EAA&7A^ |z{|na 5  (D( %(697D@( ?E& 5d.73&"D.97 %& +.!@9!5AD&A 79@6EK 9E .AAEA %& DE6&E@A666.@&AE,6&7D&5"@9.AA7"&%&ADE6&E@Ama E&D5a^|zz‚na&>&7%7D^ "&A@(AE5DDAA97DAE3&DA"97D@9I&@A&A"@ ',5'31-^5"9j"E5DE@&%&+.!@9!5AD&A79@6EK 5.6.D&5>@95.+(@D.97%&"&55E5&A%&"@".796&m5!&@,&D5a^|z{|na&A+.!@9!5AD&A"D.I(A A97D EAA. ">!5&A %& A&"@(D&@ %&A "LD94.7&A &55&Aj6)6&A &7 6&AE@& %& @&"@ED&@ %&A "&55E5&A %E ALAD'6& .66E7.D.@&a &A "&55E5&A A97D  5 +9.A AA9".(&A  5d(5.6.7D.97 %&A "&55E5&ADE69@5&A6.AEAA."&@D.7AA>&"DA%&5>@9,@&AA.97DE69@5&a.7A.5A("@(D.97 %& "-&694.7&A &D %d.7D&@5&E4.7&A >@ 5&A "&55E5&A %E ALAD'6& .66E7.D.@& .7+.5D@(&A A97D ">!5&A %& AD.6E5&@ 5d7,.9,(7'A& &D %& 69%.+.&@ 5 @(>97A& .66E7.D.@& 7D.jDE69@5& mE6A4. &D 5a^ |zz_ &@!& &D 5a^ |zzzna d7,.9,(7'A& &AD EAA. E7& 69%.+."D.97 .6>9@D7D& %E 6."@9&7I.@977&6&7D DE69@5a 7A 9KL,'7& &D [email protected]&7DA >>9@D(A >@ 5& A7,^ 5&A "&55E5&A DE69@5&A A97D .7">!5&A %& +9@6&@ E7& DE6&E@ %& >5EA %& |"6 %& %.6'D@& m95467^ |zzna 7 "97"5EA.97^ 5& 6."@9&7I.@977&6&7D DE69@5 >@D.".>& "D.I&6&7DE%(I&59>>&6&7D&D5>@9,@&AA.97DE69@5&a  7 >5EA %& "&A %&EK +"D&E@A              n^ 5 >5       (,5&6&7D  5 >@9,@&AA.97 DE69@5&a  %.++(@&7".D.97 "&55E5.@&&ADD@%.D.97&55&6&7D"97A.%(@(&"966&E7>@9"&AAEA .@@(I&@A.!5&aE"9E@A%&A %.K %&@7.'@&A 77(&A^ "& %9,6&  (D( 6.A  65 >@ %&A (DE%&A >9@D7D AE@ 5 D@7A%.++(@&7".D.97^5%(j%.++(@&7".D.97&D5@&>@9,@66D.97"&55E5.@&m 9"-&%5.7,&@s 5D-^|zzƒ_674s5E^|z{zna&A(DE%&A97D>[email protected]%&697D@&@?Ed.5&AD>9AA.!5&%& @&>@9,@66&@ %&A +.!@9!5AD&A &7 "&55E5&A >5E@.>9D&7D&A &7 AE@&K>@.67D 5&A +"D&E@A %& D@7A"@.>D.97 "Dj}i~^ 9K|^ ~ 7% "jL" m4-A-. s 674^ |zz€na & 67.'@& A.6.5.@&^5d&K>@&AA.97%& |&Dƒ^%&A                            D 5&E@ %(j%.++(@&7".D.97 &D 5&E@ >&@6&D%&@&+9@6&@%&A,57%&A666.@&Aa7A5&"97D&KD&%&5>@9,@&AA.97DE69@5& 666.@&^"&A+"D&E@AA97D&K>@.6(A%7A5&A"&55E5&A6(DADD.?E&Ajj|}{&DA97D &AA&7D.&55&A %7A 5& 6.7D.&7 %&A >@9>@.(D(A DE69@.,(7.?E&A &D 6(DADD.?E&A m E9 &D 5a^

 {€ |z{|na&>5EA^5d.7%E"D.97%dE7>@9,@66&%&D@7A.D.97(>.D-(5.9j6(A&7"-L6D&EA&mI9.@ j 5 D@7A.D.97 (>.D-(5.9j6(A&7"-L6D&EA&n %7A %&A "&55E5&A D@7A+9@6(&A +[email protected]& 5&E@ >9D&7D.&5 DE69@.,(7.?E& m7. &D 5a^ |zz‚_ 9@&5 &D 5a^ |zz‚na &A %.++(@&7D&A %977(&A 697D@&7D ?E& 5 D@7A+9@6D.97 &D 5           &D ?E& 5     %&"&55E5&A"7"(@&EA&A>9E@@.D%.@&"D&6&7D>@&7%@&>@DEKA %&D@7A+9@6D.97"&55E5.@&A&D%&>@9,@&AA.97DE69@5&a 

 ($      "      )&#   5&K.AD&%&D@'A796!@&EK69%'5&A"&55E5.@&A>9E@5d(DE%&%&5D@7A+9@6D.979E %&5>@9,@&AA.97DE69@5&a&A69%'5&A"&55E5.@&A>&EI&7D>@9I&7.@%&"&55E5&A79@65&A^ "d&ADjj%.@&.A95(&A>@D.@%dE7>@(5'I&6&7Da@5AE.D&^"&A"&55E5&AA97D69%.+.(&AA9.D >@%&AD@.D&6&7DAA9.D>@5AE@j&K>@&AA.97%d97"9,'7&Aa&69%'5&%(I&59>>(>@a &.7!&@,(I9?E(%7A5>@D.&>@("(%&7D&+.D>@D.&%&"&DD&"D(,9@.&adED@&A69%'5&A A97D .AAEA %& >@(5'I&6&7DA DE69@EK [email protected] 9E -E6.7Aa &A 69%'5&A >@9I.&77&7D %& DE6&E@AA>97D7(&A"-&M5d-966&6.A>&EI&7DEAA.>@9I&7.@%&DE6&E@A.7%E.D&A"-&M5 [email protected]&A69%'5&AA97D"&7A(A@&>@(A&7D&@E>5EA>@'A5&A"@"D(@.AD.?E&A%&5DE6&E@ .7.D.5&a7A5&"%@&%E"7"&@%EA&.7^.5&K.AD&>5EA.&E@A%.M.7&A%&5.,7(&A-E6.7&A.7A. (D!5.&A ?E. @&+5'D&7D 5 %[email protected]( %&A "7"&@A %E A&.7` .7A. %&A 69%'5&A "&55E5.@&A &K>@.67D 5& @("&>D&E@ EK 9&AD@9,'7&A mjn^ AE@j&K>@.67D j|^ 9E &7"9@& %&A 5.,7(&AD@.>5&A7(,D.I&Amjj|}{na   &A69%'5&A"&55E5.@&A>&EI&7DEAA.@&+5(D&@5&AD%&%&>@9,@&AA.97%&565%.&a d&AD5&"A%E69%'5&~{a&69%'5&"&55E5.@&&AD"96>9A(%&?ED@&5.,7(&A"&55E5.@&A %(@.I7D%dE7&DE6&E@666.@&A>97D7(&"-&M5[email protected]5!i"mA54A97s.55&@^{ƒƒ|na 5EA.&E@A A9EAj>9>E5D.97A %& "&55E5&A 97D (D( &7AE.D& @(.73&"D(&A %& 67.'@& AL7,(7.?E& %7A 5&A "9EAA.7&DA %.>&EK 666.@&A &D %.++(@&7D&A ">".D(A 6(DADD.?E&A 97D (D( "@"D(@.A(&A &7 +97"D.97 %& 5 A9EAj>9>E5D.97 .73&"D(&a &5  >[email protected] %d(D!5.@ ~ 5.,7(&A "&55E5.@&A`5&A€^{€‚^~z&D~{a&A%.++(@&7D&A5.,7(&AA97DD9ED&A">!5&A%& +9@6&@E7&DE6&E@>@.6.@&a&>&7%7DD9ED&A"&A5.,7(&A97D%&A>@9>@.(D(A6(DADD.?E&A %.++(@&7D&Aa &A € 7& +9@6&7D 36.A %& 6(DADA&A^ 5&A {€‚ A97D ">!5&A

 { %d&7I-.@ 5&A ,7,5.97A 5L6>-D.?E&Aa &A ~z +9@6&7D %&A 6."@96(DADA&A &D 5&A ~{ +9@6&7D%&A6"@96(DADA&Am$'%41!L!3MnmA54A97s.55&@^{ƒƒ|na 

 '%41!L?&#,-38/!2 !234+!412+++'1!2OP>JOQ>MIP!3MJ@ ',%3!34,(-412/1"2*G',(!3'-, !N7JIN!**4*!2 ,2*%*, !+++'1! !2-41'2 *B>*!234+!412/1'+'1!2!3*!2/-4+-,2-,3#3#/1#*!5#2!3',*42!,/1$$',!/-414, !7+!,&'23-*-%'04!@44,!+#3232!/4*+-,'1!,A#3#-2!15#!&!9*!22-41'2 ',(!3#!25!*!2!**4*!2OP!3JOQ>!/!, ,3 !2+'1-+#3232!2/4*+-,'1!2!3 +1-+#3232!2-,3#3#-2!15#!2&!9*!22-41'2',(!3#!2!35!*!2!**4*!2MIP!3MJ E$*"&!2F@A/1"243!131!!3*,!1!2!1&KIJI@   &69%'5&%&>@9,@&AA.97DE69@5&>@(A&7D&5dI7D,&%d)D@&E769%'5&',5'5-9E 5&A"&55E5&A>&EI&7D)D@&.73&"D(&A"-&M%&A[email protected]a7ED@&I7D,&63&E@&AD?E&"&A %.++(@&7D&A 5.,7(&A A97D AL7,(7.?E&A^ "& ?E. >&@6&D %& Ad++@7"-.@ %E +97% ,(7(D.?E&a , 5'31-^"&A5.,7(&A97D%&A>@9>@.(D(A?E.@&+5'D&7D5&E@A>-(79DL>&A',5'5-a@&K&6>5&5&A "&55E5&A~{A97D">!5&A%&6.,@&@&D%&%.,(@&@E7&6D@."&%&"955,'7&59@A?E&5&A "&55E5&A€7d&7A97D>A">!5&Aa

 {‚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

 {ƒ   )$         @6.5&A"@"D(@.AD.?E&A%("@.D&A>@ 7-7&D&.7!&@,^97%.AD.7,E&5">".D( %&A "&55E5&A DE69@5&A  "@90D@&  %.AD7"& %E A.D& .7.D.5&6&7D DD&.7Da &DD& "@9.AA7"& 77&K& &AD >>&5(& 6(DADA&a . 5&A %.I&@A D@.D&6&7DA &D 5d!5D.97 %& 5 DE6&E@ >@ "-.@E@,.& 97D >[email protected] %dE,6&7D&@ A.,7.+."D.I&6&7D 5& DEK %& AE@I.& %&A >D.&7DA^ 5 +9@6D.97 %& 6(DADA&A &AD E7 +"D&E@ %& 6EI.A >@979AD."a  %.AA(6.7D.97 %&A 6(DADA&A&D5&E@@(A.AD7"&EKD@.D&6&7DA&AD5d9@.,.7&%&>5EA%&ƒz†%&569@D5.D( AE.D&  E7 "7"&@a  +9@6D.97 %& 6(DADA&A >>&5(& EAA. >@9,@&AA.97 DE69@5& &AD E7 >@9"&AAEA "96>5&K& ?E. >&ED )D@& @(AE6( >@ 5 g"A"%& 6(DADD.?E&h m$'%41! NF@  "A"%&6(DADD.?E&>&ED.7D&@I&7.@D;D%7A5&%(I&59>>&6&7D%&5DE6&E@>@.6.@&^9E E "97D@.@& >5EA D@%.I&6&7D m7A.&E &D 5a^ |zz‚na 7A E7 >@&6.&@ D&6>A^ ?E&5?E&A "&55E5&AI97DA&%(D"-&@%&5DE6&E@>@.6.@&^6.,@&@&D&7I-.@5&D.AAE%&A9ED.&7a'A 59@A^&55&A%&I.&77&7D">!5&A%d&7D@&@^>E.A%&A9@D.@%&5".@"E5D.97A7,E.7&a&A"&55E5&A >&EI&7D &7AE.D& @&AD&@  5d(DD %9@67D 9E +9@6&@ %&A 6(DADA&Aa & >@9"&AAEA %& "9597.AD.97&7D@.7&E7&@&j>@95.+(@D.97%&A"&55E5&Aa&>@9"&AAEA%&>@9,@&AA.97DE69@5& &D %d.7IA.97 &AD "@"D(@.A( >@ 5 >&@D& %d&K>@&AA.97 %& 5 j"%-(@.7&a &DD& @(,E5D.97 AA9".(&  "&@D.7&A 6EDD.97A AE@ 5& ,'7& %& 5 j"%-(@.7& m {n^ A97D +@(?E&66&7D 9!A&@I(&A %7A 5 >@9,@&AA.97 %&A "@".796&A -E6.7A m&@K s I7 9L^ |zzƒna  j "%-(@.7& &AD E7& >@9D(.7& "5( %&A 397"D.97A %-(@&7D&A 7("&AA.@& E 6.7D.&7 %E >-(79DL>&(>.D-(5.5a 5&AD!.&7(D!5.?EdE7&+9@D&&K>@&AA.97%&5j"%-(@.7&E7&++&D 7D,97.AD& AE@ 5&A >@9"&AAEA %d.7IA.97 &D %& +9@6D.97 %& 6(DADA&Aa 7 >5EA %& 5 @(,E5D.97%&5j"%-(@.7&^E7&!.AA&%d&K>@&AA.97%&A,'7&A"9%7DA>9E@%&A>@9D(.7&A %d%-(A.97"&55E5&Aj"&55E5&A9E"&55E5&Aj6D@."&&KD@"&55E5.@&.7A.?EdE7&E,6&7DD.97%& 5d&K>@&AA.97 %&A ,'7&A "9%7DA %&A >@9D(.7&A %d%-(A.97 AA9".(&A  5 6.,@D.97^ A97D 9!A&@I(A E "9E@A %& 5 >@9,@&AA.97 DE69@5&a 7 %&A >@9"&AAEA 63&E@ .6>5.?E( %7A 5d(D!5.AA&6&7D%&5>@9,@&AA.97DE69@5&&AD5D@7A.D.97(>.D-(5.9j6(A&7"-L6D&EA&a   D@7A.D.97 (>.D-&5.9j6(A&7"-L6D&EA& mn &AD E7 >@9"&AAEA %& D@7A%.++(@&7".D.97 E "9E@A %E?E&5 %&A "&55E5&A (>.D-(5.5&A I97D (I95E&@ &7 "&55E5&A 6(A&7"-L6D&EA&Aa &5 A& "@"D(@.A& >@ 5 >&@D& %&A "@"D(@.AD.?E&A (>.D-(5.5&A &D 5& %(I&59>>&6&7D %&A >@9>@.(D(A 6(A&7"-L6D&EA&Aa & >@9"&AAEA  (D( 5@,&6&7D %("@.D %7A %&A "97%.D.97A >-LA.959,.?E&A "966&^ >@ &K&6>5&^ E "9E@A %E %(I&59>>&6&7D

 |z &6!@L977.@&^ 9E "-&M 5d%E5D& 59@A %& 5 "."[email protected]D.97^ 6.A EAA. %7A %&A "97%.D.97A >D-959,.?E&A 59@A %& 5 >@9,@&AA.97 DE69@5& %&A "@".796&Aa E 7.I&E 695("E5.@&^ 5  &AD "@"D(@.A(& >@ E7 >@9,@66& D@7A"@.>D.977&5 %.@.,( >@ %&A +"D&E@A %& D@7A"@.>D.97 "966& 5&A >@9D(.7&A 7.5^ &! 9E J.ADa 7 >5EA %& "&A >@9,@66&A D@7A"@.>D.977&5A^ 5  &AD (,5&6&7D @(,E5(& E 7.I&E %& 5 "-@96D.7&^ >@ %&A 6("7.A6&A+.A7D.7D&@I&7.@%&A6."@9A&D%&5d(>.AA,&5D&@7D.+a.7A.5&ADE7 &K&6>5&%&>@9"&AAEA+.A7D.7D&@I&7.@D9EA5&A7.I&EK%&5@(,E5D.97%&A,'7&Aa 7A "& "->.D@&^ >@'A I9.@ !@.'I&6&7D @&>5"( 5  %7A A97 "97D&KD& >-LA.9j >D-959,.?E&^3&%(D.55&@.5&A%.++(@&7D&A"9E"-&A%&@(,E5D.97%&A,'7&A>@D.".>7D"& >@9"&AAEAa

      



      

! #  

"   '%41!N? 2 !+#3233'04!@ /3#! A/1"2&!!*H!',!1%KIJJ       

 |{  ? .)/$0$*),$0# '$*? /)#5(01/;*)040,#5/$*?,0#*'*"$-1<    d(DE%& %& 5 >@9,@&AA.97 DE69@5& >&ED A& @(5.A&@ &7 (DE%.7D %&A 69%'5&A "&55E5.@&A @&+5(D7D 5&A "@"D(@.AD.?E&A %E D.AAE DE69@5a &>&7%7D^ 97 >&ED (,5&6&7D (DE%.&@ %&A >@9"&AAEA ?E. A97D ED.5.A(A >@ 5&A "&55E5&A 65.,7&Aa  D@7A.D.97 (>.D-'5.9j 6(A&7"-L6D&EA& &AD E7 &K&6>5& %& >@9"&AAEA >-LA.959,.?E& ?E. .7D&@I.&7D  %.++(@&7DA 696&7DA%E%(I&59>>&6&7D&D?E.&AD"9@@96>EE"9E@A%&5>@9,@&AA.97DE69@5&a  

  21 '! '%'%&'$(&   E"9E@A%&5D@7A.D.97(>.D-(5.9j6(A&7"-L6D&EA&5"&55E5&(>.D-(5.5&I>&@%@& 5d&K>@&AA.97 %& >@9D(.7&A A>(".+.?E&6&7D (>.D-(5.5&A^ &7 >@D."E5.&@ 5&A "97AD.DE7DA %&A 397"D.97A .7D&@"&55E5.@&A^ &D &K>@.6&@ %&A 6@?E&E@A 6(A&7"-L6D&EK D&5A ?E& %&A >@9D(.7&A>&@6&DD7D5%(,@%D.97%&556&!A5&a&DD&%.++(@&7".D.97Ad""96>,7& %&"-7,&6&7DA69@>-959,.?E&A"@"D(@.AD.?E&A^"966&5&@&69%&5,&%E"LD9A?E&5&DD& I&"5>&@D&%&5d"D.7&"9@D."5&&D5+9@6D.97%&56&55.>9%&A^.7A.?E&5d"?E.A.D.97%& >@9>@.(D(A%&6.,@D.97&D%d.7IA.97a&>@9"&AAEA>&ED)D@&@(I&@A.!5&59@A%&5D@7A.D.97 6(A&7"-L69j(>.D-(5.5& mn E "9E@A %& 5 +9@6D.97 %&A 6(DADA&Aa @6. 5&A 6@?E&E@A.6>5.?E(A%7A5^979!A&@I&5>&@D&%d&K>@&AA.97%&A>@9D(.7&A"966&5 j"%-(@.7&m%-{n9E5d9""5E%.7&m"57nm397"D.97A"&55E5.@&A(>.D-(5.5&An&D5dE,6&7DD.97 %&5j"%-(@.7&m%-|nm"&55E5&A6(A&7"-L6D&EA&An&D%&A+"D&E@A%&D@7A"@.>D.97"5(A %&5mj>9E@+"D&E@A%&D@7A"@.>D.97AA9".(A5n"966&5&A>@9D(.7&A 7.5A^&!A&DJ.ADm-.&@L^"59?E&^ E7,^s.&D9^|zzƒna&A6@?E&E@AA97D5&A>5EA "9E@7DA 6.A !&E"9E> %dED@&A >&EI&7D )D@& ED.5.A(Aa 7 &++&D^ %&A (DE%&A @("&7D&A ED.5.A7D 5& A(?E&7$,& 6AA.+ +97D (DD %& >5EA %& }zzz ,'7&A @(,E5(A 59@A %& 5  m->.@9&D5a^|z{{na   

 || 31"#'!!$(    &ADE7>@9"&AAEA!.959,.?E&?E.7&A&@&AD@&.7D>A5+9@6D.97DE69@5&a 7 &7 %.AD.7,E& >5EA.&E@A DL>&Aa   %& DL>& { .7D&@I.&7D 59@A %E %(I&59>>&6&7D &6!@L977.@&a 55& &AD .7%.A>&7A!5& D@'A D;D E "9E@A %E %(I&59>>&6&7D 59@A %& 5 +9@6D.97%&5d&7%9%&@6&>@.(D5^%&5,AD@E5D.979E5%(56.7D.97%&5"@)D&7&E@5&a %&DL>&|&ADAA9".(&EK>@9"&AAEA%&"."[email protected]D.97D.AAE5.@&"-&M5d%E5D&&D%&A >-(796'7&A .7+566D9.@&Aa   %& DL>& } &AD AA9".(&  5 >@9,@&AA.97 DE69@5& m 55E@.s&.7!&@,^|zzƒna  &AD%97"%d!9@%E7>@9"&AAEA797>D-959,.?E&?E.>&@6&D5d"?E.A.D.97%& >@9>@.(D(A 6.,@D9.@&A &D .7IA.I&A >@ %&A "&55E5&A (>.D-(5.5&A ?E. A97D 7("&AA.@&A E %(I&59>>&6&7D&D5d-96(9ADA.&D.AAE5.@&a&5.&7&7D@&&D>@9,@&AA.97DE69@5& "966&7"( >@ 5d9!A&@ID.97 %& 6(DADA&A %& "@".796&A 6.A +GD 597,D&6>A %(!DDE^ 3EA?Ed56.A&&7(I.%&7"&%&5',5'5-m@!5&DM&D5a^|zz{na%.++."E5D(%&6&DD@& &7(I.%&7"&5',5'5-&AD>@9!!5&6&7D%E&>5EA.&E@A"EA&Aa@&6.'@&6&7D^%7A5&A DE6&E@A^A&E5E7>&D.D796!@&%&"&55E5&A&AD">!5&%&@(5.A&@5^"d&ADE7>@9"&AAEA D@'A>&E&++.""&a&A"&55E5&AA97D5d.7D&@+"&&7D@&5DE6&E@&D5&D.AAE%&A9ED.&7a55&A A("@'D&7D%&A"LD94.7&A^%&A>@9D(A&A&D@&69%'5&7D5&AD@96m7,.9,&7'A&^@&69%&5,& %& 5 6D@."& &KD@"&55E5.@&na & AD@96 A("@'D& %&A +"D&E@A^ ?E. &7 @&D9E@^ +[email protected]&7D 5  %& "&A "&55E5&Aa &EK.'6&6&7D^ 5  ', 5'5- %7A 5&A DE6&E@A >&ED )D@& D9D5& 9E >@D.&55&^"&?E.5@&7%%.++.".5&6&DD@&&7(I.%&7"&a@9.A.'6&6&7D^5&AD@(I&@A.!5&^ "&?E.D(69.,7&%&5>5AD.".D(%&A"&55E5&ADE69@5&Aa&A"&55E5&A6(DADD.?E&A>&EI&7D @&,,7&@E7>-(79DL>&(>.D-(5.5>&@6&DD7D5&E@>@95.+(@D.97a&DD&@([email protected]%&5 @&7+9@"&5d&++."".D(%&A"&55E5&A"7"(@&EA&A+9@6&@%&A6(DADA&A',5'5-m"8&D5a^ |z{|_A.^97-&@^E@>-L^-E^s7,^|z{|na     &AD %97" E7 >@9"&AAEA D@7A.D9.@& &D 5&A +"D&E@A %& D@7A"@.>D.97 "5(A .6>5.?E(A%7A5%.++(@&7".D.97m7.5A^!A^I9.@>@D.&AE.I7D&nA97DD@'A+@(?E&66&7D AE@&K>@.6(A %7A >5EA %& %.K DL>&A %& "@".796&A m 7"-&Mj.55: &D 5a^ |z{|na 7 >5EA %& 5d"?E.A.D.97%&">".D(A6.,@D@."&A&D.7IA.I&A^5&AD.6>5.?E(&%7A5@(A.AD7"& 5d>9>D9A&&D5d,.9,(7'A&m-.&@L&D5a^|zzƒna(,5&6&7D(D(>@9>9A(&"966& E76("7.A6&">!5&%&,(7(@&@%&A"&55E5&AA9E"-&A"7"(@&EA&Am7.&D5a^|zz‚na

 |}  7 "97"5EA.97^ %7A 5&A "&55E5&A "7"(@&EA&A^ 5  &AD 5 @("D.ID.97 &D 5 "9@@E>D.97 %dE7 >@9,@66& "&55E5.@& 79@65 >@ 5&A "&55E5&A DE69@5&Aa & >@9"&AAEA D(69.,7&%&5>5AD.".D(%&A"&55E5&ADE69@5&A%d9@.,.7&(>.D-(5.5&Aa 

41 ('! % &'! #'!1 & + '(&     >&ED )D@& .7%E.D& >@ E7& 5@,& I@.(D( %& A.,7EK "966& 5&A +"D&E@A %& "@9.AA7"&>@5d.7D&@6(%..@&%&A.7D(,@.7&A^%&A"LD94.7&A9EA9EA5d&++&D%&%(+9@6D.97 6("7.?E&a@6.5&A"LD94.7&A.7%E"D@."&A%&5^97@&D@9EI&5& X^5d ^5d ^5&  ^5& ^5& .7A.?E&%&A"96>9A7DA%&AI9.&A%&A.,75.AD.977Di!&Dj"D(7.7&^ &%,&-9,^4!&D9D"-a@&K&6>5&^%7A5I9.&"797.?E&%E X^5&5.,7% XA& 5.&E7"96>5&K&%&@("&>D&E@A+9@6(A>@%&EK>@9D(.7&A{&D%&EK>@9D(.7&A|a &DD& 5..A97 .7.D.& E7& "A"%& %& A.,75.AD.97A .7"5E7D 5 >-9A>-9@L5D.97 %& { >@ |a & %96.7&  m@."-& &7 @(A.%EA ,5L".7& &D A(@.7&n %& { >-9A>-9@L5( >&@6&D 5d.7D&@"D.97 I&" 5&A >@9D(.7&A 6%| &D 6%} I. 5&E@A %96.7&A  |a &A >@9D(.7&A 6%|&D6%}A97D&7AE.D&>-9A>-9@L5(&A>@{E69D.+A.DE(5d&KD@(6.D(j [email protected]5 %E %96.7&  | m49 &D 5a^ {ƒƒn m$'%41! €na &A >@9D(.7&A 6%| &D 6%} .7A.>-9A>-9@L5(&AA97D">!5&A%&"-7,&@%&"97+9@6D.97&D%&A&5.&@6%~m-"49 &D 5a^ |zz~na & "96>5&K& &AD 59@A @&59"5.A( %7A 5& 79LEa & A.,75 %& 59"5.AD.97 7E"5(.@&mn&ADA.DE(%7A5&%96.7& {%&A>@9D(.7&A6%Aa&%96.7&"97+'@& EK >@9D(.7&A 6%} &D 6%~ 5&E@ ">".D(  5.&@ 5d m-. &D 5a^ {ƒƒ‚n^ 59@A ?E& 5 >@9D(.7&6%|7d&AD>A">!5&%&5.&@5dm&775&@&D5a^{ƒƒƒna&A>@9D(.7&A6%} &D6%~A&5.&7DA>(".+.?E&6&7D%&A(5(6&7DA%&@(>97A&m6%.7%.7,5&6&7DnE 7.I&E%&A>@969D&E@A%&A,'7&A".!5&Am$'%41!€na&>&7%7D55..A97%&A6%A5d &AD%&+.!5&++.7.D(&D%&+.!5&A>(".+.".D(a&>5EA"&A>@9D(.7&A.7D&@,.AA&7D%&67.'@& A>(".+.?E& &D &++.""& I&" %& 796!@&EK >@D&7.@&A 7E"5(.@&A &D 79D66&7D %&A @(,E5D&E@A"9jD@7A"@.>D.977&5Aa@6."&A"9j@(,E5D&E@AD@7A"@.>D.977&5A^97@&D@9EI&%&A 6&6!@&A %& 5 +6.55& 9@4-&% D&5A ?E& 9K { m &@6.7 &D 5a^ |zzzn 9E  9K{^} &D ~ m&97& &D 5a^ |zz~na 7 @&D@9EI& (,5&6&7D 5d -(5."A&  m>€‚n m@7&@ &D 5a^ |zz~na 

 |~ 

'%41!O?-'! !2'%,*'23'-, !2/1-3#',!2+ 2!,1#/-,2!4=@ !31!/1#2!,3!,3 !2/1-3#',!2-C1#%4*31'!2!*31,21'/3'-,@  & =&ADE7>E.AA7D.7%E"D&E@%&5&D+[email protected]&5d&K>@&AA.97%dE7>7&5%& +"D&E@A %& D@7A"@.>D.97 "5(A "966& 5&A +"D&E@A 7.5A &D &!A I. 5 I9.& "797.?E& %E  Xa&AI9.&A797"797.?E&AA97D7(769.7A7("&AA.@&A>9E@5@(5.AD.97%dE7& "96>5'D& &D 5d"D.ID.97 %& "&@D.7&A 6(D559>@9D(.7A&A ma a -7,^ |zzƒna &  X &AD (,5&6&7D ">!5& %d.7%E.@& 9E %d.7-.!&@ 5d&K>@&AA.97 %d 797 "9%7D 9E %& >@9D(.7&A .7D&@I&77D  %.++(@&7DA 7.I&EK %& "97D@;5&a @ &K&6>5&^ 5&  jX &AD ">!5& %d"D.I&@ 5d&K>@&AA.97 %d 6(D-L5jD@7A+(@A& m69%.+."D.97 (>.,(7(D.?E&n^ %d.7%E.@& 9E %d.7-.!&@

 | 5d&K>@&AA.97%&6."@9A9E%&@(>@.6&@5d&K>@&AA.97%&+"D&E@A%d(>.AA,&a&A%.++(@&7DA 7.I&EK%&@(,E5D.97%&5A&@97D%("@.DA%7A5>@D.&AE.I7D&a 



 ? .)/$0$*),$0# '$*? /)#5(01/;( )$/(/(*' 1'$./   & @;5& %&A +"D&E@A %& D@7A"@.>D.97 %7A 5  &AD !.&7 (D!5.a &@D.7A +"D&E@A @(>@.6&7D %& 67.'@& %.@&"D& 5d&K>@&AA.97 %& 5 j"%-(@.7& m {n &7 A& +.K7D AE@ A97 >@969D&E@a &DD& >&@D& %& 5 j"%-(@.7& &AD E7 (I(7&6&7D +97%6&7D5 "@ "d&AD E7 6@?E&E@%&5>&@D&%&A397"D.97A%-(@&7D&AmA>(".+.?E&AEK"&55E5&A(>.D-(5.5&An(D>& &AA&7D.&55& %7A 5 ^ 6.A EAA. >@"& ?E& "&DD& >&@D& .7%E.D E7& "A"%& %& A.,75.AD.97A+[email protected]7D5m-&&59"4s 9-7A97^|zz}na7>5EA%&5j"%-(@.7&^.5&AD (D!5. ?E& "&A +"D&E@A %& D@7A"@.>D.97 "97D@;5&7D %&A >@9,@66&A D@7A"@.>D.977&5A 7("&AA.@&A5a&>E.A?E&5?E&A77(&A^.5>>@0D"5.@&6&7D?E&57d&AD>A A&E5&6&7DE7>@9,@66&,9EI&@7(>@%&A+"D&E@A%&D@7A"@.>D.97a&796!@&EKED@&A 7.I&EK %& @(,E5D.97 %&A ,'7&A "966& 5&A 6."@9A^ 5d(>.AA,& 5D&@7D.+ &D 5&A 69%.+."D.97A >9ADjD@%E"D.977&55&A^ A97D .6>5.?E(A %7A 5 a  &A 6("7.A6&A %& @(,E5D.97%&5A97D"96>5&K&A>5EA.&E@A>9.7DA%&IE&Aa@&6.'@&6&7D^.5&K.AD&%&A @(,E5D.97A"@9.A(&A&7D@&+"D&E@A%&D@7A"@.>D.97a&@D.7A+"D&E@A%&D@7A"@.>D.97A97D ">!5&A %d&7 "D.I&@ %dED@&A "& ?E. @&7+9@"& 5 a 5A A97D (,5&6&7D ">!5&A %& A& @(,E5&@&EKj6)6&A&7AdED9j"D.I7D9E&7AdED9j.7-.!7Da&EK.'6&6&7D^5&A%.++(@&7DA 7.I&EK %& @(,E5D.97 A97D .7D.6&6&7D "977&"D(Aa &A +"D&E@A %& D@7A"@.>D.97 A97D 69%.+.(A>9ADjD@7A"@.>D.977&55&6&7D++&"D7D.7A.5&E@AD!.5.D(a&A6."@9A>@D.".>7D 5".!5&7D5+9.A%&A+"D&E@A%&D@7A"@.>D.97"5(A&D%&A&++&"D&E@A%&5a&A %.++(@&7DA7.I&EK%&@(,E5D.97A97D%(D.55(A%7A5&AA9EAj>@D.&AAE.I7D&Aa     

 |€ 21('! '% &%#'!    &A +"D&E@A %& D@7A"@.>D.97 .6>5.?E(A %7A 5  97D %d!9@% (D( .%&7D.+.(A E "9E@A%E%(I&59>>&6&7D&6!@L977.@&a 5A>&EI&7D)D@&"5AA(A&7%&EK"D(,9@.&A`%dE7& >@D"&EK?E.@(,E5&7D%&67.'@&%.@&"D&5d&K>@&AA.97%&5j"%-(@.7&&7A&+.K7DAE@A97 >@969D&E@ &D %dED@& >@D "&EK ?E. @(,E5&7D 5 j"%-(@.7& %& 67.'@& .7%.@&"D&a &A +"D&E@A7.5A^&!A^}&D ‚A&+.K&7D%.@&"D&6&7DAE@5&>@969D&E@%&j"%-(@.7&&D @(>@.6& A97 &K>@&AA.97a &A +"D&E@A J.AD{^ 9K|^ ,99A&"9.%^ ~^  { &D { A&6!5&7D">!5&A%&%("5&7"-&@5A7AA&+.K&@%.@&"D&6&7DAE@5&>@969D&E@%&5j "%-(@.7&a&>5EAD9EA"&A+"D&E@A%&D@7A"@.>D.97A97D">!5&A%d.7%E.@&9F%&@(>@.6&@ E7,@7%796!@&%&,'7&A.6>5.?E(A%7A5&A397"D.97A"&55E5.@&A9E%7A5d"?E.A.D.97%&5 69D.5.D(&D5%(,@%D.97%&56D@."&&KD@"&55E5.@&m3*!{na    

*!J? '23!2 !2/1','/47%",!2'*!2 !2$3!412 !31,21'/3'-, !*@ ,1-4%!*C ',!', '04,3*1! -, ,! A3'-, !2C@,*!4*!23!+C '**4231,3*!21#%4*3'-,21-'2#!2 !2$3!412 !31,21'/3'-, !*A@ /3# A/1"2 &'!118H*>!**>KIIR!3,&!9C'**-H*> >KIJK@        

 |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|‚ {a{j&A+"D&E@A7.5A   &A +"D&E@A %& 5 +6.55& 7.5^ 7.5{ m7.5n^ 7.5| m5E,n &D 7.5} m6E"n "97D.&77&7D%&A%96.7&A%9.,D%&M.7"&7[email protected]5%&5>@9D(.7&?E.5&E@>&@6&DD&7D %& 5.&@ 5d AE@ %&A %96.7&A j!9K m  na 7 [email protected]5^ 97 %.AD.7,E& E7 %96.7&  ?E.5&E@>&@6&D%d.7D&@,.@I&"%&A>@9D(.7&A%&69%.+."D.97%&5"-@96D.7&9E %&A "96>5&K&A "9jD@7A"@.>D.977&5A "966& {j"9&AD^ {i| 9E 3E!jj| mI9.@ 3*! | >9E@ 5 5.AD& %&A .7D&@"D7DA %&A jna &A +"D&E@A 7.5{ &D 7.5| @(>@.6&7D5d&K>@&AA.97%&>@9D(.7&A%&A397"D.97A"&55E5.@&A"966&5&A"5E%.7&Am397"D.97A %-(@&7D&An 5d9""5E%.7& m397"D.97A A&@@(&An &D "D.I&7D 5c&K>@&AA.97 %& 5 I.6&7D.7&^ 5 +.!@97&"D.7&&D%&A6(D559>@9D(A&AmI9.@3*!{na&A>@9D(.7&A7.5{&D7.5|@(,E5&7DE7 ,@7%796!@&%&,'7&A"966E7A"966&5&@("&>D&E@5I.D6.7&m@@.!&D5a^|zzƒn 6.A"&@D.7&A&K>(@.&7"&A%(697D@&7D?E&?E&5?E&A,'7&AA97D@(,E5(A>@5dE7&9E>@ 5dED@&m9@&79jE&79&D5a^|zz€na d"D.I.D(%&A+"D&E@A%&D@7A"@.>D.97%&5>&ED)D@&69%.+.(&>@%&A69%.+."D.97A >9ADjD@%E"D.977&55&Aa  >-9A>-9@L5D.97 %& 7.5{ mA(@.7& {{ &D A(@.7& ƒ|n &D 7.5| mA(@.7&~n+[email protected]&7D5&E@A"D.I.D(AD@7A"@.>D.977&55&A59@A?E&5>-9A>-9@L5D.97>@ A4}j X.7-.!&5&E@A"D.I.D(Aa7.5{&D7.5|>&EI&7D)D@&(,5&6&7DE!.?E.D.7.5(&A&D%(,@%(&Aa 

{a|j&A+"D&E@AA   &A+"D&E@A%&5+6.55&^{mY{n^|m {n&D}A97D%&A>@9D(.7&A "97D&77D%&EK,@9E>&A%&%96.7&A%9.,D%&M.7"A(>@(A>@E7-969j%96.7&&DE7 %96.7&%&5..A97EK6%Aa5..A975d&AD6(%.(&>@5&E@A%96.7&A%9.,D%& M.7"ad.7D&@"D.97%&A>@9D(.7&AI&"D&ADE7&6@?E&%&@(>@&AA.97%&6)6&?E& 5d.7D&@"D.97&7D@&{&D.>€z^ {&D.@D{m>@9D(.7&A%E@&69%&5,&"-@96D.7.&7na& 6)6&5d.7D&@"D.97%&|I&"5&"96>5&K&E^&ADAA9".(&5@(>@&AA.97m 7"-&Mj .55: &D 5a^ |z{{na { &D | 7d.7D&@,.AA&7D %97" >A I&" 5&A 6)6&A "96>5&K&A @(>@&AA.+Am3*!|na5d.7I&@A&^5d.7D&@"D.97&7D@&5&A>@9D(.7&AA&D>}zzm-.AD97&"(DL5j D@7A+(@A&n &AD AA9".(&  5d"D.ID.97 D@7A"@.>D.977&55& m 7"-&Mj.55: &D 5a^ |z{{na &A >@9D(.7&A ^ D9ED "966& 5&A >@9D(.7&A 7.5A^ >>@.AA&7D D@'A >@9"-&A +97"D.977&55&6&7Da&>&7%7D^{&D|A97D.7%E.DA&7@(>97A&E X6.AA&6!5&7D  |ƒ I9.@%&A69%&A%d"D.979>>9A(A%7A"&"97D&KD&a&A%&EK>@9D(.7&A.7D&@,.AA&7DI&" 5&A >@9D(.7&A 6%A &D 6%} &7 >@D."E5.&@a { .7D&@,.D 6%} &D >}zz &D >>@0D "966&E7"D.ID&E@%&5D@7A"@.>D.97a&67.'@&9>>9A(&^|.7D&@,.DI&"D&D @(>@.6&5d&K>@&AA.97%&A,'7&Am9AD.,9^&>>^L59@^s @955^|zz}na7A"&"97D&KD&5&A >@9D(.7&A A A&6!5&7D 39E&@ %&A @;5&A %.++(@&7DAa &DD& %E5.D( &AD "97D@9I&@A(& "@ 5 %(>5(D.97%&{%7A%&A"&55E5&AE .7%E.D&E X.7%E.DE7&E,6&7DD.97%& 5d"D.I.D( %E >@969D&E@ %& 5 j"%-(@.7&^ "& ?E. AE>>9A& ?E& { &AD E7 @(>@&AA&E@ D@7A"@.>D.977&5 m-.@4.-@^ .D9-^ s .LM979^ |zzna & >5EA^ { &D | A97D ">!5&A%d.7D&@,.@I&">}zzmI7 @E7AI&7&D5a^|zz€na  7 >5EA %& @(>@.6&@ 5&A "%-(@.7&A (>.D-(5.5&A^  5&A +"D&E@A ^ A97D (,5&6&7D ">!5&A%&@(>@.6&@%&A6@?E&E@A%&5>95@.D((>.D-(5.5&A"966&@E6!A}^.7A.?E& %&A "96>9A7DA %&A %&A69A96&A m&A69>54.7&^ >549>-.5.7& }n^ %&A 397"D.97A A&@@(&A m""5E%.7&^ "5E%.7&^ }n^ %&A 397"D.97A "966E7."7D&A m977&K.7& |€i}{n &D %d.7%E.@& 5d&K>@&AA.97%&6@?E&E@A6(A&7"-L6D&EK"966&5I.6&7D.7&&D5j"%-(@.7&m3*! {na 

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

 }z {a}jJ.AD&D5&AED@&A+"D&E@A%&D@7A"@.>D.97%&5   J.AD &AD E7 +"D&E@ %& D@7A"@.>D.97  %96.7& !   m>9E@ !A."i-&5.Kj599>j-&5.Kn ?E. 5E. >&@6&D %& A& +.K&@  5d^ &D  %96.7& gDJ.AD !9Kh A.DE( &7 [email protected]5 ?E. 5E. >&@6&D %& +.K&@ %&A @(,E5D&E@A "9jD@7A"@.>D.977&5Aa J.AD &AD ">!5& %& +.K&@ %&A "96>5&K&A%d"D.ID.97&D%&@(>@&AA.97%&5D@7A"@.>D.97"966&>}zz9EEm3*!|n >9E@"D.I&@mj"%-(@.7&n9E@(>@.6&@mj"%-(@.7&n5&A6@?E&E@A%&5a&AED@&A +"D&E@A %& 5  97D (D( .%&7D.+.(A >@.7".>5&6&7D AE@ 5&E@A ">".D(A  @(>@.6&@ %& 67.'@&%.@&"D&9E.7%.@&"D&5j"%-(@.7&a   6[email protected]( %&A +"D&E@A %& D@7A"@.>D.97 .%&7D.+.(A %7A 5  "97D@;5&7D &D +[email protected]&7D 5& >-(79DL>& 6(A&7"-L6D&EKa &A +"D&E@A A97D .7%E.DA 59@A %& 5 a (769.7A^"&@D.7A+"D&E@A%&D@7A"@.>D.97>>@.AA&7D@(>@.6(AE"9E@A%&5&D A&6!5&7D A>(".+.?E&A EK "&55E5&A (>.D-(5.5&Aa &A +"D&E@A }^  &D  | A97D @(>@.6(AEK"9E@A%&5&DA97D@(j&K>@.6(AE"9E@A%E>@9"&AAEA.7I&@A&^5a m-4@!@D.&D5a^|z{|_.&>5L&D5a^|z{|naE"9E@A%&5^.5L%97"E7&!57"&%& +"D&E@A%&D@7A"@.>D.97"@"D(@.A(&>@E7&!.AA&%d&K>@&AA.97%&A+"D&E@A+[email protected]7D5& >-(79DL>& (>.D-(5.5 &D E7& E,6&7DD.97 %& 5d&K>@&AA.97 %&A +"D&E@A +[email protected]7D 5& >-(79DL>&6(A&7"-L6D&EKa 

{a~j(,E5D.97"@9.A(&%&A+"D&E@A%&D@7A"@.>D.97  &A+"D&E@A%&D@7A"@.>D.97A97D">!5&A%&A&@(,E5&@&7D@&&EKa7.5{>>@.D "966&5&+"D&E@&K>@.6(5&>5EA>@("9"&6&7D%7A5D@7A.D.97a7.5|^&!A&DJ.ADA97D &K>@.6(A >5EA D@%.I&6&7D &D >9E@@.&7D 39E&@ E7 @;5& %7A 5& 6.7D.&7 %E >-(79DL>& 6(A&7"-L6D&EKa7.5{&AD">!5&%d.7%E.@&5d&K>@&AA.97%&7.5|&D%&A>@9D(.7&AA m-&7, s 7,^ |z{}na & "97D@;5& %& &!| >@ 7.5{ &AD @(5.A( >@ 5 @(,E5D.97 %dE7 D@7A"@.D7D.A&7Am&5D@7&D5a^|zz‚na@(,E5D.97"@9.A(&%&Aj&ADEAA.6(%.(& >@5@(,E5D.97%&6."@9mI9.@>@D.&AE.I7D&na 



 }{   31"& %! &   &A6."@9Am6.nA97D%&AA797j"9%7DA%&||7E"5(9D.%&A,.AA7D"966& %&A@(,E5D&E@A>9ADjD@7A"@.>D.977&5A%&5d&K>@&AA.97%&A,'7&Aa&,(796&-E6.7"9%& >9E@ &7I.@97 {zzz 6.a 5EA %& €z† %&A ,'7&A -E6.7A >&EI&7D )D@& %&A ".!5&A %&A 6."@9Am@.&%67^@-^E@,&^s@D&5^|zzƒna-&M5&A7.6EK^5&A6.AA&+.K&7DAE@ 5 >@D.& }d 797 D@%E.D& %&A  6&AA,&@Aa d>>@.&6&7D &7D@& 5& 6. &D 5d6 &AD >@D.&5^A&E5&E7&A(?E&7"&gA&&%h%&€‚7E"5(9D.%&A&AD>@+.D&6&7D"96>5(6&7D.@&a & 6."@9^ I&" 5&A >@9D(.7&A %& 5 +6.55& @,97ED&^ m n I97D %(,@%&@ 5d6 m"D.I.D(@.!97E"5(A& n&D.7-.!&@AD@%E"D.97a   &A6."@9>&EI&7D)D@&>@9%E.DA>@D.@%dE7,'7&9E>@D.@%dE7&A(?E&7"& .7D@97.?E&(>.AA(&a&A,'7&A%&6."@9AA97DD@7A"@.DA>@5>95L6(@A& m&&&D5a^ |zz~na &A D@7A"@.DA A97D ">(A &D >95Lj%(7L5(Aa &A D@7A"@.DA >9AA'%&7D E7& AD@E"DE@& D.,&j!9E"5& %d&7I.@97 ‚z 7E"5(9D.%&A &D A97D >>&55(A 6."@9 >@.6.@& m>@.j6.Ana  AD@E"DE@& %9E!5& [email protected] %c &AD @&"977E& >@  ‚ mA-n ?E. A& 5.&  @9A-^ E7& &7ML6&?E.I"5.I&@5&>@("E@A&E@>9E@+9@6&@5&@&j6.m @&,9@L&D5a^|zz€na&DD& (D>&&AD7E"5(.@&a&>@&j6.&AD&K>9@D(%7A5&"LD9>5A6&&D&AD"5.I(>@."&@>9E@ +9@6&@5&%E>5&K&6DE@&%&||7E"5(9D.%&AmE7%s-5!&@,^|zz€na  

|a{j6."@9A&D"7"&@a   &A 6."@9A >&EI&7D )D@& &K>@.6(A A>(".+.?E&6&7D %7A "&@D.7A D.AAEA 9E %& 67.'@& E!.?E.D.@&a 7A 5&A DE6&E@A^ .5  (D( 9!A&@I( E7& %.6.7ED.97 ,59!5& %& 5d&K>@&AA.97 %&A 6."@9A mE &D 5a^ |zzna Ej%&5 %& "&DD& 9!A&@ID.97^ "&@D.7A 6."@9AA97DAE@j&K>@.6(A%7A5&ADE6&E@A"966&6.j|{&D6 j{%7A5&A"7"&@A %E A&.7 &D %E >9E697 m&7"-&I s IM9.&^ |z{}na d&K>@&AA.97 %&A 6."@9A "-7,& %97" >@9+97%(6&7D 59@A %& 5 >@9,@&AA.97 DE69@5&a d5D(@D.97 %& 5d&K>@&AA.97 %&A 6."@9A Ad&K>@.6&  >5EA.&E@A 7.I&EKa .7A.` {j E 7.I&E "-@969A96.?E&^ 5&A ,'7&A "9%7D 5&A 6."@9A >&EI&7D A& D@9EI&@ AE@ %&A @(,.97A ,(796.?E&A 6>5.+.(&A^ D@7A59?E(&A 9E %(5(D(&A m&K&6>5& %& 6.j{i{€j{ m5.7 &D 5a^ |zz|nna |j E 7.I&E  }| (>.,(7(D.?E&^ %& 5 6)6& 67.'@& ?E& 5&A ED@&A ,'7&A^ "&@D.7A ,'7&A "9%7D 5&A 6."@9A A97D AA9".(A  %&A .59DA >  %97D 5 @(,E5D.97 %(>&7% %& 5 6(D-L5D.97 %& 5dmL49^|zzna}jE7.I&E%&5D@7A"@.>D.97a&796!@&EK6."@9AA97D@(,E5(A >@ %&A +"D&E@A %& D@7A"@.>D.97 39E7D E7 @;5& AE>>@&AA&E@ %& DE6&E@ 9E E7 @;5& %d97"9,'7&"966&}^9Em @95^9&%.,&^s.5.>9J."M^|z{zna~jE7.I&E%&5 !.9,(7'A& %&A 6."@9A 9F 5&A >@9D(.7&A "5(A %& 5 6"-.7&@.& %& 5 !.9,(7'A& %&A 6."@9A"966&@9A-&D."&@A97D%(@(,E5(&A%7A%&ADE6&E@Am E6@^E^&@"&@^ 95E!^s "4A^|zzna  7 >5EA %& 5&E@ @;5& +97"D.977&5 %7A 5  mI9.@ A9EAj>@D.& AE.I7D&n^ 5&A 6."@9A A97D .6>5.?E(A %7A >5EA.&E@A ED@&A "@"D(@.AD.?E&A %E "7"&@ "966&  5 >@95.+(@D.97m&K`6.j{€ma.E&D5a^|zz‚nn^5A(7&A"&7"&m&K`6.j}~mMJ^AE"-.L^ ME6.L^s4,6^|zznn^5d7,.9,(7'A&m&K`6.j{zm64E"-.&D5a^|z{znn9E5&A %966,&A5dm&K`6.j€zm.9^.7^E9^E9^s7,^|z{{nna 

|a|j6."@9A&D   &A6."@9A>&EI&7D)D@&&K>@.6(A%&67.'@&AD.AAEAA>(".+.?E&A&D"97D@;5(A>@ 5d&K>@&AA.97 %& +"D&E@A %& D@7A"@.>D.97 D.AAEA A>(".+.?E&Aa &A 6."@9A >&EI&7D )D@&A &K>@.6(A%&67.'@&!&@@7D&59@A%&5"9@@E>D.97%&A>@9,@66&AD@7A"@.>D.977&5AE "9E@A%&5>@9,@&AA.97DE69@5&m&K`5D@7A.D.97(>.D-(5.9j6(A&7"-L6D&EA&naE"9E@A%& 5^979!A&@I&E7&%.6.7ED.97%&A6."@9A&K>@.6(A%7A5&AD.AAEA(>.D-(5.EK&D E7& E,6&7DD.97 %&A 6."@9A &K>@.6(A %7A 5&A D.AAEA 6(A&7"-L6D&EKa &5 "9@@&A>97% E "-7,&6&7D %d&K>@&AA.97 %& +"D&E@A %& D@7A"@.>D.97 E "9E@A %& "& >@9"&AAEAa&A6."@9A.6>5.?E(A%7A5>&EI&7D)D@&"5AA(A&7%&EK"D(,9@.&A`{j &A6."@9A".!57D5&A+"D&E@A%&D@7A"@.>D.97%&5&D|j5&A6."@9A".!57D5&A &++&"D&E@A %& 5 a &A 6."@9A ".!5&7D %&A "96>9A7DA %E "LD9A?E&5&DD&^ 9E %&A 6@?E&E@A(>.D-(5.EK9E6(A&7"-L6D&EKm3*!}na   

 }}  '@#%4*3'-, !2$3!412 !31,21'/3'-,   >@&6.'@&+6.55&%&6.A.6>5.?E(&%7A5&AD5+6.55&6.j|zza55&&AD &K>@.6(&>@D.@%&%&EK"5EAD&@AAE@5&"-@969A96&{m6.j|zz^6.j|zz!&D6.~|ƒn&D AE@ 5& "-@969A96& {| m6.j|zz" &D 6.j{~{na d&K>@&AA.97 %& "&A 6.A &AD AA9".(& E >-(79DL>&(>.D-(5.5&D%.6.7E&%97"59@A%&5m@4^ E@^&7,L&5^s&D&@^|zz‚na 5d.7I&@A&5AE@j&K>@&AA.97%&6.j|zz77E5&5&>-(79DL>&6(A&7"-L6D&EKa&DD&+6.55& %&6.jA".!5&7D5&A+"D&E@A{&D|?E.A97D&EK6)6&A">!5&A%&5&A@(>@.6&@ m @&,9@L&D5a^|zz‚na7&++&D^{&D|A&+.K&7DAE@5A(?E&7"&j!9K%&A>@969D&E@A %&A "5EAD&@A %& 5 +6.55& 6.j|zza &A +"D&E@A A97D EAA. ".!5(A >@ %dED@&A 6."@9A m3*!}na  6)6&!9E"5&%&@(D@9"97D@;5&&AD9!A&@I(&>9E@7.5{&D6.j}~`6.j}~".!5& 7.5{ ?E. @(>@.6& A97 &K>@&AA.97 &7 A& +.K7D AE@ 5 A(?E&7"& j!9K %& A97 >@969D&E@ m.&6&7A&D5a^|z{{na7.5{&AD(,5&6&7D".!5(>@6.|ƒ!&D6.j}za  9EA 5&A 6."@9A ".!57D 5&A +"D&E@A 7.5 9E  I9.&7D 5&E@A &K>@&AA.97A %.6.7E(&AE"9E@A%&5a&DD&%.6.7ED.97&ADEAA.@&5&I7D&%7A5&A"7"&@A>E.A?E& "&A6."@9A39E&7DE7@;5&>>@&7D%&AE>>@&AA&E@%&DE6&E@a@&K&6>5&6.j|zz&AD A9EAj&K>@.6( %7A 5&A "7"&@A %& 5 I&AA.&^ ,AD@.?E&^ "@".796& %& 5 D)D& &D %E "9E^ %& 5d9I.@&ma E&D5a^|zzƒ_-.79M4.&D5a^|z{z_.45E7%&D5a^|z{{_.%@&D5a^|z{{n&D 6.j}~%7A5&A"7"&@A%EA&.7^%E>9E697^%E"9597^%E>7"@(A^&D%&5d9I.@&m9@7&L&D 5a^|z{z_9,D&D5a^|z{{_a7,&D5a^|z{{na 

 }~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

   ''C#%4*3'-, !2!$$!3!412   &A 6&6!@&A %dE7& 6)6& +6.55& %& 6."@9 >@D,&7D %&A ".!5&A "966E7&Aa 7%.I.%E&55&6&7D^E76."@9>&ED".!5&@%&796!@&EK6%dE7&I9.&%&A.,75.AD.97^ "&?E.@&7+9@"&5&E@A&++&DAmE!@67L6s5&559"-^|z{{na7A"&"97D&KD&^6.j|zz?E. @(,E5& 5&A +"D&E@A A^ @(,E5& EAA. 5d&K>@&AA.97 %dED@&A ,'7&A &7 >@D."E5.&@ "&EK .6>5.?E(A%7A5+9@6D.97%&56D@."&&KD@"&55E5.@&m"-5.&4&567&D5a^|z{{na@6.

 } "&A 6."@9A %97D 5c&K>@&AA.97 &AD %.6.7E(& E "9E@A %& 5 ^ "&@D.7A ".!5&7D %&A 6@?E&E@A 6(A&7"-L6D&EK "966& 5 j"%-(@.7& m6.j{ƒ~n 9E 5 I.6&7D.7& m6.j{}‚n m$'%41!nadED@&A6."@9AA97D%("@.DA"966&AE@j&K>@.6(AE"9E@A%&5ad&AD5& "A>9E@6.jƒ?E.".!5&%.@&"D&6&7D5j"%-(@.7&&D?E.+[email protected]&56.,@D.97&D5d.7IA.97 m&D5a^|z{zna 

 

'%41!P?+'1-2'+/*'04#2 ,2*@ A/1"2 +-4'**!>!3H*KIJL411!,3-/','-,',!**'-*-%8             }€ 97"5EA.97   @7A.D.97 >.D-(5.9j(A&7"-L6D&EA& &AD AA9".(&  5 >@9,@&AA.97 DE69@5& mI,7&@^ 6%^ s -.&@L^ {ƒƒ_ -.&@L^ |zz|na  @(>@&AA.97 %& 5 j"%-(@.7& >@ 5& +"D&E@%&D@7A"@.>D.977.5{&AD"97A.%(@(&"966&E7%9,6&%&>E.AA979!A&@ID.97E %(!ED%&A77(&A|zzza&>E.A"&DD&6.A&&7(I.%&7"&^5%("9EI&@D&%dED@&A+"D&E@A%& D@7A"@.>D.97 .7%E.A7D 5   @&7+9@"( 5d.%(& ?E& "& >@9"&AAEA &AD "97D@;5( >@ E7 >@9,@66& D@7A"@.>D.977&5a d&7A&6!5& %& "&A +"D&E@A %& D@7A"@.>D.97 @(,E5& D9EA 5&A >-(79DL>&A9!A&@I(A59@A%&5`>&@D&%&A397"D.97A(>.D-(5.5&A^5d"?E.A.D.97%&5 69D.5.D(6.AEAA.5@(A.AD7"&5A(7&A"&7"&^5d>9>D9A&^&D5@(A.AD7"&EK@%.9j&D "-.6.9D-(@>.&Am-.&@L&D5a^|zzƒna  7 >5EA %& "& >@9,@66& D@7A"@.>D.977&5^ %& 796!@&EA&A (I.%&7"&A @&5.&7D 5&A 6."@9AA5@(,E5D.97%&5a966&%("@.D%7A"&DD&>@D.&^5&A6."@9AA97D 5+9.A%&A@(,E5D&E@A%&A+"D&E@A%&D@7A"@.>D.97^%&A@(,E5D&E@A%&5&DA97DEAA. ">!5&A %& @(,E5&@ 5&A &++&"D&E@A %& 5  "966& 5&A D@7A"@.DA "9%7DA >9E@ %&A >@9D(.7&A .6>5.?E(&A %7A %&A 397"D.97A "&55E5.@&A^ 5 6.,@D.97^ 5& "LD9A?E&5&DD& %d"D.7& &D"ba d.6>"D %&A 6."@9A AE@ 5 ^ &D 5& +.D ?Ed.5A ".!5&7D E7 ,@7% 796!@& %& D@7A"@.DA^ +.D %&A 6."@9A %&A @(,E5D&E@A EAA. .6>9@D7DA ?E& 5&A +"D&E@A %& D@7A"@.>D.97a  E"9E@A%&"&A%&@7.'@&A77(&A^%&796!@&EA&A69%.+."D.97A%d(>.AA,&97D(D( 9!A&@I(&A E "9E@A %& 5 >@9,@&AA.97 DE69@5&a 7 >@D."E5.&@^ .5  (D( .%&7D.+.( E7 >@9,@66& %d(>.AA,& 5D&@7D.+ E "9E@A %& 5  "97D@;5( >@ %.++(@&7DA +"D&E@A %d(>.AA,&"966&&D!+9K|a7A5>@D.&AE.I7D&^3&"966&7"&@.>@%(D.55&@5&A 6("7.A6&A %& @(,E5D.97 %& 5 @("D.97 %d(>.AA,& &D %& 5d(>.AA,& 5D&@7D.+a 7A E7 A&"97%D&6>A^3&%("@.@.5&A(I.%&7"&A%&A69%.+."D.97A%d(>.AA,&5D&@7D.+E"9E@A%&5 >@9,@&AA.97DE69@5&&D>@D."E5.'@&6&7DE"9E@A%&5a        } .0$ ;,$//"'0.)0$!0,.*".//$*)01(*.'   &A(?E&7$,&%E,(796&-E6.7E%(!ED%&A77(&A|zzz>[email protected]%d&AD.6&@?E&"&5E.j". &AD"96>9A(%d&7I.@97|zzzz,'7&Aa&"-.++@&&AD"&>&7%7D.7+(@.&E@EK>@(I.A.97A%& 5d(>9?E&?E.D!5.&7DAE@}zzzzz,'7&Aa&+.D^5&,(796&-E6.7"97D.&7DE7796!@&%& ,'7&AAA&MA.6.5.@&%dED@&A&A>'"&A"966&5[email protected]^5&>9.AA97M'!@&9E5%@9A9>-.5&a & ,(796& -E6.7 "96>@&7% &7I.@97 ƒ† %& ,'7&A %.A"97D.7EA^ "d&ADjj%.@& ?E& 5d.7+9@6D.97"9%7D5&A>@9D(.7&Am&AA&7D.&55&6&7D"9%(&A>@5&A&K97An&ADA(>@(&>@%& 5@,&A>[email protected],(796.?E&A,(7(@5&6&7D797j"9%7D&A>>&5(&A.7D@97Aa&A&K97AA97D%& "9E@D&AA(?E&7"&A%97D5D.55&&AD"96>@.A&&7D@&z&D|z7E"5(9D.%&A&7I.@97m69L&77& {~na&A.7D@97A^A(?E&7"&A.7D@,(7.?E&A,(7(@5&6&7D797"9%7D&A^97D(D(.%&7D.+.(A 5+.7%&A77(&Azm&@,&D^99@&^s-@>^{ƒna&E@D.55&&AD&769L&77&{z+9.A>5EA ,@7%& ?E& "&55& %dE7 &K97^ "& ?E. A.,7.+.& ?Ed.5A "97AD.DE&7D >5EA %& ƒz† %& 5dE7.D( %& D@7A"@.>D.97a &A 6 A97D D@7A"@.DA >@ 5 >95L6(@A& a d(>.AA,& &AD E7 >@9"&AAEA 7E"5(.@&E"9E@A%E?E&55&A.7D@97AA97D&K".A(A&D6.A!9ED!9EDa&>@9"&AAEA.7D&@I.&7D "9j9E>9ADjD@7A"@.>D.977&55&6&7Da  & >@&6.&@ 9!3&"D.+ %& "&DD& A&"97%& >@D.& &AD %& %(+.7.@ 5d(>.AA,& &D 5d(>.AA,& 5D&@7D.+&D%&>@(".A&@5&A6("7.A6&A695("E5.@&A%&@(,E5D.97%&5d(>.AA,&5D&@7D.+a & %&EK.'6& 9!3&"D.+ &AD %& 6&DD@& &7 (I.%&7"& 5& @;5& %& 5d(>.AA,& 5D&@7D.+ 59@A %& 5 >@9,@&AA.97DE69@5&&D79D66&7D59@A%&5a 

? . 0$*)= ,$//"   &AA.D&A%d(>.AA,&A97D>&E"97A&@I(A&DA97D"96>9A(A%dE7A.D&%977&E@^E7A.D& ""&>D&E@^E7>9.7D%&!@7"-&6&7D&DE7&A(?E&7"&@."-&&7>[email protected].%.7&m$'%41!‚na @("D.97%d(>.AA,&&AD9>(@(&>@5&A>5."&9A96&^E7"96>5&K&6"@9695("E5.@&%&>5EA %&|zz>@9D(.7&Am &,&5&&D5a^|z{|na&A>&D.D&A@.!97E"5(9>@9D(.7&A7E"5(.@&AmA7n A97D%&A"96>9A7D&A%EA>5."&9A96&a@("D.97%d(>.AA,&+.D>>&5A7A`{^|^ ~^&D€a{AdAA9".&EA.D&d%d(>.AA,&&D|AdAA9".&E>9.7D%&!@7"-&6&7D^5& D9ED"97AD.DE&5&>@(jA>5."&9A96&m"96>5&K&na~&D€+97D5&>97D&7D@&{&D|_ >&@6&D 5 5..A97 I&" 5& A.D& }d %d(>.AA,& &D 5& D9ED "97AD.DE& 5& A>5."&9A96& 6DE@&  }‚ m"96>5&K&na%.AA9".D.97%&5d%&~&D€.7%E.D%&A"-7,&6&7DA%&"97+9@6D.97^ "& ?E. @&7% 5& "96>5&K& "D.+ &D >&@6&D %&A @("D.97A %& D@7A&AD(@.+."D.97a  >@&6.'@& @("D.975.&E&7D@&5&|d %E>9.7D%&!@7"-&6&7D&D5397"D.97%EA.D&d"&?E.+9@6& E7&!9E"5&%d>>&5(&5&5@.DaA&"97%&@("D.975.&E&7D@&5&A397"D.97Ad&D}da& 796!@&EA&A>@9D(.7&A.7D&@I.&77&7D%7A"&>@9"&AAEAa7%.AD.7,E&79D66&7D-E.D -(5."A&A?E."D5LA&7D5&A@(@@7,&6&7DAj&Dj@9D(.7&`E!|i€^@>^ @>|‚^@@|^@>|^@>{€^@>||^&D@>~m9@%.7^ -7^s&,,A^|z{|na 

 

'%41!Q?'3! A#/'22%!-,2!,242!31#3'-, A#/'22%!@

d>@'Aa 9@7!5.-DDs5aDE@&&I.&J95&"E5@&55.959,La 

 }ƒ ? )$/(. "1'0$*)'= ,$//"'0.)0$! 

  21('! #%&'(%&/#&&   &AA.D&A%d(>.AA,&A97D%([email protected](A>@%&AA(?E&7"&A%d(>.AA,&?E.>&EI&7D)D@& >5EA9E69.7A(59.,7(&A%&5A(?E&7"&"97A&7AEA?E.&AD5AE.I7D&` i >9E@5

5..A97 &K97i.7D@97^  >9E@ 5 !90D& %& !@7"-&6&7D &D 7 i  >9E@ 5 5..A97 .7D@97i&K97a7A.D&%d(>.AA,&g+9@Dh&AD>5EA>@9"-&%&5A(?E&7"&"97A&7AEA?EdE7A.D& %.Dg+.!5&ha+@(?E&7"&%dED.5.AD.97%dE7A.D&%d(>.AA,&&AD%([email protected](&&7>@D.&>@5 +9@"&[email protected]'?E&%&"-?E&A.D&a&>&7%7D5&AA(?E&7"&A%d(>.AA,&A97DD@'A%(,(7(@(&A &D7&AE++.A&7D>A%(+.7.@E7&K97a&A+"D&E@A%d(>.AA,&A&+.K&7DAE@5&A(5(6&7DA".Aj @(,E5D&E@AmA(?E&7"&A&K97.?E&A9E.7D@97.?E&An&DA97D">!5&A%&+[email protected]&@9E%d.7-.!&@ 5&@&"@ED&6&7D%EA>5."&9A96&&D&7>@D."E5.&@5&AA7Am$'%41!ƒna 5&K.AD&%&EK+6.55&A %& +"D&E@A %d(>.AA,& ?E. A& +.K&7D AE@ "&A 69D.+A` 5&A >@9D(.7&A  m&@i@, @."-n &D 5&A >@9D(.7&A -7 m-&D&@9,&79EA 7E"5&@ @.!97E"5&9>@9D&.7n m3*! E>{na &A >@9D(.7&A  +[email protected]&7D5@&"977.AA7"&%EA.D&%d(>.AA,&&7+".5.D7D5&@&"@ED&6&7D%&{&D|a 5d.7I&@A&^5&A>@9D(.7&A-797D,(7(@5&6&7DE7@;5&@(>@&AA&E@&76A?E7D5&AA.D&A %d(>.AA,&mEA"-&D5a^|z{}na&A%977(&A5&A>5EA@("&7D&A%&55.DD(@DE@&697D@&7D?E& 5&A >@9D(.7&A  >&EI&7D (,5&6&7D @(>@.6&@ 5d(>.AA,& 59@A ?E& 5&A >@9D(.7&A -7A >&EI&7D 5& +[email protected]&@ &D ,.@ %& "97"&@D >9E@ @(,E5&@ "& >@9"&AAEAa &E@A @(,E5D.97A %(>&7%&7D(,5&6&7D%&5>9A.D.97%&5&E@A@&"@ED&6&7DAAE@5&A(5(6&7DA".Aj@(,E5D&E@A >@@>>9@D5d&K97m E&5,&D5a^|z{|_7%.D&D5a^|z{}_.DD&7s5&^|z{{nad(>.AA,& 5D&@7D.+ &AD %97" @(,E5( %& 67.'@& "96!.7D9.@& &D "96>(D.D.I& >@ 5&A +"D&E@A %d(>.AA,&a  "&A%&EK+6.55&AAd39ED&7D%&A+"D&E@A%d(>.AA,&AD.AAEA9E"&55E5&AA>(".+.?E&Aa @ &K&6>5&^ 5&A >@9D(.7&A j{ &D j| A97D &K>@.6(&A %7A 5& "&@I&E m@7&55^ &5&7^ 5&^ s @49^ |zzn^ { &D | A97D &K>@.6(&A %7A 5&A "&55E5&A (>.D-(5.5&A m@M&"-^ D9^ !&D^ 9,&7&A"-^ s @AD&7A^ |zzƒna 7 +97"D.97 %& 5 >9A.D.97 %& "&A >@9D(.7&A^&55&A>&EI&7D+[email protected]&@9E.7-.!&@5d(>.AA,&a&A>@9D(.7&A"97D@;5&7D%&67.'@& A>(".+.?E&%&A>@9,@66&A%d(>.AA,&D.AAEA9E"&55E5&AA>(".+.?E&Aa

 ~z  .7A.^5@(,E5D.97%&5d(>.AA,&5D&@7D.+&ADE7&"96>(D.D.97&7D@&A.D&A%d(>.AA,& &D%(>&7%%&5>@(A&7"&&D%&5>9A.D.97%&AA(?E&7"&A".Aj@(,E5D@."&A.7A.?E&%E7.I&E %d&K>@&AA.97%&A+"D&E@A%d(>.AA,&a 

 '%41!R?#04!,!'2C1#%4*31'! !*A#/'22%!!3$3!412 A#/'22%!@ A/1"2@ -1,*'&33H*@341!!5'!6-*!4*1!**'-*-%8@E7-,'/*'',% ,&,!1F> E ,31-,'/*'',%,&,!1F>E7-,'/*'',%'*!,!1F!3 E ,31-,' /*'',%'*!,!1F   @ "&EK(DE%&A5@,&("-&55&^.5&AD>9AA.!5&^&7D&77D"96>D&%&5>9A.D.97%&A (5(6&7DA @(,E5D&E@A %& 5d(>.AA,& &D %& 5d&K>@&AA.97 %&A +"D&E@A %d(>.AA,&^ %& >@(%.@& 5d.7"5EA.97 9E 5d&K"5EA.97 %dE7 &K97 %7A E7 D.AAE %977( m@A- &D 5a^ |z{zna & "9%& %& 5d(>.AA,&(D(ED.5.A(>9E@I5.%&@%&A>@9,@66&A%d(>.AA,&A>(".+.?E&A"&@D.7AD.AAEA m@A-&D5a^|z{zna&>&7%7D^E7&%&A5.6.D&A%&"&"9%&%&5d(>.AA,&&AD?Ed.57&>@&7% >A &7 "96>D& 5&A ED@&A 7.I&EK %& @(,E5D.97 %& 5d(>.AA,& "966& 5d.7+5E&7"& %& 5 D@7A"@.>D.979E%&5"-@96D.7&a



  31 ( '% &%#'! '%! '    &7%7D 597,D&6>A 5 D@7A"@.>D.97 &D 5d(>.AA,& 97D (D( "97A.%(@(A "966& %&A >@9"&AAEA.7%(>&7%7DA6.A.5>>@0D55E6.'@&%&A(DE%&A%&A(?E&7$,&-ED%(!.D m -9%9@ &D 5a^ |z{{n ?E& 5d(>.AA,& &AD @(5.A( "9jD@7A"@.>D.977&55&6&7D &D ?E& "&A %&EK 6("7.A6&A A97D "9E>5(Aa & +.D^ 5 695("E5& %d >@(j6&AA,&@ m>@(j6n &AD E7&  ~{ 695("E5& I.@DE&55& "@ 5&A &K97A A97D (>.AA(A E +E@ &D  6&AE@& %& 5&E@ D@7A"@.>D.97a . 5d(>.AA,& %d&K97A "97AD.DED.+A >>@0D "9jD@7A"@.>D.977&5^ 5d(>.AA,& 5D&@7D.+ >&ED %7A "&@D.7A "A .7D&@I&7.@ >9ADjD@7A"@.>D.977&55&6&7D m@,A &D 5a^ |z{{na & >5EA^ %&A +"D&E@A%&D@7A"@.>D.97^%&A"9j@(,E5D&E@AD@7A"@.>D.977&5AmE!9&E+^DA"-(^ED&@D@&^ E"-@%D^ s d55&L^ |zzn &D %&A >@9D(.7&A %E @&69%&5,& &D %& 5 AD@E"DE@& %& 5 "-@96D.7& mE"9^ 559^ "-9@^ 9@7!5.-DD^ s .AD&5.^ |z{z_ E"9^ 7^ &D 5a^ |z{zn A97D (,5&6&7D ">!5&A %d.7+5E&7"&@ 5&A "-9.K %d(>.AA,&a &EK 6("7.A6&A "96>5(6&7D.@&A A97D@&A>97A!5&A%&"&"9E>5,&&7D@&D@7A"@.>D.97&D(>.AA,&m+.,E@&{zna@&6.'@&6&7D^ 5".7(D.?E&%d(597,D.97%&5>95L6(@A& .7+5E&7"&%.@&"D&6&7D5d(>.AA,&a7&++&DE7& %.6.7ED.97 %& 5 I.D&AA& %d(597,D.97 5.AA& %I7D,& %& D&6>A  5d.7D@97^ D9ED 3EAD& AL7D-(D.A(>9E@)D@&@&"977E>@5&A>5."&9A96&m&K`K97}}%&5+.!@97&"D.7&m 9@7!5.-DD &D5a^{ƒƒn&D%97"+[email protected]&@A97.7"5EA.97a5d.7I&@A&^E7&I.D&AA&%d(597,D.97>5EA(5&I(& 5.AA&69.7A%&D&6>A%&A@(,E5D&E@Am>9A.D.+A9E7(,D.+An>9E@A&+.K&@AE@5&D@7A"@.D&7 "9E@A %& AL7D-'A& &D %97" %& +[email protected]&@ 5d.7"5EA.97 9E 5d&K"5EA.97 %d&K97 m 9@7!5.-DD &D 5a^ |z{}na &EK.'6&6&7D^ 5& @&"@ED&6&7D %&A +"D&E@A %d(>.AA,& >@ 5 6"-.7&@.& D@7A"@.>D.977&55&&AD E7 >@9"&AAEA "5( %7A 5&A "-9.K %d(>.AA,&a 7 &++&D %& 796!@&EK +"D&E@A%d(>.AA,&A97D@&"@ED(A>@5&%96.7&m%96.7&"@[email protected]5&n%&5>95 mI.%^9L7&^.55-9EA&^s75&L^|z{{n&D%&A@(,E5D&E@A%&5D@7A"@.>D.97"966& |}A&5.&7D%&A+"D&E@A%d(>.AA,&m&K`-7m E7,&D5a^|z{|nna   & 6("7.A6& %& "9E>5,& D@7A"@.>D.97i(>.AA,& &AD >@D."E5.'@&6&7D .6>9@D7D %7A 5& "%@& %& 5 >@9,@&AA.97 DE69@5& "@ %& 796!@&EK D@.D&6&7DA 7D.j"7"(@&EK ED.5.A(A "9E@66&7D A97D %&A .7-.!.D&E@A %& 5d(597,D.97 %& 5 >95L6(@A& m&K` 6>D-9D-(".7&^9K9@E!.".7&nad-L>9D-'A&&AD?E&"&AD@.D&6&7DA>9E@@.&7D69%.+.&@5& @(>&@D9.@& %&A &K97A &K>@.6(A^ "& ?E. >&ED &7D@.7&@ 5 >@9%E"D.97 %d.A9+9@6&A I&" %&A >@9>@.(D(A%.++(@&7D&A^>9D&7D.&55&6&7D%&6EI.A>@979AD."9E.6>5.?E(A%7A5@(A.AD7"& EKD@.D&6&7DAa 

 ~| Pol II

exon exon



                

  

'%41!JI? ,$*4!,! !*31,21'/3'-,241*A#/'22%!*3!1,3'$@ - '$'# A/1"2@ -1,*'&33H*@341!!5'!6-*!4*1!**'-*-%8@



 .I&@A&A (DE%&A 97D EAA. >9.7D( 5& @;5& %& 5 "-@96D.7& &D %&A 69%.+."D.97A (>.,(7(D.?E&A %7A 5 @(,E5D.97 %& 5d(>.AA,& 5D&@7D.+a 9ED %d!9@%^  5&A 69%.+."D.97A %d-.AD97&A>&EI&7D.7+5E&7"&@5I.D&AA&%d(597,D.97%&5>95L6(@A&&D@(,E5&@5d(>.AA,& 5D&@7D.+ %& 5 6)6& 67.'@& ?E& %("@.D& %7A 5& >@,@>-& >@("(%&7Dm&5.A"-^ 559^ 9@7!5.-DD^s"-9@^|zzƒna&A69%.+."D.97A%d-.AD97&7&A97D>A%.AD@.!E(&A5(D9.@&6&7D 5&597,%&A,'7&Aad&AD5&"A%&569%.+."D.97 } ƒ6&|?E.&AD&7@."-.&E7.I&E%&A &K97Aa&A>@9D(.7&A%&5"-@96D.7&m&K` {9EA.>{i5&%,+>|nA97D">!5&A%&A& +.K&@ A>(".+.?E&6&7D AE@ "&DD& 69%.+."D.97 &D %& @&"@ED&@ %&A +"D&E@A %d(>.AA,& m@&A>&"D.I&6&7D&D{n>9E@.7+5E&7"&@5%(".A.97%c(>.AA,&m$'%41!{{nmE"9^7^ &D5a^|z{z_@%&&>^ED-&@57%^5&^ @.6&A^s."469@&^|z{|na  &A7E"5(9A96&A^>>@.AA&7D&EKEAA.>9A.D.977(A797j5(D9.@&6&7D5&597,%&A ,'7&Aa 5A &7@9E5&7D E7& A(?E&7"& %d %d&7I.@97 {z >.@&A %& !A&A^ E7& D.55& D@'A A.6.5.@&  "&55& %dE7 &K97a &A 7E"5(9A96&A A97D &7@."-.A E 7.I&E %&A &K97A &D "& >9A.D.977&6&7D  (D( >@9>9A( "966& >@D.".>7D  5 %(+.7.D.97 %&A &K97A m"-J@DM^ &A-9@&@^sAD^|zzƒ_>.&A^.&5A&7^%,&DD^sE@,&^|zzƒ_.5,7&@&D5a^|zzƒna

 ~}  

              

 

'%41!JJ?7!+/*! A',$*4!,! !+- '$'3'-,2 !*&1-+3',!241*A#/'22%!*3!1,3'$@ - '$'# A/1"2@ -1,*'&33H*@341!!5'!6-*!4*1!**'-*-%8@  

41 ( &&'%('(%&&! %&.   &A+"D&E@A%d(>.AA,&.7D&@,.AA&7DI&"5d,@ "&%.++(@&7DA69D.+A>@9D(.?E&A m&K`5&A%96.7&A m-7 -96959,Ln^m@&"9,7.D.9769D.+An9E5&A69D.+A %9.,D %& M.7"na &A 69D.+A @&"977.AA&7D %&A A(?E&7"&A %d A.6>5& [email protected] %& |  {z 7E"5(9D.%&A mEJ&D&@^ !&@AD@AA^ s E^ |zz€n m!5& E> {na &>&7%7D^ 5&A 695("E5&A %d >&EI&7D A& AD@E"DE@&@ &D %9>D&@ %& 796!@&EA&A "97+9@6D.97A ,@ "&  %&A >>@.&6&7DA .7D@695("E5.@&A m$'%41! {|na  +9@6D.97 %& AD@E"DE@&A A&"97%.@&A >&ED %.@&"D&6&7D .7+5E&7"&@ 5 %(".A.97 %d(>.AA,&a 7 &++&D^ .5  (D( 697D@( ?E& 5&A AD@E"DE@&A A&"97%.@&AAD!5&AA97D&7@."-.&AE7.I&E%&A&K97A5D&@7D.+AI&"E7DEK@."-&&7  ma 7, &D 5a^ |z{{na @'A @("&66&7D^ .5  (D( 697D@( >@ 5d(?E.>& %& a  ?E& 5 +9@6D.97%&AD@E"DE@&A&"97%.@&&ADD@'A.6>9@D7D&>9E@5d(>.AA,&5D&@7D.+%&5d&K97{| %E,'7& a7&++&D^"&DD&(?E.>&.%&7D.+.(E7&@(,.97"D.ID@."&%&5d(>.AA,&A.DE(&7

 ~~ I5%&5d&K97?Ed&55&@(,E5&a>@D."E5@.D(&AD?Ed&55&&ADA.DE(&>5EA%&zz>!%EA.D&d %d(>.AA,&m9I".&D5a^|z{}na+9@6D.97%&AD@E"DE@&AA&"97%.@&A>>@0D%97""966& E7(5(6&7D@(,E5D&E@%&5%(".A.97%d(>.AA,&a  &A AD@E"DE@&A A&"97%.@&A %& 5d >&EI&7D 6A?E&@ E7 A.D& %d(>.AA,&^ 9E E7& A(?E&7"&".Aj@(,E5D@."&^"&?E.&7D@.7&E7%(+ED%&@&"@ED&6&7D%E+"D&E@%d(>.AA,&aE "97D@.@&^"&@D.7&AAD@E"DE@&AA&"97%.@&A>&@6&DD&7D5&@>>@9"-&6&7D&7D@&5d&K97&DE7& A(?E&7"& ".Aj@(,E5D@."& &D [email protected]&7D 5 @("D.97 %c(>.AA,&a &A AD@E"DE@&A A&"97%.@&A .7+5E&7"&7DEAA.5&"-9.K%&A&K97A6EDE&55&6&7D&K"5EA.+Am$'%41!{|nma7,&D5a^|z{{na &A  -(5."A&A^ A97D %&A &7ML6&A ">!5&A %& 69%.+.&@ 5&A AD@E"DE@&A A&"97%.@&A %& 5da&796!@&EA&A-(5."A&A.7D&@I.&77&7D%7A5@("D.97%d(>.AA,&&D>@D.".>&7D E@&"L"5,&%&AA7Am&DD&@^&!&@^&7&"4&^s-5^|zzƒna&D{A97D%&EK -(5."A&A%&5+6.55&%&A!9K>9E@5&A?E&55&A.5(D(697D@(?Ed&55&A@(,E5&7D 5d(>.AA,&5D&@7D.+&769%.+.7D%&AAD@E"DE@&AA&"97%.@&A%&5dmI9.@@D.&}na 

 '%41!JK?7!+/*! !2314341!2!-, '1! !*A',$*4!, ,3*!2 #'2'-,2 A#/'22%!@ EF314341!$$!3,3*!1!143!+!,3 !$3!41 A#/'22%!@EF2314341!/!1+!33,3*! 1//1-&!+!,3 !2#04!,!2'2C1#%4*31'!2@EF-+/#3'3'-,!,31!2314341!22!-, '1!2@  /3# A/1"2 ',H*'-*-%8KIJJ@  

 ~ ?,$//"'0.)0$!0,.*".//$*)01(*.'    d5D(@D.97%&5d(>.AA,&5D&@7D.+&ADE7>-(796'7&@&D@9EI(%7A%&796!@&EA&A >D-959,.&A^ 79D66&7D "&@D.7&A 65%.&A ,(7(D.?E&A &D 5& "7"&@a 5EA.&E@A (DE%&A 97D >[email protected]%d9!A&@I&@5%(@(,E5D.97%&5d(>.AA,&%7A5&A"7"&@A&D%&6&DD@&&7(I.%&7"& 5d6>5&E@ %& "&A %(@',5&6&7DAa d.6>9@D7"& %& "& >-(796'7& %7A 5 >@9,@&AA.97 DE69@5&  (D( A9E5.,7(& E D@I&@A %& %&EK A>&"DAa @&6.'@&6&7D^ %&A (DE%&A  5@,& ("-&55&^@(5.A(&A79D66&7DEA&.7%&79D@&(?E.>&^97D>[email protected]%&6&DD@&&7(I.%&7"&%&A "-7,&6&7DA63&E@A&7D&@6&%d(>.AA,&&7D@&%&AD.AAEADE69@EK&D%&AD.AAEAA.7A9E %&A5(A.97A!(7.,7&A^9E&7"96>@7D%.++(@&7DA69%'5&A%&>@9,@&AA.97DE69@5&a&>5EA %&A5D(@D.97A%d&K>@&AA.97.7A.?E&%&A6EDD.97A%&+"D&E@A%d(>.AA,&E"9E@A%&5 >@9,@&AA.97 DE69@5& 97D (,5&6&7D >E )D@& 9!A&@I(&Aa &EK.'6&6&7D 5 +97"D.97 %& "&@D.7AI@.7DA@&D@9EI(A5D(@(AE"9E@A%&5>@9,@&AA.97DE69@5&(D("@"D(@.A(&a &A(DE%&A%(697D@&7D?E&"&AI@.7DA97D%&A+97"D.97A>@99E7D.jDE69@5&a  

&$      %            & 796!@&EA&A (DE%&A  5@,& ("-&55& 97D >[email protected] %& %(+.7.@ 5d6>5&E@ %E %(@',5&6&7D %E >@9"&AAEA %d(>.AA,& 5D&@7D.+ E "9E@A %& 5 >@9,@&AA.97 DE69@5&a d(?E.>&%7A5?E&55&3d.&++&"DE(6D-'A&@(5.A(E7&(DE%&5@,&("-&55&>9E@(DE%.&@ 5&D@7A"@.>D96&%&DE6&E@A>@9I&77D%E69%'5&~{m>@(A&7D(%7A5>@&6.'@&>@D.&%& 697 .7D@9%E"D.97na d %& DE6&E@A >@9I&77D %& DE6&E@A >@.6.@&A .7%E.D&A >@ 5d.73&"D.97%&"&55E5&A€^{€‚^~z&D~{(D(75LA(5d.%&%&A>E"&A&K97A %d++L6([email protected]&A%&EK>@.7".>5&A"97"5EA.97A%&"&D@I.5.7%.?E&7D?E&%dE7&>@D.5&AD >9AA.!5& %& "@"D(@.A&@ 5d,@&AA.I.D( %& "&A DE6&E@A &7 75LA7D 5&A &K97A %.++(@&7D.&55&6&7D@(,E5(Aa&5AE,,'@&^%dED@&>@D^?E&5d&K>@&AA.97%d&K975D&@7D.+&AD AA9".(  5 ">".D( %&A DE6&E@A >@.6.@&A  +9@6&@ %&A 6(DADA&Aa & >5EA "&A &K97A %.++(@&7D.&55&6&7D@&D@9EI(A%7A%&ADE6&E@A+9@67D9E>A%&A6(DADA&A"-&M5[email protected] A97DAA9".(AE7&AE@I.&A7A6(DADA&A"-&M%&A>D.&7D&ADD&.7D&A%&"7"&@%EA&.7 mED&@D@&&D5a^|z{zna&DL>&%d(DE%&@(5.A(%7A5d(?E.>&AE@E769%'5&"&55E5.@&[email protected]  ~€ (,5&6&7D(D(6&7(%.@&"D&6&7DI&"%&A("-7D.5597A%&>D.&7D&Aad(?E.>&%&a7%@( .7A."96>@(5&D@7A"@.>D96&%&{|z("-7D.5597A%&"7"&@%EA&.7~5(A.97A!(7.,7&Aa &DD& (DE%&  .7A. >[email protected] %& %(+.7.@ ƒz &K97A %.++(@&7D.&55&6&7D &K>@.6(A &7D@& ("-7D.5597A DE69@EK &D ("-7D.5597A !(7.7A m7%@( &D 5a^ |zzƒna &A (DE%&A^  5@,& ("-&55&^%(697D@&7D?E&5d(>.AA,&5D&@7D.+>&ED)D@&E76@?E&E@>9E@5>@(%."D.97%&5 +9@6D.97%&6(DADA&A"-&M5d 966&a55&A>&@6&DD&7D(,5&6&7D%d(D!5.@E7@(>&@D9.@& %d&K97A%.++(@&7D.&55&6&7D@(,E5(AE"9E@A%&5>@9,@&AA.97DE69@5&a&A"7%.%DA%&A (DE%&A+97"D.977&55&A>&EI&7D)D@&59@A%(+.7.A?E.>&@6&DD@97D%&%([email protected]&@5+97"D.97 &D5@&5&I7"&%&"-?E&I@.7D%d(>.AA,&%7A5>@9,@&AA.97DE69@5&a   & DL>& %d75LA&  5@,& ("-&55& >&@6&D EAA. 5d9!A&@ID.97 %&A %.++(@&7"&A %c&K>@&AA.97%&+"D&E@A%d(>.AA,&ad(DE%&6&7(&>@a7%@(^"96>@7D%&ADE6&E@A%E A&.7  %&A 5(A.97A !(,7.7&A  .7A. 6.A &7 (I.%&7"& ?E& 5&A +"D&E@A 9E 5&A >@9D(.7&A .6>5.?E(A%7A5d(>.AA,&>>@.AA&7D(,5&6&7D%(@(,E5(A%7A5>@9,@&AA.97DE69@5&a dED@&A(DE%&A97D697D@(%&A5D(@D.97A%d&K>@&AA.97%&+"D&E@A%d(>.AA,&ad&AD5&"A %&{mi|n"97A.%(@("966&E7>@9D9j97"9,'7&m7"ME4:J&D5a^|z{|na&+"D&E@ %d(>.AA,& "97D@;5& 79D66&7D 5d(>.AA,& %E >@9D9j97"9,'7& a  AE@j&K>@&AA.97 %& {&7D@.7&5d&K"5EA.97%&5d&K97{{%&^"&?E."97%E.D5>@9%E"D.97%&V^E7& +9@6& "97AD.DED.I&6&7D "D.I& %& "&DD& [email protected]& 4.7A& ">!5& %& +[email protected]&@ 5 6.,@D.97 m -.,7&D5a^|zzna&@D.7&A-7Am{&D|nA9EA5&"97D@;5&%&"j@(,E5&7D5&A &K97A6EDE&55&6&7D&K"5EA.+A%& m>L@EID&4.7A&na  7 >5EA %&A 5D(@D.97A %E 7.I&E %d&K>@&AA.97^ "&@D.7A +"D&E@A %d(>.AA,& >>@.AA&7D 6ED(A %7A %&A ("-7D.5597A DE69@EKa d&AD >@ &K&6>5& 5& "A %E +"D&E@ }{ %7A 5&A 5&E"(6.&A 5L6>-91%&A "-@97.?E&A mna }{ >>@0D D@'A +@(?E&66&7D 6ED(%7A5&A&D"&A6EDD.97AA97DAA9".(&AE7&65%.&>5EA,@&AA.I&&DE7DEK %&AE@I.&>5EA"9E@D%&A>D.&7DAmE&A%&D5a^|z{|na 





 ~  '$     %         #   & @;5& %& 5d(>.AA,& 5D&@7D.+ %7A 5& "7"&@ &D 5 >@9,@&AA.97 DE69@5& &AD (,5&6&7D6.A&7I7D>@%&A(DE%&A+97"D.977&55&AAE@%.++(@&7DAI@.7DA%d(>.AA,&a&A (DE%&A97D>[email protected]%&697D@&@?E&"&@D.7A,'7&A>&EI&7D"9%&@>9E@%&AI@.7DAL7D%&A +97"D.97A!.959,.?E&A%.++(@&7D&A%7A5&"%@&%&5>@9,@&AA.97DE69@5&a&A%.++(@&7"&A %d(>.AA,& 9!A&@I(&A %7A 5&A "7"&@A >&EI&7D )D@& A9.D 5 "97A(?E&7"& %& 6EDD.97A ++&"D7D5&AA(?E&7"&A".Aj@(,E5D@."&A%&5d(>.AA,&^A9.D5"97A(?E&7"&%&5d"D.ID.979E 5@(>@&AA.97!&@@7D&%&+"D&E@A%d(>.AA,&a  5EA.&E@A"@"D(@.AD.?E&A%&A"&55E5&ADE69@5&A%("@.D&A>@ 7-7&D&.7!&@, A97D AA9".(&A  %&A I@.7DA %d(>.AA,& ?E. "97D@.!E&7D E >-(79DL>& AA9".(  "&A "@"D(@.AD.?E&Aa@&K&6>5&^%&796!@&EK,'7&A%&AI9.&A>9>D9D.?E&AA97D">!5&A%& >@9%E.@&%&A.A9+9@6&A>@99E7D.j>9>D9D.?E&A"966&>@&K&6>5&j9EAm.E@^ E3.!E"-.^s779^|z{|na&,'7& j&AD(,5&6&7D">!5&%&>@9%E.@&%&EKI@.7DA %d(>.AA,& EK +97"D.97A 9>>9A(&A >@9 9E 7D.j7,.9,(7.?E& m5.I&.@ s I@.79^ |z{|n m$'%41! {}na &A +9@6&A 7D.j>9>D9D.?E& &D >@9j7,.9,(7.?E& A97D @&D@9EI(&A %I7D,& &K>@.6(&A%7A5&A"&55E5&A"7"(@&EA&Am9775^.,&I7.^s5" @"&5^|z{|nI9.@$'%41!{}a >@9%E"D.97%&5d.A9+9@6& |%7A5&A"&55E5&A"7"(@&EA&A5&E@>&@6&D%&@(5.A&@E7& ,5L"95LA& 7(@9!.& &D %97" %& [email protected]@& %7A %&A "97%.D.97A 7(@9!.?E&A -9AD.5&A m&++&D @!E@,n D@'A +@(?E&66&7D 9!A&@I(&A %E@7D 5 >@9,@&AA.97 DE69@5& mI.%^ -&7^ AA7-^7955^s75&L^|z{z_ A@&5A&7&D5a^|z{}na & >5EA^ %dED@&A ,'7&A +9@D&6&7D .6>5.?E(A %7A 5 >@9,@&AA.97 DE69@5& "966& }^ @%{9E5&@("&>D&E@EK7%@9,'7&Amn^"9%&7D>9E@%.++(@&7DAI@.7DA%d(>.AA,&EK +97"D.97A%.++(@&7D&Am a-&7s&.AA^|z{~na  .7A. >5EA.&E@A "@"D(@.AD.?E&A %&A "&55E5&A "7"(@&EA&A %("@.D&A >@ 7-7 &D &.7!&@,^ "966& 5 @(A.AD7"&  5 69@D "&55E5.@&^ 5& %(@',5&6&7D %E 6(D!95.A6&^ 5d.7%E"D.97%&5d7,.9,(7'A&^5d"D.ID.97%&5d.7IA.979E5+9@6D.97%&6(DADA&AA97D @(,E5(A^&7>@D.&^>@%&A(I(7&6&7DA%d(>.AA,&a

 ~‚  

'%41!JL?7!+/*!2 ! '$$#1!,3!2'2-$-1+!2/1- 4'3!2/1#/'22%!*3!1,3'$1!*!5,3!24 -412 !*/1-%1!22'-,34+-1*!@ A/1"2-,,*H*@341!!5'!614%'2-5!18    &A%&@7.'@&A77(&A^%&796!@&EA&A(DE%&A97D%("@.D5&A5D(@D.97A%d(>.AA,&E "9E@A %& 5 >@9,@&AA.97 DE69@5& "& ?E.  >[email protected] %& A9E5.,7&@ 5d.6>9@D7"& %& "& >-(796'7&a&>&7%7DAA&M>&E%&I@.7DA97D(D(@(&55&6&7D"@"D(@.A(A>9E@39E&@E7 @;5&%7A5>D-959,.&a@6.5&A"&7D.7&A%d&K97A@(,E5(A&7D@&E7D.AAE"7"(@&EK&DE7 D.AAE A.7^ 5& >@9"-.7 9!3&"D.+ A&@ %& %(+.7.@ 5&A?E&5A >@D.".>&7D &++&"D.I&6&7D E >-(79DL>&"7"(@&EKa&A&"97%"-55&7,&A&@%d.%&7D.+.&@5&A@(,E5D&E@A%&"&A5D(@D.97A +.7%&>@9>9A&@%&AAD@D(,.&A%&D@.D&6&7Da  >@D.& AE.I7D& &AD "97A"@(& EK (I(7&6&7DA %d(>.AA,& @&D@9EI(A E "9E@A %& 5 D@7A.D.97(>.D-(5.9j6(A&7"-L6D&EA& 

 ~ƒ  ? "1'0$*)'= ,$//")/'0.)/$0$*) ,$0# '$*?( /)#5(01/<   5&AD!.&7(D!5.?E&5&A>@9"&AAEA%&%.++(@&7".D.97"&55E5.@&A97D"97D@;5(A>@%&A +"D&E@A %& D@7A"@.>D.97 m&K` L9 %7A 5 %.++(@&7D.D.97 6EA"E5.@&na &A (I(7&6&7DA %d(>.AA,& @(,E5(A EK "9E@A %& "&A >@9"&AAEA %& %.++(@&7D.D.97 A97D A>(".+.?E&6&7D AA9".(A  5 +97"D.97 %E D.AAEa @ &K&6>5&^ 5&A (>.AA,&A 5D&@7D.+A 9!A&@I(A A>(".+.?E&6&7D %7A 5& "&@I&E A97D "97D@;5(A >@ E7 +"D&E@ %d(>.AA,& A>(".+.?E& %E "&@I&Emn&D++&"D&7D%&AA"9%7DA>9E@%&A+97"D.97A7&E@5&Am5&&D5a^|zzna  d(DE%&%&A.A9+9@6&A%& |^>{|z"D&7.7&D m&7n>[email protected]%&6&DD@& &7(I.%&7"&5d(>.AA,&5D&@7D.+%7A5m 9AJ6.&D5a^|zzƒ_ &.@A&!.5"4&D5a^{ƒƒ‚_ I,7&@^55(A^ 9E77&E^6%^s-.&@L^{ƒƒ~na7A.D>@&K&6>5&?E&5&A&K97A 6EDE&55&6&7D&K"5EA.+A%& |"9%&7D>9E@%&A@("&>D&E@AI&"%&A++.7.D(A%.++(@&7D&A >9E@5&5.,7% ma-7,&D5a^{ƒƒƒn&D?E&5+9@6&6(A&7"-L6D&EA&%&5>{|zj "D&7.7 @(%E.D 5d"D.I.D( %& -9^ E,6&7D7D .7A. 5&A ">".D(A .7IA.I&A %&A "&55E5&A m7,.AJ &D 5a^ |zz‚na 7+.7^ 97 A.D ?E& 5&A %&EK .A9+9@6&A %&   9E  m665.7 9@D-959, 9+ D-& @9A9>-.5 >@9D&.7 n @(,E5&7D 5&A .7D&@"D.97A &7D@& 5&A +.56&7DA%d"D.7&5d&KD@(6.D(%&A56&55.>9%&A6.A?E&A&E5&5+9@6&6(A&7"-L6D&EA& +[email protected]&56.,@D.97&D5d.7IA.97%&A"&55E5&Am9EAA9A^5A69^&D5a^|z{{na  d@@.I(&%&D&"-7959,.&A>&@6&DD7D5d75LA&%&5d(>.AA,&5D&@7D.+5@,&("-&55& "966&5&A>E"&A9E5&A(?E&7$,&^>[email protected]%&6.&EK"@"D(@.A&@5&A(I(7&6&7DA.7%E.DA 59@A %& 5  6.A EAA. %& "@"D(@.A&@ 5&A @(,E5D&E@A %& "&A (I(7&6&7DA %d(>.AA,&a @6.5&A@(,E5D&E@A.%&7D.+.(A^97%.AD.7,E&5&A>@9D(.7&A{&D|m&>.D-&5.5A>5.".7, @&,E5D9@L>@9D&.7{i|n?E.@(,E5&7D5d(>.AA,&%& |&D?E.A97DE7.?E&6&7D&K>@.6(A %7A5&A"&55E5&A(>.D-(5.5&Am 9I-77.AL7s@AD&7A^|zzna7&++&D5&E@&K>@&AA.97&AD @(>@.6(&>@5&A+"D&E@A%&D@7A"@.>D.97"5(A%&5^7.5{^{&DJ.ADm 9@.,E"-.&D 5a^|z{|_&.74&^E^s-&7,^|z{|_->.@9&D5a^|z{{nadED.5.AD.97%&>E"&A397"D.97A >[email protected] %& 697D@&@ ?E& "&A >@9D(.7&A @(,E5&7D %.++(@&7DA DL>&A %d(>.AA,& 5D&@7D.+ m@M&"-&D5a^|zzƒn&D?E&5&>@9,@66&%d(>.AA,&&AD!@9,(59@A%&5m@M&"- &D5a^|z{zna{&D|@(,E5&7D(,5&6&7D%&D@'A796!@&EK(I(7&6&7DA%d(>.AA,& %&A D@7A"@.>DA "9%7D %&A >@9D(.7&A .6>5.?E(&A %7A 5d"?E.A.D.97 %E >-(79DL>& 6(A&7"-L6D&EK m>@9D(.7&A %E "LD9A?E&5&DD&^ %& +.KD.97  5d"D.7&^ %& 5 6D@."& &KD@j

 z "&55E5.@&^%&A397"D.97A"&55E5&Aj"&55E5&A^%&-9 A&9E%&A"96>9A7DA%&5I9.& 4.7A&nm@M&"-&D5a^|z{znm'%41!{~D!5&E>|na  7 ED@& +"D&E@ %d(>.AA,& .6>5.?E( %7A 5  &AD 5& +"D&E@ !+9K| mƒn m&7!5&A&D5a^|z{}_&7!5&A^.,75^,-%.,E.7^9@D^sM.^|z{|na 5&AD(,5&6&7D "977E >9E@ )D@& .6>9@D7D %7A 5 6"-.7&@.& %d(>.AA,& %&A "&55E5&A A9E"-&A &6!@L977.@&Am&9&D5a^|zzƒna97&K>@&AA.97&AD5(,'@&6&7DAE>(@.&E@&%7A5&A"&55E5&A 6(A&7"-L6D&EA&A ?E& %7A 5&A "&55E5&A (>.D-(5.5&Aa 5 @(,E5& E7 &7A&6!5& %d&K97A >@D.".>7D5%(+.7.D.97%&5"&55E5&6(A&7"-L6D&EA&a&796!@&EK&K97A@(,E5(AA97D "966E7AI&"5&A".!5&A%&{i|m6.A@(,E5(A%7A5&A&7A.7I&@A&nm@&ED.,6&D5a^ |z{}_&7!5&A&D5a^|z{}!nad75LA&5@,&("-&55&%&A".!5&A%&{&D|.7A. ?E&!9K|>[email protected]%&%(+.7.@E7"9%&%d(>.AA,&@&5D.+"&A%&EK+"D&E@Am$'%41!{~na 9@A?E& 5&A >@9D(.7&A A 9E !+9K| A& +.K&7D AE@ 5d.7D@97 7^ "&5 +[email protected]& 5d.7"5EA.97 %& 5d&K97 7a  5d.7I&@A&^ A. 5&A >@9D(.7&A A A& +.K&7D AE@ 5d&K97 7 9E AE@ 5d.7D@97 7j{^ "&5 +[email protected]&5d&K"5EA.97%&5d&K977m$'%41!{~na  &>5EA^5&A%.++(@&7DAD@IEKAE@A&D+9K|97D697D@(?E&5AE@j&K>@&AA.97 9E5d&KD.7"D.97%&"&A>@9D(.7&A&ADAE++.A7D&>9E@(D!5.@E7&>@D.&55&&7ED.5.A7D%&A 6@?E&E@A >-(79DL>.?E&A "966& 5 6.,@D.97 "&55E5.@&^ 5& @&69%&5,& %E "LD9A?E&5&DD& %d"D.7&9E5d&K>@&AA.97%&>@9D(.7&A(>.D-(5.5&A9E6(A&7"-L6D&EA&Aa&5%(697D@&?E& 5d(>.AA,& 39E& E7 @;5& 63&E@ %7A 5 D@7A.D.D.97 (>.D-(5.9j6(A&7"-L6D&EA&a 7A 5& "97D&KD&%E"7"&@%EA&.7^5&A>@9D(.7&AAA97DD@'A>&E&K>@.6(&A%7A5&A9EAjDL>& %&A5.,7(&Ag5E%.7&j59Jhm!A5n?E.A97D5&A>5EA,@&AA.I&Am>E4&D5a^|z{z_&I&&D 5a^ |zz€na &>&7%7D^  "& 39E@^ E"E7& (DE%& 7d AA9".( 5d&K>@&AA.97 %& "&A >@9D(.7&A @(,E5D@."&A%&5d(>.AA,&%7A55d,@&AA.I.D(%7A%&ADE6&E@A-E6.7&Aa  dED@&A +"D&E@A %d(>.AA,& 97D (D( .6>5.?E(A %7A 5  "966& { ?E. "97D@;5&79D66&7D5d(>.AA,&%&m -.,7&D5a^|zzn&D5-9 A&"{mD9As ^|zzƒna{>&ED)D@&@(,E5(>@E7ED@&+"D&E@%d(>.AA,&^€‚&7@(>97A&5I9.&  a&A+"D&E@AA97D&K>@.6(A%&+$97>5EAE!.?E.D.@&Aa&A+"D&E@A%d(>.AA,&"966& {^ 9E %&A 6&6!@&A %& 5 +6.55& %&A -7A A97D (,5&6&7D .6>5.?E(A %7A 5 @(,E5D.97%&5m->.@9&D5a^|z{{na  

 { 







 '%41!JM?EF-,3'-,!3*-*'23'-, !2'*!2 !@

EF- ! A#/'22%!1!*3'$47/1-3#',!2!3$-7K@ S', ',%2*'',%!,&,!1> S', ',%2/*'',%','&''3-1>S-7K ', ',%2/*'',%!&,!1@A/1"219!&H123!,!+',12',,!1'-*-%8KIJK!3! 1!,!!3!17341!!5'!6,!1KIJL@ 

 | 97"5EA.97   d(>.AA,& 5D&@7D.+ &AD E7 >@9"&AAEA ?E. >&@6&D %& 69%.+.&@ 5& 6&AA,& %& 5d&K>@&AA.97%&A,'7&Aa6[email protected](%&A,'7&A-E6.7AA97DA9E6.A"&DD&@(,E5D.97"&?E. +.D%&5d(>.AA,&5D&@7D.+E7&%&A(D>&A63&E@&A%&5d&K>@&AA.97%&A,'7&Aa&>@9"&AAEA &AD "97D@;5(^ &7 >@D.&^ >@ %&A +"D&E@A %d(>.AA,& ?E. >&EI&7D I9.@ E7& &K>@&AA.97 %.++(@&7D.&55& &7 +97"D.97 %&A D.AAEA 9E %E DL>& "&55E5.@&a 5 &AD .7%.A>&7A!5& %7A %&A >@9"&AAEA"&55E5.@&A"5(A^D&5A?E&5%.++(@&7D.D.976.A(,5&6&7D%7A%&796!@&EA&A

>D-959,.&Aa  E "9E@A %& 5 %&@7.'@& %("&77.&^ %& 796!@&EA&A (DE%&A 97D >[email protected] %& 697D@&@ 5d&K>@&AA.97!&@@7D&%dE7,@7%796!@&%d"D&E@A%&5d(>.AA,&5D&@7D.+%7A%&AD.AAEA "7"(@&EK&7"96>@.A97EKD.AAEAA.7Aa&".@(I'5&?Ed.5Ad,.D%dE7>@9"&AAEA@(,E5(E "9E@A%&5>@9,@&AA.97DE69@5&a7.%&7D.+.7D"&@D.7A@(,E5D&E@A%&"&A(I(7&6&7DA^.5 >>@0Dj%&>E.AA&E5&6&7D?E&5?E&A77(&Aj?E&5d(>.AA,&5D&@7D.+&ADE7>@9,@66& %d&K>@&AA.97%&A,'7&A@(,E5(E"9E@A%&5>@9,@&AA.97DE69@5&&D&7>@D."E5.&@>&7%7D 5a&A+"D&E@A{i|&D@!+9K|>>@.AA&7D"966&5&A@(,E5D&E@A63&E@A%&5 9!A&@I(A%7A%&A69%'5&A"&55E5.@&A"7"(@&EKa   >@D.& AE.I7D& A&@ "97A"@(& EK  -(5."A&A  69D.+  !9K  &D { &D  5&E@A +97"D.97A %7A 5&A %.++(@&7D&A (D>&A %E 6(D!95.A6& %&A Aa d&K>5.".D&@.79D66&7D5&E@A+97"D.97A%7A5@(,E5D.97%&5d(>.AA,&&D5&E@A+97"D.97AE "9E@A%&5>@9,@&AA.97DE69@5&a    



  

 } .0$ ; / # '$//KA,LNB0GMA,MHB<   &A  -(5."A&A A97D %&A &7ML6&A ">!5&A %& 5.&@ &D %cED.5.A&@ 5d >9E@ @&69%&5&@ 5&A AD@E"DE@&A A&"97%.@&A %& 5d &D %&A "96>5&K&A j@9D(.7&a 97D@.@&6&7D EK  -(5."A&A^ 5&A  -(5."A&A 7& >&EI&7D 69%.+.&@ ?E& %& "9E@D&A A(?E&7"&A m&7I.@97 {z>!n ma 7, s 749JA4L^ |zz€na &A  -(5."A&A A97D D@'A "97A&@I(&A&D39E&7DE7@;5&%7A56[email protected](%&A(D>&A%E6(D!95.A6&%&5d^%&>E.A5 D@7A"@.>D.97 3EA?Ed 5 D@%E"D.97 m 747&"-D^ |z{zna >@'A E7& .7D@9%E"D.97 AE@ 5&A  -(5."A&A^&D&7>@D."E5.&@AE@5&A-(5."A&A69D.+^5d9!3&"D.+%&"&DD&>@D.&A&@ %& %(D.55&@ 5&A @;5&A %& %&EK %& "&A  -(5."A&A^  &D {^ %7A 5&A %.++(@&7DA >@9"&AAEA%E6(D!95.A6&%&5da@5AE.D&^3&>@(".A&@.5&A(I.%&7"&A?E.@&5.&7D5&A -(5."A&A&D{E%(I&59>>&6&7DDE69@5&D5a 

 ? / # '$//(*0$!   &A  -(5."A&A A97D 9@,7.A(&A &7 A.K ,@7%&A +6.55&A &7 +97"D.97 %& 5&E@A AD@E"DE@&A&D%&5&E@A"D.I.D(A"&55E5.@&Aa&A-(5."A&A69D.+^>>@D.&77&7D 5 %&EK.'6& +6.55&a &DD& %&@7.'@& &AD "96>9A(& %& >5EA %& } 6&6!@&A "-&M 5d 966& %97D  &D { m.@67jJ.55.6A s E&7D-&@^ |z{zna &A  -(5."A&A %& "&DD& +6.55& >9AA'%&7D %&A 69D.+A %& +.KD.97  5d &D %d-L%@95LA&  5d &KD@)6&6&7D "97A&@I(Am$'%41!{na&A69D.+A^ ^ &D A97D.6>5.?E(A%7A5d-L%@95LA&%&5dad".%& 6.7(  m".%& A>@D.?E&n %E 69D.+ "@"D(@.AD.?E&  .7D&@,.D I&" 5d.97 6,7(A.E6 "96>5&K(I&"5da&DD&.7D&@"D.97&AD&AA&7D.&55&>9E@5d(D!5.AA&6&7D%dE7&>9"-&?E.^ &7AA9".D.97I&"5&A.7D&@"D.97A.7D@695("E5.@&A&7D@&5&A69D.+A &D ^5&A69D.+A &D  &D 5&A 69D.+A  &D  ^ >&@6&D 5d(D!5.AA&6&7D %dE7 A.D& "D.+ &7ML6D.?E& @&?E.A >9E@ 5d-L%@95LA& %& 5d m9@%.7 &D 5a^ |z{|_ -.^ 9@%.7^ .7%&@^ .7%&@^ s E^ |zz~na d".%& 6.7(  m5LA.7&n %E 69D.+  (D!5.D E7& 5..A97 -L%@9>-9!& I&" 5& Xj>-9A>-D& %& 5d m$'%41! {na 9ED& 6EDD.97 >97"DE&55& AE@ "& @(A.%E !95.D 5d"D.I.D( A& %& 5 >@9D(.7&m@ED-&@As" L^|zz|_9@%.7^77&@^.7%&@^s7@9?E&A^|zz~na&6ED7D >9@D7D "&DD& 6EDD.97  &AD ED.5.A( 59@A %&A %.++(@&7D&A &K>(@.&7"&A %& @&ADE@D.97 +97"D.977&55&@(5.A(&A%7A5&A%.++(@&7DA@D."5&A>@(A&7D(A%7A5>@D.&@(AE5DDa   ~  '%41!JN?+-3'$2 !2&#*'2!2@ EF2&#+ !* '231'43'-, !2+-3'$2 EF#04!,! !2+-3'$2@,1-4%!*!2+-3'$2'+/*'04#2 ,2*A&8 1-*82! !*A>!,*!4*!2 +-3'$2'+/*'04#2 ,2**''2-,5!*A@EF&#+ !2*''2-,2&8 1-/&-!2!,31!*!2 ' !2+',#2 !2+-3'$2> > !3 !3*A@ $*"&!5!13!', '04!*!1#2' 4*82',!+43# $', A-*'1*A3'5'3#2! !N!3JP@ /3# A/1"2 ', !1H ,)-62)8> >KIJJ!343,+@H ,)-62)8>>KIJL@   &A  -(5."A&A %& 5 %&EK.'6& +6.55& A97D AA9".(&A  D9ED&A 5&A (D>&A %E 6(D!95.A6&%&5dm$'%41!{€na&A,@7%&A(D>&A%E6(D!95.A6&%&5d7("&AA.D&7D 5& @&"@ED&6&7D %& 796!@&EA&A @.!97E"5&9>@9D(.7&A &D %& 796!@&EK @(@@7,&6&7DA AD@E"DE@EK ?E. A97D 6(%.(A >@ "&A  -(5."A&A m&K` 5& A>5."&9A96&^ 5 6"-.7&@.& %& D@%E"D.97na    

 

'%41!JO?!/1#2!,33'-, !2/1-!2242!**4*'1!2'+/*'04,3*!2&#*'2!2+-3'$2 @  /3# A/1"2 ', !1H ,)-62)8>>KIJJ@ !2&#*'2!22-,3,#!22'1!2 *31,21'/3'-,>*'-%#,"2! !21'-2-+!2!3*A#/'22%!> !2#3/!2,4*#'1!2@ !2/1-3#',!22-,3422''+/*'04#!2 ,2 !2#3/!283-/*2+'04!2-++!*31 43'-,-4 * #%1 3'-, !22@ 

 



 

 €  ?*)0$*)/(*' 1'$.// # '$//K0GM<   &A-(5."A&A&D{A97D%&EK6&6!@&ADL>.?E&A%&5+6.55&69D.+A a 55&A A97D .6>5.?E(&A %7A 5 D@7A"@.>D.97^ 5d(>.AA,&^ 5d&K>9@D^ 5 6DE@D.97 %&A 6."@9A^ &D 5 %(,@%D.97 %&A Aa  &D { 97D E7& "D.I.D( A& &D A97D ">!5&A %& %(@9E5&@ m 747&"-D^ |z{zna  &D { A97D %&EK >@9D(.7&A >@59,E&Aa {(D(.%&7D.+.(&>9AD(@.&E@&6&7D,@ "&A97-96959,.&I&""&DD&>@9D(.7&a &%96.7&-(5."A&&D5&A7&E+A%96.7&Am^ ^ ^ !^ ^ ^ ^&D nA97D.%&7D.?E&A>5EA %&ƒz†&7D@&"&A%&EK>@9D(.7&Aa&A%96.7&Aj&Dj[email protected]EKA97D^?E7D&EK^>5EA %.I&@,&7DAI&"@&A>&"D.I&6&7D€z†&D}z†%d.%&7D.D(mI9.@+.,E@&{na  











'%41!JP?314341!!3&-+-*-%'! !2&#*'2!2 7N!3 7JP@ A/1"24**!1C/!H--1!@4341!,-*@KIJJ@   21" &'% &%#'!    >E.A{.7D&@,.AA&7DI&"%&A+"D&E@A%&D@7A"@.>D.97&DA97D">!5&A%& 69%E5&@ >9A.D.I&6&7D 9E 7(,D.I&6&7D 5&E@A "D.I.D(Aa  &D { A97D %97" %&A "9j @(,E5D&E@AD@7A"@.>D.977&5A?E..7D&@,.AA&7DI&"5>95L6(@A& m9AA9Js 747&"-D^ |zz}na  >@&6.'@& (I.%&7"& %E @;5& %&  %7A 5 @(,E5D.97 %& 5 D@7A"@.>D.97  (D(

  %(697D@(&ED@I&@AA97@;5&%&"9j"D.ID&E@%E@("&>D&E@EK&AD@9,'7&AmWnm7%9-&D 5a^{ƒƒƒna&@;5&(,5&6&7D(D(%("@.D>9E@{mD7!&&D5a^|zz{n&D(D&7%E %dED@&A@("&>D&E@A7E"5(.@&A"966&5&@("&>D&E@EK7%@9,'7&Amnm5@4&D5a^|zz‚n &D 5& @("&>D&E@  5 I.D6.7&  m,7&@ &D 5a^ |z{|na   &AD E7 "9j"D.ID&E@ %E AE>>@&AA&E@%&DE6&E@>}^&D"97D@;5&5D@7A"@.>D.97%&,'7&A.6>5.?E(A%7A5d@@)D%E "L"5& "&55E5.@& &D 5d>9>D9A& m."95 &D 5a^ |z{|n  &Di9E { A97D (,5&6&7D .6>5.?E(A %7A 5 %.++(@&7".D.97 "&55E5.@& 9F .5A 39E&7D 5& @;5& %& "9j"D.ID&E@A %& %.++(@&7DA+"D&E@A%&D@7A"@.>D.97"966&E7K|m &7A&7&D5a^|zz‚n9EL9m@&DD.&D 5a^|zz€nm3*!~na&@;5&%&&D{>>@0D>@D."E5.'@&6&7D.6>9@D7D>9E@5 6.A&&7>5"&%&A>@9"&AAEA%&%.++(@&7D.D.97"97D@;5(A>@"&A+"D&E@A%&D@7A"@.>D.97a .7A.  &Di9E { A97D 7("&AA.@&A  5 %.++(@&7D.D.97 9AD(9!5AD.?E& &D  5 %.++(@&7D.D.97 %E 6EA"5& A?E&5&DD.?E&a 7 D@I.5^ >E!5.( @("&66&7D %7A 5d(?E.>&^  697D@( ?E&  &D { A97D %&A "9j@(,E5D&E@A %& ^ +"D&E@ %& D@7A"@.>D.97 .6>5.?E( %7A 5 6.,@D.97 "&55E5.@& m &@677 &D 5a^ |z{|na  "9j@(,E5& (,5&6&7D 6%}m@7&@&D5a^|zz~n6.A5&A,'7&A".!5&A%&i6%}7d97D>A(D(.%&7D.+.(Aa .7D&@,.D(,5&6&7DI&"5Xj"D(7.7&mI9.@>@D.&}a}na&A+97"D.97A%&"9j"D.ID&E@A D@7A"@.>D.977&5A %&  &D { .6>5.?E&7D 5& @&"@ED&6&7D %dED@&A "9j"D.ID&E@A "966&i9E>}zzim%&EK"(DL5jD@7A+(@A&Ana   &D { 39E&7D (,5&6&7D 5& @;5& %& "9j@(>@&AA&E@A D@7A"@.>D.977&5A &7 @&"@ED7D %&A -.AD97&A %("(DL5A& m n {^ | &D } m@D&@^ .7^ .E^ 7,^ s .E^ |z{z_ "9!A &D 5a^ |zz_ .5A97 &D 5a^ |zz~na &A %&EK +97"D.97A m%& "9j"D.ID&E@ 9E "9j @(>@&AA&E@n A97D >@9!!5&6&7D %.@.,(&A >@ E7& !57"& &7D@& "(DL5D@7A+(@A&A &D 5&A -.AD97&A %("(DL5A&Aa 5  (,5&6&7D (D( 697D@( ?E&  AD!.5.A& 5& "96>5&K& "9-&A.7iI&"5d797"9%7D>@D.".>7D.7A.5+97"D.97%d.7AE5D.97m9&D 5a^|z{zna   &D { A97D %97" %&A >@9D(.7&A ?E. 39E&7D E7 @;5& .6>9@D7D %7A 5& @&"@ED&6&7D%&"96>9A7DA&AA&7D.&5A%&56"-.7&@.&D@7A"@.>D.977&55&&7+.A7D5&5.&7 &7D@&5&+"D&E@%&D@7A"@.>D.97&D%&A>@9D(.7&A"9j@(,E5D@."&A"966&%&A69%.+."D&E@A %&5"-@96D.7&a55E6.'@&%&D9ED&A"&A(DE%&A^.5>>@0D?E&5&A+97"D.97AA&&D -(5."A& 7& A97D >A @&?E.A&A %7A 5&E@ +97"D.97 %& "9j@(,E5D&E@ D@7A"@.>D.977&5 mE55&@j "&^|z{}_ 747&"-D^|z{zn"&?E.AE,,'@&?E&"&A>@9D(.7&AA97D6E5D.+97"D.977&55&A&D

 ‚ ?E&5&E@A%.++(@&7DA%96.7&A>@9D(.?E&AA97D%.++(@&7D.&55&6&7D.6>5.?E(A%7A>5EA.&E@A >@9"&AAEA"&55E5.@&Aa



Facteurs de Transcription Autres facteurs influençant la transcription ERα HDAC1/2/3+ AR SRC-1 VDR SRA p53 MAML1/Notch1 Ddx5 Runx2 CBP/p300 NF-κBTBP Smad3 TFIIA E2F1 Brg-1 β-catenin P/CAF ERα SRC-1, SRA, CBP, p300, P/CAF Ddx17 p53 HDAC1/2/3 MyoD Ddx5/Ddx17 NFAT5  !5&~`"D&E@A%&D@7A"@.>D.97&D>@9D(.7&A@(,E5D@."&A%&5D@7A"@.>D.97.7D&@,.AA7D I&"&D{a977(&A"96>.5(&A>@D.@%&4**!1C/!H--1!@4341!,-*@ KIJJ@

 

  31" &/#&&   &D{39E&7DE7@;5&.6>9@D7D%7A5d(>.AA,&79D66&7D,@ "&5&E@A +97"D.97AA&&D-(5."A&m.7^7,^7,^ E7,^s.E^|zzna  &A 75LA&A >@ A>&"D@96(D@.& %& 6AA&^ 97D >[email protected] %d.%&7D.+.&@  "966& "96>9A7D%EA>5."&9A96&m&E!E&@&D5a^{ƒƒ‚n&D>5EA>@(".A(6&7D%E>@(jA>5."&9A96& m @D6ED-&D5a^|zz|na&A(DE%&A!.9"-.6.?E&A97D697D@(?E&7&>@D.".>&>A5 @&"977.AA7"&%EA.D&d%d(>.AA,&>@5&A7{6.A.%&5%.AA9".D.97%EA7{ E 7.I&E %E A.D& d %d(>.AA,& >&@6&DD7D .7A. 5& >AA,& %E >@(jA>5."&9A96& E A>5."&9A96& m.7 &D 5a^ |zz_ a .E &D 5a^ |zz‚_ a .E^ |zz|na d"D.I.D( -(5."A& &AD 7("&AA.@&5%.AA9".D.97%&{6.A"&DD&>@9>@.(D(7d&AD>A@&?E.A&>9E@5&@&"@ED&6&7D %E[email protected]~^€ia55E6.'@&%&"&A%.++(@&7DAD@IEK@(5.A(A>@5&,@9E>&%&a .E^>>@0D"966&E7"96>9A7D&AA&7D.&5%EA>5."&9A96&a 

 ƒ   &D { A97D .6>5.?E(&A %7A 5 @(,E5D.97 %& 5d(>.AA,& 5D&@7D.+ %& >5EA.&E@A D@7A"@.DAa @ &K&6>5&^  @(,E5& 5d(>.AA,& 5D&@7D.+ %& -j@A I&" %dED@&A +"D&E@A %d(>.AA,& m E.5^ DD97.^ D(I&7.7^ "-j&5.A^ s D&^ |zz}na 55& &AD (,5&6&7D ">!5&%&%(@9E5&@E7&AD@E"DE@&A&"97%.@&E7.I&E%EA.D&d%d(>.AA,&%&5d&K97{z%& E>&@6&DD7D.7A.5@&"977.AA7"&%EA.D&d>@{A7m @&D5a^|z{{nad(>.AA,& %& 5d&K97  %& 5 >@9>97.7&  "@%.?E&  &AD EAA. @(,E5( >@  ?E. +[email protected]& 5 @&"977.AA7"&%dE7A.D&%&+.KD.97>9E@5&+"D&E@%d(>.AA,&{%7AE769%'5&%& %LAD@9>-.& 6L9D97.?E& mE@&7D &D 5a^ |z{|na &A %&EK %&@7.&@A &K&6>5&A >&@6&DD&7D %& 697D@&@ ?E&  @(,E5& 5d(>.AA,& I. A97 @;5& %7A 5 69%.+."D.97 %& 5 AD@E"DE@& A&"97%.@&%&5da{&AD">!5&%&@(,E5&@5d(>.AA,&5D&@7D.+%&~~m 97.7,&D 5a^ |zz|na 5 Ad,.D %& 5dE7.?E& (DE%& 697D@7D 5d.6>5."D.97 %& { %7A 5d(>.AA,& 5D&@7D.+a   &AD %97" E7 @(,E5D&E@ %& 5d(>.AA,& ?E. ,.D  >5EA.&E@A 7.I&EK %& 5 @(,E5D.97`.%&5@&"977.AA7"&%EA.D&d^%.AA9".D.97%E{A7&D69%.+."D.97%& 5d""&AA.!.5.D(%dED@&A+"D&E@A%d(>.AA,&a  

  41" & '(%'! & %! &   &796!@&EA&A(I.%&7"&A@&5.&7D&D{56DE@D.97%&A6."@9Aa7 &++&D^ %&A 75LA&A >@ A>&"D@96(D@.& %& 6AA&^ 97D >[email protected] %d.%&7D.+.&@  "966& "96>9A7D %E "96>5&K& @9A- &D 5 %(>5(D.97 %&  9E %& { "97%E.D  %&A 5D(@D.97A%E6(D!95.A6&%&796!@&EK6."@9AmE4E%&D5a^|zz_a7,^ E7,^ s E^ |z{|na d"D.I.D( -(5."A& %&  &D { &AD @&?E.A& >9E@ 5 6DE@D.97 %&A 6."@9Aa7&++&D^.5(D(697D@(?E&"&DD&+97"D.97>&@6&D%&@&7%@&5&>@.j6."@9 ""&AA.!5&>9E@5&"5.I,&>@@9A-mE4E%&D5a^|zzna&>5EA5d.7D&@"D.97&7D@& &D@9A-.7A.?E&>}&D5&A>@9D(.7&A6%AA97D(,5&6&7D7("&AA.@&A56DE@D.97 %& "&@D.7A 6."@9A "966& 6.j|{ mI.A^ .5L@%^ ,7^ s D^ |zz‚_ EME4. &D 5a^ |zzƒna &A %977(&A @&7+9@"&7D 5d.%(& %dE7& .7D&@"D.97 +97"D.977&55& &7D@&  &D 5&A >@9D(.7&A6%Am@7&@&D5a^|zz~n&D>}mD&A&D5a^|zzna39E&(,5&6&7DE7 @;5& %7A 5 6DE@D.97 %E >@(j6."@9 &7 .%7D  5 +9@6D.97 %E "96>5&K&   m5M67^-E!&@Dj95&67^sE@7&EK^|zzna7+.7&ADEAA.@&?E.A&>9E@%(@9E5&@  €z 5&>@("E@A&E@%E6."@95&Dj',5'31-m5M67&D5a^|zzn&D>@D.".>&56DE@D.97%& 6."@9AD&5A?E&6.j|{&D6.j{‚|?E.97D>9E@".!5&A%&A"96>9A7DA%E"LD9A?E&55&DD& %d"D.7&ma7,&D5a^|z{|na.7A.^.5&AD>@9>9A(?E&"97D@.!E&E@(@@7,&6&7D %E "LD9A?E&55&DD& %d"D.7& I. A +97"D.97 @(,E5D@."& %7A 5 6DE@D.97 %&A 6."@9A L7D>9E@".!5&5&A"96>9A7DA%E"LD9A?E&5&DD&ma7,&D5a^|z{|na   d&7A&6!5&%&"&A(DE%&A@(I'5&5d.6>5."D.97%&&D{%7A56DE@D.97 %&A 6."@9A^  6.A .5 A& >&ED (,5&6&7D ?E& "&A >@9D(.7&A .7D&@I.&77&7D &7 697D^ 79D66&7DE7.I&E%&5D@7A"@.>D.97%&"&A6."@9AA9E%&67.'@&.7%(>&7%7D& 5&E@"D.I.D(%d-(5."A&a  

  51! '! &6'27 &&('%&#%!&&(&( '!&  /    @6. 5&A ED@&A +97"D.97A %&  &D { %7A 5& 6(D!95.A6& %& 5d^ 97 %.AD.7,E& 5&E@ .6>5."D.97 %7A 5 6DE@D.97 %& 5d&KD@(6.D( }d %&A A m"5.I,& &D >95L%(7L5D.97na 7A "& >@9"&AAEA^  .7D&@,.D I&" j ?E. >@D.".>& E "5.I,& %& 5d>@(j6&AA,&@&7AA9".D.97I&"&D>~7@!m &.C5&@^5D6&L&@^D&.7^-5677j "-.++5&@^sD-5^|z{}na-&M5%@9A9>-.5&^@(,E5&5&@&5@,,&%&5>95L6(@A&%&5 "-@96D.7& mEAM"M4 s >@%5.7,^ |zz€n &D "-&M 5 5&IE@&^ 5 %(>5(D.97 %&  m|n "97%E.D  %& 6E5D.>5&A %(+EDA %& 6DE@D.97 %&A A 79D66&7D E 7.I&E %& 5 6DE@D.97%&A&KD@(6.D(A}dm59ED.&@^^,EL&7^s@7^|z{|na   59"5.AD.97%&&D{7d&AD>A@&AD@&.7D&E79LEm66^."95^E55&@j "&^ s 697%^ {ƒƒ€n^ >E.A?E& %& 796!@&EA&A (DE%&A @(I'5&7D 5&E@ 59"5.AD.97 "LD9>5A6.?E& m9AA9J s 747&"-D^ |zz}_ -.7^ 9AA9J^ @7%&^ s 747&"-D^ |zz_ a 7,^.7^s.E^|zz€na7>@D."E5.&@^&ADE7&>@9D(.7&">!5&%&+.@&57I&DD& &7D@& 5& 79LE &D 5& "LD9>5A6& m a 7,^ 9^ E7,^ 7,^ s .E^ |zzƒn9F &55& &AD 7("&AA.@&5d&K>9@D%d6&AA,&@m&K`|nm97D&D5a^|z{|na&>5EA^.7D&@,.D I&"}%7A5&"LD9>5A6&E"9E@A%&5>-A& |im-9.s&&^|z{|n&DE7&(DE%& 6&7(& >@ 5& ,@9E>& %&  .E^  @(I'5& ?E&  .7D&@,.D I&" 5& "5".E6 &D 5  €{ "569%E5.7&E7.I&E%&A56&55.>9%&A%&A"&55E5&A6.,@D9.@&Aad>@'A5&AED&E@A^"&DD& +97"D.97 %&  >9E@@.D I9.@ E7 .6>"D AE@ 5 6.,@D.97 %&A "&55E5&A E "9E@A %& 5 >@9,@&AA.97DE69@5&ma7,&D5a^|z{|na   -&M 5 5&IE@&^  .7D&@,.D I&" {^ E7&  -(5."A& 39E7D E7 @;5& 63&E@ %7A 5&  m797jA&7A 6&%.D&% %&"Ln &D 5 %(>5(D.97 %&  m>%|n "97%E.D  5d""E6E5D.97 %d 797jA&7A m97%^ 7,EA^ &^ s "9!A97^ |zz{na 7A %&A "&55E5&A -E6.7&A^&D{.7D&@,.AA&7DI&"}&DA97D">!5&A%&@(,E5&@5AD!.5.D(%& 5&E@AAm &.C5&@&D5a^|z{}na&A@(AE5DDA&K>5.?E&7D&7>@D.&>9E@?E9.5%(>5(D.97%& 5dE7&%&"&A>@9D(.7&A"97%E.D5dE,6&7DD.97%&A97>@59,E&a   &D{A97D(,5&6&7D.6>5.?E(&A%7A56DE@D.97%&5d@.!9A96.?E& "-&M55&IE@&&D%7A%&A"&55E5&A-E6.7&Am97%&D5a^|zz{_ 55^-5677j"-.++5&@^s D-5^ |zzna  %(>5(D.97 %& "&A >@9D(.7&A "97%E.D  %&A %(+EDA %& 6DE@D.97 %&A  @.!9A96.?E&A&DE7&>&@DE@!D.97%&AAD@E"DE@&A7E"5(95.@&Am 55&D5a^|zzna    +97"D.97 -(5."A& %&  &Di9E { &AD @&?E.A& EK "9E@A %& "&A %.++(@&7DA >@9"&AAEA?E.7("&AA.D&7D5&@&69%&5,&%&5AD@E"DE@&A&"97%.@&%&5dm6DE@D.97%&A @.!9A96.?E&n9E5&@&69%&5,&%&"96>5&K&A@.!9j>@9D(.?E&AmDE@D.97&7}d^^ 6DE@D.97%&A@.!9A96.?E&^&K>9@D7E"5(.@&na

 €| 

 +     (,        ++    

   

      *+  +  (, (-+$-)-  &  #' #( #) ) #*

     "    +  +  (,  (,   ' ( ) 

' ( ) *    ' ) *     !%   +      + #   )    + "  

       

'%41!JQ?&#+ !2 '$$#1!,321.*!2 !N!3JP ,2*+3413'-, ! '$$#1!,32 2@ +'*!+!,3/13%#/1'+-,+,@ 

  61 !'! #!&''%('! &    &D { >@D.".>&7D  %& 796!@&EA&A +97"D.97A "&55E5.@&A %E 6(D!95.A6& %&A A &D 5&A 69%.+."D.97A >9ADjD@%E"D.977&55&A A97D E7 A>&"D .6>9@D7D %& 5 @(,E5D.97%&5d"D.I.D(%&"&A>@9D(.7&Aa  &D{>&EI&7D)D@&E!.?E.D.7L5(&A`>95LjE!.?E.D.7L5(&>9E@mEA&I." &D 5a^ |zz{n &D 6979jE!.?E.D.7L5(& >9E@ { m997&L^ 9&5^ dAA9@9^ 5.A!E@L^ s 747&"-D^|z{zna&D{A97DE!.?E.D.7L5(&AAE@E7&5LA.7&59"5.A(&&7[email protected]5&

 €} m }>9E@&D z>9E@{na&D{>&EI&7D(,5&6&7D)D@&L5(&A m "9!A&D5a^|zz_997&L^ @7%&^5.A!E@L^s 747&"-D^|z{zna&A%&EK69%.+."D.97A >9ADjD@%E"D.977&55&A69%.+.&7D5AD!.5.D(%&&D{&D++&"D&7D5&E@A"D.I.D(A%& D@7A"D.ID.97%&5D@7A"@.>D.97mm997&L^ @7%&^&D5a^|z{znm$'%41!{‚na  >&ED(,5&6&7D)D@&>-9A>-9@L5(&^"&?E..6>"D&>5EA.&E@A%&A&A+97"D.97Aa >-9A>-9@L5D.97%&>@5>@9D(.7&4.7A&m nmE&5D^ 5.%%&7^sD9@6^{ƒƒ~_ 9A&7!&@,&@ &D 5a^ |zz|n .6>"D& 5d"D.I.D( -(5."A& &D 5 ">".D( %&   +.K&@ 5d mE&5D{ƒƒ~&D7,|zz~na>-9A>-9@L5D.97@(,E5&%97"A97"D.I.D(-(5."A&a&AD (,5&6&7D >-9A>-9@L5(& &7 @(>97A& E   >@ "j5! AE@ 5 [email protected]& ƒ} "& ?E.  E7 .6>"DAE@A">".D(97"9,(7.?E&a&DA>&"DA&@%(D.55(%7A5>@D.&D@.D7D%E@;5& %&%7A5D@7A.D.97(>.D-(5.9j6(A&7"-L6D&EA&a{7&>9AA'%&>A%&[email protected]& &7 >9A.D.97 ƒ} &D 7& >&ED %97" >A )D@& >-9A>-9@L5(&a  >&ED (,5&6&7D )D@& >-9A>-9@L5(& >@ 5 >@9D(.7& 4.7A& >}‚ m >}‚n AE@ 5 A(@.7& {ƒa &DD& 69%.+."D.97 ++&"D&559"5.AD.97%&&D.6>"D&A97@;5&%7A56DE@D.97%&A6."@9Am 97, &D5a^|z{}na 

 '%41!JR?- '$'3'-,/-23C31 43'-,,!**! !N!3JP?'+/*'3'-,241*!412 $-,3'-,2@ A/1"24**!1C!'-&'+'!3'-/&82'3KIJL@ 

 €~  ? +'/ /   # '$// K 0 GM )/ ' ,.*".//$*) 01(*.' 0 ' 0.)/$0$*) ,$0# '$*?( /)#5(01/<   &A6E5D.>5&A+97"D.97A%&&D{&7+97D%&A>@9D(.7&A&AA&7D.&55&A%7A5& %(I&59>>&6&7Da -&M 5 [email protected] 5 AE>>@&AA.97 %&  9E { &AD 5(D5&  5d(DD &6!@L977.@&mn9E%&EK39E@A>@'A57.AA7"&m{nmE4E%&D5a^|zzna-&M 55&IE@&9E5%@9A9>-.5&^&D{39E&7DE7@;5&>@.69@%.5%7A5&"L"5&"&55E5.@& &D5AE@I.&m@Ds ,,9^{ƒƒ_EAM"M4s>@%5.7,^|zz€!_ ,,9&D5a^{ƒƒ{nad&K>@&AA.97 &Di9E 5d"D.I.D( %& "&A >@9D(.7&A A97D 69%.+.(&A %7A %& 796!@&EK "7"&@Aa  >5E>@D %E D&6>A^  &D { 97D %&A >@9>@.(D(A 97"9,(7.?E&A >@9j>@95.+(@D.I&A &D >@9j 6.,@D9.@&A 6.A %7A "&@D.7A "A &D &7 +97"D.97 %E "97D&KD&^ &55&A >&EI&7D I9.@ %&A +97"D.97A7D.j>@95.+(@D.I&A&DAE>>@&AA&E@%&DE6&E@a  

  21 #'! 6'27 &#%!%&&! '( !%   &796!@&EA&A(DE%&A%("@.I&7D&D{"966&%&A>@9D9j97"9,'7&Aa&5 &AD !A( AE@ %&EK A>&"DAa @&6.'@&6&7D  &Di9E { A97D AE@&K>@.6(&A %7A %.++(@&7DA "7"&@A -E6.7Aa &EK.'6&6&7D 5&E@A +97"D.97A 695("E5.@&A +[email protected]&7D 5 >@95.+(@D.97&D5AE@I.&"&55E5.@&m 747&"-D^|z{zna   d&K>@&AA.97%&&D{&AD69%.+.(&%7A%&796!@&EKDL>&A%&"7"&@Aa&A >@9D(.7&A A97D AE@&K>@.6(&A %7A 5& "7"&@ %E A&.7 &D %E "9597 mEA&I." &D 5a^ |zz{_ 9@D-6 &D 5a^ |zzƒna  &AD (,5&6&7D AE@&K>@.6(& %7A 5&A "7"&@A %& 5 >@9ADD& m5@4&D5a^|z{}n&D%&5>&Ema a7,^-7,^9E^s-.^|z{|n&D>95LjE!.?E.D.7L5(&&D >-9A>-9@L5(& %7A %&A DE6&E@A %E A&.7 ma 7,^ .7^ s .E^ |zzn^ %E "9597 &D %E "&@I&Eam&L s .E^ |z{|n  >95LE!.?E.D.7D.97 %&  !59?E& A %(,@%D.97 >@ 5& >@9D(A96&a   7A 5& "7"&@ %E A&.7^  &D { A97D AE@&K>@.6(&A %7A }z  z† %&A DE6&E@AA&5975&A(DE%&Aa‚z†%&"&A"7"&@A&K>@.6&7D%&A6@?E&E@A-9@697EK"966& 5& @("&>D&E@ EK &AD@9,'7&A mWn &D  &D { A97D %&EK "9j@(,E5D&E@A  € D@7A"@.>D.977&5A!.&7"@"D(@.A(A%E@("&>D&E@EK&AD@9,'7&Am7%9-&D5a^{ƒƒƒ_E55&@j "&s5.^|zz‚_D7!&&D5a^|zz{_9@D-6&D5a^|zzƒna.7A.5AE@&K>@&AA.97%& &D{>9E@@.D+[email protected]&@5>@9,@&AA.97%E"7"&@%EA&.7I.5&E@A@;5&A%7A5 @(,E5D.97D@7A"@.>D.977&55&%&Wa&>5EA^.5(D(@("&66&7D%("@.D?E&+[email protected]&5& >AA,&%&5>-A& {%E"L"5&"&55E5.@&&D%97"5@(>5."D.97%&5r^&7,.AA7D "966&"9j"D.ID&E@%E+"D&E@%&D@7A"@.>D.97|{%7A%&A"&55E5&A%&"7"&@%EA&.7a 7A"&DD&6)6&(DE%&^5&AED&E@A97D697D@(?E&5&59"EA%&&AD6>5.+.(%7A%&A ("-7D.5597A&D%&A5.,7(&A%&"7"&@%EA&.7mME@&4&D5a^|z{|na  dED@&A %977(&A %& 5 5.DD(@DE@& %7A %&A 5.,7(&A %& "7"&@ %E A&.7 697D@&7D E7& AE@j &K>@&AA.97 %&  %7A 5&A 5.,7(&A !A5&A &7 "96>@.A97 %&A 5.,7(&A 5E6.75&Aa &DD& AE@&K>@&AA.97"97%E.D5@(,E5D.97%&6."@9A".!57D%&A"96>9A7DA%E@&69%&5,& %E"LD9A?E&5&DD&%d"D.7&ma7,&D5a^|z{|na.7A.^%7A5&A"7"&@A%EA&.79E%7A5&A 5.,7(&A %& "7"&@ %E A&.7^  &D { 97D E7& "D.97 >@9jDE69@5&  "EA& %& 5&E@ "D.I.D(%&@(,E5D&E@"9jD@7A"@.>D.977&56.A(,5&6&7D"EA&%&5&E@A.6>5."D.97A%7A 56DE@D.97%&A6."@9Aa   &67.'@&AA&MA.6.5.@&^&ADE7"9j"D.ID&E@%E@("&>D&E@EK7%@9,'7&A mnm5@4&D5a^|zz‚^|z{}n&D{.7D&@,.DI&"a966&W%7A5&A"7"&@A%E A&.7^39E&E7@;5&.6>9@D7D%7A5d(D.959,.&%E"7"&@%&5>@9ADD&aAE@&K>@&AA.97%& %7A%&A"7"&@A%&5>@9ADD&>9E@@.D39E&@E7@;5&%7A5>@9,@&AA.97DE69@5&&7 +[email protected]7D5d"D.I.D(%&a7AE769%'5&%&"7"&@%&5>@9ADD&&AD(,5&6&7D"9j "D.ID&E@%&5Xj"D(7.7&m5@4&D5a^|z{}na79ED@&^%7A5&"7"&@%&5>@9ADD& (D( @&D@9EI( +EA.977( I&" ~^ E7 +"D&E@ %& D@7A"@.>D.97 %& 5 +6.55& %&A  m|€ D@7A+9@6D.97jA>&".+."n%97D5+97"D.975@,&6&7D(D(%("@.D&%7A5&ADE6&E@A65.,7&A m 7&D5a^|zz‚na   7A5&"7"&@%E";597^>5EA.&E@A(I.%&7"&A@&5.&7D5&A+97"D.97A%&&D{ I&" 5 >@9,@&AA.97 DE69@5&a 55&A A97D AE@&K>@.6(&A %& +$97 "@9.AA7D& &7 AE.I7D 5&A AD%&A%&5>@9,@&AA.97%&A"7"&@A"959@&"DEK%EAD%&>95L>&EAD%&%(796&>E.A %(79"@".796&a & >5EA^ "966& 79EA 5dI97A IE >5EA D;D  &D { A97D %&A "9j "D.ID&E@A %& 5 Xj"D(7.7& &D D9ED& %(>5(D.97 %& "&A >@9D(.7&A "97%E.D  5d.7-.!.D.97 %&A

 €€ >@9D9j97"9,'7&A"jL"^L"5.7&{^"j3E7&D+@j{%97D5d&K>@&AA.97%(>&7%%&5Xj"D(7.7& m-.7 &D 5a^ |zzna  &AD 79@65&6&7D >-9A>-9@L5(& %7A %&A 5.,7(&A %& "7"&@A "959@&"DEK "96>@(&A  %&A 5.,7(&A "9@@&A>97%7D&A .AAE&A %& D.AAEA A.7Aa &DD& >-9A>-9@L5D.97 &AD AA9".(&  5 >@95.+(@D.97 "&55E5.@& ma 7,^ .7^ s .E^ |zz!na  >-9A>-9@L5D.97 %&  &7 >9A.D.97 ƒ} &7 @(>97A& E   +[email protected]& 5  >@ %.++(@&7DA 6("7.A6&A mI9.@ }a}a|n 59@A ?E& 5 >-9A>-9@L5D.97 AE@ 5&A >9A.D.97A €~ &D ~~€ >@ >}‚  4.7A& &7 @(>97A& E D@.D&6&7D >@ 9K5.>5D.7& +[email protected]& 5d>9>D9A& .7%E.D& >@ 5d9K5.>5D.7& m&L s .E^ |z{|na -&M 5 [email protected]^ .5  (D( 697D@( ?E&  &D {+9@6&7DE7&!9E"5&%&@(D@9"97D@;5&I&""jL"&D5&A6."@9A6.j{~}&D6.j {~^ %&EK 6."@9A 39E7D %&A @;5&A %& AE>>@&AA&E@ %& DE6&E@a  &D { >@D.".>&7DI&" ‚&D@9A-56DE@D.97%E"5EAD&@6.j{~}i{~a&A6."@9A @(,E5&7D  &D { %& 67.'@& %.@&"D& 9E .7%.@&"D& m494 &D 5a^ |z{|na 7A "& 69%'5&^5@(>@&AA.97%&&D{.6>"D&(,5&6&7DAE@5D@7A"@.>D.97%&A,'7&A"j L"^L"5.7&{&D"j3E7a  7AE7&5.,7(&%&"7"&@"959@&"D5^5d(?E.>&%&aa.E(,5&6&7D697D@(?E&  .7D&@,.D I&" 5 "569%E5.7& &D 5&A 6."@9DE!E5&A %7A 5& "LD9>5A6& %& "&55E5&A &7 6.,@D.97a&DD&.7D&@"D.97&AD7("&AA.@&56.,@D.97%&A"&55E5&Aa&>5EA^E7>&>D.%& .7-.!7D 5d.7D&@"D.97 &7D@&  &D 5 "569%E5.7& @(%E.D 5& 796!@& %& 6(DADA& %7A %&EK69%'5&A%d559j9EK(79,@&++&m€|zm-966&n&D~{[email protected]m a7,^ 9^7,^ s.E^|z{}na   5d.7I&@A&^,.D"966&"9j"D.ID&E@%&>}%7A5&"7"&@%E>9E697mD&A &D5a^|zzn&D5d&K>@&AA.97%&{&ADAA9".(&E7>@979AD."+I9@!5&%7A5&"7"&@%E A&.7 m9@D-6 &D 5a^ |zzƒna &A %977(&A "97D@%."D9.@&A AE,,'@&7D ?E& 5&A +97"D.97A %& &D%&{>&EI&7D)D@&%.++(@&7D&A&D%(>&7%7D&A%E"97D&KD&"&55E5.@&a   7 @(AE6(^ %7A 5& "7"&@ %E "9597^  &D { >@D.".>&7D  5 >@9,@&AA.97 DE69@5&6[email protected].@&6&7DEKD@I&@A%&5&E@A+97"D.97A%&"9j@(,E5D&E@AD@7A"@.>D.977&5A &D%&6DE@D.97%&A6."@9Aa  

 €   31"6 &% &'! #'!1 & + '(&.   5EA.&E@A (I.%&7"&A @&5.&7D 5 +97"D.97 %&  %7A 5 a &A (DE%&A 97D (D( @(5.A(&A>@5d(?E.>&%&aa.E%7A%&A69%'5&A%&"7"&@%E";597^E769%'5&9E5 &ADE7>@9"&AAEA@&5&I7D%7A5&"%@&%&5>@9,@&AA.97DE69@5&ma7,&D5a^|zz!na 7A"&69%'5&&AD>-9A>-9@L5(&>@"j5!&7>9A.D.97ƒ}&7@(>97A&E "& ?E..7%E.D5a&DD&>-9A>-9@L5D.97>&@6&D5@&59"5.AD.97%&5Xj"D(7.7&%7A5& 79LEa>-9A>-9@L5(&&6>)"-&5>-9A>-9@L5D.97%&5Xj"D(7.7&>@  }jXaXj "D(7.7&!A&7D&%E79LE7d&AD>A">!5&%d"D.I&@A&A,'7&A".!5&A>9E@.7%E.@&5 ma 7, &D 5a^ |zz€na &>&7%7D^ "&A @(AE5DDA A&6!5&7D )D@& %(>&7%7DA %&A 5.,7(&A ED.5.A(&AmDE"4&^ 9@A&A^s 9+677^|zz‚na  +9@6& >-9A>-9@L5(& %&  "97%E.D (,5&6&7D  5 D@7A"@.>D.97 %& 7.5{a  >-9A>-9@L5(&%.AA9".& {%E>@969D&E@%&7.5{&D.7D&@,.DI&"5&"96>5&K&%&%&j "(DL5D.97}`.j|iE&7@(>97A&E m@D&@&D5a^|z{zna  &>5EA^E7"@.!5&%9E!5&j-L!@.%&>[email protected]%d.%&7D.+.&@"966&E7"9j"D.ID&E@ %&6%}m@7&@&D5a^|zz~na&@(AE5DD5.AA&>&7A&@?E&>9E@@.D39E&@E7@;5& %7A5A.,75.AD.97%E,+jX?E.>&ED)D@&E7.7%E"D&E@%&5a

  

   !     ! !             

     

   





!   

   

   

'%41!KI?-,3'-, !N ,2*@ EFN' !*1!*-*'23'-, !*=C3#,',! ,2*!,-84@EFN/!1+!3* '22-'3'-, ! J 4/1-+-3!41 !,'*@EFN!234,-C3'53!41 !+ L@   €‚ 97"5EA.97   &D{A97D%&EK6&6!@&A%&5+6.55&%&A-(5."A&A69D.+a 966& 5&A ED@&A 6&6!@&A %& "&DD& +6.55&^ "&A >@9D(.7&A .7D&@I.&77&7D %7A %.++(@&7DA >@9"&AAEA%E6(D!95.A6&%&AA&D79D66&7D%7A5@(,E5D.97D@7A"@.>D.977&55&^5 6DE@D.97 %&A 6."@9A &D 5d(>.AA,&a &@D.7&A %& "&A +97"D.97A @&?E.'@&7D 5d"D.I.D( -(5."A& m%(>&7%7D& %& 5d-L%@95LA& %& 5dn &D %dED@&A 797a &A >@9D(.7&A A97D %97" 6E5D.+97"D.977&55&A^ >E.A?Ed&55&A A&@I&7D  5 +9.A %& >@9D(.7&A %>DD@."&A m@;5& %& "9j @(,E5D&E@D@7A"@.>D.977&5n&D%&69%.+."D&E@A%&AAD@E"DE@&Aj9Ej>@9D(.7&a 9E@"&DD&%&@7.'@&+97"D.97^5d"D.I.D(-(5."A&%(>&7%7D&%&5d&AD@&?E.A&a  AE@&K>@&AA.97%&&D{&AD9!A&@I(&%7A%&796!@&EK"7"&@AaE@5& >57 695("E5.@&^ &55&A ,.AA&7D &7 D7D ?E& "9j@(,E5D&E@ D@7A"@.>D.977&5 &D 5&E@A .6>5."D.97A%7A56DE@D.97%&A6."@9AA97D@&5.(&A5>@9,@&AA.97DE69@5&aI7D 6D-'A&^E"E7&>E!5."D.977dI.D6.D&7(I.%&7"&5&5.&7&7D@&5&@;5&%&&D{ %7A5d(>.AA,&&D5>@9,@&AA.97DE69@5&a                 

 €ƒ %0$!/

   &D{A97D%&EK-(5."A&A69D.+A?E.39E&7D%&796!@&EK@;5&A %7A 5& 6(D!95.A6& %& %.++(@&7DA A %&>E.A 5&E@ D@7A"@.>D.97 3EA?Ed 5&E@ (I&7DE&5 D@%E"D.97a&>E.A>5EA%&|z7A^5&E@A+97"D.97A695("E5.@&AA97D(DE%.(&A&7%(D.5Aa&A >@9D(.7&AD@'AA.6.5.@&AA97DAE@&K>@.6(&A%7A%&796!@&EK"7"&@A&D97D(D(>@9>9A(&A "966&%&A97"9,'7&A>9D&7D.&5Am 747&"-D^|z{zna&E@A">".D(A97"9,(7.?E&A97D(D( @&5.(&A  %&EK %& 5&E@A +97"D.97A 695("E5.@&A` {j 97"D.97A %& "9j@(,E5D&E@A D@7A"@.>D.977&5A %& >5EA.&E@A +"D&E@A %& D@7A"@.>D.97 39E7D %&A @;5&A "5(A %7A 5 >D-959,.& "7"(@&EA& m&K` @W^ ^ Xj"D(7.7& mE55&@j"&^ |z{}na |j ;5&A %7A 5 6DE@D.97 %&A 6."@9A^ &7 >@D."E5.&@ %7A 5& "7"&@ %E A&.7 ma 7, &D 5a^ |z{|na 5EA.&E@A D@IEK 6&DD&7D &7 (I.%&7"& 5& @;5& %&  &D { %7A 5 @(,E5D.97 %& 5d(>.AA,& 5D&@7D.+a @ "& >@9"&AAEA >>@0D %(@(,E5( EK "9E@A %& 5 >@9,@&AA.97 DE69@5&aE"9E@A%&697D@I.5%&D-'A&^3&6&AE.A+9"5.A(AE@5@(,E5D.97%&5d(>.AA,& 5D&@7D.+>@&D{>&7%7D5>@9,@&AA.97DE69@5&a  5&AD!.&7(D!5.?E&&D{@(,E5&7D5D@7A"@.>D.97^5d(>.AA,&5D&@7D.+&D 5 6DE@D.97 %&A 6."@9Aa &A %.++(@&7DA 6("7.A6&A A97D 5&A !A&A %E "97D@;5& %& 5d&K>@&AA.97%&A,'7&A&D5&E@@(,E5D.97&AD"@.D.?E&>9E@5%.++(@&7".D.97"&55E5.@&6.A EAA.>9E@5>@9,@&AA.97DE69@5&a.7A.%7AE7%&EK.'6&D&6>A^3&6&AE.A.7D(@&AA(5 "99@%.7D.97&D5@(,E5D.97%&5D@7A"@.>D.97^%&5d(>.AA,&5D&@7D.+&D56DE@D.97%&A 6."@9A >@ 5&A  -(5."A&A  &D { %7A E7 69%'5& %& D@7A.D.97 >.D-(5.9j (A&7"-L6D&EA&.7%E.D>@5& jXa  





 z  /1'00/

 







 .*

      

 (  











  &A@(+(@&7"&A".D(&A%7A"&D@D."5&A97D@(>&@D9@.(&A5+.7%&5d@D."5&a  

 {  7A"&D@D."5&^3d.D9ED%d!9@%6.A&7(I.%&7"&5AE@&K>@&AA.97%&%K{%7A5&A "&55E5&A6(DADD.?E&A~{&DA9E5.,7(5&@;5&%&%K{&D%&A97>@59,E&%K%7A5 6.,@D.97 &D 5d.7IA.97 %& "&A "&55E5&Aa @ 5 AE.D&^ 3& 6& AE.A .7D(@&AA( E @;5& %& "&A >@9D(.7&A%7A5@(,E5D.97%&I@.7DA%d(>.AA,&A>9EI7D39E&@E7@;5&%7A5&>-(79DL>& 6.,@D9.@& &D .7IA.+a d. >E 697D@&@ ?E& %K &D %K{ @(,E5&7D 5d(>.AA,& %& 5d-.AD97& 6"@9 |{a &A %&EK I@.7DA %d-.AD97& 97D E7 &++&D 9>>9A( AE@ 5d&K>@&AA.97 %& ,'7&A .6>5.?E(A%7A56.,@D.97&D5d.7IA.97"966&5&,'7&}a97D@I.5"d&AD+9"5.A(AE@ 5&@;5&%&%&%K&D%K{%7A5d(>.AA,&%&6"@9 |{a&DAE@5d75LA&+97"D.977&55&%&A %&EKI@.7DA%d(>.AA,&ad75LA&%&5d&K>@&AA.97%&ADE6&E@A(D(+.D&&7"955!9@D.97a 

 % !'

 

% !'

 

"!" "!!   $ !  &'            

# #

       - "*! !1#%4*3'-, 4/&#,-38/!+'%13-'1!!3',52'$/1*!2 '$$#1!,3!2'2-$-1+!2 ! *A&'23-,!+1- KJ@

 | ARTICLES

Splicing switch of an epigenetic regulator by RNA promotes tumor-cell invasiveness

Etienne Dardenne1,5, Sandra Pierredon2,3,5, Keltouma Driouch4, Lise Gratadou1, Magali Lacroix-Triki2,3, Micaela Polay Espinoza1, Eleonora Zonta1, Sophie Germann1, Hussein Mortada1, Jean-Philippe Villemin1, Martin Dutertre1, Rosette Lidereau4, Stéphan Vagner2,3 & Didier Auboeuf1

Both epigenetic and splicing regulation contribute to tumor progression, but the potential links between these two levels of - expression regulation in pathogenesis are not well understood. Here, we report that the mouse and human RNA helicases Ddx17 and Ddx5 contribute to tumor-cell invasiveness by regulating alternative splicing of several DNA- and chromatin-binding factors, including the macroH2A1 histone. We show that macroH2A1 splicing isoforms differentially regulate the transcription of a set of involved in redox metabolism. In particular, the SOD3 gene that encodes the extracellular superoxide dismutase and plays a part in cell migration is regulated in an opposite manner by macroH2A1 splicing isoforms. These findings reveal a new regulatory pathway in which splicing factors control the expression of histone variant isoforms that in turn drive a transcription program to switch tumor cells to an invasive phenotype.

Modifications of gene-expression programs resulting from altera- factors control the splicing programs and thus the alternative splicing tions of epigenetic, transcription and pre-mRNA–splicing regula- of gene networks. Although many differences at the splicing level have tion largely contribute to the development of various cancers. At been described for cancer cells compared to normal cells, one major the epigenetic level, a plethora of chromatin alterations affecting challenge is to better understand how splicing programs and splic- canonical histone-tail modifications appear to be involved in tumor ing factors may contribute to specific cancer-associated phenotypes, malignancy1–3. There is also an increasing interest in analyzing including cellular invasion and migration. Indeed, deciphering the the role in cancer of histone variants of the H2A family, including genetic programs involved in cancer progression will allow researchers H2A.Z, H2AX, macroH2A1 and macroH2A2. Indeed, the exchange to better understand how cancer cells can evolve and form metastases, © 2012 Nature America, Inc. All rights reserved. America, Inc. © 2012 Nature of canonical histones for histone variants is one of the most extensive which are the main cause of death in cancer. Another major perspec- epigenetic regulation events that affects genomic functions such as tive is to determine whether and how cross-talk between the different cell-cycle control, response to DNA damage, heterochromatin silenc- layers of gene-expression regulation, such as epigenetic and splicing npg ing and transcriptional regulation3–8. Changes in the expression of programs, contribute to tumor progression. H2A.Z, H2AX or macroH2A1 histones in cancer can affect DNA The 4T1 mouse model of tumor progression comprises four repair, thereby leading to alterations of segregation or syngeneic tumor cell lines that can give rise to primary tumors with cell proliferation3–7. Understanding the role in physiological or patho- a spectrum of metastatic phenotypes when implanted into mouse logical states of histone variants, whose diversity can also be increased mammary fat pads10,16,17. Using splicing-sensitive microarrays to ana- by alternative splicing, remains a major issue. lyze the 4T1 mouse model, we and others have reported that several Numerous studies have also reported changes in the expression of alternative-splicing events are associated with the ability of primary splicing variants in cancers9–12. Alternative splicing is the main mech- tumors to disseminate10,17. Many of the splicing variants differen- anism responsible for increasing the diversity of the proteome. Indeed, tially expressed between primary tumors that do or do not develop 90% of human genes generate several splicing variants, producing pro- into metastases are also expressed in normal tissues and have been tein isoforms that can have different and sometimes opposing biological conserved during evolution, which suggests that they are under the activities11–14. Differential selection of alternatively spliced exons control of conserved splicing regulators10. relies on the recognition of the splicing sites by the spliceosome and To characterize the role of splicing factors in tumor progression, on splicing-regulatory sequences, located within exons and introns, we focus on the Ddx17 RNA (also known as p72) that we that are recognized by splicing factors13–15. In a way akin to how tran- observed to be overexpressed in invasive tumor cells and that we scriptional programs are controlled by transcription factors, splicing show to contribute to tumor-cell migration and invasion together

1Institut National de la Santé et de la Recherche Médicale, U1052, Centre de Recherche en Cancérologie de Lyon, Lyon, France. 2Institut National de la Santé et de la Recherche Médicale, U563, Institut Claudius Regaud, Toulouse, France. 3Institut National de la Santé et de la Recherche Médicale, U981, Institut de Cancérologie Gustave Roussy, Villejuif, France. 4Laboratory of Oncogenetics, Institut Curie–Hôpital René Huguenin, St-Cloud, France. 5These authors contributed equally to this work. Correspondence should be addressed to S.V. ([email protected]) or D.A. ([email protected]). Received 9 November 2011; accepted 24 August 2012; published online 30 September 2012; doi:10.1038/nsmb.2390

NATURE STRUCTURAL & MOLECULAR BIOLOGY VOLUME 19 NUMBER 11 NOVEMBER 2012 } 1139 ARTICLES

Figure 1 The Ddx5 and Ddx17 RNA helicases a b c 100 4T1 siCtrl contribute to cell migration and invasion. siDdx5 siDdx17 (a) Relative expression level of Ddx17 mRNA 67NR 168F 4T07 4T1 4T1 siDdx5-17 measured by RT-qPCR in mouse primary tumors Metastasis – – + + 75 3 with different metastatic capabilities, * Size ** Ddx5 Ddx17 Ddx5-17 * siCtrl si si si (kDa) as indicated. (b) Western blot analysis of Ddx5, 50 2 90 *** *** Ddx17 and B-actin, 48 h after the transfection Ddx17 *** of mouse 4T1 cells with a control siRNA (siCtrl) level Ddx5 25 or siRNAs targeting Ddx5 (siDdx5), Ddx17 1 60 Relative cell number *** (siDdx17) or both Ddx5 and Ddx17 (siDdx5-17). -actin Relative expression 40 *** (c) Relative number of migrating and invasive Migration Invasion 4T1 cells after transfection with siCtrl, siDdx5, siDdx17 or siDdx5-17. Data are shown as mean number of migrating and invasive cells transfected with each siRNA relative to those of cells transfected with the control siRNA. In a and c, histograms represent the average of at least three independent experiments. Error bars, s.e.m.; *P < 0.05; **P < 0.01; ***P < 0.001 (t test).

with its paralog Ddx5 (p68). Ddx17 and Ddx5 are transcriptional To understand how Ddx5 and Ddx17 regulate cell migration and co-regulators of the estrogen receptor and the p53 transcriptional invasion, we used Affymetrix exon arrays to analyze the transcriptome factor18–21 and are also splicing regulators known to control alterna- of 4T1 cells at the exon level, following siRNA-mediated depletion of tive splicing of the genes encoding CD44, H-ras and tau18,19,22–24. Our Ddx5 and Ddx17. There were 1,136 genes predicted to be affected at genome-wide analysis at the exon level revealed that Ddx5 and Ddx17 the global gene-expression level (Supplementary Table 1) and 1,325 regulate an alternative-splicing network that affects several DNA- genes predicted to be affected at the exon level (Supplementary and chromatin-binding factors, including the macroH2A1 histone Table 2). The effect of Ddx5 and Ddx17 depletion was validated gene (mH2A1; also called H2AFY). We demonstrate that the mH2A1 by RT-PCR for 31 out of 42 splicing variants (74% validation rate) splicing isoforms are differentially expressed in primary tumors that (Supplementary Fig. 2). either give rise to metastases or do not. Furthermore, these isoforms analysis through the Panther website25 indicated differentially regulate breast cancer–cell invasiveness through the that genes regulated at the splicing level as compared to genes regu- transcriptional regulation of genes involved in redox metabolism, lated at the global level were enriched in ‘binding activity’ (P = 2.8 × including SOD3. Thus, our work reveals a new cascade of altered 10−13), in particular ‘nucleic acid binding’ (P = 2.9 × 10−11), ‘RNA gene-expression events participating in tumor progression, wherein binding’ (P = 9 × 10−8), ‘chromatin binding’ (P = 1.9 × 10−5) and splicing factors (Ddx5 and Ddx17) control the alternative splicing ‘DNA binding’ (P = 5 × 10−4). This observation suggests that the of an epigenetic factor (mH2A1), leading in turn to transcriptional set of alternative exons that are co-regulated by Ddx5 and Ddx17 alterations of genes involved in a specific cellular program. are enriched in functionally related genes and that Ddx5 and Ddx17 regulate the alternative splicing of DNA- and chromatin-binding RESULTS factors that could in turn affect the transcription of genes involved Ddx5 and Ddx17 contribute to cell migration and invasion in tumor progression. While analyzing the transcriptome of the 4T1 mouse model of mam- © 2012 Nature America, Inc. All rights reserved. America, Inc. © 2012 Nature mary tumor progression10, we observed that Ddx17 was overexpressed Ddx5 and Ddx17 control mH2A1 alternative splicing in primary tumors that are more likely to give rise to metastases (4T1 To test this hypothesis, we focused on the mH2A1 histone gene cells) or micrometastases (4TO7 cells) as compared to tumors that are because recent reports have described a role for it in cancer4,7,8. npg not able to disseminate (67NR cells) or that disseminate to the lymph The mouse mH2A1 gene generates two splicing isoforms (mH2A1.1 nodes (168FARN cells). We validated this result by reverse-trancription and mH2A1.2) through the use of two mutually exclusive exons quantitative PCR (RT-qPCR) on RNAs extracted from mouse tumors (Fig. 2a). As predicted by the exon-array analysis, depletion of Ddx5 (Fig. 1a) and also showed that Ddx17 was overexpressed at both the and Ddx17 increased the level of the mH2A1.1 isoform and decreased RNA and the levels in the invasive 4T1 cell line as compared that of the mH2A1.2 isoform but had no effect on the global mH2A1 to the noninvasive 67NR cell line (Supplementary Fig. 1a). To test level (Fig. 2b and Supplementary Fig. 3a,b). These the potential role of Ddx17 in tumor-cell invasiveness, we knocked results were confirmed at the protein level (Fig. 2c) and were extended down its expression by using a specific siRNA (Fig. 1b). Because Ddx5 in human HeLa cells (Fig. 2d and Supplementary Fig. 4). and Ddx17 often have redundant functions18,19 and although Ddx5 To perform rescue experiments, siRNAs targeting the untrans- did not seem to be misregulated (Supplementary Fig. 1a), we also lated regions (UTRs) of the endogenous Ddx5 and Ddx17 mRNAs used siRNAs against either Ddx5 or against a conserved region that were designed. Transfection of these siRNAs led to an increase in the targeted both Ddx5 and Ddx17 (Fig. 1b). Both cell migration and inva- ratio of mH2A1.1 to mH2A1.2 (Fig. 2e, lane 2 compared to lane 1). sion were inhibited by Ddx5 or Ddx17 depletion, and this inhibition Furthermore, cell transfection with either Ddx5 or Ddx17 expres- was enhanced when both Ddx5 and Ddx17 were depleted (Fig. 1c). sion vectors (to produce mRNAs without UTRs that are therefore Similar effects were obtained with either the single siRNA target- refractory to the UTR-targeting siRNAs) rescued to a large extent ing both Ddx5 and Ddx17 or a mixture of siRNAs separately target- the splicing pattern generated by endogenous Ddx5 and Ddx17 pro- ing Ddx5 and Ddx17 in 4T1 cells, whereas depletion of Ddx5 and tein depletion (Fig. 2e, lane 2 compared to lane 3 and to lane 5). Ddx17 in the noninvasive 67NR cells had no effect (Supplementary Even though overexpression of Ddx5 or Ddx17 did not fully rescue Fig. 1b,c). These results demonstrate that Ddx5 and Ddx17 have a the splicing pattern induced by Ddx5 and Ddx17 depletion, our data role in cell invasiveness because depletion of Ddx5 and/or Ddx17 suggest that Ddx5 and Ddx17 have similar functions, as depletion neither increased cell apoptosis nor significantly decreased cell-cycle of either Ddx5 or Ddx17 affected mH2A1 splicing (Supplementary progression (Supplementary Fig. 1d,e). Fig. 4c). Notably, overexpressing mutated Ddx5 or Ddx17

1140 VOLUME 19 NUMBER 11 NOVEMBER 2012 NATURE STRUCTURAL & MOLECULAR~ BIOLOGY ARTICLES

HeLa Size a b c d e 123456(kDa) mH2A1.2 4T1 4T1 HeLa Ddx5 * 60 88 87 ATG Stop Ddx17 80 Ddx5-17Size (bp) Ddx5-17Size (kDa) Ddx5-17Size (kDa) 1.2 siCtrl si siCtrl si siCtrl si I II III IV V VIII IX X 130 (1.1) 16 mH2A1.1 40 Ddx5 60 1.1 87, 88 (1.2) 12 NS NS mH2A1.2 40 Ddx17 * 80 36 (1.1) 8 130 36 mH2A1.1 40 * mH2A1.1 (%) 20.6 71.4 mH2A1 40 ** mH2A1.2 40 4

mH2A1.1 Ratio 1.1:1.2 0.5 0.4 -actin 40 mH2A1.2 (%) 79.4 28.6 mH2A1 40 siUTRDdx5-Ddx17 –+++++ Figure 2 The Ddx5 and Ddx17 RNA 0.5 0.4 -actin 40 Ddx5 WT ––+––– helicases control alternative splicing of Ddx5 MUT –––+–– DNA- and chromatin-binding factors. (a) Schematic representation of the Ddx17 WT ––––+– Ddx17 MUT –––––+ structure of the mH2A1 gene, which contains two mutually exclusive HeLa 4T1 exons (1.1 and 1.2). The expected sizes of Msp1-digested PCR products f -globin FN FN -globin are indicated to distinguish mH2A1.1 and mH2A1.2 splicing variants. SV40 Ndel Ddx5-17 Ddx5-17 (b) PCR products, digested with the Msp1 restriction enzyme, Ex2 Ex3 Ex3 siCtrl si siCtrl si corresponding to mH2A1.1 and mH2A1.2 splicing variants as indicated, + 1.1 + 1.1 1.1 300 – 1.1 300 – 1.1 obtained from 4T1 cells transfected with siCtrl or siDdx5-17. Relative 1.2 + 1.2 + 1.2 – 1.2 expression levels (% of total o s.e.m.) of the mH2A1 splicing variants in each 200 – 1.2 200 experimental condition are indicated. (c) Western blot analysis of mH2A1.1, mH2A1.2, mH2A1 and B-actin, 48 h after the transfection of mouse 4T1 cells with siCtrl or siDdx5-17. (d) Western blot analysis of Ddx5, Ddx17, mH2A1.1, mH2A1.2, mH2A1 and B-actin, 48 h after transfection of human HeLa cells with siCtrl or siDdx5-17. The asterisk indicates p82, an isoform generated from an alternative ATG start site. (e) Western blot analysis of Ddx5 and Ddx17 after transfection of HeLa cells with a control siRNA or an siRNA targeting Ddx5 and Ddx17 UTRs (siUTRDdx5-Ddx17), together with control expression vectors or wild-type Ddx5 (Ddx5 WT), wild-type Ddx17 (Ddx17 WT), mutated Ddx5 (Ddx5 MUT) or mutated-Ddx17 (Ddx17 MUT) expression vectors (top). Tagged Ddx5 protein is marked by an asterisk. Relative expression levels of mH2A1.1 and mH2A1.2 mRNA, as determined by RT-qPCR in the same experimental conditions as described above, are indicated. Histograms represent the average of three independent experiments corresponding to the fold change of the mH2A1.1-to-mH2A1.2 ratio. Error bars, s.e.m.; *P < 0.05; **P < 0.01; NS, not significant (t test). (f) Schematic representation of the mH2A1 minigenes (left). Ex2, exon 2; Ex3, exon 3; FN, fibronectin. RT-PCR analysis using primers in the minigene exons indicated by arrows on the left panel, in HeLa and 4T1 cells transfected with siCtrl or siDdx5-17 (right). Inclusion or exclusion of the 1.1 and 1.2 exon are shown as +1.1 and +1.2, or −1.1 and −1.2, respectively.

that lacked the RNA helicase activity did not significantly rescue mH2A1.1 isoform was expressed at a higher level in the noninvasive the mH2A1 splicing pattern (Fig. 2e, lane 4 compared to lane 3 and 67NR cells than in the invasive 4T1 cells, whereas the mH2A1.2 iso- lane 6 compared to lane 5). form was expressed at a higher level in 4T1 cells, both at the RNA To directly assay the effect of Ddx5 and Ddx17 on mH2A1 alterna- and protein levels (Fig. 3a,b). Whereas 67NR cells expressed similar tive splicing, exons 1.1 or 1.2 and ~200 nucleotides of their surround- levels of mH2A1.1 and mH2A1.2 isoforms, 4T1 cells expressed ing intronic sequences were cloned into a reporter minigene (Fig. 2f). much less mH2A1.1 than mH2A1.2 (Fig. 3a,b and Supplementary As expected, Ddx5 and Ddx17 depletion increased exon 1.1 inclu- Fig. 3c,d). This result was confirmed in tumor samples, as a lower sion in HeLa and 4T1 cells transfected with the minigene. Ddx5 and ratio of mH2A1.1 to mH2A1.2 was observed in mouse primary tumors © 2012 Nature America, Inc. All rights reserved. America, Inc. © 2012 Nature Ddx17 depletion did not increase exon 1.2 skipping (Fig. 2f ), which that gave rise to metastasis (Supplementary Fig. 3e). suggested that sequences not included in the minigene are required Supporting the hypothesis that the mH2A1 splicing isoforms have to regulate exon 1.2. Alternatively, Ddx5 and Ddx17 may regulate different roles in cell migration and invasion, specific depletion of the npg exon 1.1 splicing, which in turn could affect exon 1.2 splicing. These mH2A1.2 isoform in the invasive 4T1 cells (Supplementary Fig. 3c) results demonstrate that Ddx5 and Ddx17 regulate mH2A1 alterna- inhibited cell migration and invasion (Fig. 3c), whereas the depletion of tive splicing and that the RNA helicase activity of Ddx5 and Ddx17 is the mH2A1.1 isoform in the noninvasive 67NR cells (Supplementary required for their effects on splicing. Fig. 3c) increased cell migration and invasion (Fig. 3d). These results, which were confirmed by using different siRNAs (Supplementary mH2A1 alternative splicing participates in cell invasiveness Fig. 5a), demonstrate that the mH2A1.2 isoform favored breast We next investigated a possible role of mH2A1 splicing isoforms cancer–cell invasiveness, whereas the mH2A1.1 isoform had the oppo- in tumor-cell invasiveness. In this context, we observed that the site effect. Transfection of 4T1 cells with siRNA targeting the mH2A1.1

Figure 3 Alternative splicing of the macroH2A1 a b c siCtrl si1.2 d siCtrl si1.1 histone variant is involved in tumor progression. Size Size 4T1 67NR (a) PCR products, digested with the Msp1 4T1 67NR (bp) 4T1 67NR (KDa) 100 200 *** 130 (1.1) restriction enzyme, corresponding to mH2A1.1 mH2A1.1 40 150 *** 87, 88 (1.2) 75 *** *** mH2A1.2 40 and mH2A1.2 splicing variants. Template is 50 100 RNA isolated from 67NR or 4T1 cells. 36 (1.1) 40 mH2A1 25 50 Relative expression levels (% of total o s.e.m.) mH2A1.1 (%) 23.2 51.6 -actin Relative cell number of the mH2A1 splicing variants in each  2.9  6.3 40 Relative cell number experimental condition are indicated. mH2A1.2 (%) 76.8 48.4  2.9  6.3 Migration Invasion Migration Invasion (b) Western blot analysis of mH2A1.1, mH2A1.2, mH2A1 and B-actin, expressed in 4T1 or 67NR cells. (c) Relative number of migrating and invasive 4T1 cells after transfection with siCtrl or an siRNA targeting mH2A1.2 (si1.2). (d) Relative number of migrating and invasive 67NR cells after transfection with siCtrl or a siRNA targeting mH2A1.1 (si1.1). In c and d, histograms represent the average of at least three independent experiments. Error bars, s.e.m.; ***P < 0.001 (t test).

NATURE STRUCTURAL & MOLECULAR BIOLOGY VOLUME 19 NUMBER 11 NOVEMBER 2012  1141 ARTICLES

Figure 4 Alternative splicing of the macroH2A1 histone variant a Size b contributes to the Ddx5-Ddx17 effect on cell migration and invasion. 4T1 (kDa) Ddx5 60 siCtrl (a) Western blot analysis of Ddx5, Ddx17, mH2A1.1, mH2A1.2, mH2A1 90 Ddx17 siDdx5-17 and B-actin, 48 h after transfection of mouse 4T1 cells with siCtrl, siDdx5-17 + si1.1 40 siDdx5-17, si1.1 or si1.2 isoforms (top). Relative expression levels of mH2A1.1 siDdx5-17 + si1.2 40 mH2A1.1 and mH2A1.2 measured by RT-qPCR in the same experimental mH2A1.2 4T1 conditions as described above are indicated (bottom). Histograms mH2A1 40 100 represent the average of three independent experiments corresponding -actin 40 to the fold change of the mH2A1.1 to mH2A.2 mRNA ratio. (b) Relative siDdx5-17 –+++ 75 number of migrating and invasive 4T1 cells after transfection with siCtrl, si1.1 ––+– siDdx5-17 and either si1.1 or si1.2. (c) Motility of 4T1 cells transfected si1.2 –––+ 50 ** with siCtrl or with siDdx5-17 alone or together with si1.1, measured 40 *** ** by the wound healing assay at 8, 16 or 24 h. Histograms represent the 30 20 25 average of at least three independent experiments. Error bars, s.e.m.; *** Relative cell number **P < 0.01; ***P < 0.001 (t test). 10 Ratio 1.1:1.2 1234 Migration Invasion

4T1 siCtrl isoform had no effect, as the mH2A1.1 isoform is scarcely expressed c siDdx5-17 100 siDdx5-17 + si1.1 in 4T1 cells, and depletion of the mH2A1.2 isoform had no effect in Ddx5-171.1 75 the noninvasive 67NR cells (Supplementary Fig. 5b). Ddx5-17 si siCtrl si + si *** To further test whether mH2A1 alternative-splicing regulation 50 *** contributes to the effect of Ddx5 and Ddx17 on tumor cell invasive- 0 h 25 ** ness, Ddx5- and Ddx17-depleted 4T1 cells were transfected with wound closure) 24 h siRNAs that specifically targeted each of the mH2A1 splicing variants Motility (percentage of (Fig. 4a). Notably, depletion of the mH2A1.1 isoform but not the 8 h 16 h 24 h mH2A1.2 isoform reduced the inhibitory effect of Ddx5 and Ddx17 depletion on cell invasion and migration (Fig. 4b). This result was Ddx5- and Ddx17-dependent regulation of the invasive phenotype. further confirmed by the depletion of the mH2A1.1 isoform in Ddx5- For this purpose, we compared the transcriptome of 4T1 cells that and Ddx17-depleted cells, which reduced the inhibitory effect of Ddx5 had been depleted of either mH2A1.1 or mH2A1.2 in the context of and Ddx17 depletion on cell motility (Fig. 4c, siDdx5-17 compared to Ddx5 and Ddx17 depletion. Manipulating mH2A1 isoform expression siDdx5-17 + si1.1). These results support a model in which mH2A1 had only weak effects, with changes in the expression levels of only alternative splicing mediates, at least in part, the inhibition by Ddx5 53 genes by more than 1.5 fold (Fig. 5a and Supplementary Table 3). and Ddx17 depletion of tumor-cell invasiveness. This was anticipated because several reports have indicated that mH2A1 has only a slight impact on steady-state levels of gene expression26–29. mH2A1 splicing isoforms control redox metabolism genes Reducing the threshold to 1.2-fold revealed 633 genes that were differ- We next performed a high-throughput gene-expression analysis to entially regulated when comparing mH2A1.2- to mH2A1.1-depleted identify genes whose expression might be differentially regulated cells (Fig. 5a and Supplementary Table 3). We validated by RT-qPCR by the mH2A1 splicing isoforms and thus could be involved in the 17 of the 21 tested genes that had fold changes ranging from 1.2 to © 2012 Nature America, Inc. All rights reserved. America, Inc. © 2012 Nature 2.6 (Supplementary Fig. 6a and Supplementary Table 3). Among a siDdx5-17 + si1.2 vs. siDdx5-17 + si1.1 the 633 genes, 501 genes were upregulated and 132 genes were down- regulated in mH2A1.2-depleted cells compared to mH2A1.1-depleted Upregulation 37 16 npg Downregulation 464 (≥1.5) 116 (≥1.5) cells, which suggests that the mH2A1.1 and mH2A1.2 isoforms have (≥1.2) (≥1.2) a differential impact on gene transcriptional activity. mH2A1 b c 67NR vs. 4T1 Of the genes that were differentially regulated by splic- Genes regulated by mH2A1 200 *** ing isoforms, 45 were also affected by Ddx5 and Ddx17 depletion 100 and were also differentially expressed in 67NR cells as compared to 15 *** 123 45 102 4T1 cells (Fig. 5b), and more than one-third of these are involved in redox metabolism (Supplementary Table 3). Notably, meta- 168 genes differentially 147 genes 10 regulated between regulated by static cells may come from cancer cells that have avoided oxidative 67NR and 4T1 cells Ddx5 and Ddx17 SOD3 5 HAO1 ** **

RFESD Relative expression level 1 Figure 5 Regulation of mH2A1 splicing isoforms by Ddx5 and Ddx17 GFOD1 affects regulation of genes involved in redox metabolism. (a) Number

SOD3 HAO1 of genes regulated at the global expression level, as determined by RFESDGFOD1 comparing 4T1 cells transfected with siDdx5-17 and with si1.2 or si1.1. SOD3 HAO1 RFESD GFOD1 d (b) Venn diagram corresponding to mH2A1-regulated genes that were ** ** 3 ** 2 4 ** ** regulated by Ddx5 and Ddx17 and differentially expressed in 67NR as 2 ** * * compared to 4T1 cells. (c) Relative expression level of SOD3, HAO1, 3 2 RFESD and GFOD1 mRNAs, as determined by RT-qPCR, comparing 1 2 1 nonmetastatic 67NR cells with metastatic 4T1 cells. (d) Relative

Relative 1 expression level of SOD3, HAO1, RFESD and GFOD1 mRNAs measured 1

expression level by RT-qPCR in 4T1 cells transfected with siCtrl or siDdx5-17 alone or in combination with either si1.1 or si1.2. Histograms represent the average siDdx5-17 – +++ –+++ –+++ –+++ si1.1 – –+– ––+– ––+– ––+– of at least three independent experiments. Error bars, s.e.m.; *P < 0.05; si1.2 – ––+ –––+ –––+ –––+ **P < 0.01; ***P < 0.001 (t test).

1142 VOLUME 19 NUMBER 11 NOVEMBER 2012 NATURE STRUCTURAL & MOLECULAR€ BIOLOGY ARTICLES

Figure 6 Ddx5 and Ddx17–regulated SOD3 a b c d expression is involved in cell migration and siCtrl siDdx5-17 + siCtrl Ctrl siDdx5-17 + Ctrl invasion. (a) Relative number of migrating siSOD3 siDdx5-17 + siSOD3 DMNQ siDdx5-17 + DMNQ and invasive 67NR cells after transfection 67NR 4T1 67NR 4T1 with siCtrl or siRNA against SOD3 (siSOD3), 300 400 400 300 as determined by comparing with the number ** *** of cells transfected with siCtrl that migrated *** ** 300 and invaded. (b) Relative number of migrating 300 200 200 and invasive 4T1 cells after transfection with *** siDdx5-17 alone or in combination with siSOD3. 200 *** 200 ***

(c) Relative number of migrating and invasive 100 *** 100 67NR cells in the absence or presence of DMNQ. 100 100 Relative cell number Relative cell number Relative cell number (d) Relative number of migrating and invasive Relative cell number 4T1 cells in the absence or presence of DMNQ. Histograms represent the average of at least three n independent experiments. Error bars, s.e.m.; Invasion Invasio Invasion Invasion **P < 0.01; ***P < 0.001 (t test). Migration Migration Migration Migration

damage from excess reactive oxygen species (ROS) in the primary SOD3 regulation by mH2A1 contributes to cell invasiveness tumor site30,31. We therefore focused on four genes involved in To directly examine the impact on tumor-cell invasiveness of Ddx5- redox metabolism: the extracellular superoxide dismutase (SOD3), and Ddx17-mediated regulation of mH2A1 alternative splicing and hydroxyacid oxidase 1 (HAO1), Rieske (Fe-S) domain containing the downstream mH2A1-dependent transcriptional regulation, we (RFESD) and glucose-fructose oxidoreductase domain containing 1 focused on SOD3, which converts extracellular superoxide radicals GFOD1 •– 32–35 ( ). We observed by RT-qPCR that 67NR cells that expressed a ([O2] ) into hydrogen peroxide (H2O2) and is known to be high mH2A1.1 to mH2A1.2 ratio as compared to metastatic 4T1 cells involved in invasion32–35. As shown in Figure 6a, SOD3 depletion also expressed higher levels of SOD3, HAO1, RFESD and GFOD1 (Supplementary Fig. 5d) indeed led to an increased invasiveness mRNAs than metastatic 4T1 cells (Fig. 5c). Furthermore, Ddx5 of the noninvasive 67NR cells. Notably, transfection of Ddx5- and and Ddx17 depletion increased SOD3, HAO1, RFESD and GFOD1 Ddx17-depleted 4T1 cells with an siRNA targeting SOD3 increased mRNA expression (Fig. 5d). Finally, depletion of mH2A1.1, but not cell migration and invasion (Fig. 6b). This result suggests that Ddx5 of mH2A1.2, abrogated the increase mediated by Ddx5 and Ddx17 and Ddx17 depletion increased the mH2A1.1 isoform level (Fig. 2), depletion (Fig. 5d). Altogether, these data show that mH2A1 splicing which in turn increased SOD3 expression level (Fig. 5d) and thereby isoforms mediate a part of Ddx5- and Ddx17-induced regulation reduced cell migration and invasion (Fig. 6b). Further supporting of gene expression in 4T1 cells, including that of genes involved in this model, directly adding SOD3 protein to 4T1 cell-culture medium redox metabolism. inhibited cell migration (Supplementary Fig. 5e), whereas adding 2,3-dimethoxy-1,4-naphthoquinone (DMNQ), a redox-cycling agent that induces superoxide anion formation and mimics SOD3 deple- tion, increased the migration and invasion capability of 67NR cells a c ABC SOD3 mRNA pA and Ddx5- and Ddx17-depleted 4T1 cells (Fig. 6c,d).

© 2012 Nature America, Inc. All rights reserved. America, Inc. © 2012 Nature 4T1 67NR 0242 4,142 5,706 The differential effects of the mH2A1 splicing variants on the 2 *** 2 67NR 4TI gene-expression level could be explained if the two isoforms act on ABCdifferent genes or if they act on the same genes but have opposite 1 1 1.00 npg effects on gene transcription activity. Consistent with the second hypo-

Relative 0.75 *** thesis, mH2A1.2 depletion in metastatic 4T1 cells led to an increasing

expression level 0.50 amount of SOD3 mRNA, which suggested that mH2A1.2 repressed Input (%) 0.25 1.1 1.2 1.1 1.2 siCtrlsi si siCtrlsi si SOD3 expression (Fig. 7a). Conversely, depletion of mH2A1.1 in non- IgG +– +– +– +– +– +– metastatic 67NR cells decreased the amount of SOD3 mRNA, which mH2A1 –+ –+ –+ –+ –+ –+ b SOD3 pre-mRNA 4T1 67NR A BC 1,500 2 2 *** 1,000 Figure 7 mH2A1 associates with the SOD3 gene, and its splicing variants 200 regulate SOD3 transcription activity in an opposite manner. (a) Relative 1 1 SOD3 mRNA expression level measured by RT-qPCR in 4T1 or 67NR 100 Input (%) Relative *** cells transfected with siCtrl, si1.1 or si1.2. (b) Relative SOD3 pre-mRNA

expression level expression level measured by RT-qPCR, in 4T1 or 67NR cells transfected IgG +– +– +– +– +– +– with siCtrl, si1.1, or si1.2. (c) Schematic representation of the SOD3 1.1 1.2 1.1 1.2 siCtrlsi si siCtrlsi si H3 –+ –+ –+ –+ –+ –+ gene structure and localization of the primers used to analyze mH2A1 d e SOD3 mRNA incorporation (upper panel). qPCR analysis with primers targeting the mH2A1 ChIP mH2A1 ChIP 67NR SOD3 gene after ChIP using control antibodies (against IgG) or antibodies 2.0 2.0 4T1 si1.1 67NR si1.1 1.0 against mH2A1 (middle panel) or histone H3 (lower panel). (d) Effect si1.2 si1.2 of mH2A1.1 depletion (si1.1) or mH2A1.2 depletion (si1.2) on the 1.5 1.5 0.5 ** incorporation of mH2A1 in the SOD3 gene in 4T1 cells or in 67NR 1.0 1.0 Relative cells. Primers A, B, and C were as shown in c. (e) Relative SOD3 mRNA expression level

Fold effect expression level measured by RT-qPCR in 67NR cells transfected with Fold effect 0.5 ****** 0.5 *** 1.1 1.2 siCtrl, si1.1 alone or in combination with si1.2. Histograms represent the siCtrlsi ABC ABC ABC ABC + si average of at least three independent experiments. Error bars, s.e.m.; 1.1 si **P < 0.01; ***P < 0.001 (t test).

NATURE STRUCTURAL & MOLECULAR BIOLOGY VOLUME 19 NUMBER 11 NOVEMBER 2012  1143 ARTICLES

a siCtrl b siCtrl c siCtrl d siCtrl e Ctrl f g siDdx5-17 siDdx5-17 si1.2 si1.2 SOD3 mH2A1.1 mH2A1.2 SOD3 Ratio 1.1:1.2 SOD3 4 100 100 100 3 * 1.0 100 100 *** ** High High 75 75 75 (n = 128) (n = 42) 2 75 75 2 2 2 1 50 50 50 50 1 0.5 *** *** 50 Low Low (n = 315) *** 1 1 (n = 296) 25 25 25 25 25 P = 0.021, n = 443 P = 0.0033, n = 438 Relative cell number Relative cell number Relative cell number Relative expression level Metastasis-free survival (%) Relative expression level 0 Metastasis-free survival (%) 0 0 45 90 135 180 0 45 90 135 180 Time (months) Time (months)

Figure 8 mH2A1 alternative splicing controlled by Ddx5 and Ddx17 in MDA-MB-231 cells. (a) Relative number of migrating MDA-MB-231 cells after transfection with siCtrl or siDdx5-17. (b) Relative expression level of mH2A1.1 and mH2A1.2 mRNAs measured by RT-qPCR in MDA-MB-231 cells transfected with siCtrl or siDdx5-17. (c) Relative number of migrating MDA-MB-231 cells after transfection with siCtrl or si1.2. (d) Relative expression level of SOD3 mRNA measured by RT-qPCR in MDA-MB-231 cells transfected with siCtrl or si1.2. (e) Relative number of migrating MDA-MB-231 cells after addition of recombinant SOD3 protein to the cell-culture medium. Histograms represent the average of at least three independent experiments. Error bars, s.e.m.; *P < 0.05; **P < 0.01; ***;P < 0.001 (t test). (f) Kaplan-Meier curves for 180-month outcome in patients with breast cancer (n = 443) based on the mH2A1.1-to-mH2A1.2 splicing ratio. (g) Kaplan-Meier curves for 180-month outcome in patients with breast cancer (n = 438) based on SOD3 mRNA expression level.

suggested that mH2A1.1 stimulated SOD3 expression (Fig. 7a). These isoform (Fig. 8b), as had been observed in the 4T1 mouse cell line. results were confirmed by the SOD3 pre-mRNA levels (Fig. 7b), so As expected, depletion of the mH2A1.2 isoform in MDA-MB-231 we concluded that mH2A1 isoforms had opposite effects on SOD3 cells also inhibited cell migration (Fig. 8c) and increased the SOD3 gene transcription activity. mRNA expression level (Fig. 8d). Therefore, the mH2A1.2 isoform We next tested whether the mH2A1 histone was incorporated into may favor cancer-cell migration by repressing SOD3 expression, as the SOD3 gene in 67NR and 4T1 cells. Chromatin immunoprecipita- addition of SOD3 in MDA-MB-231 cell-culture medium inhibited tion (ChIP) was performed with an antibody that recognizes both cell migration (Fig. 8e). mH2A1 isoforms, as isoform-specific antibodies did not give any sig- Further supporting the physiopathological significance of our find- nal above background (data not shown). PCR analysis was performed ings, analyses of a cohort of 443 patients with primary breast tumors by using primers at several locations along the SOD3 gene (Fig. 7c). with a well-documented clinical follow-up (Supplementary Table 5) As shown in Figure 7c, mH2A1 was detected all along the SOD3 gene, indicated that low mH2A1.1-to-mH2A1.2 expression ratio was both in 67NR and 4T1 cells. A slightly higher level of mH2A1 was associated with shorter metastasis-free survival (Fig. 8f). Notably, detected on the SOD3 promoter in 4T1 as compared to 67NR cells a high level of mH2A1.2 isoform expression was associated with (Fig. 7c), but there was also more nucleosome in the SOD3 promoter shorter metastasis-free survival, which represents a prognostic fac- in 4T1 as compared to 67NR cells, as demonstrated by H3 histone tor for disease severity that is independent of other known markers ChIP (Fig. 7c). Moreover, mH2A1.2 depletion but not mH2A1.1 (Supplementary Fig. 7). Notably, the mH2A1.2 splicing isoform that © 2012 Nature America, Inc. All rights reserved. America, Inc. © 2012 Nature depletion strongly decreased the mH2A1 histone level incorporated represses SOD3 expression in 4T1 and MDA-MB-231 cells (Figs. 7a in the SOD3 gene in 4T1 cells (Fig. 7d). Thus, the mH2A1.2 isoform, and 8d) correlated negatively with the expression level of SOD3 which is the main mH2A1 isoform expressed in 4T1 cells, is incor- (P = 0.00019; Spearman rank correlation test) in the primary breast- npg porated in the SOD3 gene (Fig. 7d) and contributes to SOD3 gene tumor collection. Finally, a high expression level of SOD3 was associ- repression in these metastatic cells (Fig. 7a). ated with longer metastasis-free survival (Fig. 8g), in agreement with In contrast, in the nonmetastatic 67NR cells that expressed a simi- the inhibitory effect of SOD3 on in vitro migration (Figs. 6 and 8e). lar level of the mH2A1.1 and mH2A1.2 isoforms, depletion of either the mH2A1.1 or the mH2A1.2 isoform induced a similar, statistically DISCUSSION insignificant decrease in the level of the mH2A1 incorporated into RNA helicases, which comprise more than 60 enzymes that use the the SOD3 gene (Fig. 7d). Because mH2A1.1 depletion in 67NR cells energy of ATP hydrolysis to remodel RNA or RNA-protein complexes, decreased SOD3 mRNA and pre-mRNA levels and mH2A1.2 deple- are thought to be required at all stages of cellular RNA metabo- tion had opposite effects in 4T1 cells (Fig. 7a,b), we hypothesized that lism18,19,36. However, their target mRNAs and their cellular func- the mH2A1.2 isoform replaced the mH2A1.1 isoform on the SOD3 tions are poorly defined. In this context, whereas Ddx5 and Ddx17 gene following mH2A1.1 depletion in 67NR cells. This hypothesis was have been shown to play a part in the splicing process, their effect supported by the fact that SOD3 gene repression in the 67NR cells that on alternative splicing has been reported only for a few cases22–24. resulted from mH2A1.1 depletion was reversed by the co-depletion To the best of our knowledge, we provide the first evidence that of the mH2A1.2 isoform (Fig. 7e). Collectively, these results suggest RNA helicases can have a widespread impact on alternative splicing that both competition between the mH2A1.1 and mH2A1.2 isoforms (Supplementary Table 2 and Supplementary Fig. 2). Notably, we for their incorporation into the SOD3 gene and a higher mH2A1.2-to- observed that the 5` splicing sites of exons that are skipped upon mH2A1.1 ratio contribute to breast cancer–cell invasiveness through Ddx5 and Ddx17 depletion are within GC-rich regions as compared SOD3 gene repression. to the 5` splicing sites of exons that are included following depletion To further challenge this model, the human invasive MDA-MB- (Supplementary Fig. 2b). Because we show that the helicase activity 231 breast cancer–cell line was transfected with an siRNA that tar- of Ddx5 and Ddx17 is critical for their ability to regulate mH2A1 geted both Ddx5 and Ddx17, which inhibited cell migration (Fig. 8a) splicing (Fig. 2e), this result suggests that these proteins regulate and decreased the mH2A1.2 isoform and increased the mH2A1.1 alternative splicing by affecting RNA secondary structure24.

1144 VOLUME 19 NUMBER 11 NOVEMBER 2012 NATURE STRUCTURAL & MOLECULAR‚ BIOLOGY ARTICLES

Ddx5 and Ddx17 have been reported to have an impact on cellular model of tumor progression (Supplementary Fig. 6b), these results phenotypes, owing to their functions as transcriptional coregula- suggest that mH2A1 expression might be altered in different ways tors18,19,37. For example, we recently showed that Ddx5 and Ddx17 depending on tumor types, which may result in alteration of different are transcriptional coactivators of the NFAT5 promigratory transcrip- gene-expression programs. tion factor and that they participate in controlling the transcriptional In addition, the role of mH2A1 in cancer may also rely on post- activity of genes involved in cell migration and invasion, including translational modifications. In this context, it is important to under- S100A4 (ref. 38). We now show that the splicing activity of Ddx5 score the fact that the macrodomain of the mH2A1.1 splicing isoform, and Ddx17 contributes to their role in cell migration and invasion, but not that of the mH2A1.2 isoform, binds ADP-ribose and related by modulating the splicing of the mH2A1 gene, whose product in NAD metabolites. It has been proposed that mH2A1.1 but not turn regulates genes involved in cell migration and invasion, such as mH2A1.2 recognizes ADP-ribosylated chromatin components to pro- SOD3. As NFAT5 and mH2A1 do not control the expression of the mote rearrangement of chromatin loops28,29,49. This specific feature same genes (Supplementary Fig. 8), our data demonstrate that both of the mH2A1.1 splicing isoform is particularly relevant with respect the transcriptional and the splicing activities of Ddx5 and Ddx17 to our observation that mH2A1.1 and mH2A1.2 differentially regu- influence tumor progression by affecting different gene-expression late genes involved in redox metabolism. Indeed, ADP-ribosylation programs. Notably, a recent report also demonstrates that Ddx5 is a consequence of PARP activation under stress conditions, par- affects the expression of another set of genes that are also involved ticularly in the context of redox metabolism alteration and exposure in tumor progression (for example, cofilin and profilin) by modu- to ROS30,50–52. In addition, NAD metabolites have a central role lating the expression of micro RNAs (miRNAs)39. In sum, these in redox and ROS metabolism, and increasing evidence indicates results demonstrate that the different activities of Ddx5 and Ddx17 that redox metabolism in turn affects epigenetic regulation30,50–52. in transcription, splicing and miRNA processing affect different gene- Therefore, nonmetastatic 67NR cells that express a higher mH2A1.1- expression programs, which collectively contribute to their effects on to-mH2A1.2 ratio than metastatic 4T1 cells might be able to adapt tumor progression. their genetic programs to redox metabolism by expressing genes In agreement with a proposed role in cancer, several reports have such as SOD3, which would in turn affect cell invasiveness. Indeed, shown that Ddx5 or Ddx17 are overexpressed in different tumor the SOD3 gene codes for the extracellular superoxide dismutase that 18,19,40–42 •– types . Although we found that Ddx17 was overexpressed converts extracellular [O2] into H2O2. Notably, it has been recently in the metastatic 4T1 cells and tumors (Fig. 1a and Supplementary shown that the overexpression of SOD3 inhibits invasion by breast Fig. 1), a high level of Ddx17 (and/or Ddx5) mRNA did not correlate cancer cells32–35, which is consistent with the findings of this study. •– with a poor metastasis-free survival rate in the analyzed set of human Indeed, by converting extracellular [O2] into H2O2, SOD3 attenuates tumors (data not shown). However, analyzing Ddx5 and Ddx17 expres- oxidative fragmentation of the extracellular-matrix components such sion in human tumors is very challenging because Ddx5 or Ddx17 as heparan sulfate, thereby decreasing invasion and migration, and it alterations might occur at the protein level or at a post-translational may prevent other effects of ROS on cell motility30,32–35,53–56. In sup- modification level, as suggested by several reports43–47. port of a role for SOD3 in tumors, we report that SOD3 is an indica- Our work also revealed a role for mH2A1 histone pre-mRNA alter- tive factor for good prognosis (Fig. 8g). Notably, as well, it has been native splicing in breast cancer progression. Notably, it has been pre- recently shown that the mH2A1.2 isoform but not the mH2A1.1 iso- viously shown that the mH2A1.1 isoform inhibits lung cancer–cell form can interact with and regulate HER2, which is a major player in proliferation and that patients with low mH2A1.1 levels in lung tumor breast cancer initiation and progression57. Therefore, further experi- © 2012 Nature America, Inc. All rights reserved. America, Inc. © 2012 Nature samples are more likely to relapse than those with strong nuclear ments are required to decipher the precise mechanisms by which mH2A1.1 staining7,8. Similar results were recently reported in colon mH2A1.1 and mH2A1.2 differentially contribute to cancer. cancer48. Notably, the proportion of mH2A1.1 isoform (as a percent- In conclusion, this study reveals a role of tumor-associated alter- npg age of total mH2A1) decreased in tumors relative to normal samples native splicing of an epigenetic factor in tumor progression, which in several cancer types including breast cancer8. In the tumor samples leads in turn to transcriptional alterations of genes involved in a that we analyzed, we observed that the mH2A1 global level strongly specific cellular program. Because epigenetic factors can also have correlates with the level of mH2A1.2 isoform (Supplementary Fig. 7b), an impact on alternative splicing (reviewed in ref. 58), it is likely which we showed favors cancer-cell invasiveness (Fig. 3). On the that cross-talk between splicing and epigenetic networks may basis of the data obtained from cancers in lung7, colon48 and breast contribute to the control of cell phenotypes during both physiological (this study), one model is that the decrease in the mH2A1.1 isoform– and pathological states. expression level in tumor cells as compared to normal cells could favor cell proliferation, and overproduction of the mH2A1.2 isoform in METHODS tumor samples could further increase cell aggressiveness. Alteration Methods and any associated references are available in the online of mH2A1 alternative splicing during tumor initiation and progres- version of the paper. sion could be mediated by several splicing factors, as the QKI splicing factor, which is downregulated in cancer compared to normal cells, Accession codes. Microarray data have been deposited in the Gene has also been shown to affect mH2A1 splicing8. Expression Omnibus database under accession code GSE40737. Notably, a recent study demonstrated that the loss of mH2A1 histone Note: Supplementary information is available in the online version of the paper. is positively correlated with an increasing malignant phenotype of melanoma cells in culture and human tissue samples and that the ACKNOWLEDGMENTS suppressive effects of mH2A1 on melanoma progression is medi- We thank M. Buschbeck and V.A. Raker for critically reading the manuscript and ated through the regulation of CDK8 (ref. 4). Because we did not L. Mhamdi for technical assistance. We thank C. Vanbelle and C. Boucharon from mH2A1 CeCILE - SFR Sante Lyon-Est for technical assistance and helpful discussion. observe a general downregulation of the gene associated with This work was supported by the Institut National du Cancer (2008-1-PL BIO 01 breast cancer aggressiveness (data not shown) and because we did not to D.A. and S.V.), Agence Nationale de la Recherche (BLAN07-3_186592 to D.A. find any change in CDK8 gene expression in the 4T1 breast cancer and S.V.), Association d’Aide à la Recherche Cancérologique de Saint-Cloud (to R.L.)

NATURE STRUCTURAL & MOLECULAR BIOLOGY VOLUME 19 NUMBER 11 NOVEMBER 2012 ƒ 1145 ARTICLES

and Fondation Recherche Médicale (Equipe FRM DEQ20110421278). S.G. was 26. Gamble, M.J., Frizzell, K.M., Yang, C., Krishnakumar, R. & Kraus, W.L. The histone supported by Association pour la Recherche sur le Cancer; L.G. and S.P. variant macroH2A1 marks repressed autosomal chromatin, but protects a subset by Agence Nationale de la Recherche; E.Z., H.M. and J.-P.V. by FRM; E.D. of its target genes from silencing. Genes Dev. 24, 21–32 (2010). by Ligue Nationale Contre le Cancer; and M.P.E. by Association Française 27. Buschbeck, M. et al. The histone variant macroH2A is an epigenetic regulator of key developmental genes. Nat. Struct. Mol. Biol. 16, 1074–1079 (2009). contre les Myopathies. 28. Timinszky, G. et al. A macrodomain-containing histone rearranges chromatin upon sensing PARP1 activation. Nat. Struct. Mol. Biol. 16, 923–929 (2009). AUTHOR CONTRIBUTIONS 29. Gamble, M.J. & Kraus, W.L. Multiple facets of the unique histone variant macroH2A: E.D., S.P., M.D., R.L., S.V. and D.A. designed the experiments; S.V. and from genomics to cell biology. Cell Cycle 9, 2568–2574 (2010). D.A. supervised the project; E.D., S.P., K.D., L.G., M.L.-T., E.Z. and S.G. conducted 30. Grek, C.L. & Tew, K.D. Redox metabolism and malignancy. Curr. Opin. Pharmacol. 10, the experiments; M.P.E., H.M. and J.-P.V. performed all bioinformatics analysis. 362–368 (2010). All authors performed the data analysis and wrote the manuscript. 31. Pani, G., Galeotti, T. & Chiarugi, P. Metastasis: cancer cell`s escape from oxidative stress. Cancer Metastasis Rev. 29, 351–378 (2010). COMPETING FINANCIAL INTERESTS 32. Chaiswing, L. & Oberley, T.D. Extracellular/microenvironmental redox state. Antioxid. Redox Signal. 13, 449–465 (2010). The authors declare no competing financial interests. 33. Mori, K., Shibanuma, M. & Nose, K. Invasive potential induced under long-term oxidative stress in mammary epithelial cells. Cancer Res. 64, 7464–7472 (2004). Published online at http://www.nature.com/doifinder/10.1038/nsmb.2390. 34. Teoh, M.L., Fitzgerald, M.P., Oberley, L.W. & Domann, F.E. Overexpression of Reprints and permissions information is available online at http://www.nature.com/ extracellular superoxide dismutase attenuates heparanase expression and inhibits reprints/index.html. breast carcinoma cell growth and invasion. Cancer Res. 69, 6355–6363 (2009). 35. Svensk, A.M., Soini, Y., Paakko, P., Hiravikoski, P. & Kinnula, V.L. Differential 1. Füllgrabe, J., Kavanagh, E. & Joseph, B. Histone onco-modifications. Oncogene 30, expression of superoxide dismutases in lung cancer. Am. J. Clin. Pathol. 122, 3391–3403 (2011). 395–404 (2004). 2. Portela, A. & Esteller, M. Epigenetic modifications and human disease. 36. Abdelhaleem, M. Helicases: an overview. Methods Mol. Biol. 587, 1–12 (2010). Nat. Biotechnol. 28, 1057–1068 (2010). 37. Yang, L., Lin, C. & Liu, Z.R. P68 RNA helicase mediates PDGF-induced epithelial 3. Dalvai, M. & Bystricky, K. The role of histone modifications and variants in regulating gene mesenchymal transition by displacing Axin from B-catenin. Cell 127, 139–155 expression in breast cancer. J. Mammary Gland Biol. Neoplasia 15, 19–33 (2010). (2006). 4. Kapoor, A. et al. The histone variant macroH2A suppresses melanoma progression 38. Germann, S. et al. Dual role of the /ddx17 RNA helicases in the control of the through regulation of CDK8. Nature 468, 1105–1109 (2010). pro-migratory NFAT5 transcription factor. Oncogene 618, 1038–1048 (2012). 5. Srivastava, N., Gochhait, S., de Boer, P. & Bamezai, R.N. Role of H2AX in DNA 39. Wang, D. et al. RNA helicase DDX5 regulates microRNA expression and contributes damage response and human cancers. Mutat. Res. 681, 180–188 (2009). to cytoskeletal reorganization in basal breast cancer cells. Mol. Cell Proteomics 11, 6. Valdés-Mora, F. et al. Acetylation of H2A.Z is a key epigenetic modification M111.011932 (2012). associated with gene deregulation and epigenetic remodeling in cancer. Genome 40. Shin, S., Rossow, K.L., Grande, J.P. & Janknecht, R. Involvement of RNA helicases Res. 22, 307–321 (2012). p68 and p72 in colon cancer. Cancer Res. 67, 7572–7578 (2007). 7. Sporn, J.C. et al. Histone macroH2A isoforms predict the risk of lung cancer 41. Mooney, S.M., Grande, J.P., Salisbury, J.L. & Janknecht, R. Sumoylation of p68 recurrence. Oncogene 28, 3423–3428 (2009). and p72 RNA helicases affects protein stability and transactivation potential. 8. Novikov, L. et al. QKI-mediated alternative splicing of the histone variant macroH2A1 Biochemistry 49, 1–10 (2010). regulates cancer cell proliferation. Mol. Cell. Biol. 31, 4244–4255 (2011). 42. Wortham, N.C. et al. The DEAD-box protein p72 regulates ERA-/oestrogen-dependent 9. Venables, J.P. et al. Cancer-associated regulation of alternative splicing. Nat. Struct. transcription and cell growth, and is associated with improved survival in Mol. Biol. 16, 670–676 (2009). ERA-positive breast cancer. Oncogene 28, 4053–4064 (2009). 10. Dutertre, M. et al. Exon-based clustering of murine breast tumor transcriptomes 43. Yang, L., Lin, C. & Liu, Z.R. Phosphorylations of DEAD box p68 RNA helicase are reveals alternative exons whose expression is associated with metastasis. associated with cancer development and cell proliferation. Mol. Cancer Res. 3, Cancer Res. 70, 896–905 (2010). 355–363 (2005). 11. David, C.J. & Manley, J.L. Alternative pre-mRNA splicing regulation in cancer: 44. Carter, C.L., Lin, C., Liu, C.Y., Yang, L. & Liu, Z.R. Phosphorylated p68 RNA helicase pathways and programs unhinged. Genes Dev. 24, 2343–2364 (2010). activates Snail1 transcription by promoting HDAC1 dissociation from the Snail1 12. Dutertre, M., Vagner, S. & Auboeuf, D. Alternative splicing and breast cancer. promoter. Oncogene 29, 5427–5436 (2010). RNA Biol. 7, 403–411 (2010). 45. Causevic, M. et al. Overexpression and poly-ubiquitylation of the DEAD-box RNA 13. Stamm, S. et al. Function of alternative splicing. Gene 344, 1–20 (2005). helicase p68 in colorectal tumours. Oncogene 20, 7734–7743 (2001).

© 2012 Nature America, Inc. All rights reserved. America, Inc. © 2012 Nature 14. Pan, Q., Shai, O., Lee, L.J., Frey, B.J. & Blencowe, B.J. Deep surveying of alternative 46. Jacobs, A.M. et al. SUMO modification of the DEAD box protein p68 modulates splicing complexity in the human transcriptome by high-throughput sequencing. its transcriptional activity and promotes its interaction with HDAC1. Oncogene 26, Nat. Genet. 40, 1413–1415 (2008). 5866–5876 (2007). 15. Witten, J.T. & Ule, J. Understanding splicing regulation through RNA splicing maps. 47. Mooney, S.M., Goel, A., D`Assoro, A.B., Salisbury, J.L. & Janknecht, R. Pleiotropic Trends Genet. 27, 89–97 (2011). effects of p300-mediated acetylation on p68 and p72 RNA helicase. J. Biol. Chem. npg 16. Aslakson, C.J. & Miller, F.R. Selective events in the metastatic process defined by 285, 30443–30452 (2010). analysis of the sequential dissemination of subpopulations of a mouse mammary 48. Sporn, J.C. & Jung, B. Differential regulation and predictive potential of MacroH2A1 tumor. Cancer Res. 52, 1399–1405 (1992). isoforms in colon cancer. Am. J. Pathol. 180, 2516–2526 (2012). 17. Bemmo, A. et al. Exon-level transcriptome profiling in murine breast cancer reveals 49. Kustatscher, G., Hothorn, M., Pugieux, C., Scheffzek, K. & Ladurner, A.G. Splicing splicing changes specific to tumors with different metastatic abilities. PLoS ONE 5, regulates NAD metabolite binding to histone macroH2A. Nat. Struct. Mol. Biol. 12, e11981 (2010). 624–625 (2005). 18. Fuller-Pace, F.V. & Moore, H.C. RNA helicases p68 and p72: multifunctional 50. Krishnakumar, R. & Kraus, W.L. The PARP side of the nucleus: molecular actions, proteins with important implications for cancer development. Future Oncol. 7, physiological outcomes, and clinical targets. Mol. Cell 39, 8–24 (2010). 239–251 (2011). 51. Giansanti, V., Dona, F., Tillhon, M. & Scovassi, A.I. PARP inhibitors: new tools to 19. Janknecht, R. Multi-talented DEAD-box proteins and potential tumor promoters: protect from inflammation. Biochem. Pharmacol. 80, 1869–1877 (2010). p68 RNA helicase (DDX5) and its paralog, p72 RNA helicase (DDX17). 52. Cyr, A.R. & Domann, F.E. The redox basis of epigenetic modifications: from mechanisms Am. J. Transl. Res 2, 223–234 (2010). to functional consequences. Antioxid. Redox Signal. 15, 551–589 (2011). 20. Bates, G.J. et al. The DEAD box protein p68: a novel transcriptional coactivator of 53. Acharya, A., Das, I., Chandhok, D. & Saha, T. Redox regulation in cancer: a double- the p53 tumour suppressor. EMBO J. 24, 543–553 (2005). edged sword with therapeutic potential. Oxid. Med. Cell. Longev. 3, 23–34 21. Endoh, H. et al. Purification and identification of p68 RNA helicase acting as a (2010). transcriptional coactivator specific for the activation function 1 of human estrogen 54. Weaver, A.M. Regulation of cancer invasion by reactive oxygen species and Tks receptor alpha. Mol. Cell. Biol. 19, 5363–5372 (1999). family scaffold proteins. Sci. Signal. 2, pe56 (2009). 22. Hönig, A., Auboeuf, D., Parker, M.M., O`Malley, B.W. & Berget, S.M. Regulation of 55. Svineng, G., Ravuri, C., Rikardsen, O., Huseby, N.E. & Winberg, J.O. The role of alternative splicing by the ATP-dependent DEAD-box RNA helicase p72. Mol. Cell reactive oxygen species in integrin and matrix metalloproteinase expression and Biol. 22, 5698–5707 (2002). function. Connect. Tissue Res. 49, 197–202 (2008). 23. Guil, S. et al. Roles of hnRNP A1, SR proteins, and p68 helicase in c-H-ras 56. Radisky, D.C. et al. Rac1b and reactive oxygen species mediate MMP-3-induced alternative splicing regulation. Mol. Cell Biol. 23, 2927–2941 (2003). EMT and genomic instability. Nature 436, 123–127 (2005). 24. Kar, A. et al. RNA Helicase p68 (DDX5) regulates tau exon 10 splicing by modulating 57. Li, X. et al. The atypical histone macroH2A1.2 interacts with HER-2 in cancer a stem-loop structure at the 5` splice site. Mol. Cell. Biol. 31, 1812–1821 (2011). cells. J. Biol. Chem. 287, 23171–23183 (2012). 25. Thomas, P.D. et al. PANTHER: a library of protein families and subfamilies indexed 58. Luco, R.F., Allo, M., Schor, I.E., Kornblihtt, A.R. & Misteli, T. Epigenetics in by function. Genome Res. 13, 2129–2141 (2003). alternative pre-mRNA splicing. Cell 144, 16–26 (2011).

1146 VOLUME 19 NUMBER 11 NOVEMBER 2012 NATURE STRUCTURAL & MOLECULAR‚z BIOLOGY ONLINE METHODS The bottom chamber was filled with complete medium. After 24 h, cells at the top Cell culture, treatment and transfection. 67NR, 168FARN, 4T07 and 4T1 side of the filter were removed with a cotton swab. The remaining cells at the bottom cells were kindly provided by F. Miller (Michigan Cancer Foundation, Detroit, side were washed with PBS, fixed with 10% TCA and stained with Amido black. Michigan, USA). Cells were cultured in DMEM supplemented with 10% FBS Cells were counted from three randomized fields per well under a Nikon inverted at 37 °C. Transient transfection of siRNAs (25 nM) was performed by using microscope. A cell-invasion assay was performed in similar conditions, except Lipofectamine RNAiMAX (Invitrogen) following the manufacturer’s instruc- that wells were precoated with matrigel at a 1:3 dilution for 4T1 or a 1:12 dilution tions. Cells were harvested 48 h after transfection. For rescue experiments, HeLa for 67NR (BD Biosciences). For wound-healing assays, 2 d after transfection, cells were transfected with siRNA targeting the Ddx5 and Ddx17 UTR regions. a wound space of a confluent cell monolayer was produced by scratching the Twenty-four hours later, wild-type and either mutated Ddx5 or Ddx17 expres- plate with a p200 pipette tip. The percentage of migration was defined at different sion vectors were transfected by using Lipofectamine 2000 (Invitrogen). Cells times (8, 16, or 24 h) by measuring the lengthwise migration three times with the were harvested 48 h later. SiRNA sequences are provided in Supplementary program ImageJ. Mitomycin C was added 1 h before wounding and maintained Table 4. SiDdx5-17 is an siRNA that targets a conserved sequence that simul- during migration. Recombinant SOD protein (100 units; S9697, Sigma) were taneously affects Ddx5 and Ddx17 (and not a mixture of siDdx5 and siDdx17) added after cells were plated in the upper chamber. (Supplementary Table 4). Affymetrix exon array. One microgram of total RNA purified with TRIzol was Minigene assay and PCR digestion. hH2AFY exon 6, with 200 bp and 174 bp of labeled with Ambion reagents and hybridized to Affymetrix GeneChip Mouse flanking introns, was amplified by PCR with the primers 5`-GGAATTCCATAT Exon 1.0 ST arrays. Affymetrix exon-array data were normalized with quantile GGAATTCCCTAGTTTGCTTGC-3` and 5`-GGAATTCCATATGGAATTCAC normalization. The background correction was performed by using antigenomic ACTTGGAATGGC-3` that contained the NdeI restriction site. hH2AFY exon 7, probes. Only probes targeting exons annotated from full-length cDNA were used with 106 bp and 212 bp of flanking introns, was amplified by PCR with the for analysis. Among these preselected probes, poor-quality probes (for example, primers 5`-GGAATTCCATATGGAATTCAATGTGCCTGTGCGT-3` and 5`- probes labeled as ‘cross-hybridizing’ probes by Affymetrix) and probes with GGAATTCCATATGGAATTCCATTTGTGAGCTGC-3` that contained the NdeI intensity signals that were too low when compared to antigenomic background restriction sites. NdeI-digested PCR products were cloned into pTBminigene. probes with the same GC content were removed from the analysis. Only probes Primers used for minigene assay are described in Supplementary Table 4. with a DABG P value of a 0.05 in at least half of the chips were considered for further statistical analysis. Paired statistical analyses were performed by using the RNA preparation and RT-qPCR. Total RNA was extracted by using TRIzol Student’s paired t-test on the splicing index to analyze the exon-array data. The (Invitrogen). One microliter of Glycoblue (Ambion) was added before RNA precip- splicing index corresponds to a comparison of gene-normalized exon intensity itation with isopropanol. RNA (1 Mg) was treated with DNase I (DNAfree, Ambion) values between the two analyzed experimental conditions. for 30 min at 37 °C and reverse-transcribed (RT) by using SuperScript II and random primers (Invitrogen). Before PCR, all RT reaction mixtures were Clinical samples, quantitative PCR and statistical analysis. Samples of 443 diluted to contain 2.5 ng/Ml of initial RNA. PCR reactions were performed using breast cancer tumors excised from women at the Institut Curie/Hôpital René 5 Ml of the diluted cDNAs and GoTaq polymerase (Promega). Quantitative PCR Huguenin (Saint-Cloud, France) from 1978 to 2008 were analyzed. This study (qPCR) was performed using 2.5 Ml of the diluted cDNAs and SYBR Green I was approved by the local ethical committee (Breast Group of Hôpital René Master mix (Roche). Primer sequences are provided in Supplementary Huguenin), which provided informed consent. Standard prognostic factors Table 4. The relative RNA levels were determined on the basis of the threshold are reported in Supplementary Figure 4. Immediately after surgery, the tumor cycle for each PCR product (Ct). samples were flash frozen and stored in liquid nitrogen until RNA extraction. The samples were examined histologically for the presence of at least 60% tumor cells. ChIP. Cells were cross-linked with 1% formaldehyde for 10 min and lysed in The patients (mean age, 61.7 yrs; range, 35–91 yrs) all met the criteria of having buffer containing 50 mM Tris-HCl, pH7.5, 150 mM KCl, 5 mM EDTA, 1% NP40, primary unilateral breast carcinoma for which complete clinical, histological and 0.1% SDS, 0.5% sodium deoxycholate, 50 mM sodium fluoride and phosphatase biological data were available; 378 patients received adjuvant therapy, consisting © 2012 Nature America, Inc. All rights reserved. America, Inc. © 2012 Nature and protease inhibitors. Chromatin was sheared by sonication with a Bioruptor of chemotherapy alone in 90 cases, hormone therapy alone in 175 cases and both (Diagenode) to generate ~200-bp DNA fragments and incubated with antibodies treatments in 93 cases. The median follow-up was 8.9 years (range 0.5–29.0). against mH2A1 (5 Mg, Abcam cat. no. 37264), H3 (1 Mg, Abcam cat. no. 1791) A total of 168 patients relapsed at distant sites. npg or control immunoglobulins (Santa Cruz) and magnetic beads overnight at All RT-qPCR reactions were performed by using an ABI Prism 7700 Sequence 4 °C. Beads were washed five times with buffer containing 50 mM Tris-HCl, Detection System and the SYBR Green PCR Core Reagents kit (PerkinElmer pH 7.5, 150 mM KCl, 1% NP40 and 0.25% sodium deoxycholate, and two times Applied Biosystems). TATA-box-binding protein (TBP) transcripts were used with TE buffer59. Antibody-bound chromatin was reverse cross-linked overnight as an endogenous RNA control, and each sample was normalized on the basis at 65 °C and treated with proteinase K (Qiagen) before DNA purification on of its TBP content. To determine whether the mH2A1 splicing-variant ratio was column (Qiagen). Purified DNA was diluted three times in water and analyzed associated with patient clinical outcome, metastasis-free survival distribution by qPCR. was estimated by the Kaplan-Meier method. To select cutoff expression levels to classify each patient in one of two risk groups, sensitivity and specificity was Western blot analysis. Protein extracts were obtained by using NP-40 buffer explored by using a receiver-operating curve (ROC) analysis. Briefly, the area (50 mM Tris-HCl, pH 8, 0.4 M NaCl, 5 mM EDTA, pH 8, 1% NP40, 0.2% SDS and under the ROC curve (with 95% confidence interval) was calculated, and a test 1 mM DTT) and protease inhibitors (Sigma). Ten micrograms of total proteins for the null hypothesis that the area under the curve was 50% was performed. were separated by SDS-PAGE, and western blotting was subsequently performed The ROC analysis provided the threshold expression value to balance sensitivity with antibodies against Ddx5 (PAb204, Upstate/Millipore,1:1,000), Ddx17 and specificity for the detection of life-threatening cancer, and this cut point was (ab24601, Abcam, 1:250), mH2A1.1 (4160, Cell Signaling, 1:1,000), mH2A1.2 used in the Kaplan-Meier analysis. Kaplan-Meier curves estimate metastasis-free (4287, Cell Signaling, 1:1,000), mH2A1 (ab37264, Abcam, 1:1,000) and B-actin survival from 0 to 180 months after breast cancer diagnosis. The Spearman rank (Mab1501, Millipore, 1:1,000). correlation test provided correlation between continuous variables.

Cell-migration, invasion and wound-healing assays. For cell-migration assay, 59. Bittencourt, D. et al. Cotranscriptional splicing potentiates the mRNA production 4 4 cells (4 × 10 for 67NR and 2 × 10 for 4T1) were plated, in serum-free medium from a subset of estradiol-stimulated genes. Mol. Cell. Biol. 28, 5811–5824 48 h after transfection, in the upper chamber of 8-Mm pore-size ThinCerts. (2008).

doi:10.1038/nsmb.2390 NATURE STRUCTURAL & MOLECULAR‚{ BIOLOGY 4T1 a b siCtrl * 120 siddx5-17 67NR siddx5 + siddx17 2 4T1 90 ddx17 80 70 1 ddx5 60

level (RT-qPCR) level ȕ-actin 50 Relative expression 30 *** *** ddx5 ddx17 Relative cell number c 67NR d 4T1 siCtrl 100% subG1 siddx5 -17 G2 S 100 75% G1

75 50% 50 25% 25 Relative cell number Relative cell number 0% siCtrl siddx5 siddx17 siddx5-17 e 150 f

150 si1.1 100 si1.2 100 50 50 (compared to siCtrl) to siCtrl) (compared Relative cell number (compared to siCtrl) to siCtrl) (compared Relative cell number

24 h 48 h 24 h 48 h Supplementary Figure 1: Effect of ddx5 and ddx17 in cell migration and proliferation in the 4T1 mouse model of mammary tumor progression. (a) ddx17, but not ddx5, was overexpressed in the metastatic 4T1 cells, both at the RNA (left panel) and at the protein level (right panel), as compared to non-metastatic 67NR cells. (b) The siRNA siddx5-17 corresponds to a conserved sequence that allows ddx5 and ddx17 to be targeted simultaneously. Using siddx5-17 had similar effects on cell migration as did a mixture of siddx5 and siddx17 (siddx5 + siddx17). (c) ddx5 and ddx17 depletion (using siddx5-17) did not affect cell migration of the 67NR non- invasive cells. (d) Under the tested conditions, ddx5 and/or ddx17 depletion in 4T1 cells did not affect subG1 cell proportion, indicating that it did not affect cell apoptosis. In addition, ddx5 and/or ddx17 depletion in 4T1 cells did not increase the G1- to S-phase ratio, indicating that it did not decrease cell proliferation; this was substantiated by DNA flow cytometry analysis. (e) Depletion of ddx5 and ddx17 did not affect 4T1 cell proliferation, as assessed by the sulforhodamine B growth assay at 24 h or 48 h after siddx5-17 transfection. (f) Depletion of mH2A1.1 or mH2A1.2 isoforms did not affect 4T1 cell proliferation, as assessed by thet sulforhosulforhodaminedamine B ggrowthgrowrowth assaaassayssayy at 24 h or 48 h after si1.1 or si1.2 transfection. Nature Structural & Molecular Biology: doi:10.1038/nsmb.2390  ‚|  a - + %          )     & &*  + 2  2  2&*  + *, +   2&)%-    2&)%/        -  ) % 4        )  

   c   &  p 2  )  0  2&)%-    SI=2.5 SI=2.5  2&*%1       SnaSnapc4 6 -  a %           ) 

 & ,/ at2 2 Mat2a Mat2a M SI= 1. SI= 1.6  ,/  2&)%-       

 )     2&)%1 *+      .         %) )% + + & ) 2     2 +      2+%)       2&,%-   2&)%.  ) )*+  4  0        G     )% & $ BK 

 )     $    2&*%1

2  SI=2.4 SI=2. 

   IKBKG IK 2&)%0   1  +          &*     $ =2.5  2  OXR1 OXR1 OXR  2&* SI SI=2.5

$+          !,

 2&*%-       

    - *     %  vk * & Mvk Mvk M     SI=1.7 SI=1.7

   2 "* *  2&*%/     2&*%-

3      ) id

        p

*%  & ! al    SI=1.7 SI=1.7 Talpid3T

2  !  !0 2&*%)    2&*%,      - /     +  *     

   "  2)% +% "/ )%0  2)%- &    )

2     )   2&)%0  2&+%*      ) + b )%.     **  2&)%.  ! !" "*      ! ! !" $   

        -    $ "

 2&)%/      ""     "  # #"$!       "   "   2&)%1     """"-       Supplementary Figure 2: Validation of splicing events regulated by ddx5 and ddx17 (a) Validation by RT-PCR of the splicing events predicted by exon array analysis to be regulated by ddx5 and ddx17. 4T1 cells were transfected for 48 h with a control siRNA (siCtrl) or an siRNA targeting both ddx5 and ddx17 (siddx5-17). Depletion of ddx5 and ddx17 induced both exon inclusion and exon skipping. SI: Splicing Index calculated from the exon array dataset. (b) Comparison of GC content at the 5’ splicing site of included or skipped exons upon ddx5-17 depddeppletionletion.. Nature Structural & Molecular Biology: doi:10.1038/nsmb.2390  ‚}  a 4T1 c mH2A1.1 mH2A1.2     mH2A1.1 mH2A1.2 8   1 6   4 0.5 Relative Relative 2 expression level expression level

siddx5-17 - + - +

Ratio 1.1/1.2 2 mH2A1 30 d     20   1    

Relative 10

expression level *** siddx5-17 - + - +   *** *** b 4T1 *** Ratio 1.1/1.2 Relative expression level

20 ** ** 15 10 Relative 5 Ratio 1.1:1.2

expression level e siCtrl 67NR 168F 4T07 4T1 siddx5-17 3 *** ** 2 mH2A1 siddx5 + siddx17 2

1 Relative

Relative 1 expression level expression level

Supplementary Figure 3: Regulation of the expression of the mH2A1.1 and mH2A1.2 isoforms in the 4T1 mouse model of tumor progression. (a) Depletion of ddx5 and ddx17 in mouse 4T1 cells increased the mH2A1.1 isoform level and decreased mH2A1.2 isoform level, which resulted in the increase in the mH2A1.1 to mH2A1.2 ratio without affecting the global expression level of the mH2A1 gene. (b) Siddx5-17, which targets ddx5 and ddx17 simultaneously, had similar effect on mH2A1 splicing as a mixture of siddx5 and siddx17 (siddx5 + siddx17). (c) Validation of si1.1 and si1.2 in 4T1 and 67NR cells, by measuring mH2A.1 and mH2A1.2 splicing variants by RT-qPCR. (d) RT-qPCR analyses of the mH2A1 total mRNA levels after transfection with siCtrl, si1.1, or si1.2 in 4T1 cells or 67NR cells. (e) Relative expression levels of mH2A1.1 and mH2A1.2 mRNAs, as measured by RT-qPCR, in mouse primary tumors with different metastatic capability. Histograms represent the average of three independent experiments corresponding to the fold change of the mH2A1.1 to mH2A1.2 ratio.

Nature Structural & Molecular Biology: doi:10.1038/nsmb.2390  ‚~  a

    

           

 

                               

b c mH2A1.1 mH2A1.2 Ratio 1.1:1.2 15 8 1 *** 6

4 0.5 10 Relative 2 expression level Relative 5 siddx5-17 - + - + expression level ** ** Ratio 1.1:1.2 mH2A1

15 - + - + 1 siUTRddx5 siUTRddx17 - - + + 10 0.5 Relative 5 expression level

siddx5-17 - + - +

Supplementary Figure 4: Regulation of the expression of the mH2A1.1 and mH2A1.2 isoforms in HeLa cells. (a) The mH2A1 gene is highly conserved during evolution. Alignment of human and mouse sequences of exon 1.1, exon 1.2, and 100 nucleotides of the surrounding introns revealed 75% to 100% conservation. UI: upstream intron; DI: downstream intron. (b) Depletion of ddx5 and ddx17 in human HeLa cells increased the mH2A1.1 isoform level and decreased mH2A1.2 isoform level, resulting in increase in the mH2A1.1 to mH2A1.2 ratio without affecting the global expression level of the mH2A1 gene. (c) Depletion of ddx5 (siUTRddx5) or ddx17 (siUTRddx17) in human HeLa cells increased the mH2A1.1 to mH2A1.2 ratio, and depletion of both ddx5 and ddx17 worked synergistically to further increase this ratio. Nature Structural & Molecular Biology: doi:10.1038/nsmb.2390  ‚  a siCtrl siCtrl b 4T1 si1.2#1 67NR si1.1#1 siCtrl siCtrl 4T1 67NR si1.2#2 si1.1#2 si1.1 si1.2 100 160 ** * 100 100 *** *** 75 120

50 50 80 50

25 40 Relative cell number Relative cell number Relative cell number Relative cell number

siCtrl siCtrl d c siUTRSOD3 SOD3 mRNA siddx5-17 siddx5 + siddx17 SOD3 mRNA migration 1 150 6 ** ** 0.8 100 4 0.4

** level

Relative 0.4 2 50

expressionlevel 0.2 Relative expression Relative cell number

Ctrl siCtrl e 4T1 f 4T1 SOD3 siSOD3 100 ** 100 75

50 50

25 Relative cell number Relative cell number

Supplementary Figure 5: Effect of mH2A1 splicing isoforms and SOD3 on 67NR and 4T1 cell migration (a) Two different siRNAs targeting the mH2A1.2 isoform inhibited 4T1 cell migration (left panel), two different siRNAs targeting the mH2A1.1 isoform increased 67NR cell migration (right panel). (b) Depletion of the mH2A1.1 isoform in 4T1 cells, and of mH2A1.2 in 67NR cells, had no effect on cell migration. (c) siddx5-17 targeting simultaneously ddx5 and ddx17 had similar effects on SOD3 expression as a mixture of siddx5 and siddx17 (siddx5 + siddx17). (d) siRNAs targeting SOD3 mRNA UTR (siUTRSOD3; left panel) increased 67NR cell migration (right panel). (e) Addition of SOD3 protein in culture media inhibited 4T1 cell migration. (f) SOD3 depletion had no effect on 4T1 cell migration, as expected considering the low expression of SOD3 in these cells.

Nature Structural & Molecular Biology: doi:10.1038/nsmb.2390  ‚€  a 5 4T1 siddx5-17+si1.2 vs. 4T1 siddx5-17+ si1.1 * 4 ** 3 * * * * * 2 * * *

Relative Relative * * * 1

expression level level expression * ** *** *** AQP1 MTTP ATF3 CD53 HAO1 HAO1 PHEX SOD3 TXNIP TXNIP LOXL3 LOXL3 RFESD GFOD1 GFOD1 HMOX1 HMOX1 GPNMB GPNMB APCDD1 SEMA7A SEMA7A SLC7A11 CDKN1A

b Comparison 67NR vs. 4T1 primary tumors 3 *

2

1 Relative expression Relative level (real-time RT-PCR) RT-PCR) level(real-time CDK8 mH2A2 H2A.Z SOD3

Supplementary Figure 6: Gene expression regulation by mH2A1 splicing isoforms. (a) Validation by RT-qPCR of the expression level of genes predicted by the microarray analysis to be differentially expressed when 4T1 cells transfected with siddx5-17 and si1.2 were compared to 4T1 cells transfected with siddx5-17 and si1.1. Histograms represent the average of at least three independent experiments. Error bars represent S.E.M. (* P <0.05, **P <0.01, ***P <0.001 (t-test)). (b) Relative expression levels measured by RT-qPCR of CDK8, mH2A2, H2A.Z, and SOD3 RNAs, when comparing non-metastatic 67NR to metastatic 4T1 primary tumors. Only SOD3 mRNA was more highly expressed in non-metastatic 67NR as compared to the metastatic 4T1 primary tumors.

Nature Structural & Molecular Biology: doi:10.1038/nsmb.2390  ‚  a mH2A1.2 b 100 mH2A1.2 60 mH2A1.1 Low 2 75 (n=270) R = 0.9 40 50    25 20 R2= 0.48 p=0.017 of mH2A1.1 of mH2A1.1 or mH2A1.2 Relative expression level

Metastasis-free survival(%) 0 0 45 90 135 180 20 40 60 Time (month) Relative expression level of mH2A1

c Variables Metastasis-free survival HR (95% CI)a p-value Lymph node status 0 1 1-3 1.57 (1.24-1.98) 0.00014 >3 2.46 (1.55-3.90) SBR histological grade 1 1 2 1.52 (1.18-1.95) 0.0011 3 2.31 (1.40-3.80) Macroscopic tumor size ”PP 1 0.015 >25 mm 1.56 (1.08-2.09)

mH2A1.2 expression low expression 1 0.043 high expression 1.37 (1.01-1.87)

Supplementary Figure S7: Expression of mH2A1 splicing isoforms in breast cancer patients. (a) Kaplan-Meier curves for 180-month outcome in breast cancer patients (n = 443), based on the mH2A1.2 expression level. (b) The global expression level of mH2A1 histone correlates with the expression level of the mH2A1.2 splicing isoform in human breast tumors. (c) A high expression level of the mH2A1.2 isoform correlates with the prognostic value, independently of other known markers of disease severity, including lymph node status, Scarff Bloom Richardson (SBR) histological grade, and macroscopic tumor size, as demonstrated by multivariable Cox regression analysis of metastasis-free survival in breast cancer patients. The significance of differences between survival rates was ascertained using the log-rank test. Cox proportional hazards regression model was used to assess prognostic significance. a: hazard ratio and 95% confidence interval.

Structural & Molecular Biology: doi:10.1038/nsmb.2390  ‚‚  a MDA-MB-231 siCtrl si1.2 S100A4 mRNA siNFAT5 SOD3 mRNA 1.5 3 *

1 2

0.5 ** 1 Relative expression Relative expression level (real-time RT-PCR) (real-time level RT-PCR) (real-time level

b 4T1 siCtrl si1.2 S100A4 mRNA siNFAT5 SOD3 mRNA 1.5 3

1 2 **

0.5 1 *** Relative expression Relative expression level (real-time RT-PCR) (real-time level RT-PCR) (real-time level

Supplementary Figure 8: SOD3 and S100A4 gene expression regulation by NFAT5 and mH2A1.2 in 4T1 and MDA-MB-231 breast cancer cells. NFAT5, whose transcriptional activity depends on ddx5-ddx17, controls the expression of different set of genes as the mH2A1.2 isoform, whose splicing is regulated by ddx5-ddx17. Indeed, NFAT5 depletion inhibited S100A4 expression (which favors cell migration) but had no effect on SOD3 expression, while depletion of mH2A1.2 increased SOD3 expression but had no effect on S100A4 expression, in both MDA-MB-231 cells (a) and 4T1 cells (b).

Nature Structural & Molecular Biology: doi:10.1038/nsmb.2390  ‚ƒ  Number of Patients Number (%) of metastases N= 443 N=168 (37.9) p-valuea Age d50 92 38 (41.3) NS >50 351 130 (37.3) SBR histological gradeb,c 54 7 (13) I 226 84 (37.2) 0.000051 II 154 73 (47.4) III Lymph node statusd 0 114 34 (29.8) 1-3 231 76 (32.9) 0.00000044 >3 97 58 (59.8) Macroscopic tumor sizee d25mm 214 61 (28.5) 0.000015 >25mm 222 106 (47.7)

RE (estrogen receptor) status 114 51 (44.7) 0.0054 Negative 329 117 (34.6) Positive

RP (progesteron receptor) status 191 85 (44.5) 0.00094 Negative 252 83 (32.9) Positive

ERBB2 Negative 346 127 (36.7) NS Positive 97 41 (42.3)

Supplementary Table 5: Characteristics of the 443 primary breast tumors a Log-rank test. NS : not significant, b Scarff Bloom Richardson classification. c Available information about 434 patients. d Available information about 442 patients. e Available information about 436 patients.

Nature Structural & Molecular Biology: doi:10.1038/nsmb.2390  ƒz 



   

 .+







         %       





       &A@(+(@&7"&A".D(&A%7A"&D@D."5&A97D@(>&@D9@.(&A5+.7%&5d@D."5&a

 ƒ{  7A "&D @D."5&^ 3& 6& AE.A .7D(@&AA( E @;5& %&  &D { %7A 5 D@7A.D.97 (>.D-(5.9j6(A&7"-L6D&EA&a 9ED %d!9@% 3d. %(697D@( ?E& 5&A  -(5."A&A  &D {"97D@;5&7DE7>@9,@66&D@7A"@.>D.977&5I&"5&A>@9D(.7&A6%A>9E@"97D@;5&@ 5d&K>@&AA.97%&A+"D&E@A A^%&A@(,E5D&E@A63&E@A%&5d^.7A.?E&5d&K>@&AA.97%& 6."@9A A>(".+.?E& %& 5 %.++(@&7D.D.97a 9@A %E >@9"&AAEA %& %.++(@&7D.D.97^ "&A 6."@9A@(,E5&7D5d&K>@&AA.97%&&D{%7AE7&!9E"5&%&@(,E5D.977(,D.I&a 7+.7 5 !.AA& %d&K>@&AA.97 %&  &D { "97D@;5& E7 >@9,@66& %d(>.AA,& ?E. "97D@.!E&  5c"?E.A.D.97 %& >-(79DL>&A 69@>-959,.?E&A A>(".+.?E&A %& 5 D@7A.D.97 (>.D-(5.9j6(A&7"-L6D&EA&a &A %.++(@&7DA @(AE5DDA 97D (,5&6&7D (D( @&D@9EI(A %7A 5 %.++(@&7".D.97 6L9,(7.?E&a 5A AE,,'@&7D ?E& 5&A  -(5."A&A  &D { A97D %&A 9@"-&AD@D&E@A%&5%.++(@&7".D.97&7"97D@;57D%&67.'@&%L76.?E&%.++(@&7DA7.I&EK %& 5d&K>@&AA.97 %&A ,'7&Aa 97 D@I.5 "d&AD +9"5.A( &AA&7D.&55&6&7D AE@ 5& 69%'5& %& 5 D@7A.D.97>.D-(5.9j6(A&7"-L6D&EA&&DAE@5&A@(,E5D.97AE7.I&E%&5d(>.AA,&%7A5&A %.++(@&7DA69%'5&Aa

Diff medium DDX5 DDX17 Myod Transcription Splicing

miR-1/206 Bnip2, Gripap1, Cttn, Myotubes Myoblasts Myog, Mef2c, Myh2… Tbc1d1, Tjp1… Gene expression Gene expression Lamellipodia program A program B Epithelial cell SNAIL1/2, CDH1, OCLN... ENAH, ARFGAP2, INSR, Mesenchymal FNIP1, CTTN... Loss of epithelial junctions miR-181b cell

Transcription Splicing SMADs DDX5 TGFβ DDX17

 1&!2313'-, 8,+'04! !2 '$$#1!,32,'5!47 !1#%4*3'-, !2%",!2/1*!2 &#*'2!2N!3JP4-412 !*A!3* '$$#1!,'3'-,+8-%#,'04!@ 

 ƒ| Title: RNA helicases dynamically orchestrate transcription, microRNA and splicing programs in cell differentiation      Authors: Etienne Dardenne1-4†, Micaela Polay Espinoza1-4†, Laurent Fattet1-4†‡, Lise Gratadou1-4, Helen Neil1-4, Mathieu Deygas1-4, Eleonora Zonta1-4, Samaan Samaan1-4, Sophie Germann1-4, Fatima Zahra Chakrama1-4, Martin Dutertre1-4, Ruth Rimokh1-4, Cyril F. Bourgeois1-4* and Didier Auboeuf1-4*      Affiliations: 1INSERM U1052, Centre de Recherche en Cancérologie de Lyon, F-69008 Lyon, France ; 2CNRS UMR5286, Centre de Recherche en Cancérologie de Lyon, F-69008 Lyon, France ; 3Université Lyon 1, F-69000 Lyon, France ; 4Centre Léon Bérard, F-69008 Lyon, France

*Correspondence to: [email protected] and [email protected] Centre de Recherche en Cancérologie de Lyon, Centre Léon Bérard, 28 Rue Laënnec, 69008 Lyon, France. Phone: +33.4.26.55.67.46

p-&A&ED-9@A"97D@.!ED&%&?E55LD9D-.AJ9@4 qE@@&7D%%@&AA`&>@D6&7D9+-@6"959,L^[email protected]9+5.+9@7.^7.&,9mn

 ƒ} /0.0; &7& &K>@&AA.97 >@9,@6A^ @&,E5D&% D 6E5D.>5& 5&I&5A !L D@7A"@.>D.97^

5D&@7D.I&A>5.".7,^7%6."@9A^%&+.7&D-&7DE@&7%+E7"D.97A9+7L,.I&7"&557%

@&%&&>5L69%.+.&%%[email protected],"&55%.++&@&7D.D.97a 9JD-&6E5D.>5&5L&@A9+,&7&&K>@&AA.97

@& "99@%.7D&% .A >99@5L E7%&@AD99%a &@& J& A-9J .7 DJ9 %.AD.7"D A>&".&A 7%

%.++&@&7D.D.97 >@9"&AA&A D-D D-& -.,-5L "97A&@I&%  !9K  -&5."A&A  7%

{^+.@AD6&%.D&%.++&@&7D.D.97jA>&".+."D@7A"@.>D.97>@9,@6A7%%.@&"D5L"97D@95D-&

&K>@&AA.97 9+ 4&L %.++&@&7D.D.97 6.Aa &"97%^ %[email protected], D-& &7AE.7, %.++&@&7D.D.97

>@9"&AA^D-&A&6.A%9J7@&,E5D&7%{.77&,D.I&+&&%!"4599>a-.@%^

D-& @&AE5D.7, %9J7@&,E5D.97 9+  7% { %.@&"D5L &5.".DA %.++&@&7D.D.97jA>&".+."

A>5.".7, >@9,@6A D-D +E@D-&@ "97D@.!ED& D9 D-& "?E.A.D.97 9+ 69@>-959,."5 >-&79DL>&Aa

-&@&+9@&^ J& .%&7D.+.&%  7&J DL>& 9+ %.++&@&7D.D.97 +"D9@A^ "9.7&% e6AD&@

9@"-&AD@D9@Af^ D-D @& 79D D-&6A&5I&A "&55 DL>&jA>&".+." !ED %.@&"D5L "97D@95 A&I&@5

%.++&@&7D.D.97jA>&".+." 6AD&@ @&,E5D9@A^ 7% 9@"-&AD@D& ,&7&D." ".@"E.D@.&A "-@"[email protected]&%

!L"A"%&9+D@7A"@.>D.97^6.7%A>5.".7,>@9,@6Aa



$)040;

@7A"@.>D96& @&>@9,@66.7, %[email protected], "&55 %.++&@&7D.D.97 .7I95I&A 6E5D.>5& 5L&@A 9+

@&,E5D.979+,&7&&K>@&AA.97.7"5E%.7,D@7A"@.>D.97^5D&@7D.I&A>5.".7,D-D&K>7%AD-&

,&7&6&AA,&^7%6.J-."-@&,E5D&6&AA&7,&@AD!.5.DL7%D@7A5D.97mJ>Kn^L&D

-9JD-&A&%.++&@&7D5L&@A@&"99@%.7D&%.AAD.55>99@5LE7%&@AD99%a&I&@5!9K

-&5."A&A 5.4&  7% .DA >@59, { @& -.,-5L "97A&@I&% >@9D&.7A D-D +E7"D.97 .7

69AD AD&>A 9+ D-& ,&7& &K>@&AA.97 >@9"&AA^ &I&7 D-9E,- D-&.@ D@,&DA 7% "D.I.D.&A .7

"96>5&K!.959,."5>@9"&AA&A@&79DJ&55"-@"[email protected]&%mLCNna9D->@9D&.7A@&"9@&,E5D9@A

9+ A&I&@5 D@7A"@.>D.97 +"D9@A^ .7"5E%.7,^ D-& 6AD&@ @&,E5D9@ 9+ 6EA"5& %.++&@&7D.D.97

 ƒ~ mOn^7%>@9D&.7AmPn^J-."-6&%.D& !j.7%E"&%&>.D-&5.5D96&A&7"-L65

D@7A.D.97mn^'@!@D-&D@7A%.++&@&7D.D.979+&>.D-&5.5"&55AD9+.!@9!5ADAmQna7%

{ @& 5A9 "96>97&7DA 9+ D-& A>5."&9A96& 7% @&,E5D& 5D&@7D.I& A>5.".7, mRCJJna

.755L^  7% { "97D@95 D-& !.9,&7&A.A 9+ A>&".+." 6.A^ .7 >@D I. D-&.@

.7D&@"D.97J.D-D-&@9A-i ‚"96>5&KmJK>JLna 7D-.A@&>9@DJ&D&AD&%D-&-L>9D-&A.A

D-D  7% { "D.I&5L "99@%.7D& D-&A& AD&>A 9+ ,&7& &K>@&AA.97 %[email protected], "&55

%.++&@&7D.D.97a



6 27 '% ''! 1&#'% &%#'! #%!% &

9 %%@&AA D-& +E7"D.97 9+ i{ .7 ^ J& A.5&7"&% D-&.@ &K>@&AA.97 .7

{z&>.D-&5.5"&55A^EA.7,7A.D@,&D.7,"97A&@I&%A&?E&7"&.7!9D-N7%

JPD@7A"@.>DA7%J-."-%.%79D++&"D>@9D&.7&K>@&AA.97m+.,a{nai{

479"4%9J7 >@.9@ D9  ! D@&D6&7D .7-.!.D&%^ D !9D-  7% >@9D&.7 5&I&5A^ D-&  !j

.7%E"&%7%|i}j%&>&7%&7D&K>@&AA.979+ J7% K>DJ96AD&@@&,E5D9@A9+

mQnm.,a{7%+.,a{na7%{.7D&@"D&%J.D->@9D&.7A7%

JA@&"@E.D&%97 J7% K>@969D&@A.7@&A>97A&D9 !m+.,a{^^na."@9@@L

75LA.A @&I&5&% D-D  5@,& 7E6!&@ 9+  !j@&,E5D&% ,&7&A J&@& %&>&7%&7D 97

i{ &K>@&AA.97 m.,a {^ ^^ !5&A { 7% |na 697, D-& ,&7&A D-D J&@& 79D

>@9>&@5L@&,E5D&%!L !.7i{j%&>5&D&%"&55A^A&I&@5@&.7I95I&%.7&>.D-&5.5

"&553E7"D.97Am!@%@^""5E%.77%j%[email protected]^.,a{7%+.,a|na97A.AD&7D5L^D-& j

%&>&7%&7D 59AA 9+ &>.D-&5.5 "&55 3E7"D.97A .7%E"&% !L  ! mJMn JA "96>@96.A&% .7

i{j%&>5&D&% "&55A m.,a { 7% +.,a |na -EA^  7% { @& 4&L

D@7A"@.>D.975"9@&,E5D9@A9+A6&%.D.7,D-&.7%E"D.979+!L !a

 ƒ .4&J.A&^ J& %%@&AA&% D-& j@&5D&% +E7"D.97 9+  7% { .7 A4&5&D5

6L9!5AD %.++&@&7D.D.97 EA.7, D-& 69EA& >5E@.>9D&7D |{| "&55 5.7&a KD&7%.7, >@&I.9EA

9!A&@ID.97A mOn^ %K .7D&@"D&% J.D- L9% .7 7E"5&. m+.,a }na 7%^ 5D-9E,- .D %.% 79D

++&"D L9% &K>@&AA.97 m+.,a }n^ D-& 479"4%9J7 9+  7NB 7JP .7-.!.D&% D-& L9%j

%&>&7%&7D&K>@&AA.979+8-%7%!$K,&7&A>DJ96AD&@@&,E5D9@A9+6L9,&7&A.AmJNn m.,a{7%+.,a}na."@9@@L75LA.A7%.66E795!&5.7,AALA@&A>&"D.I&5L@&I&5&%

D-D %Ki%K{ %&>5&D.97 .6>.@&% D-& &K>@&AA.97 9+  5@,& AE!A&D 9+ ,&7&A @&,E5D&%

%[email protected],6L9!5AD%.++&@&7D.D.97m.,a{^^ n7%"96>@96.A&%6L9DE!&+9@6D.97m.,a{ n^

"[email protected],D-&.@4&L+E7"D.97.76L9,&7&A.AmOna



%('!%+!!# )!)  %! & 6027

I&7 D-9E,-  7% { J&@& @&?E.@&% +9@ "&55 %.++&@&7D.D.97 m.,a {n^ D-&.@

>@9D&.7 5&I&5A %&"@&A&% %[email protected], !9D- |{| 7% {z %.++&@&7D.D.97 m.,a |na

-L59,&7&D."55L "97A&@I&% !.7%.7, A.D&A +9@ DJ9 6.A 479J7 D9 !& E>@&,E5D&% %[email protected],

6L9,&7&A.A^ 6.j{ 7% 6.j|z€ mJOn^ @& >@&%."D&% J.D-.7 D-& }d E7D@7A5D&% @&,.97 m}d

n 9+ !9D- -E67 7% 69EA& N 7% JP 6A m+.,a ~^ ^na [email protected],5L^

9I&@&K>@&AA.979+>@&j6.j{9@>@&j6.j|z€.7|{|"&55A@&%E"&%%K7%%K{>@9D&.7

5&I&5A m.,a |^ 5&+D >7&5na E".+&@A& AALA "97+.@6&% D-D D-.A &++&"D JA %&>&7%&7D 97

%.@&"D !.7%.7, 9+ 6.j{i|z€ 97D9 >@&%."D&% A.D&A m@.,-D >7&5 9+ .,a | 7% +.,a ~na

&6@4!5L^D-&.7%E"D.979+6DE@&6.j{7%6.j|z€7%9+D-&.@>@&"E@A9@A>@.j6.j{{

7% >@.j6.j|z€ %[email protected], |{| "&55 %.++&@&7D.D.97 mJPn JA >@&I&7D&% !L %Ki%K{

%&>5&D.97m.,a|n^.7%."D.7,D-DD-&>@.6@L&++&"DJAD@7A"@.>D.975.7-.!.D.979+D-&A&

6. ,&7&Aa -& >@D.5 &++&"D 97 >@.j6.j|z€ AE,,&AD&% D-D %Ki%K{ "9E5% 5A9

"97D@95D-&>@9"&AA.7,9+6.j|z€>@&"E@A9@A^.7,@&&6&7DJ.D->@&I.9EA@&>9@DAmJK>JLna  ƒ€ [email protected],5L^ D-& %&"@&A&% &K>@&AA.97 9+ i{ %[email protected],  !j.7%E"&% {z

D@7A%.++&@&7D.D.97m.,a|n.A6&%.D&%!LA.6.5@+&&%!"4@&,E5D.97599>a 7%&&%^6.j

{‚{!^D-D.A.7I95I&%.7 !A.,75.7,mJQn^%9J7@&,E5D&%>@9D&.7&K>@&AA.97m.,a

|n I.  %.@&"D !.7%.7, D9 N }d m.,a | 7% +., ~na .j{‚{! 5A9 @&>@&AA&%

&7%9,&79EA { &K>@&AA.97 m.,a |n^ &I&7 D-9E,- 79 6.j{‚{! !.7%.7, A.D& JA

>@&%."D&%J.D-.7JP}dm+.,a~n^AE,,&AD.7,&.D-&@797j"797."5.7D&@"D.979@7

.7%.@&"D @&,E5D.97 !L 6.j{‚{!a 9@&9I&@^  !j.7%E"&% mJQn^ |i}j%&>&7%&7D 6.j

{‚{! &K>@&AA.97 .7 {z "&55A m.,a | 7% +.,a ~n JA >@&I&7D&% !L i{

%&>5&D.97m.,a|7%+.,a~na9D-7%D-&"9jm~nJ&@&@&"@E.D&%97D-&

>@969D&@9+D-&6.j{‚{!j{,&7&>@969D&@m.,a|n^+E@D-&@%&697AD@D.7,%.@&"D@95&9+

.76.j{‚{!.7%E"D.97a

5D9,&D-&@^ 9E@ @&AE5DA @&I&5&%  7&,D.I& +&&%!"4 599> 9+ @&,E5D.97 %[email protected], !9D-

6L9!5AD %.++&@&7D.D.97 7%  !j.7%E"&% ` i{ >@969D& D-& &K>@&AA.97 9+

A>&".+."6.A^J-."-.7DE@7%.@&"D5L%9J7@&,E5D&i{&K>@&AA.97a 7%%.D.97^

D-&A& 6.A 6L -I&  !@9%&@ .6>"D 97 %.++&@&7D.D.97^ A A-9J7 +9@ 6.j{i|z€ .7

6L9!5ADA mJRna ""9@%.7,5L^ D-& D@7A"@.>DA %9J7@&,E5D&% .7  %Ki%K{j%&>&7%&7D

677&@%[email protected],|{|"&55%.++&@&7D.D.97@&&7@."-&%.7>@&%."D&%6.j{i|z€!.7%.7,A.D&A m+.,a~na-&%.@&"D@&,E5D.979+!9D-"9%.7,m.,a{n7%797j"9%.7,m.,a|n6AD&@,&7&A

9+ %.++&@&7D.D.97 !L i{ &6>-A.M&A D-&.@ 4&L "99@%.7D.7, @95& .7 D-& "96>5&K

"&55E5@>@9"&AAa



!* %('! !6027%)&% ''! 1&#&# #%!% &

.7"&  7% { @& 5A9 .7I95I&% .7 A>5.".7,^ J& D&AD&% J-&D-&@ D-&.@

%9J7@&,E5D.97 "9E5% ++&"D 5D&@7D.I& A>5.".7, .7 |{| 7% {z "&55Aa >5.".7, 9+  ƒ A&I&@5 -E7%@&%A 9+ "AA&DD& &K97A JA 5D&@&% E>97 i{ 479"4%9J7 .7

E7%.++&@&7D.D&%|{|9@{z"&55Am.,a}7%+.,a^!5&A}7%~n^.7"5E%.7,

AE!,@9E>9+9@D-959,9EA&K97A@&,E5D&%A.6.5@5L.7!9D-"&55DL>&Am.,a}na

9%&697AD@D&D-&%.@&"D.7I95I&6&7D9+7%{.7D-& A>5.".7,9+&K97A

@&,E5D&% .7 !9D- |{| 7% {z "&55A^ @&A"E& &K>[email protected]&7DA J&@& >&@+9@6&% EA.7,

A.AD@,&D.7,D-&}d9+&7%9,&79EAN7%JPD@7A"@.>DA.7AD!5&"&55

5.7&AmJJ>KIna-&A>5.".7,5D&@D.97.7%E"&%!Li{%&>5&D.97JA@&A"E&%E>97

&K>@&AA.97 9+ J.5%jDL>&^ !ED 79D -&5."A&j6EDD&%  9@ { m.,a } 7% +.,a €na

&KD^ D@7A+&"D&% 6.7.,&7&A "97D.7.7, D-& -E67 J9@ Q "AA&DD& &K97A

@&">.DE5D&%D-&&7%9,&79EAA>5.".7,>DD&@77%J&@&A.6.5@5L@&,E5D&%!Li{ m+.,a €na -.@%^ jD,,&% m.,a }n 9@ &7%9,&79EA  m+.,a €n "9E5% !&

.66E79>@&".>.DD&% D9,&D-&@ J.D- .DA D@,&D A^ >@&+&@&7D.55L D D-& 5&I&5 9+ @&,E5D&%

&K97Aa

7D&@&AD.7,5L^AE!A&D9+D-&&K97A@&,E5D&%%[email protected],6L9,&7&A.A.79E@ADE%Lm.,a}n

7% .7  >@&I.9EA ADE%L mKJn^ J&@& @&,E5D&% A.6.5@5L !L  7NB 7JP A.5&7".7, .7

E7%.++&@&7D.D&%|{|"&55Am.,a}7%+.,ana.4&J.A&^AE!A&D9+D-&&K97A@&,E5D&%

%[email protected],  !j.7%E"&%  m.,a }n mKK> KLn J&@& @&,E5D&% A.6.5@5L !L i{

%&>5&D.97 .7 E7D@&D&% {z "&55A m.,a } 7% +.,a na D-&@ %.++&@&7D.D.97jA>&".+."

I@.7DAJ&@&E7++&"D&%!LNBJP479"4%9J7m.,a}^7%+.,a^naE>>[email protected],

 69%&5 J-&@& 6.j.7%E"&% i{ %9J7@&,E5D.97 %@.I&A  A>&".+." A>5.".7,

>@9,@6.7%.++&@&7D.D.7,"&55A^&"D9>."9I&@&K>@&AA.979+6.j{9@6.j|z€.7|{|"&55A m.,a}7%+.,an^9@6.j{‚{!.7{z"&55Am.,a}7%+.,an^.7%E"&%A.6.5@

A>5.".7,>DD&@7D-7"&55%.++&@&7D.D.977%i{%&>5&D.97a

 ƒ‚ 7D&@&AD.7,5L^  AE!A&D 9+ ,&7&A J-9A& A>5.".7, JA A.6.5@5L @&,E5D&% E>97

i{ %9J7@&,E5D.97 9@ "&55 %.++&@&7D.D.97^ @& .7I95I&% .7 "D.7 "LD9A4&5&D97

%L76."A m+.,a ‚na &6@4!5L^ i{ %&>5&D.97 .7 !9D- |{| 7% {z "&55A

.7%E"&% D-& +9@6D.97 9+ 56&55.>9%.^ "&55E5@ >@93&"D.97A D-D @&?E.@&  >@9+9E7%

@&69%&5.7, 9+ D-& "LD9A4&5&D97 %L76."A m.,a }  7% 9I.&A {jna 6&55.>9%. @&

.7I95I&%.7D-&6.,@D.977%+EA.979+6L9!5ADAmKMn^7%@&5A9&AA&7D.5+9@7%

6&A&7"-L65 "&55 69D.5.DL mKNna 7 A>.D& 9+ 9E@ &++9@DA^ J& %.% 79D AE""&&% .7 &5.".D.7, D-&

+9@6D.97 9+ 56&55.>9%. !L 67.>E5D.7, A.7,5& A>5.".7, &I&7DAa D-&@^ 9E@ %D AE,,&AD

D-Di{%9J7@&,E5D.97"97D@.!ED&AD956&55.>9%.+9@6D.979J.7,.7>@DD9D-&

"97"&@D&% @&,E5D.97 9+  7&DJ9@4 9+ 5D&@7D.I& &K97A m+.,a ‚n^ 5D-9E,- .D .A 5.4&5L D-D

D@7A"@.>D.97j7%6.j6&%.D&%&++&"DA@&5A9.7I95I&%a

5D9,&D-&@^9E@@&AE5DAE7@I&5%L76."9@"-&AD@D.979+%.++&@&7D,&7&&K>@&AA.97

5L&@A!LD-&-&5."A&A7%{%[email protected],"&55%.++&@&7D.D.97a-&i{j

%&>&7%&7D,&7&D."".@"E.D@.&AD-DJ&"-@"[email protected]&%.7DJ9%.++&@&7D!.959,."5>@9"&AA&A7%

.7|%.++&@&7DA>&".&A@&@&6@4!5LA.6.5@m.,a~n7%J&>@9>9A&D9%&+.7&AE"-+"D9@AA

6AD&@9@"-&AD@D9@A9+%.++&@&7D.D.97aD-&@-&5."A&AAE"-A}^~i7%

 ƒi D-D@&5A9.7I95I&%D6E5D.>5&5&I&5A9+,&7&&K>@&AA.977%"97D@.!ED&D9

I@.&DL9+!.959,."5>@9"&AA&AmL>KO>KPn^6L5A9-I&!@9%+E7"D.97.7"99@%.7D.7,D-&

,&7&D.">@9,@6AD-DE7+95%%[email protected],!.959,."5D@7A.D.97A9@"&55%.++&@&7D.D.97>@9"&AA&Aa



$"1.,0$*)/

$"<G<*."1'0$*)*!0.)/.$,0$*),.*".(/5K)GM1.$)"$!!.)0$0$*)<

ABK>@&AA.979+ J7% K6A.7-E67{z"&55A+D&@A.5&7".7,9+K

7% LmA.KBLn 9@ N 7% JPmA.NBJPn^ +9559J&% 9@ 79D !L  |~ -  ƒƒ D@&D6&7DJ.D- !aeA."D@5f`"97D@95A.aABE6!&@9+,&7&A@&,E5D&%%[email protected], !j

.7%E"&%  D D-& ,59!5 &K>@&AA.97 5&I&5a AB 5.%D.97 9+ 6."@9@@L %Da AB

66E795!&55.7,9+{z"&553E7"D.97AJ.D-77D.jj"%[email protected]7D.!9%La"5&!@`|z

Ž6aABK>@&AA.979+8-%7%!$K6A.7|{|69EA&6L9!5AD"&55A+D&@A.5&7".7,

9+8- mA.8- n9@ 7N7% 7JPmA. 7NBJPn^+9559J&%!L.7"E!D.97.7,@9JD-mj.++n

9@ %.++&@&7D.D.97 m.++n 6&%.E6a eA. "D@5f` "97D@95 A.a AB E6!&@ 9+ ,&7&A @&,E5D&%

%[email protected],6L9!5AD%.++&@&7D.D.97DD-&,59!5&K>@&AA.975&I&5aAB5.%D.979+6."@9@@L

%DaA B 66E795!&55.7,9+|{|"&55AJ.D-77D.j6L9A.7-&IL"-.7m n7D.!9%La

"5& !@ ` |z Ž6a .AD9,@6A .7 ^ ^  7%  @&>@&A&7D 6&7 I5E&A 9+ D 5&AD D-@&&

.7%&>&7%&7D&K>[email protected]&7DAm,}n79@65.M&%D9D-&"97D@95A6>5&‹m>.@&%DE%&7DdA3j

D&AD`oŠzaz_ooŠzaz{na



$"GM 4,.//$*)

1.$)"$!!.)0$0$*)<

ABK>@&AA.979+7%{%[email protected],%.++&@&7D.D.97D.6&j"9E@A&9+|{|"&55Am5&+D

>7&5n7% !j.7%E"&%9+{z"&55Am@.,-D>7&5na"D.7`59%.7,"97D@95aAB&+D

>7&5`&K>@&AA.979+%K7%%K{.7|{|"&55A9I&@&K>@&AA.7,>@&j6.j{^>@&j6.j|z€

9@"97D@95>@&j6.a.,-D>7&5`5E".+&@A&AALA6&[email protected],D-&&++&"D9+6.j{7%6.j

|z€97J.5%jDL>&7%6EDD&%69EA& 7N}da .AD9,@6A@&>@&A&7D6&79+D5&AD

}.7%&>&7%&7D&K>[email protected]&7DA‹mE7>.@&%DE%&7DdA3jD&AD`oŠzaz_ooŠzaz{_oooŠ zazz{naABK>@&AA.979+6.j{7%6.j|z€.7|{|"&55AaDE@&6.mL&559Jn9@>@.j

6.m9@7,&nJ&@&?E7D.+.&%|~-+D&@%.++&@&7D.D.979+"&55A>@&jD@&D&%J.D-"97D@959@

 7NBJPA.a5E&AJ&@&79@65.M&%D9E7%.++&@&7D.D&%"&55Aa‰}‹_oŠzaz_oo

Šzaz{m>.@&%3jD&ADnaAB&+D>7&5`&K>@&AA.979+7%{.7 !jD@&D&%{z

 {zz "&55A 9I&@&K>@&AA.7, >@&j6.j{‚{! 9@  "97D@95 >@&j6.a .,-D >7&5` 5E".+&@A& AALA

6&[email protected],D-&&++&"D9+6.j{‚{!97J.5%jDL>&7%6EDD&%-E67N}d.7 !j

D@&D&% "&55Aa&D.5A@&A.7aABK>@&AA.979+6DE@&6.j{‚{!m5.,-D!5E&n7%

>@.j6."@9j{‚{!m%@4!5E&n.7{z"&55AD@&D&%9@79DJ.D- !+9@|~-a&D.5A@&A

.7aAB-@96D.7.66E79>@&".>.DD.97&K>[email protected]&7DA-9J.7,~7%!.7%.7,D9

D-& 6.j{‚{! >@969D&@a -& D&AD&% @&,.97 &7"96>AA&A  >@&%."D&%  !.7%.7, A.D&

J.D-.7D-&6.j{‚{!{j&7"9%.7,,&7&>@969D&@m5&+D%.,@6na&AE5DA@&&K>@&AA&%AD-&

6&7>&@"&7D,&9+.7>EDm,‰}‹n79@65.M&%D9D-&"97D@95A6>5&a



$"<I<K>GM*)0.*'$!!.)0$0$*)?/,$!$/,'$$)",.*".(/<

AB&77%.,@6A-9J.7,D-&7E6!&@9+5D&@7D.I&&K97A@&,E5D&%!L7%{.7

|{|"&55AmL&559J7%@&%n7%i9@.7{z"&55Am5.,-D7%%@4!5E&naE555.ADA9+&K97A

@&,.I&7.7!5&A}7%~aAB5.%D.979+6."@9@@L%Da75L&K97A@&,E5D&%.7D-&

A6&677&@.7!9D-"&555.7&A@&A-9J7a9@&"-&I&7D^D-&&K977E6!&@.AA-9J7.7@&%

+9@ D-& 69EA& ,&7& m5&+Dn 7% .7 !5E& +9@ D-& -E67 ,&7& m@.,-Dna AB >5.".7, @&A"E&

&K>[email protected]&7DAaD!5&"&55AJ&@&D@7A+&"D&%J.D-A.NiJPm}dn7%D@&D&%

J.D- %9KL"L"5.7 D9 .7%E"& D-& &K>@&AA.97 9+ J.5%jDL>& 9@  -&5."A& 6ED7D j  m57&A{j€n9@ j{m57&Aj{|na75L97&@&>@&A&7DD.I&&K>[email protected]&7D.AA-9J7m,‰}na

E7D.+."D.97 .A A-9J7 A  >&@"&7D,& 9+ 5D&@7D.I& A&?E&7"& .7"5EA.97 m>A.na AB 

.66E79>@&".>.DD.97&K>[email protected]&7DA-9J.7,D-&@&5D.I&!.7%.7,9+ jD9I@.9EA&K97A

.7 AD!5&  "&55Aa j ` E7.7%E"&% _ ‡ ` %9KL"L"5.7jD@&D&%a -& @&,.97 75LM&%

9I&@5>AD-&A>5.".7,%979@A.D&9+&"-D&AD&%&K97a5E&&K97A"9@@&A>97%D9i{j

@&,E5D&%&K97A^J-.5&!5"4&K97A@&E7@&,E5D&%&K97A+@96D-&A6&,&7&a-&@&AE5DA mD&"-7."5@&>5."D&A^,‰}‹n@&&K>@&AA&%A>&@"&7D,&9+.7>EDaAB5D&@7D.I&

 {z{ A>5.".7, 9+ &K97A E>97 %Ki%K{ %9J7@&,E5D.97 %[email protected], %.++&@&7D.D.97 9+ |{| "&55Aa

&+D` "&55A .7%E"&% D9 %.++&@&7D.D& 9@ D@&D&% J.D- A.NBJPa .,-D` "&55A &"D9>."55L

&K>@&AA.7,6.j{9@6.j|z€A%&A"@.!&%.7.,E@&|aE7D.+."D.97.AA-9J7AD-&+95%

"-7,& 9+ >A. @&5D.I& D9 D-& "97D@95 A6>5&a AB 5D&@7D.I& A>5.".7, 9+ &K97A E>97

i{%9J7@&,E5D.97%[email protected],9+{z"&55Aa&+D`"&55AD@&D&%J.D- !9@

J.D- A.NBJPa .,-D` "&55A &"D9>."55L &K>@&AA.7, 6.j{‚{! A %&A"@.!&% .7 .,E@& |a

&D.5A@&A.7aAB9@6D.979+56&55.>9%..7i{j%&>5&D&%"&55Aa9>`-A&j

"97D@AD.6,&A9+{z"&55AD@7A+&"D&%J.D-"97D@959@NBJPA.7%D@&D&%9@

79DJ.D- !a9DD96`|{|"&55AD@7A+&"D&%J.D-"97D@959@NBJPA.a"5&!@`

zŽ6a 7A&D`6,7.+.&%I.&J9+56&55.>9%.mJ-.D&@@9JAn.7 !jD@&D&%{z"&55A

7%.7!9D-i{j%&>5&D&%"&555.7&AmA"5&!@`|Ž6na



$"< J< .#/0.0 ."1'0$*) *! ") 4,.//$*) 5 K ) GM 1.$)" ''

$!!.)0$0$*)a

-&D9>m@&%n7%!9DD96m!5E&nA"-&6&A@&>@&A&7DD-&%.++&@&7D.D.979+69EA&6L9!5ADA m|{|n .7D9 6L9DE!&A 7% D-&  !j.7%E"&%  9+ {z "&55A^ @&A>&"D.I&5La >97

.7%E"D.97 9+ %.++&@&7D.D.97^  7% { "9@&,E5D& j 9@ j%&>&7%&7D

D@7A"@.>D.975>@9,@6A"[email protected],4&L&++&"D9@A9+%.++&@&7D.D.97m&K6>5&A@&.7.D5."Ana

>&".+." E>@&,E5D&% 6.A %.@&"D5L @&>@&AA D-& &K>@&AA.97 9+  7% {^ J-."-

"97D@.!ED&AD9D-&&AD!5.A-6&7D9+"&55jA>&".+."5D&@7D.I&A>5.".7,>@9,@6.7!9D-"&55

DL>&Am&K6>5&A@&.7.D5."Ana7&A-@&%"-@"[email protected]."9+!9D-i{j%&>5&D&%"&55

5.7&A.AD-&+9@6D.979+56&55.>9%.^J-."-"9E5%@&AE5D+@96D-&"96!.7&%@&,E5D.979+

A&I&@5A>5.".7,&I&7DA.7,&7&A.7I95I&%.7"LD9A4&5&D97%L76."Aa



 {z| 

!.)/) *0/;

{a aa!&@D^aa-@>a!**GJO^{j|~m|z{|na

|a a 5A9D@^aa99>&@a3@!5@!,!3@GH^{j|ƒm|z{{na

}a a.7%&@^a 749JA4La3@!5@-*@!**@'-*@GH^zj{€m|z{{na

~a a 747&"-Da+@ @1,2*@!2@H^||}j|}~m|z{zna

a aaE55&@j"&^ aa99@&a4341!,-*@M^|}ƒj|{m|z{{na

€a a@&DD.!3*@!5@!**GG^~j€zm|zz€na

a aa@7&@!3*@'-&!+@'-/&82@!2@-++4,@IHJ^zj€m|zz~na

‚a aa-.&@L^ a"59?E&^aa E7,^aa.&D9a!**GIO^‚{j‚ƒzm|zzƒna

ƒa a.7^a7,^ a a7,^a E7,^aa.Ea-*@!**@'-*@HK^~‚~j~ƒ}m|zzna

{za a@%&77&!3*@3@3143@-*@'-*@GO^{{}ƒj{{~€m|z{|na

{{a a &@677!3*@,-%!,!IG^~}€j~~ƒm|z{|na

{|a aE4E%!3*@3@!**'-*@O^€z~j€{{m|zzna

{}a aaI.A^aa .5L@%^ a,7^a Da341!JKJ^€j€{m|zz‚na

{~a a79!3*@3@!**'-*@H^€j‚}m|zzzna

{a a@E7^a ED&5a3@!5@-*@!**@'-*@GH^}~ƒj}€{m|z{{na

{€a a a .55.6A^ a .E^ a I7 99.3^ a a 5A97a 411@ /',@ !** '-*@ HG^ ~€{j~€ƒ

m|zzƒna

{a aa9495a411@-/@!5@'-*@OO^ƒj‚m|z{|na

 {z} {‚a a7,!3*@,-%!,!HO^{‚j{ƒm|z{zna

{ƒa a 9537&4j-LA55!3*@ @!**'@GHK^}ƒzj}€zzm|z{|na

|za a97D!3*@4*!'' 2!2@JG^~j€~m|z{}na

|{a aa57%!3*@4*!'' 2!2@IN^€{j€€~m|z{zna

||a aa@M&"-!3*@ @HO^}|‚€j}}zzm|z{zna

|}a aa->.@9!3*@ -!,!3@M^&{zz||{‚m|z{{na

|~a a9"-5.7^aE^a9L^a aL5.&Aa!5@'-*@IJG^€€j‚}m|z{zna

|a a.56M^ a[email protected][email protected],!1!3232'2!5@HN^{j}}m|zzƒna

|€a a!%&5-5&&6a*',@'-&!+@IN^~ƒƒjz}m|zzna

|a aA49a'-&'+@'-/&82@3GNHO^‚{zj‚{€m|z{}na

|‚a aDD&D!3*@ @!**'@GHL^}{}j}|}m|z{}na























 {z~ &)*3'"()0/

&@&,@D&+E5D9a[email protected],7% a@.&+9@D-&.@"966&7DA97D-&67EA"@.>D^7%D955

9E@"955&,E&A+9@.7A>[email protected],%.A"EAA.97Aa&D-74a7!&55&7%a9E"-@%97m& j

7D(L97jADn7%@ 9*5"-E&@m@9+.5&>&@Dj+".5.DL^L97^@7"&n+9@D-&.@

AA.AD7"& J.D- .6,.7, 7% 6."@9@@L 75LA&A^ @&A>&"D.I&5La -& %D @&>9@D&% .7 D-.A

@D."5&@&D!E5D&%.7D-&6.7>>&@7%.7D-&E>>5&6&7D@L975.7&D&@.5Aa-.AJ9@4

JA AE>>9@D&% !L ,@7DA +@96 7A&@6^ ^ ^ 7%  7% ^ +@96 %9"D9@5

+&559JA-.>A+@96maa7%aan^maaan^D-&+@&7"-.7.AD'@&%&5r7A&.,7&6&7D

E>(@.&E@&D%&5&"-&@"-&maan7%maan^7%+@96>9ADj%9"D9@5+&559JA-.>A+@96

D-&&7D@&(97(@@%m aaan7%97%D.97ma a7%aaana



 {z 1,,'()0.5 0.$'/

0.$'/) 0#*/

''1'01.

|{| "&55A mn J&@& "E5DE@&% .7 &.D-&@ ,@9JD- 6&%.E6 mE5!&""9dA 9%.+.&% ,5&dA

&%.E6 mn AE>>5&6&7D&% J.D- {z† +&D5 !9I.7& A&@E6^ {† 7D.!.9D."An 9@

%.++&@&7D.D.97 6&%.E6 m AE>>5&6&7D&% J.D- |† -9@A& A&@E6^ {† 7D.!.9D."Ana -&

{z 666@L &>.D-&5.5 "&55 5.7& JA 9!D.7&% +@96 D-& 6&@."7 L>& E5DE@&

955&"D.97a{z"&55AJ&@&"E5DE@&%.7{`{mni j{|6&%.E6m.+&&"-7959,.&An

AE>>5&6&7D&%J.D-†-9@A&A&@E6m6!@&Kn^{z7,i65-E67&>.%&@65,@9JD-+"D9@ m@969&55n^ za Z,i65 -L%@9"[email protected]& m.,6n^ {z Z,i65 .7AE5.7 m.,6n 7% {zz 7,i65

"-95&@ D9K.7 m.,6na  "&55A J&@& "E5DE@&% .7 {`{ i j{| 6&%.E6 m.+&

&"-7959,.&An AE>>5&6&7D&% J.D- {z† +9&D5 !9I.7& A&@E6 mn^ {z 7,i65 -E67

&>.%&@65 ,@9JD- +"D9@ m@969&55n^ za Z,i65 -L%@9"[email protected]& m.,6n 7% {z Z ,i65

.7AE5.7 m.,6na &"96!.77D  !j{ m&>@9D&"-n JA EA&% D |^ 7,i65a 9 !59"4 7L

A.,75.7, @.A.7, +@96 ED9"@.7& >@9%E"D.97 9+  X^ &K>[email protected]&7DA J.D-9ED  !j{ J&@&

>&@+9@6&% .7 D-& >@&A&7"& 9+ {z Ž X  4.7A& .7-.!.D9@ j~}{~| m.,6n +9@ D-&

.7%."D&%D.6&Aa &7%"&55AJ&@&"E5DE@&%.7AE>>5&6&7D&%J.D-{z†

D}vaj ^ {~~j ^ j{7% j{ {~|J&@&"597&%.7D9D-&

>|-L,I&"D9@m597D&"-nD9,&7&@D&.7%E".!5&ji&Dj7AD!5&"&555.7&Am597D&"-na

&A.AD7D "597&A J&@& A&5&"D&% J.D- -L,@96L".7 m}zz Z,i65^ 597D&"-n 7% >@9D&.7

&K>@&AA.97JA"-&"4&%+D&@%9KL"L"5.7&mnD@&D6&7Dm{Z,i65n+9@~‚-a





  {z€ .)/!0$*))./14,.$()0

@7A.&7D D@7A+&"D.97 9+ A.A 9@ >@&j6.A m@&j6. 6. @&"E@A9@A^ @&+&@&7"& w{{zz 9@ @&j6. 7&,D.I& "97D@95 w{^ &+&@&7"& w{{{z^ .+& &"-7959,.&An JA

>&@+9@6&% D  +.75 "97"&7D@D.97 9+ {z 7 EA.7, 5.>9+&"D6.7& . m.+&

&"-7959,.&AnA%&A"@.!&%>@&I.9EA5LmJJ>KIna&?E&7"&A9+A.A&?E&7"&A@&>@9I.%&%

.7E>>5&6&7D@L!5&na-&NBJPA.D@,&DAA&?E&7"&"97A&@I&%.7D-&|,&7&A

N7%JPa

9@ @&A"E& &K>[email protected]&7DA^ ji&Dj7  "&55A J&@& D@7A+&"D&% +9@ |~ - J.D- {z 7

A.D@,&D.7,N9@JPd9@}dmE>>5&6&7D@L!5&nEA.7,.>9+&"D6.7&

.m.+&&"-7959,.&Ana-&7&KD%L^"&55AJ&@&D@&D&%J.D-%9KL"L"5.7+9@~‚-.7

9@%&@D9.7%E"&D-&&K>@&AA.979+&>.D9>&jD,,&%7%{a



((1)*!'1*./)

&55AJ&@&+.K&%.7~†>@+9@65%&-L%&+9@|z6.7a+D&@}JA-&A.7{K^"&55AJ&@&

>&@6&!.5.M&%.7za|†@.D97j{zz+9@}z6.77%5&+D+9@{-.7!59"4.7,A95ED.97m{K^

{†A&@E6^za{†@.D97j{zzna@.6@L7D.!9%.&A` mj}|^.,6n^m{j{zz^

.+& &"-7959,.&An^ j"%[email protected] m€{z{‚{^ n 7% !j"D&7.7 m7D "@EMna @9D&.7A J&@&

@&I&5&%EA.7, j"973E,D&%7D.j69EA& , m.,6na

6,&"?E.A.D.97JA>&@+9@6&%EA.7,&.AA‚z9@&.""97+9"56."@9A"9>& m€}K9!3&"D.I&n7%.6,&AJ&@&>@9"&AA&%J.D-D-& 6,& >@9,@6a



&'(,.*4$($05'$"0$*)//5

|{| "&55A J&@& D@7A+&"D&% +9@ ~‚ - J.D- &.D-&@ "97D@95 9@ NBJP A.a +D&@

D@7A+&"D.97^,@9JD-6&%.E6JA@&>5"&%!L%.++&@&7D.D.976&%.E6+9@|~-a-&',2'34

 {z >@9K.6.DL5.,D.97AALmnJA"97%E"D&%EA.7,D-&E95.74  .Dm5.74.9A".&7"&nA

%&A"@.!&%>@&I.9EA5LmKIna



$*($.*/*,5

{z9@|{|"&55AJ&@&D@7A+&"D&%J.D-"97D@959@NBJPA.m+9@{7%|~-^

@&A>&"D.I&5Lna-&7.7%."D&%^{zJ&@&5A9D@&D&%+9@}}-J.D- !aE5DE@&>5D&A

J&@& D-&7 >5"&% .7 D-& .7"E!D.97 "-6!&@ m}v 7% † z|n 9+ 7   |zz

6."@9A"9>&a."DE@&AJ&@&D4&7&I&@L6.7+9@|~-m{zn9@ƒ-m|{|n+9@&"-+.&5%

7%+.75AD"4&%>."DE@&AJ&@&>@9"&AA&%EA.7, 6,& a



.*0$)) )'5/$/

@9D9D9"95A +9@ >@9D&.7 &KD@"D.97^  &KD@"D.97^ j 7% ?E7D.DD.I& j J&@&

%&A"@.!&% >@&I.9EA5L mJJ> KIna @.6&@ A&?E&7"&A +9@  7% ? @& >@9I.%&% .7

E>>5&6&7D@L !5& a @.6@L 7D.!9%.&A EA&% +9@ J&AD&@7j!59DD.7,`  m!{z|€{^

!"6n^ { m!|~€z{^ !"6n^ } mA"jƒ{}^ 7D @EMn^ | mA"j}{||^ 7D

@EMn^  { mA"j|‚{ƒƒ^ 7D @EMn^  |i  mA"jƒ‚^ 7D @EMn^ j"%[email protected] m€{z{‚{^ n^  m{j{zz^ .+& &"-7959,.&An^  m}{|^ 49n^   m}ƒ^

49n^  mz{z^9I5!n9@"D.7m j{ƒA"{€{€^7D@EMna

9@ 75LM.7, 6DE@& 6.A^ |z 7, 9+ D9D5  JA @&I&@A&jD@7A"@.!&% 7% ?E7D.+.&%

EA.7,D-&6.7.I&@A5"AL7D-&A.A7%u @&&74.DmK.?97n^EA.7,

D-&A>&".+.">@.6&@AA&DA+9@-Aj6.j{^-Aj6.j|z€7%-Aj6.j{‚{!j>^+9559J.7,

67E+"DE@&@dA .7AD@E"D.97Aa j? I5E&A J&@& 79@65.M&% @&5D.I& D9 € A7 m5A9

6&AE@&%EA.7,>@.6&@AA&D^K.?97na



 {z‚ #.*(0$))  ((1)*,.$,$00$*)

-@96D.7 7%  "-@96D.7 .66E79>@&".>.DD.97 m-  7%  n J&@& >&@+9@6&%

&AA&7D.55LA>@&I.9EA5L%&A"@.!&%mKI>KQnEA.7,7D.!9%.&A,.7ADm~Ž,^}zzj|}^

&D-L5n^~mA"j{~^7D@EMn9@"97D@95 , m7D@EMna@.6&@AEA&%+9@7%

.66E79>@&".>.DD.97@&%&A"@.!&%.7E>>5&6&7D@L!5&a



*?$((1)*,.$,$00$*)

"&55AJ&@&D@7A+&"D&%J.D-&K>@&AA.97I&"D9@A&7"9%.7,5,j}7%5,j~ m4.7%,.+D9+a&7.34&^&.%&7^&D-&@57%An+9@~‚-a&55AJ&@&D@&D&%J.D- !j{9@

79Dmj~}{~|D@&D6&7Dn+9@|-!&+9@&5LA.A.7!E++&@"97D.7.7,z6@.Ak 5> ‚az^

{z65^{†~z^{6^>@9D&A&7%>-9A>-DA&.7-.!.D9@Am.,6na5&@&%

"&55&KD@"DAJ&@&.7"E!D&%9I&@7.,-DJ.D-~Z,9+7D.!9%.&A!9E7%D9>@9D&.7 6,7&D."

!&%Am.+&&"-7959,.&Ana-&>@9D&.7j!9E7%6,7&D."!&%AJ&@&JA-&%D-@&&D.6&AJ.D-

{K^@&AEA>&7%&%.759%.7,!E++&@+9@J&AD&@7j!59DD.7,a 7>ED"9@@&A>97%AD9†9+

D-&I95E6&9+"5&@&%"&555LAD&Aa



1$!.///5/

-&+9559J.7,}d+@,6&7DAJ&@&j6>5.+.&%+@96"9@,&796."7%"597&%

.7D-&>A.  j|>5A6.%m@96&,n%9J7AD@&69+D-&@&7.555E".+&@A&,&7&`6 7N m.,a |!n^ +@,6&7D ƒj}{{ mlzz‚~zn _ 6 7JP  m.,a |!n ` +@,6&7D {}|j|||| ml{|‚z€n _ -N  m.,a |%n ` +@,6&7D ‚|j{€€ >5EA } %%.D.975 7E"5&9D.%&A

%9J7AD@&6 9+ D-& &7% 9+ }d  mlzz~}ƒ€n _ -JP  m.,a |%n {~ƒ|j|~}€ ml{zƒ‚z~naED7D%[email protected]D.I&AJ&@&9!D.7&%!L%&5&D.979+~7E"5&9D.%&AJ.D-.76.j

{i|z€ 7% 6.j{‚{! !.7%.7, A.D&A A .7%."D&% .7 E>>5&6&7D@L .,A ~ 7% ~! !L

 {zƒ %.@&"D&% 6ED,&7&A.Aa -& A6& >5A6.% 5A9 "97D.7A D-& +.@&+5L 5E".+&@A& ,&7& EA&% +9@

.7D&@7579@65.MD.979+D-&%Da

9@ D-& 75LA.A 9+ 6.j{i|z€ &++&"D^ & "&55A J&@& A&&%&% .7 |~jJ&55 >5D&A 7%

D@7A+&"D&%+9@|~-J.D-z7,9+5E".+&@A&@&>9@D&@597,J.D-{z7>@&j6.>@&"E@A9@ m"97D@95 9@ >@&j6.j{ 9@ >@&j6.j|z€^ .+& &"-7959,.&An^ EA.7, .>9+&"D6.7& |zzz m.+&

&"-7959,.&An +9559J.7, 67E+"DE@&@dA .7AD@E"D.97Aa 9@ 6.j{‚{!^ D-& >@9D9"95 JA D-&

A6& &K"&>D D-D  "&55A J&@& EA&% 7% >@&jD@&D&% +9@ |~ - J.D-  ! >@.9@ D9

D@7A+&"D.97a|~-+D&@D@7A+&"D.97^"&55AJ&@&5LA&%7%5E".+&@A&AALAJ&@&"@@.&%9ED

EA.7, D-& E5jE".+&@A&u &>9@D&@ AAL LAD&6 m@96&,n^ A @&"966&7%&% !L D-&

67E+"DE@&@aDJ&@&&K>@&AA&%AD-&@D.99+@&7.55mD&ADnD9+.@&+5Lm.7D&@75"97D@95n

5E6.7&A"&7"&^ 7% 79@65.M&% D9 D-& @D.9 9!D.7&% J.D- D-& "97D@95 >@&j6. m@!.D@@.5L

%3EAD&%D9{na



$.*..5)'5/$/

6>5&A EA&% +9@ 6."@9@@L 75LA&A J&@& "955&"D&% +D&@  - 9+  ! 9@ j~}{~|

D@&D6&7Dm+9@{z"&55An^9@+D&@|-.7,@9JD-9@%.++&@&7D.D.976&%.E6m+9@|{|

"&55Ana 7& Ž, 9+ D9D5  >E@.+.&% J.D-  M95 J&@& 5!&5&% J.D- 6!.97 @&,&7DA 7%

-L!@.%.M&%D9[email protected] &7&-.>9EA&m+9@|{|"&55An9@ E67m{z"&55AnK97

{az@@LAa[email protected]&K97@@L%DJ&@&79@65.M&%EA.7,?E7D.5&79@65.MD.97a-&

!"4,@9E7% "9@@&"D.97 JA >&@+9@6&% EA.7, 7D.,&796." >@9!&Aa 75L >@9!&A D@,&D.7,

&K97A 779DD&% +@96 +E55j5&7,D- " A&?E&7"&A J&@& EA&% +9@ 75LA.Aa 697, D-&A&

>@&A&5&"D&%>@9!&A^>99@?E5.DL>@9!&Am&a,a^>@9!&A5!&5&%Ae"@9AAj-L!@.%.M.7,f>@9!&A

!L [email protected]^ 7% >@9!&A J.D- .7D&7A.DL A.,75A D-D J&@& D99 59J J-&7 "96>@&% D9

7D.,&796."!"4,@9E7%>@9!&AJ.D-D-&A6& "97D&7D^J&@&@&69I&%+@96D-&75LA.Aa

 {{z 75L >@9!&A J.D-    /jI5E& Πzaz .7 D 5&AD -5+ 9+ D-& "-.>A J&@& "97A.%&@&% +9@

+E@D-&@ ADD.AD."5 75LA.Aa .@&% ADD.AD."5 75LA&A J&@& >&@+9@6&% EA.7, D-& DE%&7DrA

>.@&% 3jD&AD 97 D-& A>5.".7, .7%&K D9 75LM& D-& &K97 @@L %Da -& A>5.".7, .7%&K m n

"9@@&A>97%A D9  "96>@.A97 9+ ,&7&j79@65.M&% &K97 .7D&7A.DL I5E&A !&DJ&&7 D-& DJ9

75LM&%&K>[email protected]&7D5"97%.D.97Aa



$*$)!*.(0$/)/00$/0$')'5//

K97j.7D@97AD@E"DE@&A9+69EA&7%-E67,&7&AJ&@&9!D.7&%EA.7,D-&AD&@A&@I&@ m-DD>`ii+AD&@%!a5L97aE7."7"&@[email protected]%&Ka>->na 9 .%&7D.+L i{j%&>&7%&7Dj,&7&A

A-9J7 .7 .,E@& {^ D-& %.++&@&7D.D.97 52 E7%.++&@&7D.D&% +95% "-7,& mn JA "96>ED&%

&.D-&@.7D-&>@&A&7"&9@.7!A&7"&9+i{a.++&@&7D.D.97j@&,E5D&%,&7&AJ&@&

"97A.%&@&% A %&>&7%&7D 97 i{ J-&7 D-&  JA A.,7.+."7D5L @&%E"&%^ J.D- 

D-@&A-95%{a}7%J.D-/jI5E&Œzaza

9.%&7D.+Li{j%&>&7%&7Dj&K97AA-9J7.7.,E@&}^D-& m>5.".7, 7%&Kn9+&"-

&K97 JA "96>ED&% .7 !A&7"& 9+ %.++&@&7D.D.97a     {a J.D-  /jI5E& Œ zaz JA

"97A.%&@&%AA.,7.+."7Da9.%&7D.+L"96697-E677%69EA&i{j%&>&7%&7Dj

&K97A^9@D-959,A9+&K97AA.,7.+."7D5L@&,E5D&%.797&"&555.7&J&@&+.@AD.%&7D.+.&%.7D-&

9@,7.A6 "9@@&A>97%.7, D9 D-& 9D-&@ "&55 5.7& EA.7, D-& .+DI&@ D995 m-DD>`ii,&796&aE"A"a&%Ei",.j!.7i-,.+DI&@n^ 7% 5'! 5!12a -& I@.D.97 9+

.7"5EA.97i&K"5EA.979+D-9A&&K97AJAD-&7"96>ED&%.7D-&A&"97%"&55E5@69%&5AD-&†

%.++&@&7"& 9+ ,&7&j79@65.M&% &K97 .7D&7A.DL .7 D-& DJ9 "97%.D.97Aa  I@.D.97 9+

.7"5EA.97i&K"5EA.97{z†J.D-/jI5E&ŒzazJA"97A.%&@&%AA.,7.+."7Da

-&@,&DA"7A9+DJ@&m-DD>`iiJJJaD@,&DA"7a9@,inJAEA&%D9.%&7D.+L>@&%."D&%6.j

{i|z€ D@,&DA .7 69EA& }d Aa -& 7E6!&@ 9+ >@&%."D&% A.D&A J.D-.7 i{j

 {{{ %&>&7%&7D ,&7&A mE>j 9@ %9J7@&,E5D&% %[email protected], %.++&@&7D.D.97 9+ |{| "&55An JA

"96>@&%D9D-&7E6!&@9+>@&%."D&%A.D&A.755AI.5!5&.7D-&@,&DA"7%D!A&a

@&%."D&%A.D&AJ&@&"9E7D&%A{A.7,5&A.D&J-&7D-&.@7E6!&@JA{J.D-.7D-&A6&

D@7A"@.>Da7@."-6&7D75LA&AJ&@&>&@+9@6&%J.D-A9+DJ@&m-DD>`ii"@7a@j>@93&"Da9@,in

EA.7,!.796.5D&ADa





1,,'()0.5!$"1./

$"<Gai{@&}"9j@&,E5D9@A7%@&,E5D&D-&&K>@&AA.979+ {i|aAB

5.%D.97 !L J&AD&@7j!59DD.7, 9+ D-& &++&"D 9+ A.A ,.7AD NBJP m97& A.7,5& A.

D@,&D.7,"97A&@I&%@&,.97!&DJ&&7D-&|,&7&Ana7%KBLm6.KDE@&9+|A.A^

A&&&D-9%A+9@D-&A&?E&7"&9+A.Ana-&&K>@&AA.979+|i}i~>@9D&.7AJA79D

++&"D&%!LD-&A.NBJPD@&D6&7Da  7%I.7"E5.7J&@&EA&%A59%.7,"97D@95AaAB

&AD&@7j!59D75LA.A9+ {7% |m n&K>@&AA.97+D&@~‚-9+i{

%&>5&D.977%}}-9+ !D@&D6&7DaAB9.66E79>@&".>.DD.97AALAa "&55AJ&@&

D@7A+&"D&%J.D-5,j}9@5,j~>5A6.%A7%D@&D&%9@79D+9@|-J.D- !a

{z†9+.7>EDJ&@&EA&%A"97D@957%{zz†9+.66E79>@&".>.DD.97JA59%&%a9D-

7%{>@9D&.7AJ&@&%&D&"D&%+D&@5,m}9@~n.66E79>@&".>.DD.97aAB

-@96D.7 .66E79>@&".>.DD.97 A-9J.7, D-& !.7%.7, 9+  7% ~ D9  J 7%

 K>@969D&@@&,.97Aa 66E79>@&".>.DD.979+9@~>@9D&.7AJA>&@+9@6&%

+D&@|-9+ !D@&D6&7Da-&6>5."97A75LM&%!Lj?"9@@&A>97%D9@&,.979+

 J7% K>@969D&@A"97D.7.7,!.7%.7,&5&6&7Da&AE5DA"9@@&A>97%D9

6&79+}.7%&>&7%&7D&K>[email protected]&7DA7%@&>@&A&7DD-&>&@"&7D,&9+@&"9I&@&%+D&@

@&5D.I&D9.7>ED"-@96D.7m7‰}‹na

 {{| 

$"< Ha &,E5D.97 9+ &>.D-&5.5 3E7"D.97A !L i{a AB &AD&@7 !59D 75LA.A 9+ j

%[email protected] 7% ""5E%.7 &K>@&AA.97a A &K>&"D&%  ! D@&D6&7D .7%E"&%  %@6D."

%&"@&A& 9+ D-& &K>@&AA.97 9+ !9D- >@9D&.7A^ !ED >@&jD@&D6&7D J.D- NBJP A.

"96>@96.A&% D-.A %&"@&A&a AB {z .66E79AD.7.7, 9+ &>.D-&5.5 3E7"D.97A  EA.7,

""5E%.7 7% Xj"D&7.7 7D.!9%.&A +9559J.7, i{ %&>5&D.97 7%  ! D@&D6&7Da

E"5&.J&@&AD.7&%J.D- 9&A"-Da



$"< Ia %Ki%K{ +E7"D.97 A L9% D@7A"@.>D.975 "9@&,E5D9@Aa AB &D&"D.97 9+

&7%9,&79EA%KjL9%"96>5&K&A!L.7A.DE@9K.6.DL.,D.97AALAmna.K&%|{|

"&55A J&@& .7"E!D&% J.D- 7D.!9%.&A ,.7AD %K 7% L9%a 7D&@"D.97A J&@& @&I&5&%

EA.7, A&"97%@L 7D.!9%.&A "9E>5&% D9 A>&".+."   >@9!&A D-D -L!@.%.M&% 7% J&@&

&7ML6D."55L39.7&%J-&759"D&%.7"59A&>@9K.6.DLa+D&@@955.7,".@"5&6>5.+."D.97^&"-

.7D&@"D.97 ,&7&@D&%  +5E9@&A"&7D A>9D D-D JA 75LM&% !L "97+9"5 6."@9A"9>La A

7&,D.I&"97D@95A^&K>[email protected]&7DAJ&@&>&@+9@6&%.7|{|"&55AD@7A+&"D&%J.D- 7NBJP

9@8- A.AaAB&AD&@7!59D75LA.AA-9J.7,D-&&K>@&AA.979+%K^%K{7%L9%

+@96 |{| "&55A D@7A+&"D&% J.D- "97D@95  7NBJP9@8-  A.Aa "D.7 JA EA&% A 

59%.7,"97D@95aAB&AD&@7!59D75LA.A9+L9,7%"D.7|{|"&55A+D&@D@7A+&"D.97

J.D-"97D@95 7NBJP9@8- A.+9559J&%9@79D!LD-&.7%E"D.979+%.++&@&7D.D.97+9@

~‚-a



$"<Ja&,E5D.979+7%{&K>@&AA.97!L6.AaAB@&%."D&%6.!.7%.7,

A.D&A.7D-&N}da9>>7&5`A"@&&7">DE@&+@96D-&@,&D"7J&!A.D&A-9J.7,

D-&>@&%."D&%A.D&A+9@6.j{^6.j|z€7%6.j{‚{!.7D-&N}ra9D&D-DD-&A&

 {{} A.D&AJ&@&>@&%."D&%!L9D-&@+@&&5L""&AA.!5&A9+DJ@&AAE"-A6.7%9@."@m%D

79DA-9J7na&59J@&A-9J7%&D.5A!9EDD-&>@&%."D&%!A&j>[email protected],!&DJ&&7D-&%.++&@&7D

6.A 7% D-&.@ @&A>&"D.I& D@,&D A.D&Aa A D-& &++&"DA 9+ 6.j{i|z€ 7% 6.j{‚{! J&@&

ADE%.&%@&A>&"D.I&5L.769EA&|{|"&55A7%-E67{z"&55A^D-&.@!A&j>[email protected],J.D-

D-& N }d  JA A-9J7 J.D-.7 D-&.@ @&A>&"D.I& A>&".&A "97D&KDa -& .%&7D."5

7E"5&9D.%&A !&DJ&&7 -.,-5L @&5D&% 6.j{ 7% 6.j|z€ @& !9K&% .7 !5E&^ .7%."D.7, D-&

>&@+&"D"97A&@ID.97DD-&5&I&59+D-&d&7%j59"D&%A&&%@&,.97^J-."-.A"@.D."5+9@D-&

!A&j>[email protected], J.D- .DA 6 D@,&D m!9K&% .7 ,@&&7na J9 !.7%.7, A.D&A +9@ 6.j{‚{! J&@&

>@&%."D&% J.D-.7 D-& }d  9+ -E67 N D@7A"@.>DA^ !ED D-& A&"97% A.D& .A 5&AA

&I95ED.97@.5L"97A&@I&%7%JA79DD&AD&%.79E@&K>[email protected]&7DAa-&~j7E"5&9D.%&%&5&D.97A

D-D J&@& 6%& J.D-.7 D-& N  5E".+&@A& "97AD@E"DA D9 %&AD@9L D-& >@&%."D&% 6.j

{i|z€7%6.j{‚{!D@,&DA.D&A.75E".+&@A&AALA@&A-9J7DD-&!9DD969+&"-!9Ka

AB@&%."D&%6.!.7%.7,A.D&A.7D-&JP}da&D.5A@&A.7a9D&D-DD-&@&

6L!&A&"97%E7"-@"[email protected]&%6.j{i|z€!.7%.7,A.D&.7D-&JP^A6.j{7%

6.j|z€A5.,-D5L%&"@&A&%D-&5E".+&@A&"D.I.DL+@96D-&6ED7DJP@&>9@D&@m.,a|^

@.,-D >7&5na 9 !.7%.7, A.D& JA >@&%."D&% +9@ D-& 6.j{‚{ +6.5La AB E".+&@A& AAL

6&[email protected],D-&&++&"D9+6.j{7%6.j|z€97J.5%jDL>&7%6EDD&%69EA& 7JP}da

A.7.,E@&|aABK>@&AA.979+6DE@&6.j{‚{!JA6&AE@&%|~-+D&@D-&.7%E"D.97

!L !.7{z"&55A>@&jD@&D&%J.D-"97D@959@KBLA.aABE7D.+."D.979+

>@.j6.j{‚{!j|^D-&>@&"E@A9@&K>@&AA&%+@96D-&A&"97%6.j{‚{!j"9%.7,,&7&^|~-+D&@

D-&.7%E"D.97!L !.7{z"&55A>@&jD@&D&%J.D-"97D@959@NBJPA.aAB

7@."-6&7D9+6.j{i|z€!.7%.7,A.D&A.7,&7&A%9J7@&,E5D&%.7i{j%&>&7%&7D

677&@%[email protected],%.++&@&7D.D.979+|{|"&55Aa-&"97D@95.7"5E%&A55D-&}dAI.5!5&

+@96D-&@,&DA"7J&!A.D&a-&,&7&AE>@&,E5D&%.7i{j%&>&7%&7D677&@

 {{~ %[email protected],%.++&@&7D.D.97"97D.75&AA>@&%."D&%6.j{i|z€!.7%.7,A.D&A!EDD-&@&AE5DJA79D

ADD.AD."55LA.,7.+."7Da



$"<Ka>5.".7,@&,E5D.97!Li{aABj75LA.A9+5D&@7D.I&&K97Am.7@&%n

+D&@ 7NBJPA.5&7".7,.7|{|"&55Aa9@&"-&I&7D^D-&,&7&.AA-9J797D-&5&+DaABj

75LA.A9+5D&@7D.I&&K97Am.7!5E&n+D&@NBJPA.5&7".7,.7{z"&55Aa9@&"-

&I&7D^D-&,&7&.AA-9J797D-&5&+Da



$"< La .@&"D @&,E5D.97 9+ A>5.".7, !L i{a AB &A"E& &K>[email protected]&7DA 9+

i{j%&>&7%&7D A>5.".7, &I&7DAa A .7 .,E@& }a AB 97A&@ID.97 9+ A>5.".7,

@&,E5D.97 .7 I@.9EA "&55E5@ "97D&KDAa >>&@ >7&5 ` A"-&6D." @&>@&A&7DD.97 9+ D-&

6.7.,&7&AEA&%D9ADE%LD-&A>5.".7,9+i{j%&>&7%&7D&K97Aa-&J&K97{z

7% Q&K97~J&@&.7A&@D&%597,J.D-!9ED|zz!>9+D-&.@@&A>&"D.I&AE@@9E7%.7,

.7D@97." A&?E&7"&A .7 7 .7D@97 +@96 D-& +.!@97&"D.7 ,&7&a @.6&@A EA&% +9@ D-& 6.7.,&7&

AALJ&@&.7%."D&%!L@@9JAmA&?E&7"&A@&,.I&7.7!5&na9J&@>7&5`&7%9,&79EA

A>5.".7,@&,E5D.979+D-&A&DJ9&K97AJAD&AD&%!Lj.7|{|^{z^7%

& "&55A ~‚ - +D&@ %&>5&D.97 9+ i{a -& @.,-D  >7&5 A-9JA D-D +D&@

D@7A+&"D.97 9+ J 7%  Q 6.7.,&7&A .7 & "&55A^ D-&.@ @&A>&"D.I& &K97A J&@&

@&,E5D&% A D-& "9@@&A>97%.7, &7%9,&79EA &K97A !L i{a AB 

.66E79>@&".>.DD.97 &K>[email protected]&7D A-9J.7, D-& @&5D.I& !.7%.7, 9+ &7%9,&79EA  D9

i{j@&,E5D&%&K97A.7"&55Aa 66E79>@&".>.DD.97JA"@@.&%9ED+D&@"&55

"@9AAj5.74EA.7,77D.j7D.!9%L9@J.D-"97D@95 , 7%JA?E7D.+.&%!Lj?

EA.7, >@.6&@A D-D 6>5.+.&%  @&,.97 9I&@5>>.7, D-& A>5.".7, %979@ A.D& 9+ D-&

"9@@&A>97%.7, &K97a K97A .7%."D&% .7 !5E& "9@@&A>97% D9 i{j@&,E5D&% &K97A^

 {{ J-.5&&K97A.7!5"4"9@@&A>97%D9E7@&,E5D&%&K97A+@96D-&A6&,&7&a-&@&AE5DAm7‰}

‹n@&&K>@&AA&%A>&@"&7D,&9+.7>EDajI5E&AJ&@&"5"E5D&%!&DJ&&7D-&

 7%D-& ,  m>.@&%3jD&AD`oŠzaz_ooŠzaz{_`797A>&".+."^I5E&A

.7%."D&%na



$"<Ma5D&@7D.I&A>5.".7,@&,E5D.97!Li{%[email protected],%.++&@&7".D.97aAB>5.".7,9+

&K97A@&,E5D&%.7D-&A6&677&@%[email protected],%.++&@&7D.D.979+|{|"&55A7%E>97 7NBJP

A.5&7".7,aAB>5.".7,9+&K97A@&,E5D&%.7D-&A6&677&@%[email protected], !j.7%E"&%9+

{z"&55A7%E>97NBJPA.5&7".7,aAB>5.".7,9+&K97A@&,E5D&%%[email protected],|{|"&55

%.++&@&7D.D.97 !ED 79D E>97  7NBJP A.5&7".7,a AB >5.".7, 9+ &K97A @&,E5D&% .7  !j

D@&D&% "&55A !ED 79D E>97 NBJP A.5&7".7,a AB >5.".7, 9+ &K97A @&,E5D&% .7 D-& A6&

677&@ %[email protected], |{| "&55 %.++&@&7D.D.97^ E>97  7NBJP A.5&7".7, 7% E>97 &"D9>."

9I&@&K>@&AA.97 9+ 6.j{ 9@ 6.j|z€a AB >5.".7, 9+ &K97A @&,E5D&% .7 D-& A6& 677&@

%[email protected],  !j.7%E"&%  9+ {z "&55A^ E>97 NBJP A.5&7".7, 7% E>97 &"D9>."

9I&@&K>@&AA.979+6.j{‚{!a&D.5A@&A.7.,E@&A}&j}+a



$"<Na"-&6&9+@&,E5D.979+"LD9A4&5&D979@,7.MD.97!L7%{a-&A"-&6&

@&>@&A&7DAD-&  >D-JL6>&7D.D5&%e&,E5D.979+"D.7"LD9A4&5&D97f^%9J759%&%

+@96 D-& L9D9 7"L"59>&%. 9+ &7&A 7% &796&A J&!A.D& m-DD>`iiJJJa,&796&a3>i4&,,ina &7&A D-D J&@& @&,E5D&% D D-& A>5.".7, 5&I&5 !L

i{%[email protected],%.++&@&7D.D.979+|{|"&55Am@&%!9K&An^%[email protected], !j.7%E"&%9+

{z"&55Am!5E&!9K&An9@.7!9D->@9"&AA&Am,@&&7!9K&AnJ&@&%%&%D9D-&A"-&6&a

@@9JA .7%."D& %.@&"D .7D&@"D.97A 9@ +E7"D.975 5.74A !&DJ&&7 i{j@&,E5D&%

 {{€ ,&7&A7%&5&6&7DA9+D-&>D-JLa&D.5A!9EDD-&+E7"D.9755.74A!&DJ&&7D-&,&7&A

7%"LD9A4&5&D97@&69%&55.7,7%i9@56&55.>9%.+9@6D.97@&,.I&7.7D-&D!5&!&59Ja



'Ga &7&A++&"D&%DD-&,59!55&I&5!L7%{%&>5&D.97.7{z"&55A D@&D&%9@79DJ.D- !a ' H< &7&A ++&"D&% D D-& ,59!5 5&I&5 !L %K 7% %K{ %&>5&D.97 .7 |{| "&55A D@&D&%9@79DJ.D-%.++&@&7D.D.976&%.E6a 'IaAA&DD&&K97A++&"D&%!L%K7%%K{%&>5&D.97.7|{|"&55Aa 'JaAA&DD&&K97A++&"D&%!L7%{%&>5&D.97.7{z"&55Aa 'Ka.AD9+A.A7%>@.6&@AEA&%.7D-.AADE%La  *2$ Ga .6&j5>A& I.%&96."@9A"9>L 9+ {z "&55A D@&D&% J.D- "97D@95 A. .7 D-& !A&7"&9+ !a *2$H`.6&j5>A&I.%&96."@9A"9>L9+{z"&55AD@&D&%J.D-"97D@95A.7%J.D-  !a *2$I`.6&j5>A&I.%&96."@9A"9>L9+{z"&55AD@&D&%J.D-NBJPA..7D-& !A&7"&9+ !a *2$J`.6&j5>A&I.%&96."@9A"9>L9+|{|"&55AD@&D&%J.D-"97D@95A.a *2$K`.6&j5>A&I.%&96."@9A"9>L9+{z"&55AD@&D&%J.D-%Ki{A.a   $0$*)'10#*.)*0/

10#*.*)0.$10$*)

aa7%aa>&@+9@6&%&K>[email protected]&7DA.7{z"&55AJ.D-.7>ED+@96aaaaa>&@+9@6&%

55!.9.7+9@6D."75LA&A7%A96&&K>[email protected]&7DA.7|{|"&55Aaaa7%aaa>&@+9@6&%

A96&9+j7%j?75LA&A.79D-&@"&555.7&Aaa a,&7&@D&%D-&6."@9@@L%D

7%>&@+9@6&%A96&9+|{|&K>[email protected]&7DAaaa>&@+9@6&%D-&.66E79>@&".>.DD.97

&K>[email protected]&7DA 9+ .,a }% 7% "597&% D-& 6.7.,&7&A 9+ E>>5&6&7D@L .,a €!a aaa

>&@+9@6&%D-&.66E79>@&".>.DD.977%j?9+E>>5&6&7D@L.,a€"aa a7%  {{ aa 6%& D-& .7%E".!5& "&55 5.7&A 7% >&@+9@6&% A>5.".7, @&A"E& &K>[email protected]&7DAa aaa

>&@+9@6&%D-&AALaaa>&@+9@6&%5E".+&@A&AALA^6.?E7D.+."D.977%A96&

j75LA&A^J.D-.7>ED+@96aa7%aaaaa^aaa7%aa%&A.,7&%D-&ADE%L

7% .7D&@>@&D&% D-& @&AE5DA J.D- .7>ED +@96 55 D-& ED-9@Aa aaa 7% aa J@9D& D-&

67EA"@.>D^J.D-.7>ED+@9655D-&ED-9@Aa



DJ.55!&%&>9A.D&%.7D-& &7&K>@&AA.9767.!EAE>97""&>D7"&9+D-&67EA"@.>D^

!ED@&I.5!5&97@&?E&ADa



-&ED-9@A%&"5@&79"96>&D.7,+.77".5.7D&@&ADAa







 

 {{‚ Figure 1

A MCF10a E C2C12 SNAI1 SNAI2 Myog Mef2c  *  *  * ** ** *   **          

Relative expression Relative   expression Relative   TGFβ - +++ - +++ Diff - +++ - +++ si ctrl ++- - ++- - si ctrl ++ - - ++- - siSMAD2/3 --+ - --+ - siMyod --+ - --+ - siDDX5/17 ---+ ---+ siDdx5/17 ---+ ---+

B MCF10aF C2C12 804 regulated genes 1033 regulated genes

137 59 368 240 333 263 143 294

196 activated 608 repressed 596 activated 437 repressed

DDX5/17-dependent genes Ddx5/17-dependent genes

C si ctl si ctl +TGFβ siDDX5/17 +TGFb G - Diff si ctl Diff si ctl Diff siDdx5/17 * *

*       *  

n *** n *

** *                  

Relative expressio     Relative expressio     FAP ITGB6 WIPF1 OCLN Myh2 Ccdc25 Pstpip1 Aqp1

Negative controls Negative controls NS NS    NS  NS  

   NS     NS NS    NS        Relative expression Relative     Relative expression     RASSF8 PLD5 PKP2 TUBB Mafk Tec Wdhd1 Tpx2

D -TGFβ + TGFβ H -Diff Differentiation si ctrl si ctrl siDDX5/17 si ctrl si ctrl siDdx5/17 MHC E-cadherin

  {{ƒ Figure 2

ADMCF10a ctrl miR-181b C2C12 MCF10a  pre-miR  Diff TGFβ 012345 023  ** (days) (days) ctrl 181b  DDX5 DDX5 

DDX17 DDX17 activity Luciferase   Actin Actin wt mut DDX5 UTR

pri- B C2C12 ctrl miR-1 miR-206 E miR-181b-1  miR-181b  *  * pre-miR    

ctrl 1 206 **  ***   Relative Ddx5 Relative  expression expression

 expression    Ddx17 activity Luciferase  TGFβ - ++ TGFβ - ++ wt mut si ctrl ++- si ctrl ++- Actin Ddx5 UTR siDDX5/17 --+ siDDX5/17 --+

C miR-1 pri-miR-1a1 F * ChIP MCF10a    

on on Ctrl TGF  β   miR-181b1

expressi miR-206expressi pri-miR-206 +1 * *  % input    Relative  Relative     IgG SMAD4 DDX5 Diff - ++ - ++ si ctrl ++ - ++ - siDdx5/17 --+ --+









  {|z Figure 3

A C C2C12 MCF10a DDX5 DDX5 KA DDX17 DDX17 KR Dox --+ --+ --+ --+ siDDX5/17 - ++- ++ - ++- ++ E4 included 142 46 773 CHTF8

43 2 36 48 1 3 30 3 62 23 1 2 E5 SEC24B

47 16 43 50 11 16 55 40 53 54 19 20 skipped 217 69 1069 E6 TCERG1

13 0 5 12 1 2 28 8 9 19 2 1 E11 B CTTN C2C12 MCF10a 87 72 85 89 70 80 91 71 91 91 61 83 lanes 1 2 3 4 5 6 7 8 9 10 11 12 siDDX5/17 - + - + siDDX5/17 E20 E24 TJP1 D inducible MCF7 cell line E10 E10 REPS1 -DOX + DOX

E4 E4 CHTF8    

E3 E5    SEC24B    

E32 % input E32 RIF1     E18 E13 CDC42BPA     E6 E6 E3 E4 E10 E5 E16 E6 E15 E11 TCERG1 CHTF8 SEC24B TCERG1 CTTN E11 E11 CTTN

EFC2C12 MCF10a pre-miR pre-miR Diff - + - TGFβ - + - siDDX5/17 --+ ctrl 1 206 siDDX5/17 --+ ctrl 181b

Bnip2 ARFGAP2

1 1.4 1.7 1 1.9 2.0 1 0.6 0.6 1 0.7

Gripap1 ITSN1

1 2.1 3.6 1 3.9 4.9 12.44.5 14.9

Atp5c1 INSR

10.50.7 10.50.6 1 0.8 0.8 1 0.6

Tbc1d1 CA12

12.32.2 15.77.3 1 0.5 0.5 1 0.6

R3hdm1 PLEKHA1

11.20.9 Negative 11.31.0 Negative controls controls Anxa7 ATP11C

12.20.9 10.91.0

 {|{  Figure 3G

MCF10a si ctrl

si ctrl + TGFβ

siDDX5/17

C2C12

si ctrl

siDDX5/17

 {||  Figure 4

Diff medium DDX5 DDX17 Myod Transcription Splicing

miR-1/206 Bnip2, Gripap1, Cttn, Myotubes Myoblasts Myog, Mef2c, Myh2… Tbc1d1, Tjp1… Gene expression Gene expression Lamellipodia program A program B Epithelial cell SNAIL1/2, CDH1, OCLN... ENAH, ARFGAP2, INSR, Mesenchymal FNIP1, CTTN... Loss of epithelial junctions miR-181b cell

Transcription Splicing SMADs DDX5 TGFβ DDX17







 {|} Figure S1 MCF10a cells

AB siDDX5/17 - + TGFβ - + - + DDX5 siDDX5/17 - - + + siSMAD2/3 - + DDX17 SNAIL1 SMAD2 GAPDH SNAIL2/SLUG SMAD3 SMAD2 GAPDH Vinculin SMAD3

SMAD4

C Input IP Flag Input IP Flag

Flag-SMAD3 - ++- ++ Flag-SMAD4 - ++- ++ TGFβ --+ --+ TGFβ --+ --+

DDX5 DDX5

DDX17 DDX17

Flag- Flag- SMAD3 SMAD4

D SNAI1 SNAI2  

   

% input  % input  

  + - + - + TGFβ + - + - + TGFβ IgG DDX5 SMAD4 IgG DDX5 SMAD4

 {|~  Figure S2 MCF10a cells

A TGFβ - ++ siDDX5/17 --+ E-Cadherin

Occludin

GAPDH

B si ctrl si ctrl + TGFβ siDDX5/17 + TGFβ

Occludin

b-catenin





  {| Figure S3 C2C12 cells

A Proximity Ligation Assay Ddx5-Myod

si ctrl siMyod siDdx5/17

B C Diff - +++ siDdx5/ 17 - + siMdMyod -- + - siMyod - + siDdx5/17 -- -+ Ddx5 Myod Myog Ddx17 Actin Actin Myod

Actin

 {|€  Figure S4 A TargetScan Predictions of miRNA binding in the DDX5 3' UTR http://www.targetscan.org

seed mmu-miR-206 3' GGUGUGUGAAGGAAUGUAAGGU 5' mmu-miR-1 3' UAUGUAUGAAGAAAUGUAAGGU 5' Human 5' AAGUUGGAUAUUUCUCUACAUUCCUGAA 3' Rhesus 5' AAGUUGGAUAUUUCUCUACAUUCCUGAA 3' miR-1/206 Mouse 5' AAGUUGGAUAUUUCUCUACAUUCCUGAA 3' site Chicken 5' AAGUUGaAUAUUUCUCUACAUUCCUGAA 3' Xenopus 5' AAGUUGacUgU--aUCUACAUUCCUGAg 3' Zebrafish 5' cAccUuuuUgUUUaguUguAUUuuUuuA 3'

mmu-miR-1 3' UAUGUAUGAAGAAAUGUAAGGU 5' mutant UTR 5' AAGUUGGAUAUUUCUCUAC----CUGAA 3'

seed hsa-miR-181b 3' UGGGUGGCUGUCGUUACUUACAA 5'

miR-181b Human 5' GUCACA-CUGGGAGUCAUGAAUGUCUU 3' site 1 Rhesus 5' GUCACA-CUGGGAGUCAUGAAUGUCUU 3' Mouse 5' GcCACAcCUGGaAGUCAUGAAUGUCUU 3'

hsa-miR-181b 3' UGGGUGGCUGUCGUUACUUACAA 5' mutant UTR 5' GUCACA-CUGGGAGUCAUG----UCUU 3' seed miR-181b hsa-miR-181b 3' UGGGUGGCUGUCGUUACUUACAA 5' site 2 Human 5' UAGCUAGAAGUGGGUCUGAAUGUUUU 3' Rhesus 5' UAGCUAGAAGUGGGUCUGAAUGUUUU 3' Mouse 5' UgGCUAuAAGUaGGUCUaAAUGUcUU 3'

 {|  Figure S4

B TargetScan Predictions of miRNA binding in the DDX17 3' UTR http://www.targetscan.org

seed mmu-miR-206 3' GGUGUGUGAAGGAAUGUAAGGU 5' mmu-miR-1 3' UAUGUAUGAAGAAAUGUAAGGU 5' Human 5' CUUGCUGGUCUCCAGACACAUUCCUGUU 3' miR-1/206 Rhesus 5' CUUGCUGGUCUCCAaACA--UUCCUGUU 3' site Mouse 5' CUUGCUGGUCUCuccACACAUUCCUGUc 3' Chicken 5' uUUGCUGGUCUCugaACACAUUCCUGUU 3'

mmu-miR-1 3' UAUGUAUGAAGAAAUGUAAGGU 5' mutant UTR 5' CUUGCUGGUCUCCAGACAC----CUGUU 3'

  {|‚ Figure S4

C HeLa

  * **  ctrl  miR-1  miR-206  Luciferase activity  wt mut Ddx17 UTR

DE MCF10a MCF10a

mature pri-  miR-181b  miR-181b-2

  lative lative

ression  ression  e e R R exp  exp 

  TGFβ -++ TGFβ -++ si ctrl ++- si ctrl ++- siSMAD2/3 --+ siDDX5/17 --+

F

p=0.039 

p=0.932  ctrl (n=16959) up (n=263)  down (n=143)

% UTR with % UTR with miR- 1/206 binding site binding 1/206



 {|ƒ  Figure S5

A C2C12

siDdx5/17 - + siDdx5/17 - + siDdx5/17 - + Dcun1d2 Sorbs1 Ablim1

Meaf6 Stradb Clta

Dnm1 Dguok Msi2

Sla Rock2 H13

Tnnt3 Fndc1 Map4k3

Cul7 Svep1 Itgb1

Oxr1

B MCF10a

siDDX5/17 - + siDDX5/17 - + siDDX5/17 - + PLEKHM2 ABI1 MICAL3

APP FGFR1OP MAGI1

ASXL1 GIT2 SNX14

MYO18A GOLGA2 MLPH

LOXL2 APH1B NEK1

PBRM1 ARFGAP3 CLSTN1

KITLG

 {}z  Figure S6 A DDX5 DDX5-KA DDX17 DDX17-KR Dox --+ --+ --+ --+ inducible MCF7 siDDX5/17 - ++- ++ - ++- ++ cell lines REPS1 90 61 81 95 66 63 59 37 61 63 50 49

RIF1 80 55 69 77 51 45 70 29 70 74 35 40

CDC42BPA

77 43 66 80 37 40 67 32 80 66 23 51 B

pTB plasmid (6003 bp)

SV40 α-globin fibronectinNdeI fibronectin α-globin

E2 E3 E3

NdeI NdeI intron 9 intron 10 REPS1 Exon 10 intron 3 intron 4 CHTF8 Exon 4

HeLa C2C12 MCF10a MCF7 endogene minigene

CHTF8

REPS1 siDDX5/17 - + - + - + - + - +

C

  ** * RNA-IP MCF7   =0.063) endogenous DDX5 p   NS ( IgG   DDX5 

ratio IP/input IP/input (%) ratio *  

  E3 E4 E15 E11 CHTF8 CTTN  {}{     

          -+-  -+-  -+-   --+   --+      --+ %& * * #(     &&"  (  ( (     !# $+ $+ %)     #'  #(  #(

      "( )+ )+     !  !



         β -+- β -+- β -+-   --+   --+     --+   )  )     

(-       

 (  ,  , * .    

 ) 

 (* 

,( 

*   {}|  Figure S8 INSR TBC1D1

REPS1 ARID4B

RALGPS2 BNIP2

GRIPAP1 ENAH

Tight junctions TJP1 MPDZ ADD3 GMIP

ARFGAP2 TJP1

CAST CTTN

STX16 CSNK1G3

   

"&#3-($' !($#3''$(%&$(##*$!*#(''"!-$('%(&#3(##(+$& 0&)!&$! # )#9 #!!"$(!(-

" '(*(#%&$(#$&7 7$#(&$!'7'#!#($&)!(!! "&($#

 $ ))#($(#9(&#'&%($#!$&%&''$& #9')'(*&$!#"('((%&$&''$#$&'(#&/ ( #(&('+(#!8($%&*#(

 "  82#$*&)'77< #(&(#%&$(#8 #*$!*#!!!$#($#0#(&('+(:8/' #'

 !%'((#/!%##($& $!# "-$!'()'$# 0!%#'&#*$!*#!!'%&#/!"!!%$$&"($47'#!#5

 !# '# #'7/""9 '# #'7%&$(#'%$'%$&-!(#β3(##

$#(&)('($($&#.($#$((#-($' !($##!!'(&)()&0%!-'&$!#(&)!($#$!!  $&((# "&($## #(#*'*#''$#&!!'

(#3''$(%&$(##*$!*#'*&!%&$'''%##($#-($' !($#&"$!##!!%$!&(-  #! $"$!$4&$'$%!5 ')'!"!!%$!#!$%$!-#"'#"&(#!!'

  3#(&(#%&$(# $3'(*(#"3#(&(#%&$(#

 ! '3')##,#($&4'5 $)!(''3'(*(-

 #')!#&%($& %($&(-&$'# #'(((*(' '*&!'#!#%(+-'

 )!(%!$"#%&$(# $!.#(( )#($#'0#(&('+($3

 " !+( $"## 9##"$(8 $)!('!')#($#1!'#($&$''#!##"($&$&""&#(& #

! !73''$(%'$"#3$#(##7 -$$&#((($#'$(*(&%($&'#!3''

!% -#(,#7; "&#(& #

!! '3(*(#%&$(#435 #*$!*#((& ##(&#'!$($#$ :3$#(##*'!'##')!#3'(")!(!)$')%( 

)!($#$!!"&($#-(&(#:8($(!#$"&(#!!'0!$!.#  ! ( )#($#&$(#74375 !"!!%$#$&#!&$!'(+$)#"$!  {}}  $/1//$*)





"!  , *-   '    &A-(5."A&A&D{A97D%&A>@9D(.7&A6E5D.+97"D.977&55&A?E.^"966& %dED@&A-(5."A&A^A97D.6>5.?E(&A%7A%&796!@&EA&A(D>&A%&56DE@D.97%&A Aa &A %&EK  -(5."A&A A97D AE@j&K>@.6(&A %7A %.++(@&7DA "7"&@A 79D66&7D 5& "7"&@%EA&.7&D%E"9597mE55&@j"&s5.^|zz‚_ 747&"-D^|z{z_-.7&D5a^|zzna7A 5& "7"&@ %E "9597^ 5d&K>@&AA.97 %&  &D { E,6&7D& +9@D&6&7D E "9E@A %& 5 >@9,@&AA.97DE69@5&%EAD%&%&>95L>&EAD%&%d%(796&&D%d%(79"@".796&m-.7&D 5a^|zzna&>5EA^%7AE769%'5&[email protected]%E"7"&@%EA&.7^3d.%("@.D5AE@j&K>@&AA.97%& {%7A5&A"&55E5&A~{^%&A"&55E5&A6(DADD.?E&A%E"7"&@%EA&.7^>@@>>9@DEK "&55E5&A €^ E7& 5.,7(& AL7,(7.?E& 797 6(DADD.?E&a  +97"D.97 %& "&A %&EK  -(5."A&A&7D7D?E&"9j@(,E5D&E@AD@7A"@.>D.977&5A(D(!.&7%("@.D&^79D66&7D>9E@5& @("&>D&E@EK9&AD@9,'7&A%7A5&"7"&@%EA&.7mD7!&&D5a^|zz{n&D%&5Xj"D(7.7& %7A5&"7"&@%E"9597m-.7&D5a^|zzna7A5&"7"&@%EA&.7^79EAI97AEAA.697D@( ?E&  &D { A97D %&A "9j@(,E5D&E@A D@7A"@.>D.977&5A %E +"D&E@ >@9j6.,@D9.@& m &@677&D5a^|z{|nmI9.@77&K&na   7A"&"97D&KD&^&D{"9j"D.I&7D5&A,'7&A".!5&A%&.6>5.?E(A %7A56.,@D.97&D5d.7IA.97"966&{zz~9E|m &@677&D5a^|z{|naE"9E@A%& 6D-'A&^.5(,5&6&7D(D(>E!5.(?E&@(,E5&5d&K>@&AA.97%&6."@9Am6.j|{&D 6.j{‚|n?E."97D@;5&7D5&E@D9E@5d&K>@&AA.97%&,'7&A.6>5.?E(A%7A5&@&69%&5,&%E "LD9A?E&5&DD&%d"D.7&^"966&5"9+.5.7&&D5>@9+.5.7&^"&?E.>9E@@.DI9.@E7.6>"DAE@ 5&>-(79DL>&6.,@D9.@&&D.7IA.+%&A"&55E5&A%E"7"&@%EA&.7ma7,&D5a^|z{|na&A A-(5."A&A&D{97D(D(@>>9@D(&A"966&"97D@;57D5d(>.AA,&5D&@7D.+%& ?E&5?E&A,'7&A"966&E^~~9E j@Am E.5&D5a^|zz}_ 97.7,&D5a^|zz|_ @&D5a^ |z{{n 6.A I7D 6 D-'A&^ 5d.6>"D^  5@,& ("-&55&^ %&  &D { AE@ 5d(>.AA,& 5D&@7D.+7dI.D>A(D((DE%.(a&>5EA^5&@;5&+97"D.977&5%&AI@.7DA%d(>.AA,&"97D@;5(A  {}~ >@&D{7d(D.D>A"977Ead(DE%&6&7(&%7A5&69%'5&~{697D@&?E&"&A A-(5."A&A"97D@;5&7DE73&E%(+.7.%d&K97A5D&@7D.+Aa@6.5&A&K97A@(,E5(A>@%K &D %K{^ 97 D@9EI& 5&A &K97A 6EDE&55&6&7D &K"5EA.+A € &D  %& 5d-.AD97& 6"@9 |{ ?E. "9%&7D@&A>&"D.I&6&7D>9E@5+9@6&6"@9 |{a|&D6@9 |{a{a&A%&EK.A9+9@6&A97D %&A&++&DA9>>9A(AAE@5&>-(79DL>&6.,@D9.@&&D.7IA.+%&A"&55E5&A~{a d.%(697D@(?E& "&A I@.7DA 97D E7 @;5& 9>>9A( &D "97D@;5&7D >9A.D.I&6&7D 9E 7(,D.I&6&7D %&A ,'7&A "966&}^?E.A97D.6>5.?E(A%7A56.,@D.97&D5d.7IA.97I.5&"97D@;5&%&A&A>'"&A @("D.I&A%&5d9KL,'7&a   & D@I.5 697D@& ?E& 5 @(,E5D.97 %& 5d(>.AA,& 5D&@7D.+ >@  &D { "97D@.!E&  5 >@9,@&AA.97 DE69@5&a & 6("7.A6& 695("E5.@& Ad39ED&  5 @(,E5D.97 D@7A"@.>D.977&55& &D  5 6DE@D.97 %& 6."@9A "966& (D7D "97D@;5( >@  &D {&D39E7DE7@;5&E"9E@A%&5>@9,@&AA.97DE69@5&a&5@&7+9@"&(,5&6&7D5& @;5& 63&E@ %& "&A A -(5."A&A %7A 5 >@9,@&AA.97 DE69@5& "@ &55& ++&"D& %&A >-(79DL>&A "&55E5.@&A %& 5 D@7A+9@6D.97^ "966& 5 >@95.+(@D.97 m79D66&7D I. 5&E@ +97"D.97 %& "9j"D.ID&E@A %& @("&>D&E@A 7E"5(.@&An 9E 5 6.,@D.97 m79D66&7D I. 5&E@ "97D@;5&%&5d(>.AA,&%&6"@9 |{9E%&"9j"D.ID&E@%&na 5&AD.7D(@&AA7D%& 79D&@?E&&D{@(,E5&7D5d&K>@&AA.97%&A,'7&A&7+97"D.97%E"97D&KD&"&55E5.@&a 7 &++&D^  5&A ,'7&A %(>&7%7DA %& "&A A -(5."A&A %7A 5& "97D&KD& %& 5 @(,E5D.97 9&AD@9,(7.?E&7&A97D>AA.6.5.@&AEK,'7&A?E.%(>&7%&7D%&&D{%7A5& "97D&KD& %& "9j@(,E5D.97 >@  9E  "&EK @(,E5(A >@ 5&A I@.7DA %& 5d-.AD97& 6"@9 |{a&DD&9!A&@ID.97Ad&K>5.?E&&7>@D.&>@5&+.D?E&5&A>@9D(.7&A"9j@(,E5(&A >@&D{A97D&K>@.6(&A&7+97"D.97%E"97D&KD&"&55E5.@&a   &A %.++(@&7D&A %977(&A %(697D@&7D ?E& "&A %&EK -(5."A&A %r @(,E5&7D 5d&K>@&AA.97,(7.?E&%.++(@&7DA7.I&EK_,@"&5&E@A+97"D.97A%7A5"9@(,E5D.97%&5 D@7A"@.>D.97&D%&5r(>.AA,&.7A.?E&56DE@D.97%&A6.^&D{"97D@;5&7D >5EA.&E@A>@9,@66&A"&55E5.@&A"97D@.!E7D.7A.5>@9,@&AA.97DE69@5&a  &A D@IEK %& D-'A& >9@D7D AE@ 5d(DE%& %& 5d.6>"D %& %K &D %K{ AE@ 5 @(,E5D.97 %& 5d(>.AA,& 5D&@7D.+ %7A %&A 69%'5&A DE69@EK [email protected] @&7+9@"&7D 5 "96>@(-&7A.97%E@;5&63&E@%&"&A-(5."A&AA%7A5>@9,@&AA.97DE69@5&&D"&5 +.D%&"&A>@9D(.7&A%&A".!5&AD-(@>&ED.?E&A>9D&7D.&55&Aa7&++&D^5&E@A+97"D.97A%7A5

 {} "9j@(,E5D.97D@7A"@.>D.977&55&@&>9A&7DAE@%&A.7D&@"D.97A>-LA.?E&AI&"%.I&@A+"D&E@A %&D@7A"@.>D.97>9E@5&A?E&5A5d"D.I.D(-(5."A&%(>&7%7D&%&&D{7d&AD >A @&?E.A& m@&DD. &D 5a^ |zz€_ D7!& &D 5a^ |zz{na  5d.7I&@A&^ 5& @;5& %&  &D {%7A5@(,E5D.97%&5d(>.AA,&&AD5@,&6&7D%(>&7%7D&%&5&E@"D.I.D(-(5."A& m.7&D5a^|zzna&%(D.5&AD.6>9@D7D>E.A?E&5&%96.7&%&5..A975d&ADE7&>9"-& -L%@9>-9!&^ ?E. "97AD.DE& E7& ".!5& >-@6"&ED.?E& A9EI&7D >5EA &++.""& ?E& 5 69%E5D.97 %&A .7D&@"D.97A %E DL>& >@9D(.7&j>@9D(.7&a 7A "& "97D&KD&^ E7 "@.!5,& %d>D6'@&A7E"5(.?E&A%7A%&ADE6&E@A%E"9597"-&M5[email protected]>[email protected]%d.%&7D.+.&@E7& 695("E5& ".!57D A>(".+.?E&6&7D  &D %.6.7E7D A97 "D.I.D( A& m. &D 5a^ |z{zna &DD& %("9EI&@D&  >[email protected] 5d9EI&@DE@& %& 79EI&55&A >&@A>&"D.I&A_ .5 A&@.D 6.7D&77D .7D(@&AA7D%d(DE%.&@5d(>.AA,&%&A,'7&A".!5&A%&&D{%7A%&A"&55E5&A9E%&A DE6&E@AgD@7A+&"D(&AhI&""&D>D6'@&^>E.A%d(DE%.&@5&>-(79DL>&&D&7>@D."E5.&@5 6.,@D.97^5d.7IA.979E5+9@6D.97%&6(DADA&Aa  Ej%&5 %& 5d.7D(@)D >@9>@& E ".!5,& >-@6"&ED.?E& %&  &D { "966& E7&>9D&7D.&55&D-(@>.&7D.jDE69@5&^5d&7A&6!5&%&"&A%977(&A6&D&7I7D5&@;5&%&5 %(@(,E5D.97%&5d(>.AA,&5D&@7D.+&7D7D?E&>@9,@66&^.7A.?E&5&A!&@@D.97A%&A +D&E@A@(,E57D5d(>.AA,&5D&@7D.+E"9E@A%&5>@9,@&AA.97DE69@5&a7A"&"97D&KD&^ .5&AD.7D(@&AA7D%&6&7D.977&@?E&5&+"D&E@%d(>.AA,&}!>>@0D+@(?E&66&7D6ED( "-&M %&A >D.&7DA DD&.7D %& 5&E"(6.& "-@97.?E& 5L6>-9"LD.?E&^ ?E. >@(A&7D&7D %&A 5D(@D.97A%d(>.AA,&m &,&5&&D5a^|z{|_E&A%&D5a^|z{|na.7A.5&".!5,&%&+"D&E@A %d(>.AA,&>>@0D"966&E7&>>@9"-&D-(@>&ED.?E&7D.j"7"(@&EA&9@.,.75&a   &A@;5&A%&&D{%7A5D@7A"@.>D.97^56DE@D.97%&A6."@9A&D 5d(>.AA,&5D&@7D.+A97D.6>9@D7DA>9E@5>@9,@&AA.97DE69@5&a&A%.++(@&7D&A(D>&A%& 5d&K>@&AA.97 %&A ,'7&A A97D >@.69@%.5&A >9E@ 5 >5AD.".D( "&55E5.@&a -"E7 %& "&A >@9,@66&A"&55E5.@&A+.D5d9!3&D%&796!@&EA&A(DE%&A?E.97D%("@.D5&E@@;5&5+9.A %7A %&A ALAD'6&A %& %.++(@&7".D.97 >-LA.959,.?E& &D %7A %&A A.DED.97A %& >5AD.".D( "&55E5.@& 5.(&A  5 >@9,@&AA.97 DE69@5&a 7 @&I7"-&^ 5 "99@%.7D.97 %& "&A %.++(@&7DA >@9"&AAEA%7AE76)6&ALAD'6&"&55E5.@&&ADD@'A>&E"977E&a E "9E@A %& 6 D-'A&^ 3& 6& AE.A +9"5.A( AE@ 5d(DE%& %& 5 D@7A.D.97 (>.D-(5.9j 6(A&7"-L6D&EA&mnE7>@9"&AAEA>-LA.959,.?E&?E.Ad9!A&@I&59@A%E%(I&59>>&6&7D &6!@L977.@&&D?E.79D66&7DA&@("D.I&E"9E@A%&5>@9,@&AA.97DE69@5&a d..7A.

 {}€ >E %(697D@&@ ?E& 5&A A -(5."A&A  &D { A97D 7("&AA.@&A  5d.7%E"D.97 %& 5 %.++(@&7".D.97 &7 (D7D "9j"D.ID&E@A %&A +"D&E@A %& D@7A"@.>D.97 Aa & 67.'@& AE@>@&77D&^79EAI97A9!A&@I(E7&%.6.7ED.97%&5d&K>@&AA.97%&&D{E"9E@A %&5&D79EAI97A>@9>9A(E769%'5&>9E@&K>5.?E&@"&DD&%.6.7ED.97>@5d.7%E"D.97 %dE7 6."@9 A>(".+.?E& %& 5 D@7A.D.97 ?E. ".!5& "&A >@9D(.7&A^ &D %97D 5d&K>@&AA.97 %(>&7%%&&D{ad&K>@&AA.97%&"&A>@9D(.7&A&AD%97"%(>&7%7D&%dE7&!9E"5& %&@(D@9j"97D@;5&a%.6.7ED.97%&&D{&7D@.7&5d&K>@&AA.97%dE7>@9,@66& %d(>.AA,&A>(".+.?E&>&@6&DD7D%&"96>5(D&@5D@7A.D.97(>.D-(5.9j6(A&7"-L6D&EA&a7 &++&D^5&A".!5&A%d(>.AA,&%&&D{?E.A97D(,5&6&7D@(,E5(&AE"9E@A%&5 %.++(@&7D.D.97A97D%&A,'7&A79D66&7D.6>5.?E(A%7A5&@&69%&5,&%E"LD9A?E&5&DD& %d"D.7&&D5+9@6D.97%&56&55.>9%&Aa  &A@(AE5DDA697D@&7D?E&5&A-(5."A&A&D{"99@%977&7D%&+$97 %L76.?E& %.++(@&7DA 7.I&EK %& @(,E5D.97 %& 5d&K>@&AA.97 %&A ,'7&A E "9E@A %& 5 D@7A.D.97 (>.D-(5.9j6(A&7"-L6D&EA&a 5 &AD .7D(@&AA7D %& 79D&@ ?E& %&A 9!A&@ID.97A .%&7D.?E&A 97D (D( (,5&6&7D @(5.A(&A E "9E@A %dE7 ED@& >@9"&AAEA >-LA.959,.?E&` 5 %.++(@&7".D.976L9!5AD.?E&^"&?E.@&7+9@"&5&@;5&%&"-&+%d9@"-&AD@&39E(>@"&A%&EK %.++(@&7DA7.I&EK%E@(A&E"96>5&K&%&@(,E5D.97%&5d&K>@&AA.97,(7.?E&a   7A55.DD(@DE@&^5AE@j&K>@&AA.97%&A>@9D(.7&A&D{(D(@&>9@D(&E "9E@A%&5>@9,@&AA.97DE69@5&^79D66&7D%7A5&"7"&@%EA&.7a@&K&6>5&^5d75LA& E 7.I&E >@9D(.?E& %& 5d&K>@&AA.97 %&  %7A %.++(@&7D&A 5.,7(&A %& "7"&@ %E A&.7 697D@& E7& 7&DD& E,6&7DD.97 %E DEK >@9D(.?E& %&  %7A 5&A "&55E5&A !A5&A "96>@(&A EK "&55E5&A 5E6.75&Aa &A "&55E5&A !A5&A^ >5EA ,@&AA.I&A^ >@(A&7D&7D E7 >-(79DL>& 6(A&7"-L6D&EK "97D@.@&6&7D EK "&55E5&A 5E6.75&A ?E. >@(A&7D&7D E7 >-(79DL>&(>.D-(5.5a966&3&5d.%("@.D%7A5>@D.&}^5&7.I&E%d&K>@&AA.977&@&+5'D& >AD9E39E@A5d"D.I.D(%&&D{^"&A%&EK>@9D(.7&A>9EI7D)D@&69%.+.(&A>9ADj D@%E"D.977&55&6&7Da  &A%&EK69%'5&A%&>@9,@&AA.97DE69@5&?E&3d.(DE%.(AE"9E@A%&6D-'A&A97D AA&M%.++(@&7DA&D>9A&7D5?E&AD.97%&5d.6>"D%&AI@.D.97A%d&K>@&AA.97%&A>@9D(.7&A &D{E"9E@A%&5>@9,@&AA.97DE69@5&a7&++&D^3d9!A&@I&E7&AE@j&K>@&AA.97 %& %K{ E "9E@A %& 5 >@9,@&AA.97 DE69@5& %7A 5& 69%'5& ~{ &D E7& %.6.7ED.97 %& &D{E"9E@A%&5D@7A.D.97(>.D-(5.9j6(A&7"-L6D&EA&a9E@!.&7"96>@&7%@&

 {} "&DD&%.++(@&7"&^5d&K>@&AA.97%&&D{%&I@.D)D@&(DE%.(&%&AD&6>AD@'AD@%.+A %&%.++(@&7".D.97^>@'A>5EA.&E@A39E@A^>9E@AI9.@A.5%.6.7ED.97%&&D{&AD D@7A.D9.@&9E>&@67&7D&a  &>&7%7D^%7A5&A%&EK69%'5&A%d(DE%&^3d.@&5.(5&7.I&E%d&K>@&AA.97%& 9E {  %&A 5D(@D.97A 695("E5.@&A 9E >-(79DL>.?E&Aa 7A 5& 69%'5& ~{^ 5 AE@&K>@&AA.97%&%K{%7A5&A"&55E5&A~{&AD5.(&5+9.A5@(,E5D.97%&5d(>.AA,&%& 5d-.AD97&6"@9 |{&D56.,@D.97&D5d.7IA.97%&"&A"&55E5&Aa&6)6&%7A5&69%'5& {z^ 5 %.6.7ED.97 %&  &D { &7D@.7& E7 "-7,&6&7D %d(>.AA,& %& ,'7&A .6>5.?E(A %7A 5& @&69%&5,& %E "LD9A?E&5&DD& %d"D.7& 79D66&7Da & >5EA^ A. 5 %.6.7ED.97 %&  &D { %7A 5  A&6!5& .6>9@D7D& >9E@ "97D@;5&@ E7 >@9,@66& %d(>.AA,&^ 5d!95.D.97 D9D5& %&  &D { 7& >&@6&D >A %& @(5.A&@ 5 aI@.D.97%E7.I&E%d&K>@&AA.97%&&D{A&6!5&%97"D@'A.6>9@D7D& >9E@5a  7A 5&A %&EK 69%'5&A %d(DE%&A^ %K &D %K{ @(,E5&7D %&A "&7D.7&A %& ,'7&A EAA. !.&7 E 7.I&E %& 5d&K>@&AA.97 ,59!5&^ ?EdE 7.I&E %& 5d(>.AA,& 5D&@7D.+a &A +97"D.97A %& %K &D %K{ %7A 5 D@7A"@.>D.97 &D 5d(>.AA,& AE,,'@&7D E7& @(,E5D.97 %.@&"D&%&"&A-(5."A&AAE@"&A".!5&Aa&>&7%7D^"&@D.7A%&"&A,'7&A>&EI&7D)D@& @(,E5(A%&67.'@&.7%.@&"D&^79D66&7D%7A5&69%'5&%&9F3&697D@&?E&%K&D %K{@(,E5&7D {&D |^%&EK+"D&E@A%&D@7A"@.>D.97?E.^5&E@D9E@^@(,E5&7DE7 >@9,@66&D@7A"@.>D.977&5a>@&6.'@&>@D.&%&697D@I.5%&@&"-&@"-&Ad&AD+9"5.A(& AE@5d(>.AA,&%&56"@9 |{&D%dE7&%&A&A".!5&A^}a69%E5D.97%&5d(>.AA,&%& 6"@9 |{&D%&5d&K>@&AA.97%&}>&@6&D%&@(">.DE5&@E7&>@D.&%E>-(79DL>&%&5 %(>5(D.97%&&D{a&>&7%7D^5&A&++&DA!.&7?E&A.,7.+."D.+A^A97D@&5D.I&6&7D 69%&AD&Aa &A @(AE5DDA A97D DD&7%EA %7A 5 6&AE@& 9F  &D { @&,E5&7D %&A "&7D.7&A %& ,'7&A &D 5d-L>9D-'A& &AD ?EdE7 "&@D.7 796!@& %d&7D@&j&EK >@D.".>& E >-(79DL>&9!A&@I(a 57d&AD7(769.7A>A(D9777D?EdE7A&E5(I(7&6&7D%d(>.AA,&.DE7 &++&D >-(79DL>.?E&a @ &K&6>5&^ E7 %&A (I(7&6&7DA %d(>.AA,& @&D@9EI( @(,E5( %7A 5 &D>@&D{&AD5d&K97{|%& a&A%&EKI@.7DA>@9%E.DA>@"&,'7& .7+5E&7"&7D %.++(@&7D.&55&6&7D 5 6.,@D.97 "&55E5.@& m9EAA9A^ 5A69^ &D 5a^ |z{{!_ 9EAA9A^ 9AJ6.^&D5a^|z{{na  

 {}‚ !   !   )!      !   ,   *-'   & @;5& %&  %7A 5d(>.AA,& &AD @&5.(^ %7A 5 5.DD(@DE@&^  A97 .7D&@"D.97 E 7.I&E %E A.D& d %d(>.AA,& I&" 5& "96>5&K& @.!97E"5(9>@9D(.?E& { >&@6&DD7D 5 %.AA9".D.97%EA7{%&5d>@(j6&AA,&@a&D{A97DEAA..6>5.?E(&A%7A 5d(>.AA,&5D&@7D.+%&?E&5?E&A&K97Aa7A56[email protected](%&A"A%("@.DA%7A55.DD(@DE@&^  "99>'@& I&" %dED@&A +"D&E@A %d(>.AA,& >9E@ @(,E5&@ 5d(>.AA,& 5D&@7D.+a  .7D&@,.D9E>&@6&D5+.KD.97%&%.++(@&7DA+"D&E@A"966& 7{^~9E!75{E 7.I&E %E A.D& d %d(>.AA,&a 7A 5&A %&EK %&@7.&@A "A^ 5 @(,E5D.97 %& 5d(>.AA,& &AD %(>&7%7D& %E @&69%&5,& %& AD@E"DE@&A A&"97%.@&A >@ a  >@(%."D.97 &D 5 I5.%D.97 %& AD@E"DE@&A A&"97%.@&A A97D D@'A "96>5&K&A &D 5&A "977.AA7"&A %&A 6("7.A6&A%&@(,E5D.97%&5d(>.AA,&5D&@7D.+>@&D{A97DD@'A.7"96>5'D&Aa &5&AD79D66&7D%GE+.D?E&"&A(DE%&AA97D@(5.A(&AA&E5&6&7DAE@?E&5?E&A&K97A 69%'5&Aa   d>>@9"-&5@,&("-&55&ED.5.A(&%7A5c(?E.>&^%7A%.++(@&7DA69%'5&A"&55E5.@&A m~{^{z&Djn^5AE.D&%&5%(>5(D.97%&&D{^>[email protected]%d.%&7D.+.&@ %&796!@&EK&K97A5D&@7D.+A@(,E5(A>@"&A>@9D(.7&Aa&A%.++(@&7D&A(DE%&A97D>[email protected] %& %(+.7.@ %&A "@"D(@.AD.?E&A "966E7&A EK &K97A @(,E5(A >@  &D {a 7I.@97 %&EKD.&@A%&A&K97AA97D&K"5EA&DE7D.&@A&AD.7"5EA>@'A5%(>5(D.97%&&D{a 9EA"&A&K97A>>@.AA&7D%(>&7%7DA%&5d"D.I.D(-(5."A&%&9E%&{a&A &K97A 97D (,5&6&7D "966& "@"D(@.AD.?E& "966E7& %dI9.@ E7& @(,.97 d %d(>.AA,& +.!5&>@@>>9@D%dED@&A&K97A5D&@7D.+Aa9EA>@9>9A97AE769%'5&%.++(@&7D>9E@5&A &K97A.7"5EA9E&K"5EA>@5%(>5(D.97%&&D{a&A&K97A&K"5EA97DE7A.D&d %d(>.AA,& D@'A @."-& &7 a &A A(?E&7"&A +9@6&7D >5EA +".5&6&7D %&A AD@E"DE@&A A&"97%.@&A AD!5&Aa & >5EA %&A >@(%."D.97A !.9j.7+9@6D.?E&A 97D 697D@( E7 &7@."-.AA&6&7D &7 j?E%@E>5&K&Aa &A AD@E"DE@&A A97D +9@6(&A >@ %&A >5D&EK %& ~ ,E7.7&Am n^EAA.>>&5(Ag?E@D&Dhad&6>.5&6&7D>@55'5&&D.7.7D&@@96>E%dE69.7A| ?E@D&DA^.7D&@"5(A>@E7"D.976979I5&7DmA9%.E69E>9DAA.E6nAD!.5.A7D5AD@E"DE@& "97AD.DE&5& j?E%@E>5&K&a&AAD@E"DE@&A97D(D(6.A&A&7(I.%&7"&E7.I&E%EA.D&d%& "&@D.7A &K97A @(,E5(A >@  &D {a &A &K97A &K"5EA >@ 5 %(>5(D.97 %&  &D

 {}ƒ {A97DEAA.@(,E5(A>@ 7 i^%&A+"D&E@A%d(>.AA,&?E.A&+.K&7DAE@%&A@(,.97A @."-&A&7 a&D{.7D&@,.AA&7DI&" 7@7> &D5%(>5(D.97%&&D{ %.6.7E&5&@&"@ED&6&7D%& 7 AE@5&AA.D&Ad%d(>.AA,&%&A&K97A@(,E5(A>@&D {a .7A. %7A "& 69%'5&^  &D { "99>'@&@.&7D I&" 5& +"D&E@ %d(>.AA,& 7@7> i&7%(@9E57D%&AAD@E"DE@&AA&"97%.@&A%d+[email protected]7D.7A.A97@&"@ED&6&7Da  &A &K97A .7"5EA^ AE.D&  5 %(>5(D.97 %&  &D {^ >@(A&7D&7D %&A "@"D(@.AD.?E&A%.++(@&7D&Aa&A&K97A97D%&AA(?E&7"&A@."-&AE7.I&E%&5&E@AA.D&A d%d(>.AA,&a7@&I7"-&^5d&K977j{E7&@(,.97d%d(>.AA,&@."-&&7 a9EAI97A (,5&6&7D9!A&@I(E7&"99>(@D.97E7.I&E%&5@(,.97d%d(>.AA,&%&5d&K977j{&7D@&  &D 7@7> a d-L>9D-'A& &AD ?E& 5 %(>5(D.97 %& 5dE7 %& "&A +"D&E@A %.6.7E& 5 ".7(D.?E&%d(>.AA,&&D5.AA&5&D&6>A%dED@&A+"D&E@A%d(>.AA,&^"966&.{?E.A&+.K& AE@ %&A A(?E&7"&A @."-&A &7 ^ %d.7"5E@& 5d&K97 7 >9AA(%7D %&A A.D&A +.!5&Aa & 69%'5& ".7(D.?E&@&>9A&(,5&6&7DAE@5&+.D?E&"&A&K97AA97DE7.I&E%&!A&D@'A>&E.7"5EA &D 5&E@ DEK %d.7"5EA.97 E,6&7D& 59@A?E& 5d97 @5&7D. 5 I.D&AA& %d(597,D.97 %& 5 D@7A"@.>D.97a

+DDX5/17 5’5 s U1 U2 w H/F Exon n-1 U1 Exon n DDX5/17 U2 Exon n+1 33’s GC rich -DDX5/17

+DDX5/17 s U1 5’5 H/F 3’w 5’ww AT rich DDX5/17 1 ALU U2 3 U2 TIA1 GC rich TIAL1 -DDX5/17 U1

 - "*! !1#%4*3'-, !*A#/'22%!*3!1,3'$/1*!22&#*'2!2N!3JP   &DD& (DE%&  %97" >[email protected] %& 6&DD@& &7 (I.%&7"& E7 79EI&E 6("7.A6& %& @(,E5D.97%&5d(>.AA,&5D&@7D.+>A5&A-(5."A&A&D{a d.>9E@6>@D "97D@.!E(I5.%&@5&A&K97A&K"5EA>@5%(>5(D.97%&&D{&D 7@7> i&D I5.%&@5d.7D&@"D.97&7D@&^{&D 7@7> a d.(,5&6&7DI5.%(5>@(A&7"&%& j

 {~z ?E%@E>5&K&A&7ED.5.A7D%&A%@9,E&AAD!.5.A7D"&AAD@E"DE@&A&D&7(DE%.7D5d&++&D%&"&A %@9,E&AAE@5&A(I(7&6&7DA%d(>.AA,&%(>&7%7D%&&D{a 

 )! !!  , *-'   &796!@&EA&A>E"&A&K97A97D(D(@(5.A(&A%7A5d(?E.>&5AE.D&%&5%(>5(D.97 %&  &D {a &A (DE%&A^  5@,& ("-&55&^ 97D (D( 6&7(&A %7A %.++(@&7DA 69%'5&A "&55E5.@&A "966& 5&A "&55E5&A ~{^ "+j^ 7">^ {z &D |{|a &A 5.,7(&A 97D %&A 9@.,.7&A D@'A %.++(@&7D&Aa &A "&55E5&A 57"> A97D %&A 5.,7(&A -E6.7&A %& 5 >@9ADD&a &A |{|A97D%&A6L9!5AD&A[email protected]&A~{^j&Dj{zA97D%&A5.,7(&A9@.,.7.@&A%E A&.7a &A %.++(@&7D&A (DE%&A 97D >[email protected] %& %([email protected]&@ E7 6("7.A6& %& @(,E5D.97 %& 5d(>.AA,& 5D&@7D.+ >@ 5&A  -(5."A&A  &D { &D 79D66&7D %& 6&DD@& &7 (I.%&7"&5"99>(@D.97I&"5&+"D&E@ 7@7> %7A5&69%'5&%("@.D".j%&AAEAa&>&7%7D .5 &AD .7D(@&AA7D %& "96>@&@ 5&A &K97A @(,E5(A >@  &D { %7A "&A %.++(@&7DA 69%'5&Aa7>&EDA(>@&@5&A&K97A@(,E5(A>@&D{&7%&EK"D(,9@.&AadE7& >@D^  .5 &K.AD& E7 "&@D.7 796!@& %d&K97A @(,E5(A >@  &D { ?E&5 ?E& A9.D 5& "97D&KD&"&55E5.@&adED@&>@D^"&@D.7A&K97AA97D@(,E5(A>@&D%K{%&67.'@& %(>&7%7D&%E"97D&KD&"&55E5.@&a@&K&6>5&^5&A&K97A@(,E5(AE"9E@A%&5%7A 5&A"&55E5&A{z%&67.'@&%(>&7%7D&%&&D{7&A97D>A@(,E5(A>@"&A >@9D(.7&A E "9E@A %& 5 %.++(@&7D.D.97 %&A |{| 9E %7A 5&A "&55E5&A ~{ 9E ja & >5EA^ "&A &K97A A97D .6>5.?E(A %7A 5& 6)6& >@9"&AAEA^ 5 @(,E5D.97 %E "LD9A?E&5&DD& %d"D.7&a  &D{"97D@;5&7D%97"E7A9EAj&7A&6!5&%d&K97A?E.A97D.6>5.?E(A%7A 5d(D!5.AA&6&7D %dE7 >-(79DL>& >@D."E5.&@a &A >@9D(.7&A "97D@;5&7D %&A >@9,@66&A %d(>.AA,& ?E. A97D %(>&7%7DA %E "97D&KD& "&55E5.@&a 7A 5&A "&55E5&A |{|^ %K &D %K{ @(,E5&7D %&A &K97A .6>5.?E(A %7A 5 %.++(@&7".D.97 6EA"E5.@& ?E. 7& A97D >A @&D@9EI(A %7A 5&A ED@&A 69%'5&A "&55E5.@&Aa 7A 5&A "&55E5&A +j &D 7>^  "&A >@9D(.7&A@(,E5&7D%&A&K97A%&,'7&A.6>5.?E(A%7A5@(,E5D.97%&AI9.&A%&A.,75.AD.97 %&A-9@697&AAD(@91%.&77&Aa&DD&@(,E5D.97%(>&7%7D&%E"97D&KD&>&EDAd&K>5.?E&@&7 >@D.&>@5&+.D?E&5&A,'7&A"97D@;5(AE7.I&E%&5d(>.AA,&>@&D{A97D &K>@.6(AA>(".+.?E&6&7D%7A"&A69%'5&Aa

 {~{  9EA 697D@97A E7 69%'5& %d(>.AA,& %& "99>(@D.97 &7D@& i{ &D 7@7> i E 7.I&E %E A.D& d %d(>.AA,& ?E. &AD @."-& &7 a &>&7%7D^ &7 +97"D.97 %E "97D&KD&"&55E5.@&^%dED@&A+"D&E@A%d(>.AA,&L7DE7&++.7.D(>9E@%&AA.D&A%&+.KD.97 @."-&A &7  >9E@@.&7D "99>(@&@ I&" 5&A  -(5."A&Aa 7A 5& 69%'5& {z 79EA I97A %&A @(AE5DDA >@(5.6.7.@&A ?E. 697D@&7D ?E& "&@D.7A &K97A @(,E5(A >@  &D { E "9E@A %& 5  A97D (,5&6&7D %(>&7%7DA %& 7@7> i &D 5&E@A @(,E5D.97A >9E@@.&7D"9@@&A>97%@&E69%'5&?E&3d.%(3%("@.DadED@&A&K97A7&A97D"&>&7%7D >A %(>&7%7DA %& 7@7> ia 9E@ "&A &K97A^ 79EA I97A %&A @(AE5DDA >@(5.6.7.@&A ?E. 697D@&7D?Ed.5AA97D%(>&7%7DA%&A+"D&E@AA@>{&DA@>|^%&A+"D&E@A%d(>.AA,&!.&7 "@"D(@.A(AE"9E@A%&5a&>5EA^3d.697D@(?E&A@>{&D.7D&@,.AA&7D%7A 5&A "&55E5&A {za .7A.^  &D { >9E@@.&7D @(,E5&@ %&A &K97A 5D&@7D.+A (,5&6&7D &7 +97"D.97 %& 5d&K>@&AA.97 %& +"D&E@A %d(>.AA,& &K>@.6(A %7A 5&A 69%'5&A (DE%.(Aa              





 {~| '/1,,' ()0$./





 

 {~} 

 



*!4//*#+!,3'1!J? '23! !2/1','/47$3!412 A#/'22%! !2$+'**!2-4&, ',2'04!*!2$3!412 A#/'22%!!7/1'+#2 !+,'"1!3'22422/#'$'04!2@ A/1"2&!,H,*!8KIIR@ /3#/1@-*82/',-9@   

 {~~        





*!4//*#+!,3'1!K? '23! !2/1','/4731,21'321#%4*#2/1#/'22%!*3!1,3'$4 -412 !*1,2'3'-,/'3&#*'-C#2!,&8+3!42!@ -,,#!2-+/'*#!2/13'1 !!,*!2H*@KIJK>19!&H123!,KIJL> 1!43'%+H*@,-%!,!KIJL>&/'1-H*@ -%!,!3'2KIJJ!3'+-,3'H*@  KIJK@       

 {~  ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 7? 7? 7? 7?   # >7!'>   !  7@ 7? 3(,3%.,.%$  %"4."('.(.!)%,& '(&)%2,   ., )!(,)!(+3%."(' !(+3%."'+,)(',.( 8>  "'6&3"' %/'."0".3  (%( (/,+ "(' (%((/,+ +&(%/%+ !("'"' ,."%"4."('(&"' !("'"'  " "".36" 8%"$(&"' " "".36" 7? %%(1, (+&(++" "."'+'!"' "."'+'!"' %%(1, (+&(++"  6 .,,/%%/%+%(%"4."(' 7? B; B; @; B; B; 8.+&"'% 8.+&"'% ? 8.+&"'% +).(1',.+& )!(,)!(+3%."(',"..!.&3",+/).."'"'"' )!(,)!(+3%."(',"..!.&3",+/).."'"'"' 7?  " 8%"$(&"' 3.()%,&"(&"' %('.8.+&"'/,  8%"$(&"'B9 8: )!(,)!(+3%."('&(."  (+$.> .()%,&"(&"'6!' ,8.+&"'/,,*/' .()%,&"(&"'6!'  "%."0."('( '.7-8.'"'.+',.  &"'6'())."&3'!'"'.+."('1".!0%@  "' %/'!(&("&+"4."('(+"'.+."('1".! A'   ("%8("%(&"'6&3,)" 3"'.+."(')+.'+, '(3(1',.+&2('",+*/"+ (+"'.+."('1".!? $  #  !$  !(,)!(.3+(,"'"'"'(&"'6.!'())."(' +,%( '(,)).")+(2"&%.(.! "+,.(&"'6)+."%%3 ('% "('6+.,!0"%3 2.+8%%/%+&&+')+(2"&%+ 0%()&'. #/'."(' +.( )+(%"'+"!(&"'  (+&."(' './%."&.2('1".!,.()(('6" +'."%%3)!(,) ,$%.(' 2(3.(,", (,$%.(' (+ '"4."(' .1'22&(." ')%$,.+"'!(&(%( 3(& %%+!"../+(&"'6'!', (%!,"( 2"%%"'8"'"' &&&+' '(,)).".1'.!.1((&"',  .(+."0".3 %%)(%+".3 .6'(3.(,", (''!,"(' ,$%.('%%&(."(' (,$%.('%%&(."(' (,$%.('%%&(."(' (,$%.('%%&(."(' )+(."'," '%"' ).!13 '%"' )+(."'," ( %%!,"(''," '%"' '%"' ( %%!,"(''," ("%8("%(&"'68.+&"'% ,)(+.6)+"&+3&.(%")+(,, ,+ /%.(+."0".36."'3.(,$%.(' (+."0".3"'.+%%/%+)+(."'.+',)(+.6%%&(."(' !3+(8%3,."0".3 .+',+")."(' .(+."0".3 .+',+")."(' .(+."0".3 /%.(+( ."'3'&", 3(&"'6('."', '7,)!(&(%( 8."'+(,,8%"'$"')+(."' (&"'6 ('.+"/.,.(.!!"' ,.!."',3.(,$.%)+(."',(2"(+/.,."0".3%%!,"(' %$,.+"'!(&(%(3(&"' '0(%0"'."0."('( '.," '%"' '%"' '0(%0"'."0."('( '.," +".("%("%( +',&&+'+).(+.3+(,"'$"', (' +,%",) '"'"' ." !.#/'."('%%!,"('." !.#/'."(' !.#/'."('%%!,"('." ." %)!,//'".( "'.+"'5%&"'"'+).(+ !.!"''3 " +).(+"'"' '/+(%( "%,3,.&)+(,, '/+(%( +).(+"'"' !(8,)" "/'"''/%(."2!'  .(+ 8.+&"'/,6&3 !3+(%,."0".36.+',&&+'.+',)(+.+."0".3 ?=8%"$$"',1".!+(%"')+(&(."' %%&(."%".3 ?=8%"$$"',1".!+(%"')+(&(."' +". '%"' ).!13 8)+(."'(/)%+).(+)+(."'," '%"' ).!13 +',+")."(' .(+"'0(%0"''.," " +'. "',.!.&".,8-' ," '%.+',/."(' "',.!.&".,8-' ," )." ,.%",!&'.(+&"'.''( !+(&."'+!"../+ %%8,/+  %3()+(."'"'0(%0"'%%!,"(''&" +." %3()+(."'"'0(%0"'%%!,"(''&" %%8,/+  ,3')."0,"%2(3.(,",6"'.+%%/%+)+(."'.+',)(+.6 ,.+/./+%(',."./'.( 3.(,$%.('6."'"'"' ."'3.( ,.+/./+%(',."./'.( 3.(,$%.('6."'"'"' ."'3. ,.+/./+%(',."./'.( 3.(,$%.('6."'"'"' ,8."0."' )+(."'"'0(%0"'&"'.''( .!."!. ,8."0."' A? .(+)+(."'"'0(%0"'%&%%")("' "%()(" .+',+")."(' .(+."0".36+).(+"'"' 6.+',+")."('  .+',+")."(' .(+."0".36+).(+"'"' ,&%%,+ /%.(+."0".3"'.+%%/%+)+(."'.+',)(+ ,&%%,+  (%"')+(."',,(".1".!(&)%2,..!"''+)%,  /'"''/%(."2!'  .(+"'0(%0"'3.(,$%.('+ /'"''/%(."2!'   )+(."'.!."'.+.,1".!+/&,(&)%2.(+ /%.  )+(."'.!."'.+.,1".!+/&,(&)%2.(+ (&)%2)+(."'"&)%"."'&'3+(%,"'%/"'%%&"+." ,.+/./+%(',."./'.( 3.(,$%.('6."'"'"' ."'3.( ,.+/./+%(',."./'.( 3.(,$%.('6."'"'"' ."'3. ,.+/./+%(',."./'.( 3.(,$%.('6."'"'"' ."'3. ,.+/./+%(',."./'.( 3.(,$%.('6."'"'"' ."'3. ,.+/./+%(',."./'.( 3.(,$%.('6."'"'"' (/%8,.+'"'"'' &,(+&0%()&'.6&/,%(+ 8)+(."'(/)%+).(+."0".38)+(."'(/)%+).(+ %%!,"(''," '%.+',/."('%.8.'"'6+ /%.(+ '%.+',/."('%.8.'"'6+ %%!,"(''," (',."./'.( 3.(,$%.('6&"+(.//%60,"%8&"..+' ,.+/./+%(',."./'.( 3.(,$%.('6)+(."'"'"' 6,&%% ,.+/./+%(',."./'.( 3.(,$%.('6)+(."'"'"' ,&%%,+ /%.(+."0".36 /'3%8'/%(."2!'  . ./&(+"'0,"('0""'!""."('( 8!+"'  ?+."(' ./&(+,/))+,,(+6"'!"".(+( %%&"   8.+&"'%"'. (' ,."%"4."('( 8!+"'.!+',#/'."(', 3+(,"' "',).(+ ..'  ? ,? .0> AA >?>?  ? '!'&'.( %%&"+."(''"'0,"(' )".!%"%%%&(+)!(',",'."0."('   /%."('( ."'+&(%"' +  @ '.," '%"' ('",. '. @   8?!.#/'."(','  (%"')+(."',,(".1".!." (+&."('',."%".3( %%#/'."(',+@ !(&(8(&"'!(& ($E C   > ? '/."('(   ? ? '%> ,."%"4."('( !+',#/'."(', >? > ., .+',+")."('%."0."('( &,'!3&%.+    ? "'.+'%   %!    #  >   @ '/."('(   >  @ > + /%."('( ."'3.(,$%.('   ? >?C >>  ?  >+."(' ./&(+%%&(."%".3'&"  ?  >> >@  ? )+(&(.,)".!%"%%%8%%!,"(' '.).!13+ /%."(' D ?9 A:  9 ': @ D9$>:  >9")>:  ?9 +,)B@: A> B9&(>:  >D9"!>: >9)>?=8.'"':  >>98!(:   #      "#      #!    #     {~€  & "      )  +*   '

  7A 5& >@&6.&@ @D."5& %& 6 D-'A&^ 3d. @(5.A( E7& (DE%& +97"D.977&55& AE@ 5&A I@.7DA%d(>.AA,&%&5d-.AD97&6"@9 |{a+97"D.97%&"&DD&-.AD97&&D5&@;5&%&A&A %&EKI@.7DAA97D>&E"977EAad9!3&"D.+%&"&DD&77&K&&AD%&%("@.@&5AD@E"DE@&%&"&A %&EKI@.7DA^%&%.A"ED&@%&5d.6>5."D.97%&"&DD&AD@E"DE@&%7A5@(,E5D.97%&A,'7&A&D %&%(D.55&@5&A(I.%&7"&A?E.@&5.&7D"&AI@.7DA%d(>.AA,&5>@9,@&AA.97DE69@5&a  d-.AD97& 6"@9 |{ &AD "96>9A(& %& [email protected] %96.7&A %.AD.7"DAa 7 %96.7& j [email protected]5 |j5.4&m&7!5&EAE@5+.,E@&{n^E7%96.7&5.74&@797AD@E"DE@(m&7I&@Dn&DE7 %96.7&6"@9&7[email protected]5m&7@9E,&na&%96.7&6"@9&AD%&EK+9.A>5EA,@7%?E&5& %96.7& |j5.4& &D 7d&AD >A .7"5EA %7A 5& 7E"5(9A96&^ E "97D@.@& .5 A9@D %& 5 "-@96D.7& mI9.@ +.,E@& |na &EK &K97A "9%7D E7& >@D.& %E %96.7& 6"@9 A97D "977EA >9E@)D@&6EDE&55&6&7D&K"5EA.+"&?E.>&@6&D5>@9%E"D.97%&A.A9+9@6&A6"@9 |{a{&D 6"@9 |{a|a&A%&EK.A9+9@6&A%.++'@&7D>@E7&"9E@D&A(?E&7"&%d".%&A6.7(AmI9.@ +.,E@& {n ?E. ++&"D& 5 D.55& &D 5d-L%@9>-9!.".D( %dE7& >9"-& %& 5..A97a 5  .7A. (D( %(697D@(?E&A&E5&5d.A9+9@6&6"@9 |{a{&AD">!5&%&+.K&@%&A6(D!95.D&A%(@.I(A%E ‡m EADDA"-&@^ 9D-9@7^E,.&EK^"-&++M&4^s%E@7&@^|zznad.A9+9@6&6"@9 {a| 7d(D7D >A ">!5& %& +.K&@ "&A 6(D!95.D&A^ "&DD& .A9+9@6& >>@.D %7A "& "97D&KD& "966&E7%96.77D7(,D.+ad-.AD97&6"@9 |{&AD%97""97A.%(@(&"966&E7A&7A&E@ %& 6(D!95.D&A E 7.I&E %& 5 "-@96D.7&a 5EA.&E@A >@9D(.7&A A97D ">!5&A %& +.K&@ 5& %96.7&6"@9%&6"@9 |{&D5+.KD.97%&6(D!95.D&>@5d.A9+9@6&6"@9 |{a{ >9E@ "97A(?E&7"& %& 69%.+.&@ 5 "97+9@6D.97 %E %96.7&^ "&5  (D( AE,,(@( "966& >9EI7D 69%.+.&@ %&A .7D&@"D.97A >@9D(.7&j>@9D(.7&A m9AI&"^ .6.7AM4L^ s EA"-!&"4^ |z{}na  d-.AD97& 6"@9 |{  %d!9@% (D( "97A.%(@(& "966& E7 @(>@&AA&E@ %& 5 D@7A"@.>D.97a55&(D(@&5.(&5d.7"D.ID.97%E"-@969A96&"-&M5+&66&m9AD7M.s &-@A97^{ƒƒ‚naE@5&AED9A96&A^"&DD&-.AD97&9""E>&%&A@(,.97A,(796.?E&AAA9".(&A  5 @(>@&AA.97 D@7A"@.>D.977&55& m 6!5&^ @.MM&55^ 7,^ @.A-74E6@^ s @EA^ |z{zna &>&7%7D^ %&A (DE%&A @("&7D&A 97D AE,,(@( ?E& "&DD& -.AD97& &AD .6>5.?E(& (,5&6&7D %7A5d"D.ID.97%&"&@D.7A,'7&Am@&>>&&D5a^|z{|_ 6!5&&D5a^|z{zna

 {~                  .,E@&{`.++(@&7"&A!.9"-.6.?E&A&DAD@E"DE@5&A%&A%&EK.A9+9@6&A%&5d-.AD97& 6"@9 |{_d>@'A 4233&!1!3*@>KIJIa          .,E@&|`.++(@&7"&A%&AD@E"DE@&A&7D@&5&AI@.7DA6"@9 |{a{&D6"@9 |{a|&D "97A(?E&7"&A>9E@5+.KD.97%&6(D!95.D&Aad>@'A-25!!3*@!**@-*@ '$!'@> KIJL@ 

 {~‚  &796!@&EA&A(I.%&7"&A@&5.&7D"&DD&-.AD97&5>@9,@&AA.97DE69@5&mI9.@+.,E@& ~na&>&7%7D^%&796!@&EA&A(DE%&A7&>@&77&7D>A&7"96>D&5&A%.++(@&7D&A.A9+9@6&A %&"&,'7&a&A(DE%&A?E.D.&77&7D"96>D&5d(>.AA,&%&"&I@.7D@(I'5&7D%&>@9+97%&A %.++(@&7"&A +97"D.977&55&A &7D@& 5d.A9+9@6& 6"@9 |{a{ &D 6"@9 |{a|a & I@.7D %d(>.AA,& 6"@9 |{a{ &AD AA9".(  E7 @;5& %& DE6&E@ AE>>@&AA&E@a d&K>@&AA.97 %& "& I@.7D "9@@'5& %& 67.'@& .7I&@A& I&" 5 >@95.+(@D.97 %7A %&A DE6&E@A %E A&.7 &D %E >9E697m >9@7&D5a^|zzƒna&67.'@&>5EA,(7(@5&^5d&K>@&AA.97%&6"@9 |{a{&AD %.6.7E( %7A %&A D.AAEA "7"(@&EK %& >9E697^ I&AA.&^ D&AD."E5&^ 9I.@&^ "9597^ A&.7 &7 "96>@.A97 %& D.AAEA A.7A "9@@&A>97%7DA m9I.49I &D 5a^ |z{{_ E%.D-  >9@7 s E7,^ |z{|na  5d.7I&@A&^ 5d&K>@&AA.97 &D 5 +97"D.97 %& 6"@9 |{a| A&6!5&7D %(>&7%7D&A %E "97D&KD& &D %E "7"&@ (DE%.(a 7A 5& 6(5796&^ 6"@9 |{a| @(%E.D 5& >9D&7D.&5 6(DADD.?E& %&A "&55E5&A m >99@ &D 5a^ |z{zna  5d.7I&@A& %7A 5& "7"&@ %E A&.7^ &7 >@D."E5.&@ %7A %&A "&55E5&A j| >9A.D.I&A^ 6"@9 |{a| .7D&@,.D &D @(,E5& j| &D +[email protected]& 5 6.,@D.97 &D 5d.7IA.97 %&A "&55E5&A DE69@5&A m. &D 5a^ |z{|na E 7.I&E 695("E5.@&^ "&A %&EK I@.7DA 97D (D( %("@.DA "966& @(,E57D >9A.D.I&6&7D 9E 7(,D.I&6&7D5D@7A"@.>D.97%&,'7&AAA9".(A5>@9,@&AA.97DE69@5&a 

             

 {~ƒ  

 .,E@&}`.AD&%&A.7D&@"D7DA"977EA%&5d-.AD97&6"@9 |d>@'A42&!)!3' 1-!>KIJI  

.,E@&~`!A&@ID.97A.6>5.?E7D5d-.AD97&6"@9 |%7A5&A"7"&@A_d>@'A,31',- !3*@,!1 !33!12>KIJL@

 {z  

   E"9E@A%&6D-'A&^3d.(,5&6&7D>@D.".>(%.++(@&7DAD@IEK%&5d(?E.>&a7A 5&A"&55E5&Ajj|}{^79EAI97A697D@(?E&&D{A97D%&A@(,E5D&E@A%E +"D&E@ %& D@7A"@.>D.97 ^ E7 +"D&E@ %& D@7A"@.>D.97 .6>5.?E( %7A 5 6.,@D.97 &D 5d.7IA.97a 7 &++&D^ "&A  -(5."A&A A97D %&A "9j@(,E5D&E@A D@7A"@.>D.977&5A %&  6.A A97D (,5&6&7D @&A>97A!5&A %& 5d.7"5EA.97 %& 5d&K97  %&  ?E. .7%E.D  5 %(,@%D.97%&5d%&a7A"&>@93&D^3d.@(5.A(5&A%.++(@&7D&A&K>(@.&7"&A%& 6.,@D.97&D%d.7IA.97>9E@%(697D@&@5&@;5&%&^{&D%7A56.,@D.97 &D5d.7IA.97"&55E5.@&%&A"&55E5&Ajj|}{a   7A&"97%>@93&D%(I&59>>(%7A5d(?E.>&Ad&AD+9"5.A(AE@5&@;5&%&&D{ %7A5@(,E5D.97%&AI9.&A%&A-9@697&AAD(@91%.&77&Aa9EAI97AED.5.A(%&A"&55E5&A%& "7"&@%EA&.7"+j?E.@(>97%&7DEK9&AD@9,'7&A&D5&A"&55E5&A%E"7"&@%&5>@9ADD& 57> ?E. @(>97%&7D  5 D&AD9AD(@977&a 7A "&A %&EK 69%'5&A^ 79EA I97A 697D@( ?E& &D{A97D%&A@(,E5D&E@A63&E@A%&AI9.&A9&AD@9,(7.?E&AadE7&>@D"&A -(5."A&A @(,E5&7D E 7.I&E D@7A"@.>D.977&5 &7I.@97 z† %&A ,'7&A @(,E5(A >@ 5&A -9@697&AAD(@91%.&77&Aa&>5EA&D{@(,E5&7DE7.I&E%&5d(>.AA,&%&A,'7&A "9%7D >9E@ %&A @(,E5D&E@A %E @("&>D&E@ EK 9&AD@9,'7&A &D EK 7%@9,'7&Aa 7A "& >@93&D^3d.@(5.A(5&A&K>(@.&7"&AI&"5&69%'5&7>a   &A%&EK(DE%&A@&7+9@"&7D5&@;5&%&A-(5."A&A&D{E"9E@A%&5 >@9,@&AA.97DE69@5&%7A%&A69%'5&A%&"7"&@%EA&.7&DA9E5.,7&5d.6>9@D7"&%&5&E@ +97"D.97%7A5d(>.AA,&%7A"&>@9"&AAEAa 

 {{ Oncogene (2012) 31, 4536–4549 & 2012 Macmillan Publishers Limited All rights reserved 0950-9232/12 www.nature.com/onc ORIGINAL ARTICLE Dual role of the ddx5/ddx17 RNA helicases in the control of the pro-migratory NFAT5 transcription factor

S Germann1,2,3,4,5, L Gratadou1,2,3,4,5, E Zonta1,2,3,4,5, E Dardenne1,2,3,4,5, B Gaudineau6, M Fouge` re6, S Samaan1,2,3,4,5, M Dutertre1,2,3,4,5, S Jauliac6 and D Auboeuf1,2,3,4,5

1Universite´ de Lyon, Lyon, France; 2Inserm U1052, Lyon, France; 3CNRS UMR5286, Lyon, France; 4Centre de Recherche en Cance´rologie de Lyon, Lyon, France; 5Universite´ Lyon 1, Lyon, France and 6CNRS UMR7212, INSERM U944, Universite´ Paris Diderot, Institut d’He´matologie, Hoˆpital Saint-Louis, Paris, France

Ddx5 and ddx17 are two highly related RNA helicases directly linked to the migratory and invasive phenotype involved in both transcription and splicing. These of cancer cells (Friedl and Wolf, 2003). Invasion is a proteins coactivate transcription factors involved in complex process relying on the capacity of the cells to cancer such as the estrogen receptor alpha, p53 and migrate and to destroy and reorganize the extracellular beta-catenin. Ddx5 and ddx17 are part of the splicing matrix. It is now well established that tumor progression machinery and can modulate alternative splicing, the relies on the alteration of transcriptional programs main mechanism increasing the proteome diversity. controlling specific cellular programs (Hanahan and Alternative splicing also has a role in gene expression Weinberg, 2000). In this context, the NFAT family of level regulation when it is coupled to the nonsense- transcription factors is gaining increasing interest in mediated mRNA decay (NMD) pathway. In this work, breast cancer. This family comprises five genes, we report that ddx5 and ddx17 have a dual role in the NFAT1 to NFAT5. NFAT1, NFAT2, NFAT3 and control of the pro-migratory NFAT5 transcription NFAT4 have been first identified as T-cell transcription factor. First, ddx5 and ddx17 act as transcriptional factors, whereas NFAT5 has been involved in the coactivators of NFAT5 and are required for activating cellular response to osmotic stress (Macian, 2005; Burg NFAT5 target genes involved in tumor cell migration. et al., 2007; Mancini and Toker, 2009; Muller and Rao, Second, at the splicing level, ddx5 and ddx17 increase 2010). It is now well documented that NFAT proteins the inclusion of NFAT5 exon 5. As exon 5 contains a are also present in non-immune cells and regulate a pre-mature translation termination codon, its inclusion variety of signaling pathways involved in cell growth and leads to the regulation of NFAT5 mRNAs by the NMD development (Baksh et al., 2002; Chuvpilo et al., 2002; pathway and to a decrease in NFAT5 protein level. Mancini and Toker, 2009; Muller and Rao, 2010). Therefore, we demonstrated for the first time that a Importantly, members of the NFAT family, in particular transcriptional coregulator can simultaneously regulate NFAT5, have recently been involved in the migratory the transcriptional activity and alternative splicing of a capacity of breast cancer cells (Jauliac et al., 2002; Ayers transcription factor. This dual regulation, where ddx5 et al., 2004; Yoeli-Lerner et al., 2005; Mancini and and ddx17 enhance the transcriptional activity of Toker, 2009; Fougere et al., 2010; Muller and Rao, NFAT5 although reducing its protein expression level, 2010). suggests a critical role for ddx5 and ddx17 in tumor Increasing evidences indicate that alterations of cell migration through the fine regulation of NFAT5 splicing programs contribute to tumor progression pathway. (Blencowe, 2003; David and Manley, 2010; Warzecha Oncogene (2012) 31, 4536–4549; doi:10.1038/onc.2011.618; et al., 2010; Dutertre et al., 2010b). Indeed, most of the published online 23 January 2012 human genes can generate different splicing variants coding for different protein isoforms having slightly Keywords: ddx5; ddx17; NFAT5; cancer; transcriptional different activities or even opposite biological activities coregulator; splicing (Stamm et al., 2005; David and Manley, 2010; Dutertre et al., 2010b). In addition, alternative splicing can have an important role in the regulation of gene expression level. Indeed, about one third of alternative exons Introduction contain a pre-mature translation termination codon resulting in mRNA regulation by the nonsense- Breast cancer is a leading cause of morbidity in women mediated mRNA decay (NMD) pathway. Therefore, worldwide because of metastasis formation, which is the coupling of alternative splicing with NMD may provide a general means of decreasing protein expres- Correspondence: Dr D Auboeuf, Centre de Recherche en Cance´rologie sion level. This mechanism that has been referred to de Lyon, Centre Le´on Be´rard, 28 Rue Lae¨nnec, Lyon F-69008, France. E-mail: [email protected] as ‘regulated unproductive splicing and translation’ Received 14 July 2011; revised 1 November 2011; accepted 28 November or RUST has been involved in the downregulation of 2011; published online 23 January 2012 several cancer-related proteins (Lewis et al., 2003;  {|

 Regulation of NFAT5 by ddx5/ddx17 S Germann et al 4537 Barbier et al., 2007; Chang et al., 2007; Neu-Yilik and Results Kulozik, 2008; Gardner, 2010). Although transcriptional programs are under the Ddx5 and ddx17 are required for mediating the pro- control of transcriptional factors and coregulators, migratory effect of NFAT5 splicing programs are under the control of splicing Ddx5 and ddx17 have been reported to be transcrip- factors (Blencowe, 2003; David and Manley, 2010; tional coactivators of several key transcriptional factors, Warzecha et al., 2010; Dutertre et al., 2010b). An including estrogen receptor alpha, p53 and beta-catenin increasing number of proteins have been involved in and are likely to have a role in tumor initiation and/or both transcription and splicing (Auboeuf et al., 2007; progression (Janknecht, 2010; Fuller-Pace and Moore, Allemand et al., 2008). This is the case of the highly 2011). As ddx17 has been co-purified with NFAT5 related ddx5 and ddx17 RNA helicases (also known as (Chen et al., 2007), we tested whether ddx17 and its p68 and p72, respectively) that act both in transcription paralog, ddx5, co-immunoprecipitate with NFAT5 in and splicing. Indeed, ddx5 and ddx17 are transcriptional human MDA-MB-231 breast cancer cells. As shown on coregulators of the estrogen receptor alpha, p53, beta- Figure 1a (left panel), FLAG-ddx17 protein was catenin and MyoD transcription factors among others specifically detected after the immunoprecipitation of a (Watanabe et al., 2001; Bates et al., 2005; Caretti et al., Myc-NFAT5 protein. Conversely, Myc-NFAT5 protein 2006; Fuller-Pace and Ali, 2008). It is believed that ddx5 was specifically detected after the immunoprecipitation and ddx17 are recruited on target gene promoters by of FLAG-ddx17 (right panel, Figure 1a). Likewise, these transcriptional factors and in turn recruit the Myc-NFAT5 protein was detected after the immuno- RNA polymerase II or enzymes with histone acetylase precipitation of endogenous ddx5 protein (left panel, or deacetylase activities (Metivier et al., 2003; Rossow Figure 1b) or after the immunoprecipitation of a HA- and Janknecht, 2003; Wilson et al., 2004; Janknecht, ddx5 protein (right panel, Figure 1b). Finally, immuno- 2010; Dutertre et al., 2010a; Fuller-Pace and Moore, precipitation of the endogenous NFAT5 protein, co- 2011). In addition, ddx5 and ddx17 copurify with the immunoprecipitate the endogenous ddx5 protein but not splicing machinery or spliceosome and can change when cells were first transfected with a siRNA targeting alternative splicing-site selection in transcripts produced NFAT5 (Figure 1c). from the H-ras, CD44 and Tau genes (Auboeuf et al., Because we could not immunoprecipitate endogenous 2002; Honig et al., 2002; Guil et al., 2003; Camats et al., ddx17 with endogenous NFAT5 protein, the ability of 2008; Clark et al., 2008; Kar et al., 2011). There are now endogenous NFAT5 protein to associate with endogen- many reports indicating that these multifunctional ous ddx17 and ddx5 proteins was further assessed with proteins have important implications for cancer devel- the in situ proximity ligation assay (PLA), which generates opment, as recently reviewed (Janknecht, 2010; Fuller- a signal when two proteins are in close proximity (B40 nm) Pace and Moore, 2011). For example, on one hand, as to each other. As shown on Figure 1d, prominent signals transcriptional coactivators of estrogen receptor alpha, were detected in fixed MDA-MB-231 cells between they may contribute to the proliferative effect of anti-NFAT5 and anti-ddx5, and anti-NFAT5 and anti- estradiol on breast cancer cells (Wortham et al., 2009; ddx17 antibodies but not in control experiments. Dutertre et al., 2010a) and, as coregulators of beta- Numbers of signals were on average of 14 per cells after catenin, they may contribute to the epithelial to incubation of anti-NFAT5 and anti-ddx5 antibodies, mesenchymal transition, which has been associated with and of 11 per cells after incubation of anti-NFAT5 and breast cancer progression (Yang et al., 2006). On the anti-ddx17 antibodies (Figure 1e). Moreover, the other hand, as coactivators of p53 and Smad, ddx5/ number of signals markedly decreased in MDA-MB- ddx17 may exert tumor suppressor functions (Warner 231 cells after transfection with siRNAs targeting either et al., 2004; Bates et al., 2005). However, oncogenic NFAT5 or both ddx5 and dd17 (Figure 1f). functions or tumor suppressor roles of ddx5 and ddx17 In addition, the over-expression of ddx5 and ddx17 are still a matter of debate and could be context- enhanced the transcriptional activity of NFAT5 as dependent (Fuller-Pace and Moore, 2011). measured using a luciferase reporter gene driven by an In this work, we report that ddx5 and ddx17 have a NFAT5-responsive promoter suggesting that ddx5 dual role in the control of the pro-migratory NFAT5 and ddx17 are NFAT5 transcriptional co-activators transcription factor. First, ddx5 and ddx17 act as (Figure 1g). To test further this hypothesis, we next transcriptional coactivators of NFAT5 and are required analyzed the effect of ddx5 and ddx17 on NFAT5 pro- for activating NFAT5 target genes involved in tumor migratory function. cell migration. Second, at the splicing level, ddx5 and Indeed, it has been shown that NFAT5 is a ddx17 favor the inclusion of the human NFAT5 exon 5 transcriptional factor that stimulates cell migration that contains a pre-mature translation termination (Jauliac et al., 2002; O’Connor et al., 2007). As expected, codon, which then results in the synthesis of the NFAT5 over-expression in MDA-MB-231 cells in- unproductive (that is, not translated) NFAT5 mRNAs creased cell migration (Figure 2a), whereas NFAT5 and in the reduction of the NFAT5 protein level. This depletion decreased it (Figure 2b). The depletion of dual activity of ddx5 and ddx17 may have a role in the either ddx5 or ddx17 had almost no effect on cell fine regulation of the NFAT5 pathway and may migration (Figure 2c). However, the depletion of either contribute to the context-dependent role of ddx5 and ddx5 or ddx17 further inhibited cell migration mediated ddx17 in cancer. by the decrease in NFAT5 expression level (Figure 2d,

 {} Oncogene Regulation of NFAT5 by ddx5/ddx17 S Germann et al 4538 Supplementary Figure S1). In addition, the depletion of Collectively, these results suggest that ddx5 and ddx17 both ddx5 and ddx17 strongly decreased cell migration have a role in cell migration by co-activating NFAT5. (Figure 2e) and, finally, NFAT5 depletion did not To further test this hypothesis, the ddx5 and ddx17 inhibit cell migration in the absence of ddx5 and ddx17 effect on NFAT5 endogenous target genes was next (comparing lane 2 to lane 4, Supplementary Figure S2). investigated.

IP IP IP IP Myc Myc IN IN FLAG FLAG IN IN 260 260 IB Myc IB Myc

80 80 IB Flag IB Flag

Myc-NFAT5 + - + - Myc-NFAT5 +++ + FLAG-ddx17 + + + + FLAG-ddx17 + - + -

IP IP IP ddx5 IgG IN HA HA IN IN 260 260 IB Myc IB Myc IB HA 60 IB ddx5 Myc-NFAT5 + + + + 60 HA-ddx5 + - + - Myc-NFAT5 + + +

IP IP 260 NFAT5 NFAT5 IN 260 IB

NFAT5 IB ddx5 actin NFAT5 60 40 siN5 -+ siN5 -+-+ siN5 - +

In situ interaction Controls NFAT5 ddx5 NFAT5 ddx5

NFAT5 ddx17 NFAT5 ddx17

16 6 *** *** 12 14 10 *** 5 12 *** 4 10 8 *** ** 8 6 3 6 4 2 Relative firefly

PLA signals/cell 4 PLA signals/cell 2 luciferase activity 1 2

siddx5/17 - - + - - + NFAT5 - + -+- + ddx5 siN5 - + --- + ddx5 - - + + - - ddx17NFAT5 ddx17 - - - - + + NFAT5 NFAT5 NFAT5-ddx5 NFAT5-ddx17 ddx5 ddx17

Oncogene  {~

 Regulation of NFAT5 by ddx5/ddx17 S Germann et al 4539 It has been reported that NFAT5 regulates the which is a member of the S100 family of calcium-binding transcriptional activity of the S100A4 gene in a direct proteins, has been shown to have an important role in manner (Chen et al., 2009, 2011). S100A4 or metastasin, breast cancer progression (Helfman et al., 2005; Garrett

CTRL NFAT5 siCTRL siN5 260 260 siCTRLsiddx5 siCTRLsiddx17 80

ddx5 60 ddx17 NFAT5 NFAT5 160 160 actin actin actin actin 40 40 40 40 200 500 200 ** 400 150 150 300 100 100 ** 200 Number of 50 Number of Number of migrating cells

migrating cells 100 50 migrating cells siddx5 --+ NFAT5 - + siN5 - + siddx17 -+-

200 *** *** 200 150 *** siCTRL siddx5/17 150

100 ddx5 60 80 100 ddx17 Number of 50 Number of 50 migrating cells migrating cells actin 40 siN5 - + --+ + siddx5/17 - + siddx5 - - + - + - siddx17 - - - + - + Figure 2 NFAT5 and ddx5/ddx17 have a role in cell migration. (a) WB analysis (upper panel) of NFAT5 and actin, as a control, 48 h after transfection of MDA-MB-231 cells with a control or NFAT5 expression vectors. Number of migrating cells (lower panel). (b)WB analysis (upper panel) of NFAT5 and actin 48 h after transfection of MDA-MB-231 cells with a control (siCTRL) or a siRNA targeting NFAT5 (siN5). Number of migrating cells (lower panel). (c) WB analysis (upper panel) of ddx5, ddx17 and actin, 48 h after transfection of MDA-MB-231 cells with a control siRNA or siRNAs targeting either ddx5 (siddx5) or ddx17 (siddx17). Number of migrating cells (lower panel). (d) Number of migrating MDA-MB-231 cells after transfection with a control siRNA, siRNAs targeting either ddx5 (siddx5) or ddx17 (siddx17) and/or a siRNA targeting NFAT5 (siN5). (e) WB analysis (left panel) of ddx5, ddx17 and actin, 48 h after transfection of MDA-MB-231 cells with a control siRNA or a single siRNA (siddx5/17) targeting a conserved region in the ddx5 and ddx17 mRNAs. Number of migrating cells (right panel). Histograms represent the average of at least three independent experiments. Error bars represent s.e.m. (**Po0.01, ***Po0.001).

Figure 1 ddx5 and ddx17 co-immunoprecipitate with NFAT5. (a) Protein immunoblot (IB) analysis of Myc-NFAT5 and FLAG- ddx17 in input (IN) or after immunoprecipitation (IP) with either Myc (left panel) or FLAG (right panel) antibodies. MDA-MB-231 cells were transfected with Myc-NFAT5 and/or FLAG-ddx17 expression vectors, as indicated. (b) Protein immunoblot (IB) analysis of endogenous ddx5 or HA-ddx5 and Myc-NFAT5 after IP with ddx5 and IgG (left panel) or HA (right panel) antibodies. MDA-MB-231 cells were transfected with Myc-NFAT5 expression vector (left panel) or with Myc-NFAT5 and/or HA-ddx5 expression vectors (right panel), as indicated. (c) Protein immunoblot (IB) analysis of endogenous NFAT5 after IP with NFAT5 antibody after transfection of MDA-MB-231 cells either with a control (siCTRL) or a siRNA targeting NFAT5 (siN5). Protein IB of ddx5 in the same conditions (left panel). (d) Detection of endogenous NFAT5-ddx5 and NFAT5-ddx17 complexes by in situ PLA. Fixed MDA-MB-231 cells were incubated either with antibodies against NFAT5, ddx5 and/or dd17 as indicated. (e) Quantification of the number of PLA signals per cell as distinct fluorescent red spots in the experimental conditions described in panel D (n450 cells). (f) Quantification of the number of PLA signals per cell. After transfection with a control siRNA or siRNAs targeting either NFAT5 (siN5) or both ddx5 and ddx17 (siddx5/17), fixed MDA-MB-231 cells were incubated either with antibodies against NFAT5 and ddx5 or antibodies against NFAT5 and ddx17 (n450 cells). (g) NFAT5-luciferase activity assay performed with MDA-MB-231 cells transfected with NFAT5-luc reporter gene, and with NFAT5, ddx5 and/or ddx17 expression vectors. Experiments are representative of at least three independent experiments. Error bars represent s.e.m. (**Po0.01, ***Po0.001).

 { Oncogene

 Regulation of NFAT5 by ddx5/ddx17 S Germann et al 4540 et al., 2006). We have recently reported that lipocalin 2 contain exons 4 and 5 (E3/E6, Figure 4d). This result was (LCN2), which also plays a role in breast cancer further supported after sequencing the PCR products progression (Yang et al., 2009; Leng et al., 2011), is (not shown) and by quantifying NFAT5 splicing variants regulated by NFAT transcriptional factors (Fougere et al., by RT-Quantitative (q)PCR. As expected from PCR 2010). As expected, the depletion of NFAT5 decreased the results (Figure 4d), ddx5/ddx17 depletion increased the S100A4 and LCN2 mRNA levels (Figure 3a). In addition, splicing variants lacking exons 4 and 5 (E3/E6, chromatin immunoprecipitation (ChIP) assay using Figure 4e), whereas decreasing the level of the splicing NFAT5 antibody indicated that NFAT5 was bound to variants containing exon 5 (E5/E6, Figure 4e). The level the S100A4 and LCN2 promoters (Figure 3b, Supple- of the splicing variants containing exon 4 but not exon 5 mentary Figure S3), as expected (Chen et al., 2009). (E4/E6) was not affected in HeLa and MCF-7 cells and We next tested the role of ddx5 and ddx17 on the increased in the MDA-MB-231 cells (E4/E6, Figure 4e). regulation of S100A4 and LCN2 genes. The depletion of These results indicated that ddx5/ddx17 depletion both ddx5 and ddx17 decreased the S100A4 and LCN2 induced the skipping of exon 5 and this was confirmed mRNA levels (Figure 3c). Furthermore, the sub-optimal by using primers that allowed to specifically quantify depletion of NFAT5 together with ddx5 and ddx17 each splicing variant containing or not exon 5. Indeed, showed an additive effect (Figure 3d, Supplementary the level of exon 5 containing transcripts decreased after Figure S1). To test whether the observed effects were ddx5/ddx17 depletion when compared with the level of direct transcriptional effects, ChIP assay was performed transcripts that did not contain exon 5, either in the after MDA-MB-231 transfection with either HA-ddx5 presence or in the absence of exon 4 (Figure 4f). or FLAG-ddx17 expression vectors. Both ddx5 and These data suggested that ddx5/ddx17 depletion ddx17 specifically bound to the S100A4 and LCN2 impacts on alternative splicing of the human NFAT5 promoters and, importantly, NFAT5 depletion reduced exon 5. This possibility was first confirmed by using their recruitment (Figures 3e and f). Together with the another set of siRNAs targeting the UTRs of ddx5 or interaction and transactivation studies (Figure 1), these ddx17 (siUTRddx5/17) that indeed increased the level results demonstrate that ddx5 and ddx17 are bona fide of NFAT5 splicing variants that did not contain exon 5 transcriptional coactivators of the NFAT5 pro-migra- in HeLa cells (E3/E6 vs E5/E6 and E4/E6 vs E5/E6, tory transcriptional factor. Figure 5a). In addition, cell co-transfection with wild- type ddx5 or ddx17 cDNAs that do not contain UTR sequences and that are therefore not affected by Ddx5 and ddx17 regulate alternative splicing of the siUTRddx5/17, partially rescued the siUTRddx5/17 NFAT5 gene effect on NFAT5 splicing (Figure 5a). Importantly, We tested whether the effects of ddx5/ddx17 depletion mutated ddx5 and ddx17 cDNAs encoding proteins on cell migration and gene expression might be due to a lacking the helicase activity did not rescue the splicing decrease in NFAT5 protein level. Remarkably, ddx5/ phenotype (Figure 5a). Furthermore, ddx5 and ddx17 ddx17 depletion in MDA-MB-231 cells did not decrease depletion resulted in a strong increase in NFAT5 NFAT5 protein level, but instead strongly increased it protein expression level in HeLa cells (Figure 5b), as (left panel, Figure 4a). Similar results were obtained in observed in MDA-MB-231 and MCF-7 cells the human MCF-7 breast cancer cells (right panel, (Figure 4a); this was rescued by ddx5 or ddx17 over- Figure 4a) and in the human HeLa cells (see below). We expression (Figure 5b). demonstrated that the protein detected by immunoblot Going a step forward, we next tested whether ddx5 with the NFAT5 antibody after ddx5/ddx17 depletion and/or ddx17 immunoprecipate the NFAT5 pre- was indeed NFAT5, by co-transfecting MDA-MB-231 mRNA. HeLa cells were transfected either with HA- cells with an siRNA targeting NFAT5 (Figure 4b). ddx5 or FLAG-ddx17 expression vectors or with the Therefore, the effects described above of ddx5/ddx17 corresponding empty expression vectors as controls that depletion on migration and NFAT5 target genes are did not affect NFAT5 pre-mRNA level (IN, Figure 5c). clearly not explained by a decrease in NFAT5 protein After formaldehyde-mediated crosslinking of HeLa level. In fact, the inhibition of NFAT5 target genes cells, HA and FLAG antibodies immunoprecipitated expression resulting from ddx5/ddx17 depletion (Fig- NFAT5 pre-mRNA in the HA-ddx5- and FLAG- ure 3) despite the increase in NFAT5 protein level in ddx17-transfected cells, respectively, but not in the these conditions (Figure 4a) is probably due to the control cells (Figure 5c). Similar results were obtained requirement of ddx5 and ddx17 in mediating NFAT5 using inducible stable MCF-7 cells expressing HA-ddx5 transcriptional effects (Figures 1 and 3). or FLAG-ddx17 proteins (Figure 5d). We next analyzed the mechanisms by which ddx5 and To further test the potential role of ddx5 and ddx17 ddx17 control NFAT5 expression level. Remarkably, the in NFAT5 alternative splicing, a minigene containing depletion of ddx5 and ddx17 did not affect the total the human NFAT5 exon 5 and about 200 nucleotides NFAT5 mRNA level in any of the tested cell lines of the surrounding introns was generated (Figure 5e). As (Figure 4c). However, RT–PCR using primers in exons 3 expected, ddx5/ddx17 depletion in HeLa cells induced and 6 that flank known alternative exons 4 and 5 (Dalski the skipping of the minigene exon 5 (Figure 5e). et al., 2002; Maouyo et al., 2002) revealed that ddx5/ Furthermore, the over-expression of wild-type but not ddx17 depletion in MDA-MB-231, HeLa and MCF-7 mutated ddx5 and ddx17 proteins in HeLa cells cells favored the shorter NFAT5 product that did not antagonized the splicing effect of ddx5/ddx17 depletion

Oncogene {€ Regulation of NFAT5 by ddx5/ddx17 S Germann et al 4541

S100A4 LCN2 S100A4 LCN2 IP 0.2 1 N5 IgG IN *** *** * * 100 0.1

0.5 S100A4 of siN5 % INPUT

Fold effect 100

LCN2 IP IgG IP IgG siN5 - + - + N5 N5

S100A4 LCN2 S100A4 LCN2 1 ** * 1 * * 0.5 0.5 * *** Fold effect Fold effect of siddx5/17

siddx5/17 - + - + siddx5/17 -+-+ -+-+ siN5 - - + + - - + +

IP IP IP IP HA INHA IN FLAG IN FLAG IN

S100A4 S100A4 100 100

LCN2 LCN2 100 100 HA-ddx5 - - + + FLAG-ddx17 -+- +

IP HA IP FLAG 0.3 ** ** S100A4 1 * 0.2 LCN2 ** 0.5 0.1 % INPUT % INPUT

HA-ddx5 - +-+ FLAG-ddx17 -++-

IP FLAG IP HA S100A4 LCN2 1 1 * *

0.5 0.5 *** of siN5

of siN5 *** Fold effect Fold effect

siN5 -++- siN5 -++- Figure 3 ddx5 and ddx17 regulate NFAT5 target genes. (a) RT–qPCR analysis of S100A4 and LCN2 mRNA levels after transfection of MDA-MB-231 cells with a control siRNA or siN5. (b) PCR and qPCR analysis with primers targeting S100A4 and LCN2 promoters after immunoprecipitation of NFAT5 (IP N5)-containing chromatin in MDA-MB-231 cells and compared with control IgG. (c) RT–qPCR analysis of S100A4 and LCN2 mRNA levels after transfection of MDA-MB-231 cells with a control siRNA or siddx5/17. (d) RT–qPCR analysis of S100A4 and LCN2 mRNA levels after transfection of MDA-MB-231 cells with a control siRNA, siN5 and/or siddx5/17. (e) PCR and qPCR analysis with primers targeting S100A4 and LCN2 promoters after immunoprecipitation of HA-ddx5 (IP HA) or FLAG-ddx17 (IP FLAG)-containing chromatin in MDA-MB-231 cells transfected with HA-ddx5 or FLAG- ddx17 expression vectors compared with MDA-MB-231 cells transfected with empty expression vectors. PCR gels are representative of at least three independent experiments. (f) qPCR analysis with primers targeting S100A4 and LCN2 promoters after the immunoprecipitation of HA-ddx5 (IP HA) or FLAG-ddx17 (IP FLAG)-containing chromatin in MDA-MB-231 cells transfected with HA-ddx5 or FLAG-ddx17 expression vectors and siN5 compared with MDA-MB-231 cells transfected with HA-ddx5 or FLAG-ddx17 expression vectors and a control siRNA. Histograms represent the average of at least three independent experiments. Error bars represent s.e.m. (*Po0.05, **Po0.01, ***Po0.001).

 { Oncogene Regulation of NFAT5 by ddx5/ddx17 S Germann et al 4542 MDA-MB-231 MCF-7 260 260 NFAT5

NFAT5 160 160 160 NFAT5 OE actin actin 40 40 actin siN5 40 ddx5 60 siCTRL siCTRL ddx5 ddx17 80 60 siddx5/17 80 ddx17 siddx5/17 - + siddx5/17 - +

2

1

MDA- Fold effect

of siddx5/17 MB-231 HeLa MCF-7 300 MDA-MB-231 HeLa MCF-7 3 4 5 6 200 3 4 6 and 3 5 6 ddx5 100 3 6 ddx17 siddx5/17 -++ - + - actin siddx5/17 -+ -+ -+

E5/E6 E3/E6 E4/E6 E3/E4/E5/E6 vs. E3/E4/E6 3 1 * E3/E5/E6 vs. E3/E6

2 *

0.5 1 ** * *** *** * *** *** *** *** Fold effect of siddx5/17 Fold effect of siddx5/17 MDA- HeLa MCF-7 MDA- HeLa MCF-7 MB-231 MB-231 Figure 4 ddx5 and ddx17 control the expression level of NFAT5-splicing variants. (a) WB analysis of NFAT5, ddx5, ddx17 and actin, 48 h after transfection of human MDA-MB-231 and MCF-7 cells with a control siRNA or a siRNA targeting both ddx5 and ddx17 (siddx5/17). OE, over-exposed. (b) WB analysis of NFAT5 and actin, 48 h after transfection of human MDA-MB-231 cells with a control siRNA (siCTRL) or siddx5/17 and/or a siRNA targeting NFAT5 (siN5). (c) RT–qPCR of total NFAT5 transcripts (upper panel) and WB analysis (lower panel), 48 h after transfection of MDA-MB-231, HeLa and MCF-7 cells with a control siRNA or siddx5/17. (d) RT–PCR analysis of NFAT5 transcripts using a forward primer in exon 3 and a reverse primer in exon 6 in the same experimental conditions as in panel C. (e) RT–qPCR analysis of NFAT5 transcripts, in the same experimental conditions as in panel C, using primers to amplify transcripts containing exon 5 (E5/E6), neither exon 5 nor exon 4 (E3/E6) or exon 4 alone (E4/E6). (f) RT– qPCR analysis of NFAT5 transcripts, in the same experimental conditions as in panel C, using primers to amplify transcripts containing exons 3, 4, 5 and 6 (E3/E4/E5/E6), transcripts containing exons 3, 4 and 6 (E3/E4/E6), transcripts containing exons 3, 5 and 6 (E3/E5/E6), or transcripts containing exons 3 and 6 only (E3/E6). Histograms represent the average of at least three independent experiments and represent the fold effect of the siRNA targeting ddx5 and ddx17 compared with a control siRNA. Error bars represent s.e.m. (*Po0.05, **Po0.01, ***Po0.001).

(Figure 5f). Similarly, the over-expression of wild-type suggested that NFAT5 transcripts containing this exon ddx5 or ddx17 in MDA-MB-231 cells favored exon 5 might be unproductive (that is, not-translated) tran- inclusion in the minigene assay, whereas the over- scripts and regulated by the NMD pathway. To test this expression of mutated ddx5 or ddx17 form induced hypothesis, MDA-MB-231 and HeLa cells were treated exon 5 skipping (Figure 5g). Collectively, these results for 4 and 6 h with the translation inhibitor cyclohex- demonstrated that ddx5 and ddx17 favor NFAT5 exon 5 imide, which inhibits mRNA degradation by the NMD inclusion in an RNA helicase activity-dependent manner. pathway that requires on-going translation. As shown on Figure 6b, cycloheximide treatment increased the level of the NFAT5 splicing variants containing exon 5 NFAT5 protein level is regulated by splicing coupled to (E5/E6, Figure 6b). Therefore, the ratio of transcripts NMD containing exon 5 to transcripts that do not contain On analyzing NFAT5 sequence, we observed that exon 5 was increased by cycloheximide treatment NFAT5 exon 5 contains two in-frame stop codons in both MDA-MB-231 and HeLa cells (Figure 6c). that are conserved during evolution (Figure 6a). This Similarly, cell transfection with a siRNA targeting

Oncogene  {‚

 Regulation of NFAT5 by ddx5/ddx17 S Germann et al 4543 UPF1, a major actor of the NMD pathway, increased possibly contributing to tumor invasion (Garrett et al., by B20 to 50% the ratio of transcripts containing exon 2006; Li and Bresnick, 2006). Therefore, S100A4 can 5 to transcripts that do not contain exon 5 in MDA- promote metastasis and crossing mice that overexpress MB-231 cells (MDA-MB-231, Figure 6d). A stronger S100A4 in the mammary epithelium with mouse models increase by B100% was observed in HeLa cells (HeLa, of metastasis dramatically increases the incidence of Figure 6d). Therefore, the data described on Figures 5 metastasis (Davies et al., 1996). Finally, S100A4 and 6 indicated that the regulation by ddx5 and ddx17 expression is associated with a metastatic phenotype in of NFAT5 exon 5 inclusion coupled with NMD multiple types of carcinoma, including breast cancer provides a means for regulating NFAT5 protein (Garrett et al., 2006). Similarly, elevated levels of LCN2 expression level. expression have been reported in various cancers, To further test this model, siRNAs targeting either including breast cancer. Remarkably, there is a strong all NFAT5 splicing variants (siN5 targeting sequences association between LCN2 and metastasis (Yang et al., in exons 7 and 9) or specifically the splicing variants 2009). LCN2 is a member of the lipocalin family containing human exon 5 (siE5) were generated. As composed of small extracellular proteins that transport expected, cell transfection with siN5 or siE5 decreased and present ligands to cell surface receptors, and form the global level of all NFAT5 transcripts (Total, macromolecular complexes, thus playing important Figure 6e) and the level of the splicing variants roles in cell regulation, proliferation and differentiation containing E5 (E5/E6, Figure 6e). As expected, siE5 (Flower, 1994). LCN2 can enhance tumor growth and did not affect the level of the transcripts that do not metastasis by protecting matrix metalloproteinase-9 contain exon 5 in contrast to siN5 (E3/E6, Figure 6e), from degradation and by increasing angiogenesis and/ which demonstrated the specificity of the siE5 siRNA. or by promoting the epithelial to mesenchymal transi- More importantly, siN5, but not siE5, strongly de- tion, which is associated with breast cancer progression creased the NFAT5 protein level (Figure 6f). This result (Leng et al., 2009; Yang et al., 2009). Therefore, this indicated that NFAT5 transcripts containing exon 5 did study identified a novel gene regulatory pathway not significantly contribute to NFAT5 protein synthesis. (NFAT5 stimulation of S100A4 and LCN2 genes) that This result was confirmed by using an other siRNA is coactivated by ddx5/ddx17 and that, together with targeting NFAT5 exon 2 (siE2) that has also been previously reported ddx5 and ddx17 target genes like shown to be alternatively spliced (Dalski et al., 2002). cyclin D1 and Snail (Shin et al., 2007; Yang et al., 2007; Indeed, although siE2 had a similar effect to siE5 on the Carter et al., 2010), strengthens the role of ddx5 and global NFAT5 mRNA level (Total, Figure 6e), siE2 but ddx17 in tumor progression. not siE5 decreased NFAT5 protein level (Figure 6g). We also demonstrated that ddx5 and ddx17 regulate Collectively, these results demonstrated that ddx5 and NFAT5 pre-mRNA splicing in addition to acting as ddx17 decreased NFAT5 protein level by favoring NFAT5 transcriptional coactivators. Indeed, ddx5 and exon 5 inclusion, which resulted in the synthesis of ddx17 splicing activity, relying on their helicase activity, unproductive (that is, not translated) mRNAs. favored the inclusion of the human NFAT5 exon 5 (Figures 4 and 5). To the best of our knowledge, this is the first demonstration that a transcriptional coregula- tor modulating the activity of a transcriptional factor Discussion can also control its expression at the splicing level. It must be underlined that ddx5, that coactivates p53, has Ddx5 and ddx17 are two highly related RNA helicases been shown to selectively regulate the expression of the that have been involved in several steps of the gene delta133p53 isoform (Moore et al., 2010). However, this expression process (Fuller-Pace and Ali, 2008; Jan- isoform is not generated by alternative splicing but by knecht, 2010). In particular, ddx5 and ddx17 act as transcription from an internal promoter in intron 4 of transcriptional coregulators of transcription factors like the p53 gene. Remarkably, the effect of ddx5 and ddx17 estrogen receptor alpha, p53 and beta-catenin that have on NFAT5 splicing has a very important role in NFAT5 important roles in cancer initiation and progression. protein expression level. Indeed, ddx5 and ddx17 Although the role of ddx5 and ddx17 in cancer cell favored the inclusion of the NFAT5 exon 5, which proliferation has been reported, our data identified generated unproductive transcripts that were not a role for ddx5/ddx17 in cell migration (Figure 2). translated (Figure 6). Interestingly NFAT5 exon 5 is This new function of ddx5 and ddx17 is likely to be one of the 111 ‘ultraconserved elements’ identified in mediated, at least in part, by the regulation of the human protein-coding genes (Bejerano et al., 2004; expression of the S100A4 and LCN2 genes that McGlincy and Smith, 2008). It has been reported that are involved in tumor progression and that are target many ultraconserved elements are associated with exons genes of the NFAT5 transcription factor (Figures 1 inducing NMD ((Bejerano et al., 2004; McGlincy and and 3), which itself is known to favor cell migration Smith, 2008). In this context, the ddx5/ddx17 effect on (Jauliac et al., 2002). NFAT5 expression was observed in different human cell Indeed, it has been shown that, in breast cancer, lines (Figure 4) as well as in a mouse cell line intracellular S100A4 can induce cell motility in part (Supplementary Figure S4). Collectively, these observa- through its ability to interact with -IIA and tions strengthen the key role of exon 5 in NFAT5 gene extracellular S100A4 can stimulate MMP-13 activity, expression regulation. However, we cannot exclude that,

 {ƒ Oncogene

 Regulation of NFAT5 by ddx5/ddx17 S Germann et al 4544 in some cellular contexts, the NFAT5 mRNA bearing volved in NFAT5 protein level regulation in hyper- exon 5 may escape NMD. However, many reports have osmotic conditions. Similarly, NFAT5 protein level is indicated that NFAT5 protein level is tightly regulated, regulated during development and NFAT5 has a role in particularly during differentiation and in response to muscle differentiation (O’Connor et al., 2007). As ddx5 hyper-osmotic stresses (Woo et al., 2000; Ferraris et al., and ddx17 also have a role in muscle differentiation 2002). It has been shown that NFAT5 protein level is (Caretti et al., 2006), it will be interesting to test the regulated by several mechanisms involving protein interplay between NFAT5 and ddx5/ddx17 in muscle stability and miRNA-mediated mRNA silencing (Dahl development. et al., 2001; Irarrazabal et al., 2010; Levy et al., 2010). In conclusion, our data indicate that, on one hand, We now report that NFAT5 protein level can also be ddx5 and ddx17 transcriptional activity enhances regulated by the coupling of splicing to the NMD NFAT5-mediated effects and, on the other hand, pathway (Figures 4, 5 and 6). It will be interesting to test ddx5- and ddx17-splicing activity decreases NFAT5 whether this ddx5/ddx17-dependent mechanism is in- protein level. These dual functions of ddx5 and ddx17

1 2 3 4 5 6 * ddx5 60 HeLa 80 ddx17 HeLa * 260 8 * 6 NFAT5 4 160

2 actin 40 E3/E6 vs. E5/E6 * 4 ddx5 60

ddx17 80 2 siUTRddx5/17 - +++ - -+-

E4/E6 vs. E5/E6 ddx5 WT ddx17 WT ---+ siUTRddx5/17 -+++ + + ddx5 WT - - + - - - ddx5 MUT ---+ -- ddx17 WT - - - - + - ddx17 MUT - - - - - +

IP IP IP IP MCF-7 HA-ddx5 HA IN HA IN FLAG IN FLAG IN ddx5 200 60 E5 100 MCF-7 FLAG-ddx17 80 HA-ddx5 - --+ + FLAG-ddx17 - + + ddx17 IP HA IP FLAG dox - + 20 15 20 10 15

% INPUT 10 5 % INPUT 5 HA-ddx5 - + - - FLAG-ddx17 --- + rIgG HA mIgG FLAG

SV40 α-globin fibronectin fibronectin α-globin NdeI HeLa siCTRL siddx5/17 ex1 ex2 3 ex3 300 +E5 -E5 200 ex5

300 +E5 -E5 200 +E5 300 siUTRddx5/17 -+++ ++ -E5 200 ddx5 WT - - + - - - ddx5 MUT -- -+ ------+ - CTRL ddx17 WT ddx5 WT ddx5 MUTddx17 WT ddx17 MUT - - - - - + ddx17 MUT

Oncogene {€z Regulation of NFAT5 by ddx5/ddx17 S Germann et al 4545

6 E5/E6 E3/E4/E5/E6 vs. E3/E4/E6 *** * E3/E6 10 E3/E5/E6 vs. E3/E6 4 E4/E6 * 8 * * 6 Relative 2 *

expression level 4 Relative * * 2 Cyclo 4h6h 4h 6h expression level

MDA-MB-231 HeLa MDA-MB-231 HeLa

1.5 siN5 siCTRL siUPF1 siCTRL siUPF1 siE2 1 siE5 MDA-MB-231 HeLa level 0.5 E3/E4/E5/E6 vs. E3/E4/E6

E3/E5/E6 vs. E3/E6 Relative expression 150 Total E3/E6 E1/E2 E5/E6

100

* level (%) 50 160 NFAT5 ** NFAT5 160 Relative expression actin 40 actin 40 MDA-MB-231 HeLa siE5 siN5 siE2 siE5 siN5 siCTRL siCTRL Figure 6 Regulation of NFAT5 protein level by coupling alternative splicing and the NMD pathway. (a) Conservation during evolution of the human NFAT5 exon 5 sequence bearing two in-frame stop codons (*). (b) RT–qPCR analysis of splicing variants containing exon 5 (E5/E6), none of exon 4 and exon 5 (E3/E6), or exon 4 alone (E4/E6) after 4 and 6 h of cycloheximide treatment (10 mg/ml) in MDA-MB-231 and HeLa cells. (c) RT–qPCR analysis of NFAT5 transcripts after 6 h of cycloheximide treatment using primers to amplify transcripts containing exons 3, 4, 5 and 6 (E3/E4/E5/E6), transcripts containing exons 3, 4 and 6 (E3/E4/E6), transcripts containing exons 3, 5,and 6 (E3/E5/E6), or transcripts containing exons 3 and 6 (E3/E6). (d) RT–PCR using UPF1 specific primers (upper panel) and RT–qPCR (lower panel) as described in panel C after transfection of MDA-MB-231 and HeLa cells with a control siRNA (siCTRL) or a siRNA targeting UPF1 (siUPF1). Data are represented as % of increase measured in three independent experiments. (e) RT–qPCR analysis of all NFAT5 mRNAs (Total), splicing variants containing none of exon 4 and exon 5 (E3/E6), splicing variants containing exon 2 (E1/E2) or exon 5 (E5/E6) after transfection of MDA-MB-231 cells with a control siRNA, siN5 targeting all NFAT5 splicing variants, siE2 targeting the NFAT5 splicing variants containing exon 2 or siE5 targeting the NFAT5 splicing variants containing exon 5. (f) WB analysis of NFAT5 and actin 48 h after transfection of MDA-MB-231 cells with a control siRNA (siCTRL), siN5 or siE5. (g) WB analysis of NFAT5 and actin 48 h after transfection of MDA-MB-231 cells with a control siRNA (siCTRL), siN5, siE2 or siE5. WBs and PCR gels are representative of at least three independent experiments. Histograms represent the average of at least three independent experiments. Error bars represent s.e.m. (*Po0.05, **Po0.01, ***Po0.001).

Figure 5 Regulation of NFAT5 alternative splicing by ddx5 and ddx17. (a) WB analysis of ddx5 and ddx17 after transfection of HeLa cells with a control siRNA or siRNAs targeting ddx5 or ddx17 UTRs (siUTRddx5/17) together with control expression vectors or wild- type ddx5 (ddx5 WT), wild-type ddx17 (ddx17 WT), mutated ddx5 (ddx5 MUT), or mutated ddx17 (ddx17 MUT) expression vectors (upper panel). Tagged ddx5 protein is marked by an asterisk. RT–qPCR analysis of NFAT5 transcripts, in the same experimental conditions as described above, using primers to amplify transcripts containing exon 5 (E5/E6), none of exon 5 and exon 4 (E3/E6) or exon 4 alone (E4/E6). Histograms represent the average of the E3/E6 to E5/E6 ratio and the E4/E6 to E5/E6 ratio obtained in three independent experiments. (b) WB analysis of NFAT5, ddx5, ddx17 and actin, 48 h after transfection of HeLa cells with a control siRNA or siRNAs targeting ddx5 or ddx17 UTRs (siUTRddx5/17) together with control expression vectors or wild-type ddx5 or ddx17 expression vectors. (c) RT–PCR and RT–qPCR analysis of NFAT5 pre-mRNAs precipitated using HA or FLAG antibodies after transfection of HeLa cells with HA-ddx5, FLAG-ddx17 or empty expression vectors. Histograms correspond to the average level of precipitated NFAT5 pre-mRNA compared with the input obtained in three independent experiments. (d) WB analysis of HA-ddx5 and Flag-ddx17 in the presence or absence of doxycycline using inducible MCF-7/Tet-On stable cell lines. RT–qPCR analysis of NFAT5 pre-mRNAs precipitated using HA, FLAG or control (IgG) antibodies after induction by doxycycline (dox) of HA-ddx5 or FLAG-ddx17 expression in stable MCF-7 cell lines. Histograms correspond to the average level of precipitated NFAT5 pre-mRNA compared with the input obtained in three independent experiments. (e) Schematic representation of the NFAT5 exon 5 minigene (left panel). RT–PCR analysis using primers in the minigene exons indicated by arrows on the left panel, after HeLa cells transfection with a control siRNA (siCTL) or siddx5/ddx17 (right panel). (f) RT–PCR using primers in the minigene exons after transfection of HeLa cells in the same experimental conditions as described in panel A. (g) RT–PCR using primers in the minigene exons after transfection of MDA-MB-231 cells with the minigene together with wild-type or mutated ddx5 or ddx17 expression vectors. WBs and PCR gels are representative of at least three independent experiments. Histograms represent the average of at least three independent experiments. Error bars represent s.e.m.

 {€{ Oncogene  Regulation of NFAT5 by ddx5/ddx17 S Germann et al 4546 probably have an important role that is conserved have different impact on cell behavior and tumor during evolution in the NFAT5 pathway (Supplemen- progression. tary Figure S4). Increasing the level of ddx5/ddx17 would enhance NFAT5-mediated pro-migratory effects and would be antagonized by a decrease in NFAT5 Materials and methods protein level mediated by ddx5-and ddx17-splicing Cell culture, plasmid constructions and stable cell lines activity. Meanwhile, decreasing the level of ddx5/ Cells were maintained in DMEM supplemented with 10% fetal ddx17 would increase NFAT5 protein level but repress bovine serum, 1% glutamine, 1% penicillin and streptomycin the expression level of ddx5/ddx17-dependent NFAT5 antibiotics and 4,5 g/l glucose (HeLa, MCF-7) or 1 g/l glucose target genes like S100A4 and LCN2. Interestingly, all (MDA-MB-231). HA-ddx5 and FLAG-ddx17 were cloned NFAT5 target genes might not be ddx5/ddx17-depen- in pTRE2-hyg vector to generate inducible MCF-7/Tet-On dant genes. Indeed, among the tested NFAT5 target stable cell lines (Clontech, Saint-Germain-en-Laye, France). genes, we observed that the SMIT (sodium/myo-inositol Resistant clones were selected with hygromycin (300 mg/ml, cotransporter) gene, a known NFAT5 target gene was Clontech) and protein expression was checked after doxycy- m downregulated after NFAT5 depletion, but upregulated cline treatment (1 g/ml). NFAT5 exon 5 with 245 and 173 bp after ddx5/17 depletion, likely owing to the increase in of flanking introns was amplified by PCR with primers containing NdeI restriction site: NFAT5_in4S (GGAATTCC NFAT5 protein level (Supplementary Figure S5). Even ATATGGAATTCCCTACCACTTCCAGC) and NFAT5_ though the interplay between ddx5/ddx17 and the in5AS (GGAATTCCATATGGAATTCCTTTATTGCCTCA NFAT5-signaling pathway is likely complex, ddx5 and GC). NdeI-digested PCR product was cloned into pTB ddx17 seem to have a critical role in the regulation of minigene. Primers used for minigene assay are: pTBalpha a subset of NFAT5 target genes involved in cell migra- (CAACTTCAAGCTCCTAAGCCACTGC) and pTBbra tion as ddx5/ddx17 depletion inhibited this process (GGTCACCAGGAAGTTGGTTAAATCA). (Figure 2). The precise and respective role of ddx5 and ddx17 in the NFAT5 pathway will also require further Cell transfection experiments. Indeed, we systematically observed that the Cells were transiently transfected with siRNAs (Table 1) at a co-depletion of ddx5 and ddx17 had a stronger effect on final concentration of 26.6 nM (except in Figure 3d, 10 nM siddx5/17 and 20 nM siN5) using RNAiMax (Invitrogen, NFAT5 pathway and NFAT5 gene expression regulation Carlsbad, CA, USA). Cells were harvested 48 h after transfec- than the depletion of either ddx5 or ddx17 alone (Figure 2; tion. Transfections were performed using Lipofectamine LTX Supplementary Figures S6 and S7). Even though these (Invitrogen) in immunoprecipitation and minigene assays. results suggest that ddx5 and ddx17 have similar activities, HeLa cells were plated and transfected with siRNAs targeting the data obtained could be explained either by ddx5 and ddx5 and ddx17 UTRs and transfected 24 h later with siRNAs ddx17 having full redundant functions when co-expressed and expression vectors using Lipofectamine 2000 (Invitrogen) in cells (meaning that they work together in the same in rescue experiments. complexes) or by their replacement by each other when RNA, RT–PCR and RT–qPCR one or the other is missing. m Finally, it is interesting to integrate the dual impact of Total RNAs were extracted using TRIzol (Invitrogen). 1 lof Glycoblue (Ambion/Applied Biosystems, Austin, TX, USA) ddx5 and ddx17 on the pro-migratory NFAT5 tran- was added before RNA precipitation with isopropanol. RNAs scription factor in the context of the known role of (1 mg) were treated with DNase I (DNAfree, Ambion) 30 min ddx5 and ddx17 in cancer. Indeed, although there are at 37 1C and reverse-transcribed using SuperScript II and increasing evidences that these proteins have an random primers (Invitrogen). Before performing PCR, all RT important role in cancer, their effects are likely to be reaction mixtures were diluted to contain 2.5 ng/ml of initial context-dependent, as recently reviewed (Fuller-Pace RNA. PCRs were performed using 5 ml of the diluted cDNAs and Moore, 2011). Ddx5 and ddx17 have been reported and GoTaq polymerase (Promega, Madison, WI, USA). to have either oncogenic functions or tumor cosupressor qPCR was performed using 2.5 ml of the diluted cDNAs and roles (Fuller-Pace and Moore, 2011). One intriguing SYBR Green I Master mix on a LightCycler (Roche, possibility is that, depending on the cellular context, Mannheim, Germany). Primer sequences are provided in Table 1. The relative amounts of mRNAs were determined ddx5/ddx17-splicing activity or their transcriptional on the basis of the threshold cycle for each PCR product (Ct) activity could dominate each other and this could and were normalized with Actin RNA levels (CCTCCCTGG participate in the dual function of ddx5 and ddx17 in AGAAGAGCTA; CCAGACAGCACTGTGTTGG). Primers cancer. In this context, several reports have indicated used to analyze S100A4 and LCN2 were: S100A4-E2S (GATG that the splicing and transcriptional activities of ddx5 TGATGGTGTCCACCTTC);S100A4-E3AS (GTACTCTTG and ddx17 can be modulated by different post-transla- GAAGTCCACCTC); LCN2-E2S (GGTAGGCCTGGCAGGG tional modifications that depend on cellular signaling AATG); LCN2-E3AS (CTTAATGTTGCCCAGCGTGAAC). pathways (Yang et al., 2006, 2007; Jacobs et al., 2007; Primers used to analyze NFAT5 splicing were: NFAT5-E3S2 Carter et al., 2010; Mooney et al., 2010a, b). In addition, (AATGAGTCAGACAAGCGGTG) and NFAT5-E6AS2 several post-translational modifications of ddx5 or (TGGAAGAGGTGGTAAAGGAG). Primers used to analyze UPF1 were: UPF1-S (ACACCAAGCTCTACCAGGAGG) ddx17 have been shown to be altered in different and UPF1-AS (ACAATGATGACGCCATACCTTG). cancers (Causevic et al., 2001; Yang et al., 2005). Therefore, different cellular signaling pathways acti- Western blot (WB) vated in different cellular contexts may favor one of the Protein extracts were obtained using NP-40 buffer (50 mM molecular functions of ddx5 and ddx17, which could Tris–HCl pH 8, 0.4 M. NaCl, 5 mM EDTA pH 8, 1% NP40,

Oncogene  {€|

 Regulation of NFAT5 by ddx5/ddx17 S Germann et al 4547 Table 1 Sequences of the siRNAs and primers used to quantify NFAT5 mRNAs by qPCR SiRNAs Sequences siCTRL CGUACGCGGAAUACUUCGAdTdT siddx5/17 GGCUAGAUGUGGAAGAUGUdTdT siUTRddx5 AACCGCAACCAUUGACGCCAUdTdT siUTRddx17 UCAUGCAGAUUAGUUAGAAdTdT; CAGCAGACUUAAUU ACAUUdTdT siN5 GAAAGGAGCUGAAGAUAGUdTdT; GUAGGGAUAUUGAA AUUGAdTdT siE2 GUUACACCCAUCACAGAAUdTdT; GAAUUUUCAUAGAG CUGGAdTdT siE5 GACCUGUAGUUCUCUGCUUdTdT; GCAGGGAGUGUCU GCAUUAdTdT

Primers Forward primer (50–30) Reverse primer (50–30)

Total human NFAT5 AGTGGACATTGAAGGCACTAC TTGGAACCAGCAATTCCTATTC E5/E6 GATTTGCCTCTGAAGCAGGG CATAGCCTTGCTGTCGGTGAC E3/E6 AGCTGTTGTTGCTGCTGATGCa CATAGCCTTGCTGTCGGTGAC E4/E6 CTGCATTCTCATGTGCATGATCC GGGAGCTGAAGAAGCATCCTTa E3/E4/E5/E6 AGCTGTTGTTGCTGCTGGGAa GGGAGCTGAAGAAGCATCTATa E3/E5/E6 AGCTGTTGTTGCTGCTGGATTa GGGAGCTGAAGAAGCATCTATa E1/E2 GATGCCCTCGGACTTCATCTC CTGTGATGGGTGTAACTTCAGAG Total mouse NFAT5 GTGGACATTGAAGGTACCAC CTTCAACATCGGCATTCCTC þ mE4 GATTTGCCTCTGAAGCAGGG CATAGCCTTGCTGTCGGTGAC mE4 AGCTGTAGTTGCTGCTGATGCa CATAGCCTTGCTGTCGGTGAC aPrimers overlapping two exons.

0.2% SDS, 1 mM DTT) and protease inhibitors (Sigma 2001). MDA-MB-231 were plated in 10-cm plates and Aldrich, St Louis, MO, USA). 20 mg of proteins were analyzed transfected with firefly luciferase reporter gene (3 mg), renilla by WB. Used antibodies were: NFAT5 (PA1-023, Thermo luciferase control plasmid (100 ng), NFAT5 (4 mg), ddx5 and Fischer Scientific, Waltham, MA, USA), DDX5 (ab10261, ddx17 expression vectors (2 mg). After 24 h, cells were plated in Abcam, Cambridge, MA, USA), DDX17 (ab24601 and 6-well plates in triplicates. Harvested cells were lysed with lysis ab52826 for the in situ PLA, Abcam), c-Myc (9E10, buffer 5X to measure firefly and renilla luciferase activities sc-40, Santa-Cruz Biotechnology, Santa Cruz, CA, USA), using Dual Luciferase Reporter Assay System (Promega). Anti-FLAG M2 (Sigma Aldrich), Anti-HA (3F10, Roche) or Actin (I19, sc-1616, Santa-Cruz Biotechnology) antibodies. ChIP and RNA-ChIP Cells were cross-linked with 1% formaldehyde for 10 min and Co-immunoprecipitation lysed in buffer containing 50 mM Tris–HCl pH 7.5, 150 mM MDA-MB-231 cells were lysed 24 h after transfection in buffer KCl, 5 mM EDTA, 1% NP40, 0.1% SDS, 0.5% sodium containing 50 mM Tris–HCl pH 8, 150 mM NaCl, 1% NP40, deoxycholate, 50 mM sodium fluoride, phosphatase and 1mM PMSF, protease and phosphatase inhibitors (Sigma). protease inhibitors. Chromatin was sheared by sonication Cleared cell extracts were incubated overnight with 4 mgof with Bioruptor (Diagenode, Lie` ge, Belgium) to generate antibodies bound to protein G magnetic beads (Invitrogen). B200-bp DNA fragments and incubated with antibodies and The protein-bound magnetic beads were washed three times magnetic beads overnight at 4 1C. Beads were washed five with PBS1X, resuspended in loading buffer before WB times with buffer containing 50 mM Tris–HCl pH7.5, 150 mM analysis. Inputs correspond to 5% volume of the cleared cell KCl, 1% NP40 and 0.25% sodium deoxycholate, and two lysates. times with TE buffer. Antibody-bound chromatin was reverse- crosslinked overnight at 65 1C and treated with proteinase K In situ PLA (Qiagen, Valencia, CA, USA) before DNA purification on The in situ PLA was conducted using the Duolink II Kit (Olink column (Qiagen). Purified DNA was diluted three times in Bioscience, Uppsala, Sweden) according to the manufacturer’s water and analyzed by PCR using the following primers: instructions. Proximity ligation signals were detected with a hS100A4-pS1 (CAGAGTCCTGCCCTCAAGGAA); hS100A4- Zeiss fluorescence microscope (63 objective) and images pAS1 (TCTCCGTGGTCATTCCCATC); hLCN2-pS1 (CCCA were processed with ImageJ software program. TGCAAGGAGGGAAATC); and LCN2-pAS1 (CCTCATGGG AGGTGGTGTTG). RNA-ChIP was performed using NFAT5- E5S2 (TTGCCTCTGAAGCAGGGAGTG) and NFAT5-I5AS Luciferase assay (ATGCTGGTGGTCCACATTCAA) primers as previously The osmotic stress-responsive luciferase reporter plasmid ORE described (Bittencourt et al., 2008). Luc, which incorporates 132 bp of the human aldose reductase gene enhancer containing three consensus binding sites for NFAT5, was made by PCR amplifying genomic DNA (50- Migration assay TTACATGGAAAAATATCTGGGCTAG-30;50-CTGGTAG Cells were transfected with NFAT5 and b-Gal constructs TGACTCAAGCAC-30) and cloning the resulting PCR frag- (1 mg) or with siRNAs (siNFAT5, siddx5/17, siddx5, siddx17). ment 50 of the minimal SV40 promoter in the luciferase Migration assay was performed as previously described reporter plasmid pGL3 (Promega) (Lopez-Rodriguez et al., (Jauliac et al., 2002).

 {€} Oncogene  Regulation of NFAT5 by ddx5/ddx17 S Germann et al 4548 Conflict of interest Emanuele Buratti for providing pTB minigene construct. We thank Coralie Poulard and Muriel Le Romancer-Cherifi for The authors declare no conflict of interest. their help with PLA. This work was supported by EC (NoE EURASNET), Institut National du Cancer, Association pour la Recherche sur le Cancer, Ligue Nationale Contre le Cancer and Agence Nationale de la Recherche. S Germann, B Acknowledgements Gaudineau and M Fouge` re were supported by Association pour la Recherche sur le Cancer; L Gratadou by Agence We thank S Kato and J Marie for providing wild-type and Nationale de la Recherche; E Zonta by the French Ministry of mutant ddx5 and ddx17 expression vectors. We thank Education; E Dardenne by Ligue Nationale Contre le Cancer.

References

Allemand E, Batsche E, Muchardt C. (2008). Splicing, transcription, Chen M, Sinha M, Luxon BA, Bresnick AR, O’Connor KL. (2009). and chromatin: a menage a trois. Curr Opin Genet Dev 18: 145–151. Integrin alpha6beta4 controls the expression of genes associated Auboeuf D, Batsche E, Dutertre M, Muchardt C, O’Malley BW. with cell motility, invasion, and metastasis, including S100A4/ (2007). Coregulators: transducing signal from transcription to metastasin. J Biol Chem 284: 1484–1494. alternative splicing. Trends Endocrinol Metab 18: 122–129. Chen Y, Schnetz MP, Irarrazabal CE, Shen RF, Williams CK, Burg Auboeuf D, Honig A, Berget SM, O’Malley BW. (2002). Coordinate MB et al. (2007). Proteomic identification of proteins associated regulation of transcription and splicing by steroid receptor with the osmoregulatory transcription factor TonEBP/OREBP: coregulators. Science 298: 416–419. functional effects of Hsp90 and PARP-1. Am J Physiol Renal Ayers M, Symmans WF, Stec J, Damokosh AI, Clark E, Hess K et al. Physiol 292: F981–F992. (2004). Gene expression profiles predict complete pathologic Chuvpilo S, Jankevics E, Tyrsin D, Akimzhanov A, Moroz D, Jha response to neoadjuvant paclitaxel and fluorouracil, doxorubicin, MK et al. (2002). Autoregulation of NFATc1/A expression and cyclophosphamide chemotherapy in breast cancer. J Clin Oncol facilitates effector T cells to escape from rapid apoptosis. Immunity 22: 2284–2293. 16: 881–895. Baksh S, Widlund HR, Frazer-Abel AA, Du J, Fosmire S, Fisher DE Clark EL, Coulson A, Dalgliesh C, Rajan P, Nicol SM, Fleming S et al. (2002). NFATc2-mediated repression of cyclin-dependent et al. (2008). The RNA helicase p68 is a novel androgen receptor kinase 4 expression. Mol Cell 10: 1071–1081. coactivator involved in splicing and is overexpressed in prostate Barbier J, Dutertre M, Bittencourt D, Sanchez G, Gratadou L, de la cancer. Cancer Res 68: 7938–7946. Grange P et al. (2007). Regulation of H-ras splice variant expression Dahl SC, Handler JS, Kwon HM. (2001). Hypertonicity-induced by cross talk between the p53 and nonsense-mediated mRNA decay phosphorylation and nuclear localization of the transcription factor pathways. Mol Cell Biol 27: 7315–7333. TonEBP. Am J Physiol Cell Physiol 280: C248–C253. Bates GJ, Nicol SM, Wilson BJ, Jacobs AM, Bourdon JC, Wardrop J Dalski A, Hebinck A, Winking H, Butzmann U, Schwinger E, Zuhlke et al. (2005). The DEAD box protein p68: a novel transcriptional C. (2002). Complete cDNA sequence, expression, alternative coactivator of the p53 tumour suppressor. Embo J 24: 543–553. splicing, and genomic organization of the mouse Nfat5 gene. Bejerano G, Pheasant M, Makunin I, Stephen S, Kent WJ, Mattick JS Cytogenet Genome Res 97: 179–182. et al. (2004). Ultraconserved elements in the . Science David CJ, Manley JL. (2010). Alternative pre-mRNA splicing 304: 1321–1325. regulation in cancer: pathways and programs unhinged. Genes Bittencourt D, Dutertre M, Sanchez G, Barbier J, Gratadou L, Dev 24: 2343–2364. Auboeuf D. (2008). Cotranscriptional splicing potentiates the Davies MP, Rudland PS, Robertson L, Parry EW, Jolicoeur P, mRNA production from a subset of estradiol-stimulated genes. Barraclough R. (1996). Expression of the calcium-binding protein Mol Cell Biol 28: 5811–5824. S100A4 (p9Ka) in MMTV-neu transgenic mice induces metastasis Blencowe BJ. (2003). Splicing regulation: the cell cycle connection. of mammary tumours. Oncogene 13: 1631–1637. Curr Biol 13: R149–R151. Dutertre M, Gratadou L, Dardenne E, Germann S, Samaan S, Burg MB, Ferraris JD, Dmitrieva NI. (2007). Cellular response to Lidereau R et al. (2010a). Estrogen regulation and physiopathologic hyperosmotic stresses. Physiol Rev 87: 1441–1474. significance of alternative promoters in breast cancer. Cancer Res Camats M, Guil S, Kokolo M, Bach-Elias M. (2008). P68 RNA 70: 3760–3770. helicase (DDX5) alters activity of cis- and trans-acting factors of the Dutertre M, Vagner S, Auboeuf D. (2010b). Alternative splicing and alternative splicing of H-Ras. PLoS One 3: e2926. breast cancer. RNA Biol 7: 403–411. Caretti G, Schiltz RL, Dilworth FJ, Di Padova M, Zhao P, Ogryzko V Ferraris JD, Williams CK, Persaud P, Zhang Z, Chen Y, Burg MB. et al. (2006). The RNA helicases p68/p72 and the noncoding RNA (2002). Activity of the TonEBP/OREBP transactivation domain SRA are coregulators of MyoD and skeletal muscle differentiation. varies directly with extracellular NaCl concentration. Proc Natl Dev Cell 11: 547–560. Acad Sci U S A 99: 739–744. Carter CL, Lin C, Liu CY, Yang L, Liu ZR. (2010). Phosphorylated p68 Flower DR. (1994). The lipocalin protein family: a role in cell RNA helicase activates Snail1 transcription by promoting HDAC1 regulation. FEBS Lett 354: 7–11. dissociation from the Snail1 promoter. Oncogene 29: 5427–5436. Fougere M, Gaudineau B, Barbier J, Guaddachi F, Feugeas JP, Causevic M, Hislop RG, Kernohan NM, Carey FA, Kay RA, Steele Auboeuf D et al. (2010). NFAT3 transcription factor inhibits breast RJ et al. (2001). Overexpression and poly-ubiquitylation of the cancer cell motility by targeting the Lipocalin 2 gene. Oncogene 29: DEAD-box RNA helicase p68 in colorectal tumours. Oncogene 20: 2292–2301. 7734–7743. Friedl P, Wolf K. (2003). Tumour-cell invasion and migration: Chang YF, Chan WK, Imam JS, Wilkinson MF. (2007). Alternatively diversity and escape mechanisms. Nat Rev Cancer 3: 362–374. spliced T-cell receptor transcripts are up-regulated in response to Fuller-Pace FV, Ali S. (2008). The DEAD box RNA helicases p68 disruption of either splicing elements or reading frame. J Biol Chem (Ddx5) and p72 (Ddx17): novel transcriptional co-regulators. 282: 29738–29747. Biochem Soc Trans 36: 609–612. Chen M, Sastry SK, O’Connor KL. (2011). Src kinase pathway is involved Fuller-Pace FV, Moore HC. (2011). RNA helicases p68 and p72: in NFAT5-mediated S100A4 induction by hyperosmotic stress in colon multifunctional proteins with important implications for cancer cancer cells. Am J Physiol Cell Physiol 300: C1155–C1163. development. Future Oncol 7: 239–251.

Oncogene {€~ Regulation of NFAT5 by ddx5/ddx17 S Germann et al 4549 Gardner LB. (2010). Nonsense-mediated RNA decay regulation by Mooney SM, Goel A, D’Assoro AB, Salisbury JL, Janknecht R. cellular stress: implications for tumorigenesis. Mol Cancer Res 8: (2010a). Pleiotropic effects of p300-mediated acetylation on p68 and 295–308. p72 RNA helicase. J Biol Chem 285: 30443–30452. Garrett SC, Varney KM, Weber DJ, Bresnick AR. (2006). S100A4, a Mooney SM, Grande JP, Salisbury JL, Janknecht R. (2010b). mediator of metastasis. J Biol Chem 281: 677–680. Sumoylation of p68 and p72 RNA helicases affects protein stability Guil S, Gattoni R, Carrascal M, Abian J, Stevenin J, Bach-Elias M. and transactivation potential. Biochemistry 49: 1–10. (2003). Roles of hnRNP A1, SR proteins, and p68 helicase in c-H- Moore HC, Jordan LB, Bray SE, Baker L, Quinlan PR, Purdie CA ras alternative splicing regulation. Mol Cell Biol 23: 2927–2941. et al. (2010). The RNA helicase p68 modulates expression and Hanahan D, Weinberg RA. (2000). The hallmarks of cancer. Cell 100: function of the Delta133 isoform(s) of p53, and is inversely 57–70. associated with Delta133p53 expression in breast cancer. Oncogene Helfman DM, Kim EJ, Lukanidin E, Grigorian M. (2005). The 29: 6475–6484. metastasis associated protein S100A4: role in tumour progression Muller MR, Rao A. (2010). NFAT, immunity and cancer: a and metastasis. Br J Cancer 92: 1955–1958. transcription factor comes of age. Nat Rev Immunol 10: 645–656. Honig A, Auboeuf D, Parker MM, O’Malley BW, Berget SM. (2002). Neu-Yilik G, Kulozik AE. (2008). NMD: multitasking between Regulation of alternative splicing by the ATP-dependent DEAD- mRNA surveillance and modulation of gene expression. Adv Genet box RNA helicase p72. Mol Cell Biol 22: 5698–5707. 62: 185–243. Irarrazabal CE, Gallazzini M, Schnetz MP, Kunin M, Simons BL, O’Connor RS, Mills ST, Jones KA, Ho SN, Pavlath GK. (2007). A Williams CK et al. (2010). Phospholipase C-gamma1 is involved in combinatorial role for NFAT5 in both myoblast migration and signaling the activation by high NaCl of the osmoprotective differentiation during skeletal muscle myogenesis. J Cell Sci 120: transcription factor TonEBP/OREBP. Proc Natl Acad Sci U S A 149–159. 107: 906–911. Rossow KL, Janknecht R. (2003). Synergism between p68 RNA Jacobs AM, Nicol SM, Hislop RG, Jaffray EG, Hay RT, Fuller-Pace helicase and the transcriptional coactivators CBP and p300. FV. (2007). SUMO modification of the DEAD box protein p68 Oncogene 22: 151–156. modulates its transcriptional activity and promotes its interaction Shin S, Rossow KL, Grande JP, Janknecht R. (2007). Involvement of with HDAC1. Oncogene 26: 5866–5876. RNA helicases p68 and p72 in colon cancer. Cancer Res 67: 7572–7578. Janknecht R. (2010). Multi-talented DEAD-box proteins and potential Stamm S, Ben-Ari S, Rafalska I, Tang Y, Zhang Z, Toiber D et al. tumor promoters: p68 RNA helicase (DDX5) and its paralog, p72 (2005). Function of alternative splicing. Gene 344: 1–20. RNA helicase (DDX17). Am J Transl Res 2: 223–234. Warner DR, Bhattacherjee V, Yin X, Singh S, Mukhopadhyay P, Jauliac S, Lopez-Rodriguez C, Shaw LM, Brown LF, Rao A, Toker A. Pisano MM et al. (2004). Functional interaction between Smad, (2002). The role of NFAT transcription factors in integrin-mediated CREB binding protein, and p68 RNA helicase. Biochem Biophys carcinoma invasion. Nat Cell Biol 4: 540–544. Res Commun 324: 70–76. Kar A, Fushimi K, Zhou X, Ray P, Shi C, Chen X et al. (2011). Warzecha CC, Jiang P, Amirikian K, Dittmar KA, Lu H, Shen S et al. RNA Helicase p68 (DDX5) regulates tau exon 10 splicing by (2010). An ESRP-regulated splicing programme is abrogated during modulating a stem-loop structure at the 50 splice site. Mol Cell Biol the epithelial-mesenchymal transition. Embo J 29: 3286–3300. 31: 1812–1821. Watanabe M, Yanagisawa J, Kitagawa H, Takeyama K, Ogawa S, Leng X, Ding T, Lin H, Wang Y, Hu L, Hu J et al. (2009). Inhibition Arao Y et al. (2001). A subfamily of RNA-binding DEAD-box of lipocalin 2 impairs breast tumorigenesis and metastasis. Cancer proteins acts as an estrogen receptor alpha coactivator through the Res 69: 8579–8584. N-terminal activation domain (AF-1) with an RNA coactivator, Leng X, Wu Y, Arlinghaus RB. (2011). Relationships of lipocalin 2 with SRA. Embo J 20: 1341–1352. breast tumorigenesis and metastasis. JCellPhysiol226: 309–314. Wilson BJ, Bates GJ, Nicol SM, Gregory DJ, Perkins ND, Fuller-Pace Levy C, Khaled M, Iliopoulos D, Janas MM, Schubert S, Pinner S FV. (2004). The p68 and p72 DEAD box RNA helicases interact et al. (2010). Intronic miR-211 assumes the tumor suppressive with HDAC1 and repress transcription in a promoter-specific function of its host gene in melanoma. Mol Cell 40: 841–849. manner. BMC Mol Biol 5: 11. Lewis BP, Green RE, Brenner SE. (2003). Evidence for the widespread Woo SK, Dahl SC, Handler JS, Kwon HM. (2000). Bidirectional coupling of alternative splicing and nonsense-mediated mRNA regulation of tonicity-responsive enhancer binding protein in response decay in humans. Proc Natl Acad Sci U S A 100: 189–192. to changes in tonicity. Am J Physiol Renal Physiol 278: F1006–F1012. Li ZH, Bresnick AR. (2006). The S100A4 metastasis factor regulates Wortham NC, Ahamed E, Nicol SM, Thomas RS, Periyasamy M, cellular motility via a direct interaction with myosin-IIA. Cancer Jiang J et al. (2009). The DEAD-box protein p72 regulates Res 66: 5173–5180. ERalpha-/oestrogen-dependent transcription and cell growth, and Lopez-Rodriguez C, Aramburu J, Jin L, Rakeman AS, Michino M, is associated with improved survival in ERalpha-positive breast Rao A. (2001). Bridging the NFAT and NF-kappaB families: cancer. Oncogene 28: 4053–4064. NFAT5 dimerization regulates cytokine gene transcription in Yang J, Bielenberg DR, Rodig SJ, Doiron R, Clifton MC, Kung AL response to osmotic stress. Immunity 15: 47–58. et al. (2009). Lipocalin 2 promotes breast cancer progression. Proc Macian F. (2005). NFAT proteins: key regulators of T-cell develop- Natl Acad Sci U S A 106: 3913–3918. ment and function. Nat Rev Immunol 5: 472–484. Yang L, Lin C, Liu ZR. (2005). Phosphorylations of DEAD box p68 Mancini M, Toker A. (2009). NFAT proteins: emerging roles in cancer RNA helicase are associated with cancer development and cell progression. Nat Rev Cancer 9: 810–820. proliferation. Mol Cancer Res 3: 355–363. Maouyo D, Kim JY, Lee SD, Wu Y, Woo SK, Kwon HM. (2002). Yang L, Lin C, Liu ZR. (2006). P68 RNA helicase mediates PDGF- Mouse TonEBP-NFAT5: expression in early development and induced epithelial mesenchymal transition by displacing Axin from alternative splicing. Am J Physiol Renal Physiol 282: F802–F809. beta-catenin. Cell 127: 139–155. McGlincy NJ, Smith CW. (2008). Alternative splicing resulting in Yang L, Lin C, Zhao S, Wang H, Liu ZR. (2007). Phosphorylation of nonsense-mediated mRNA decay: what is the meaning of nonsense? p68 RNA helicase plays a role in platelet-derived growth factor- Trends Biochem Sci 33: 385–393. induced cell proliferation by up-regulating cyclin D1 and c-Myc Metivier R, Penot G, Hubner MR, Reid G, Brand H, Kos M et al. expression. J Biol Chem 282: 16811–16819. (2003). Estrogen receptor-alpha directs ordered, cyclical, and Yoeli-Lerner M, Yiu GK, Rabinovitz I, Erhardt P, Jauliac S, Toker A. combinatorial recruitment of cofactors on a natural target (2005). Akt blocks breast cancer cell motility and invasion through promoter. Cell 115: 751–763. the transcription factor NFAT. Mol Cell 20: 539–550.

Supplementary Information accompanies the paper on the Oncogene website (http://www.nature.com/onc)

 {€ Oncogene Nucleic Acids Research Advance Access published November 25, 2013 Nucleic Acids Research, 2013, 1–11 doi:10.1093/nar/gkt1216

The Ddx5 and Ddx17 RNA helicases are cornerstones in the complex regulatory array of steroid hormone-signaling pathways Samaan Samaan1,2,3,4,Le´ on-Charles Tranchevent2,3,4,5, Etienne Dardenne2,3,4,5, Micaela Polay Espinoza2,3,4,5, Eleonora Zonta1,2,3,4, Sophie Germann2,3,4,5, Lise Gratadou2,3,4,5, Martin Dutertre2,3,4,5 and Didier Auboeuf2,3,4,5,*

1Universite´ de Paris Diderot-Paris 7, F-75013 Paris, France, 2Inserm U1052, F-69008 Lyon, France, 3CNRS UMR5286, F-69008 Lyon, France, 4Centre de Recherche en Cance´ rologie de Lyon, 69008 Lyon, France and 5Universite´ de Lyon 1, F-69100 Villeurbanne, France Downloaded from

Received August 30, 2013; Revised October 25, 2013; Accepted November 4, 2013

ABSTRACT sensitive tissues. Many lines of evidence have implicated steroid hormones as etiologic factors in the origin and http://nar.oxfordjournals.org/ Estrogen and androgen receptors (ER and AR) play progression of various malignancies (1,2). Perturbation key roles in breast and prostate cancers, respect- of the estrogen-signaling pathway is associated with two- ively, where they regulate the transcription of large thirds of breast cancers that express the estrogen receptor arrays of genes. The activities of ER and AR are alpha (ERa), which is considered as a good prognosis controlled by large networks of protein kinases marker. The androgen receptor (AR) is activated by the and transcriptional coregulators, including Ddx5 binding of testosterone or its physiologically active metab- and its highly related paralog Ddx17. The Ddx5 and olite, 5a-dihydrotestosterone (DHT) and is involved in

Ddx17 RNA helicases are also splicing regulators. prostate cancer initiation and metastasis. at Centre Léon Bérard on January 15, 2014 a Here, we report that Ddx5 and Ddx17 are master ER and AR are members of the large superfamily of regulators of the estrogen- and androgen-signaling nuclear receptors and act as ligand-activated transcription factors. The canonical model of steroid receptor action pathways by controlling transcription and splicing implies a ligand-specific conformational change triggering both upstream and downstream of the receptors. its phosphorylation, homodimerization and binding to First, Ddx5 and Ddx17 are required downstream of hormone responsive elements located in promoters or regu- ER and AR for the transcriptional and splicing regu- latory regions of target genes (3). Activation of the estrogen lation of a large number of steroid hormone target and androgen pathways requires the concerted action of a genes. Second, Ddx5 and Ddx17 act upstream of plethora of factors. Some of them are involved in posttrans- ER and AR by controlling the expression, at the lational modifications of the hormone receptors, impacting splicing level, of several key regulators of ER and for example on their subcellular localization or stability (4– AR activities. Of particular interest, we demonstrate 6). Other factors, the so-called transcriptional coregulators, that Ddx5 and Ddx17 control alternative splicing of are recruited by hormone receptors on target promoters to a the GSK3b kinase, which impacts on both ER and AR mediate their effects on transcription (7,8). For both ER and AR, dozens of coregulators, including coactivators and protein stability. We also provide a freely available corepressors, have been identified. However, most of them online resource which gives information regarding have been studied in the context of a few target genes, and splicing variants of genes involved in the estrogen- therefore it is currently unknown whether they contribute and androgen-signaling pathways. to the hundreds of gene regulations induced by hormones. Among these, the DEAD-box RNA helicase Ddx5 (p68) and its highly related paralog, Ddx17 (p72), are tran- INTRODUCTION scriptional coregulators of ERa and AR (9,10). Ddx5 and The sex steroid hormones, estrogen and testosterone, Ddx17 interact directly with ERa and AR and were shown influence normal physiology, reproduction and behavior. in a few cases to be recruited to target promoters (11,12) Their biological functions are mediated through cognate where they might modulate RNA polymerase II recruit- nuclear receptors that govern gene expression in hormone- ment. However, the extent of ERa and AR endogenous

*To whom correspondence should be addressed. Tel: +334 26 55 67 46; Fax: +334 78 78 27; Email: [email protected]

ß The Author(s) 2013. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.

 {€€ 2 Nucleic Acids Research, 2013 target genes on which Ddx5/Ddx17 act as transcriptional added before RNA precipitation. Nuclear fractionation coregulators is not known. In addition to being transcrip- was performed as previously described (20). Reverse tran- tional coregulators, Ddx5 and Ddx17 are bona fide com- scription (RT) was performed with 1–2 mg of total RNA ponents of the splicing machinery, the spliceosome (13), using M-MLV Reverse Transcriptase (LifeTechnologies) and play a role in the regulation of alternative splicing that and random primers. The RT reactions were diluted and leads to the production from the same gene of several used either for PCR analysis using GoTaq Flexi DNA splicing variants coding for different protein isoforms Polymerase (Promega) or in qPCR using SYBR Green I with different and sometimes opposite biological activities Master Mix (Roche) on a Roche LightCycler 480 II. (14–17). Alternative splicing is the rule, not the exception, Primer sequences are provided in Supplementary Table as 90% of human genes produce several splicing variants. S3. The relative RNA levels were determined on the Alternative splicing is the main mechanism increasing the basis of the threshold cycle (Ct) for each qPCR product diversity of the proteome coded by a limited number of and normalized to 18S ribosomal RNA levels. genes (18). In this context, the Ddx5 and Ddx17 multi- functional proteins could coordinate transcription and Affimetrix exon array splicing allowing the production of the proper isoform One microgram of total RNA was processed with the from hormone target genes as previously suggested by GeneChip WT Sense Target Labeling kit and hybridized Downloaded from using minigene reporter assay (19). However, whether to GeneChip Human Exon 1.0 ST arrays. Affymetrix Ddx5 and Ddx17 regulate splicing of endogenous exon-array data were normalized with quantile normaliza- hormone target genes is not known. tion. Antigenomic probes were used to perform the back- In this work, we report using large-scale approaches for ground correction. Only probes targeting exons annotated

first time that Ddx5 and Ddx17 are master regulators of from full-length cDNA were retained for analysis. Cross http://nar.oxfordjournals.org/ the estrogen and androgen-signaling pathways. Indeed, hybridizing probes and probes with lower signals intensity these proteins are not only required for regulating the than anti-genomic background probes showing the same expression of a large number of endogenous estrogen- GC content were removed. Only probes with a DABG and androgen-target genes both at the transcriptional P-value 0.05 in at least half of the arrays were con- and splicing level but remarkably, they are also acting sidered for further statistical analysis. Arrays were per- upstream of the estrogen and androgen receptors by formed in four independent replicates. The strategies controlling the expression, at the splicing level, of several adopted to identify E2-, DHT-regulated and Ddx5/17- key regulators of the hormone-signaling pathways. dependent genes are described in Supplementary Figure S1. Briefly, the median intensity of all constitutive at Centre Léon Bérard on January 15, 2014 exonic probes was calculated for each gene in each MATERIALS AND METHODS sample, and the experimental samples and control Cell culture and stable cell lines groups were compared using a Student’s paired t-test. Paired statistical analyses were performed using the MCF-7 cells were grown in DMEM and LNCaP cells in Student’s paired t test on the splicing index (SI) to RPMI-1640 (Gibco/LifeTechnologies). Both mediums analyze the Exon Array data at the exon level (SI > 1.5, were supplemented with 10% fetal bovine serum, 1% P < 0.05). The SI corresponds to a comparison of gene- glutamine and 1% penicillin/streptomycin. Cells were normalized exon intensity values between the two maintained at 37 C in a humidified atmosphere of 95% analyzed experimental conditions. air and 5% CO2. Wild-type Ddx5-HA and Ddx17-HA and mutated Ddx5-HA K144A and Ddx17-HA K142R Western blot analysis were cloned into pTRE2-hyg vectors to generate inducible MCF-7/Tet-On stable cell lines (Clontech). Resistant Total protein extracts were obtained using NP-40 buffer clones were selected with hygromycin (300 mg/ml, (50 mM Tris–HCl pH 8, 0.4 M. NaCl, 5 mM EDTA pH 8, Clontech) and protein expression was checked after 1% NP40, 0.2% SDS, 1 mM DTT) supplemented with Doxycycline treatment (1 mg/ml) for 48 h. Protease and Phosphatase Inhibitors (Roche). An amount of 20 mg of proteins were separated by Õ Cell transfection and treatment NuPAGE Novex 3-8% Tris-Acetate or 12% bis-tris- Acetate Gels (LifeTechnologies). Membranes were 6 A total of 3 10 cells were transiently transfected with incubated with specific primary antibodies against Ddx5 26.6 nM siRNAs (Supplementary Table S3) using Lipo- (ab10261, Abcam), Ddx17 (ab24601 Abcam), SMRT (H- fectamine RNAiMax (LifeTechnologies). Twenty-four 300, sc-20778 Santa Cruz), AR (441, sc-7305), ERa (F-10, hours before treatment, MCF-7 and LNCaP cells, were sc-8002 Santa Cruz), GSK3b (27C10, Cell signaling) and cultured in red phenol-free medium supplemented with 2% b-Actin (I-19, sc-1616 Santa Cruz). charcoal-treated FBS. Cells were treated with E2 (10 nM; Sigma) for 1 or 10 h or DHT (10 nM; Sigma) for 24 h. Chromatin-immunoprecipitation Control cells received equal volumes of vehicle (ethanol). ERa chromatin-immunoprecipitation (ChIP) assay was performed in MCF-7 cells using anti-ERa antibody RNA preparation, RT–PCR and RT–qPCR (HC-20, sc-543 Santa Cruz). Ddx5-HA ChIP assay was Total RNAs were prepared using TRIpure Isolation performed in inducible MCF-7 cell line as previously Reagent (Roche), and 1 ml of Glycoblue (Ambion) was described (21) using anti-HA antibody (3F10, Roche).

 {€  Nucleic Acids Research, 2013 3

MCF-7 or MCF-7 inducible cell lines were maintained in transcription, but also for the production of specific red phenol-free medium supplemented with 2% charcoal- isoforms from, at least, a subset of endogenous E2- treated FBS. Cells were treated with E2 (10 nM; Sigma) regulated genes. However, as most of the genes regulated for 1 h before fixation. Primers used are detailed in at the splicing level by Ddx5 and Ddx17 were not regulated Supplementary Table S3. at the transcriptional level by E2 treatment (Figure 2A), this suggested that Ddx5/17 effects on splicing and E2 effects on transcription were mainly independent. RESULTS Ddx5 and Ddx17 are master regulators of the estrogen- Ddx5 and Ddx17 regulate splicing of several genes signaling pathway involved in the estrogen-signaling pathway In order to assess the extent to which Ddx5 and Ddx17 Among 193 genes encoding for mediators or regulators of participate in gene expression regulation in response to the estrogen-signaling pathway, 26 were predicted to be estradiol (E2), ER-positive MCF-7 breast cancer cells regulated by Ddx5/17 at the splicing level (Figure 2C and were treated with E2 or vehicle for 10 h after being trans- Supplementary Figure S3A). For example, several genes

fected with a control siRNA (siCTRL) or an siRNA coding for protein kinases involved in ERa phosphoryl- Downloaded from (siDdx5/17) targeting a conserved region shared by ation, including CDK2, p38 MAPK (ERK2) and GSK3b, Ddx5 and Ddx17 (Figure 1A). Whole-transcriptome were affected at the splicing level when Ddx5 and Ddx17 analysis was performed using Affymetrix Exon Arrays, were knocked down (Figure 2D and Supplementary which allowed us to assess expression at the global gene Figure S3A). Likewise, several ERa transcriptional level as well as at the exon level. The median intensity of coregulators were affected at the splicing level by Ddx5/ exonic probes of each gene was computed in each sample 17 silencing (Figure 2D and Supplementary Figure S3A). http://nar.oxfordjournals.org/ from four independent experiments. Using cutoffs of 1.5 This included transcriptional coactivators like CREB- for fold change and 0.05 for P-value, 354 (67%) and 173 binding protein (CBP) and the Mediator Subunit 1 (33%) genes were predicted to be activated and repressed (MED1) that plays a crucial role to anchor liganded by E2, respectively (Figure 1B). Comparative analysis of ERa to the transcriptional pre-initiation complex (24), control and Ddx5/17-depleted cells (Supplementary Table as well as several corepressors such as the nuclear S1) revealed that Ddx5 and Ddx17 are required for the corepressor 1 (N-CoR1) and its paralog N-COR2, also regulation of 186 (53%) of E2-activated and 98 (57%) of known as SMRT (Silencing Mediator of Retinoic acid

E2-repressed genes (Figure 1B and Supplementary Table and Thyroid hormone receptor) (25). The identity of at Centre Léon Bérard on January 15, 2014 S1) as validated by RT–qPCR for a large number of cases some splicing variants was verified by sequencing (Figure 1C). Similar results were observed in a subset of (Supplementary Figure S4). selected genes using another set of siRNAs targeting Ddx5 To further confirm the effects of Ddx5 and Ddx17 on and Ddx17 (Figure 1D and Supplementary Figure S2A). splicing, inducible stable MCF-7 cell lines expressing wild The observed effects likely occurred at the transcrip- type or helicase-mutated Ddx5 were transfected with a tional level as Ddx5/17 depletion abrogated E2-mediated control siRNA or a mixture of siRNAs targeting DDX5 effects at the pre-mRNA level and did not significantly and DDX17 UTRs. The re-expression of wild-type, but not affect the half-life of the tested E2-regulated mRNAs mutated Ddx5, partially rescued the effect of the siDdx5/17 (Figure 1E and Supplementary Figure S2C). UTR on the splicing pattern (Figure 2E, left panel). Similar Furthermore, ERa and Ddx5 were detected on a subset results were obtained in inducible stable MCF-7 cell lines of E2-target genes and their recruitment was increased expressing wild-type Ddx17 or the mutant K142R lacking after E2 treatment as demonstrated by ChIP assay the helicase activity (Figure 2E, right panel). This observa- (Figure 1F). These results are in agreement with tion raised the possibility that Ddx5 and Ddx17 previous reports (9,11,22,23) indicating that Ddx5 and contributed to the ERa-signaling pathway not only as Ddx17 are bona fide ERa transcriptional coregulators ERa transcriptional coregulators but also by regulating and are therefore required for mediating E2 effects on the splicing pattern of key actors of the pathway. transcription. However, we observed that Ddx5/17 deple- To address the functional relevance of splicing regula- tion also decreased ERa protein (but not mRNA) levels tion of key actors of the ERa-signaling pathway by Ddx5 (Figure 1G), suggesting that Ddx5 and Ddx17 may also and Ddx17, we focused on the SMRT transcriptional contribute to the estrogen-signaling pathway by another corepressor. As shown in Figure 3A, RT–qPCR analysis mechanism (see below). Of note, Ddx5/17 depletion did confirmed the RT–PCR data (Figure 2D) and showed not affect other signaling pathways (Supplementary that Ddx5/17 depletion induced exon 42 skipping Figure S2D). (SMRTÁE42). Interestingly, exon 42 skipping is predicted In addition to their role in transcription, Ddx5 and to introduce a premature stop codon in exon 43 Ddx17 have been shown to play a role in alternative (Figure 3B, upper panel), suggesting that SMRTÁE42 splicing regulation (14–17). The analysis of Exon Array transcripts might be unproductive (that is, not translated) data revealed that Ddx5 and Ddx17 silencing changed and regulated by the nonsense-mediated mRNA decay the splicing pattern of 65 genes among the E2-regulated (NMD) pathway. Supporting this hypothesis, incubation genes (Figure 2A and Supplementary Table S1), as of MCF-7 cells with the translation inhibitor validated by RT–PCR (Figure 2B). Therefore, Ddx5 and cycloheximide, that blocks mRNA degradation by the Ddx17 are not only required for the E2-mediated effects on NMD pathway, increased the ratio of SMRTÁE42

 {€‚ 4 Nucleic Acids Research, 2013

A BC

D E Downloaded from http://nar.oxfordjournals.org/

F at Centre Léon Bérard on January 15, 2014

G

Figure 1. Ddx5 and Ddx17 are master regulators of the estrogen-signaling pathway. (A) Western blot analysis of Ddx5, Ddx17 and b-Actin as loading control, using total protein extract from MCF-7 cells transfected with either a control siRNA or siDdx5/17 in the presence or absence of estradiol (E2) for 10 h. (B) In silico prediction of Exon array analysis of genes regulated at the global expression level by E2 and dependent of Ddx5 and Ddx17 (in gray). (C) Hormone fold change induced by E2 treatment for 10 h of the expression level of a subset of mRNAs as determined by RT–qPCR in MCF-7 cells transfected with siCTRL or siDdx5/17 (genes upregulated by E2 in the upper panel and genes downregulated by E2 in the lower panel). (D) Hormone fold change induced by E2 treatment for 10 h of the expression level of a subset of mRNAs as determined by RT–qPCR in MCF-7 cells transfected with siCTRL or a mixture of siRNAs-targeting Ddx5 and Ddx17 UTRs (siDdx5/17 UTR). (E) Hormone fold change as assessed by RT-qPCR measuring the unspliced pre-mRNAs of E2-regulated genes in the nuclear extracts. MCF-7 cells were transiently transfected with Control siRNA or siDdx5/17 then treated with E2 or vehicle for 1 h. (F) qPCR analysis using primers spanning promoter regions of a subset of E2-regulated genes on genomic DNA in a ChIP assay using MCF-7 for ERa (left panel) and MCF-7 stable cell lines expressing Ddx5-HA (right panel). (G) Western blot analysis of ERa and b-Actin as loading control using MCF-7 cells transfected with either a control siRNA or siDdx5/17 in the presence or absence of estradiol (E2) for 10 h (left panel). Relative expression level of ERa mRNA as assessed by RT–qPCR analysis (right panel). Histograms represent the average of at least three independent experiments. Error bars, s.e.m.; *P < 0.05; **P < 0.01; ***P < 0.001 (t-test). transcripts to transcripts containing exon 42 decrease of SMRT protein expression levels (Figure 3C, (SMRT+E42) (Figure 3B, lower panel). In addition and left panel). Finally, MCF-7 cells transfected with siRNA as expected from the increased NMD-degraded targeting SMRT mRNA showed a significantly reduced SMRTÁE42 production, Ddx5/17 depletion induced a E2 effect on a subset of downregulated transcripts

 {€ƒ Nucleic Acids Research, 2013 5

A B

CD E Downloaded from http://nar.oxfordjournals.org/

Figure 2. Ddx5 and Ddx17 regulate splicing of a subset of E2-target genes and genes involved in estrogen-signaling pathway. (A) in silico prediction of Exon array analysis. Venn diagram representing E2-regulated genes and genes that were regulated by Ddx5 and Ddx17 at the splicing level. (B) Validation by RT–PCR of splicing variants produced from E2-target genes and regulated by Ddx5 and Ddx17. Exons are represented by white boxes and primers used for PCR by black arrows. (C) Diagram representing ER-signaling regulators genes that were regulated by Ddx5 and Ddx17 at the splicing level. (D) Validation by RT–PCR of splicing events regulated by Ddx5/17. MCF-7 cells were transfected for 48 h with a Control siRNA at Centre Léon Bérard on January 15, 2014 (siCTRL) or siDdx5/17, then treated for 10 h with E2 (10 nM). Depletion of Ddx5 and Ddx17 induced both exon inclusion and exon skipping. (E) Rescue experiments in inducible stable cell lines. Re-expression of wild-type or mutated Ddx5 (left panel) and Ddx17 (right panel) forms in MCF-7 inducible stable cell lines after transfection with siRNAs targeting DDX5 and DDX17 UTRs was induced by Doxycyclin (1 mg/ml) treatment.

compared with control cells (Figure 3C, right panel and splicing was affected by Ddx5/17 depletion (Figure 4D 3D). Altogether, these data indicated that the role of Ddx5 and Supplementary Table S2), as validated by RT–PCR and Ddx17 in the regulation of the estrogen-signaling (Figure 4E). Finally, as shown in Figure 4F, several (24 pathway is not limited to their function as ER transcrip- out of 144) Ddx5/17-regulated splicing variants are tional coregulators on estrogen-regulated genes, but also produced from genes involved in the regulation of the pointed to a previously undocumented regulation of androgen-signaling pathway (Figure 4G and coregulated splicing events directly linked to this pathway. Supplementary Figure S3B). Of particular interest, and similarly to our observation on ERa, Ddx5/17 depletion in LNCaP cells, reduced Ddx5 and Ddx17 play a major role in the androgen- AR protein (but not mRNA) level (Figure 4H). signaling pathway Therefore, Ddx5 and Ddx17 likely contribute to both As for ERa, Ddx5 and Ddx17 are transcriptional estrogen and androgen-signaling pathways through coregulators of AR. To study the genome-wide contribu- several mechanisms. tion of Ddx5 and Ddx17 to the androgen-signaling a pathway, we applied the same experimental strategy Ddx5 and Ddx17 stabilize ER and AR by modulating b using the human LNCaP prostate cancer cell line that GSK3 splicing was treated with DHT for 24 h (Figure 4A). Affymetrix Comparing the data obtained in both MCF-7 and LNCaP Exon array analysis predicted that out of a total of 1573 cells, we observed that several genes regulated at the DHT-regulated genes, 1319 (84%) were activated and 254 splicing level by Ddx5/17 are involved in both ERa and (16%) repressed (Figure 4B). In silico analysis predicted AR-signaling pathways (e.g. CBP, MED1, N-COR1, that 868 (66%) of DHT-activated and 103 (40%) of DHT- SMRT and GSK3b)(Figure 2D, 4G and Supplementary repressed genes were regulated in a Ddx5/17-dependent Figure S3). Interestingly, it has been shown that the manner (Figure 4B and Supplementary Table S2), as GSK3b kinase phosphorylates and stabilizes both ERa validated by RT–qPCR (Figure 4C). (5,26) and AR (6,27). Two GSK3b isoforms, resulting When analyzing Exon Array data at the exon level, we from exon 9 alternative splicing, have been reported in identified 371 DHT-regulated genes (among 1573) whose rodents (28,29) and human (30). The conventional

 {z 6 Nucleic Acids Research, 2013

AB C

D Downloaded from

Figure 3. Ddx5 and Ddx17 control SMRT protein expression level. (A) RT–qPCR analysis measuring SMRT splicing variant expression level in MCF-7 cells transfected for 48 h with a control siRNA (siCTRL) or siDdx5/17. Specific primers were designed to amplify transcripts containing or http://nar.oxfordjournals.org/ not exon 42, SMRT+E42 and SMRTÁE42, respectively. (B) Upper panel, skipping of exon 42 results in a premature stop codon in exon 43. Lower panel, RT–qPCR analysis measuring SMRT splicing variant ratio in MCF-7 cells after 6 h of cycloheximide treatment (10 mg/ml). (C) Western blot analysis of SMRT protein expression level and b-Actin as a loading control in MCF-7 cells transfected with a control siRNA, siDdx5/17 or siRNA- targeting SMRT mRNA. (D) Hormone fold change induced by E2 treatment for 10 h of the expression level of a subset of mRNAs as determined by RT-qPCR in MCF-7 cells transfected with siCTRL or siSMRT. Histograms represent the average of at least three independent experiments. Error bars represent s.e.m.; **P < 0.01; ***P < 0.001 (t test).

isoform GSK3b1 that regulates ERa and AR protein sta- pharmacological inhibition (Figure 5E, bottom panel), as at Centre Léon Bérard on January 15, 2014 bility does not contain exon 9 in contrast to the GSK3b2 previously reported (6,27,34,35). isoform (Figure 5A, upper panel). Interestingly, exon 9 Therefore, Ddx5/17 directly participate to the estrogen- encodes for a 13 amino acid peptide within the kinase and androgen-signaling pathways as ERa and AR tran- catalytic domain, which reduces the kinase activity of scriptional coregulators but also indirectly by regulating the GSK3b2 isoform compared with the canonical alternative splicing of other transcriptional coregulators GSK3b1 isoform (28,30,31). or other actors of the signaling pathways (Figure 5F, We first validated the effect of Ddx5/17 depletion on and see ‘Discussion’ section). GSK3b exon 9 splicing by RT–qPCR analysis (bottom panel of Figure 5A). This confirmed that in the absence of Ddx5/17, exon 9 inclusion augments, indicating that the DISCUSSION production of GSK3b2 splicing variant is increased while Activation of the estrogen and androgen-signaling the GSK3b1 splicing variant is decreased in MCF-7 cells. pathways requires the concerted action of a plethora of Accordingly, a western blot analysis showed that Ddx5/17 factors that either affect post-translational modifications silencing resulted in the appearance of a slower migrating of ERa and AR or are recruited on hormone target genes protein band detected by the GSK3b antibody (Figure to mediate their transcriptional effects. In this work, we 5B), consistent with the detection of the GSK3b2 report that the Ddx5 and Ddx17 RNA helicases, that have isoform (32), while the GSK3b1 isoform decreased. been shown to be ERa and AR transcriptional As previously reported (5,26,33), GSK3b depletion and coregulators (see ‘Introduction’ section), are master regu- pharmacological inhibition resulted in ERa protein lators of the estrogen and androgen-signaling pathways by level decrease (Figure 5C). Remarkably, Ddx5/17 deple- controlling transcription and splicing both upstream and tion had the same effect (Figure 5C) and this effect was downstream of receptors (Figure 5F). Indeed, Ddx5 and further enhanced in MCF-7 cells transfected with specific Ddx17 regulate the expression of a large number of siRNAs (Figure 5D, upper panel) targeting the GS3K3b1 estrogen- and androgen-target genes both at the transcrip- but not the GSK3b2 isoform (Figure 5D, bottom panel). tional and splicing levels (Figures 1–4). Therefore, Ddx5 Collectively these results support a model in which Ddx5/ and Ddx17 intervene downstream of the signaling 17 silencing reduces the production of the canonical pathways for the transcriptional and splicing regulation GSK3b1 isoform, resulting in a decreased ERa protein of hormone-targeted genes. This result observed on a level. Similar results were obtained in LNCaP cells: large scale on endogenous hormone target genes Ddx5/17 depletion decreased the production of the supports a model where at least some transcriptional conventional GSK3b1 isoform (Figure 5E, upper panel) coregulators are not only involved in the quantitative as well as AR protein level, as did GSK3b depletion or regulation of endogenous gene expression but also in

 {{  Nucleic Acids Research, 2013 7

A BC

D E Downloaded from http://nar.oxfordjournals.org/

F G at Centre Léon Bérard on January 15, 2014

H

Figure 4. Ddx5 and Ddx17 play a major role in the androgen-signaling pathway. (A) Western blot analysis of Ddx5, Ddx17 and b-Actin as loading control, using total protein extract from LNCaP cells transfected with either a control siRNA (siCTRL) or siDdx5/17 in the presence or absence of DHT for 24 h. (B) In silico prediction of exon array analysis of genes regulated at the global expression level by DHT and dependent of Ddx5 and Ddx17 (in gray). (C) Hormone fold change induced by DHT treatment for 24 h of the expression level of a subset of mRNAs as determined by RT– qPCR in LNCaP cells transfected with siCTRL or siDdx5/17 (genes upregulated by DHT in the upper panel and genes downregulated by DHT in the lower panel). (D) In silico prediction of Exon array analysis. Venn diagram representing DHT-regulated genes and genes that were regulated by Ddx5 and Ddx17 at the splicing level. (E) Validation by RT–PCR of splicing variants produced from DHT-target genes and regulated by Ddx5 and Ddx17. LNCaP cells were transfected with siCTRL or siDdx5/17 then treated by DHT (10 nM) for 24 h. (F) Diagram representing AR-signaling regulators genes that were regulated by Ddx5 and Ddx17 at the splicing level. (G) Validation by RT–PCR of splicing events regulated by Ddx5/ Ddx17 in the same experimental condition than F panel. (H) Western blot analysis of Ddx5, Ddx17, AR proteins level. b-Actin was used as loading control (left panel). LNCaP cells were transfected with either a control siRNA or siDdx5/17 in the presence or absence of DHT for 24 h. RT–qPCR analysis of the relative expression level of AR mRNA (right panel). Histograms represent the average of at least three independent experiments. Error bars represent s.e.m.; *P < 0.05; **P < 0.01 (t-test).

their qualitative regulation by allowing the production of level, of several key regulators of the hormone-signaling specific splicing variants in response to stimuli. pathways. (Figures 1–4 and 5F). Of particular interest, we Furthermore, we show that Ddx5 and Ddx17 act showed that Ddx5 and Ddx17 controlled the expression upstream of the estrogen and androgen-signaling level of the SMRT transcriptional coregulator. Indeed, pathways by controlling the expression, at the splicing Ddx5/17 depletion favored the production of a SMRT

 {| 8 Nucleic Acids Research, 2013

A B D

C Downloaded from E F http://nar.oxfordjournals.org/

Figure 5. Ddx5 and Ddx17 stabilize ERa and AR protein by modulating GSK3b splicing. (A) Schematic representation of GSK3b primary tran- at Centre Léon Bérard on January 15, 2014 scripts (upper panel). RT–qPCR analysis measuring the GSK3b1 and GSK3b2 ratio in MCF-7 cells transfected with a control siRNA or siDdx5/17. (B) Western blot analysis of GSK3b and b-actin as a loading control in MCF-7 cells transfected with a control siRNA or siDdx5/17. (C) Western blot analysis of ERa protein level and b-actin as a loading control in MCF-7 cells transfected with a control siRNA, siDdx5/17 or siGSK3b (left panel) or treated with 20 mM of the SB-216673 GSK3b inhibitor for 24 h (right panel). (D) RT–qPCR analysis measuring GSK3b1 and GSK3b2 splicing variant expression level in MCF-7 cells transfected with specific siRNAs targeting each isoform (i.e. siGSK3b1 and siGSK3b2, upper panel). Western blot analysis of ERa protein level and b-Actin as a loading control in MCF-7 cells transfected with a control siRNA, siDdx5/17 and siGSK3b1 or siGSK3b2 (lower panel). (E) RT–qPCR analysis measuring the GSK3b1 and GSK3b2 ratio in LNCaP cells transfected with a control siRNA or siDdx5/17 (upper panel). Western blot analysis of AR protein level and b-actin as a loading control in LNCaP cells transfected with a control siRNA, siDdx5/17 or siGSK3b or treated with 20 mM of the SB-216673 GSK3b inhibitor for 24 h (lower panels). (F) Ddx5 and Ddx17 that are recruited by Steroid Hormone Receptors (HR) on Hormone Responsive Elements (HRE) act as transcriptional coregulators and regulate alternative splicing of a least a subset of hormone-regulated genes (left panel). Meanwhile, Ddx5 and Ddx17 control the expression at the splicing level of key regulators of the steroid hormone-signaling pathways. Effectors of the steroid hormone-signaling pathways regulated at the splicing level by Ddx5/17 can either be transcriptional coregulators (e.g. SMRT) or proteins modulating HR post-translational modifications, like GSK3b that controls AR and ERa protein expression levels. splicing variant that is degraded by the NMD pathway, It has been proposed that, these proteins may help the leading to the decrease of SMRT protein level (Figure 3). binding of splicing regulators on their target pre- Furthermore, we demonstrated that Ddx5 and Ddx17 mRNAs owing to their RNA helicase activity (14,15,39). regulated alternative splicing of GSK3b, thereby Interestingly, Ddx5 and Ddx17 interact with RNA poly- controlling AR and ERa protein levels (Figure 5). Thus, merase II (RNAPII) and could affect transcription our data show that Ddx5/17 control steroid hormone-sig- elongation (40,41). Because transcription elongation can naling pathways on a large scale, by acting both upstream have an impact on alternative splicing regulation (42), and downstream of hormone receptors. Ddx5 and Ddx17 could impact on splicing owing to One interesting remaining question is whether Ddx5 their effects on transcription elongation. However, as and Ddx17 effects on transcription and splicing are most of the genes regulated at the splicing level by somehow connected. At the transcriptional level, it Ddx5 and Ddx17 were not regulated at the transcriptional has been proposed that Ddx5 and Ddx17 are recruited level by E2 or DHT treatment (Figures 2A and 4D), at the promoter level by transcriptional factors and this suggested that Ddx5/17 effects on splicing and E2 serve as a bridge between transcription factors and or DHT effects on transcription were mainly independent. other transcriptional coregulators, in particular, histone However, this does not exclude a link between tran- modifiers CBP (36,37) and HDACs (38). The effects scription and splicing. For example, Ddx5 and Ddx17 of Ddx5 and Ddx17 in splicing are less characterized. may have to be first recruited by the transcriptional

 {}  Nucleic Acids Research, 2013 9 machinery before being loaded onto the nascent RNA which was associated with reduced kinase activity molecules to impact on splicing; the recruitment by the (30,31). Changes in the balance of GSK3b2/GSK3b1 transcriptional machinery may not necessarily be measur- isoforms due to alternative splicing regulation could able by an effect on gene transcription activity. have a broad impact in cancer initiation or progression, Even though further experiments are required to better not only owing to an effect on ER and AR protein level, as understand the mechanisms by which Ddx5 and Ddx17 reported in this work but also owing to the key role of affect transcription and splicing, their broad effects on GS3Kb in the b-catenin homeostasis and its cross-talk the estrogen and androgen-signaling pathways has with WNT-signaling pathway. several physio-pathological consequences. First, the Because there is increasing evidence that the interplay misregulation of Ddx5 and/or Ddx17 expression level between transcription and splicing participates in signal- and/or activity owing to posttranslational modifications ing pathway regulation and outcomes, we created a freely that was reported in many cases of breast and prostate available website, named SSAS-DB (http://fasterdb.lyon. cancer (10,23,43–45), may have a broad impact on the unicancer.fr/ssas-db/home.pl) that provides genome-wide estrogen- and androgen-signaling pathways that play a information on the impact of estrogen, androgen and major role in the etiology and progression of those Ddx5/17 on gene expression in breast and prostate cancers. For example, the Ddx5 gene is fused in frame cancer cell lines, at both whole-gene (transcriptional) Downloaded from to the ETV4 gene in prostate tumors (43). Based on our and exon- (alternative splicing) levels. This website is a data, it can be anticipated that such a translocation may valuable tool to analyze the splicing variants of have a broad impact on the androgen-signaling pathway hormone-regulated genes or genes involved in hormone- in these tumors. signaling pathways (see Supplementary Figure S5 for Second, we observed that many of the Ddx5/17 splicing- further information on SSAS-DB). http://nar.oxfordjournals.org/ regulated genes have been reported to play a direct role in In conclusion, our work provides a new understanding tamoxifen resistance, one of the major endocrine therapies of steroid hormone-signaling pathways owing to the iden- of breast cancer. This includes several transcriptional tification of two related proteins, Ddx5 and Ddx17, that coregulators such as SMRT, MED1 and NCoR1 not only mediate the regulation at both transcriptional involved in resistance to tamoxifen (46–48) and several and splicing levels of a large subset of steroid hormone kinases such as GSK3b, CDK2 and RET which partici- target genes, but also control a set of splicing events in pate, directly or indirectly, to ERa phosphorylation and genes that, in turn, control steroid receptor abundance thus, the hormone-independent activation of ER and tam- and activity. Thus, we propose that Ddx5 and Ddx17 oxifen resistance (49–51). Interestingly as well, Ddx5 are master regulators of estrogen and androgen-signaling at Centre Léon Bérard on January 15, 2014 expression correlates with that of HER-2/neu epidermal pathways by acting both upstream and downstream of growth factor receptor (23,45). It has been reported that receptors. overexpression of HER2/neu is associated with increased ER phosphorylation on ser-118 residue (52) which could play a role in resistance to tamoxifen by permanent ERa SUPPLEMENTARY DATA activation. Interestingly, even though further experiments Supplementary Data are available at NAR Online, are required because of potential contradictory reported including [55]. data, high Ddx5 expression level could be associated with higher tumor grade or poor prognosis in ER-positive breast cancer patients (23,53). Furthermore, a recent AKNOWLEDGEMENTS report indicate that, Ddx5 and Ddx17 proteins level change significantly according to breast cancer subtype The authors thank Mark McCarron for helpful reading (45) and is highly dependent on certain miRNAs abun- of the paper and Marie-Pierre Lambert for drawing dance and regulation, in particular miR-206 that has figures. been reported to play a major role in breast cancer and estrogen pathway (54). Therefore, changes in the expres- sion of Ddx5 and Ddx17, in particular under the control FUNDING of altered miRNAs in breast cancer, could play a role in The Institut National du Cancer (INCa); Agence resistance to tamoxifen, in particular owing to their effects Nationale de la Recherche (ANR); Fondation Recherche on splicing of genes involved in resistance (see above). It Me´dicale (FRM); Ligue Nationale Contre le Cancer (to would be particularly interesting to test whether resistance S.S. and E.D.); Association Franc¸aise contre les to endocrine therapy is associated with mis-regulation of Myopathies (to M.P.E.), FRM (to E.Z.) and Association splicing of genes involved in steroid hormone-signaling pour la Recherche sur le Cancer (to S.G.). Funding for pathways. open access charge: Institut National du Cancer (INCa) While many researches have been concentrated on the Agence Nationale de la Recherche (ANR) Fondation canonical GSK3b1, ubiquitously distributed in organs, Recherche Me´dicale (FRM) Ligue Nationale Contre le minor attention was attributed to GSK3b2 isoform, Cancer (LNCC) Association Franc¸aise contre les whose substrate preference and physiological significance Myopathies (AFM) Association pour la Recherche sur b remain unclear. In GSK3 2 long isoform, exon 9 transla- le Cancer (ARC) tion produces a 13 amino acid insert of unknown function in an external loop near the catalytic domain (28–30) Conflict of interest statement. None declared.

 {~  10 Nucleic Acids Research, 2013

REFERENCES 18. Kelemen,O., Convertini,P., Zhang,Z., Wen,Y., Shen,M., Falaleeva,M. and Stamm,S. (2013) Function of alternative 1. Madhunapantula,S.V., Mosca,P. and Robertson,G.P. (2010) splicing. Gene, 514, 1–30. Steroid hormones drive cancer development. Cancer Biol. Ther., 19. Auboeuf,D., Honig,A., Berget,S.M. and O’Malley,B.W. (2002) 10, 765–766. Coordinate regulation of transcription and splicing by steroid 2. Green,S.M., Mostaghel,E.A. and Nelson,P.S. (2012) Androgen receptor coregulators. Science, 298, 416–419. action and metabolism in prostate cancer. Mol. Cell Endocrinol., 20. Barbier,J., Dutertre,M., Bittencourt,D., Sanchez,G., Gratadou,L., 360, 3–13. de la Grange,P. and Auboeuf,D. (2007) Regulation of H-ras 3. Kato,S., Yokoyama,A. and Fujiki,R. (2011) Nuclear receptor splice variant expression by cross talk between the p53 and coregulators merge transcriptional coregulation with epigenetic nonsense-mediated mRNA decay pathways. Mol. Cell Biol., 27, regulation. Trends Biochem. Sci., 36, 272–281. 7315–7333. 4. Le Romancer,M., Poulard,C., Cohen,P., Sentis,S., Renoir,J.M. 21. Zonta,E., Bittencourt,D., Samaan,S., Germann,S., Dutertre,M. and Corbo,L. (2011) Cracking the estrogen receptor’s and Auboeuf,D. (2013) The RNA helicase DDX5/p68 is a key posttranslational code in breast tumors. Endocr. Rev., 32, factor promoting c-fos expression at different levels from 597–622. transcription to mRNA export. Nucleic Acids Res., 41, 554–564. 5. Grisouard,J., Medunjanin,S., Hermani,A., Shukla,A. and 22. Endoh,H., Maruyama,K., Masuhiro,Y., Kobayashi,Y., Goto,M., Mayer,D. (2007) Glycogen synthase kinase-3 protects estrogen Tai,H., Yanagisawa,J., Metzger,D., Hashimoto,S. and Kato,S. receptor alpha from proteasomal degradation and is required for (1999) Purification and identification of p68 RNA helicase acting

full transcriptional activity of the receptor. Mol. Endocrinol., 21, as a transcriptional coactivator specific for the activation function Downloaded from 2427–2439. 1 of human estrogen receptor alpha. Mol. Cell. Biol., 19, 6. Schutz,S.V., Cronauer,M.V. and Rinnab,L. (2010) Inhibition of 5363–5372. glycogen synthase kinase-3beta promotes nuclear export 23. Wortham,N.C., Ahamed,E., Nicol,S.M., Thomas,R.S., of the androgen receptor through a CRM1-dependent Periyasamy,M., Jiang,J., Ochocka,A.M., Shousha,S., Huson,L., mechanism in prostate cancer cell lines. J. Cell. Biochem., 109, Bray,S.E. et al. (2009) The DEAD-box protein p72 regulates 1192–1200. ERalpha-/oestrogen-dependent transcription and cell growth, and 7. O’Malley,B.W., Malovannaya,A. and Qin,J. (2012) Minireview: is associated with improved survival in ERalpha-positive breast http://nar.oxfordjournals.org/ nuclear receptor and coregulator proteomics–2012 and beyond. cancer. Oncogene, 28, 4053–4064. Mol. Endocrinol., 26, 1646–1650. 24. Zhang,X., Krutchinsky,A., Fukuda,A., Chen,W., Yamamura,S., 8. Rosenfeld,M.G., Lunyak,V.V. and Glass,C.K. (2006) Sensors and Chait,B.T. and Roeder,R.G. (2005) MED1/TRAP220 exists signals: a coactivator/corepressor/epigenetic code for integrating predominantly in a TRAP/ Mediator subpopulation enriched in signal-dependent programs of transcriptional response. Genes RNA polymerase II and is required for ER-mediated Dev., 20, 1405–1428. transcription. Mol. Cell, 19, 89–100. 9. Watanabe,M., Yanagisawa,J., Kitagawa,H., Takeyama,K., 25. Varlakhanova,N., Snyder,C., Jose,S., Hahm,J.B. and Ogawa,S., Arao,Y., Suzawa,M., Kobayashi,Y., Yano,T., Privalsky,M.L. (2010) Estrogen receptors recruit SMRT and N- Yoshikawa,H. et al. (2001) A subfamily of RNA-binding CoR corepressors through newly recognized contacts between the DEAD-box proteins acts as an estrogen receptor alpha corepressor N terminus and the receptor DNA binding domain. at Centre Léon Bérard on January 15, 2014 coactivator through the N-terminal activation domain Mol. Cell Biol., 30, 1434–1445. (AF-1) with an RNA coactivator, SRA. EMBO J., 20, 26. Medunjanin,S., Hermani,A., De Servi,B., Grisouard,J., Rincke,G. 1341–1352. and Mayer,D. (2005) Glycogen synthase kinase-3 interacts 10. Clark,E.L., Coulson,A., Dalgliesh,C., Rajan,P., Nicol,S.M., with and phosphorylates estrogen receptor alpha and is Fleming,S., Heer,R., Gaughan,L., Leung,H.Y., Elliott,D.J. et al. involved in the regulation of receptor activity. J. Biol. Chem., (2008) The RNA helicase p68 is a novel androgen receptor 280, 33006–33014. coactivator involved in splicing and is overexpressed in 27. Schutz,S.V., Schrader,A.J., Zengerling,F., Genze,F., prostate cancer. Cancer Res., 68, 7938–7946. Cronauer,M.V. and Schrader,M. (2011) Inhibition of glycogen 11. Metivier,R., Penot,G., Hubner,M.R., Reid,G., Brand,H., Kos,M. synthase kinase-3beta counteracts ligand-independent activity of and Gannon,F. (2003) Estrogen receptor-alpha directs ordered, the androgen receptor in castration resistant prostate cancer. cyclical, and combinatorial recruitment of cofactors on a natural PLoS One, 6, e25341. target promoter. Cell, 115, 751–763. 28. Wood-Kaczmar,A., Kraus,M., Ishiguro,K., Philpott,K.L. and 12. Wong,H.Y., Demmers,J.A., Bezstarosti,K., Grootegoed,J.A. and Gordon-Weeks,P.R. (2009) An alternatively spliced form of Brinkmann,A.O. (2009) DNA dependent recruitment of DDX17 glycogen synthase kinase-3beta is targeted to growing neurites and other interacting proteins by the human androgen receptor. and growth cones. Mol. Cell Neurosci., 42, 184–194. Biochim. Biophys. Acta, 1794, 193–198. 29. Mukai,F., Ishiguro,K., Sano,Y. and Fujita,S.C. (2002) Alternative 13. Liu,Z.R. (2002) p68 RNA helicase is an essential human splicing splicing isoform of tau protein kinase I/glycogen synthase kinase 0 factor that acts at the U1 snRNA-5 splice site duplex. Mol. Cell. 3beta. J. Neurochem., 81, 1073–1083. Biol., 22, 5443–5450. 30. Schaffer,B., Wiedau-Pazos,M. and Geschwind,D.H. (2003) Gene 14. Kar,A., Fushimi,K., Zhou,X., Ray,P., Shi,C., Chen,X., Liu,Z., structure and alternative splicing of glycogen synthase kinase 3 Chen,S. and Wu,J.Y. (2011) RNA helicase p68 (DDX5) regulates beta (GSK-3beta) in neural and non-neural tissues. Gene, 302, tau exon 10 splicing by modulating a stem-loop structure at the 73–81. 50 splice site. Mol. Cell Biol., 31, 1812–1821. 31. Castano,Z., Gordon-Weeks,P.R. and Kypta,R.M. (2010) The 15. Guil,S., Gattoni,R., Carrascal,M., Abian,J., Stevenin,J. and neuron-specific isoform of glycogen synthase kinase-3beta is Bach-Elias,M. (2003) Roles of hnRNP A1, SR proteins, and required for axon growth. J. Neurochem., 113, 117–130. p68 helicase in c-H-ras alternative splicing regulation. Mol. Cell 32. Kwok,J.B., Hallupp,M., Loy,C.T., Chan,D.K., Woo,J., Biol., 23, 2927–2941. Mellick,G.D., Buchanan,D.D., Silburn,P.A., Halliday,G.M. and 16. Germann,S., Gratadou,L., Zonta,E., Dardenne,E., Gaudineau,B., Schofield,P.R. (2005) GSK3B polymorphisms alter transcription Fougere,M., Samaan,S., Dutertre,M., Jauliac,S. and Auboeuf,D. and splicing in Parkinson’s disease. Ann. Neurol., 58, 829–839. (2012) Dual role of the ddx5/ddx17 RNA helicases in the control 33. Kim,H.M., Kim,C.S., Lee,J.H., Jang,S.J., Hwang,J.J., Ro,S. and of the pro-migratory NFAT5 transcription factor. Oncogene, 31, Choi,J. (2013) CG0009, a novel glycogen synthase kinase 3 4536–4549. inhibitor, induces cell death through cyclin D1 depletion in breast 17. Dardenne,E., Pierredon,S., Driouch,K., Gratadou,L., cancer cells. PLoS One, 8, e60383. Lacroix-Triki,M., Espinoza,M.P., Zonta,E., Germann,S., 34. Liao,X., Thrasher,J.B., Holzbeierlein,J., Stanley,S. and Li,B. Mortada,H., Villemin,J.P. et al. (2012) Splicing switch of (2004) Glycogen synthase kinase-3beta activity is required for an epigenetic regulator by RNA helicases promotes tumor-cell androgen-stimulated gene expression in prostate cancer. invasiveness. Nat. Struct. Mol. Biol., 19, 1139–1146. Endocrinology, 145, 2941–2949.

 {  Nucleic Acids Research, 2013 11

35. Mazor,M., Kawano,Y., Zhu,H., Waxman,J. and Kypta,R.M. 46. Graham,J.D., Bain,D.L., Richer,J.K., Jackson,T.A., Tung,L. and (2004) Inhibition of glycogen synthase kinase-3 represses Horwitz,K.B. (2000) Thoughts on tamoxifen resistant breast androgen receptor activity and prostate cancer cell growth. cancer Are coregulators the answer or just a red herring? Oncogene, 23, 7882–7892. J. Steroid Biochem. Mol. Biol., 74, 255–259. 36. Rossow,K.L. and Janknecht,R. (2003) Synergism between p68 47. Nagalingam,A., Tighiouart,M., Ryden,L., Joseph,L., Landberg,G., RNA helicase and the transcriptional coactivators CBP and p300. Saxena,N.K. and Sharma,D. (2012) Med1 plays a critical role in Oncogene, 22, 151–156. the development of tamoxifen resistance. Carcinogenesis, 33, 37. Shin,S. and Janknecht,R. (2007) Concerted activation of the 918–930. Mdm2 promoter by p72 RNA helicase and the coactivators p300 48. van Agthoven,T., Sieuwerts,A.M., Veldscholte,J., Meijer-van and P/CAF. J. Cell Biochem., 101, 1252–1265. Gelder,M.E., Smid,M., Brinkman,A., den Dekker,A.T., 38. Wilson,B.J., Bates,G.J., Nicol,S.M., Gregory,D.J., Perkins,N.D. Leroy,I.M., van Ijcken,W.F., Sleijfer,S. et al. (2009) CITED2 and and Fuller-Pace,F.V. (2004) The p68 and p72 DEAD box RNA NCOR2 in anti-oestrogen resistance and progression of breast helicases interact with HDAC1 and repress transcription in a cancer. Br. J. Cancer, 101, 1824–1832. promoter-specific manner. BMC Mol. Biol., 5, 11. 49. Caldon,C.E., Sergio,C.M., Kang,J., Muthukaruppan,A., 39. Laurent,F.X., Sureau,A., Klein,A.F., Trouslard,F., Gasnier,E., Boersma,M.N., Stone,A., Barraclough,J., Lee,C.S., Black,M.A., Furling,D. and Marie,J. (2012) New function for the RNA Miller,L.D. et al. (2012) Cyclin E2 overexpression is associated helicase p68/DDX5 as a modifier of MBNL1 activity on with endocrine resistance but not insensitivity to CDK2 inhibition expanded CUG repeats. Nucleic Acids Res., 40, 3159–3171. in human breast cancer cells. Mol. Cancer Therapeut., 11,

40. Clark,E.L., Fuller-Pace,F.V., Elliott,D.J. and Robson,C.N. (2008) 1488–1499. Downloaded from Coupling transcription to RNA processing via the p68 DEAD 50. Oyama,M., Nagashima,T., Suzuki,T., Kozuka-Hata,H., box RNA helicase androgen receptor co-activator in prostate Yumoto,N., Shiraishi,Y., Ikeda,K., Kuroki,Y., Gotoh,N., cancer. Biochem. Soc. Trans., 36, 546–547. Ishida,T. et al. (2011) Integrated quantitative analysis of the 41. Clark,E.L., Hadjimichael,C., Temperley,R., Barnard,A., phosphoproteome and transcriptome in tamoxifen-resistant breast Fuller-Pace,F.V. and Robson,C.N. (2013) p68/DdX5 supports cancer. J. Biol. Chem., 286, 818–829. beta-catenin & RNAP II during androgen receptor mediated 51. Plaza-Menacho,I., Morandi,A., Robertson,D., Pancholi,S.,

transcription in prostate cancer. PLoS One, 8, e54150. Drury,S., Dowsett,M., Martin,L.A. and Isacke,C.M. (2010) http://nar.oxfordjournals.org/ 42. Kadener,S., Fededa,J.P., Rosbash,M. and Kornblihtt,A.R. (2002) Targeting the receptor tyrosine kinase RET sensitizes breast Regulation of alternative splicing by a transcriptional enhancer cancer cells to tamoxifen treatment and reveals a role for RET through RNA pol II elongation. Proc. Natl Acad. Sci. USA, 99, in endocrine resistance. Oncogene, 29, 4648–4657. 8185–8190. 52. Yamashita,H., Nishio,M., Toyama,T., Sugiura,H., Kondo,N., 43. Han,B., Mehra,R., Dhanasekaran,S.M., Yu,J., Menon,A., Kobayashi,S., Fujii,Y. and Iwase,H. (2008) Low phosphorylation Lonigro,R.J., Wang,X., Gong,Y., Wang,L., Shankar,S. et al. of estrogen receptor alpha (ERalpha) serine 118 and high (2008) A fluorescence in situ hybridization screen for E26 phosphorylation of ERalpha serine 167 improve survival in transformation-specific aberrations: identification of DDX5-ETV4 ER-positive breast cancer. Endocrine-related Cancer, 15, 755–763. fusion protein in prostate cancer. Cancer Res., 68, 7629–7637. 53. Dutertre,M., Gratadou,L., Dardenne,E., Germann,S., Samaan,S., Lidereau,R., Driouch,K., de la Grange,P. and Auboeuf,D.

44. Janknecht,R. (2010) Multi-talented DEAD-box proteins and at Centre Léon Bérard on January 15, 2014 potential tumor promoters: p68 RNA helicase (DDX5) and its Estrogen regulation and physiopathologic significance of paralog, p72 RNA helicase (DDX17). Am. J. Transl. Res., 2, alternative promoters in breast cancer. Cancer Res., 70, 223–234. 3760–3770. 45. Wang,D., Huang,J. and Hu,Z. (2012) RNA helicase DDX5 54. Kondo,N., Toyama,T., Sugiura,H., Fujii,Y. and Yamashita,H. regulates microRNA expression and contributes to cytoskeletal (2008) miR-206 Expression is down-regulated in estrogen reorganization in basal breast cancer cells. Mol. Cell. Proteom., receptor alpha-positive human breast cancer. Cancer Res., 68, 11, M111 011932. 5004–5008.

 {€ $'$*".,#$

9D&` ("&66&7D^  >5EA.&E@A >E!5."D.97A >@9I&77D %& 5d(?E.>& a D9 97D (D( @&D.@(&A >9E@+@E%&AA".&7D.+.?E&Aa&5"96>@&7%79D66&7D5d@D."5&, -&!3*@JRRR?E.&AD5 >@&6.'@& 6.A& &7 (I.%&7"& %E @;5& %& %K "966& "9@(,E5D&E@ %E @("&>D&E@ EK &AD@9,'7&Aa&>&7%7D^5>5E>@D%&A@(AE5DDA9!D&7EA>@"&DD&(?E.>&>@9>9A%E@;5& %&K&D%K{"966&"9@(,E5D&E@AD@7A"@.>D.97&5A^9E@(,E5D&E@A%&A6."@9A97D (D("97+.@6(A>@%dED@&AED&E@Aa

7"ME4:J^ a^ 9A&7!&@,^ a a^ 4&@67^ a^ A^ a^ -7^ a^ @7.^ a^ b @.7&@^ a a m|z{|na -& A>5.".7, +"D9@ { @&,E5D&A >9>D9A.A 7% >@95.+&@D.97 D9 >@969D& 666@L&>.D-&5.5"&55D@7A+9@6D.97a341!314341*H-*!4*1'-*-%8^JRm|n^ ||zk‚a%9.`{za{z}‚i7A6!a||z

7%@(^a^."-.&5A^a^&AA&7^a^"9DD^a^E".E^a^M7^a^b&559,&^am|zzƒnaK97." &K>@&AA.97>@9+.5.7,9+!@&AD"7"&@7%!&7.,75&A.97A`@&D@9A>&"D.I&75LA.Aa&! ,!3,-*-%8^JIm~n^}‚{kƒza%9.`{za{z{€i{~zj|z~mzƒnzz|~j

7A.&E^ a^ AD.%^ a^ 9@&E^ a^ 9@&5^ aja^ 9E"-&D^ a a^ -96A^ a^ b E.A.&EK^ a m|zz‚na 7%E"D.97 9+  !L DJ.AD >@9D&.7A A  "955D&@5 &++&"D 9+ DE69@j>@969D.7, .7"D.ID.97 9+ >@&6DE@& A&7&A"&7"&a ,!1 !**^ JMm{n^ ƒk‚ƒa %9.`{za{z{€i3a""@a|zz‚az€azz

A54A97^a a^s.55&@^aam{ƒƒ|na&5&"D.I&I&7DA.7D-&&DADD."@9"&AA&+.7&%!L 75LA.A 9+ D-& &?E&7D.5 .AA&6.7D.97 9+ E!>9>E5D.97A 9+  9EA& 66@L E69@&5&"D.I&I&7DA.7D-&&DADD."@9"&AA&+.7&%!L75LA.A9+D-&&?E&7D.5 .AA&6.7D.979+E!>9>E5D.97A9+^{}ƒƒk{~za

E!9&E+^a^DA"-(^a^ED&@D@&^a^E"-@%D^a^sd55&L^aam|zzna9@&,E5D9@A` D@7A%E".7, A.,75 +@96 D@7A"@.>D.97 D9 5D&@7D.I& A>5.".7,a 1!, 2 ', , -1',-*-%8 , !3-*'2+?^JQm}n^{||kƒa%9.`{za{z{€i3aD&6a|zzaz|azz}

EJ&D&@^ a a^ !&@AD@AA^ a a^ s E^ aja m|zz€na &?E&7"&jA>&".+." !.7%.7, 9+ A.7,5&j AD@7%&% ’` .A D-&@&  "9%& +9@ @&"9,7.D.97’]^ LMm{n^ ~ƒ~}k~ƒƒa %9.`{za{zƒ}i7@i,45€|z  { @A-^a^5@"9^ a^ 9^a^7^a^7,^a^-.^a^b@&L^a am|z{zna&".>[email protected], D-&A>5.".7,"9%&a341!^MONm|ƒ~n^}kƒa%9.`{za{z}‚i7DE@&zƒzzz

@D^ a^ s ,,9^ a m{ƒƒna ED9@&,E5D.97 9+ &K>@&AA.97 9+ D-& L&AD !>|> ej!9Kf >@9D&.7 .A 6&%.D&% !L A&?E&7"&A .7 D-& "97A&@I&% | .7D@97a &!  -41,*^ JMm{n^ }‚zzk‚a &D@.&I&% +@96 -DD>`iiJJJa>E!6&%"&7D@5a7.-a,9Ii@D."5&@&7%&@a+",.]@D.%‰}ƒ~~~sD995‰>6"&7D@&Ms @&7%&@DL>&‰!AD@"D

D&A^ a a^."95^aa^.5A97^a a^ "9!A^ajaa^9E@%97^ aja^@%@9>^ a^bE55&@j "&^aam|zzna-&!9K>@9D&.7>€‚`79I&5D@7A"@.>D.975"9"D.ID9@9+D-& >} DE69E@ AE>>@&AA9@a &!  -41,*^ KMm}n^ ~}k}a %9.`{za{z}‚iA3a&6!93a€zzz

&5D@7^ a^ E.,^ a^ &8^ a^ @"/^ a a^ 5I@&M^ a a^ &8^ a^ b %& &@@&@9A^ a a m|zz‚na  7DE@5 7D.A&7A& D@7A"@.>D @&,E5D&A &!|i.>{ ,&7& &K>@&AA.97 %[email protected], 7.5{j.7%E"&% &>.D-&5.5j6&A&7"-L65 D@7A.D.97a !,!2 H !5!*-/+!,3^ KKm€n^ €k €ƒa%9.`{za{{z{i,%a~z‚

&@,&D^ a a^ 99@&^ a^ s -@>^ a a m{ƒna >5."&% A&,6&7DA D D-& d [email protected] 9+ %&79I.@EA|5D&6a{ƒa!5'!62',! '*'1-*-%8^JIm€n^}€k€|_%.A"EAA.97 }k€a&D@.&I&%+@96-DD>`iiJJJa7"!.a756a7.-a,9Ii>E!6&%i{{{{~z

&@K^ a^ s I7 9L^ a m|zzƒna 7I95I&6&7D 9+ 6&6!&@A 9+ D-& "%[email protected] AE>&@+6.5L .7 "7"&@a -*  /1',% 1-1 !12/!3'5!2 ', '-*-%8^ Jm€n^ zz}{|ƒa %9.`{za{{z{i"A->&@A>&"Dazz}{|ƒ

-9J6."4^a^-LD.5^a^5.&D-^a^ 9@A4^aa^E697D^a^->>&55^a^b9A&A^ aa m|zz~na j!&DA.,75.7,.7+.!@9!5ADA69%E5D&AD-&97"9,&7.">9D&7D.59+%3"&7D &>.D-&5.a'!,!E!6-1)>@@F^LILm€ƒn^‚~‚k{a%9.`{za{{|€iA".&7"&a{zƒzƒ||

-9J6."4^aa^&.5A97^a a^s9A&A^ aam|zz~na7%>@9,@&AA.97^MLKm9I&6!&@na

 {‚ 97%^ a a^ 7,EA^ a a^ &^ a^ s "9!A97^ a m|zz{na !A&7"& 9+ !>|> 5D&@A 9D- 97A&7A&j&%.D&%6&"L7%@@9"&AA.7,!A&7"&9+!>|>5D&@A9D- 97A&7A&j&%.D&%6&"L7%@@9"&AA.7,a%9.`{za{{|‚ia|{a|{a}€€

9775^ a^ .,&I7.^ a^ s 5" @"&5^ a m|z{|na -& A>5."&9A96& A  D@,&D 9+ 79I&5 7D.DE69E@ %@E,Aa 341! !5'!62@ 14% '2-5!18^ JJm{{n^ ‚~kƒa %9.`{za{z}‚i7@%}‚|}

@!5&DM^ a^ E7,^ ^ &E^ a^ 9@M7&@^ a^ 5E!&4^ a^ E7Mj"-E,-@D^ a ^ b .@"-7&@^ a m|zz{na @.!5& !&Dj"D&7.7 &K>@&AA.97 .7 "959@&"D5 "7"&@A .7%."D&A DE69@ >@9,@&AA.97%@.I&7!LD-&DE69@&7I.@976&7Da1-!! ',%2-$3&!3'-,* !+8-$ '!,!2 -$ 3&! ,'3!  33!2 -$ +!1'^ RQm{‚n^ {z}€k€{a %9.`{za{z}i>7Aa{{€{z~ƒ‚

@&ED.,6^ a^ ,9^ a^ 954&^ ^ 5%6&.&@^ a^ [email protected]+9@.^ a^ s .7D&@^ a m|z{}na -& j!.7%.7, >@9D&.7 !+9K|` 7 &AA&7D.5 @&,E5D9@ 9+ j%@.I&7 5D&@7D.I& A>5.".7, 7%  6&%.D9@ 9+ "&55E5@ .7IA.97a ,-%!,!^ m 7E@Ln^ {k{{a %9.`{za{z}‚i97"a|z{}az

E&5D^a a^ 5.%%&7^a a^sD9@6^aam{ƒƒ~na@3^|ƒ}€k|ƒ}za

EAM"M4^ a^ s >@%5.7,^ a a m|zz€na -& @9A9>-.5 €‚  -&5."A& @&,E5D&A D@7A"@.>D.975 %&"D.ID.97 !L >@969D.7,  @&5&A& +@96 "-@96D.7a !,!2 H !5!*-/+!,3^KIm‚n^ƒk‚ƒa%9.`{za{{z{i,%a{}ƒ€}z€

EAM"M4^ a^ s >@%5.7,^ a a m|zz€!na -& @9A9>-.5 €‚  -&5."A& @&,E5D&A D@7A"@.>D.975 %&"D.ID.97 !L >@969D.7,  @&5&A& +@96 "-@96D.7a !,!2 H !5!*-/+!,3^KIm‚n^ƒk‚ƒa%9.`{za{{z{i,%a{}ƒ€}z€

5.7^ aa^E6.D@E^aa^-.6.ME^a^."-.^a^E>9^a^9"-^a^b@9"&^aam|zz|na @&?E&7D %&5&D.97A 7% %9J7j@&,E5D.97 9+ 6."@9j  ,&7&A 6.{ 7% 6.{€ D {}?{~ .7 "-@97." 5L6>-9"LD." 5&E4&6.a 1-!! ',%2 -$ 3&! 3'-,*  !+8 -$ '!,!2 -$ 3&! ,'3!  33!2 -$ +!1'^ RRm|~n^ {|~kƒa %9.`{za{z}i>7Aa|~|€z€ƒƒ

 {ƒ @&DD.^ a^"-.5DM^aa^.5J9@D-^a a^.%9I^a^-9^a^,@LM49^a^b@D9@&55.^a m|zz€na -&  -&5."A&A >€‚i>| 7% D-& 797"9%.7,   @& "9@&,E5D9@A 9+ L9 7% A4&5&D5 6EA"5& %.++&@&7D.D.97a !5!*-/+!,3* !**^ JJm~n^ ~k€za %9.`{za{z{€i3a%&I"&5a|zz€az‚azz}

@D&@^aa^.7^a^.E^aja^7,^a^s.E^ajam|z{zna-9A>-9@L5D&%>€‚-&5."A& "D.ID&A 7.5{ D@7A"@.>D.97 !L >@969D.7, { %.AA9".D.97 +@96 D-& 7.5{ >@969D&@a,-%!,!^KRm}ƒn^~|k}€a%9.`{za{z}‚i97"a|z{za|€

@ED-&@A^ aa^s" L^aam|zz|na &5."A&AD@E"DE@&7%6&"-7.A6a411!,3/','-, ', 314341* '-*-%8^ JKm{n^ {|}k}}a &D@.&I&% +@96 -DD>`iiJJJa7"!.a756a7.-a,9Ii>E!6&%i{{‚}ƒ~ƒƒ

EA&I."^a^ .A59>^a a^ &@79-7^aa^@&L^a^ L^a^D&&5&^a a^sE55&@j"&^ aam|zz{naI&@&K>@&AA.977%>95LjE!.?E.DL5D.979+D-&j!9K-&5."A&>€‚ .7"959@&"D5DE69E@Aa,-%!,!^KIm}n^}~k~}a%9.`{za{z}‚iA3a97"a{|z~ƒ€

-4@!@D.^a^ J7,^ a^7%@&A57"9^a^&.^a^E4#.B.7^a^9679^aja^b 7,^ a m|z{|na 5+ .7-.!.DA D-& &>.D-&5.5j6&A&7"-L65 D@7A.D.97 .7 666@L ,57% %&I&59>6&7D 7% !@&AD "7"&@ 6&DADA.A !L D@7A"@.>D.9755L @&>@&AA.7, 7.5|a 341!!**'-*-%8^JMm{{n^{|{|k{|||a%9.`{za{z}‚i7"!|€z

-&7^ a^s&.AA^aam|z{~na5D&@7D.I&A>5.".7,.7"7"&@`.6>5."D.97A+9@!.959,L7% D-&@>La,-%!,!^m"D9!&@|z{}n^{k{~a%9.`{za{z}‚i97"a|z{}az

-&7^a^s@6.,.7.^ am|zzna&Dj75LA.A9+{7%|>&7&D@7"&a -41,*-$ *','*,-*-%8T?$$''* -41,*-$3&!+!1',-'!38-$*','*,-*-%8^KNm{{n^ {}|ƒk}}a%9.`{za{|zzi a|zz€azƒa{z€€

-.7^ a^&@.&A^a^@.%5L7%^ a^>&5567^aa^9L%A,E>D^a^ E9^aja^b @L^ aa m|zz€na &796." 7% D@7A"@.>D.975 !&@@D.97A 5.74&% D9 !@&AD "7"&@ >D-9>-LA.959,.&Aa,!1!**^JIm€n^|ƒk~{a%9.`{za{z{€i3a""@a|zz€a{zazzƒ

-9.^aj a^s&&^aj am|z{|na-&j!9K-&5."A&}.7D&@"DAJ.D-^"9j 59"5.M&AJ.D-.D.7D-&"LD9>5A6%[email protected],D-& |i>-A&9+D-&"L"5&^7%++&"DA.DA

 {‚z A-EDD5.7, %[email protected], 6 &K>9@Da -41,* -$ !**4*1 '-&!+'2318^ JJLm}n^ ƒ‚kƒ€a %9.`{za{zz|i3"!a|}~|‚

.&>5L^a^.5&L^a^.+&@^aa^.%6&L&@^ a^%%.A97^ aa^ I79I^a^b@.A"-^aa m|z{|na E>>@&AA.97 9+ D-& &>.D-&5.5j6&A&7"-L65 D@7A.D.97 !L @.7L-&%j5.4&j|a ,!1!2!1&^PKmƒn^|~~zk}a%9.`{za{{‚izzz‚j~|aj{{j~z}‚

5@4^ a a^ 9E5A97^ a^ 5,5.&A-^ a^ 37^ a^ ."95^ a a^ 5&6.7,^ a^ b 9!A97^ a a m|zz‚na -&  -&5."A& >€‚ .A  79I&5 7%@9,&7 @&"&>D9@ "9"D.ID9@ .7I95I&% .7 A>5.".7, 7% .A 9I&@&K>@&AA&% .7 >@9ADD& "7"&@a ,!1 !2!1&^ OQm{ƒn^ ƒ}‚k~€a %9.`{za{{‚izzz‚j~|ajz‚jzƒ}|

5@4^aa^ %3.6."-&5^a^&6>&@5&L^a^@7@%^a^E55&@j"&^a^s9!A97^aa m|z{}na >€‚i% AE>>9@DA Xj"D&7.7 s   %[email protected], 7%@9,&7 @&"&>D9@ 6&%.D&% D@7A"@.>D.97 .7 >@9ADD& "7"&@a *- ,!^ Qm{n^ &~{za %9.`{za{}{i39E@75a>97&azz~{z

59ED.&@^ a a^ ^ a a^ ,EL&7^ a a^ s @7^ a a m|z{|na -& j!9K  -&5."A& !>| "977&"DA  ?E5.DL "97D@95 J.D- @&>@&AA.97 9+ !&@@7D D@7A"@.>D.97a &! -41,*-$'-*-%'*&!+'2318^KQPm}{n^|€{k€€a%9.`{za{z~i3!"a{{|a}‚}z

9@%.7^a^ -7^a^s&,,A^ aam|z{|naD@E"DE@&^+E7"D.977%@&,E5D.979+A>5."&9A965  -&5."A&Aa 411!,3 /','-, ', !** '-*-%8^ KMm}n^ ~}{k‚a %9.`{za{z{€i3a"&!a|z{|az}azz~

9@%.7^a^77&@^a a^.7%&@^a^s7@9?E&A^ am|zz~na-&7&J5L%.A"9I&@&%69D.+9+ j!9K  -&5."A&A @&,E5D&A j!.7%.7, 7% -&5."A& "D.I.DL^ KLm{}n^ |~‚k |~‚a%9.`{za{z}‚iA3a&6!93a€zz||

9@7&L^aa^ J7,^aj a^D9A9^a^9,D^a^5&A4&7j.4.D.7^a^ 9%J.7^a a^b.4.D.7^ a a m|z{zna @&?E&7D %9J7@&,E5D.97 9+ 6.j}~ +6.5L .7 -E67 9I@.7 "7"&@Aa *','* ,!1 !2!1&T? , $$''* -41,* -$ 3&! +!1', 22-'3'-, $-1 ,!1 !2!1&^JOm~n^{{{ƒk|‚a%9.`{za{{‚i{z‚jz~}|ajzƒj|€~|

9AD7M.^a^s&-@A97^ aam{ƒƒ‚na9++&65&6665A^OKQm{ƒƒn^{ƒƒk{ƒƒƒa

 {‚{ @&>>&^a^ 7."-^a^7D@.89^a^9,E&@^a^5&@9^a^EAE5(7^a^bEA"-!&"4^a m|z{|na "@9 |{ @&,E5D&A D-& !57"& !&DJ&&7 A&5+j@&7&J5 7% %.++&@&7D.D.97 "966.D6&7D.7&6!@L97."7%%E5DAD&6"&55Aa-*!4*1, !**4*1'-*-%8^LKm‚n^ {~~|k|a%9.`{za{{|‚iaz€}|}j{{

@7&55^aa^ &5&7^a^5&^ a^s@49^am|zznaI95ED.979+9Ij&>&7%&7D>5.".7, &,E5D.97.7D-&@.7^Lm{zna%9.`{za{}{i39E@75a>,&7azz}z{}

I.%^ a a^ 9L7&^ a a^ .55-9EA&^ a a^ s 75&L^ a a m|z{{na "D.ID.97 D-@9E,- @&"@E.D6&7D9+|€ˆ@>{ƒ"96>5&K-&>95L6&@A& [email protected]5%96.7 >@969D&AA>5.".7,"D.ID.97D-@9E,-@&"@E.D6&7D9+|€k@>{ƒ"96>5&K^ƒ|k ƒ‚}a%9.`{za{{z{i,%a|z}‚z{{

I.%^ a a^ -&7^ a^ AA7-^ a^ 7955^ a^ s 75&L^ a a m|z{zna 7 >@9D&.7A "97D@955&% !L "jL" %&@&,E5D& >L@EID& 4.7A& 6 A>5.".7, .7 "7"&@a 341!^ MOLm|ƒn^}€~k‚a%9.`{za{z}‚i7DE@&z‚€ƒ

I.A^ a a^ .5L@%^ a a^ ,7^ a^ s D^ a m|zz‚na  >@9D&.7A "97D@95  j 6&%.D&%6."@96DE@D.97a341!^MNMm|zzn^€k€{a%9.`{za{z}‚i7DE@&zz‚€

&L^ a^ s .E^ aja m|z{|na -9A>-9@L5D.97 9+ >€‚  -&5."A& !L >}‚  4.7A& "97D@.!ED&A D9 "9597 "7"&@ "&55A >9>D9A.A .7%E"&% !L 9K5.>5D.7a  !** '-*-%8^ JLm{n^|a%9.`{za{{‚€i{~{j|{|{j{}j|

ED&@D@&^a^ @D%9E^a^@%&77&^a^ &@677^a^67^a^.%&@&E^a^bE!9&E+^ a m|z{zna AD@9,&7 @&,E5D.97 7% >-LA.9>D-959,." A.,7.+."7"& 9+ 5D&@7D.I& >@969D&@A .7 !@&AD "7"&@a ,!1 !2!1&^ PImƒn^ }€zkza %9.`{za{{‚izzz‚j ~|ajzƒj}ƒ‚‚

5&7!A^ a m|zz{na E67 !@&AD "7"&@ "&55A ,&7&@D&% !L 97"9,&7." D@7A+9@6D.97 9+ >@.6@L 666@L &>.D-&5.5 "&55Aa !,!2 H !5!*-/+!,3^ JNm{n^ zk€a %9.`{za{{z{i,%a‚|‚ƒz{

7%9-^ a^ @EL6^ a^ AE-.@9^ a^ 9!LA-.^ a^ 9D9^ a^ .^ a^ b A-.69D9^ a m{ƒƒƒnaE@.+."D.977% %&7D.+."D.979+>€‚ &5."A&"D.7,A@7A"@.>D.975

 {‚| 9"D.ID9@ >&".+." +9@ D-& "D.ID.97 E7"D.97 { 9+ E67 AD@9,&7 &"&>D9@ W E@.+."D.97 7% %&7D.+."D.97 9+ >€‚  &5."A& "D.7, A  @7A"@.>D.975 9"D.ID9@>&a

.@67jJ.55.6A^aa^s E&7D-&@^am|z{zna{7%|-&5."A&A’`+6.5L6DD&@A^}{}k }|~a%9.`{za{z{€i3aA!.a|z{zaz}az{{

5!&@,^a^ EI&7^ a^I"-&749^a^I59I^a^ A-E!^a^M&4&5L^a^s 5&.7^ am|z{|na -& @"-.D&"DE@& 9+ +.!@9!5AD 69795L&@A 9+ %.++&@&7D 9@.,.7 %.++&@&7D.55L .7+5E&7"&A DE69@ "&55 ,@9JD-a ,3!1,3'-,* -41,* -$ ,!1@ -41,* ,3!1,3'-,* 4 ,!1^ JLJm{zn^||~k‚}a%9.`{za{zz|i.3"a||{

95467^ am|zzna7,.9,&7&A.A`79@,7.M.7,>@.7".>5&+9@%@E,%.A"9I&@L]341!!5'!62@ 14%'2-5!18^Om~n^|}k‚€a%9.`{za{z}‚i7@%|{{

@.&%67^aa^@-^ a aj a^E@,&^aa^s@D&5^aam|zzƒna9AD6665.76A @& "97A&@I&% D@,&DA 9+ 6."@9Aa !,-+! !2!1&^ JRm{n^ ƒ|k{za %9.`{za{{z{i,@az‚|z{a{z‚

E4E%^a^6,D^ a^E3.L6^a^DAE69D9^a^ 9A-.%^ a^9A-.6E@^ a^b D9^a m|zzna j!9K  -&5."A& AE!E7.DA 9+ D-& @9A- "96>5&K @& @&?E.@&% +9@ >@9"&AA.7, 9+ @ 7%  AE!A&D 9+ 6."@9Aa 341! !** '-*-%8^ Rmn^ €z~k{{a %9.`{za{z}‚i7"!{

E55&@j"&^aam|z{}na-&!9K>@9D&.7Am>€‚n7%{m>|n`E5D.jDA4.7, D@7A"@.>D.975 @&,E5D9@Aa '-&'+' !3 '-/&82' 3^ Na %9.`{za{z{€i3a!!,@6a|z{}az}azz~

E55&@j"&^a^s5.^am|zz‚na-&!9K-&5."A&A>€‚m%Kn7%>|m%K{n` 79I&5D@7A"@.>D.975"9j@&,E5D9@Aa'-&!+'*-'!381,23'-,2^LOmD~n^€zƒk{|a %9.`{za{z~|iz}€z€zƒ

6!5&^a a^@.MM&55^ aa^7,^a^ @.A-74E6@^a^s @EA^aam|z{zna-&-.AD97& I@.7D6"@9 |{6@4A@&>@&AA&%ED9A965"-@96D.7^!ED>@9D&"DAAE!A&D9+.DA

 {‚} D@,&D ,&7&A +@96 A.5&7".7,a !,!2 H !5!*-/+!,3^ KMm{n^ |{k}|a %9.`{za{{z{i,%a{‚€{{z

&.C5&@^a^5D6&L&@^a^D&.7^a^-5677j"-.++5&@^ a^sD-5^ am|z{}na-&-&5."A& %Ki>€‚!.7%AD9->+}7%&7-7"&A9+%K{i>|7%6,6a4*!' ' 2!2!1&^MJm{€n^‚k‚‚a%9.`{za{zƒ}i7@i,4D}‚

&@677^a^ @D%9E^a^97D^a^@%&77&^a^ E%.7&E^a^9E,'@&^a^bE!9&E+^ a m|z{|na E5 @95& 9+ D-& %%Ki%%K{  -&5."A&A .7 D-& "97D@95 9+ D-& >@9j 6.,@D9@L  D@7A"@.>D.97 +"D9@a ,-%!,!^ LJm~|n^ ~}€k~ƒa %9.`{za{z}‚i97"a|z{{a€{‚

-.,7^a^ .9@%79^a^-&7^ a^&7I&7ED9^a^AD.,5.97.^a^969,5.9^aa^b.697D.^ am|zzna&5569D.5.DL.A"97D@955&%!L|iD-@9E,-5D&@7D.I&A>5.".7,9+D-&97 >@9D997"9,&7&a-*!4*1!**^KIm€n^‚‚{kƒza%9.`{za{z{€i3a695"&5a|zza{zaz|€

9AJ6.^a^-.5.>>@^a^E7^a^DA.59E^a^I@-6^ a^7,^a^b97%&&5.A^ aa m|zzƒna %&7D.+."D.97 9+ .7IA.97 A>&".+." A>5."& I@.7DA 9+ D-& "LD9A4&5&D5 >@9D&.7 &7>@&A&7D.7666@LDE69@"&55A%[email protected],.7IA.97.7I.I9a*','*H7/!1'+!,3* !3232'2^KOm|n^{}kƒa%9.`{za{zziA{z‚jzz‚jƒ||j‚

@&,9@L^a^&@D^a a^D&@A97^aa^@@L^aa^AL4.7^a^@A-.%^ a^b 99%55^ a a m|zz‚na-&6.j|zz+6.5L7%6.j|z@&,E5D&&>.D-&5.5D96&A&7"-L65D@7A.D.97 !LD@,&D.7,{7% {a341!!**'-*-%8^JImn^ƒ}k€z{a%9.`{za{z}‚i7"!{||

E.5^ a^ DD97.^ a^ D(I&7.7^ a^ "-j&5.A^ a^ s D&^ a m|zz}na &5."A& .7 "j j @A 5D&@7D.I&>5.".7,&,E5D.9795&A9+-7{^@9D&.7A^7%>€‚ &5."A&.7"j j@A5D&@7D.I&>5.".7,&,E5D.979a%9.`{za{{|‚ia|}a‚a|ƒ|

E9^a^.^a^97,^a^ E^a^-E^a^E^a^s7,^am|z{|naY{%9J7j@&,E5D&AjW &K>@&AA.977%"97+&@AD69K.+&7@&A.AD7"&.7!@&AD"7"&@a*-,!^Pm{|n^&|}‚za %9.`{za{}{i39E@75a>97&azz|}‚z

 {‚~ -7^aa^9E7D&@^aa^E7%!&@,^a^&.3&@A!&@,&7^aa^@994A^aa^s&.7!&@,^ a a m{ƒƒƒna @&D.97 9+ -E67 DE69E@ "&55A J.D- %&+.7&% ,&7&D." &5&6&7DAa 341!^ MIIm€~}n^~€~k‚a%9.`{za{z}‚i||‚z

7^a^&-@^a^-7A&4@7^aa^E^ a^&797^a^97.,@9^a a^b-.77.L7^aa m|zz‚na  +5E9@&A"&7"& .7 A.DE -L!@.%.MD.97 A"@&&7 +9@ |€ D@7A+9@6D.97jA>&".+." !&@@D.97A` .%&7D.+."D.97 9+ j~ +EA.97 >@9D&.7 .7 >@9ADD& "7"&@a ,!1 !2!1&^OQm{‚n^€|ƒk}a%9.`{za{{‚izzz‚j~|ajz‚j|z{~

7-7^a^s9EAA&7A^aam|z{|na""&AA9@.&AD9D-&"@.6&`+E7"D.97A9+"&55A@&"@E.D&% D9 D-& DE69@ 6."@9&7I.@976&7Da ,!1 !**^ KJm}n^ }zƒk||a %9.`{za{z{€i3a""@a|z{|az|az||

7-7^ a^ s &.7!&@,^ a a m|z{{na 556@4A 9+ "7"&@` D-& 7&KD ,&7&@D.97a !**^ JMMmn^€~€k€~a&D@.&I&%+@96-DD>`iiJJJa7"!.a756a7.-a,9Ii>E!6&%i|{}€|}z

7-7^ a^ &.7!&@,^ a a^ s @7".A"9^ a m|zzzna -& 556@4A 9+ 7"&@ &I.&J [email protected]9+5.+9@7.D7@7".A"9^JII^kza

@D6ED-^ a^@5E!^ a^9@759"-&@^ aja^.55^aa^ &7DM&5^a^.56^a^sH-@677^a m|zz|na @9D&.7 "96>9A.D.97 9+ -E67 >@&A>5."&9A96&A .A95D&% !L  D9!@6L".7 ++.7.DLjA&5&"D.97 6&D-9%a 1-!! ',%2 -$ 3&! 3'-,*  !+8 -$ '!,!2 -$ 3&! ,'3! 33!2-$+!1'^RRm|€n^{€{ƒk|~a%9.`{za{z}i>7Aa|€|~‚}‚ƒƒ

&,&5&^a^ 6!E@9I^a^ @9AA677^a^[email protected]^a^9J@9^a^&.677^a^bD&5M5^a m|z{|na L76." >@9D&.7j>@9D&.7 .7D&@"D.97 [email protected], 9+ D-& -E67 A>5."&9A96&a -*!4*1!**^MNm~n^€k‚za%9.`{za{z{€i3a695"&5a|z{{a{|az}~

&5."A&^aaaaa^ <7.,^a^E!9&E+^a^@4&@^aa^&@D^a^55&L^a^b55&L^ a a a m|zz|na &,E5D.97 9+ 5D&@7D.I& >5.".7, !L D-& &,E5D.97 9+ 5D&@7D.I& >5.".7, !L D-& j&>&7%&7D j9K  &5."A& >| @7% 9a %9.`{za{{|‚ia||a{€a€ƒ‚

9"-&%5.7,&@^ a^ s 5D-^ a m|zzƒna >.,&7&D." @&>@9,@66.7, 7% .7%E"&% >5E@.>9D&7"La !5!*-/+!,3E+1' %!>,%*, F^JLOm~n^zƒk|}a%9.`{za{|~|i%&Iaz|z‚€

 {‚ 97,^a^9-^ a^-&7^ a^%.^a^7^a a^E^aja^b E7,^am|z{}na.,75.7,!L >}‚ D.6E5D&AE"5&@9"5.MD.979+D-&."@9>@9"&AA9@96>97&7D>€‚+9@ @9"&AA.7, 9+ &5&"D&% @.6@L ."@9Aa '!,! '%,*',%^ Om|€€n^ @{€a %9.`{za{{|€iA".A.,75a|zz}z€

9@.,E"-.^ a^469D9^ a^ 9.7E6^a^&6!^ a^ 79E&^^ 79E&^a^b.D9-^am|z{|na  jX%@.I&A&>.D-&5.5j6&A&7"-L65D@7A.D.97D-@9E,-Y{j6&%.D&%%9J7@&,E5D.97 9+a,-%!,!^LJm|€n^}{ƒzk|z{a%9.`{za{z}‚i97"a|z{{a~ƒ}

9I-77.AL7^ a a^ s @AD&7A^ a a m|zzna  9I&5 7D@97." ".A 5&6&7D ^  i j} ^ &,E5D&A D .!@9!5AD @9JD- "D9@ &"&>D9@ | >5.".7, D-@9E,- "D.ID.97 9+ 7 >AD@&6 K97 7% &>@&AA.97 9+  9J7AD@&6 K97 97D.7.7,  97"797."5 @7"-9.7D&?E&7"&9I&5 7D@97."".A5&6&7Da%9.`{za{{|‚ia|a{a|z

E^a^9^ a^@@955^a a^&@&69J."M^a^-&7^ a^@@A"9^a^b95L4^ am|zz‚na &,E5D.97 9+ .7 A.DE D9 .7IA.I& !@&AD "@".796 D@7A.D.97a ,!1 !**^ JLmn^ }ƒ~k ~z€a%9.`{za{z{€i3a""@a|zz‚az}azz

E^a^"%975%^aa^ E&DD7&@^aa^&7,^a^5?^ aa^"-J@M^ a a^b7,^a m|zzƒna6.j|zz6."@9"5EAD&@A>@9,79AD."6@4&@.7%I7"&%9I@.7"7"&@a 8,!-*-%',-*-%8^JJMm}n^~k€~a%9.`{za{z{€i3aL,L79a|zzƒazaz||

E7,^a^.^a^9^a^.7^aj a^.7^ aja^.7,^a^b7,^ am|z{|na&%.D9@"96>5&K @&,E5D&A 5D&@7D.I& 6 >@9"&AA.7, I. D-& |} AE!E7.Da -*!4*1 !**^ MNm~n^ ~ƒk€ƒa%9.`{za{z{€i3a695"&5a|z{{a{|az||

E&5,^aa^E^aa^@795%^ aa^.7,^aa^.E^aa^7^aa^b&9^ aam|z{|na 7D&,@D.I&,&796&jJ.%&75LA.A@&I&5A"99>&@D.I&@&,E5D.979+5D&@7D.I&A>5.".7, !L-7>@9D&.7Aa!**!/-132^Jm|n^{€k‚a%9.`{za{z{€i3a"&5@&>a|z{|az|azz{

,,9^aa^ 6.&A97^a a^"7&.55^aa^9ED-,D&^ a^">-&D^ a^s7&D^aam{ƒƒ{na >€‚  &5."A&’` %&7D.+."D.97 9+  E"5&95@ 9@6 7% 597.7, 9+ &5D&% &7&A 97D.7.7,97A&@I&% 7D@97.7&ADAa%9.`{za{{|‚ia{{a}a{}|€a>%D&%

 {‚€ A@&5A&7^ a a^ LD97^ a a^ I.%A97^ a a^ .A4&^ a a^ 9A.9A^ a a^ &55.7,&@^ a^ b 7%&@ &.%&7^ a a m|z{}na  | .A9+9@6jA>&".+." %&5&D.97 @&I&5A  %.++&@&7D.5 @&?E.@&6&7D +9@ >L@EID& 4.7A& .7 DE69@ "&55Aa !**^ JNNm|n^ }ƒk~zƒa %9.`{za{z{€i3a"&55a|z{}azƒaz|

"9!A^jaa^."95^aa^ .A59>^a a^ ++@L^a a^ L^aa^sE55&@j"&^aam|zzna 69%.+."D.979+D-&!9K>@9D&.7>€‚69%E5D&A.DAD@7A"@.>D.975"D.I.DL 7% >@969D&A .DA .7D&@"D.97 J.D- {a ,-%!,!^ KOm~zn^ ‚€€k€a %9.`{za{z}‚iA3a97"a{|{z}‚

55^a^-5677j"-.++5&@^ a^sD-5^ am|zzna&%E7%7D@95&9+!9K>@9D&.7A>€‚ m%Kn7%>|i>‚|m%K{n.7@.!9A96&!.9,&7&A.A7%"&55>@95.+&@D.97a4*!'' 2 !2!1&^LNm{{n^}ƒzk€z{a%9.`{za{zƒ}i7@i,46z‚

747&"-D^am|z{znaE5D.jD5&7D&%j!9K>@9D&.7A7%>9D&7D.5DE69@>@969D&@A`>€‚ -&5."A&mn7%.DA>@59,^>|-&5."A&m{na +!1', -41,*-$ 1,2*3'-,* !2!1&^ Km}n^ ||}k}~a &D@.&I&% +@96 -DD>`iiJJJa>E!6&%"&7D@5a7.-a,9Ii@D."5&@&7%&@a+",.]@D.%‰|‚ƒ|~z}sD995‰>6"&7D@&M s@&7%&@DL>&‰!AD@"D

&7A&7^aa^.E^a^@&DD.^ a^."95^aa^&>5LE4^a^D&.7^ aa^b&AD&7%9@+^ a a m|zz‚na >€‚ m%Kn .7D&@"DA J.D- E7K| 7% @&,E5D&A 9AD&9!5AD %.++&@&7D.D.97a -41,*-$!**4*1'-&!+'2318^JILmn^{~}‚k{a%9.`{za{zz|i3"!a|{|€

55E@.^ a^ s &.7!&@,^ a a m|zzƒna &I.&J A&@.&A -& !A."A 9+ &>.D-&5.5j6&A&7"-L65 D@7A.D.97^JJRm€na%9.`{za{{|i  }ƒ{z~a{~|z

>99@^a^ 95%!&@,^aa^E6!&@57%^a a^D74E6@^ a^&,E@^aa^67E&5^aa^ b &@7AD&.7^ a m|z{zna -& -.AD97& I@.7D 6"@9 | AE>>@&AA&A 6&5796 >@9,@&AA.97 D-@9E,- @&,E5D.97 9+  ‚a 341!^ MOQm}|n^ {{zkƒa %9.`{za{z}‚i7DE@&zƒƒz

@^a^EA-.6.^ a^-9E^a^L^a^-.^a^-&7^a^bE^ aam|z{{na-&5."A&>€‚ mn @&,E5D&A DE &K97 {z A>5.".7, !L 69%E5D.7,  AD&6j599> AD@E"DE@& D D-& d A>5."&A.D&a-*!4*1, !**4*1'-*-%8^LJmƒn^{‚{|k|{a%9.`{za{{|‚iaz{{~ƒj{z  {‚ &.@A&!.5"4^a^977&^a^D&A^ a^ &7,&5^ a7^955&D^a^&L795%A^a^s9L^a7a m{ƒƒ‚na95&"E5@597.7,9+D-& E67>{|z"D7D&7.7 &7&m{n`K>@&AA.97 9+E5D.>5&5D&@7D.I&5L>5."&% A9+9@6A^JMO^{|ƒk{~€a

-9%9@^aa^9%@.,E&M^ a^!@EMM.^ aa^7,^aa^ .^ aaa^s9A!A-^am|z{{na A"&7DjA&? .7%."D&A J.%&A>@&% "9D@7A"@.>D.975 >@&j6 A>5.".7, .7 @9A9>-.5^ |z|k|{|a%9.`{za{{z{i,%a{‚ƒ€|a{{{a

9@7!5.-DD^ a a^ "-9@^ a a^ 55:^ a^ E3@%.7^ a^ [email protected]^ a^ s E89M^ a a m|z{}na 5D&@7D.I& A>5.".7,`  >.I9D5 AD&> !&DJ&&7 &E4@L9D." D@7A"@.>D.97 7% D@7A5D.97a 341!!5'!62@-*!4*1!**'-*-%8^JMm}n^{}k€a%9.`{za{z}‚i7@6}|

@95^ a^ 9&%.,&^ a^ s .5.>9J."M^ a m|z{zna -& J.%&A>@&% @&,E5D.97 9+ 6."@9 !.9,&7&A.A^ +E7"D.97 7% %&"La 341! !5'!62@ !,!3'2^ JJmƒn^ ƒk€{za %9.`{za{z}‚i7@,|‚~}

E6@^ a a^ E^ a^ &@"&@^ a a^ 95E!^ a a^ s "4A^ a m|zzna 6>.@&% 6."@9 >@9"&AA.7,&7-7"&A"&55E5@D@7A+9@6D.977%DE69@.,&7&A.Aa341!!,!3'2^LRmn^ €}ka%9.`{za{z}‚i7,|zz}

EADDA"-&@^ a^ 9D-9@7^ a^ E,.&EK^ a^ "-&++M&4^ a^ s %E@7&@^ a a m|zzna >5.".7, @&,E5D&A6&D!95.D&!.7%.7,D9-.AD97&6"@9 |a341!314341*H-*!4*1 '-*-%8^JKmn^€|~ka%9.`{za{z}‚i7A6!ƒ€

66^ aa^."95^aa^E55&@j"&^a^s697%^ am{ƒƒ€na>|`-E677E"5&@  !9K >@9D&.7 -.,-5L @&5D&% D9 >€‚a 4*!' ' 2 !2!1&^ KMm{ƒn^ }}ƒk~a &D@.&I&% +@96 -DD>`iiJJJa>E!6&%"&7D@5a7.-a,9Ii@D."5&@&7%&@a+",.]@D.%‰{~€{€‚sD995‰>6"&7D@&Ms @&7%&@DL>&‰!AD@"D

>E4^a^@@^ a^ 44E5^a^&%@9^ a^-DD"-@L^a^E@%96^a^b @L^ aam|z{zna K97j5&I&56."@9@@L75LA&A.%&7D.+L5D&@7D.I&A>5.".7,>@9,@6A.7!@&AD"7"&@a -*!4*1,!1!2!1&T?^Qmn^ƒ€{k~a%9.`{za{{‚i{~{j‚€ajzƒjz|‚

 {‚‚ @@.!^ a a^ @D/7j.55@^ a^ @"/^ a a^ &@&.@^ a^ &8^ a^ %& &@@&@9A^ a a^ b E89M^am|zzƒna7.5|"99>&@D&AJ.D-7.5{.7D-&@&>@&AA.979+I.D6.7@&"&>D9@ .7"9597"7"&@a1',-%!,!2'2^LIm‚n^{~ƒk€‚a%9.`{za{zƒ}i"@".7i!,>{~z

E@&7D^aja^E@&E^a^ 5&.7^aa^@9EA5@%^a^ A7.&@^a^[email protected],^a^s@.&^ am|z{|na &J +E7"D.97 +9@ D-&  -&5."A& >€‚i A  69%.+.&@ 9+ { "D.I.DL 97 &K>7%&%   @&>&DAa 4*!' ' 2 !2!1&^ MImn^ }{ƒk{a %9.`{za{zƒ}i7@i,4@{||‚

M@9I^a^ E!9^a^.^a^3&&^a^@ED7.^a^.7^a^b -I@.^aam|zz|na ~ "9&K>@&AA.97 J.D- A ,&7&@D&A 65.,77D -E67 &>.%&@65 DE69@.,&7&A.Aa 341! ! '',!^Qm{zn^{{zk{~a%9.`{za{z}‚i76ƒ

&&^ a^ .6^ a^ 7^ a^ &96^ aj a^ &&^ a^ &4^ a a^ s .6^ a a m|zz~na ."@9 ,&7&A @& D@7A"@.!&% !L  >95L6&@A& a &!  -41,*^ KLm|zn^ ~z{k€za %9.`{za{z}‚iA3a&6!93a€zz}‚

.^a^ E7,^ a^-&7^a^3&@^aa^&5A97^aa^@AJ@^ a^b EJ%^a am|z{|na -& DL>."5 -.AD97& 6"@9 |{a| .7D&@"DA J.D- j| >@9D&.7 .7 "7"&@ "&55Aa &! -41,*-$'-*-%'*&!+'2318^KQPm|n^|}{{k‚}a%9.`{za{z~i3!"a{{|a}ƒ~{|

.7^a^7,^a^7,^ a a^ E7,^a^s.E^am|zznaA&i &5."A&"D.I.D.&A9+>€‚ &5."A&@&&?E.@&%+9@@&j6>5.".7,!ED9D+9@AA&6!5L9+D-&>5."&9A96& A&i &5."A&"D.I.D.&A9+>€‚ &5."A&@&&?E.@&%+9@@&j6>5.".7, !ED9D+9@AA&6!5L9+D-&>5."&9A96&a%9.`{za{{|‚ia|a{a~‚~

.E^a^E^ a^E7^a^-7,^ a^.&^a^-E^ a^b-&7,^am|zz‚na6.j{€+6.5L.7%E"&A "&55"L"5&@@&AD!L@&,E5D.7,6E5D.>5&"&55"L"5&,&7&Aa4*!'' 2!2!1&^LOm{€n^ }ƒ{k~z~a%9.`{za{zƒ}i7@i,47||

.E^am|zz|na>€‚ &5."A& A7AA&7D.5 E67>5.".7,"D9@-D"DADD-&{ A7j ‘ >5."& .D& E>5&K >€‚  &5."A& A 7 AA&7D.5 E67 >5.".7, "D9@ -D"DADD-&{A7j[>5."&.D&E>5&K^KKm{na%9.`{za{{|‚ia||a{a~~}

 {‚ƒ 9I".^aa^ -7&6^a^@@^ a^@795%^ a^ &&^a^@@^a^b&9^ aam|z{}na!+9K >@9D&.7A @&,E5D& 5D&@7D.I& 6 A>5.".7, D-@9E,- &I95ED.97@.5L "97A&@I&%  !@.%,&Aa 341! 314341* H -*!4*1 '-*-%8^ KIm{|n^ {~}~k~|a %9.`{za{z}‚i7A6!a|€ƒƒ

E^ a^ &DM^ a^.A4^a^5I@&MjI&%@^a^6!^ a^&"4^a^b 95E!^aam|zzna ."@9 &K>@&AA.97 >@9+.5&A "5AA.+L -E67 "7"&@Aa 341!^ MLNmz~}n^ ‚}~k‚a %9.`{za{z}‚i7DE@&z}z|

E"9^aa^559^a^"-9@^ aa^ 9@7!5.-DD^aa^s.AD&5.^am|z{zna&I.&J>.,&7&D."A.7 5D&@7D.I&@&j6>5.".7,a!**^JMMm{n^{€k|€a%9.`{za{z{€i3a"&55a|z{za{{az€

E"9^aa^7^a^96.7,^ a^5&7"9J&^a a^&@&.@j6.D-^aa^s.AD&5.^am|z{zna &,E5D.97 9+ 5D&@7D.I& A>5.".7, !L -.AD97& 69%.+."D.97Aa '!,! E!6 -1)> @@F^ LKPmƒ€‚n^ƒƒ€k{zzza%9.`{za{{|€iA".&7"&a{{‚~|z‚

E7%^ a^ s -5!&@,^ a a m|zz€na E!AD@D& A&5&"D.I.DL 9+ &K>[email protected]  7% ."&@ .7 D-& !.9,&7&A.A9+6."@9Aa-* /1',% 1-18+/-2'-,4,3'33'5!'-*-%8^PJ^ƒk €€a%9.`{za{{z{iA?!a|zz€a{azz

E7%!&@,^aa^7%&55^a a^D&J@D^a^5&7!A^a^ @DJ&55^ a^@994A^aa^b -7^ a a m|zz|na 669@D5.MD.97 7% D@7A+9@6D.97 9+ >@.6@L -E67 .@JL &>.D-&5.5 "&55A !L ,&7& D@7A+&@a ,-%!,!^ KJm|ƒn^ ~k‚€a %9.`{za{z}‚iA3a97"a{|zz

L49^am|zzna&D-L5D.979+ E67."@9 &7&A.79@657%&9>5AD."&55A  &LJ9@%A  ^mLn^{zz{k{zza

^a^9E7,^ a^@!-5^ a^7^a^&AD%,-^a^ED-^a^b&.7!&@,^aam|z{zna 6.jƒ^ij"D.ID&%6."@9^@&,E5D&Aj"%[email protected]7%"7"&@6&DADA.Aa 341!!**'-*-%8^JKm}n^|~k€a%9.`{za{z}‚i7"!|z|~

7.^ a ^ E9^ a^ .9^ aj a^ D97^ a a^ LL77^ a^ -9E^ a a^ b &.7!&@,^ a a m|zz‚na-&&>.D-&5.5j6&A&7"-L65D@7A.D.97,&7&@D&A"&55AJ.D->@9>&@D.&A9+AD&6 "&55Aa!**^JLLm~n^z~k{a%9.`{za{z{€i3a"&55a|zz‚az}az|

 {ƒz D9A^a^s^a am|zzƒna7D,97.AD.">@9D&.7A@&,E5D&5D&@7D.I&A>5.".7,9+DE69@j @&5D&% "{! %9J7AD@&6 9+ D-&  }j4.7A& 7% 7D >D-JLA^ JQm{ƒn^ }€ƒ€k}za %9.`{za{zƒ}i-6,i%%>}{

ME@&4^ a^ E9^ a^ @A7.DM^ a^ ."4A^ a^ 9J&@A^ a a^ s D.5567^ a m|z{|na  @&,E5D&A  @&>5."D.97 7% .A @&?E.@&% +9@ "&55 >@95.+&@D.97 .7  AE!A&D 9+ !@&AD "7"&@"&55Aa,!1'2-5!18^Kmƒn^‚{|k|a%9.`{za{{‚i|{ƒj‚|ƒzaj{|jz{{€

.^ a^.E^a^!!7.^aa^7,^a^@!7^ a a^E55&7,&@^a^s5@L^aam|z{zna 7 I.I9 A&5&"D.97 9+ DE69@jD@,&D.7,  69D.+Aa 341! &!+'* '-*-%8^ Om{n^ ||k~a %9.`{za{z}‚i7"-&6!.9a|

.E@^ a^E3.!E"-.^a^s779^am|z{|na   ^KIJK^|{|k|{a

997&L^aa^ 9&5^a^dAA9@9^aa^5.A!E@L^ aa^s 747&"-D^am|z{zna5&.9D@9>." &++&"DA 9+ >}zzj6&%.D&% "&DL5D.97 97 >€‚ 7% >|  -&5."A&a &! -41,* -$ '-*-%'*&!+'2318^KQNm~zn^}z~~}k|a%9.`{za{z~i3!"a{{za{~}ƒ|

997&L^aa^ @7%&^ aa^5.A!E@L^ aa^s 747&"-D^am|z{znaE69L5D.979+>€‚7% >|-&5."A&A++&"DA>@9D&.7AD!.5.DL7%D@7A"D.ID.97>9D&7D.5a'-&!+'2318^ MRm{n^{k{za%9.`{za{z|{i!.ƒz{|€}6

9@&5^aja^.'I@&^a^-96A^a^ .745^ a^7A.&E^a^sE.A.&EK^am|zz‚na &7&@D.97 9+!@&AD"7"&@AD&6"&55AD-@9E,-&>.D-&5.5j6&A&7"-L65D@7A.D.97a*-,!^Lm‚n^ &|‚‚‚a%9.`{za{}{i39E@75a>97&azzz|‚‚‚

9@&79jE&79^ a^ E!.559^ a^ @@.:^ a^ &.7%9^ a^ 9%@/,E&Mj.7.55^ a a^ .55^ a^ b 79^ a m|zz€na &7&D." >@9+.5.7, 9+ &>.D-&5.5 "&55A &K>@&AA.7, j"%[email protected] @&>@&AA9@A @&I&5A  %.AD.7"D @95& +9@ 7.5^ 5E,^ 7% ~ +"D9@A .7 &>.D-&5.5j6&A&7"-L65 D@7A.D.97a,!1!2!1&^OOm{ƒn^ƒ~}k€a%9.`{za{{‚izzz‚j~|ajz€jz~ƒ

&DD&@^a^&!&@^ a^&7&"4&^ a^s-5^aam|zzƒnaE7"D.975AD!.5.MD.979+7 @&"9,7.D.9769D.+!L797"797."5[email protected]5&K>7A.97aE!6-1)>@@F^JNmn^ {}zk{}a%9.`{za{|€{i@7a{}ƒƒzƒ

 {ƒ{ &E!E&@^ a^ .7,^^>>A.5!&@^ a^5I.9^a^DA97^a^3E-^a^b77^am{ƒƒ‚na AA A>&"D@96&D@L 7% j%D!A& A&@"-.7, 559JA "-@"[email protected]D.97 9+ D-& 6E5D.j >@9D&.7A>5."&9A96&"96>5&Ka341!!,!3'2^KIm{n^~€kza%9.`{za{z}‚i{zz

&I&^aa^-.7^ a^@.%5L7%^ a^&-^ a^&-7&@^aa^&I@^a^b @L^ aam|zz€na "955&"D.97 9+ !@&AD "7"&@ "&55 5.7&A +9@ D-& ADE%L 9+ +E7"D.9755L %.AD.7"D "7"&@ AE!DL>&Aa,!1!**^JIm€n^{k|a%9.`{za{z{€i3a""@a|zz€a{zazz‚

."95^aa^@L^aa^&@&45"4^ a^[email protected]@&^a^@.,-D^a a^7&^aa^bE55&@j "&^aam|z{|na-&-&5."A&>€‚mn.AA&5&"D.I&5L@&?E.@&%+9@D-&.7%E"D.97 9+>}j%&>&7%&7D>|{&K>@&AA.977%"&55j"L"5&@@&AD+D&@%6,&a,-%!,!^ OQm E5Ln^{kƒa%9.`{za{z}‚i97"a|z{|a~|€

9I.49I^ a^ @4^ a a^ -&7^ a^ 5&@67^ a^ 559-^ a a^ s 6!5&^ a a m|z{{na  j 6&%.D&% 5D&@7D.I& A>5.".7, 9+ D-& -.AD97& I@.7D "@9 |{ @&,E5D&A "7"&@ "&55 >@95.+&@D.97a -*!4*1 ,  !**4*1 '-*-%8^ LJm|zn^ ~|~~ka %9.`{za{{|‚iaz|~~j{{

E6A4.^ a^ D7!&^ a^ EME4.^ a^ 4-A-.^ a^ 46E@^ a^ "55.AD&@^ a^ b A4.^ a m|zzna j{ &7-7"&A D-& 7&D 7,.9,&7." "D.I.DL 7% .7 I.I9 ,@9JD- 9+ -E67797jA655"&555E7,"7"&@.7 6."&D-@9E,->@969D.7,j|j%&>&7%&7D 7,.9,&7&A.Aa -41,* -$ ++4,-*-%8 E*3'+-1!>  @T? JRNIF^ JPNmƒn^ €{k‚ƒa &D@.&I&%+@96-DD>`iiJJJa7"!.a756a7.-a,9Ii>E!6&%i{€|}{{

"8^a a^:@"95&A^a^!@^a^9@&79jE&79^ a^"59?E&^ a^&,^a^b.&D9^aa m|z{|na&DADD.""9597.MD.97@&?E.@&AD-&@&>@&AA.979+D-&&>.D-&5.5j6&A&7"-L65 D@7A.D.97.7%E"&@@@K{a,!1!**^KKm€n^zƒk|~a%9.`{za{z{€i3a""@a|z{|a{zaz{|

+^a^@9D&.7^aa^s"D9@A^am|z{}na   7   ^ Lm{n^{k{|a%9.`{za{zz|iJ@7a{zza 

5.I&.@^aaaa^sI@.79^aam|z{|na ,&7&5D&@7D.I&A>5.".7,’`>@9j7%7D.j 7,.9,&7.".A9+9@6A.7"7"&@^}€}k}za%9.`{za{zziAzz~}|jz{{j{z}j|

 {ƒ| 7%.D^ a^ -9E^ a^ -.E&^ a^ 9ED.7-9j7A+.&5%^ a^ .^ a^ .E^ a^ b E^ aja m|z{}na &796&jJ.%& 75LA.A @&I&5A  >@9D&.7 "99>&@D.97 7% "96>&D.D.97 .7 @&,E5D&% A>5.".7,a-*!4*1!**^NIm|n^||}k}a%9.`{za{z{€i3a695"&5a|z{}az}azz{

@4^aja^ E@^aa^&7,L&5^a^s&D&@^aam|zz‚na-&6.j|zz+6.5L%[email protected]&AD-& &>.D-&5.5 >-&79DL>& 9+ "7"&@ "&55A !L D@,&D.7, D-& j"%[email protected] @&>@&AA9@A { 7% |a!,!2H!5!*-/+!,3^KKmn^‚ƒ~kƒza%9.`{za{{z{i,%a{€~z€z‚

&5.A"-^ a^ 559^ a^ 9@7!5.-DD^ a a^ s "-9@^ a a m|zzƒna &E@975 "&55 %&>95@.MD.97 .7%E"&A .7D@,&7." "-@96D.7 69%.+."D.97A ++&"D.7,  5D&@7D.I& A>5.".7,^ JIOm{{na

&7"-&I^ a^ s IM9.&^ a a m|z{}na 97D@95 9+ 6&DADD." >@9,@&AA.97 !L 6."@9 @&,E5D9@L7&DJ9@4Aa341!!**'-*-%8^JNm€n^~€k~a%9.`{za{z}‚i7"!|€ƒ

9AI&"^a^.6.7AM4L^ a^sEA"-!&"4^am|z{}na"@9%96.7AA6&D!95.D&A&7A9@A 97 "-@96D.7a !**4*1 ,  -*!4*1 '$! '!,!2T?  ^ PImƒn^ {zƒk|~a %9.`{za{zziAzzz{‚jz{}j{|ƒ~j~

9AD.,9^ a ^ &>>^ a a^ L59@^ a a^ s @955^ a a m|zz}na &,E5D.97 9+ 6% A.,75.7, D-@9E,-%.++&@&7D.5@&"@E.D6&7D9+"9"D.ID9@A7%"9@&>@&AA9@A!L>@9D&.7Aa&!  -41,*^KKm{zn^|~}k€|a%9.`{za{zƒ}i&6!93i"%,||€

@%&&>^ a a^ ED-&@57%^ a a^ 5&^ a^ @.6&A^ a a^ s ."469@&^ a a m|z{|na A.>{i&%,+>|!.7%A6&D-L5D&%-.AD97& } }€7%A>5.".7,+"D9@A7%"97D@.!ED&A D9 D-& @&,E5D.97 9+ 5D&@7D.I& A>5.".7,a  - !,!3'2^ Qmn^ &{zz|{a %9.`{za{}{i39E@75a>,&7a{zz|{

E&A%^a^97%&^a^.5569@^a^@%:8&M^ aa^ @&A^a^AA,7LA^a^b:>&MjD/7^ am|z{|naK96&A&?E&7".7,.%&7D.+.&A@&"E@@&7D6EDD.97A9+D-&A>5.".7,+"D9@}{ ,&7& .7 "-@97." 5L6>-9"LD." 5&E4&6.a 341! !,!3'2^ MMm{n^ ~k|a %9.`{za{z}‚i7,a{z}|

 {ƒ} &.74&^ a a^ E^ a^ s -&7,^ a m|z{|na 7.5 @&>@&AA&A D-& A>5.".7, @&,E5D9@ &>.D-&5.5 A>5.".7,@&,E5D9@L>@9D&.7{D9>@969D&&>.D-&5.5j6&A&7"-L65D@7A.D.97a&! -41,* -$'-*-%'*&!+'2318^KQPm~}n^}€~}k~|a%9.`{za{z~i3!"a{{|a}ƒ{|

."-^ a a^ E9^ a^ "5&7%97^ a a^ .,7&@^ a a^ 7,^ a^ s 9E7D&@^ a a m|zz{na  &7&D."55L@"D!5&9%&59+ E67 5.969@6D.97 &7&D."55L@"D!5&9%&5 9+ E67 5.969@6D.97{^}€k}€za

9A&7!&@,&@^ a^ &-677^ a^ &.A&^ a^ @74&^ a^ E"-9^ a^ s E"-7&@^ a m|zz|na %&7D.+."D.97 9+  A  >@9D&.7 4.7A& 5>-j!.7%.7, >@9D&.7 .7 D-& "&55 7E"5&EAa -41,*-$!**4*1'-&!+'2318^QOm|n^}ƒ~k~z|a%9.`{za{zz|i3"!a{z|}}

9AA9J^ a a^ s 747&"-D^ a m|zz}na L7&@,.A6 !&DJ&&7 >€‚  -&5."A& 7% D-& D@7A"@.>D.975 "9"D.ID9@A  7% >}zza ,-%!,!^ KKm{n^ {{k€a %9.`{za{z}‚iA3a97"a{|z€z€

9EAA9A^ a a^ 5A69^ a^ 5+9@%^ a a^ L"49++^ a a^ 5.,[email protected]&I."^ a^ 7,^ a^ b 97%&&5.A^ aam|z{{na&7.7IA.I&m&7 n>@969D&A6E5D."&55E5@AD@&6.7, 69D.5.DL7%D@7A&7%9D-&5.56.,@D.97.769EA&69%&59+!@&AD"7"&@^|{|zk|{}{a %9.`{za{|~|i3"Aaz‚€|}{

9EAA9A^aa^ 9AJ6.^a^5A69^a^7,^a^D9!&M4.^a^%5&@^a^b4DL^a a m|z{{na&7.7IA.I&m&7m nn7%&7{{.A9+9@6A>5L%.AD.7"D@95&A.7!@&AD "7"&@ "&55 "9-&A.97 7% AA9".D.97 J.D- a *','* H 7/!1'+!,3* !3232'2^ KQm€n^{k|a%9.`{za{zziA{z‚jz{{jƒ}‚‚j€

5M67^aa^-E!&@Dj95&67^ a^sE@7&EK^ am|zzna€‚-&5."A&E7J.7%AD-& -E675&Dj6."@9>@&"E@A9@%E>5&K7%.A@&?E.@&%+9@5&Djj%.@&"D&%A.5&7".7,9+ ,&7& &K>@&AA.97a &! -41,* -$ '-*-%'* &!+'2318^ KQKm~n^ }|}kƒa %9.`{za{z~i3!"azz~|zz

 7"-&Mj.55:^ a^ .E^ a^ %& @@.9A^ a^ .5&A^ a^ 759^ a^ ED@&"AA^ a^ b 9AD.,9^ a m|z{|na j"D.ID.7, D@7A"@.>D.97 +"D9@A .7 "7"&@` !&L97%  7% DE69@ .7IA.I&7&AAa !**4*1 ,  -*!4*1 '$! '!,!2T?  ^ ORm|zn^ }~|ƒk€a %9.`{za{zziAzzz{‚jz{|j{{||j|  {ƒ~  7"-&Mj.55:^ a^ .5&A^ a^ %& @@.9A^ a^ ED@&"AA^ a^ ?E&@9^ a a^ AD&55A^ a^ s 9AD.,9^ a m|z{{na K>7%.7, @95&A 9+  +"D9@A .7 DE69@.,&7&A.A 7% DE69@ >@9,@&AA.97a +!1', -41,* -$ ,!1 !2!1&^ Jmn^ ‚ƒkƒ{|a &D@.&I&% +@96 -DD>`iiJJJa>E!6&%"&7D@5a7.-a,9Ii@D."5&@&7%&@a+",.]@D.%‰}{ƒ€|‚sD995‰>6"&7D@&M s@&7%&@DL>&‰!AD@"D

I,7&@^a^55(A^a^ 9E77&E^ a^6%^ aa^s-.&@L^ aam{ƒƒ~na5D&@7D.I& A>5.".7, .7 +.!@9!5AD ,@9JD- +"D9@ @&"&>D9@ | .A AA9".D&% J.D- .7%E"&% &>.D-&5.5j 6&A&7"-L65 D@7A.D.97 .7 @D !5%%&@ "@".796 "&55Aa -*!4*1 '-*-%8 -$ 3&! !**^ Nm‚n^ ‚{k€|a &D@.&I&% +@96 -DD>`iiJJJa>E!6&%"&7D@5a7.-a,9Ii@D."5&@&7%&@a+",.]@D.%‰}z{{z€sD995‰>6"&7D@&Ms @&7%&@DL>&‰!AD@"D

I,7&@^ a^ 6%^ a a^ s -.&@L^ a a m{ƒƒna -& M.7"j+.7,&@ >@9D&.7 A5E, "EA&A %&A69A96& %.AA9".D.97^ 7 .7.D.5 7% 7&"&AA@L AD&> +9@ ,@9JD- +"[email protected]%E"&% &>.D-&5.5j6&A&7"-L65 D@7A.D.97a &! -41,* -$ !** '-*-%8^ JLPm€n^ {~z}k{ƒa &D@.&I&% +@96 -DD>`iiJJJa>E!6&%"&7D@5a7.-a,9Ii@D."5&@&7%&@a+",.]@D.%‰|{}|~{sD995‰>6"&7D@&M s@&7%&@DL>&‰!AD@"D

"-5.&4&567^ a a^ .!!97A^ a a^ "^ a a^ @&.,-D97^ a a^ .MI.^ a a^ -7,^ a^ b 7A-^ a a m|z{{na @,&DA 9+ D-& DE69@ AE>>@&AA9@ 6.j|zz .7 @&,E5D.97 9+ D-& &>.D-&5.5j6&A&7"-L65 D@7A.D.97 .7 "7"&@a ,!1 !2!1&^ PJm|~n^ €zk‚|a %9.`{za{{‚izzz‚j~|aj{{jzƒ€~

"-J@DM^ a^ &A-9@&@^ a^ s AD^ a m|zzƒna -@96D.7 9@,7.MD.97 6@4A &K97j.7D@97 AD@E"DE@&a341!314341*H-*!4*1'-*-%8^JOmƒn^ƒƒzka%9.`{za{z}‚i7A6!a{€ƒ

->.@9^ aa^-&7,^aa^5LDM7.A^aa^5A69^a^97%&&5.A^ aa^4DL^a a^b &@D5&@^ a a m|z{{na 7 j%@.I&7 5D&@7D.I& A>5.".7, >@9,@6 9""E@A .7 -E67 !@&AD "7"&@ 7% 69%E5D&A "&55E5@ >-&79DL>&a  - !,!3'2^ Pm‚n^ &{zz||{‚a %9.`{za{}{i39E@75a>,&7a{zz||{‚

 {ƒ -.^ a^ 9@%.7^ a^ .7%&@^ a a^ .7%&@^ a^ s E^ aja m|zz~na @LAD5 AD@E"DE@& 9+ D-& -E67j%&>&7%&7DA>5.".7,7%&K>9@D+"D9@€a1-!! ',%2-$3&!3'-,*  !+8 -$ '!,!2 -$ 3&! ,'3!  33!2 -$ +!1'^ JIJm{n^ {€|‚k}}a %9.`{za{z}i>7Aaz~z‚{|{z{

-.7^a^9AA9J^ aa^ @7%&^ aa^s 747&"-D^am|zzna 7I95I&6&7D9+-&5."A&A >€‚ 7% >| .7 "9597 "7"&@a ,!1 !2!1&^ OPm{€n^ |k‚a %9.`{za{{‚izzz‚j ~|ajz€j~€|

-.79M4.^ a^ 4D7.^ a^ A-.4E^ a^ .79^ a^ A9,.^ a^ A-.4J^ a^ b E4L6^ a m|z{zna9J7@&,E5D.979+6."@9j|zz.7jAA9".D&%,AD@.""@".796a,!1 !2!1&^PIm{{n^~{ƒk|a%9.`{za{{‚izzz‚j~|ajzƒj~€|z

-.@4.-@^a^.D9-^a^s.LM979^ am|zzna.++&@&7D.5&,E5D.979+>.D-&5.57% &A&7"-L65 @4&@A !L  { @9D&.7A .7 >.D-&5.5 k &A&7"-L65 @7A.D.97

7%E"&%!L j“^JQm&>D&6!&@n^}}}k}~~a%9.`{za{zƒ{i6!"az

.&6&7A^ a^ "4AD%D^ a^ H7D&7^ a^ 55&@^ a^ &7AA&7^ a^  D9 @&,E5D& &>.D-&5.5j6&A&7"-L65 D@7A.D.97A t |z{| 7%&A .9A".&7"&a979D%.AD@.!ED&at|z{|7%&A.9A".&7"&a979D%.AD@.!ED&a^JIm|~n^ ~|€k~|{a

>.&A^ a^ .&5A&7^ a a^ %,&DD^ a ^ s E@,&^ a a m|zzƒna .A&% "-@96D.7 A.,7DE@&A @9E7% >95L%&7L5D.97 A.D&A 7% &K97Aa -*!4*1 !**^ LOm|n^ |~k~a %9.`{za{z{€i3a695"&5a|zzƒa{zazz‚

>9@7^ aa^s E7,^am|z{|na.++&@&7D.5@&,E5D.977%>@&%."D.I&>9D&7D.59+"@9 |{ .A9+9@6A .7 "9597 "7"&@a &! +!1', -41,* -$ 3&-*-%8^ JQIm€n^ |{€k|€a %9.`{za{z{€i3a3>D-a|z{|az|az|

>9@7^ aa^ EADDA"-&@^ a^ 9D-9@7^a^955%9^a^&@@79^a^E5&L^a^b%E@7&@^ a m|zzƒna .AD97& 6"@9 | .A9+9@6A >@&%."D D-& @.A4 9+ 5E7, "7"&@ @&"E@@&7"&a ,-%!,!^KQm}‚n^}~|}k‚a%9.`{za{z}‚i97"a|zzƒa|€

@&!@9J^am7a%ana-&+.!@97&"D.7D@7A"@.>D.97a

 {ƒ€ DE"4&^aa^ 9@A&A^a^s 9+677^am|zz‚na ^Pm€n^‚}zk‚}|a

E!@67L6^a^s5&559"-^am|z{{na@966."@9AD9D@,&DA`>D-JL%.A"9I&@L.7 "&55 +D& D@7A.D.97Aa 411!,3 /','-, ', !,!3'2 H !5!*-/+!,3^ KJm~n^ ~ƒ‚kz}a %9.`{za{z{€i3a,%&a|z{{az~az{{

EME4.^ a a^ 6,D^ a^ E,.69D9^ a^ J69D9^ a^ D9^ a^ s .LM979^ a m|zzƒna 9%E5D.97 9+ 6."@9 >@9"&AA.7, !L >}a 341!^ MOIm|~n^ |ƒk}}a %9.`{za{z}‚i7DE@&z‚{ƒƒ

4-A-.^ a^ s 674^ a m|zz€na 7%E"D.97 9+ >5E@.>9D&7D AD&6 "&55A +@96 69EA& &6!@L97." 7% %E5D +.!@9!5AD "E5DE@&A !L %&+.7&% +"D9@Aa !**^ JKOm~n^ €€}k€a %9.`{za{z{€i3a"&55a|zz€azaz|~

494^ a^ -.6.ME^ a^ A&,J^ a^ E"-.^ a^ .9^ a^ ,-@^ a^ b J69D9^ a m|z{|na9@"&%&K>@&AA.979+6.j{~}@&>@&AA&A i"jL"7%>€‚i>|A.,75.7,.7 "97"&@D J.D- 6.j{~ .7 ,ED DE69@A 9+ >"m.7n 6."&a *- ,!^ Pm‚n^ &~|{}a %9.`{za{}{i39E@75a>97&azz~|{}

MJ^ a^ AE"-.L^ a^ ME6.L^ a^ s 4,6^ a m|zzna ,@9JD- @@&AD D-@9E,- 69%E5D.979+D-&|>D-JL.7-E67"9597"7"&@"&55A^JIMm}ƒn^|ka

&@!&^ a^ DAE.^ a^ 9!&7j@ED-^ a^ -&7^ a^ DA97^ a^ 975%A97^ a a^ b &@M9+A4L^ aam|zzzna "&55j6&%.D&%@&>@&AA.979+DE69@.66E79AE@I&.557"&!L j{} 7% D-& j~j€ >D-JLa 341! ++4,-*-%8^ Jm€n^ {k|za %9.`{za{z}‚i‚|{

-.&@L^ a a m|zz|na >.D-&5.5j6&A&7"-L65 D@7A.D.97A .7 DE69E@ >@9,@&AA.97a 341! !5'!62@,!1^Km€n^~~|k~a%9.`{za{z}‚i7@"‚||

-.&@L^ a a^ "59?E&^ a^ E7,^ a a a^ s .&D9^ a a m|zzƒna >.D-&5.5j6&A&7"-L65 D@7A.D.97A .7 %&I&59>6&7D 7% %.A&A&a !**^ JLRmn^ ‚{kƒza %9.`{za{z{€i3a"&55a|zzƒa{{azz

 {ƒ .5,7&@^ a^.4959E^a^5D-66&@^a^66&D-^a^&D9^a^5" @"&5^ a^s E.,:^a m|zzƒnaE"5&9A96&>9A.D.97.7,A%[email protected]7D9+&K97@&"9,7.D.97a341!314341* H-*!4*1'-*-%8^JOmƒn^ƒƒ€k{zz{a%9.`{za{z}‚i7A6!a{€‚

A.^ a a^97-&@^ aa^E@>-L^a^-E^a^s7,^ am|z{|na>D.9D&6>9@5@&,E5D.97 9+ &>.D-&5.5j6&A&7"-L65 D@7A.D.97 .A &AA&7D.5 +9@ A?E69EA "&55 "@".796 6&DADA.Aa,!1!**^KKm€n^|k}€a%9.`{za{z{€i3a""@a|z{|azƒaz||

5&^ a^ 5&^ a^ >&7"&@^ a^ .55.6A^ a^ E^ aja^ 5.7&^ a^ b @7&55^ a a m|zzna 9I @&,E5D&A!@.7jA>&".+."A>5.".7,D9A->&D-&AL7>A&a341!!,!3'2^LPm‚n^‚~~k|a %9.`{za{z}‚i7,{€{z

7 @E7AI&7^a^&567^a^."-.&5A^a^>%&"6>^ a^ EL5&!@9&"4^a^s&55&+@9.%^a am|zz€na%&5D{7% {@&%.++&@&7D.55L&K>@&AA&%7%-I&9I&@5>>.7,"D.I.D.&A %[email protected],&79>EA&6!@L9,&7&A.Aa!5!*-/+!,3*8,+'2T?,$$''*4*'3'-,-$3&! +!1',22-'3'-,-$,3-+'232^KLNm€n^{~ƒ{kzza%9.`{za{zz|i%I%La|z|

@,A^ a a^ --^ a^ D.A-^ a^ &I7%9A4.^ a^ .7-^ a^ @@A^ a  a^ b L,.^ a m|z{{na .7,5&j695&"E5& .6,.7, 9+ D@7A"@.>D.9755L "9E>5&% 7% E7"9E>5&% A>5.".7,a !**^JMPmn^{z~k€a%9.`{za{z{€i3a"&55a|z{{a{zaz|~

&7!5&A^ aa^@9AA&E^ aja^ %&^ a^ 5.7"4^a^@.79A^a^&E5.&E^ aja^b-!9D^a m|z{}na | .A 7 .6>9@D7D @&,E5D9@ 9+ 6&A&7"-L65 D.AAE&jA>&".+." A>5.".7, .7 !9D- 79@65 7% "7"&@ D.AAE&Aa -*!4*1 ,  !**4*1 '-*-%8^ LLm|n^ }ƒ€k~za %9.`{za{{|‚iaz{{~j{|

&7!5&A^ aa^@9AA&E^ aja^ %&^ a^ 5.7"4^a^@.79A^a^&E5.&E^ aja^b-!9D^a m|z{}!na | .A 7 .6>9@D7D @&,E5D9@ 9+ 6&A&7"-L65 D.AAE&jA>&".+." A>5.".7, .7 !9D- 79@65 7% "7"&@ D.AAE&Aa -*!4*1 ,  !**4*1 '-*-%8^ LLm|n^ }ƒ€k~za %9.`{za{{|‚iaz{{~j{|

&7!5&A^ a a^ .,75^ a^ ,-%.,E.7^ a^ 9@D^ a^ s M.^ a m|z{|na .AAE&jA>&".+." 5D&@7D.I& A>5.".7, 9+ 4{ .A "97A&@I&% .7 %&ED&@9AD96&Aa -*!4*1 '-*-%8 ,  5-*43'-,^KRm{n^|€{kƒa%9.`{za{zƒ}i695!&Ii6A@{ƒ}

 {ƒ‚ 9,D^ a^ E7%.7,^ a^ @H7&@^ a^ .++&@A^ aja^ &@%99%D^ a^ E4^ a^ b &@6&4.7,^ a m|z{{na @&?E&7D "97"96.D7D .7"D.ID.97 9+ 6.j}~ 7% 6.j}~!i" !L >  6&D-L5D.97 .7 "959@&"D5^ >7"@&D."^ 666@L^ 9I@.7^ E@9D-&5.5^ 7% @&75 "&55 "@".796A 7% A9+D D.AAE& A@"96Aa '1&-62 1&'5T? , ,3!1,3'-,* -41,* -$ 3&-*-%8^MNQm}n^}{}k||a%9.`{za{zziAzz~|‚jz{zj{z}zj

,7&@^a^.%^a^.&@^a a^.&@^a a^.6&@^a^ .7D7&@^ a^b7%&@^ am|z{|na  .A  6E5D.+E7"D.975 "9j"D.ID9@ 9+ AD&@9.% -9@697& @&"&>D9@Aa -*!4*1 ,  !**4*1, -1',-*-%8^LOJm{j|n^‚zkƒ{a%9.`{za{z{€i3a6"&a|z{|az}az{~

7,^a^ E7,^ a^s E^am|z{|na-&5."A&@&,E5D&A6."@9&K>@&AA.977% "97D@.!ED&A D9 "LD9A4&5&D5 @&9@,7.MD.97 .7 !A5 !@&AD "7"&@ "&55Aa -*!4*1 H !**4*11-3!-+'2T?^JJm|n^{{{az{{ƒ}|a%9.`{za{z~i6">a{{{az{{ƒ}|

7,^ a^ 9^ a^ E7,^ a^ 7,^ a^ s .E^ aja m|zzƒna €‚  -&5."A& .A  7E"5&9"LD9>5A6." A-EDD5.7, >@9D&.7a !** !2!1&^ JRm{|n^ {}‚‚k~zza %9.`{za{z}‚i"@a|zzƒa{{}

7,^ a^ 9^a^7,^ a a^s.E^ajam|z{}na 7D&@"D.97!&DJ&&7>€‚-&5."A&7% |‡j"569%E5.7 >@969D&A "&55 6.,@D.97 7% 6&DADA.Aa 341! -++4,'3'-,2^ MmL|z{|n^{}~a%9.`{za{z}‚i7"966A|}~

7,^ a a^ -7,^ a^ 9E^ a^ s -.^ a a m|z{|na I&@&K>@&AA.97 9+  -&5."A& >€‚ >@9D&.7.7"ED7&9EAA?E69EA"&55"@".796a*','*, 7/!1'+!,3*!1+3-*-%8^ LPm‚n^‚‚|k‚a%9.`{za{{{{i3a{}€j||}za|z{|az~}€aK

7,^a^-&7^a^ 9^a^.^a^.^a^7^a^b &^ am|z{{na-L>&@6&D-L5D.979+ 6."@9j}~!i" -A >@9,79AD." I5E& +9@ AD,& Ⅰ 797jA655 "&55 5E7, "7"&@a ,!1 '-*-%8H&!1/8^JJmn^~ƒzk€a%9.`{za~{€{i"!Da{{aa{~z

@7&@^ a a^ -DD"-&@3&&^ a^ .7^ a^ .7,-^ a^ E4-9>%-LL^ a^ .A79^ a a^ s @&&7&^aam|zz~naE7"D.975.7D&@"D.97!&DJ&&76%^!.7%.7,>@9D&.7^7% >€‚-&5."A&a'-&!+'*, '-/&82'*!2!1&-++4,'3'-,2^LKMm{n^zk€a %9.`{za{z{€i3a!!@"a|zz~azƒaz{

 {ƒƒ @M&"-^ a a^ .7,^ a^ [email protected].7^ a^ .DD6@^ a ^ E^ a^ -&7^ a^ b @AD&7A^ a a m|z{zna 7 j@&,E5D&% A>5.".7, >@9,@66& .A !@9,D&% %[email protected], D-& &>.D-&5.5j 6&A&7"-L65 D@7A.D.97a &!  -41,*^ KRm{ƒn^ }|‚€k}zza %9.`{za{z}‚i&6!93a|z{za{ƒ

@M&"-^aa^D9^a a^!&D^a^ 9,&7&A"-^ aa^s@AD&7A^aam|zzƒna{7% |@&&>.D-&5.5"&55jDL>&jA>&".+."@&,E5D9@A9+ |A>5.".7,a-*!4*1!**^LLmn^ ƒ{k€z{a%9.`{za{z{€i3a695"&5a|zzƒaz{az|

D7!&^a^7,.AJ^ a^ .D,J^ a^4&L6^ a^,J^a^@9^a^b D9^a m|zz{na  AE!+6.5L 9+ j!.7%.7, j!9K >@9D&.7A "DA A 7 &AD@9,&7 @&"&>D9@ 5>- "9"D.ID9@ D-@9E,- D-& [email protected]5 "D.ID.97 %96.7 mj{n J.D- 7  "9"D.ID9@^a&! -41,*^KIm€n^{}~{k|a%9.`{za{zƒ}i&6!93i|za€a{}~{

-&&59"4^ a a^ s 9-7A97^ a a m|zz}na %[email protected]&%.D&% "&55E5@ A.,75.7,a 411!,3 /','-,',!**'-*-%8^JNmn^zƒk{~a%9.`{za{z{€izƒjz€~mz}nzz{z{j

.45E7%^aa^@6A&7^ aa^ E5+^a^L@A43=D^a^67D-7^a^ 7A&7^aa^b5@4^a a m|z{{na 99@%.7D&% &>.,&7&D." @&>@&AA.97 9+ D-& 6.j|zz +6.5L 7% 6.j|z .7 .7IA.I& !5%%&@ "7"&@a ,3!1,3'-,* -41,* -$ ,!1@ -41,* ,3!1,3'-,* 4 ,!1^JKQm€n^{}|k}~a%9.`{za{zz|i.3"a|~€{

.5A97^ a a^ D&A^ a a^ ."95^ a a^ @&,9@L^ a a^ [email protected]^ a a^ s E55&@j"&^ a a m|zz~na -& >€‚ 7% >|  !9K  -&5."A&A .7D&@"D J.D- { 7% @&>@&AA D@7A"@.>D.97 .7  >@969D&@jA>&".+." 677&@a  -*!4*1 '-*-%8^ N^ {{a %9.`{za{{‚€i{~{j|{ƒƒjj{{

.DD&7^ aa^s5&^ am|z{{na7%&@AD7%.7,A>5.".7,@&,E5D.97D-@9E,-A>5.".7,6>Aa 1!, 2',!,!3'2T? ^KPm}n^‚ƒkƒa%9.`{za{z{€i3aD.,a|z{za{|azz{

9@D-6^aa^-6&%^a^."95^aa^-96A^aa^[email protected]A6L^a^ .7,^ a^bE55&@j "&^aam|zzƒna-&j!9K>@9D&.7>|@&,E5D&A5>-ji9&AD@9,&7j%&>&7%&7D D@7A"@.>D.97 7% "&55 ,@9JD-^ 7% .A AA9".D&% J.D- .6>@9I&% [email protected]5 .7 5>-j >9A.D.I&!@&AD"7"&@a,-%!,!^KQm~€n^~z}k€~a%9.`{za{z}‚i97"a|zzƒa|€{

 |zz .9^ a^.7^ a^E9^a^E9^a^s7,^am|z{{na.7@&A>97A&D9AD@&AAa&! -41,*^ LIm}n^|~k}|a%9.`{za{z}‚i&6!93a|z{za}~

64E"-.^ a^ 9DD&@67^ a a^ 9^ a^ @E!7^ a a^ @.6^ a^ &7%&55^ a a^ b 9J&7AD&.7^ a a m|z{zna }j.7%E"&% 6."@9j{z .7-.!.DA j{ 7% DE69@ 7,.9,&7&A.Aa1-!! ',%2-$3&!3'-,* !+8-$'!,!2-$3&!,'3! 33!2-$ +!1'^JIPm{~n^€}}~kƒa%9.`{za{z}i>7Aazƒ{{z‚|{z

674^a^s5E^ aam|z{znaE"5&@@&>@9,@66.7,D9>5E@.>9D&7DADD&!LD-@&& >>@9"-&Aa341!^MONm|ƒƒn^z~k{|a%9.`{za{z}‚i7DE@&zƒ||ƒ

7,.AJ^a^ EI&5%D^a^ @&.7&AD^a^9-A&^aa^ 9-7^a^@4&@^aa^b9>57%^ aa m|zz‚na  >{|z D&7.7 A9+9@6 J.D"- ++&"DA -9 "D.I.DL ^ 7%E"&A E69@ &55 7IA.97^7%@&%."DA&DADD.".A&A&o“a%9.`{za{z~i3!"a‚z{{ƒ||zz

7,^ a^ .7^ a^ s .E^ a m|zzna -9A>-9@L5D.97A 9+  9K >€‚  &5."A& @& AA9".D&%J.D-7"&@&I&59>6&7D7%&55@95.+&@D.97-9A>-9@L5D.97A9+ 9K>€‚ &5."A&@&AA9".D&%J.D-7"&@&I&59>6&7D7%&55@95.+&@D.97^ }k}€}a

7,^a^.7^a^s.E^ajam|zz!na.,75.7,D9D-&!9Kjj@&,E5D.979+j!9K>€‚  -&5."A& !L >@9D&.7 >-9A>-9@L5D.97Aa !**4*1 '%,**',%^ JPm{|n^ {~ƒkz~a %9.`{za{z{€i3a"&55A.,a|zzaz}azz‚

7,^ a^ .7^ a^ s .E^ aja m|zz€na €‚  -&5."A& 6&%.D&A  j.7%E"&% &>.D-&5.5 6&A&7"-L65 D@7A.D.97 !L %.A>5".7, K.7 +@96 !&Dj"D&7.7a !**^ JKPm{n^ {}ƒka %9.`{za{z{€i3a"&55a|zz€az‚az}€

7,^a^s 749JA4L^am|zz€na-&j!9K>@9D&.7&%{E7J.7%A%E>5&K&A!L 69%& %.AD.7"D +@96 D@7A59"D.7, -&5."A&Aa 341! 314341* H -*!4*1 '-*-%8^ JLm{{n^ƒ‚{k€a%9.`{za{z}‚i7A6!{{€

7,^a^-7^a^-7,^a^E7^a^7,^a^.7^a^b .7^am|z{{naA&"97%@L AD@E"DE@&.76EDE55L&K"5EA.I&A>5.".7,a 341!314341*H-*!4*1'-*-%8^JQm|n^ {ƒk€‚a%9.`{za{z}‚i7A6!a{ƒƒ

 |z{ 9^ a^ @."4^ a^ I@@%^ a^ .9^ a^ [email protected]D&@9^ a a^ s &5A&7+&5%^ a m|z{zna &%.D.979+D@7A"@.>D.975.7AE5D.97!Lj!9Kj!.7%.7,>@9D&.7>€‚7% AD&@9.% @&"&>D9@  "D.ID9@ a !,!2 H !5!*-/+!,3^ KMm||n^ |~}ka %9.`{za{{z{i,%a{ƒ€‚{z

&9^ aa^9E+5^a a^.7,^aa^&7,^ aa^E^aja^s ,&^a am|zzƒna7"9%& +9@D-&|A>5.".7,@&,E5D9@@&I&5&%!L6>>.7,j>@9D&.7.7D&@"D.97A.7AD&6 "&55Aa341!314341*H-*!4*1'-*-%8^JOm|n^{}zka%9.`{za{z}‚i7A6!a{~

-7,^a^D>>&7!&"4^aa^-.D&^aa^I.7&^ a a^ 9@%97^ a a^s@7.DM^aam{ƒƒƒna &".>@9"5 &>.D-&5.5j6&A&7"-L65   A.,75.7, .A @&?E.@&% +9@ "&"5 %&I&59>6&7D^ {}k{‚za%9.`{za{|~|i%&Iaz|{

-7,^aam|zzƒna97j6%>D-JLA.7 j!&DA.,75.7,a!**!2!1&^JRm{n^{|‚k}ƒa %9.`{za{z}‚i"@a|zz‚a}|‚

-&7,^ a^ s 7,^ a m|z{}na E5D.5L&@ "97D@95 9+ D-&  6AD&@ @&,E5D9@Aa ,-%!,!^ m&!@E@Ln^{kƒa%9.`{za{z}‚i97"a|z{}a{|‚

.%@^a^9BD37#.#^a^ 5&^a^ 93"^a^9534^a^ 5I#^a^s@%&A^am|z{{na9J7j @&,E5D.979+6."@9A9+D-&6.j|zz+6.5L7%6.j|z^7%75D&@&%&K>@&AA.97 9+"5AA."7%%&A69A965"%[email protected].7A>.7%5&"&55"@".7969+D-&-&%7%7&"4jj -556@4 9+ &>.D-&5.5j6&A&7"-L65 D@7A.D.97a 4+, 3&-*-%8^ MKm~n^ ~‚|k‚a %9.`{za{z{€i3a-E6>D-a|z{zazaz|z

97D^a^.DD&7"9E@D^a^67^a^ &@677^a^ED&@D@&^a^sE!9&E+^am|z{|na-&  -&5."A& i>€‚ .A  4&L +"D9@ >@969D.7, "j+9A &K>@&AA.97 D %.++&@&7D 5&I&5A +@96 D@7A"@.>D.97 D9 6 &K>9@Da 4*!' ' 2 !2!1&^ MJm{n^ ~k€~a %9.`{za{zƒ}i7@i,4A{z~€

 |z|