3ca$RS`$?SRWhW\WaQVS\$=ZW\WY$c\R$B]ZWYZW\WY$;;;$ =ZW\WY$RS`$>cReWU$?OfW[WZWO\a$G\WdS`aWbjb$?l\QVS\$ 6W`SYb]`1$B`]T%$6`%$6`%$?WQVOSZ$d]\$4S`UeSZb$ $ $ $ 3YcbS$[gSZ]WaQVS$>ScYj[WS$[Wb$`SYc``S\bS\$QV`][]a][OZS\$F`O\aZ]YObW]\S\#$ 3\OZgaS$RS`$8caW]\a^`]bSW\S$c\R$Y]]^S`WS`S\RS`$[]ZSYcZO`S`$?cbObW]\S\$OcT$ RS[$ISU$hc$SW\S`$hWSZUS`WQVbSbS\#$^S`a]\OZWaWS`bS\$FVS`O^WS%$ $ $ $ $ 6WaaS`bObW]\$ $ hc[$7`eS`P$RSa$6]Yb]`U`ORSa$RS`$?SRWhW\$ O\$RS`$?SRWhW\WaQVS\$8OYcZbjb$RS`$ >cReWU$?OfW[WZWO\a$G\WdS`aWbjb$hc$?l\QVS\$ $ $ d]`USZSUb$d]\$ $ $ EOP`W\O$A^Obh$ $ $ Oca$ :WZRSaVSW[$ $ $ $ )'(/$ $ $
$
?Wb$9S\SV[WUc\U$RS`$?SRWhW\WaQVS\$8OYcZbjb$ $ RS`$>cReWU$?OfW[WZWO\a$G\WdS`aWbjb$?l\QVS\$ $
$
$
4S`WQVbS`abObbS`1$$$ $ $ B`]T%$6`%$[SR%$EbSTO\$4]VZO\RS`$
$ $ $ $ $ $
?WbPS`WQVbS`abObbS`1$$$ $ B`]T%$6`%$BSbS`$4%$4SQYS`$
$ $ $ $ $ B6$6`%$5V`Wab]^V$IOZh$
B`]T%$6`%$;`[SZO$ B`]T%$6`%$EbSTO\$7\R`Sa$ $ ?WbPSb`Scc\U$Rc`QV$RS\$$ ^`][]dWS`bS\$?WbO`PSWbS`1$$$ B6$6`%$[SR%$BVWZW^^$9`SWT$ $ $ $ 6SYO\1$ $ $ $ B`]T%$6`%$[SR%$RS\b%$DSW\VO`R$:WQYSZ$ $ $ FOU$RS`$[l\RZWQVS\$B`lTc\U1$$ '*%',%)'(/$ $ $ $ $ $ $ ! V! $ $ 0E@ANNO=OOGE?DA+;AMNE?DAMPIC+ $ $ ;QV#$ EOP`W\O$ A^Obh$ S`YZj`S$ VWS`[Wb$ O\$ 7WRSa$ abObb#$ ROaa$ WQV$ RWS$ d]`ZWSUS\RS$ 6WaaS`bObW]\$[Wb$RS[$FVS[O$ $ Z)HRQC' JUCILGPAFC' 2CRHWJGC' JGQ' OCHROOCKQCK' AFOLJLPLJ?ICK' 9O?KPILH?QGLKCK6' )K?IUPC' BCO' .RPGLKPMOLQCGKC' RKB' HLLMCOGCOCKBCO' JLICHRI?OCO' 3RQ?QGLKCK' ?RD' BCJ' ;CE'VR'CGKCO'VGCIECOGAFQCQCK6'MCOPLK?IGPGCOQCK'9FCO?MGCY' $ aSZPabj\RWU$ dS`TOaab#$ [WQV$ OcmS`$ RS`$ O\USUSPS\S\$ YSW\S`$ eSWbS`S\$ :WZTa[WbbSZ$ PSRWS\b$ c\R$ OZZS$ 7`YS\\b\WaaS#$ RWS$ Oca$ RS[$ EQV`WTbbc[$ UO\h$ ]RS`$ O\\jVS`\R$ lPS`\][[S\$ aW\R#$ OZa$ a]ZQVS$ YS\\bZWQV$ US[OQVb$ c\R$ \OQV$ WV`S`$ :S`Yc\Tb$ c\bS`$ 4ShSWQV\c\U$RS`$8c\RabSZZS$SW\hSZ\$\OQVUSeWSaS\$VOPS%$ $ ;QV$S`YZj`S$RSa$ISWbS`S\#$ROaa$RWS$VWS`$d]`USZSUbS$6WaaS`bObW]\$\WQVb$W\$UZSWQVS`$]RS`$ jV\ZWQVS`$8]`[$PSW$SW\S`$O\RS`S\$EbSZZS$hc`$7`ZO\Uc\U$SW\Sa$OYORS[WaQVS\$9`ORSa$ SW\US`SWQVb$ec`RS%$ $ $ $ $ A`b#$6Obc[$ $ $ $ $ $ G\bS`aQV`WTb$ $ $ $ $ $ $ $ $ ! W! $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ ! X! 2ID=GONQAMTAE?DIEN+ $ 3PYl`hc\UadS`hSWQV\Wa%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%$-$ BcPZWYObW]\aZWabS%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%$/$ $ ($$$7W\ZSWbc\U$ (%($3YcbS$[gSZ]WaQVS$>ScYj[WS$!3?>"%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%$0$ (%)$3?>$[Wb$53>?$38('$8caW]\%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%$(($$ (%*$3?>$[Wb$5484$?K:(($$]RS`$DG@J($DG@J(F($8caW]\$$ !548$>ScYj[WS"%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%$()$ (%*%($=]]^S`WS`S\RS$?cbObW]\S\$W\$548$>ScYj[WS\%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%$(+$ (%+$6O`abSZZc\U$RSa$8]`aQVc\Uad]`VOPS\a$c\R$7WUS\O\bSWZ$RS`$3`PSWbS\%%%%%$(,$ ! $ )$ LcaO[[S\TOaac\U%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%$(-$ *$ Ec[[O`g%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%$(/$ $ +$ A`WUW\OZO`PSWbS\%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%$))$ +%($FVS$QZObV`W\$PW\RW\U$R][OW\$]T$53>?$O\R$bVS$A?$>L$R][OW\$]T$38('$$ O`S$acTTWQWS\b$b]$W\RcQS$OQcbS$[gSZ]WR$ZScYS[WO$W\$[WQS%%%%%%%%%%%%%%%%%%%%%%%%%%$)*$ +%)$7f][S$aS_cS\QW\U$WRS\bWTWSa$`SQc``W\U$8>F*[email protected]=$[cbObW]\a$W\$$ 5]`S$4W\RW\U$8OQb]`$ZScYS[WO%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%$**$ $ $ ,$ >WbS`Obc`dS`hSWQV\Wa%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%$+)$ $ -$ 6O\YaOUc\US\%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%$+,$ $ $ $ $ $ $ $ ! Y! 3PYl`hc\UadS`hSWQV\Wa$ $ 38('$$$ 3>>$($TcaSR$US\S$T`][$QV`][]a][S$('#$]RS`$?>>F('$ 3?>$ $ 3YcbS$[gSZ]WaQVS$>ScYj[WS$ 3EJ>($ 3RRWbW]\OZ$aSf$Q][Pa$ZWYS$b`O\aQ`W^bW]\OZ$`SUcZOb]`$($ 3EJ>)$ 3RRWbW]\OZ$aSf$Q][Pa$ZWYS$b`O\aQ`W^bW]\OZ$`SUcZOb]`$)$ 53>?$$ 5ZObV`W\$OaaS[PZg$Zg[^V]WR$[gSZ]WR$ZScYS[WO$US\S$ 5484$$ 5]`S$PW\RW\U$TOQb]`$PSbO$acPc\Wb$ 548$ $ 5]`S$PW\RW\U$TOQb]`$ 54>$ $ 5PZ$^`]b]$]\Q]US\S#$7*$cPW_cWbW\$^`]bSW\$ZWUOaS$ 585$$ $ 5]Z]\g$T]`[W\U$QSZZ$ 5D$ $ 5][^ZSbS$`S[WaaW]\$!Y][^ZSbbS$DS[WaaW]\"$ 6AF(>$ 6AF($ZWYS$VWab]\S$:*=.0$[SbVgZb`O\aTS`OaS$ 7>@$ $ 7c`]^SO\$>ScYS[WO$@Sb$ 834$ $ 8`S\QV$3[S`WQO\$4`WbWaV$ 835E$$ 8Zc]`SaQS\QS$OQbWdObSR$QSZZ$a]`bW\U$!6c`QVTZcaahgb][Sb`WS"$ 8;E:$ $ 8Zc]`SahS\h$W\$aWbc$:gP`WRWaWS`c\U$ 8>F*$ $ 8?E$ZWYS$bg`]aW\$YW\OaS$ :*=.0$ :Wab]\S$*$ZgaW\S$.0$ :AJ$$$ :][S]P]f$ :E5$ $ :c[O\$abS[$QSZZ$!Vc[O\S$EbO[[hSZZS"$ :E5F$$ :S[Ob]^]WSbWQ$abS[$QSZZ$b`O\a^ZO\bObW]\$$ !Vj[Ob]^]SbWaQVS$EbO[[hSZZb`O\a^ZO\bObW]\"$ W\d$ $ ;\dS`aW]\$ ;F6$ $ ;\bS`\OZ$bO\RS[$Rc^ZWYObW]\$ =;F$ $ =;F$^`]b]$]\Q]US\S$`SQS^b]`$bg`]aW\S$YW\OaS$ ?D5$ $ ?SRWQOZ$DSaSO`QV$5]c\QWZ$ ?D6$ $ ?W\W[OZ$`SaWRcOZ$RWaSOaS$![W\W[OZS$DSabS`Y`O\Yc\U"$ ?K:(($ ?g]aW\$VSOdg$QVOW\$(($ @9E$ $ @Sfb$US\S`ObW]\$aS_cS\QW\U$ BFB@(($ B`]bSW\$bg`]aW\S$^V]a^VObOaS#$\]\$`SQS^b]`$bg^S$(($ D3E$ $ DOb$aO`Q][O$dW`OZ$]\Q]US\S$V][]Z]U$ DG@J($ Dc\b$`SZObSR$b`O\aQ`W^bW]\$TOQb]`$($ DG@J(F($ Dc\b$`SZObSR$b`O\aQ`W^bW]\$TOQb]`$($b`O\aZ]QObSR$b]$($ ! Z! AE$ $ AdS`OZZ$ac`dWdOZ$!9SaO[blPS`ZSPS\"$ b$ $ F`O\aZ]YObW]\$ F7F)$ $ FSb$[SbVgZQgb]aW\S$RW]fgUS\OaS$)$ F=6$ $ Fg`]aW\YW\OaS$6][j\S$ F=;$ $ Fg`]aW\YW\OaS$;\VWPWb]`$ H38$ $ HO`WO\b$OZZSZS$T`S_cS\Qg$!3ZZSZT`S_cS\h$RS`$^ObV]Z]UWaQVS\$HO`WO\bS"$ I7E$ $ IV]ZS$Sf][S$aS_cS\QW\U$ I:A$ $ I]`ZR$:SOZbV$A`UO\WhObW]\$ L4F4.3$ LW\Q$TW\US`$O\R$4F4$R][OW\$Q]\bOW\W\U$.3$ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ ! [! $ $ $ 6WS$d]`ZWSUS\RS$Yc[cZObWdS$6WaaS`bObW]\$c[TOaab$T]ZUS\RS$heSW$?O\caY`W^bS#$ RWS$d]`OP$^cPZWhWS`b$ec`RS\1$ $ $ $ $ $ 7K=OT+ 9 #$B]ZhS`$:#$:S`]ZR$F#$=]\abO\RW\$@B#$=aWS\hgY$4#$LSZZ[SWS`$7#$H]aPS`U$E#$ 9`OT$3#$=`SPa$E#$4Zc[$:#$:]^T\S`$=B#$=OYORWO$B?#$EQV\SWRS`$E#$6cT]c`$3#$4`OSaa$ <#$ EOcS`ZO\R$ ?5#$ 4S`RSZ$ I7#$ 4cQV\S`$ F#$ I]S`[O\\$ 4<#$ :WRRS[O\\$ I#$ E^WSYS`[O\\$ =#$ 4]VZO\RS`$ E=$ O\R$ 9`SWT$ B3$ !)'(*"$ -TLJC' 8CNRCKAGKE' 0BCKQGDGCP' 7CAROOGKE'.29&'4'('1'3RQ?QGLKP'GK'+LOC *GKBGKE'.?AQLO'2CRHCJG? %$ -GJJ@ 2$())!('"1$ (.-($0%$ $ $ 6SaV^O\RS$3<#$D]cVW$3#$>W\$K#$EbORZS`$5#$9`SWT$B3#$3`aS\W$@#$ 7K=OT+9 #$CcW\bO\WZZO$ 8S\R$>#$:]Zh[O\\$=#$:WRRS[O\\$I#$6kV\S`$=#$6kV\S`$:#$ Jc$ 9#$ 3`[ab`]\U$ E3#$ 4]VZO\RS`$E=#$4caYS$5$!)'(("$ 9FC'AI?QFOGK @GKBGKE'BLJ?GK'LD'+)23'?KB'QFC'53 2>' BLJ?GK'LD').$#'?OC'PRDDGAGCKQ'QL'GKBRAC'?ARQC'JUCILGB'ICRHCJG?'GK'JGAC! $4APFAHE= 2$$ ),!(("1$(.(/$).%$ $ $ $ $ $ $ $ $ $ $ $ $ $ $ + ! \! $+0EIGAEOPIC+ + $"$+,FPOA+HSAGJEN?DA+4APFUHEA+A,54 + $ 6WS$ 3?>$ Wab$ SW\S$ US\SbWaQV$ VSbS`]US\S#$ YZ]\OZS$ 7`Y`O\Yc\U$ Vj[Ob]^]SbWaQVS`$ H]`ZjcTS`hSZZS\$ W[$ =\]QVS\[O`Y%$ EWS$ S\babSVb$ Rc`QV$ a][ObWaQV$ S`e]`PS\S$ US\SbWaQVS$c\R$&$]RS`$S^WUS\SbWaQVS$HS`j\RS`c\US\#$eSZQVS$eWQVbWUS$B`]hSaaS$eWS$ ESZPabS`\ScS`c\U#$ B`]ZWTS`ObW]\$ c\R$ 6WTTS`S\hWS`c\U$ PSSW\b`jQVbWUS\%M(N$ 6WS$ PSb`]TTS\S$ H]`ZjcTS`hSZZS$ dS`ZWS`b$ RWS$ 8jVWUYSWb$ Ocahc`SWTS\$ !6WTTS`S\hWS`c\UaPZ]QY"$ c\R$eW`R$UZSWQVhSWbWU$c\abS`PZWQV#$aWS$bSWZb$aWQV$c\Y]\b`]ZZWS`b$!B`]ZWTS`ObW]\ad]`bSWZ"$c\R$ Sa$Y][[b$hc`$[OaaWdS\$YZ]\OZS\$7f^O\aW]\$RWSaS`$SW\hSZ\S\$RSTSYbS\$LSZZS%M)N$ 6c`QV$ RS\$ :W\hcUSeW\\$ eSWbS`S`$ ?cbObW]\S\$ eW`R$ RWS$ YZ]\OZS$ 7d]ZcbW]\$ d]`O\USb`WSPS\#$ RS`$ [OZWU\S$ =Z]\$ USeW\\b$ eSWbS`S\$ iPS`ZSPS\ad]`bSWZ$ c\R$ RO[Wb$ aQVZWSmZWQV$ROa$iPS`USeWQVb$W\$RS`$:j[Ob]^]SaS%M*#$+N$6WS$USac\RS\$LSZZS\$eS`RS\$ dS`R`j\Ub%$6Oa$YZW\WaQVS$7`aQVSW\c\UaPWZR$RS`$3?>$eW`R$Rc`QV$US\Oc$RWSaS$[OaaWdS$ [OZWU\S$ =\]QVS\[O`YW\TWZb`ObW]\$ [Wb$ RS`$ 8]ZUS$ SW\S`$ hc\SV[S\RS\$ Vj[Ob]$ ^]SbWaQVS\$;\acTTWhWS\h$PSabW[[b%$ 6WS$ W\RWdWRcSZZS$ 3?>$ hSWUbS$ XSR]QV$ SW\S$ abO`YS$ YZW\WaQVS$ :SbS`]US\Wbjb#$ eSZQVS$ 3caR`cQY$ RS`$ hcU`c\RSZWSUS\RS\$ US\SbWaQVS\$ HS`j\RS`c\U$ RWSaS`$ V]QV[OZWU\S\$ 7`Y`O\Yc\U$ Wab%$ 7\ba^`SQVS\R$ eW`R$ RWS$ 3?>$ aSWb$ )''($ O\VO\R$ RS`$ I:A$ =ZOaaWTWYObW]\$ \OQV$ []`^V]Z]UWaQVS\#$ YZW\WaQVS\$ c\R$ US\SbWaQVS\$ ?S`Y[OZS\$ c\bS`bSWZb%M,N$6WS$DWaWY]ab`ObWTWhWS`c\U$S`T]ZUb$\OQV$7[^TSVZc\US\$RSa$7>@$c\R$?D5$ [WbbSZa$hgb]US\SbWaQVS`$c\R$OcaUSejVZbS`$[]ZSYcZO`US\SbWaQVS`$HS`j\RS`c\US\$hc[$ LSWb^c\Yb$RS`$7`abRWOU\]aS$!3PP%$("%M-#$.N$$ $ ,>>"+ $+$ Oca$ M-N$ 7W\bSWZc\U$ d]\$ 3?>$BObWS\bS\$W\$R`SW$US\SbWaQVS$ EcPU`c^^S\$ O\VO\R$ RS`$ hgb]$ US\SbWaQVS\$c\R&$]RS`$[]ZSYcZO`$ US\SbWaQVS\$ HS`j\RS`c\US\$ !7>@$=ZOaaWTWYObW]\"%$ $ $ $ ! ]! 6WS$ SW\hSZ\S\$ EcPU`c^^S\$ hSWUS\$ aWU\WTWYO\bS$ G\bS`aQVWSRS$ W[$ Y`O\YVSWbaT`SWS\$ iPS`ZSPS\$c\R$9SaO[blPS`ZSPS\$!3PP%$)"%M-$/N$$ $ A B 1.0 Favorable (n = 324) 1.0 Favorable (n = 339) Intermediate-I (n = 109) Intermediate-I (n = 144) Intermediate-II (n = 123) Intermediate-II (n = 156) 0.8 Adverse (n = 90) 0.8 Adverse (n = 179) 0.6 0.6 0.4 0.4 (probability) (probability) 0.2 Survival Overall 0.2 Disease-Free Survival Disease-Free P < .001 P < .001 0 12 3 45 0 12 3 45 Time (years) Time (years) C D ,>>"+ %+$ Oca$ M0N$ iPS`ZSPS\$ d]\$ 3?>$ BObWS\bS\$ S\ba^`SQVS\R$ RS`$ dWS`$ US\SbWaQVS\$ 9`c^^S\$ RS`$ 7c`]^SO\$>ScYS[WO$@Sb$=ZOaaWTWYObW]\$Tl`$BObWS\bS\$c\bS`$-'$ $ 6WSaS$ G\bS`bSWZc\U$ VW\aWQVbZWQV$ RS[$ FVS`O^WSO\a^`SQVS\$ Wab$ S\baQVSWRS\R$ Tl`$ RWS$ 3caeOVZ$ RS`$ 4SVO\RZc\U%$ BObWS\bS\$ RS`$ ^`]U\]abWaQV$ Ul\abWUS\$ EcPU`c^^S$ S``SWQVS\$[SWab$[Wb$OZZSW\WUS`$5VS[]bVS`O^WS$SW\S$ROcS`VOTbS$DS[WaaW]\$!5D1$0- #$ *$ F`]bh$RS`$U`]mS\$4SRScbc\U$RWS$RS\$8caW]\ab`O\aY`W^bS\$ObbSabWS`b$eW`R#$hSWUb$aWQV$W\$ RS\$ dS`aQVWSRS\S\$ >ScYj[WS$?]RSZZS\$ SW\S$ aSV`$ ZO\US$ >ObS\h$ PWa$ hc`$ >ScYj[WS$ ! UT! 7\babSVc\U%$ 6WSa$ Zjaab$ dS`[cbS\#$ ROaa$ RWS$ RW`SYbS\$ 7TTSYbS$ RWSaS`$ ]\Y]US\S\$ DSO``O\US[S\ba$\WQVb$Oca`SWQVS\#$c[$SW\S$>ScYj[WS$OcahcZkaS\%$HWSZ[SV`$aQVSW\b$ RWS$3YYc[cZObW]\$hcajbhZWQVS`$US\SbWaQVS`$HS`j\RS`c\US\$\]beS\RWU$hc$aSW\%M()$(-N$$ $ $ $"%"+,54+HEO+ $"'(%"%! +1PNEJI++ $ 6Oa$ 8caW]\aUS\$ +)23 ).$# $ Wab$ ROa$ DSacZbOb$ SW\S`$ F`O\aZ]YObW]\$ heWaQVS\$ RS\$ 5V`][]a][S\$('$!38('"$c\R$(($!53>?"$Mb!('2(("!^(*$(+2_(+$)("N%$7a$eW`R$W\$QO%$ ( $ RS`$ 3?>$$ c\R$ OcQV$ 3>>$BObWS\bS\$ PSaQV`WSPS\$ c\R$ Wab$ [Wb$ SW\S`$ aSV`$ c\Ul\abWUS\$B`]U\]aS$Oaa]hWWS`b%M(.N$$ $ 7a$ Y]\\bS$ UShSWUb$ eS`RS\#$ ROa$ RWSaSa$ 8caW]\ab`O\aY`W^b$ GK' SGSL $ RWS$ [OZWU\S$ F`O\aT]`[ObW]\$OcaZkaS\$YO\\%M()#$(/N$6WS$ZO\US$>ObS\h#$eWS$OcQV$RWS$W\Y][^ZSbbS$ BS\Sb`O\h$ a^`SQVS\$ XSR]QV$ ROTl`#$ ROaa$ hcajbhZWQVS$ US\SbWaQVS$ HS`j\RS`c\US\$ Tl`$ RWS$S\RUlZbWUS$7\babSVc\U$SW\S`$>ScYj[WS$\]beS\RWU$aW\R%M()$(+N$$ $ 6S`$?SQVO\Wa[ca$RS`$ZScYj[WaQVS\ F`O\aT]`[ObW]\$Rc`QV$ +)23 ).$# Wab$hc[$FSWZ$ OPVj\UWU$ d]\$ ,59$2 %$ 6AF(>$ Wab$ SW\S$ ?SbVgZb`O\aTS`OaS$ c\R$ OcaaQVZWSmZWQV$ dS`O\be]`bZWQV$ Tl`$ RWS$ ?SbVgZWS`c\U$ d]\$ >gaW\$ .0$ W[$ :Wab]\$ :*$ !:*=.0"%$ 38('$ W\bS`OUWS`b$ lPS`$ aSW\S$ A?$>L$ 6][j\S$ [Wb$ 6AF(>%$ 6c`QV$ RWS$ 8caW]\$ [Wb$ +)23 $ Y][[b$ Sa$ hc`$ 8SVZabScS`c\U$ c\R$ SW\S`$ OPS``O\bS\$ :*=.0$?SbVgZWS`c\U$ !3PP%$ *"%$ 6AF(>$ [SbVgZWS`b$ c\bS`$ O\RS`S[$ B`][]b]`S\$ RS`$ /5<) $ !V][S]P]f$ 3"$ 5ZcabS`$ 9S\S$ c\R$ TlV`b$ RORc`QV$ hc$ SW\S`$ OPS``O\bS\#$ Y]\abWbcbWdS\$ 3YbWdWS`c\U$ RWSaS`$ 9S\S%M(*#$(+#$(0N$/5< $9S\S$a^WSZS\$SW\S$eWQVbWUS$D]ZZS$W\$RS`$:j[Ob]^]SaS$c\R$ >ScYj[]US\SaS%$ 6WS$ Tl`$ +)23").$# $>ScYj[WS\$ QVO`OYbS`WabWaQVS$ iPS`Sf^`SaaW]\$ RS`$ /5<)$ 5ZcabS`$ 9S\S$ abW[cZWS`b$ RWS$ ESZPabS`\ScS`c\U$ c\R$ RWS$ ZScYj[WaQVS$ F`O\aT]`[ObW]\$d]\$:E5%M)'N$$ $ (!&)$'#!*'&*/.-//* ,>>"+&+ $8SVZabScS`c\U$d]\$6AF(>$ Rc`QV$ RWS$ 8caW]\$ d]\$ +)23 $ c\R$ !&'*$% ).$# $[Wb$\OQVT]ZUS\RS`$OPS``O\bS`$ "$%%%%%%%%%%%""*)% .12%"0* #&("% :*=.0$?SbVgZWS`c\U $ $ ! UU! 3[$4SWa^WSZ$RS`$ +)23").$# $>ScYj[WS$Y]\\bS$hcZSbhb$SfS[^ZO`WaQV$UShSWUb$eS`RS\$ eWS$HS`j\RS`c\US\$W\$S^WUS\SbWaQVS\$?cabS`\#$W\aPSa]\RS`S$RWS$^]abb`O\aZObW]\OZS$ 5V`][ObW\[]RWTWYObW]\#$hc`$>ScYj[WS$7\babSVc\U$PSWb`OUS\%$6O[Wb$PWSbSb$aWQV$\WQVb$ hcZSbhb$ OcQV$ SW\$ Uj\hZWQV$ \ScS`$ FVS`O^WSO\aObh$ OPaSWba$ RS`$ abO\RO`RWaWS`bS\$ 5VS[]bVS`O^WS`SUW[S$ c\R$ EbO[[hSZZb`O\a^ZO\bObW]\%$ ;\VWPWb]`S\$ RS`$ ?SbVgZ$ b`O\aTS`OaS$ 6AF(>$ hSWUS\$ SW\S$ V]VS$ IW`YaO[YSWb$ W\$ +)23").$# $ >ScYj[WS$ ?]RSZZS\%$ 6c`QV$ aSZSYbWdS$ ;\VWPWbW]\$ RS`$ E$!,r$ORS\]agZ"$Z$[SbVW]\W\$ !E3?"$ 4W\Rc\UaOYbWdWbjb$RS`$?SbVgZb`O\aTS`OaS$Rc`QV$RWS$Eb`cYbc`O\OZ]UO$7BL''+...$c\R$ 7BL$,-.-$ Y][[b$ Sa$ hc$ SW\S`$ ROcS`VOTbS\$ DSRcYbW]\$ RS`$ :*=.0$?SbVgZWS`c\U#$ eSZQVS$RWS$7f^`SaaW]\$d]\$A\Y]US\S\$!c%O%$ /5<) "$SW\Rj[[b$c\R$OcT$RWSaS[$ISU$ a^ShWTWaQV$RWS$B`]ZWTS`ObW]\$RS`$[OZWU\S\$LSZZS\$c\bS`R`lQYb%M(+N$ $ $ $"&"+,54+HEO+ $#%#*(.&!! !+J@AM+ *,)-!**,)-!+! !8A=MM=ICAHAIO++ +++++++A.-1+4APFUHEA + $ ;\$(,$)' $OZZS`$ORcZbS\$ BC'KLSL $3?>$Wab$RS`$5]`S$4W\RW\U$8OYb]`$!548"$PSb`]TTS\#$ eOa$ WV\$ hc$ SW\S[$ RS`$ VjcTWUabS\$ LWSZSab`cYbc`S\$ Tl`$ >ScYj[WS$Oaa]hWWS`bS$ HS`j\RS`c\US\$[OQVb%$M)(#$))N$$ 6S`$548$Wab$SW\$VSbS`]RW[S`S`$F`O\aY`W^bW]\aTOYb]`#$hcaO[[S\USaSbhb$Oca$RS`$6@3$ PW\RS\RS$ G\bS`SW\VSWb$ DG@J($ !9S\Z]Qca$ OcT$ 5V`][]a][$ )("$ c\R$ RS[$ DG@J($ PW\RS\RS\$ 5484$ !9S\Z]Qca$ OcT$ 5V`][]a][$ (-"%$ 5484$ abOPWZWaWS`b$ c\R$ W\bS`OUWS`b$ [Wb$ DG@J($ c\R$ abSWUS`b$ RORc`QV$ RSaaS\$ 3TTW\Wbjb$ hc`$ 6@3%$ 6WSaS`$ =][^ZSf$ Wab$ SaaS\bWSZZ$Tl`$SW\S$\]`[OZS$[gSZ]WaQVS$3ca`SWTc\U%$ 6WS$ 548$ >ScYj[WS\$ RSTW\WS`S\$ aWQV$ Rc`QV$ RS\$ @OQVeSWa$ d]\$ QV`][]a][OZS\$ DSO``O\US[S\ba#$eSZQVS$hc$8caW]\aUS\S\$c\R$8caW]\a^`]bSW\S\$TlV`S\#$eSZQVS$RWS$$ ^VgaW]Z]UWaQVS$ F`O\aY`W^bW]\aTOYb]`$3YbWdWbjb$ RSa$ 548$ PSSW\b`jQVbWUS\2$ RWS$ 7f^`SaaW]\$RS`$LWSZUS\S$eW`R$c\bS`R`lQYb$!F`O\aY`W^bW]\a`S^`Saa]`"$c\R$Sa$`SacZbWS`b$ SW\$ 6WTTS`S\hWS`c\UaPZ]QY$ W[$ EW\\S$ SW\S`$ USabk`bS\$ [gSZ]WaQVS\$ 3ca`SWTc\U$ !3PP+"%M)(N$@OQV$7>@$c\R$?D5$=`WbS`WS\$hjVZb$RWSaS`$EcPbg^$hc$RS\$^`]U\]abWaQV$ Ul\abWUS\$>ScYj[WS\%M-#$.#$0N$$ $ ! UV! ,>>"+ '+ $ 6WS$ 8caW]\$ d]\$ !*2$-.&+*&9;& 5484&?K:((#$ eWS$ OcQV$ %'$(*$ DG@J(&DG@J(F($ PSSW\b`jQVbWUb$ $$$$$$#)"**$ "!"$ RWS$^VgaW]Z]UWaQVS$F`O\aY`W^bW]\a$ 666" " "666& /#-%$/&%$*$.& TOYb]`$3YbWdWbjb$ RSa$ 5482$ RWS$ 7f^`SaaW]\$ RS`$ LWSZUS\S$ eW`R$ "-#*.(+'#0+*&<5:9& c\bS`R`lQYb$ c\R$ Sa$ `SacZbWS`b$ SW\$ 8+-$,-$..+-&8+),($3& 6WTTS`S\hWS`c\UaPZ]QY$ W[$ EW\\S$ "!"$!" $ %'$(*$$$$$$$$%'$(*&*$(*&*$ SW\S`$ USabk`bS\$ [gSZ]WaQVS\$ 666" " "666& /#-%$/&%$*$.& 3ca`SWTc\U%$ $ 7\ba^`SQVS\R$RSa$XSeSWZa$hcU`c\RS$ZWSUS\RS\$DSO``O\US[S\ba$ZOaaS\$aWQV$RWS$548$ >ScYj[WS\$ W\$ heSW$ 9`c^^S\$ c\bS`bSWZS\1$ 6c`QV$ SW\S$ POZO\QWS`bS$ F`O\aZ]YObW]\$ heWaQVS\$ 5V`][]a][$ /$ c\R$ 5V`][]a][$ )($ PWZRSb$ aWQV$ ROa$ 8caW]\ab`O\aY`W^b$ 7:4<$ 7:4<$9$ $ Mb!/2)("!_))2_))"N%$ 6WSaS$ hgb]US\SbWaQVS$ 3PS``ObW]\$ Wab$ [Wb$ RS[$ 834$ EcPbg^$ ?)$ Oaa]hWWS`b%$ 6c`QV$ SW\S$ ;\dS`aW]\$ RSa$ 5V`][]a][$ (-$ MW\d!(-"!^(*2_))"N$ Phe%$ SW\S$F`O\aZ]YObW]\$ heWaQVS\$ RS\$ PSWRS\$ 5V`][]a][S\$ (-$ Mb!(-2(-"!^(*2_))"N$ PWZRSb$ aWQV$ ROa$ +*.* 3=/$$ $ 8caW]\aUS\#$ eSZQVSa$ RS\$ 834$ EcPbg^$3?>$?+$[Wb$7]aW\]^VWZWS$!834$?+S]"$RSTW\WS`b%$M)(#$))N$$ ?Oca[]RSZZS$ hSWUbS\#$ ROaa$ RWS$ 7f^`SaaW]\$ RWSaS`$ 8caW]\aUS\S#$ OZa$ 3?>$ RSTW\WS`S\RS$ HS`j\RS`c\U#$ \WQVb$ Oca`SWQVS\#$ c[$ SW\S$ >ScYj[WS$ OcahcZkaS\%$ LcajbhZWQVS$ ?cbObW]\S\$ W\$ PSabW[[bS\$ EWU\OZeSU$B`]bSW\S\#$ RWS$ a]US\O\\bS\$ B`]ZWTS`ObW]\a$6`WdS`#$ ec`RS\$ PSaQV`WSPS\%$ 6WSaS$ Y]]^S`WS`S\$ [Wb$ RS\$ ]\Y]US\S\$ 8caW]\aUS\S\#$ lPS`eW\RS\$ a]$ RS\$ IOQVabc[aPZ]QY$ c\R$ b`WUUS`\$ RWS$ [OZWU\S$ F`O\aT]`[ObW]\%M(,#$ (-#$ )(N$ 6WSa$ S\ba^`WQVb$ RS[$ YZOaaWaQVS\$ ?]RSZZ$ RS`$ BObV]US\SaS$ RS`$ 3?>#$ SW\S`$ =][PW\ObW]\$ Oca$ SW\S`$ Fg^$ ($ ?cbObW]\$ !B`]ZWTS`ObW]\aabSWUS`c\U"$ c\R$ SW\S`$ Fg^$ )$ ?cbObW]\$ !6WTTS`S\hWS`c\UaPZ]QY"%M)*N$ ;\$ 0' $OZZS`$548$>ScYj[WS$BObWS\bS\$Y]\\bS\$?cbObW]\S\$W\$ 109 #$ .29& $c\R&$]RS`$ 7)8 $ \OQVUSeWSaS\$ eS`RS\%M)+#$ ),N @WQVbaRSab]b`]bh$ ec`RS$ W\$ (' $ RWSaS`$ BObWS\bS\$ PWaZO\U$YSW\S$RS`$PSYO\\bS\$Y]]^S`WS`S\RS\$?cbObW]\S\$USTc\RS\%$ $ ! UW! $"&"$"+3JJKAMEAMAI@A+5PO=OEJIAI+EI+.-1+4APFUHEAI++ $ 6WS$ BObV]US\SaS$ d]\$ 548$ >ScYj[WS\$ aQVSW\b$ OcT$ RS\$ S`abS\$ 4ZWQY$ OcT$ SW\S\$ lPS`aQVOcPO`S\$ ?SQVO\Wa[ca$ Oca$ q\c`p$ heSW$ Y]]^S`WS`S\RS\$ ?cbObW]\S\$ hc$ PS`cVS\%M(#$ )-N$ $ Lc[$ SW\S\$ Y]\abWbcbWd$ OYbWdWS`S\RS$ ?cbObW]\S\$ W\$ Fg`]aW\YW\OaS\$ !h%4%$ .29& #$ 109 "$ ]RS`$ WV`S\$ R]e\ab`SO[$ 7TTSYb]`S\$ !h%4%$ 7)8 #$ +*2 #$ 6964$$ "#$ eSZQVS$ Tl`$ RWS$ c\USVS[[bS$ B`]ZWTS`ObW]\$ dS`O\be]`bZWQV$ aW\R$ c\R$ a]$ RS\$ iPS`ZSPS\ad]`bSWZ$ dS`[WbbSZ\%$ G\R$ hc[$ O\RS`S\$ HS`j\RS`c\US\$ d]\$ B`]bSW\S\$ eSZQVS$RWS$Vj[Ob]^]SbWaQVS$6WTTS`S\hWS`c\U$c\R&$]RS`$RWS$3^]^b]aS$PSSW\b`jQVbWUS\$ c\R$RWS$8jVWUYSWb$hc`$ESZPabS`\ScS`c\U$dS`[WbbSZ\%$6WSaS$eS`RS\$W[$8OZZ$RS`$548$ >ScYj[WS\$ Rc`QV$ RWS$ 8caW]\ab`O\aY`W^bS$ 7:4<$ 7:4<$9$ $ Phe%$ +*.* 3=/$$ $ dS`b`SbS\$c\R$abSZZS\$RWS$9`c\RZOUS$Tl`$SW\S$[OZWU\S$F`O\aT]`[ObW]\$RO`%$$ $ 7\ba^`SQVS\R$ RWSaS`$ 3\\OV[S$ USaQVWSVb$ RWS$ YZ]\OZS$ 7f^O\aW]\$ RSa$ W\WbWWS`S\RS\$ =Z]\a$ [Wb$ +*.* 3=/$$ $$ ]RS`$ 7:4<$ 7:4<$9$ $DSO``O\US[S\b$ OZa]$ lPS`$ Y]$ ]^S`WS`S\RS#$ aSYc\Rj`S$ HS`j\RS`c\US\#$ VOc^bajQVZWQV$ W\$ 1096' .29& $ c\R$ 7)8! M)+#$ ),#$).#$)/N$ $ ;\aPSa]\RS`S$ Tl`$ 548$ >ScYj[WS\$ [Wb$ 7:4<$ 7:4<$9$ $DSO``O\US[S\b$ Y]\\bS$ S`ab$Yl`hZWQV$UShSWUb$eS`RS\#$ROaa$ROa$?cbObW]\a^`]TWZ$RWSaS`$EcPU`c^^S$Y][^ZSfS`$ Wab#$OZa$c`a^`l\UZWQV$O\US\][[S\$c\R$ROa$OcQV$?cbObW]\S\$OcTb`SbS\#$RWS$\WQVb$OZa$ Fg^$($]RS`$)$YZOaaWTWhWS`b$eS`RS\$Yk\\S\%$;\$,' $RS`$BObWS\bS\$ec`RS\$?cbObW]\S\$ W\$ a]U%$ S^WUS\SbWaQVS\$ DSUcZOb]`S\$ )8<2$ #$ )8<2% $ c\R$ 9-9% $ USTc\RS\%M)0$*(N$ ISWbS`S$ )+ $ b`OUS\$ ?cbObW]\S\$ W\$ 9S\S\$ RSa$ 5]VSaW\$ =][^ZSf%M*)N$ >*9*() $ ?cbObW]\S\#$ RWS$ RS\$ Fc[]`ab]TTeSQVaSZ$ PSSW\TZcaaS\#$ ec`RS\$ W\$ )* $ OZZS`$ >ScYj[WS\$RWSaSa$EcPbg^a$USTc\RS\%M**N$ 3ca$ RWSaS[$ IWaaS\$ S`USPS\$ aWQV$ \WQVb$ \c`$ \ScS$ 9SaWQVba^c\YbS$ hc`$ DWaWY]ab`ObWTWYObW]\$ RS`$ YZW\WaQV$ aSV`$ VSbS`]US\S\$ 9`c^^S$ RS`$ 548$ >ScYj[WS\#$ a]\RS`\$ OcQV$ ?kUZWQVYSWbS\$ hc`$ ^S`a]\OZWaWS`bS\$ FVS`O^WS%$ 3Za$ 4SWa^WSZ$ aSWS\$ BObWS\bS\$ US\O\\b#$ RWS$ hcajbhZWQVS$ Bc\Yb[cbObW]\S\$ O\$ B]aWbW]\$ 6/(-$ RS`$ DShS^b]`bg`]aW\YW\OaS$=;F$b`OUS\%$8l`$ +*.* 3=/$$ $#$eWS$OcQV$ 7:4<$ 7:4<$9$ $ `SO``O\UWS`bS$>ScYj[WS$hSWUb$aWQV$VWS`$SW\$aWU\WTWYO\b$Yl`hS`Sa$DShWRWdT`SWSa$#$a]eWS$ 9SaO[blPS`ZSPS\%M*(#$*+N$;\$YZW\WaQVS\$BVOaS$;;$c\R$;;;$EbcRWS\$eW`R$Tl`$RWSaS$DWaWY]$ ! UX! =]V]`bS$ OYbcSZZ$ RS`$ 7W\aObh$ d]\$ 6OaObW\WP#$ SW\S[$ Fg`]aW\YW\OaSW\VWPWb]`#$ OZa$ 7`VOZbc\UabVS`O^WS$USbSabSb%M*,N$$ $ $"'+/=MNOAGGPIC+@AN+1JMN?DPICNQJMD=>AIN+PI@+0ECAI=IOAEG+@AM+,M>AEOAI+ 6WSaS$ 6WaaS`bObW]\$ PSaQVjTbWUb$ aWQV$ [Wb$ OcaUSejVZbS\$ QV`][]a][OZS\$ DSO``O\US[S\ba#$ WV`S`$ 3caeW`Yc\U$ OcT$ RWS$ Vj[Ob]^]SbWaQVS$ 7\beWQYZc\U#$ WV`$ LcaO[[S\a^WSZ$ [Wb$ hcajbhZWQVS\$ []ZSYcZO`US\SbWaQVS\$ HS`j\RS`c\US\$ c\R$ [kUZWQVS\$ 4SVO\RZc\Ua]^bW]\S\$ W[$ EW\\S$ SW\S`$ hWSZUS`WQVbSbS\$ FVS`O^WS$ RS`$ W\RWdWRcSZZS\$3?>%$ 6WS$S`abS$BcPZWYObW]\$PSb`OQVbSb$ROa$ +)23").$# $DSO``O\US[S\b#$eSZQVSa$SW\S$aSV`$ YZSW\S$ EcPU`c^^S$ RS`$ OYcbS\$ >ScYj[WS\$ RSTW\WS`b$ c\R$ [Wb$ SW\S`$ c\Ul\abWUS\$ B`]U\]aS$ Oaa]hWWS`b$ Wab%$ ISZQVS$ 6][j\S\$ RS`$ PSWRS\$ 8caW]\a^O`b\S`$ hc`$ >ScYj[WSS\babSVc\U$PSWb`OUS\$eO`$8`OUSabSZZc\U$RWSaS`$HS`kTTS\bZWQVc\U$c\R$ec`RS$ [WbbSZa$ Tc\YbW]\SZZS`$ 3\OZgaS\$ GK' SGQOL $ c\R$ GK' SGSL $ W[$ =\]QVS\[O`Y$ F`O\a^ZO\bObW]\a[]RSZZ$W\$RS`$?Oca$c\bS`acQVb%$ 6S`$ 7WUS\O\bSWZ$ RWSaS`$ BcPZWYObW]\$c[TOaab$ d]`$ OZZS[$ RWS$ bSQV\WaQVS$ 6c`QVTlV`c\U$ RS`$7f^S`W[S\bS%$=Z]\WS`c\U$c\R$:S`abSZZc\U$RS`$RWdS`aS\$+)23 ).$# $=]\ab`cYbS#$ 9SeW\\c\U$ d]\$ =\]QVS\[O`Y$ Oca$ E^S\RS`[jcaS\$ c\R$ =cZbWdWS`c\U$ CT' SGSL #$ `Sb`]dW`OZS$ F`O\aRcYbW]\$ d]\$ =\]QVS\[O`YahSZZS\#$ A^TS`c\U$ RS`$ S`Y`O\YbS\$ 7[^Tj\US`[jcaS$ c\R$ B`j^O`ObW]\$ d]\$ :S`hPZcb#$ =\]QVS\[O`Y$ c\R$ ?WZh#$ [Wb$ \OQVT]ZUS\RS`$ 835E$ G\bS`acQVc\U$ OcaUSejVZbS`$ ?O`YS`#$ a]eWS$ RWS$ 6c`QVTlV`c\U$ c\R$ 3caeS`bc\U$ d]\$ 585$3aaOga%$ 6WS$ bSQV\WaQVS\$ 6SbOWZa$ aW\R$ RS[$ ?SbV]RS\bSWZ$ RS`$HS`kTTS\bZWQVc\U$hc$S\b\SV[S\%$ 6WS$ heSWbS$ BcPZWYObW]\$ PSTOaab$ aWQV$ [Wb$ @9E$ ?SbV]RS\#$ W\aPSa]\RS`S$ RS`$ IV]ZS$ 7f][S$ ES_cS\hWS`c\U$ !I7E"$ hc`$ agabS[ObWaQVS\$ ;RS\bWTWYObW]\$ Y]]^S`WS`S\RS`$ ?cbObW]\S\$W\$3?>$BObWS\bS\$[Wb$ +*.* 3=/$$ $8caW]\%$ 6S`$ 7WUS\O\bSWZ$ c[TOaab$ RWS$ ?WbO`PSWb$ PSW$ RS`$ 5VO`OYbS`WaWS`c\U$ d]\$ BObWS\bS\^`]PS\#$ O\VO\R$ hgb]US\SbWaQVS`$ c\R$ []ZSYcZO`hgb]US\SbWaQVS`$ !8;E:"$ G\bS`acQVc\US\$c\R$ROa$LcaO[[S\abSZZS\$RS`$R`SW$BObWS\bS\Y]V]`bS\%$LcRS[$RWS$ ! UY! =]\hS^bW]\$ RSa$ B`]XSYba#$ B`j^O`ObW]\$ RS`$ B`]PS\$>WP`O`WSa$ c\R$ 3caeS`bc\U$ RS`$ I7E$ 6ObS\%$ 6WS$ HOZWRWS`c\U$ RS`$ 7`USP\WaaS$ [WbbSZa$ YZOaaWaQVS`$ EO\US`$ ES_cS\hWS`c\U#$ROa$EQ`SS\W\U$RS`$R`SW$3?>$=]V]`bS\$OcT$`SYc``S\bS$?cbObW]\S\$W[$ 7f]\$ (-$ RSa$ .29& $ 9S\a$ c\R$ RWS$ 4SabW[[c\U$ RS`$ H38$ [WbbSZa$ :]QVRc`QVaObhaS_cS\hWS`c\U$ !bO`USbSR$ `SaS_cS\QW\U"$ RS`$ [cbWS`bS\$ BObWS\bS\$ B`]PS\%$ISWbS`VW\$RWS$3caeS`bc\U$c\R$;\bS`^`SbObW]\$RS`$ XSeSWZa$ S`V]PS\S\$ 6ObS\$ c\R$ ROa$ EQV`SWPS\$ RSa$ ?O\caY`W^ba#$ W\YZcaWdS$ RS`$ U`OTWaQVS\$ 6O`abSZZc\U$ RS`$ SWUS\S\$7`USP\WaaS%$ 6WS$W[$heSWbS\$3`bWYSZ$PSaQV`WSPS\S\$Tc\YbW]\SZZS\$3`PSWbS\$ec`RS\$d]\$6`%$B]ZhS`$ Rc`QVUSTlV`b%$ 7`$ RS[]\ab`WS`bS$ ROa$ ZScYj[]US\S$ B]bS\bWOZ$ RS`$ .29& $ @-.-=$ ?cbObW]\$ c\R$ hSWUbS\$RWS$?kUZWQVYSWb$ SW\S`$ hWSZUS`WQVbSbS\$ FVS`O^WS$ Tl`$ RWSaS$ 3?>$ EcPU`c^^S$ OcT%$ 4SWRSa$ abSWUS`bS$ RS\$ IS`b$ RS`$ SWUS\S\$ 3`PSWb#$ a]$ ROaa$ RWS$ 7`abOcb]`aQVOTb$ RS`$ ?O\caY`W^ba#$ S\ba^`SQVS\R$ RS`$ UZSWQVeS`bWUS\$ eWaaS\aQVOTbZWQVS\$>SWabc\US\#$USbSWZb$ec`RS%$ + %++ :jcTWU$ ZWSUS\$ RS`$ 3?>$ hgb]US\SbWaQVS$ HS`j\RS`c\US\$ hcU`c\RS%$ 6WSaS$ OZZSW\$ aW\R$ XSR]QV$\WQVb$Oca`SWQVS\R$Tl`$RWS$[OZWU\S$F`O\aT]`[ObW]\#$PWSbS\$OPS`$SW\S$9`c\RZOUS$ Tl`$ RS\$ :W\hcUSeW\\$ eSWbS`S`$ ?cbObW]\S\$ c\R$ RWS$ S\RUlZbWUS$ ?O\WTSabObW]\$ RS`$ >ScYj[WS%$ LcRS[$ ZOaaS\$ QV`][]a][OZS$ HS`j\RS`c\U$ DlQYaQVZlaaS$ OcT$ FVS`O^WSO\a^`SQVS\$ c\R$ iPS`ZSPS\$ hc%$ 3cT$ RWSaS$ DWaWY]ab`ObWTWYObW]\$ Tcmb$ RWS$ 7\baQVSWRc\U$Tl`$]RS`$eWRS`$SW\S$a^ShWTWaQVS$FVS`O^WS%$ $ 6WS$S`abS$BcPZWYObW]\$PSb`OQVbSb$ROa$ +)23").$# $DSO``O\US[S\b#$eSZQVSa$SW\S$aSV`$ YZSW\S$ EcPU`c^^S$ RS`$ OYcbS\$ >ScYj[WS\$ RSTW\WS`b$ c\R$ [Wb$ SW\S`$ c\Ul\abWUS\$ B`]U\]aS$ Oaa]hWWS`b$ Wab%$ IW`$ Y]\\bS\$ hSWUS\#$ ROa$ RWS$ 8caW]\$ RS`$ 5$bS`[W\OZS\$ QZObV`W\$PW\RW\U$6][j\S$d]\$ +)23 $O\$RWS$]QbO^S^bWR$[]bWT$$$ZScQW\$hW^^S`$6][j\S$ !A?$>L"$d]\$ ).$# $M[W\W[OZ$TcaW]\N$\]beS\RWU#$c\R$UZSWQVhSWbWU$Oca`SWQVS\R$Wab#$SW\S$ ^`W[j`S$ [c`W\S$ =\]QVS\[O`YahSZZS$ hc$ b`O\aT]`[WS`S\$ c\R$ SW\S$ 3?>$ OcahcZkaS\%$ 6WS$ 5ZObV`W\$ 4W\RW\U$ DSUW]\$!OcT$ 5V`][]a][$ ((2$ +)23 "$c\R$RWS$A?$>L$6][j\S$ ! UZ! !OcT$ 5V`][]a][$ ('2$ ).$# "$ aW\R$ RWS$ OcaaQVZOUUSPS\RS\$ EQVZlaaSZabSZZS\$ RWSaSa$ V]QVOUU`SaaWdS\$ DSO``O\US[S\ba$ c\R$ S\baQVSWRS\R$ Tl`$ ROa$ b`O\aT]`[WS`S\RS$ B]bS\bWOZ$RWSaSa$8caW]\a]\Y]US\a%$$ 6WS$ d]`ZWSUS\RS$ ab`cYbc`SZZS$ 3\OZgaS$ RS`$ Tc\YbW]\SZZS\$ 6][j\S\$ d]\$ 53>?&38('$ ZSUbS$RS\$9`c\RabSW\$Tl`$SW\S$UO\hS$DSWVS$O\$8]ZUSc\bS`acQVc\US\%$6O$\c\$RWS$Tl`$ RWS$ >ScYj[WS$ eWQVbWUS\$ 6][j\S\$ TSababO\RS\#$ ec`RS$ \OQV$ ;\bS`OYbW]\a^O`b\S`\$ RWSaS`$ 6][j\S\$ USacQVb#$ c[$ RS\$ ?SQVO\Wa[ca$ RS`$ +)23").$# $ >ScYj[WS$ hc$ S\baQVZlaaSZ\%$ 7a$ TO\R$ aWQV$ RS`$ S^WUS\SbWaQVS$ DSUcZOb]`$ ,59$2 $ OZa$ :Oc^bOYbSc`%$ 3cTPOcS\R$ OcT$ RWSaS$ 7`YS\\b\WaaS$ Y]\\bS$ heWaQVS\hSWbZWQV$ [Wb$ RS`$ aSZSYbWdS$ ;\VWPWbW]\$ RS`$ E$!,r$ORS\]agZ"$Z$[SbVW]\W\$ !E3?"$ 4W\Rc\UaOYbWdWbjb$ RWSaS`$ ?SbVgZb`O\aTS`OaS$ Rc`QV$ RWS$ Eb`cYbc`O\OZ]UO$ 7BL''+...$ c\R$ 7BL$,-.-$ SW\S$ FVS`O^WS[kUZWQVYSWb$ RS`$ OPS``O\bS\$ ?SbVgZWS`c\U$ c\R$ 3YbWdWS`c\U$ d]\$ >ScYj[WS$ Oaa]hWWS`bS\$9S\S\$USTc\RS\$eS`RS\%$ $ 6WS$ heSWbS$ BcPZWYObW]\$ PSTOaab$ aWQV$ [Wb$ RS`$ agabS[ObWaQVS\$ ;RS\bWTWYObW]\$ Y]]^S`WS`S\RS`$?cbObW]\S\$W\$3?>$BObWS\bS\$[Wb$ +*.* 3=/$$ $8caW]\%$ IWS$ hcd]`$ PSaQV`WSPS\#$ Wab$ RWS$ OZZSW\WUS$ 7f^`SaaW]\$ RS`$ 8caW]\aUS\S$ \WQVb$ Oca`SWQVS\R#$ c[$ SW\S$ >ScYj[WS$ OcahcZkaS\%$ ;\aPSa]\RS`S$ W\$ 548$ >ScYj[WS\$ Y]\\bS\$ W\$PWa$ hc$ 0' $ RS`$ BObWS\bS\$ hcajbhZWQVS$ ?cbObW]\S\$W\$9S\S\#$eSZQVS$Tl`$ EWU\OZ^`]bSW\S$Q]RWS`S\$!h%4%$ 109 #$ .29& $c\R&$]RS`$ 7)8 "$c\R$\]beS\RWU$aW\R$c[$RS\$ IOQVabc[aPZ]QY$hc$lPS`eW\RS\#$USTc\RS\$eS`RS\%$@WQVbaRSab]b`]bh$USZO\U$W\$(' $ RS`$ BObWS\bS\$ PWaZO\U$ YSW\$ @OQVeSWa$ SW\S`$ RWSaS`$ PSYO\\bS\$ Y]]^S`WS`S\RS\$ ?cbObW]\S\%$ IW`$ c\bS`acQVbS\$ =\]QVS\[O`Y$ SW\Sa$ a]ZQVS\$ BObWS\bS\$ [Wb$ hgb]US\SbWaQV$ PSaQV`WSPS\S`$;\dS`aW]\$d]\$5V`][]a][$(-#$]V\S$hcajbhZWQVS$[]ZSYcZO`US\SbWaQVS$ c\R&$ ]RS`$ hgb]US\SbWaQVS$ HS`j\RS`c\US\$ W\$ RS`$ D]cbW\SRWOU\]abWY%$ ?WbbSZa$ I7E$ dS`UZWQVS\$eW`$aSW\S$6@3$!7f][S1$RS`$B`]bSW\S$Y]RWS`S\RS$3\bSWZ$RS`$US\][WaQVS\$ ES_cS\h"$ hc[$ LSWb^c\Yb$ RS`$ 7`abRWOU\]aS$ [Wb$ RS`$ 6@3$ hc[$ LSWb^c\Yb$ RS`$ =][^ZSbbS\$DS[WaaW]\$!S\ba^`SQVS\R$SW\S`$=SW[POV\Y]\b`]ZZS"%$?Wb$RWSaS[$3\aObh$ TO\RS\$ eW`$ SW\S$ \ScS$ ?cbObW]\$ W\$ RS`$ S`abS\$ Fg`]aW\YW\OaSR][j\S$ d]\$ .29& #$ US\OcS`$W\$B]aWbW]\[email protected]=#$eSZQVS$[Wb$RS`$lPZWQVS\$D]cbW\SRWOU\]abWY$\WQVb$S`TOaab$ ! U[! eW`R%$;\$SW\S`$=]V]`bS$d]\$()'$548$3?>$BObWS\bS\$Y]\\bS\$eW`$RWSaS$HS`j\RS`c\U$ W\$ W\aUSaO[b$ -$ BObWS\bS\$ !, "$ \OQVeSWaS\%$ LcRS[$ Y]\\bS\$ eW`$ W\$ Tc\YbW]\SZZS\$ 3\OZgaS\$hSWUS\#$ROaa$RWSaS$?cbObW]\$c\OPVj\UWUSa$IOQVabc[$dS`[WbbSZb$c\R$ROaa$ aWS$aS\aWPSZ$USUS\lPS`$RS\$PSYO\\bS\$Fg`]aW\YW\OaSW\VWPWb]`S\$!F=;"$Wab%$BObWS\bS\$ [Wb$RWSaS`$?cbObW]\$S`kTT\Sb$aWQV$a][Wb$SW\S$hcajbhZWQVS$4SVO\RZc\Ua]^bW]\$[Wb$F=;%$ 6WS$YZW\WaQVS$EWU\WTWYO\h$RS`$\ScS\$ .29& [email protected]=$?cbObW]\$eW`R$c\bS`ab`WQVS\$Rc`QV$ G\bS`acQVc\US\$ d]\$ :cO\U$ Sb$ OZ%$ RWS$ hSWUS\$ Y]\\bS\#$ ROaa$ ROa$ b`O\aT]`[WS`S\RS$ B]bS\bWOZ$ RS`$ .29& $ @-.-=$ ?cbObW]\$ OZZSW\$ OcQV$ GK' SGSL $ Oca`SWQVb$ SW\S$ OYcbS$ >ScYj[WS$hc$W\RchWS`S\%M*-N$ $ ;\$PSWRS\$3`PSWbS\$ec`RS$SW\$eWQVbWUS`$4SWb`OU$hc`$RS\$9`c\RZOUS\$RS`$>ScYj[WS$ 7\babSVc\U$ PSW$ BObWS\bS\$ [Wb$ a^ShWTWaQVS\$ QV`][]a][OZS\$ F`O\aZ]YObW]\S\$ USZSWabSb%$6WS$;RS\bWTWhWS`c\U$ZScYj[WS$`SZSdO\bS`$B`]bSW\R][j\S\$c\R$RS`$@OQVeSWa$ \ScS`$ ?cbObW]\S\$ W\$ aWU\OZW\U$ ^ObVeOg$ 9S\S\$ ! .29& $ @-.-="#$ RWS$ [Wb$ RS\$ QV`][]a][OZS\$ DSO``O\US[S\ba$ Y]]^S`WS`S\#$ S`kTT\S\$ \WQVb$ hcZSbhb$ \ScS$ FVS`O^WS[kUZWQVYSWbS\$W[$EW\\S$SW\S`$bO`USbSR$bVS`O^g%$ 6WS$;RS\bWTWYObW]\$RSa$W\RWdWRcSZZS\$?cbObW]\a^`]TWZa$SW\S`$3?>$Wab$\WQVb$\c`$SaaS\bWSZZ$ Tl`$ RWS$ 6WOU\]aS#$ a]\RS`\$ dWSZ[SV`$ Tl`$ RWS$ PW]Z]UWaQVS$ EcP$=ZOaaWTWYObW]\#$ RWS$ B`]U\]aS$ c\R$ DWaWY]ab`ObWTWYObW]\#$ Tl`$ RWS$ ]^bW[OZS$ bVS`O^ScbWaQVS$ Eb`ObSUWS$ !FVS`O^WSS\baQVSWRc\U"$ c\R$ ROa$ W\RWdWRcSZZS$ YZW\WaQVS$ ?]\Wb]`W\U$ RS`$ 4SVO\RZc\UaSTTWhWS\h$Rc`QV$4SabW[[c\U$RS`$?W\W[OZS\$DSabS`Y`O\Yc\U$!?D6"%$ $ $ &++9PHH=MS+ 3$ZO`US$\c[PS`$]T$`SQc``W\U#$POZO\QSR$Qgb]US\SbWQ$OP\]`[OZWbWSa$O`S$T]c\R$W\$3?>%$ 7dS\$bV]cUV$bVS$US\S$^`]RcQba$]T$[O\$]T$bVSaS$`SO``O\US[S\ba$O`S$\]b$acTTWQWS\b$b]$ W\RcQS$ZScYS[WO$Pg$bVS[aSZdSa#$bVSg$T]`[$bVS$[]ZSQcZO`$POaWa$T]`$OQQc[cZObW]\$]T$ ORRWbW]\OZ$ [cbObW]\a$ O\R$ [OZWU\O\b$ b`O\aT]`[ObW]\%$ 5gb]US\SbWQ$ QVO\USa$ O`S$ W[^]`bO\b$T]`$^`]U\]abWQ$ab`ObWTWQObW]\$O\R$W\T]`[$b`SOb[S\b$RSQWaW]\$O\R$QO\$^`SRWQb$ ac`dWdOZ%$ FVS$ ^`SaS\b$ [O\caQ`W^b$ RSaQ`WPSR$ a^SQWTWQ$ `SO``O\US[S\ba#$ bVSW`$ W[^OQb$ ]\$ ! U\! VS[Ob]^]SaWa#$ bVSW`$ Q]]^S`ObW]\$ eWbV$ ORRWbW]\OZ$ [cbObW]\a$ O\R$ b`SOb[S\b$ ]^bW]\a$ `SUO`RW\U$bO`USbSR$bVS`O^WSa%$ FVS$ +)23").$# $ TcaW]\$ US\S$ `SacZba$ T`][$ O$ QV`][]a][OZ$ b`O\aZ]QObW]\$ PSbeSS\$ QV`][]a][Sa$ ('$ O\R$ (($ Mb!('2(("!^(*$(+2_(+$)("N#$ O$ `SQc``S\b$ QV`][]a][OZ$ `SO``O\US[S\b$ ]PaS`dSR$ W\$ ^ObWS\ba$ eWbV$ ^]]`$ ^`]U\]aWa$ OQcbS$ [gSZ]WR$ ZScYS[WO$ !3?>"$ O\R$ OQcbS$ Zg[^V]PZOabWQ$ ZScYS[WO$ !3>>"#$ Sa^SQWOZZg$ F$3>>$ eWbV$ O$ F$QSZZ$ `SQS^b]`$UO[[O&RSZbO$`SO``O\US[S\b%$E]$TO`#$RWTTS`S\b$ +)23").$# $TcaW]\$b`O\aQ`W^ba$ VOdS$PSS\$RSaQ`WPSR$W\$3?>$^ObWS\ba%$F]$RSTW\S$bVS$Q]\b`WPcbW]\$]T$RWTTS`S\b$53>?$ O\R$ 38('$ ^`]bSW\$ R][OW\a$ b]$ GK' SGSL $ ZScYS[]US\SaWa$ eS$ ^S`T]`[SR$ RSbOWZSR$ ab`cQbc`S$Tc\QbW]\$abcRWSa$]T$bVS$53>?&38('$TcaW]\$^`]bSW\%$8]`$bVWa#$eS$US\S`ObSR$O$ TcaW]\$ ^`]bSW\$ !bS`[SR$ 53>?&38('$ [W\W[OZ$ TcaW]\"$ ]T$ bVS$ 5$bS`[W\OZ$ )+/$ O[W\]$ OQWRa$]T$53>?#$eVWQV$W\QZcRS$bVS$QZObV`W\$PW\RW\U$R][OW\#$O\R$bVS$]QbO^S^bWRS$[]bWT$ $$ ZScQW\S$ hW^^S`$ !A?$>L"$ R][OW\$ ]T$ 38('%$ FVWa$ [W\W[OZ$ TcaW]\$ ^`]bSW\$ [SRWObSR$ b`O\aT]`[ObW]\$QO^OPWZWbWSa$W\$Q]Z]\g$T]`[W\U$QSZZ$OaaOga$GK'SGQOL $O\R$ GK'SGSL %$ 8c`bVS`[]`S#$ ZScYS[WOa$ W\RcQSR$ Pg$ bVS$ +)23").$# $ [W\W[OZ$ TcaW]\$ `SQO^WbcZObSR$ W[^]`bO\b$ Oa^SQba$ ]T$ TcZZ$ZS\UbV$ +)23").$# $W\RcQSR$ ZScYS[WO#$ W\QZcRW\U$ S\VO\QSR$ /5<) $ QZcabS`$ Sf^`SaaW]\%$ FVWa$ Wa$ bVS$ QVO`OQbS`WabWQ$ aWU\Obc`S$ ]T$ +)23").$# $ ZScYS[WOa$ RcS$ b]$ :*=.0$ [SbVgZObW]\#$ ]\S$ ]T$ bVS$ [SQVO\Wa[a$ PSVW\R$ ZScYOS[WO$ QOcaSR$Pg$ +)23").$# $TcaW]\$O\R$OPS``O\b$6AF(>$Tc\QbW]\%$ ;\$ac[[O`g#$bVWa$abcRg$W\RWQObSa$bVOb$bVS$QZObV`W\$PW\RW\U$R][OW\$]T$53>?$O\R$bVS$ A?$>L$R][OW\a$]T$38('$O`S$acTTWQWS\b$b]$W\RcQS$3?>%$FVWa$TW\RW\U$Tc`bVS`$acUUSaba$ bVOb$ O^^`]OQVSa$ b]$ W\VWPWb$ +)23").$# $W\RcQSR$ b`O\aT]`[ObW]\$ aV]cZR$ W\Q]`^]`ObS$ ab`ObSUWSa$ b]$PZ]QY$ bVSaS$ YSg$ R][OW\a%$ DSQS\bZg#$ bVS$ 6AF(>$38('$ W\bS`OQbW]\$ eWbV$ bVS$ A?$>L$ R][OW\$ eOa$ T]c\R$ b]$ PS$ SaaS\bWOZ$ T]`$ +)23").$# $ [SRWObSR$ b`O\aT]`[ObW]\#$ SabOPZWaVW\U$ O$ RW`SQb$ ZW\Y$PSbeSS\$ OPS``O\b$ :*=.0$[SbVgZObW]\$ O\R$ ZScYS[]US\SaWa%$FVca#$RWabc`PW\U$O\$S^WUS\SbWQ$[]RWTWQObW]\$QO\$OQb$Oa$O$R`WdS`$]T$ ZScYS[WO%$$ $ FVS$aSQ]\R$^cPZWQObW]\$caSR$\Sfb$US\S`ObW]\$aS_cS\QW\U$[SbV]Ra$b]$agabS[ObWQOZZg$ WRS\bWTg$Q]]^S`ObW\U$[cbObW]\a$W\$ +*.*"3=/$$ $`SO``O\USR$ZScYS[WO%$ ! U]! 5V`][]a][OZ$`SO``O\US[S\ba$OTTSQbW\U$bVS$Q]`S$PW\RW\U$TOQb]`a$ 7:4<$ $O\R$ +*.* #$ O`S$T]c\R$W\$(, $b]$)' #$]T$ORcZb$ BC'KLSL $OQcbS$[gSZ]WR$ZScYS[WO$!3?>"$QOaSa$O\R$ O`S$Oaa]QWObSR$eWbV$O$TOd]`OPZS$^`]U\]aWa%$ 3a$ [S\bW]\SR$ PST]`S#$ Sf^`SaaW]\$ ]T$ O$ TcaW]\$ US\S$ OZ]\S$ Wa$ []ab$ ]T$ bVS$ bW[S$ \]b$ acTTWQWS\b$b]$W\RcQS$[OZWU\O\b$b`O\aT]`[ObW]\%$7a^SQWOZZg#$W\$548$ZScYS[WOa$ORRWbW]\OZ$ []ZSQcZO`$[cbObW]\a$eS`S$T]c\R$W\$c^$b]$0' $]T$^ObWS\ba%$FVSaS$OPS``ObW]\a$OTTSQb$ US\Sa$]T$aWU\OZW\U$^ObVeOga$!S%U%$ 109 #$ .29& #$ 7)8 "#$Y\]e\$Oa$^`]ZWTS`ObW]\$R`WdS`a#$ eVWQV$O`S$`S_cW`SR$b]$]dS`Q][S$bVS$PZ]QY$]T$RWTTS`S\bWObW]\$ [SRWObSR$ Pg$ bVS$ TcaW]\$ US\Sa$ 7:4<$"7:4<$9$ $ O\R$ +*.*"3=/$$ %$ :]eSdS`#$ W\$ c^$ b]$ (' $ ]T$ 548$ 3?>$ ^ObWS\ba$\]\S$]T$bVS$Y\]e\$Q]]^S`ObW\U$[cbObW]\$Q]cZR$PS$RSbSQbSR%$$ F]$ WRS\bWTg$[cbObW]\a#$ eVWQV$ [Og$ Q]ZZOP]`ObS$ eWbV$ bVS$ +*.*"3=/$$ $ TcaW]\$ Rc`W\U$ ZScYS[]US\SaWa$eS$^S`T]`[SR$Sf][S$aS_cS\QW\U$]T$O\$3?>$aO[^ZS$eWbV$O\$W\d!(-"$ eWbV]cb$ ORRWbW]\OZ$ []ZSQcZO`$ ]`$ Qgb]US\SbWQ$ OPS``ObW]\a%$ IS$ RWaQ]dS`SR$ O$ \]dSZ$ @-.-=$[cbObW]\$W\$bVS$ORS\]aW\S$b`W^V]a^VObS$!3FB"$PW\RW\U$R][OW\$!F=6("$]T$bVS$ .29& $ US\S%$ ;\$ O$ Q]V]`b$ ]T$ /+$ BC' KLSL $ 3?>$ ^ObWS\ba$ eWbV$ O$ +*.*"3=/$$ $ `SO``O\US[S\b$O\R$W\$*-$^ObWS\ba$eWbV$O$ 7:4<$"7:4<$9$ $`SO``O\US[S\b#$bVS$ .29& $ @-.-=$[cbObW]\$eOa$WRS\bWTWSR$W\$,$O\R$($^ObWS\ba#$`Sa^SQbWdSZg$!,$]T$/+$M- N2$($]T$*-$ M* N"%$FVS$ .29& [email protected]=$[cbO\b$OZ]\S$ZSORa$b]$TOQb]`$W\RS^S\RS\b$U`]ebV$W\$4O&8*$ QSZZa$O\R#$b]USbVS`$eWbV$O$Q]\Qc``S\b$ .29& $;F6$!W\bS`\OZ$bO\RS[$Rc^ZWQObW]\"#$Q]\TS`a$ `SaWabO\QS$b]$bVS$8>F*$^`]bSW\$bg`]aW\S$YW\OaS$W\VWPWb]`a$B=5+()$O\R$35))'%$9S\S$ Sf^`SaaW]\$ O\OZgaWa$ ]T$ 3?>$ ^ObWS\ba$ eWbV$ +*.*"3=/$$ $`SO``O\US[S\b$ O\R$ .29& $ @-.-=$[cbObW]\$aV]eSR$O$b`S\R$b]eO`R$O$a^SQWTWQ$Sf^`SaaW]\$^`]TWZS%$Ac`a$Wa$bVS$TW`ab$ `S^]`b$ ]T$ `SQc``W\U$ .29& $ @-.-=$ [cbObW]\a$ W\$ Q]`S$PW\RW\U$ TOQb]`$ !548"$ ZScYS[WOa$ O\R$acUUSaba$O$\Se$acPU`]c^$]T$548$ZScYS[WOa%$ 8c`bVS`$ Sf^S`W[S\ba$ W\dSabWUObW\U$ []ZSQcZO`$ [SQVO\Wa[a$ T]`$ ZScYS[]US\SaWa$ W\RcQSR$ Pg$ .29& $ @-.-=$ [cbObW]\#$ abcRWSa$ T]`$ ^`]U\]abWQ$ aWU\WTWQO\QS$ O\R$ QZW\WQOZ$ SdOZcObW]\$]T$ .29& $W\VWPWb]`a$W\$ .29& [email protected]=$^]aWbWdS$3?>$O`S$`S_cW`SR%$ 8W`ab$SdWRS\QS$T]`$W\RcQbW]\$]T$OQcbS$ZScYS[WO$Pg$bVS$\]dSZ$ .29& [email protected]=$[cbObW]\$ GK' SGSL $VOa$`SQS\bZg$PSS\$^`]dWRSR$Pg$:cO\U$Sb$OZ%%M*-N$;\$bVSW`$`S^]`b#$bVS$ .29& [email protected]=$ [cbObW]\$aSS[a$b]$VOdS$`S[O`YOPZS$ZScYS[]US\WQ$^]bS\Qg$O\R$_cOZWbg$Q][^O`OPZS$ b]$ .29& $;F6$O\R$]bVS`$ .29& $F=6$[cbObW]\a%$ ! VT! ;RS\bWTWQObW]\$ ]T$ ORRWbW]\OZ$ US\SbWQ$ OP\]`[OZWbWSa$ W\$ ^ObWS\ba$ eWbV$ +*.*"3=/$$ $ `SO``O\US[S\b$Q]cZR$PS$VSZ^TcZ$W\$^`SRWQbW\U$]cbQ][S$M*+#$*.N#$aW\QS$-' $]T$548$ ZScYS[WO$^ObWS\ba$`SZO^aS$OTbS`$abO\RO`R$QVS[]bVS`O^g%M*/#$*0N$8c`bVS`[]`S#$bV]aS$ TW\RW\Ua$[Og$^`]dWRS$bVS$POaWa$T]`$\]dSZ$bVS`O^ScbWQ$bO`USba%$ $ FOYS\$ b]USbVS`#$ RSTW\W\U$ bVS$ [cbObW]\OZ$ ZO\RaQO^S$ ]T$ 3?>$ Wa$ SaaS\bWOZ$ b]$ W[^`]dS$ PW]Z]UWQOZ$ c\RS`abO\RW\U#$ ^`SRWQbW]\$ ]T$ RWaSOaS$ ]cbQ][S#$ O\R$ bVca$ ]^bW[WhObW]\$ ]T$ b`SOb[S\b#$O\R$ZOab$Pcb$\]b$ZSOab$T]`$bVS$RSdSZ]^[S\b$]T$\]dSZ$b`SOb[S\b$O^^`]OQVSa$ O\R$W\RWdWRcOZWhObW]\$]T$bVS`O^g%$ $ + + + + + + + + + + + + + + + ! VU! '+7MECEI=G=M>AEOAI+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + ! VV! '"$+:DA+?G=ODMEI!>EI@EIC+@JH=EI+JB+.,45+=I@+ODA+75!4<+@JH=EI+JB+,1$#+=MA+ NPBBE?EAIO+OJ+EI@P?A+=?POA+HSAGJE@+GAPFAHE=+EI+HE?A+ + Leukemia (2011), 1–10 & 2011 Macmillan Publishers Limited All rights reserved 0887-6924/11 www.nature.com/leu ORIGINAL ARTICLE The clathrin-binding domain of CALM and the OM-LZ domain of AF10 are sufficient to induce acute myeloid leukemia in mice AJ Deshpande 1,2,3,9 , A Rouhi 4,9 , Y Lin 5, C Stadler 1,2 , P Greif 1,2 , N Arseni 1,2 , S Opatz 1,2 , L Quintanilla-Fend 6, K Holzmann 7, W Hiddemann 1,2 , K Do¨hner8, H Do¨hner8, G Xu 5, SA Armstrong 3, SK Bohlander 1,2 and C Buske 4 1Department of Medicine III, Klinikum Grosshadern, Munich, Germany; 2Clinical Cooperative Group Leukemia, Helmholtz Center for Environmental Health, Munich, Germany; 3Division of Hematology/Oncology, Children’s Hospital, Harvard Medical School, Boston, MA, USA; 4Institute of Experimental Cancer Research, Comprehensive Cancer Center, University Hospital of Ulm, Ulm, Germany; 5State Key Laboratory of Molecular Biology, Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences, Shanghai, China; 6Eberhard-Karls-University of Tu¨bingen, Institute of Pathology, Tu¨bingen, Germany; 7Chip Facility/ZKF, University Hospital of Ulm, Ulm, Germany and 8Department of Internal Medicine III, University Hospital of Ulm, Ulm, Germany The t(10;11)(p13-14;q14-21) translocation, giving rise to the These proteins harbor a highly conserved plant homeodomain in CALM–AF10 fusion gene, is a recurrent chromosomal rearran- the N-terminal portion and an octapeptide motif (EQLLERQW) gement observed in patients with poor prognosis acute myeloid and leucine-zipper (OM-LZ) domain in the C-terminal region. 8,9 leukemia (AML). Although splicing of the CALM–AF10 fusion transcripts has been described in AML patients, the contribu- Leukemia-associated translocations consistently fuse the tion of different CALM and AF10 domains to in vivo leukemo- C-terminal OM-LZ domain either to the MLL gene (in the case genesis remains to be defined. We therefore performed detailed of MLL–AF10 and MLL–AF17 fusions) or to CALM (in the case of structure-function studies of the CALM–AF10 fusion protein. CALM–AF10 fusions).8,10,11 Consistent inclusion of the AF10 We demonstrate that fusion of the C-terminal 248 amino acids and AF17 OM-LZ domains in leukemic fusions highlights the of CALM, which include the clathrin-binding domain, to the potential importance of these highly conserved motifs in octapeptide motif–leucine-zipper (OM-LZ) domain of AF10 generated a fusion protein (termed CALM–AF10 minimal fusion leukemogenesis. Indeed, it has been demonstrated that the (MF)), with strikingly enhanced transformation capabilities in interactions of AF10 with critical components of the chromatin colony assays, providing an efficient system for the expedi- modifying machinery are mediated by the OM-LZ domain. 12–14 tious assessment of CALM–AF10-mediated transformation. Importantly, aberrant recruitment of the histone methyltransfer- Leukemias induced by the CALM–AF10 (MF) mutant recapitu- ase DOT1L by the OM-LZ domain of AF10 is thought to be lated multiple aspects of full-length CALM–AF10-induced critical for the leukemogenesis of MLL–AF10 and CALM–AF10 leukemia, including aberrant Hoxa cluster upregulation, a 13,15,16 characteristic molecular lesion of CALM–AF10 leukemias. In fusions. Since MLL–AF10 and CALM–AF10-positive summary, this study indicates that collaboration of the clathrin- leukemias are marked by global hypomethylation of histone binding and the OM-LZ domains of CALM–AF10 is sufficient to H3 lysine 79 (H3K79), as well as HOXA cluster-specific local induce AML. These findings further suggest that future H3K79 hypermethylation, these diseases could serve as approaches to antagonize CALM–AF10-induced transformation excellent models for studying the contribution of aberrant should incorporate strategies, which aim at blocking these key epigenetic marks to neoplastic development. 17 We have domains. Leukemia advance online publication, 17 June 2011; demonstrated that the expression of the human CALM/AF10 doi:10.1038/leu.2011.153 fusion gene in murine bone marrow (BM) stem and progenitor Keywords: CALM; AF10; acute myeloid leukemia cells results in an aggressive AML in vivo .18 Similar results have also been obtained by expressing this fusion gene under the control of the hematopoietically active Vav promoter in transgenic mice.19 In a separate study, continuous expression of this fusion gene was shown to be necessary for leukemia propagation, since suppression of CALM/AF10 transcripts by Introduction shRNA knockdown significantly increased the latency of leukemia induced by the t(10;11)(p13-14;q14-21) positive 16 The CALM–AF10 fusion gene results from in-frame fusions of the U937 human leukemia cell line. Data from patients with CALM gene on chromosome 11 to the AF10 gene on CALM–AF10-positive leukemia have demonstrated that despite chromosome 10. CALM–AF10 fusions are observed in acute splicing events, key functional domains of the fusion gene are 20 myeloid leukemia (AML), acute lymphoblastic leukemia and retained. In this study, we define the leukemogenic activity of malignant lymphoma1–6 and are especially prevalent in g–d key domains of CALM–AF10 by appropriate in vitro and in vivo lineage T-acute lymphoblastic leukemias.6 AML patients with assays and demonstrate that the leukemogenic activity of this translocation have a significantly poorer prognosis as CALM–AF10 is due to the function of two regions, the compared with patients without these translocations.7 AF10 clathrin-binding domain and the OM-LZ domain. belongs to a family of proteins that includes AF17 and BR140. 8 Correspondence: Professor C Buske, Institut fu¨r Experimentelle Materials and methods Tumorforschung des CCC Ulm, Universita¨tsklinikum Ulm und Klinik fu¨r Innere Medizin III, Albert-Einstein-Allee 11, 89081 Ulm, Germany. E-mail: [email protected] Generation of CALM–AF10 mutant constructs 9These authors contributed equally to this work. The pMIG-CALM-AF10 construct has been described pre- Received 21 April 2011; accepted 3 May 2011 viously.18 A PCR amplified fragment corresponding to amino ! VW! + Key functional domains of the CALM–AF10 fusion gene AJ Deshpande 2 acids 1–648 of CALM (NCBI accession NP_009097.2) with a samples. Mice were closely monitored for signs of disease stop codon was cloned into the MSCV-IRES-YFP (pMIY) plasmid manifestation and killed when moribund. BM, peripheral blood vector for generating the CALMD30 mutant. For the and spleen cells of killed leukemic or control mice were analyzed CALM þ OM-LZ mutant, a PCR amplified fragment correspond- for morphology and flow cytometric assessment of lineage ing to amino acids 1–648 of CALM was fused to a PCR amplified markers as described. 18 Gr-1, Sca I, Ter-119, CD4, Mac1, Kit, fragment encoding amino acids 677–758 of AF10 (Accession: B220 or CD8 antibodies were used for analysis (all antibodies NP_001182555), comprising the OM þ LZ with a stop codon. from BD Biosciences, San Jose, CA, USA). Stained cells were The CALM–AF10 DOM-LZ construct was generated by ligating analyzed on a FACSCalibur flow cytometer using the CellQuest- two separate PCR fragments of CALM–AF10 such that the Pro software (BD Pharmingen, San Diego, CA, USA). Histological OM-LZ domain was excluded from between the two fragments, analysis and immunohistochemistry was performed on fixed thereby excluding amino acids 677–758 of AF10. Mutants for organs of representative leukemic mice using standard protocols. interrogation of the CALM portion were generated by fusing an EcoR1–BamH1 fragment of either amino acids 1–410 or 400–648 of CALM to the AF10 amino acids 677–758 of AF10 in Colony forming cells the MSCV-IRES-GFP vector. For in vitro CFC assays, transduced cells were sorted for GFP and directly plated in 1% myeloid-conditioned methylcellulose containing Iscove’s modified Dulbecco medium-based Mice and retroviral infection of BM cells Methocult (Methocult M3434; StemCell Technologies, Vancouver, Parental strain mice were bred and maintained at the Helmholtz Canada) at a concentration of 1000 cells/ml. Single cell Centre Munich animal facility. Donors of primary BM cells were suspensions of colonies were serially replated at the same concentration for upto 6 weeks or until the exhaustion of cell 8- to 12-week-old (C57BL/6Ly-Pep3b  C3H/HeJ) F 1 (PepC3) mice, and recipients were 8- to 12-week-old (C57BL/6J  C3H/ growth. HeJ) F 1 (B6C3) mice. Donors were treated with 5-fluorouracil and 5 days later, BM from these mice was harvested and plated in BM Immunostaining and confocal laser scanning medium (Dulbecco’s modied Eagle’s medium, 15% fetal bovine fluorescence microscopy serum, 1% Pen/Strep) þ cytokines (100 ng/ml stem cell factor, For intracellular localization studies, the human osteosarcoma 10 ng/ml interleukin 6 (IL6), 6 ng/ml interleukin 3 (IL3)). After 48 h cell line U2-OS was grown on coverslips in six-well plates and of pre-stimulation, the BM cells were transduced with different transiently transfected with pcDNA3-FLAG-AF10, pcDNA3-FLAG- viruses by overlaying them on virus-producing irradiated (40 Gy) þ CALM–AF10 or pcDNA3-FLAG-CALM (400–648) þ AF10 GP E86 producers in the presence of cytokines and protamine (677–758) plasmids using 1 mg plasmid DNA and 1.5 mg poly- sulfate (5 mg/ml). The transduction was stopped by removing the ethyleneimine (Sigma, St Louis, MO, USA). After 24 h, cells were BM cells from the GP þ E86 cells and plating them in BM fixed with phosphate-buffered saline 2% paraformaldehyde for medium with cytokines for another 48 h to allow for green 10 min, permeabilized with phosphate-buffered saline 0.1% Triton fluorescent protein (GFP) expression. Retrovirally transduced cells X for 10 min and blocked with phosphate-buffered saline 10% fetal were sorted based on expression of GFP by using a FACSVantage calf serum for 1 h. Coverslips were incubated overnight with (Becton Dickinson, Franklin Lakes, NJ, USA). Sorted GFP or monoclonal mouse FLAG antibodies (Sigma). Following extensive yellow fluorescent protein (YFP)-positive cells were used for washing with phosphate-buffered saline, Alexa555-conjugated colony forming cell (CFC) or colony forming unit-spleen (CFU-S) secondary antibodies (Invitrogen, Carlsbad, CA, USA) were added assays or injected directly into recipient mice. 0 0 for 1 h. After further washing steps, cells were stained with 4 ,6 - diamidine-2 0-phenylindole dihydrochloride (Hoechst, Frankfurt, CFU-S assay Germany) and mounted using Cytomation medium (DAKO, Glostrup, Denmark). Finally, immunostained species were ana- The CFU-S assay was performed as previously described. 21 lyzed in a confocal fluorescence laser scanning system (TCS-SP2 Transduced and sorted 5-fluorouracil-treated BM cells were scanning system and DM IRB inverted microscope; Leica, Solms, injected intravenously into lethally irradiated (800 cGy of 137Cs Germany). g-radiation) (C57BL/6J  C3H/HeJ) F1 (B6C3) mice at cell numbers adjusted to give 5 to 15 macroscopic spleen colonies. 4 Cell doses ranged from 1.5 to 5  10 sorted cells. At 12 days Western blotting after injection, animals were killed and the number of Total protein was extracted from GFP þ E86 cells to assess the macroscopic colonies on the spleen was evaluated after fixation sizes and amounts of CALM–AF10 and CALM (400–648) þ AF10 in Telleyesniczky solution (absolute ethanol, glacial acetic acid (677–758) proteins via western blotting, as described before. 18 and formaldehyde mixed in a 9:1:1 ratio, respectively). For the The membranes were incubated overnight with goat polyclonal CALM (400–648) þ AF10 (677–758) mutant, mice were injected anti-CALM antibodies A-2 and C-18 (Santa Cruz Biotechnolo- with fewer cells to ensure scoring resolution (1000 GFP sorted gies Inc., Santa Cruz, CA, USA) used at a concentration of cells per mouse). 200 ng/ml. A horseradish peroxidase conjugated donkey anti- goat antibody (100 ng/ml) was used as secondary antibody. BM transplantation and assessment of mice In all, 8- to 10-week-old recipient mice (C57BL/6J  C3H/HeJ) Gene expression profiling and microarray analysis 137 F1 (B6C3) were irradiated (800 cGy) from a Cs g-radiation Mouse BM cells were transduced with CALM–AF10, CALM source. Fluorescence activated cell sorting-purified transduced (400–648) þ AF10 (677–758), CALM–AF10 DOM-LZ and empty BM cells, or a defined ratio of transduced and untransduced vector (MIG). Three days post-transduction, GFP-expressing cells was injected into the tail vein of irradiated recipient mice. cells were sorted by fluorescence activated cell sorting and total Hematopoietic engraftment of GFP-positive cells was assessed RNA (including both long and short RNA fractions) was by flow cytometry of regularly collected peripheral blood extracted using Qiagen miRNeasy Mini Kit (Hilden, Germany). Leukemia + ! VX! + Key functional domains of the CALM–AF10 fusion gene AJ Deshpande 3 For each construct, three biological replicates were produced. used this assay to test several deletion mutants to identify protein Microarray analysis was performed using Affymetrix GeneChip domains that contribute to the aforementioned hematopoietic Mouse Gene 1.0 ST and GeneChip miRNA arrays. For the activity of the CALM–AF10 fusion. GeneChip Mouse Gene 1.0 ST Array, analyses were performed using 200 ng total RNA as starting material and 5.5 mg ssDNA The OM-LZ of AF10 is necessary and sufficient for the per hybridization (GeneChip Fluidics Station 450; Affymetrix, enhancement of CFU-S activity by CALM–AF10 Santa Clara, CA, USA). The total RNAs were amplified and We initially focused on the AF10 part of the fusion and labeled following the whole transcript sense target labeling constructed several deletion mutants of the AF10 region. assay (http://www.affymetrix.com). Labeled ssDNA was hybri- Expression of the CALM gene truncated at the breakpoint of dized to Mouse Gene 1.0 ST Affymetrix GeneChip arrays CALM–AF10 (designated CALMD30) or the CALM–AF10 fusion (Affymetrix). The chips were scanned with an Affymetrix gene with a deleted OM–LZ domain (designated CALM–AF10 GeneChip Scanner 3000 and subsequent images analyzed using DOM-LZ) gave an average of three and four d-12 CFU-S Affymetrix Expression Console Software (Affymetrix). For the colonies, respectively, per input 10 5 BM cells. These numbers GeneChip miRNA Array, 1 mg of total RNA was labeled using were similar to d-12 CFU-S colonies obtained from empty vector the FlashTag Biotin HSR labeling Kit (Genisphere, Hatfield, PA, (MIG) transduced cells, indicating that the AF10 portion of USA). In all, 21.5 ml biotin-labeled sample were hybridized to CALM–AF10, especially the OM-LZ domain (amino acids 1 GeneChip miRNA Arrays at 48 C and 60 r.p.m. for 16 h. After 677–758 of AF10) is necessary for the CFU-S enhancement washing Wash at a Fluidics Station 450 using fluidics script phenotype. BM cells transduced with a construct harboring the FS450_0003 arrays were scanned with an Affymetrix GeneChip CALM gene fused only to the OM-LZ region of AF10 Scanner 3000. Raw data were background corrected and (CALM þ OM-LZ) showed an average of 54 colonies per 10 5 quantile normalized using the miRNAQCTool from Affymetrix input cells (an 18-fold increase compared with MIG; with default parameters recommended by Affymetrix. P ¼ 0.0005). This effect was comparable to the activity of the A transcriptome analyses was performed using BRB-ArrayTools full-length CALM–AF10 fusion (CALM–AF10 vs CALM þ OM-LZ; developed by Dr Richard Simon and BRB-ArrayTools Develop- P ¼ 0.065) (Figure 2a). ment Team (http://linus.nci.nih.gov/BRB-ArrayTools.html). Raw feature data were normalized and log 2 intensity expression summary values for each probe set were calculated using robust Fusion of the clathrin-binding domain of CALM to the multiarray average 22 using Affymetrix Expression Console Soft- OM-LZ domain of AF10 enhances CALM–AF10- ware (Affymetrix). For class comparison, we identified genes/ mediated transformation miRNAs that were differentially expressed among the two classes Having established the OM-LZ domain of AF10 as the important using a two-sample t-test. Genes were considered statistically determinant of the enhancement in d-12 CFU-S, we turned our significant if their P-value was o0.05 and displayed a fold change attention to the CALM part of the fusion. The N-terminal amino between the two groups of at least 1.5-fold. acids of CALM have a high homology to AP180, the neuronal Our complete mRNA and miRNA array data are accessible at homolog of the CALM protein. These residues (amino acids a gene expression omnibus (http://www.ncbi.nlm.nih.gov/geo/ 1–413), which include the ENTH or ANTH (Epsin or AP180 query/acc.cgi?token ¼ ltmtnyguieagozi&acc ¼ GSE27514). N-terminal homology) domain, have been shown to be insuffi- cient for the targeting of CALM to clathrin-coated pits. The binding of clathrin to CALM was shown to primarily involve the C- Results terminal residues 414–652 of CALM. We therefore interrogated these N- and C-terminal portions of CALM for their contribution to CALM–AF10 significantly increases the CFU-S the oncogenicity of the CALM –AF10 fusion gene. We generated frequency of normal BM progenitors two CALM deletion mutants fused to the AF10 OM-LZ domain. Sequencing of CALM–AF10 transcripts has revealed splicing The CALM (aa 1–410) þ OM-LZ fusion generated an average of 37 events, but has shown that the C-terminal part of CALM with its d-12 CFU-S colonies, which was comparable to the CALM þ OM- clathrin-binding domain and the OM-LZ domain are present in LZ construct. Strikingly, expression of the CALM (aa 400– patients with AML.12,20 We confirmed these findings in three 648) þ AF10 (aa 677–758) mutant, henceforth referred to as the patients with CALM–AF10-positive AML, demonstrating that the CALM–AF10 minimal fusion (MF) or CALM–AF10 (MF) mutant, C-terminal clathrin-binding domain of CALM and the OM-LZ profoundly augmented the ability of BM cells to form d-12 CFU-S domain of the AF10 part in the CALM–AF10 fusion gene are not colonies, with an average of 800 colonies per 10 5 input cells ( P- spliced out (data not shown). Based on this finding, we gauged value of 0.0033 compared with MIG and o0.0001 compared the extent of the contribution of these two domains to the with CALM–AF10) (Figure 2b). leukemogenic activity of the CALM–AF10 fusion gene. The full- The dramatic increase in colony numbers by the CALM–AF10 length CALM–AF10 fusion, which was derived from the (MF) mutant suggested that this mutant might confer a significant t(10;11)(p13-14;q14-21) translocation of the U937 cell line, proliferative advantage to hematopoietic progenitors. We tested includes the amino acids 1 to 648 of CALM to amino acids the effect of CALM–AF10 (MF) expression in 5-fluorouracil-treated 81–1042 of AF10 (Figure 1a). In order to assess the impact of BM cells using methylcellulose-based CFC assays. Unlike the full- short-term CALM–AF10 expression on early hematopoietic length CALM–AF10 fusion, which fails to transform BM progeni- progenitors in vivo , we employed the day-12 CFUs in spleen tors, under defined conditions in vitro (in the presence of medium (d-12 CFU-S) assay 21 (Figure 1b). Upon injection of BM cells supplemented with 6 ng/ml IL3, 100 ng/ml stem cell factor, expressing the empty MSCV-IRES-GFP (MIG) vector, an average 10 ng/ml IL6), CALM–AF10 (MF) expression led to a rapid of 3 ±1.2 d-12 CFU-S colonies could be recovered per 10 5 input immortalization of hematopoietic precursors (Figure 3). Speci- cells, whereas sorted CALM–AF10 transduced BM cells generated fically, CALM–AF10 (MF) showed a significant increase in the an average of 81 ±5 d-12 CFU-S colonies (Figure 1b). CALM– number of secondary CFCs (32-fold vs MIG; 15.38-fold vs AF10 expression therefore increases the d-12 CFU-S activity by CALM–AF10). While CALM–AF10-expressing BM cells typically B27-fold as compared with empty vector ( Po0.0001). We then lost their replating potential in the second week of culture, Leukemia + ! VY! + Key functional domains of the CALM–AF10 fusion gene AJ Deshpande 4 1 652 1 1042 ENTH TAD NES PHD ePHD NLS AT Hook OM-LZ Q-rich CATS Clathrin Binding Binding 1 81648 1042 ENTH TAD NES ePHD NLS AT Hook OM-LZ Q-rich CATS Clathrin Binding Binding 12 days Transduced BM Injection Scoring spleen colonies 1000 p < 0.0001 100 10 1 0.1 GFP CALM-AF10 MIG CALM-AF10 Colonies From 15,000 Input Cells Day 12 CFU-S content per 105 cells Figure 1 Analysis of the CALM–AF10 fusion protein ( a) A cartoon depiction of the various domains in CALM, AF10 and the CALM–AF10 fusion. TAD, potential trans-activating domain; NES, nuclear export signal; PHD, plant homeodomain; ePHD, extended PHD domain; NLS, nuclear localization signal. The CATS and clathrin-binding domains in CALM are marked. ( b) Upper panel shows a brief schematic presentation of the d-12 CFU-S assay. Pictures of representative fixed spleen colonies are shown on the lower left panel, whereas the right panel shows d-12 CFU-S numbers normalized to input 10 5 cells plotted on a log 10 scale. CALM–AF10 (MF)-expressing BM colonies could be serially B220–, Mac1 þ /B220 þ and Mac1–/B220 þ cells (Figure 4b), replated for at least 4 weeks in CFC assays (Figure 3a) and similar to the hierarchically arranged cell populations described proliferated extensively in culture for at least 8 weeks in medium for the full-length CALM–AF10 fusion gene.18 These findings supplemented only with 6 ng/ml IL3 (data not shown). Colonies demonstrate the striking similarity in leukemia presentation, obtained from CALM–AF10 (MF)-expressing progenitors were disease latency and transplantability between CALM–AF10 and typically compact and hypercellular and composed predomi- CALM–AF10 (MF)-induced leukemias, despite the lack of larger nantly of cells with a blast-like morphology (Figure 3b). portions of the fusion gene in the CALM–AF10 (MF) construct. Induction of AML by the CALM–AF10 (MF) mutant Enhanced expression and increased nuclear localization In order to ascertain that the transformation potential of the of the CALM–AF10 (MF) mutant CALM–AF10 (MF) mutant would translate into leukemia In order to assess the localization and relative expression levels generation, we injected CALM–AF10 (MF)-expressing 5-fluor- of CALM-AF10 and the CALM-AF10 (MF) mutant, we performed ouracil-enriched BM progenitors into lethally irradiated mice. immunofluorescence microscopy using FLAG-tagged, full- CALM–AF10 (MF) expression induced leukemia in the recipients length CALM–AF10 or the CALM–AF10 (MF) mutant constructs (n ¼ 9), with a median survival of 110 days (mean 116 days). The transfected in U2-OS cells and used wild-type AF10 as a control. latencies of disease in these leukemias were not significantly As it has been demonstrated previously, 23 AF10 was predomi- different (P ¼ 0.86) from leukemias initiated by the full-length nantly nuclear, whereas CALM–AF10 could be detected in a CALM–AF10 fusion (median survival 138 days, n ¼ 8). More- punctuate pattern in the cytoplasm. Moreover, in contrast to the over, similar to the CALM–AF10 mice described previously, 18 speckled appearance of CALM–AF10, the CALM–AF10 (MF) CALM–AF10 (MF)-induced leukemias showed diffuse infiltration protein was rather homogeneously distributed throughout the of large cells with blastic chromatin and prominent nucleolus in cytoplasm, with some cells showing an increased level of the spleen, liver, kidney, intestine, lung and lymph nodes. CALM–AF10 (MF) protein expression in the nucleus (Figure 5a). Myeloperoxidase staining showed massive infiltration of the Protein expression of the CALM–AF10 (MF) was also confirmed spleen and other organs with myeloblastic cells (Figure 4a). The by western blotting (Figure 5b). disease induced by the CALM–AF10 (MF) mutant could also be transplanted into secondary recipients who died with a median latency of 31 days post-transplantation ( n ¼ 8). Analysis of BM Analysis of global gene expression changes induced by cells from leukemic CALM–AF10 (MF) mice demonstrated the the CALM–AF10 mutants presence of Gr-1/Mac1 þ myeloid blasts. Importantly, all the Our results demonstrated that the CALM–AF10 (MF) mutant mice analyzed (n ¼ 5) harbored distinct populations of Mac1 þ / can profoundly enhance the proliferative capability of BM Leukemia + ! VZ! + Key functional domains of the CALM–AF10 fusion gene AJ Deshpande 5 Day 12 CFU-S content per 105 cells 0.1 1 10 100 1000 MIG 1 648 80 1042 ENTH TAD NES ePHD NLS AT Hook OM-LZ Q-rich CALM-AF10 CATS Clathrin Binding Binding p = 0.0005 ENTH TAD NES CALM∆ 3` p = 0.065 CATS Clathrin Binding Binding ENTH TAD NES ePHD NLS AT Hook Q-rich CALM-AF10 ∆OM-LZ CATS Clathrin Binding Binding OM-LZ ENTH TAD NES OM-LZ 677 758 CALM+OM-LZ CATS Clathrin Binding Binding Day 12 CFU-S content per 105 cells 0.1 1 10 100 1000 MIG 1 648 ENTH TAD NES OM-LZ p = 0.0033 CALM+OM-LZ CATS Clathrin Binding Binding p = 0.0003 1 410 ENTH OM-LZ CALM(1-400) + OM-LZ p = 0.0005 CATS Binding 400 648 NESTAD OM-LZ CALM (410-648) + OM-LZ or Clathrin Binding CALM-AF10 (MF) Figure 2 Structure-function analysis of AF10 and CALM. Values for d-12 CFU-S colonies per 10 5 cells are plotted on a log scale for the various mutants of AF10 ( a) and CALM ( b). The mean of all values is represented by a vertical line with error bars representing the s.e.m. Schematic representations of the mutants employed in the assays are depicted on the left side of each plotted value. Each experiment includes at least four independent replicates. progenitors in short-term assays. We therefore sought to analyze cells (Figure 6). Supplementary Tables S3 and S4 list the differences in global gene expression between CALM–AF10 the differentially expressed mRNAs and miRNAs between and CALM–AF10 (MF) transduced BM cells. We performed CALM–AF10 and CALM–AF10 DOM-LZ, respectively. microarray analysis of BM cells transduced with empty MIG vector, CALM–AF10, CALM–AF10 (MF) or CALM–AF10( DOM- LZ) using Affymetrix GeneChip Mouse Gene 1.0 ST and Discussion GeneChip miRNA arrays. Transcripts from 49 mRNAs and 13 miRNA were differentially expressed between the CALM–AF10 Breakpoint heterogeneity in chromosomal translocations can be and the CALM–AF10 (MF) transduced BM cells (Supplementary instructive for the identification of domains that are crucial for Tables S1 and S2, respectively). Most notably, a number of the transformation potential of oncogenes generated by those genes known to be highly expressed in CALM–AF10 leukemias, fusions. The C-terminal portion of AF10 is retained in all especially genes of the Hoxa cluster, were significantly reported CALM–AF10 and MLL–AF10 fusions.1,20,25 This upregulated in the CALM–AF10 (MF)-expressing cells as C-terminal region includes the OM-LZ domain and a glutamine- compared with BM cells expressing the full-length CALM– rich (Q-rich) domain that are both highly conserved. The OM-LZ AF10. Figure 6 depicts the differential expression of the Hoxa domain of AF10 has been shown to interact with a number of cluster genes (coding and non-coding) among CALM–AF10, important proteins such as the Swi/Snf interacting protein CALM–AF10 (MF) or CALM–AF10 DOM-LZ and empty MIG Gas41,14 the lineage determining transcription factor Ikaros23 vector. Interestingly, the Hox cluster embedded microRNA, and the histone methyltransferase Dot1l.13 The OM-LZ domain miR-196b, which has been shown to be upregulated in the was shown to be critical for the in vitro transformation capability human CALM–AF10 T-cell acute lymphoblastic leukemia,24 of the MLL–AF10 fusion gene.15 Okada et al .16 demonstrated was also highly upregulated in CALM–AF10 (MF)-expressing that the OM-LZ region of AF10 is important for the interaction of Leukemia + ! V[! + Key functional domains of the CALM–AF10 fusion gene AJ Deshpande 6 a 150 MIG CALM-AF10 CALM-AF10 (MF) G M 100 GM B 50 Colonies per 1000 cells 0 1 2 3 4 1 2 3 4 1 2 3 4 Week Week Week b MIG CALM-AF10 CALM-AF10 (MF) Colony Morphology (40 X) Cell Morphology Figure 3 Colony forming potential of the CALM–AF10 (MF) mutant. ( a) Bar graphs (upper panel) indicate the number of colonies per 1000 sorted cells plotted on a log 10 scale. Error bars indicate s.e.m. in each case. B, Blast; G, granulocyte; GM, granulocyte–macrophage; M, macrophage. (b) Representative pictures of 1-week colonies or Wright–Giemsa-stained cytospin preparations from pooled colonies are shown in the lower panel. CALM–AF10 with Dot1l and that the deletion of this region nucleus as compared with the full-length CALM-AF10 protein. reduces the proliferative capacity of CALM–AF10 transduced We and others have observed that CALM–AF10 disrupts the BM cells. We now demonstrate that the OM-LZ domain of AF10 physiological activity of the AF10–DOT1L complex, resulting in is both necessary for the expansion of early hematopoietic global H3K79 hypomethylation and HoxA-locus-specific hy- progenitors and also sufficient for in vivo leukemic transforma- permethylation although the exact molecular mechanisms of tion by the CALM–AF10 fusion gene. these epigenetic abnormalities remain obscure. Disruption of Under conditions routinely used to assay the in vitro the H3K79 methyltransferase activity of the AF10–DOT1L myeloid transformation activity of oncogenes, the CALM-AF10 complex is believed, inter alia, to lead to the aberrant fusion gene failed to transform BM cells in vitro . This is in overexpression of the HOXA cluster genes typically observed contrast to its strong leukemogenic effect in vivo .18,19 However, in the CALM–AF10 leukemias. These Hoxa cluster genes expression of the CALM–AF10 (MF) mutant rapidly immorta- (including the microRNA miR-196b) were significantly upregu- lized BM progenitors in vitro and dramatically increased the lated after short-term expression of the CALM–AF10 (MF) recovery or d-12 CFU-S progenitors. It could be that transforma- protein, but to a much lesser extent in the full-length CALM– tion by the full-length CALM-AF10 protein is dependent on AF10 transduced BM. These results suggest that the CALM–AF10 collaboration with in vivo niche signaling such as certain growth (MF) protein might more rapidly and efficiently disrupt the factors or cellular interactions absent under the aforementioned AF10–DOT1L complex, leading to the observed in vitro myeloid in vitro culture conditions, whereas transformation transformation phenotype. driven by CALM-AF10 (MF) is independent of these factors. In Lastly, it is possible that the CALM–AF10 (MF) mutant is this regard, it is pertinent to observe that CALM–AF10 more efficient at in vitro transformation because the N-terminus transduced BM cells can proliferate extensively in media (aa 1–410) of the CALM protein, which is not present in the MF, supplemented with fetal thymic organ culture and IL3,16 but normally impairs the in vitro transformation capability of the not under more defined cytokine supplemented conditions full-length CALM–AF10 fusion. Interestingly, the expression (unpublished observations), supporting the hypothesis that of the MF mutant did not alter the latency or phenotype of the additional, hitherto unknown growth factors are required for disease in the BM transplant setting, suggesting that the lack transformation by the full-length CALM–AF10 fusion. Alterna- of pronounced transformation in vitro by the full-length tively, it is possible that the in vitro transformation potential of CALM–AF10 protein is compensated by microenvironmental CALM-AF10 (MF) is caused by a greater abundance in the factors in vivo . Leukemia + ! V\! + Key functional domains of the CALM–AF10 fusion gene AJ Deshpande 7 100 MIG CALM-AF10 CALM-AF10 (MF) 1' 50 CALM-AF10 (MF) 2' Percent survival 0 0 100 200 300 Days 10 5 10 4 Mac1 10 3 10 2 0 0 10 2 10 3 10 4 10 5 B220 Figure 4 Leukemia initiation by the CALM–AF10 (MF) mutant. ( a) (Left panel) Survival of mice injected with BM cells transduced with MIG (control vector), CALM–AF10 or CALM–AF10 (MF) plotted on a Kaplan–Meier curve. The survival curve of secondary mice injected with CALM– AF10 (MF) primary leukemias is depicted in green. (Right panel) Wright–Giemsa-stained histological preparations of various organs from leukemic primary CALM–AF10 (MF) mice are shown with respective magnifications inserted at the bottom. ( b) Flow cytometric analysis of CALM–AF10 (MF) leukemic BM cells co-stained for Mac1 and B220. Although several prior studies on CALM–AF10 have focused OM-LZ domain of AF10 (aa 677–758) is necessary and sufficient on the role of AF10, the potential contribution of the different for in vivo leukemia initiation and phenocopies the full-length CALM domains remains to be elucidated. The CALM protein, CALM–AF10 fusion strongly underline the importance of which is involved in clathrin-mediated endocytosis, has several targeting the oncogenic activity of this domain in CALM–AF10 defined domains. The N-terminal region of CALM (aa 1–300) is leukemias. Aberrant recruitment of the H3K79 methyltransferase highly homologous to the N-terminal region of Epsin and DOT1L by the oncogenic AF10 fusions has been proposed as a AP180, important components of the endocytic machinery. 26 major mechanism of leukemogenesis of AF10-rearranged This domain, which has been termed the ENTH/ANTH domain, leukemias. This is based on the observation that the DOT1L binds inositol phospholipids and contributes to the formation of protein binds to the AF10 OM-LZ domain and that genes of the clathrin coats on cell membranes. 27,28 A large part of this HoxA gene cluster show H3K79 hypermethylation in CALM– domain is excluded from the MLL–CALM fusion observed in a AF10 and MLL–AF10 leukemia cells.13,16 Exciting new findings case of infant AML. 29 Our studies show that the exclusion of the have shown that AF10 is crucial for catalyzing Dot1l-mediated N-terminal amino acids of CALM (which include the E/ANTH H3K79 methylation, since RNAi-mediated suppression of AF10 domain) from the CALM–AF10 fusion enhances the clonogenic leads to a significant reduction in this chromatin modifica- capability of the fusion protein in short-term assays. Further- tion.17,31 The importance of this process is highlighted by the more, the presence of the C-terminal amino acids 414–648 of fact that Dot1l-mediated hypermethylation of H3K79 at the CALM is sufficient to induce leukemia that recapitulates several HoxA gene cluster could be a critical leukemogenic mechanism features of full-length CALM–AF10-induced disease. The in non-AF10 leukemias as well. 32 However, the changes in C-terminal amino acids (414–648) of CALM have been shown H3K79 methylation patterns caused by AF10 containing fusion to bind clathrin. 26 A separate study demonstrated that these genes are more complicated. We have previously reported a amino acids possess transcriptional activation potential when genome-wide hypomethylation of H3K79 in MLL–AF10 and fused to a heterologous DNA-binding motif in yeast. 30 Interest- CALM–AF10-expressing leukemias. This genome-wide hypo- ingly, since the CALM–AF10 (MF) mutant showed enhanced in methylation at H3K79 could be linked to an increase in vitro transformation potential, several plausible contributions of chromosomal instability.17 the C-terminal amino acids of CALM could be hypothesized. Eight differentially expressed miRNAs were found between Either C-terminal CALM-mediated disruption of endocytosis the full-length CALM–AF10 and CALM–AF10(DOM-LZ) trans- may interfere with cytokine receptor internalization, or alter- duced BM cells. Of note is the downregulation of the natively C-terminal CALM domains may transactivate genes at pro-apoptotic and anti-proliferative miR-128 in CALM–AF10 loci targeted by the Af10–Dotll complex. Our results that the compared with CALM–AF10(DOM-LZ).33,34 There are 28 Leukemia + ! V]! + Key functional domains of the CALM–AF10 fusion gene AJ Deshpande 8 AF10AF10 CALM-AF10 CALM-AF10 (MF) DAPI Flag Merge CALM-AF10 (173 kD) Endogenous CALM (62-72 kD) CALM-AF10 (MF) (37 kD) Beta-Actin Figure 5 CALM–AF10 (MF) expression and subcellular localization. (a) Confocal laser scans of U2-OS cells transfected with FLAG-tagged AF10, CALM–AF10 or CALM (400–648) þ AF10 (677–758) probed with anti-Flag monoclonal antibodies, and visualized by Alexa555-conjugated anti- mouse secondary antibodies. Left panels show 40,6 0-diamidine-20-phenylindole dihydrochloride (DAPI) stained cell nuclei, middle panels show Alexa555-signals and overlay images are shown in the left panels. ( b) Western blots showing CALM–AF10 and CALM (400–648) þ AF10 (677–758) expression using anti-CALM antibodies A-2 and C-18 (Santa Cruz Biotechnologies Inc.). Bands for endogenous CALM and b-actin were used as loading controls. mRNAs significantly differentially expressed between CALM– CALM–AF10 and CALM–AF10 (MF) transduced BM cells AF10 and CALM–AF10(DOM-LZ) samples. Of these 28 differ- showed that interestingly, in contrast to the full-length CALM– entially expressed mRNAs, none corresponded to the Hoxa AF10 protein, short-term expression of the CALM–AF10 (MF) cluster, indicating that this genomic region is not overtly mutant significantly upregulated genes of the Hoxa cluster activated in the 3 days post-transduction of the full-length (including miR-196b). Taken together, these results indicate that CALM–AF10 BM, in vitro . However, the Hedgehog pathway the CALM–AF10 (MF) mutant is more efficient in activating gene Smo, a homolog of the Drosophila smoothened gene, 35 some of the CALM–AF10-specific leukemogenic programs was upregulated upon full-length CALM–AF10 expression but lacks the full capacity to induce a more aggressive leukemia compared with CALM–AF10(DOM-LZ), suggesting that it may in vivo compared with the full-length CALM–AF10. have a role in CALM–AF10-mediated transformation. The Since most leukemic fusion products are poor targets for Hedgehog pathway has been shown to be activated in chronic pharmacologic inhibition, biochemical changes brought about myeloid leukemia and the loss of the Smo gene was shown to by these fusions offer promising targets for rational drug design. impair chronic myeloid leukemic stem cells. However, it Even as murine models of CALM–AF10 leukemia recapitulate remains to be seen if this pathway is also activated in CALM– aspects of their corresponding human malignancies, mechan- AF10 leukemias. Comparison between the full-length istic or therapeutic studies on these leukemias could be Leukemia + ! WT! + Key functional domains of the CALM–AF10 fusion gene AJ Deshpande 9 Network (NGFNplus) (PKL-01GS0876-6) to SKB. AR was sup- -1.0 1.1 1.0 ported by a fellowship from the Canadian Institutes of Health Research (CIHR) and a grant from the Baustein Startfo¨rderung at University Hospital Ulm. PAG was supported by a stipend from the Deutsche Jose-Carreras-Leuka¨mie-Stiftung (F05/06). YL was supported by grants from the Ministry of Science and Technology of China (2005CB522400, 2006CB943900 and KSCX2-YWN- 019), the Max-Planck-Gesellschaft of Germany and the National Science Foundation of China to G-LX. OMLZ_1 OMLZ_2 OMLZ_3 D D D References 1 Bohlander SK, Muschinsky V, Schrader K, Siebert R, Schlegelberger B, Harder L et al. Molecular analysis of the CALM/AF10 fusion: MSCV-IRES-GFP_1 MSCV-IRES-GFP_2 MSCV-IRES-GFP_3 CALM-AF10 CALM-AF10 CALM-AF10 CALM-AF10_1 CALM-AF10_2 CALM-AF10_3 CALM(400-648) + -AF10 (677-758)_1 CALM(400-648) + -AF10 (677-758)_2 CALM(400-648) + -AF10 (677-758)_3 identical rearrangements in acute myeloid leukemia, acute Hoxa1 lymphoblastic leukemia and malignant lymphoma patients. Hoxa2 Leukemia 2000; 14 : 93–99. Hoxa3 Hoxa3 2 Morerio C, Rapella A, Tassano E, Rosanda C, Panarello C. MLL- Hoxa4 MLLT10 fusion gene in pediatric acute megakaryoblastic leukemia. Hoxa5 Leuk Res 2005; 29 : 1223–1226. Hoxa6 3 Morerio C, Rapella A, Rosanda C, Lanino E, Lo Nigro L, Hoxa7 Hoxa9 Di Cataldo A et al. MLL-MLLT10 fusion in acute monoblastic Hoxa10 leukemia: variant complex rearrangements and 11q proximal Hoxa11 breakpoint heterogeneity. Cancer Genet Cytogenet 2004; 152: Hoxa13 108–112. mmu-miR-196b 4 Kumon K, Kobayashi H, Maseki N, Sakashita A, Sakurai M, Figure 6 Analysis of differential regulation of the Hoxa cluster genes. Tanizawa A et al. Mixed-lineage leukemia with t(10;11)(p13;q21): Heatmap of the differentially expressed mRNA and microRNA an analysis of AF10-CALM and CALM-AF10 fusion mRNAs and transcripts associated with CALM–AF10 leukemias was generated clinical features. Genes Chromosomes Cancer 1999; 25 : 33–39. using the Affymetrix platform. Relative expression levels of Hox genes 5 Kobayashi H, Hosoda F, Maseki N, Sakurai M, Imashuku S, and the Hoxa cluster embedded microRNA miR-196b are shown for Ohki M et al. Hematologic malignancies with the t(10;11) full-length CALM–AF10, CALM–AF10( DOM-LZ) and CALM-AF10 (p13;q21) have the same molecular event and a variety of (MF) in comparison to the empty MIG vector (three independent morphologic or immunologic phenotypes. Genes Chromosomes biological replicates each). Cancer 1997; 20 : 253–259. 6 Asnafi V, Radford-Weiss I, Dastugue N, Bayle C, Leboeuf D, Charrin C et al. CALM-AF10 is a common fusion transcript in T-ALL and is specific to the TCRgammadelta lineage. Blood 2003; hampered by the prolonged latency of leukemia initiation in the 102: 1000–1006. CALM–AF10 retroviral BM transplantation model 18 and incom- 7 Dreyling MH, Schrader K, Fonatsch C, Schlegelberger B, Haase D, Schoch C et al. MLL and CALM are fused to AF10 in plete disease penetrance in the CALM–AF10 transgenic morphologically distinct subsets of acute leukemia with transloca- 19 model. In this study, we have demonstrated that the d-12 tion t(10;11): both rearrangements are associated with a poor CFU-S assay can be used as a relatively rapid in vivo assay for prognosis. Blood 1998; 91 : 4662–4667. CALM–AF10 activity. Using this method, we screened for 8 Chaplin T, Ayton P, Bernard OA, Saha V, Della Valle V, Hillion J domains essential for CALM–AF10 function and identified a et al. A novel class of zinc finger/leucine zipper genes identified potent leukemogenic mutant of CALM–AF10, which can from the molecular cloning of the t(10;11) translocation in acute leukemia. Blood 1995; 85 : 1435–1441. immortalize BM progenitors in vitro , and phenocopy leukemia 9 Saha V, Chaplin T, Gregorini A, Ayton P, Young BD. The generated by the full-length CALM–AF10 fusion in vivo . The leukemia-associated-protein (LAP) domain, a cysteine-rich motif, CALM–AF10 (MF) mutant, therefore, lends itself very well to is present in a wide range of proteins, including MLL, AF10, and expeditious assessment of transformation in the CALM–AF10 MLLT6 proteins. Proc Natl Acad Sci USA 1995; 92 : 9737–9741. leukemias from mechanistic as well as therapeutic perspectives. 10 Prasad R, Leshkowitz D, Gu Y, Alder H, Nakamura T, Saito H et al. Leucine-zipper dimerization motif encoded by the AF17 gene fused to ALL-1 (MLL) in acute leukemia. Proc Natl Acad Sci USA 1994; 91 : 8107–8111. Conflict of interest 11 Dreyling MH, Martinez-Climent JA, Zheng M, Mao J, Rowley JD, Bohlander SK. The t(10;11)(p13;q14) in the U937 cell line results The authors declare no conflict of interest. in the fusion of the AF10 gene and CALM, encoding a new member of the AP-3 clathrin assembly protein family. Proc Natl Acad Sci USA 1996; 93 : 4804–4809. 12 Silliman CC, McGavran L, Wei Q, Miller LA, Li S, Hunger SP. Acknowledgements Alternative splicing in wild-type AF10 and CALM cDNAs and in AF10-CALM and CALM-AF10 fusion cDNAs produced by the We thank Bianka Ksienzyk and Lisa Kaiser for their excellent t(10;11)(p13-14;q14-q21) suggests a potential role for truncated technical assistance and the members of the animal facility at the AF10 polypeptides. Leukemia 1998; 12 : 1404–1410. Helmholtz Centre for their tremendous support with mouse 13 Okada Y, Feng Q, Lin Y, Jiang Q, Li Y, Coffield VM et al. hDOT1L experiments. AJD and the work were supported by a grant of links histone methylation to leukemogenesis. Cell 2005; 121: 167–178. the Deutsche Krebshilfe (10-1838-Bu 2 to CB). This work was also 14 Debernardi S, Bassini A, Jones LK, Chaplin T, Linder B, supported by a grant from the Deutsche Forschungsgemeinschaft de Bruijn DR et al. The MLL fusion partner AF10 binds GAS41, DFG (SFB 684, A6) and a grant from the German Ministry of a protein that interacts with the human SWI/SNF complex. Blood Education and Research (BMBF) National Genome Research 2002; 99 : 275–281. Leukemia + ! WU! + Key functional domains of the CALM–AF10 fusion gene AJ Deshpande 10 15 DiMartino JF, Ayton PM, Chen EH, Naftzger CC, Young BD, blastic leukemia and acute myeloid leukemia. Leukemia 2000; 14 : Cleary ML. The AF10 leucine zipper is required for leukemic 100–104. transformation of myeloid progenitors by MLL-AF10. Blood 2002; 26 Tebar F, Bohlander SK, Sorkin A. Clathrin assembly lymphoid 99 : 3780–3785. myeloid leukemia (CALM) protein: localization in endocytic- 16 Okada Y, Jiang Q, Lemieux M, Jeannotte L, Su L, Zhang Y. coated pits, interactions with clathrin, and the impact of over- Leukaemic transformation by CALM-AF10 involves upregulation expression on clathrin-mediated traffic. Mol Biol Cell 1999; 10 : of Hoxa5 by hDOT1L. Nat Cell Biol 2006; 8: 1017–1024. 2687–2702. 17 Lin YH, Kakadia PM, Chen Y, Li YQ, Deshpande AJ, Buske C et al. 27 Itoh T, Koshiba S, Kigawa T, Kikuchi A, Yokoyama S, Global reduction of the epigenetic H3K79 methylation mark and Takenawa T. Role of the ENTH domain in phosphatidylinositol- increased chromosomal instability in CALM-AF10-positive leuke- 4,5-bisphosphate binding and endocytosis. Science 2001; 291: mias. Blood 2009; 114: 651–658. 1047–1051. 18 Deshpande AJ, Cusan M, Rawat VP, Reuter H, Krause A, Pott C 28 Legendre-Guillemin V, Wasiak S, Hussain NK, Angers A, et al. Acute myeloid leukemia is propagated by a leukemic stem McPherson PS. ENTH/ANTH proteins and clathrin-mediated cell with lymphoid characteristics in a mouse model of CALM/ membrane budding. J Cell Sci 2004; 117(Part 1): 9–18. AF10-positive leukemia. Cancer Cell 2006; 10 : 363–374. 29 Wechsler DS, Engstrom LD, Alexander BM, Motto DG, 19 Caudell D, Zhang Z, Chung YJ, Aplan PD. Expression of a CALM- Roulston D. A novel chromosomal inversion at 11q23 in infant AF10 fusion gene leads to Hoxa cluster overexpression and acute acute myeloid leukemia fuses MLL to CALM, a gene that encodes a leukemia in transgenic mice. Cancer Res 2007; 67 : 8022–8031. clathrin assembly protein. Genes Chromosomes Cancer 2003; 36 : 20 Narita M, Shimizu K, Hayashi Y, Taki T, Taniwaki M, Hosoda F 26–36. et al. Consistent detection of CALM-AF10 chimaeric transcripts in 30 Archangelo LF, Glasner J, Krause A, Bohlander SK. The novel haematological malignancies with t(10;11)(p13;q14) and identifi- CALM interactor CATS influences the subcellular localization of cation of novel transcripts. Br J Haematol 1999; 105: 928–937. the leukemogenic fusion protein CALM/AF10. Oncogene 2006; 21 Helgason CD, Antonchuk J, Bodner C, Humphries RK. Home- 25 : 4099–4109. ostasis and regeneration of the hematopoietic stem cell pool are 31 Mohan M, Herz HM, Takahashi YH, Lin C, Lai KC, Zhang Y et al. altered in SHIP-deficient mice. Blood 2003; 102: 3541–3547. Linking H3K79 trimethylation to Wnt signaling through a 22 Irizarry RA, Ooi SL, Wu Z, Boeke JD. Use of mixture models in a novel Dot1-containing complex (DotCom). Genes Dev Mar 15; microarray-based screening procedure for detecting differentially 24 : 574–589. represented yeast mutants. Stat Appl Genet Mol Biol 2003; 2: 32 Krivtsov AV, Feng Z, Lemieux ME, Faber J, Vempati S, Sinha AU Article1. et al. H3K79 methylation profiles define murine and human MLL- 23 Greif PA, Tizazu B, Krause A, Kremmer E, Bohlander SK. The AF4 leukemias. Cancer Cell 2008; 14 : 355–368. leukemogenic CALM/AF10 fusion protein alters the subcellular 33 Godlewski J, Nowicki MO, Bronisz A, Williams S, Otsuki A, localization of the lymphoid regulator Ikaros. Oncogene 2008; 27 : Nuovo G et al. Targeting of the Bmi-1 oncogene/stem cell renewal 2886–2896. factor by microRNA-128 inhibits glioma proliferation and self- 24 Schotte D, Lange-Turenhout EA, Stumpel DJ, Stam RW, Buijs- renewal. Cancer Res 2008; 68 : 9125–9130. Gladdines JG, Meijerink JP et al. Expression of miR-196b is not 34 Adlakha YK, Saini N. MicroRNA-128 downregulates Bax and exclusively MLL-driven but is especially linked to activation of induces apoptosis in human embryonic kidney cells. Cell Mol Life HOXA genes in pediatric acute lymphoblastic leukemia. Haema- Sci 2010; 68 : 1415–1428. tologica 2010; 95 : 1675–1682. 35 Zhao C, Chen A, Jamieson CH, Fereshteh M, Abrahamsson A, 25 Carlson KM, Vignon C, Bohlander S, Martinez-Climent JA, Blum J et al. Hedgehog signalling is essential for maintenance Le Beau MM, Rowley JD. Identification and molecular character- of cancer stem cells in myeloid leukaemia. Nature 2009; 458: ization of CALM/AF10fusion products in T cell acute lympho- 776–779. Supplementary Information accompanies the paper on the Leukemia website (http://www.nature.com/leu) Leukemia + ! WV! '"%+ 0RJHA+ NALPAI?EIC+ E@AIOEBEAN+ MA?PMMEIC+ 14:&+ 6)*)3+ HPO=OEJIN+ EI+ .JMA! -EI@EIC+1=?OJM+GAPFAHE=+ + + From bloodjournal.hematologylibrary.org at UBM Bibliothek Grosshadern on September 7, 2013. For personal use only. Regular Article MYELOID NEOPLASIA Exome sequencing identi fies recurring FLT3 N676K mutations in core-binding factor leukemia Sabrina Opatz,1-3 Harald Polzer,1,4 Tobias Herold,1,4 Nikola P. Konstandin, 1 Bianka Ksienzyk,1 Evelyn Zellmeier,1 Sebastian Vosberg,1,4 Alexander Graf,5 Stefan Krebs,5 Helmut Blum,5 Karl-Peter Hopfner,6 Purvi M. Kakadia, 1,7 Stephanie Schneider,1 Annika Dufour,1 Jan Braess,8 Maria Cristina Sauerland,9 Wolfgang E. Berdel, 10 Thomas B¨uchner, 10 Bernhard J. Woermann, 11 Wolfgang Hiddemann,1-4 Karsten Spiekermann,1-4 Stefan K. Bohlander, 4,7 and Philipp A. Greif 1-4 1Department of Internal Medicine 3, University Hospital Grosshadern, Ludwig-Maximilians-Universit¨at,Munich (LMU), Germany; 2German Cancer Consortium (DKTK), Heidelberg, Germany; 3German Cancer Research Centre (DKFZ), Heidelberg, Germany; 4Clinical Cooperative Group Leukemia, Helmholtz Zentrum M¨unchen, German Research Center for Environmental Health, Munich, Germany; 5Laboratory for Functional Genome Analysis, and 6Department of Biochemistry at the Gene Center, Munich, Germany; 7Center for Human Genetics, Philipps University Marburg, Marburg, Germany; 8Oncology and Hematology, St. John of God Hospital, Regensburg, Germany; 9Institute of Biostatistics and Clinical Research, and 10 Department of Medicine A, Hematology, Oncology and Pneumology, Universit¨at M¨unster, M¨unster, Germany; and 11 German Society of Hematology and Oncology, Berlin, Germany Key Points The t(8;21) and inv(16)/t(16;16) rearrangements affecting the core-binding factors RUNX1 and CBFB, respectively, are found in 15% to 20% of adult de novo acute myeloid leukemia • FLT3 N676K mutations (AML) cases and are associated with a favorable prognosis. Since the expression of the without concurrent internal fusion genes CBFB /MYH11 or RUNX1 /RUNX1T1 alone is not sufficient to cause leukemia, tandem duplication (ITD) are we performed exome sequencing of an AML sample with an inv(16) to identify mutations, associated with core-binding which may collaborate with the CBFB/MYH11 fusion during leukemogenesis. We factor leukemia. discovered an N676K mutation in the adenosine triphosphate (ATP)-binding domain FLT3 • N676K activates FLT3 and (tyrosine kinase domain 1 [TKD1]) of the fms-related tyrosine kinase 3 ( ) gene. In a cohort of 84 de novo AML patients with a CBFB/MYH11 rearrangement and in 36 patients downstream signaling with a RUNX1/RUNX1T1 rearrangement, the FLT3 N676K mutation was identified in pathways. 5 and 1 patients, respectively (5 [6%] of 84; 1 [3%] of 36). The FLT3-N676K mutant alone leads to factor-independent growth in Ba/F3 cells and, together with a concurrent FLT3- ITD (internal tandem duplication), confers resistance to the FLT3 protein tyrosine kinase inhibitors (PTKIs) PKC412 and AC220. Gene expression analysis of AML patients with CBFB/MYH11 rearrangement and FLT3 N676K mutation showed a trend toward a specific expression profile. Ours is the first report of recurring FLT3 N676 mutations in core-binding factor (CBF) leukemias and suggests a defined subgroup of CBF leukemias. This trial was registered at www.clinicaltrials.gov as #NCT00266136. ( Blood . 2013;122(10):1761-1769) Introduction The inversion inv(16)(p13;q22), the translocation t(16;16)(p13;q22), themselves is not suf ficient to cause leukemia, it is highly likely that and the translocation t(8;21)(q22;q22) are recurring rearrangements additional mutations are required for malignant transformation.1,3,4 in acute myeloid leukemia (AML), which result in the fusion genes Leukemogenesis is a multistep process. Mutations associated with CBFB/MYH11 or RUNX1/RUNX1T1 , respectively. These rearrange- myeloid malignancies have been found in genes involved in several ments are found in 15% to 20% of adult de novo AML cases and functional classes: signaling pathways (eg, FLT3 , KIT , RAS ), represent recognized World Health Organization entities that are transcription factors (eg, RUNX1/RUNX1T1 , CBFB/MYH11 ), epige- associated with a favorable prognosis. 1,2 netic regulators (eg, DNMT3A , IDH1 , IDH2 , TET2 ), tumor sup- CBFB and RUNX1 form the core-binding factor (CBF), a het- pressors (eg, TP53 , WT1 ), and splicing machinery (eg, SF3B1 , erodimeric transcription factor essential for normal hematopoiesis. SRSF2 ). 5 In CBF leukemia, mutations in genes coding for signaling The CBFB/MYH11 and RUNX1/RUNX1T1 fusion proteins disrupt proteins (so-called proliferation drivers) are commonly found to the physiologic activity of CBF, leading to the repression of CBF collaborate with the CBF fusion genes.6 Mutations in KIT, FLT3, target genes and resulting in a block of differentiation and impaired or NRAS/KRAS have frequently been detected in CBF leukemia. 7-9 hematopoiesis. Since knock-in mouse models have demonstrated Up to 90% of AML patients with a CBFB/MYH11 fusion have that the expression of CBFB/MYH11 and RUNX1/RUNX1T1 by either a mutation in a receptor tyrosine kinase (RTK) or in RAS. 10,11 Submitted January 18, 2013; accepted July 8, 2013. Prepublished online as The online version of this article contains a data supplement. Blood First Edition paper, July 22, 2013; DOI 10.1182/blood-2013-01-476473. The publication costs of this article were defrayed in part by page charge S.O. and H.P. contributed equally to this work payment. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 USC section 1734. Presented as an oral presentation at the 54th annual meeting of the American Society of Hematology, Atlanta, GA, December 8-11, 2012. © 2013 by The American Society of Hematology BLOOD, 5 SEPTEMBER 2013 x VOLUME 122, NUMBER 10 1761 ! WW! + + From bloodjournal.hematologylibrary.org at UBM Bibliothek Grosshadern on September 7, 2013. For personal use only. 1762 OPATZ et al BLOOD, 5 SEPTEMBER 2013 x VOLUME 122, NUMBER 10 In general, these signaling pathway mutations are mutually exclusive. 5 libraries were sequenced by performing 76-bp paired end reads on a However, about 10% of AML patients with a CBFB/MYH11 fusion Genome Analyzer IIx platform (Illumina, San Diego, CA). Sequence 15 do not carry any of the currently known mutations. alignment and variant detection was performed as described previously. To systematically identify additional collaborating mutations in CBFB /MYH11- positive AML patients, we performed exome sequenc- Sanger sequencing ing of an AML with an inv(16) without any additional known The nonsynonymous somatic variant in the FLT3 gene (detected in the genetic alterations. Using this approach, we identi fied an FLT3 AML but not in the remission sample) was veri fied by sequencing both N676K mutation. Screening a cohort of 120 CBF AML patients, DNA strands using ABI 3100-Avant technology (Applied Biosystems) after we discovered the FLT3 N676K mutation to be present in 6 of PCR ampli fication of FLT3 exon 16. PCR and sequence analysis of genomic these patients. Mutations affecting the ATP-binding pocket, in DNA was performed with forward primer 5 9-TGCAGATTGACTCTG particular position N676, resulting in variable amino acid changes AGCTG-39 and reverse primer 5 9-CACTGTGACTGAGAAAAGACAA AG-39, located in the 5 9 and 3 9 flanking introns, spanning the complete exon (N676D or N676S), were initially discovered in a screen for 16 and yielding a 327-bp PCR product corresponding to AA 649 to 685 of FLT3 resistance to tyrosine kinase inhibitors (TKIs) in internal the human FLT3 protein (National Center for Biotechnology Information 12,13 tandem duplication (ITD)-expressing Ba/F3 cells. To the best reference sequence NM_004119). The same assay was used on a total of of our knowledge, an FLT3 N676K point mutation has been reported 209 de novo AML patients ( 84 CBFB /MYH11- rearranged, 36 RUNX1/ just once before in a cytogenetically normal (CN) AML patient RUNX1T1 -rearranged, and 90 CN-AML patients). Routine diagnostic tests with an FLT3 -ITD mutation who was screened to determine the included mutation analysis at de fined positions of FLT3, KIT, KRAS, NRAS, cause of the acquired TKI resistance after PKC412 therapy. 14 In this NPM1, MLL and WT1 of all 84 AML M4eo samples (supplemental Table 3). study, we report recurring FLT3 N676K mutations at first diagnosis of CBF AML without concurrent FLT3 -ITD. Importantly, Ba/F3 cells DNA constructs and vectors expressing the FLT3 N676K mutation show factor-independent The human FLT3-wild-type (WT) and the FLT3-ITD-NPOS constructs con- growth and sensitivity toward commonly used TKIs, suggesting taining a 28 AA duplicated sequence (CSSDNEYFYVDFREYEYDLK that the presence of the FLT3 N676K mutation in CBF leukemia WEFPRENL) inserted between AA 611/612 of human FLT3-WT were patients might open up new treatment options including TKI kindly provided by Gary Gilliland (Harvard Medical School, Boston, therapy. MA). The FLT3 constructs were subcloned into the MSCV-IRES-EYFP retroviral expression vector (kindly provided by R. K. Humphries, Terry Fox Laboratory, University of British Columbia, Vancouver, BC, Canada). In vitro mutagenesis Materials and methods The N676K mutation was introduced into the FLT3-WT and the FLT3-ITD- Patient samples NPOS vectors by using the QuikChange II XL Site-Directed Mutagenesis Kit (Stratagene, La Jolla, CA) according to the manufacturer ’s instructions. A diagnostic bone marrow sample was collected from an 18-year-old patient The mutant FLT3 D835Y construct was generated by using the QuikChange diagnosed with AML M4eo according to standard French-American-British Site-Directed Mutagenesis Kit. 16 The correct sequence of all constructs was and World Health Organization criteria in December 2003. The inv(16) con firmed by sequencing. (p13;q22) was detected by standard cytogenetics analysis (karyotype: 46, XY,inv(16)(p13;q22)[10]). The CBFB/MYH11 fusion transcript was con- Cell lines, reagents, and antibodies firmed by reverse-transcriptase polymerase chain reaction (RT-PCR). No additional genetic alterations were detected at this time. The patient was Phoenix Eco cells were purchased from Orbigen (San Diego, CA) and enrolled in the AMLCG-1999 trial of the German AML Cooperative Group cultured in Dulbecco ’s modi fied Eagle medium supplemented with 10% (NCT00266136), and written informed consent was obtained in accordance fetal bovine serum and 0.5% penicillin/streptomycin. Low-passage murine with the Declaration of Helsinki. The samples were obtained under AMLCG Ba/F3 cells and WEHI-3B cells were obtained from Deutsche Sammlung von study protocols approved by the ethics committees of the participating centers. Mikroorganismen und Zellkulturen (DSMZ, Braunschweig, Germany) After induction chemotherapy and autologous peripheral blood stem cell and maintained in RPMI-1640 medium containing 10% fetal bovine serum, transplantation, complete remission was achieved ( ,5% bone marrow blasts; 0.5% penicillin/streptomycin, and 10% WEHI-3B conditioned medium as CBFB/MYH11 transcripts no longer detectable by RT-PCR). A bone a source of interleukin-3 (IL-3). Recombinant human FLT3 ligand and marrow sample at complete remission was used as normal control for recombinant murine IL-3 were obtained from Immunotools (Friesoythe, exome sequencing. Germany). FLT3 inhibitor PKC412 was obtained from Novartis (Basel, In total, bone marrow or peripheral blood samples from 84 adult Switzerland) and AC220 was obtained from SYNthesis Med Chem patients with newly diagnosed and untreated AML M4eo (CBFB/MYH11 (Cambridge, United Kingdom). fusion–positive, including the case analyzed by exome sequencing), from The following antibodies were used: anti-AKT (9272), anti-pAKT (4060), 36 patients with t(8;21) and from 90 patients with CN AML were used for anti-MAPK (9107), anti-pMAPK (9101), and anti-pSTAT5 (9351) (Cell targeted mutation screening. Signaling Technology, Danvers, MA); anti-FLT3 (sc-480), anti-pTyr (sc-7020), anti-STAT5 (sc-835), and anti-GAPDH (sc-32233) from Santa Cruz Bio- Sample preparation and high-throughput sequencing technology (Santa Cruz, CA); and CD135-PE (IM2234U) and IgG1-PE isotype control (A07796) from Immunotech (Marseille, France). Stable ’ Genomic DNA was extracted from patients bone marrow or peripheral transduction of Ba/F3 cells, western blot analysis, and detection of surface blood samples using QIAcube technology (Qiagen, Hilden, Germany). For markers were performed as described previously. 17,18 exome sequencing of the index patient, 3 mg of genomic DNA was fragmented to an average size of 150 bp by using the Bioruptor sonicator Proliferation and apoptotic cell death of Ba/F3 cells (Diagenode, Li ege,` Belgium). Paired-end sequencing libraries were prepared using DNA sample prep reagent set 1 (NEBNext). Library preparation Proliferation and apoptosis assays were carried out as described previously. 17 included end repair, adapter ligation, and PCR enrichment and was carried For long-term proliferation assays, cells were seeded at a density of 2 3 10 5/mL out as recommended by Illumina protocols. Exon-coding sequences were in growth medium containing 0.1% WEHI-conditioned medium as source then captured by using SureSelect human all exon 50Mb kit version 3 (Agilent, of murine IL-3 and as control in the presence of 10 ng/mL IL-3. After Santa Clara, CA) according to the manufacturer ’s instructions. Exome 72 hours, Ba/F3 cells were cleared from IL-3 by two centrifugation steps + + ! WX! + + From bloodjournal.hematologylibrary.org at UBM Bibliothek Grosshadern on September 7, 2013. For personal use only. BLOOD, 5 SEPTEMBER 2013 x VOLUME 122, NUMBER 10 FLT3 N676K MUTATIONS IN CBF AML 1763 Figure 1. FLT3 N676K mutations identified in CBFB /MYH11 -rearranged AML. (A) Exome data sets of a CBFB /MYH11 -positive AML sample (upper panels) and the corresponding follow-up sample from the same patient (lower panels) are displayed using the integrative genomics viewer. 24 Horizontal gray bars symbolize the 76-bp reads aligned to the reference sequence. The frequency of 24% of the mutant nucleotide T in the diagnostic leukemia sample indicates a heterozygous point mutation causing an amino acid substitution (NM_004119.2:c.2028C .A; p.N676K), whereas in the follow-up sample, only the wild-type nucleotide G is detected at this position. Read depth and base count are indicated for the affected positions, respectively. (B) Sanger sequencing confirmed the FLT3 N676K mutation found initially by exome sequencing. Chromatograms are shown for both the diagnostic AML sample and the corresponding follow-up sample at complete remission (CR) from the same patient. (C) The structure of the human FLT3 protein includes the transmembrane domain (TM), the juxtamembrane domain (JM), and TKD1 and TKD2. Amino acid positions targeted by known recurrent mutations in AML are indicated in green above the corresponding domains. N676 is indicated in blue below the TKD1 domain. (D) Frequency distribution of additional genetic aberrations in 84 CBFB /MYH11 -rearranged patients. Each column indicates one patient. Dark gray boxes indicate patients who are positive for the respective mutation; light gray boxes indicate wild-type status. Missing information is shown as a white space (N/A, not available). Gene names and types of mutations are indicated on the left. Mutation frequencies are indicated on the right. MLL -PTD, partial tandem duplications in the MLL gene. with phosphate-buffered saline and resuspended in medium without IL-3. the robust multichip average method as described by Irizarry et al. 22 The Control cells were cultivated in the presence of 10 ng/mL IL-3. Viable cells Linear Models for Microarray Data (Limma) package was used to compute were counted every day, and a cell density of 2.5 3 10 6 was not exceeded. differentially regulated probe sets by comparing patients with CBFB /MYH11 rearrangement and mutations affecting FLT3 D835, NRAS , KRAS , KIT, or FLT3-ITD to patients with CBFB/MYH11 rearrangement and FLT3 N676K. Gene expression profiling and microarray analyses Gene set enrichment analysis (GSEA) was performed with GSEA software Pretreatment bone marrow samples from 33 patients (data deposited in (Broad Institute of Massachusetts Institute of Technology and Harvard) GSE37642) were analyzed by using Affymetrix HG-U133 A/B oligonucleotide to assess signi ficant changes in gene expression levels. 23 The GSEA was run microarrays (Affymetrix, Santa Clara, CA) as described previously. 19,20 For with 1000 permutations and compared with the “c2_kegg ” collection from probes to probe set annotation, we used custom chip de finition files based the Molecular Signatures Database (MsigDB 3.0) consisting of 186 gene on GeneAnnot version 2.0, synchronized with GeneCards Version 3.04 sets. All statistical analyses were performed by using R 3.0.1 software and (http://www.xlab.unimo.it/GA_CDF/). 21 Normalization was carried out by routines from the biostatistics software repository Bioconductor. + + ! WY! + + From bloodjournal.hematologylibrary.org at UBM Bibliothek Grosshadern on September 7, 2013. For personal use only. 1764 OPATZ et al BLOOD, 5 SEPTEMBER 2013 x VOLUME 122, NUMBER 10 Results Exome sequencing of an AML M4eo patient To systematically identify mutations that may collaborate with CBFB/MYH11 during leukemogenesis, we performed exome sequencing of an AML sample with inv(16). The sample was selected on the basis of sample availability and the absence of additional genetic alterations ( FLT3 -ITD, MLL- PTD [partial tandem duplica- tion], FLT3 -TKD, NPM1 , NRAS , KRAS , KIT , and WT1 mutation negative). We sequenced the exome (protein coding regions) of the diagnostic sample and a remission sample from the same patient, generating at least 4 Gbp of sequence from each exome. This allowed us to cover more than 80% of RefSeq coding exon positions with a minimum read depth of 10 (supplemental Table 1). By comparing both exome sequences and excluding known poly- morphisms, we were able to identify somatically acquired, leukemia- speci fic sequence variants. Nonsynonymous coding mutations were con firmed by using Sanger sequencing. We found a total of 2 somatic mutations, namely an N676K missense mutation in the ATP-binding domain (TKD1) of FLT 3 (NM_004119.2: c.2028C .A; Figure 1A-C) 24 and an A251V missense mutation in the CAT gene, which encodes the cytoplasmic enzyme catalase (supplemental Table 2). Recurring FLT3 N676K mutations in CBF AML We sequenced FLT3 exon 16 (containing the codon of N676) in a cohort of 84 AML patients with CBFB/MYH11 rearrangement (71 patients with inv(16) and 13 patients with t(16;16)). Strikingly, we detected heterozygous missense mutations (N/K) at position 676 of FLT3 in 5 patients (6%) with inv(16) or t(16;16) (4 [6%] of 71 and 1 [8%] of 13, respectively). Thus, in AML with a CBFB/MYH11 fusion, FLT3 N676K mutations have a frequency similar to FLT3 D835 mutations (Figure 1D). In 36 AML samples with a t(8;21)(q22;q22) and an RUNX1/ RUNX1T1 fusion, 1 patient with an FLT3 N676K mutation could be identi fied (1 [3%] of 36). None of the CBF AML patients with an FLT3 N676K mutation had an additional FLT3 -ITD or a D835 mutation. In contrast, in 90 AML patients with normal karyotype, we detected only a single patient with an FLT3 N676K mutation, and this patient had a concurrent FLT3-ITD similar to that of the patient described by Heidel et al. 14 The incidence of FLT3 N676K without concurrent ITD in CBF AML (6/120) was compared with the incidence in CN-AML (0/90) by using a two-tailed Fisher ’s exact test ( P 5 .039). These results suggest a speci fic association between FLT3 N676K mutations and CBF leukemias. To determine whether the FLT3 N676K mutations are somatically acquired, we sequenced remission samples where available. Paired diagnostic and remission material was available only from the N676K-positive patient with t(8;21) and from 1 patient with inv Figure 2. Transforming potential of FLT3 mutants in Ba/F3 cells. All experiments (16). In both patients, the N676K mutation could be detected at were performed in triplicates. Error bars represent standard deviation of the mean. diagnosis but not in the remission sample (supplemental Figure 1). (A) Ba/F3 cells expressing indicated FLT3 constructs were seeded at a density of fi 4 3 10 4 cells per mL in the presence or absence of 10 ng/mL IL-3 and 100 ng/mL FL. Deep amplicon sequencing of N676K-positive cases con rmed Viable cells were counted by trypan blue exclusion after 72 hours. (B) Ba/F3 cells variable allele frequencies ranging from 11% to 44% indicating transduced with the indicated FLT3 constructs were seeded at a density of 2 3 10 5 clonal heterogeneity (supplemental Table 5). cells per mL in 0.1% WEHI-conditioned medium and cultured for 10 days. After 72 hours, cells were cleared from previous medium and resuspended in 0% WEHI- conditioned medium. Control cells were cultured in 10 ng/mL IL-3–supplemented Additional mutations in CBFB/MYH11-rearranged AML medium. (C) Cells were cultured in the presence or absence of 10 ng/mL IL-3 for 72 hours and stained with Annexin V and 7-aminoactinomycin D. The percentage of We analyzed mutational hotspots (see Materials and methods) of apoptotic cells was determined by fluorescence-activated cell sorter analysis. several commonly mutated genes ( FLT3 , KIT , KRAS , NRAS , NPM1 , MLL, and WT1 ) in our 84 CBFB /MYH11 -positive cohort. The + + ! WZ! + + From bloodjournal.hematologylibrary.org at UBM Bibliothek Grosshadern on September 7, 2013. For personal use only. BLOOD, 5 SEPTEMBER 2013 x VOLUME 122, NUMBER 10 FLT3 N676K MUTATIONS IN CBF AML 1765 Figure 3. Constitutive activation of FLT3 signaling by the FLT3 N676K mutant. Ba/F3 cells expressing indicated constructs were starved for 24 hours in media containing 0.3% fetal calf serum. Cells were left untreated or were stimulated with 100 ng/mL FL for 10 minutes. Crude cell lysates were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis and analyzed by western blot for phosphorylation of signaling molecules. (A) STAT5, AKT, and MAPK activation was analyzed by using phospho-specific antibodies, and then stripped and reprobed with antibodies against total STAT5, AKT, and MAPK. Ba/F3 native cells were stimulated with 100 ng/mL IL-3 for 5 minutes; control and an antibody against GAPDH were used as loading control. (B) FLT3 receptor was immunoprecipitated with polyclonal FLT3 antibody, analyzed for tyrosine phosphorylation status by immunoblotting with a phospho-tyrosin antibody, stripped, and reprobed with FLT3 antibody. mutation frequency of these genes in our cohort (Figure 1D) was (14%), more than one mutation was present. Seventy-nine percent of similar to that in previous reports. 7,8,25-27 We found KIT mutations the patients (66/84) carried a mutation in FLT3 , KRAS , NRAS , or KIT . in 18% of CBFB/MYH11-positive AMLs (14% exon 8 frameshift 7,9 mutations, known to be frequent in inv(16) AML, and 4% D816 FLT3 N676K is strongly expressed on the cell surface of missense mutations). RAS missense mutations were present in Ba/F3 cells 51% of the patients (14% KRAS , 37% NRAS ). As expected, the FLT3 -ITD mutation was rare in our cohort (2%), whereas the To analyze the transforming potential of the FLT3 N676K mutant, FLT3 -TKD (D835) mutation had a frequency of 10%. Together Ba/F3 cell lines stably expressing various FLT3 constructs were with the 6% FLT3 N676K-mutated patients, a total of 18% established. The expression of WT and mutant (mut) FLT3 receptors (15/84) of the CBFB/MYH11-positive patients had an FLT3 was con firmed by immunoblotting or flow cytometry (supplemental mutation. Figure 2). Like FLT3-WT, the FLT3 N676K receptor was highly In addition to the common signaling pathway mutations, we expressed on the cell surface (mature receptor, 160 kDa) compared found 7% of samples with WT1 mutations causing a frameshift in with FLT3-ITD and FLT3 D835Y. The FLT3-N676K-ITD double exon 7. MLL PTDs (1%) and NPM1 mutations (0%) were rare in our mutant showed the weakest cell surface expression. A weak cell CBFB /MYH11 -positive patients. In 18% of the patients, no mutation surface expression was correlated with an enhanced expression of was detected in the mutational hotspots analyzed. In 12 patients the immature receptor with a molecular weight of 130 kDa. 28 + + ! W[! + + From bloodjournal.hematologylibrary.org at UBM Bibliothek Grosshadern on September 7, 2013. For personal use only. 1766 OPATZ et al BLOOD, 5 SEPTEMBER 2013 x VOLUME 122, NUMBER 10 molecules downstream of FLT3: the mitogen-activated protein kinase (MAPK), protein kinase B (AKT), and signal transducer and activator of transcription 5 (STAT5). Protein lysates of unstimulated and FL-stimulated Ba/F3 cells expressing FLT3 and the mutants were immunoblotted (Figure 3A). MAPK was strongly activated in all FL-stimulated FLT3 or FLT3 mutant-expressing cells. In contrast to FLT3-ITD –expressing cells, STAT5 was not phosphorylated in FLT3-WT or in the FLT3 N676K or the FLT3 D835Y mutant-expressing cells. MAPK was constitutively phosphorylated in unstimulated cells of the two TKD mutants N676K and D835Y compared with WT and ITD cells. To determine the activation of the FLT3 N676K mutant receptor, the protein was immunoprecipi- tated and analyzed for tyrosine phosphorylation by immunoblotting. The FLT3 N676K receptor showed a fivefold stronger constitutive phosphorylation as well as a twofold stronger phosphorylation after ligand stimulation compared with FLT3-WT, taking into account the total protein loaded on the gel (Figure 3B). In conclusion, FLT3 N676K mutant-expressing cells showed an enhanced signaling through the MAPK pathway but no aberrant activation of STAT5. Thus, the increased MAPK activation is most likely responsible for the mutant phenotypes. FLT3 N676K-induced proliferation can be abrogated by selective PTK inhibition The FLT3 N676K mutation was previously described only in com- bination with an FLT3 -ITD to mediate resistance to PTKIs. 14 In previous studies it was not tested whether FLT3 N676K alone might be suf ficient to confer protein tyrosine kinase inhibitor resistance. To address this question, we used PKC412 and AC220 as selective FLT3 inhibitors in increasing nontoxic concentrations (Figure 4). Nontoxicity of the inhibitors was con firmed in FLT3-WT –expressing Figure 4. FLT3 N676K but not FLT3-ITD N676K is sensitive to AC220 and PKC412. Ba/F3 cells expressing indicated FLT3 variants were seeded at a density of 4 3 10 4 Ba/F3 cells (supplemental Figure 5). Both compounds potently cells per mL and counted by trypan blue exclusion after 72 hours. All experiments were inhibited FLT3-ITD –expressing cells with a half maximal inhib- performed in triplicate. Error bars represent standard deviation of the mean. itory concentration (IC50 ) of 13 nM for PKC412 and 2.5 nM for (A) Cells were treated with increasing nontoxic concentrations of selective TKI AC220. (B) Cells were treated with increasing nontoxic concentrations of TKI PKC412. AC220, respectively. FLT3 N676K-expressing cells were also sensitive to FLT3 inhibitors with an IC 50 of 7.5 nM for PKC412 and 3 nM for AC220. FLT3-ITD-N676K double mutants showed The FLT3 N676K mutant receptor leads to cytokine-independent a strong resistance to both inhibitors (IC 50 greater than 80 nM for growth and resistance to apoptosis PKC412 and greater than 16 nM for AC220). Taken together, the FLT3 N676K mutation with an ITD is very resistant to FLT3 Proliferation assays of Ba/F3 cells expressing FLT3 mutant receptors inhibitors. However, cell proliferation driven by FLT3 N676K revealed a cytokine-independent growth. As described before, alone can be inhibited rather effectively. FLT3-ITD was able to fully transform Ba/F3 cells reaching 100% – of IL-3 mediated growth. FLT3 D835Y-expressing cells reached Differential gene expression in FLT3 N676K-mutated 41%. The mutant FLT3 N676K receptor led to IL-3 and FLT3- CBFB/MYH11-rearranged AML ligand (FL) independent cell growth, and the Ba/F3 cells reached about 25% of the IL-3 reference proliferation rate at 72 hours of To assess the impact of FLT3 N676K mutations on gene expression, culture time (Figure 2A). This pro-proliferative phenotype increased we analyzed the gene expression pro files of 33 patients with CBFB / over time and, eventually, FLT3 N676K-expressing cells reached MYH11 rearranged AML. Four patients with FLT3 N676K mutations a proliferation rate similar to that of FLT3 D835Y-expressing cells were compared with 29 patients with FLT3 D835 (n 5 4), NRAS (Figure 2B). In addition to an enhanced proliferation, Ba/F3 cells (n 5 15), KRAS (n 5 3), KIT (n 5 3), WT1 (n 5 1), and FLT3 -ITD expressing the various FLT3 mutants showed a strong resistance to (n 5 2) mutations or no mutations in any of these genes (n 5 5). apoptosis after cytokine deprivation (Figure 2C). This antiapop- Some patients had no (n 5 5) or more than 1 (n 5 3) mutation. totic phenotype was strongest in FLT3-ITD –expressing cells (only Finally, all unique probe sets with P , .005 and log fold-change .1.5 4.5% apoptotic cells) followed by FLT3 D835Y (7%) and FLT3 were selected for unsupervised clustering (n 5 18). Interestingly, N676K (10%)-expressing cells. all cases with FLT3 N676K clustered together (supplemental Figure 3). Of these 18 genes, six were highly correlated with sex, Constitutive activation of FLT3 signaling in FLT3 since all cases with N676 mutation in our analysis were discovered N676K-expressing cells in male patients. Interestingly, genes with high association and elevated levels in the N676K cluster were CCNA1 (cell cycle), To determine critical pathways for the transforming potential of PRG3 (immune response), and HLA-DQA1 (immune response). the FLT3 mutants, we analyzed the activation of 3 key signaling Genes with negative correlation to the N676K cluster were MEST + + ! W\! + + From bloodjournal.hematologylibrary.org at UBM Bibliothek Grosshadern on September 7, 2013. For personal use only. BLOOD, 5 SEPTEMBER 2013 x VOLUME 122, NUMBER 10 FLT3 N676K MUTATIONS IN CBF AML 1767 Figure 5. Structural mapping of N676K. Structure of the autoinhibited FLT3 kinase (Protein Data Bank accession number 1RJB) is shown as a ribbon model with highlighted secondary structure and color-coded domains. N676 forms hydrogen bonds to the backbone of H671, stabilizing a loop at the back of the substrate and inhibitor-binding pocket (asterisk). N676K will remove these hydrogen bonds, likely destabilizing the loop and the nearby substrate-binding pocket. This structural effect can explain resistance against TKIs, which target the nearby pocket. However, the mutation could also lift the autoinhibition of FLT3, providing a possible explanation for the observation that this mutation alone shows transforming potential. (imprinting) and ARG1 (metabolism). To evaluate which pathways cohort, which was homogeneous with regard to both treatment and were associated with FLT3 N676K mutations, we compared the 4 cytogenetics, the mutation was signi ficantly associated with higher patients with this mutation to 29 patients without this mutation. leukocyte counts ( P 5 .02), elevated lactate dehydrogenase ( P 5 .02), Nine gene sets were signi ficantly enriched at a false discovery rate of and male sex ( P 5 .02) (Table 1). There was no signi ficant difference ,25% and P , .05 including metabolic, in flammation and deg- in survival of patients with FLT3 N676K (n 5 4) compared with radation pathways (supplemental Table 4). patients with FLT3 N676 wt (n 5 47) (supplemental Figure 4A). However, there was a trend toward reduced complete remission FLT3 Structural mapping of the FLT3 N676K receptor mutation rates associated with N676K mutations (Table 1). We also compared CBFB /MYH11- rearranged AML patients with Since we could demonstrate that a single point mutation in the ATP- FLT3 point mutations affecting residues N676 or D835 (n 5 9) to all binding domain is suf ficient to constitutively activate the receptor other CBFB/MYH11 -rearranged patients (n 5 42) and did not and increase downstream signaling, we performed structural modeling observe a signi ficant difference in survival (supplemental Figure 4B). of the FLT3 N676K mutant to gain further insights into the conse- quences of the mutation (Figure 5). Mapping of the FLT3 N676K onto the crystal structure of FLT3 showed that this mutation destabilizes the fold of the kinase domain between the juxtamembrane domain (JMD) and a hydro- Discussion phobic pocket that is the target of FLT3 inhibitors. The crystal Ours is the first report of recurring FLT3 N676K mutations in CBF structure of the inactive conformation of FLT3 showed that the leukemia. Despite the overall rather favorable prognosis associated JMD serves as a key autoinhibitory element regulating the kinase with CBF rearrangements, almost one third of patients relapses activity.29,30 N676K mutations might therefore interfere with the within the first year after intensive chemotherapy and only 60% FLT3 autoinhibition by reducing the stability of the JMD, thus, of CBF AML patients are still alive after 5 years. This hetero- suggesting a structural basis for the transforming activity observed geneous clinical outcome of CBF AML patients may re flect the in our experiments with Ba/F3 cells. heterogeneity of additional genetic lesions in this subgroup and Clinical characteristics associated with FLT3 N676K mutations underscores the need for further investigation. Understanding the pathogenesis of AML is challenging because of the multitude of Fifty-six of the 84 CBFB /MYH11- rearranged AML patients screened genetic events. By sequencing known mutational targets, we and for mutations in this study were enrolled in the multicenter AMLCG- others have demonstrated that between 80% and 90% of CBFB/ 1999 trial of the German AML Cooperative Group (NCT00266136). MYH11-rearranged patients have mutations that activate either Among these patients, five carried FLT3 N676K mutations. In this RAS signaling or RTK signaling ( FLT3 and KIT ), while other + + ! W]! + + From bloodjournal.hematologylibrary.org at UBM Bibliothek Grosshadern on September 7, 2013. For personal use only. 1768 OPATZ et al BLOOD, 5 SEPTEMBER 2013 x VOLUME 122, NUMBER 10 Table 1. Characteristics of CBFB /MYH11 -rearranged patients gain-of-function properties. Notably, FLT3 N676K alone has FLT3 N676K direct transforming potential in Ba/F3 cells through increased FLT3 N676 wt mut downstream signaling similar to that of the FLT3 D835 mutant Variable No. % No. % P but weaker than FLT3-ITD (Figures 3 and 4). The power of the No.ofpatients 51 5 gene expression analysis of the FLT3 N676K-mutated patients Age, years N/S is limited by the small sample size. However, FLT3 N676K-mutated Median 42 54 cases clustered together after unsupervised clustering analysis of Range 20-75 18-61 gene expression in CBFB /MYH11 -rearranged AML with different Male sex 22 43.1 5 100 .02 mutations affecting FLT3 , NRAS , KRAS , KIT , and WT1 (supple- 3 9 White blood cells 10 /L .02 mental Figure 3). These findings suggest a distinct biologic sub- Median 38.5 134 group within CBFB /MYH11 -rearranged AML characterized by FLT3 Range 1.3-316 54.1-259 N676K mutations. Hemoglobin, g/dL N/S 12 Median 8.6 9.1 On the basis of the crystal structure of FLT3, Cools et al Range 4.4-14.6 8-14.1 proposed that mutation of N676 destabilizes the conformation of Platelets 3 10 9/L N/S the hinge segment, which makes H-bonds with the lactam ring of Median 32 44 the PTKI PKC412. We suggest that a mutation at position N676 Range 0.01-370 32-47 may also activate FLT3 by disturbing its autoinhibition capacity LDH (U/L) .02 (Figure 5). Taken together, mutations at position N676 most probably Median 666 1326 have two consequences: activating FLT3 and, together with a Range 143-1870 717-2508 concurrent FLT3-ITD, conferring PTKI resistance. The additional Bone marrow blasts (%) N/S N676K mutation on the ITD background might change the con- Median 80 60 formation of the FLT3-ITD protein in a way that the binding site of Range 25-95 10-90 ECOG performance 14 29.2 1 20 N/S the PTKIs is masked, since the ITD results in an extension of the status ‡ 2 (%) JMD, which leads to a conformational change of the kinase domain. denovoAML 46 90.2 3 75 N/S The fact that the FLT3 N676K alone (without concurrent ITD) NPM1 mut 0 0 0 0 N/S does not confer PTKI resistance, might also be related to its local- FLT3 -ITD 2 3.9 0 0 N/S ization on the cell surface, in contrast to the mostly intracellular FLT3 -D835 5 9.8 0 0 N/S localization of the ITD-N676K double mutant. Hence, N676K- MLL -PTD 1 2 0 0 N/S mutated FLT3 might be exposed to higher inhibitor concentrations KRAS mut 7 13.7 0 0 N/S at the cell surface, possibly allowing ef ficient inhibition of the NRAS mut 21 41.2 0 0 .15 TKDs directly beneath the cell membrane. It was shown by others KIT mut 10 19.6 0 0 N/S WT1 that there is higher intracellular accumulation of PTKIs in the more mut 4 7.8 0 0N/S 32 Complete remission 39 76 2 40 .11 sensitive AML cells lines than in the less sensitive ones. Deceased 19 37.2 2 40 N/S In contrast to the initial report of an FLT3 N676K mutation as a late arising, disease-modifying event, detected at the time of clinical All patients were enrolled in the AMLCG-99 trial and received intensive relapse while on PKC412 monotherapy, 14 all of our N676K mu- induction treatment. Categorical clinical variables of the FLT3 N676K-mutated (mut) and FLT3 N676 wild-type (wt) cohorts were compared by Fisher’s exact tations were detectable at initial diagnosis. Since cell proliferation test. The continuous variables were compared by Mann-Whitney U test. P , .05 was driven by FLT3 N676K alone could be greatly reduced by FLT3 considered significant. ECOG, Eastern Cooperative Oncology Group; LDH, lactate inhibitors, N676K-positive patients without concurrent ITD may dehydrogenase. actually bene fit from treatment with FLT3 inhibitors. Our clinical data did not show a signi ficant impact of FLT3 N676K mutations on survival within CBFB /MYH11 -rearranged common AML-related gene mutations (eg, in NPM1, WT1, and AML patients, but there was a trend toward reduced complete MLL ) are rarely found. 7-11 The discovery of FLT3 N676K mutations remission rates (Table 1). The signi ficant association of FLT3 N676K adds another piece to the puzzle of CBF-related leukemogenesis, mutations with higher leukocyte counts, elevated lactate dehydro- suggesting that the proportion of CBF leukemia with activating genase levels, and male sex (Table 1), suggests a distinct biology RTK mutations has been underestimated. Our observation of of these leukemias. Given the small number of FLT3 N676K – concurrent activating mutations in different genes (eg, KRAS and positive patients (n 5 4) in our patient cohort, the prognostic NRAS or NRAS and KIT ; Figure 1D) suggests either clonal signi ficance of the FLT3 N676K mutation needs to be investigated in heterogeneity or multiple additive hits in synergistic pathways larger patient cohorts. within CBFB /MYH11 -rearranged AML. The varying allele frequencies of the FLT3 N676K mutation The specific occurrence of recurring FLT3 N676K mutations in ranging from 14% to 44% in those cases in which the presence of the CBFB/MYH11-rearranged AML subgroup, which accounts for the CBFB/MYH11 rearrangement was detected by fluorescence only 6% of AML, might explain why FLT3 N676K mutations have in situ hybridization in the majority of the bone marrow cells remained undetected in previous full-length FLT3 mutation screens (supplemental Table 5) indicate that the FLT3 N676K mutation did of unselected AML patients. 31 Other large studies limited FLT3 mu- not always represent the dominant leukemic clone at diagnosis. It tational screening to FLT3- ITD mutations (exon 14/15) and TKD2 would be interesting to study the clonal evolution in those cases mutations (exon 20; eg, D835) and thus would have missed FLT3 by assessing the FLT3 N676 status at relapse; unfortunately, no N676K mutations (exon 16).7,26 relapse samples from our patients were available. Even though FLT3 N676K in combination with FLT3-ITD had Although FLT3 has been known for more than a decade to be previously been shown to lead to PTKI resistance, 12,14 we show mutated in about one third of AML patients, it appears that the in this report that the FLT3 N676K mutant on its own exhibits spectrum of FLT3 mutations is still not fully understood. In + + ! XT! + + From bloodjournal.hematologylibrary.org at UBM Bibliothek Grosshadern on September 7, 2013. For personal use only. BLOOD, 5 SEPTEMBER 2013 x VOLUME 122, NUMBER 10 FLT3 N676K MUTATIONS IN CBF AML 1769 particular, de fined genetic subgroups of AML might harbor speci fic FLT3 mutations. Unbiased mutation screening using exome se- Authorship quencing allows the detection of novel sequence variations even in extensively studied genes such as FLT3 . Contribution: S.O., H.P., K.S., S.K.B., and. P.A.G. conceived and designed the experiments; S.O., H.P., N.P.K., B.K., E.Z., and S.K. performed experiments; S.O., H.P., and T.H. analyzed data; K.-P.H. performed structural modeling; S.V. and A.G. provided bioinfor- Acknowledgments matics support; H.B. managed the Genome Analyzer IIx platform; S.O., B.K., A.D., E.Z., P.M.K., S.S., J.B., S.K.B., and K.S. We thank the participating centers of the German Acute Myeloid characterized patient samples; M.C.S., J.B., W.E.B., T.B., B.J.W., Leukemia Cooperative Group 1999 (AMLCG-1999) trial. and W.H. coordinated the German Acute Myeloid Leukemia This work was supported by grant 109031 from the German Cooperative Group clinical trial; P.A.G., K.S., and S.K.B. Cancer Aid (P.A.G and S.K.B.); by National Genome Research supervised the project; and S.O., H.P., T.H., S.K.B., and P.A.G. Network Plus grant PKL-01-GS0876-6 from the Federal Ministry wrote the manuscript. of Education and Research (BMBF) (S.K.B.); and by the Collab- Con flict-of-interest disclosure: P.A.G. and S.K. received hono- orative Research Center 684 Molecular Mechanisms of Normal raria from Illumina. The remaining authors declare no competing and Malignant Hematopoiesis, projects A3, A6, A12 and knock-on financial interests. funding 2011 from the German Research Foundation (DFG) (K.-P.H., Correspondence: Philipp A. Greif, Marchioninistrasse 25, 81377 S.K.B., K.S., and P.A.G.). Munchen,¨ Germany; e-mail: [email protected]. References 1. Speck NA, Gilliland DG. Core-binding factors in 12. Cools J, Mentens N, Furet P, et al. Prediction density oligonucleotide array probe level data. haematopoiesis and leukaemia. Nat Rev Cancer. of resistance to small molecule FLT3 inhibitors: Biostatistics. 2003;4(2):249-264. 2002;2(7):502-513. implications for molecularly targeted therapy of 23. Subramanian A, Tamayo P, Mootha VK, et al. acute leukemia. Cancer Res. 2004;64(18): 2. Mr´ozek K, Bloomfield CD. Clinical significance of Gene set enrichment analysis: a knowledge- 6385-6389. the most common chromosome translocations in based approach for interpreting genome-wide adult acute myeloid leukemia. J Natl Cancer Inst 13. Pauwels D, Sweron B, Cools J. The N676D and expression profiles. Proc Natl Acad Sci USA. Monogr. 2008;(39):52-57. G697R mutations in the kinase domain of FLT3 2005;102(43):15545-15550. confer resistance to the inhibitor AC220. 3. Castilla LH, Perrat P, Martinez NJ, et al. 24. Robinson JT, Thorvaldsd´ottir H, Winckler W, Haematologica. 2012;97(11):1773-1774. Identification of genes that synergize with Guttman M, Lander ES, Getz G, Mesirov JP. Cbfb-MYH11 in the pathogenesis of acute myeloid 14. Heidel F, Solem FK, Breitenbuecher F, et al. Integrative genomics viewer. Nat Biotechnol. leukemia. Proc Natl Acad Sci USA. 2004;101(14): Clinical resistance to the kinase inhibitor PKC412 2011;29(1):24-26. 4924-4929. in acute myeloid leukemia by mutation of Asn-676 25. Bowen DT, Frew ME, Hills R, et al. RAS mutation in the FLT3 tyrosine kinase domain. Blood. 2006; 4. Schessl C, Rawat VP, Cusan M, et al. The in acute myeloid leukemia is associated with 107(1):293-300. AML1-ETO fusion gene and the FLT3 length distinct cytogenetic subgroups but does not mutation collaborate in inducing acute leukemia 15. Greif PA, Dufour A, Konstandin NP, et al. GATA2 influence outcome in patients younger than in mice. J Clin Invest. 2005;115(8):2159-2168. zinc finger 1 mutations associated with biallelic 60 years. Blood. 2005;106(6):2113-2119. CEBPA mutations define a unique genetic entity 5. Murati A, Brecqueville M, Devillier R, 26. Thiede C, Steudel C, Mohr B, et al. Analysis of of acute myeloid leukemia. Blood. 2012;120(2): Mozziconacci MJ, Gelsi-Boyer V, Birnbaum D. FLT3-activating mutations in 979 patients with 395-403. Myeloid malignancies: mutations, models and acute myelogenous leukemia: association with FAB management. BMC Cancer. 2012;12(1):304. 16. Reindl C, Bagrintseva K, Vempati S, et al. Point subtypes and identification of subgroups with poor mutations in the juxtamembrane domain of FLT3 prognosis. Blood. 2002;99(12):4326-4335. 6. Gilliland DG. Molecular genetics of human define a new class of activating mutations in AML. 27. Schnittger S, Schoch C, Dugas M, et al. leukemias: new insights into therapy. Semin Blood. 2006;107(9):3700-3707. Analysis of FLT3 length mutations in 1003 Hematol. 2002;39(4 Suppl 3):6-11. patients with acute myeloid leukemia: correlation 17. Polzer H, Janke H, Schmid D, Hiddemann W, 7. Boissel N, Leroy H, Brethon B, et al; Acute to cytogenetics, FAB subtype, and prognosis in Spiekermann K. Casitas B-lineage lymphoma Leukemia French Association (ALFA); Leuc´emies the AMLCG study and usefulness as a marker for mutants activate AKT to induce transformation Aigu¨es My´eloblastiques de l’Enfant (LAME) the detection of minimal residual disease. Blood. in cooperation with class III receptor tyrosine Cooperative Groups. Incidence and prognostic 2002;100(1):59-66. kinases. Exp Hematol. 2012;41(3):271-280. impact of c-Kit, FLT3, and Ras gene mutations 28. Schmidt-Arras DE, B¨ohmer A, Markova B, in core binding factor acute myeloid leukemia 18. Reindl C, Quentmeier H, Petropoulos K, et al. Choudhary C, Serve H, B¨ohmer FD. Tyrosine (CBF-AML). Leukemia. 2006;20(6):965-970. CBL exon 8/9 mutants activate the FLT3 pathway phosphorylation regulates maturation of receptor and cluster in core binding factor/11q deletion 8. Bacher U, Haferlach T, Schoch C, Kern W, tyrosine kinases. Mol Cell Biol. 2005;25(9): acute myeloid leukemia/myelodysplastic Schnittger S. Implications of NRAS mutations in 3690-3703. syndrome subtypes. Clin Cancer Res. 2009;15(7): AML: a study of 2502 patients. Blood. 2006; 2238-2247. 29. Griffith J, Black J, Faerman C, et al. The 107(10):3847-3853. structural basis for autoinhibition of FLT3 by the 19. Herold T, Jurinovic V, Metzeler KH, et al. An eight- 9. Paschka P, Marcucci G, Ruppert AS, et al; Cancer juxtamembrane domain. Mol Cell. 2004;13(2): gene expression signature for the prediction of and Leukemia Group B. Adverse prognostic 169-178. survival and time to treatment in chronic significance of KIT mutations in adult acute lymphocytic leukemia. Leukemia. 2011;25(10): 30. Walter M, Lucet IS, Patel O, et al. The 2.7 myeloid leukemia with inv(16) and t(8;21): 1639-1645. A crystal structure of the autoinhibited human a Cancer and Leukemia Group B Study. J Clin c-Fms kinase domain. J Mol Biol. 2007;367(3): Oncol. 2006;24(24):3904-3911. 20. Li Z, Herold T, He C, et al. Identification of 839-847. a 24-gene prognostic signature that improves the 10. Valk PJ, Bowen DT, Frew ME, Goodeve AC, European LeukemiaNet risk classification of acute 31. Fr¨ohling S, Scholl C, Levine RL, et al. L¨owenberg B, Reilly JT. Second hit mutations in myeloid leukemia: an international collaborative Identification of driver and passenger the RTK/RAS signaling pathway in acute myeloid study. J Clin Oncol. 2013;31(9):1172-1181. mutations of FLT3 by high-throughput DNA leukemia with inv(16). Haematologica. 2004; sequence analysis and functional assessment 89(1):106. 21. Ferrari F, Bortoluzzi S, Coppe A, et al. Novel of candidate alleles. Cancer Cell. 2007;12(6): definition files for human GeneChips based on 501-513. 11. Haferlach C, Dicker F, Kohlmann A, et al. AML GeneAnnot. BMC Bioinformatics. 2007;8:446. with CBFB-MYH11 rearrangement demonstrate 32. Hu S, Niu H, Minkin P, et al. Comparison of RAS pathway alterations in 92% of all cases 22. Irizarry RA, Hobbs B, Collin F, Beazer-Barclay antitumor effects of multitargeted tyrosine kinase including a high frequency of NF1 deletions. YD, Antonellis KJ, Scherf U, Speed TP. inhibitors in acute myelogenous leukemia. Mol Leukemia. 2010;24(5):1065-1069. Exploration, normalization, and summaries of high Cancer Ther. 2008;7(5):1110-1120. + + ! XU! (++4EOAM=OPMQAMTAE?DIEN $ ! UP! $DB8G(C;(AM:@C>9(A6@> ! XV! U\P! "9E ! XW! W[P! .5E7<>5N!.P!5A8!)P!"L ! XX! )++/=IFN=CPICAI+ + 3\$ RWSaS`$ EbSZZS$ [kQVbS$ WQV$ [WQV$ UO\h$ VS`hZWQV$ PSW$ OZZS\$ PSRO\YS\#$ RWS$ hc[$ 9SZW\US\$RWSaS`$3`PSWb$PSWUSb`OUS\$c\R$[WQV$OcT$RS[$ZO\US\$ISU$PSUZSWbSb$VOPS\%$ $ ?SW\$PSa]\RS`S`$6O\Y$UWZb$[SW\S[$6]Yb]`dObS`#$B`]TSaa]`$EbSTO\$4]VZO\RS`#$RS`$[W`$ RWSaS$6WaaS`bObW]\$S`[kUZWQV$VOb$c\R$[W`$XSRS`hSWb#$W\$OcmS`USekV\ZWQVS[$?OmS#$[Wb$ VWZT`SWQVS\$3\`SUc\US\$c\R$eS`bd]ZZS[$DOb$hc`$ESWbS$abO\R%$ $ ;QV$RO\YS$6`%$[SR%$BVWZW^^$9`SWT$Tl`$RWS$4Sb`Scc\U$RWSaS`$3`PSWb%$ESW\S$9SRcZR#$aSW\S$ 3cTUSaQVZ]aaS\VSWb$ c\R$ aSW\$ HS`b`OcS\$ S`[kUZWQVbS\$ Sa$ [W`$ RWS$ 8`OUSabSZZc\US\$ RWSaS`$3`PSWb$W\$RWSaS[$G[TO\U$aSZPabj\RWU$hc$S\beS`TS\$c\R$hc$PSO`PSWbS\%$ $ ?SW\$6O\Y$UWZb$hcRS[$OZZS\$=]Ocb]`S\$Tl`$RWS$UcbS$LcaO[[S\O`PSWb$hc`$BcPZWYObW]\$ RS`$US[SW\aO[S\$6ObS\#$W\aPSa]\RS`S$6`%$3\W`cRRVO$6SaV^O\RS#$6`%$[SR%$=ZOca$ ?SbhSZS`$c\R$6`%$[SR%$F]PWOa$:S`]ZR%$ $ ;QV$RO\YS$:S``\$B`]TSaa]`$I]ZTUO\U$:WRRS[O\\$Tl`$RWS$?kUZWQVYSWb$RWSaS$3`PSWb$W\$ RS`$ ?SRWhW\WaQVS\$ =ZW\WY$ c\R$ B]ZWYZW\WY$ ;;;$ RS`$ >cReWU$?OfW[WZWO\a$G\WdS`aWbjb$ ?l\QVS\$hc$dS`TOaaS\%$ $ ?SW\$6O\Y$UWZb$\WQVb$hcZSbhb$RS\$?WbUZWSRS`\$RS`$SVS[OZWUS\$==9$>ScYj[WS$c\R$RS\$ =]ZZSUS\$ Oca$ RS[$ >OP]`$ Tl`$ >ScYj[WSRWOU\]abWY%$ 6WS$ T`Sc\RaQVOTbZWQVS$ 3`PSWbaOb[]a^Vj`S$c\R$RWS$abSbS$G\bS`ablbhc\U$]P$bSQV\WaQV$]RS`$[]`OZWaQV#$VOPS\$ RWSaS$ 3`PSWb$ W\$ WV`S[$ G[TO\U$ c\R$ 7`T]ZU$ S`ab$ [kUZWQV$ US[OQVb%$ ?SW\$ UO\h$ PSa]\RS`S`$6O\Y$UWZb$VWS`$8`Oc$4WO\YO$=aWS\hgY$o$RWS$]Tb$aQV]\$d]`$[W`$ecaabS$eO\\$ WQV$:WZTS$O[$R`W\US\RabS\$P`OcQVbS%$ $ G\R$ WQV$ RO\YS$ [SW\S`$ 8O[WZWS$ c\R$ [SW\S\$ 8`Sc\RS\#$ Tl`$ RS\$ DlQYVOZb$ c\R$ RWS$ ?]bWdObW]\%$ $ 6O\YSt$ $ $ ! XY!