1
Nucleophosmin in e ac ion wi h APE1: Insigh s in o DNA epai egula ion.
Da id J. López, Ande de Blas, Mikel Hu ado1, Mikel Ga cía-Alija2, Jon Men xaka,
Igo de la A ada, Ma ía A. U baneja, Ma ián Alonso-Ma iño, Sonia Bañuelos*
Bio isika Ins i u e (UPV/EHU, CSIC) and Depa men o Biochemis y and Molecula
Biology, Uni e si y o he Basque Coun y (UPV/EHU), Leioa, Spain.
P esen add esses: 1NEIKER-Basque Ins i u e o Ag icul u al Resea ch and De elopmen , c/
Be eaga 1, E-48160 De io, Spain. 2CIC bioGUNE , E-48160 De io, Spain.
Abb e ia ions: AML: acu e myeloid leukemia; APE1: apu inic apy imidinic endonuclease
1; BER: Base excision epai ; CD: ci cula dich oism; DDR: DNA damage epai ; DSB:
double s and b eak; FT-IR: Fou ie ans o m in a ed spec oscopy; IR: ionizing adia ion;
ITC: iso he mal i a ion calo ime y; NPM1: nucleophosmin; PA: polyac ylamide; PAGE:
polyac ylamide gel elec opho esis; THF: e ahyd o u ane.
*Co esponding au ho a : Bio isika Ins i u e, Uni e si y o he Basque Coun y, Ba io
Sa iena s/n, E-48940 Leioa, Spain. E-mail add ess: [email p o ec ed]
This is he accep ed manusc ip o he a icle ha appea ed in inal o m in DNA epai 88 : (2020) // A icle ID 102809, which has been
published in inal o m a h ps://doi.o g/10.1016/j.dna ep.2020.102809. © 2020 Else ie unde CC BY-NC-ND license (h p://
c ea i ecommons.o g/licenses/by-nc-nd/4.0/)
2
Abs ac
Nucleophosmin (NPM1), an abundan , nucleola p o ein wi h mul iple unc ions
a ec ing cell homeos asis, has also been ecen ly in ol ed in DNA damage epai . The
oles o NPM1 in di e en epai pa hways emain howe e o be elucida ed. NPM1 has
been desc ibed o in e ac wi h APE1 (apu inic apy imidinic endonuclease 1), a key
enzyme o he base excision epai (BER) pa hway, which could e lec a di ec
pa icipa ion o NPM1 in his ou e. To gain insigh in o he possible ole(s) o NPM1 in
BER, we ha e explo ed he in e play be ween he subnuclea localiza ion o bo h APE1
and NPM1, he in i o in e ac ion hey es ablish, he e ec o binding o abasic DNA
on APE1 con o ma ion, and he modula ion by NPM1 o APE1 binding and ca alysis on
DNA. We ha e ound ha , upon oxida i e damage, NPM1 is eleased om nucleoli and
loca es on pa ches h oughou he ch oma in, pe haps co-localizing wi h APE1, and ha
his a ic could be media ed by phospho yla ion o NPM1 on T199. NPM1 and APE1
o m a complex in i o, in ol ing, apa om he co e domain, a leas pa o he
linke egion o NPM1, whe eas he C- e minal domain is dispensable o binding,
which explains ha an AML leukemia- ela ed NPM1 mu an wi h an un olded C-
e minal domain can bind APE1. APE1 in e ac ion wi h abasic DNA s abilizes APE1
s uc u e, as based on he mal un olding. Mo eo e , ou da a sugges ha NPM1, maybe
by keeping APE1 in an “open” con o ma ion, a ou s speci ic ecogni ion o abasic
si es on DNA, compe ing wi h o - a ge associa ions. The e o e, NPM1 migh
pa icipa e in BER a ou ing APE1 a ge selec ion as well as u no e om incised
abasic DNA.
Keywo ds: APE1, NPM1, nucleophosmin, BER, base excision epai , p o ein-p o ein
in e ac ion.
3
In oduc ion
Nucleolus, beyond i s canonical ibosome assembly unc ion, is being inc easingly
ecognized as a key hub in coo dina ing p og ession o he cell cycle and esponses o
cellula pe u ba ions, such as damage o DNA (Bois e e al., 2007; Emmo &
Hiscox, 2009; An oniali e al., 2014; Ogawa & Base ga, 2017). Upon sensing di e en
ypes o s esses, he localiza ion o se e al nucleola componen s is modula ed and
hei in ol emen in epai asks o ches a ed om his memb ane-less o ganelle (Sco
& Oe inge 2016; An oniali e al., 2014). Nucleophosmin (NPM1) is one o he
nucleola p o eins ha seem o be in ol ed in DNA damage epai (DDR) mechanisms.
NPM1 is homopen ame ic and consis s o mul iple domains: an N- e minal compac , -
s uc u ed co e domain, esponsible o oligome iza ion (Lee e al., 2007), is connec ed
by long, lexible linke s, o small, helical, C- e minal domains (G ummi e al., 2008).
Al hough no mally en iched in nucleoli, NPM1 con inuously shu les be ween
cy oplasm, nucleoplasm and nucleolus (Linds öm, 2011). NPM1 shu ling is media ed
by anspo ecep o s o he ka yophe in amily (impo in expo in CRM1)
(Linds öm, 2011; A egi e al., 2015), whe eas nucleola accumula ion p obably
depends on i s a ini y o G- ich DNA o RNA sequences adop ing special s uc u es
known as G-quad uplexes (Bañuelos e al., 2013; Chia ella e al, 2013). NPM1 is in
cha ge o mul iple unc ions ela ed o cell homeos asis, including biogenesis and expo
o ibosomes (Maggi e al., 2008), con ol o cen osome duplica ion (Wang e al.,
2005), egula ion o he s abili y o umou supp esso s such as p53 (Colombo e al
2002) and A (Colombo e al., 2005), and cell esponse o s ess. Recen expe imen al
e idence poin s o NPM1 as a key ac o di ec ly pa icipa ing in DNA damage epai
(DDR) pa hways, al hough i s ole in hese mechanisms emains o be es ablished (Sco
& Oe inge 2016; Box e al., 2016; Koike e al., 2010; Moo e e al., 2013; Zi e al.,
2014; Guillonneau e al., 2016; Yang e al., 2016).
Rega ding NPM1 ac i i y in he con ex o s ess esponse, i was ound associa ed wi h
ch oma in upon induc ion o double s and b eaks (DSBs) wi h ionizing adia ion (IR)
o e oposide (Lee e al., 2005). The pu a i e changes in subnuclea localiza ion o
NPM1 upon DSBs and i s pa icipa ion in he managemen o hese lesions seem o be
egula ed by phopsho yla ion. T199-phospho yla ed NPM1, which loca es ou side he
nucleoli, is ec ui ed o DSBs epai si es ( oci) in esponse o IR (Koike e al., 2010).
Mo eo e , NPM1 unde goes nucleoplasmic ansloca ion upon di e en geno oxic
4
insul s, such as oxida i e s ess, UV, cispla in, e c. (Colombo e al., 2011; Sco &
Oe inge 2016; Yang e al., 2016) and, apa om DSBs epai , i has been implica ed
in ye u he DNA epai pa hways, namely anslesion syn hesis (TLS) (Zi e al.,
2014) and base excision epai (BER) (Vasco o e al., 2009; Vasco o e al., 2014).
NPM1 has been desc ibed o in e ac wi h APE1 (apu inic apy imidinic endonuclease 1)
(Vasco o e al., 2009), a key enzyme o he BER pa hway, aimed o epai DNA
damage caused by oxidizing o alkyla ing agen s, including chemo he apeu ic d ugs
(Almeida & Sobol, 2007; K okan & Bjo as, 2013; Pa sons & Diano , 2013). APE1
speci ically ecognizes abasic si es in DNA and hyd olyzes he phosphodies e bond
5´o he abasic suga (F euden hal e al., 2015), gene a ed a e emo al o a modi ied /
damaged base by a speci ic glycosidase, o by spon aneous depu ina ion (Li and
Wilson, 2014). A e wa ds, a polyme ase es o es he lacking nucleo ide using he
complemen a y s and as empla e, and he phosphodies e link is e- o med by a ligase.
APE1 pe o ms addi ional unc ions in he con ex o DNA epai (being able o co ec
“non p oduc i e” ends o DNA s ands in epai in e media es) and, p obably, in RNA
managemen (as an “RNA quali y con ol”) (An oniali e al., 2014). Besides i s oles in
DDR, APE1 ac s as a edox coac i a o o di e en ansc ip ion ac o s such as p53
and NF-B. APE1 ca aly ic cycle on DNA has been desc ibed in s uc u al de ail (Mol
e al., 2000; F euden hal e al., 2015); howe e , he mechanisms egula ing APE1
unc ionali y a e la gely unknown.
NPM1 migh ha e a ole in BER epai , modula ing he localiza ion o APE1 (Vasco o
e al., 2014). In he absence o DNA damage, APE1 loca es h ough nucleoplasm and is
en iched in nucleoli, which is p obably media ed by i s in e ac ion wi h NPM1 h ough
a lexible, N- e minal egion o APE1. This egion seems o also play a egula o y ole
in APE1 binding o DNA (Fan ini e al., 2010; Vasco o e al., 2014). Fu he mo e,
ace yla ion o lysine esidues in his segmen migh modula e he in e ac ion wi h
NPM1 and hus APE1 nucleola localiza ion (Li ussi e al., 2012). The ac ha , as
men ioned, NPM1 i sel may ansi o he nucleoplasm depending on s ess condi ions
and i s own pos ansla ional modi ica ions adds u he complexi y o he egula ion o
NPM1 and APE1 dynamics. A pool o APE1 seems o loca e also in mi ochond ia, and
could ake pa in he epai o oxidized DNA in his o ganelle (Li & Wilson, 2014). On
he o he hand, NPM1 pa icipa ion in mul iple, mechanis ically di e en epai
pa hways migh be also based on i s ch oma in emodelling (his one chape oning)
5
capaci ies (Okuwaki e al., 2001). A u he , as ye unexplo ed possibili y is ha NPM1
akes pa di ec ly in he BER enzyma ic machine y. Indeed, NPM1 seems o sligh ly
s imula e APE1 incision ac i i y on abasic DNA in i o (Vasco o e al., 2009). A
di ec in ol emen o NPM1 in epai asks emains howe e o be con i med, and he
unde lying mechanism deciphe ed.
DNA epai machine ies a e aluable a ge s in cance , as long as in e e ing wi h hei
unc ioning can enhance he e iciency o an ineoplasic he apies (Pilié e al., 2018). In
his sense, APE1 is conside ed a a ge o chemosensi iza ion and adiosensi iza ion
(Wilson & Simeono , 2010; Li & Wilson, 2014). On he o he hand, NPM1 is ega ded
as an oncop o ein, being de egula ed (o e exp essed, mu a ed and/o abe an ly loca ed)
in se e al ypes o human cance (G isendi e al., 2006). In pa icula , i is mu a ed,
mis olded, and pa ly accumula ed in he cy oplasm o umou cells in a sub ype o
acu e myeloid leukemia (AML) (Falini e al., 2005; Fede ici & Falini, 2013). NPM1
binds many ligands, so ha when abe an ly loca ed as in AML, migh induce
subsequen mislocaliza ion o some o hese pa ne s. Indeed, i has been desc ibed ha
APE1 (Vasco o e al., 2014), amongs o he p o eins, may be, as a esul , disloca ed o
cy oplasm in AML, an al e a ion pe haps ela ed wi h igge ing o he disease.
To be e unde s and he ole o NPM1 in DNA epai , as well as he pa hological
consequences o i s mislocaliza ion, we ha e explo ed he subnuclea dynamics o
NPM1 and APE1 unde s ess condi ions. In addi ion, we ha e cha ac e ized he
in e ac ion be ween bo h p o eins and he e ec s o oncogenic mu a ions o NPM1 on
he binding p ope ies. Finally, we ha e e alua ed he consequences o DNA binding on
APE1 con o ma ion, and he e ec o NPM1 on APE1 binding o abasic DNA.
Al oge he , ou esul s sugges a mechanism o egula ion o APE1 ac i i y by NPM1
and shed ligh on he pu a i e ole o APE1:NPM1 in e ac ion in BER epai .
1. Ma e ial and me hods.
1.1. Cell cul u e, ans ec ion, ea men , imaging and immunoblo ing.
The clone o human APE1 used o EGFP was kindly p o ided by D . Izumi (Jackson e
al, 2005). The gene was subcloned used o C- e minal mChe y, be ween XhoI and
SmaI si es o ec o pCRY2PHR-mChe yN1, a gi om Chand a Tucke (Kennedy e
6
al., 2010). Human HeLa and HEK293T cells (DSMZ collec ion) we e g own in
Dulbecco’s modi ied Eagle’s medium, supplemen ed wi h 10% e al bo ine se um, 100
uni s/mL penicillin and 100 μg/mL s ep omycin (all om In i ogen). T ans ec ions o
cells, seeded on o s e ile glass co e slips in 12-well ays he p e ious day, we e ca ied
ou wi h X emeGENE 9 (Roche) ollowing he manu ac u e ’s p o ocol.
Induc ion o oxida i e damage and ex ac ion o soluble p o eins we e done ollowing a
p e iously desc ibed p o ocol (Campalans e al., 2013). The nex day a e ans ec ion,
cells we e ea ed wi h 0 – 300 mM KB O3 in PBS o 30 min a 37ºC, and a e wa ds
g own o 3 h longe in esh medium, in o de o “ eco e ”. Then, p io o ixa ion wi h
3.7% o maldehyde in PBS o 30 min, cells we e washed wi h 10mM PIPES pH 6.8,
100mM NaCl, 300mM glucose, 3mM MgCl2, 0.5% T i on X-100 and a cock ail o
p o ease inhibi o s (Roche) (CSK bu e , Campalans e al., 2013), o ex ac soluble
p o eins. O he ea men s we e done wi h 0.5 and 5 mM H2O2 (Sigma) o 30 min, 5
g/mL neoca zinos a in (Sigma) o 4 h o 0.5 g/mL nocodazole (Sigma) o 24 h.
Fo luo escence immunos aining, ixed cells we e u he pe meabilized wi h 0.2%
T i on X-100 in PBS o 10 min, and incuba ed o 1 h in blocking solu ion (3% BSA in
PBS). Endogenous NPM1 was de ec ed wi h mouse an i-B23 (San a C uz
Bio echnology FC82291, dilu ion 1:800 in blocking solu ion), and NPM1
phospho yla ed a Th 199 wi h abbi an i-NPM1p199 (Abcam EP1857Y, dilu ion
1:150). The seconda y an ibodies we e Alexa Fluo 594 an i-mouse (In i ogen
A11005, dilu ion 1:400) and Alexa Fluo 594 an i- abbi (In i ogen A11012, dilu ion
1:400). Washing be ween he di e en s eps was ca ied ou wi h PBS. Cells we e
inally moun ed on o mic oscope slides using Vec ashield con aining DAPI (Vec o
Labo a o ies). Cell imaging was pe o med wi h he suppo o SGIke Se ice
(Uni e si y o he Basque Coun y UPV/EHU). Rou ine images we e aken wi h Zeiss
Axioscope o Axio Obse e Apo ome 2 mic oscope. High esolu ion images (Fig. 1)
we e acqui ed in a Zeiss LSM880 wi h a 63x lens (Plan Apoch oma , 1.4 NA), using he
Fas Ai yscan supe esolu ion mode
Two days la e , a e he di e en ea men s, hey we e gen ly washed wi h cold PBS,
sc aped on ice and esuspended in 50 L o lysis bu e (Pie ce) supplemen ed wi h a
cock ail o inhibi o s (Roche). Following lysis o 30 min and cen i uga ion, samples
we e loaded in a 12.5% PAGE gel and hen ans e ed o a ni ocellulose memb ane
(GE Heal hca e) in a we appa a us (Bio-Rad). Then, he memb ane was blocked wi h
7
5% (w/ ) non‐ a d y milk powde in TBS‐T and incuba ed wi h he same p ima y
an ibodies (an i-NPM1, dilu ed 1:800 in blocking solu ion, and an i- NPM1p199,
dilu ed 1:2500) as o immunos aining. As con ols o loaded amoun , an i- ubulin
(Sigma Ald ich, T9026, dilu ion 1:5000) and an i-GAPDH (San a C uz Bio echnology,
sc-25778, dilu ion 1:2000) we e used. A e washing s eps wi h TBS-T, he memb ane
was incuba ed wi h he co esponding an i-mouse and an i- abbi HRP- conjuga ed
an ibodies and inally de ec ion was made wi h he Lumina a Fo e Wes e n HRP
Subs a e (Millipo e).
1.2. P o eins o e exp ession and pu i ica ion.
Full leng h wild ype NPM1, mu an A (NPM1Mu A), and nucleophosmin spanning
esidues 1-188 (NPM1C106), all wi h an N- e minal His-ZZ ag, we e o e exp essed
in E. coli BL21 (DE3), and pu i ied as in A egi e al. (2015). Pu i ica ion in ol ed Ni-
NTA ch oma og aphy, ags emo al wi h TEV p o ease, e e se Ni-NTA and a inal
size exclusion ch oma og aphy (SEC) s ep. His agged NPM1 co e domain was a clone
ob ained om D . Se Won Suh (Lee e al., 2007). Fo pu i ica ion o his cons uc , cells
we e dis up ed by sonica ion in 25 mM T is-HCl pH 7.5, 100 mM NaCl, 10% glyce ol,
1mM DTT, supplemen ed wi h lysozyme (20 mg/L cul u e) and a cock ail o p o ease
inhibi o s (Comple e, EDTA- ee, Roche). The cla i ied ex ac was loaded on a Ni-
NTA a ini y column (His T ap FF, GE Heal hca e) and he p o ein elu ed wi h an
imidazole g adien . Then i was u he pu i ied by gel il a ion ch oma og aphy wi h
Supe dex 200 (GE Heal hca e) in 25 mM T is-HCl pH 7.5, 100 mM NaCl, concen a ed
and ozen o s o age. APE1 ( ull leng h and APE1N33) was subcloned in he same
plasmid as NPM1 (pTGA20) and pu i ied using he same ch oma og aphic s eps. SEC
was done in bu e 25 mM T is-HCl pH 7.5, 100 mM NaCl, 1 mM DTT, 10% glyce ol.
A e pu i ica ion, all p o eins we e concen a ed, lash ozen wi h liquid ni ogen and
s o ed a -80 ºC. They we e quan i ied wi h NanoD op 2000 (The mo Scien i ic), based
on 280 as calcula ed wi h P o Pa am (Gas eige e al., 2005).
1.3. DNA oligonucleo ides.
As model o abasic DNA, we used a dumbbell-shaped oligonucleo ide o 46 esidues,
wi h a cen ally loca ed e ahyd o u ane (THF) (F euden hal e al., 2015) 5´-
CTGGAGCTTGCTCCAG CGCXCGGTCGATCGTA AGATCGACCGTGCG-3′,
whe e X ep esen s THF), syn hesized by IDT (Leu en, Belgium). Fo aniso opy
8
binding assays, a luo escein-labelled (in hymine 28) oligo wi h he same sequence was
used. To p epa e he “subs a e” (“Dumbbell S” o “Dumbbell SF” when luo escen ly
labelled), he oligos we e esuspended a 100 M in wa e , annealed by hea ing 5 min a
95 ºC and slowly cooling down in a he moblock, and liga ed wi h T4 ligase (New
England Biolabs) (4000 U ligase pe nmol o DNA). Liga ion o nea 100% o he DNA
was con i med by dena u ing polyac ylamide-u ea elec opho esis (see sec ion 2.8). Fo
ac i i y and binding assays he DNA was pu i ied om ligase and magnesium wi h he
“Nucleo ide Remo al Ki ” (QIAGEN) and e-annealed; his could no be done o CD
and FT-IR, hough, since due o he highe amoun needed, he los o yield was oo
high o be a o ded. FT-IR spec a, mos CD measu emen s and also some binding
assays we e pe o med wi h a model o p oduc DNA (“Dumbbell P” and “Dumbbell
PF”, espec i ely) (F euden hal e al., 2015), wi h he same sequence as he subs a e
bu s a ing wi h 5´ THF.
1.4. Na i e elec opho esis.
Samples con aining a ixed concen a ion o NPM1 and inc easing amoun s o APE1 in
bu e 20 mM po assium phospha e, 50 mM NaCl, 5 mM MgCl2, 2 mM DTT, 0.01%
(w/ ) Tween-20, pH 7.0 ( o al olume o 20 L) we e incuba ed o 30 min a oom
empe a u e and loaded in p ecas na i e 4-16% polyac ylamide (PA), 10 wells Bis-T is
gels (In i ogen). Running bu e and sample bu e (wi hou G-250) we e also om
In i ogen. Gels we e un o 130 min a 150 V a 4 ºC, s ained wi h Coomassie
solu ion, and pho og aphed wi h Gel Doc EZ Sys em (Bio-Rad). Densi ome y analysis
o he p o ein bands was pe o med wi h Quan i y One (Bio-Rad), and he measu ed
in ensi ies we e i ed o a quad a ic unc ion o ob ain a binding cu e. 3-5 expe imen s
we e a e aged o es ima e KD alues. Te na y mix u es NPM1/APE1/DNA we e also
analyzed by na i e elec opho esis using he same condi ions, bu in his case gels we e
also s ained wi h GelRed (Bio ium) p io o Coomassie.
1.5. Iso he mal i a ion calo ime y (ITC).
NPM1/APE1 binding was measu ed wi h a Nano-ITC (TA ins umen s) (Low Volume,
wi h a cell o 189 L). Bo h p o eins we e dyalized in bu e 20 mM po assium
phospha e, 50 mM NaCl, 5 mM MgCl2, 2 mM TCEP, pH 7.0, cen i uged and degassed
p io o he measu emen . 200-500 M APE1, was i a ed on o NPM1, ei he wild ype
9
o mu an A (5-30 M pen ame ), in 33 injec ions o 1.5 L, a 20 ºC. Dilu ion con ols
we e pe o med dilu ing APE1 in o bu e . Fo baseline co ec ion, he a e age o he
hea o he 3 las injec ions was subs ac ed om he iso he m. Iso he ms we e analyzed
wi h he Nano Analyze so wa e (TA), using a model o independen binding si es.
1.6. Ci cula dich oism (CD).
Ci cula dich oism measu emen s we e pe o med in a Jasco 720 spec opola ime e
equipped wi h Pel ie empe a u e con ol, in a cu e e o 0.2 cm pa hleng h, in bu e
20 mM po assium phospha e pH 7.0, 50 mM NaCl, and ei he 0.5 mM EDTA o 5 mM
MgCl2. Concen a ions we e 4 mM o NPM1, APE1 o oligo. Tempe a u e scans we e
done a 60 ºC /h.
1.7. Fou ie ans o m in a ed (FT-IR) spec oscopy.
Fo in a ed spec oscopy, samples con ained 100 M (o 50 M, in he mix u es wi h
DNA) APE1, ei he alone, o in he p esence o equimola amoun s o NPM1 pen ame
o DNA oligo (“p oduc ”). The co esponding con ol samples o NPM1 and DNA we e
also measu ed. Fo H/D exchange, samples we e subjec ed o 6-7 cycles o
concen a ion wi h Amicon Ul a 0.5 mL il e uni s (Millipo e) o 10 K cu o (excep
o DNA alone, whe e 3K was used), and dilu ion in D2O bu e (20 mM po assium
phospha e, 50 mM NaCl, 5 mM MgCl2, 2 mM DTT, 0.01% (p/ ) Tween-20, pD 7.0).
The inal concen a ion was checked by UV abso p ion and eadjus ed as necessa y.
Spec a we e eco ded in a The mo Nicole 5700 spec ome e equipped wi h a MCT
de ec o using a Pel ie -based empe a u e con olle wi h 25 m-pa hleng h calcium
luo ide cells (BioTools). A 25-μl sample aliquo was deposi ed on a cell ha was
sealed wi h a second cell. Typically 370 scans we e collec ed o each backg ound and
sample, and he spec a we e ob ained wi h a nominal esolu ion o 2 cm−1. Tempe a u e
was inc eased a a a e o 1 °C min−1 be ween 20 and 80 °C o all samples.
1.8. APE1 incision assays.
Incision o 5 M abasic DNA model oligonucleo ide (Dumbbell S) was assayed in
bu e 20 mM po assium phospha e, 50 mM NaCl, 5 mM MgCl2, 2 mM DTT, pH 7.0,
in he p esence o absence o 5 M NPM1, and moni o ized by dena u ing 18%
polyac ylamide (PA)-u ea gels (Adachi e al., 2015). The assay was done ei he as a
unc ion o APE1 concen a ion (0-200 pM) in eac ions o 15 min a 37 ºC, o as a
16
accoun ing o 10-15% o he pep ide bonds, migh be hough o co espond o
ea angemen s o he N- e minal lexible egion, no isible in he c ys al s uc u es.
We ha e also analyzed he CD and FT-IR spec a as well as he mal s abili y wi h and
wi hou DNA o he unca ed o m o APE1, lacking he N- e minal 33 esidues
(APE1N33), which is ca aly ically ac i e (Mol e al., 2000; Li ussi e al., 2012).
Binding o APE1N33 o he oligo Dumbbell P also esul s in simila al e a ions o he
Amide I band (Fig. 6B) and he mal s abili y, as ollowed ei he by ellip ici y a 222 nm
(no shown) o wid h o he Amide I band (Fig. S64). The e o e, he changes induced by
DNA binding canno be assigned o he N- e minal 33 esidues, bu pe haps o he
ollowing ca. 30 esidues egion, also lexible, and/o p o ein loops. O e all, and in line
wi h p e ious epo s (Yu e al., 2010) ou da a indica e ha APE1 (ei he ull leng h o
isola ed globula domain) con o ma ion o dynamics is sensi i e o abasic DNA
binding.
We ha e also explo ed he con o ma ional consequences o APE1/NPM1 complex
o ma ion, using wild ype APE1 and he unca ed mu an NPM1C106, which as
men ioned, is su icien o binding, bu lacks he mal ansi ions in he 20-80ºC ange.
A concen a ions well abo e he KD, we could no obse e any change nei he in he
spec al p ope ies o bo h p o eins, no in he he mal p o ile o APE1 (no shown).
This obse a ion suppo s he no ion ha NPM1 binds mainly o he N- e minal
segmen o APE1, as desc ibed (Vasco o e al., 2009), ha ing li le e ec on he
con o ma ion o APE1 globula , majo domain. Finally, in e na y mix u es
APE1/DNA/ NPM1C106 we could no obse e any e ec o NPM1 on he DNA-
bound APE1 con o ma ion o on he s abiliza ion b ough abou by DNA (no shown).
2.6. NPM1 a ou s he speci ic binding o APE1 o abasic subs a e DNA.
To ge insigh in o he pu a i e egula o y ole o NPM1 on APE1 ca alysis, we ha e
analyzed he e ec o NPM1 on he ecogni ion and ca aly ic incision o an abasic DNA
by APE1. Fi s we ha e con i med ha ou p epa a ion o ecombinan APE1 is highly
ac i e clea ing he model oligonucleo ide o abasic DNA (Dumbbell S), as ollowed by
dena u ing polyac ylamide-u ea DNA elec ophos esis (Fig. 7A). In line wi h he
epo ed s imula ion o APE1 incision by NPM1 (Vasco o e al., 2009), bu obse ing a
g ea e e ec han in hei epo , we ha e ound a highe endonuclease ac i i y in he
p esence o NPM1 (Fig. 7A).
17
In o de o in es iga e he molecula basis o his ac i a ion, we ha e explo ed he e ec
o NPM1 on APE1 binding o he abasic DNA subs a e. I had been p e iously
desc ibed (Mahe & Bloom, 2007; F euden hal e al., 2015) ha APE1 binds wi h high
a ini y abasic DNA and also binds, albei wi h lowe a ini y, he p oduc , nicked DNA,
which has been ela ed o a physiologically ad an ageous slow elease o cy o oxic
incised BER in e media es du ing DNA damage p ocessing (Mahe & Bloom, 2007).
We ha e measu ed he binding by moni o ing he inc ease in luo escence aniso opy o
luo escein-labelled DNA upon addi ion o inc easing amoun s o APE1. The a ini y
we ha e de e mined o APE1 binding o he oligo Dumbbell S (KD 39 ± 8 nM (Fig.
7B,C) is simila o p e iously epo ed da a wi h abasic-mimicking oligonucleo ides
(Hadi e al., 2000), bu lowe han ha de e mined by F euden hal e al. (2015) wi h he
same DNA sequence ha we ha e used. This could be explained because hey use a
bu e o lowe ionic s eng h and/o we could no ge eliable eadings a a DNA
concen a ion su icien ly below he KD. When he i a ion is pe o med in he p esence
o an excess o NPM1, no addi ional inc ease in he aniso opy is ob ained, sugges ing
ha NPM1 egula o y e ec is no media ed by o ma ion o a e na y complex DNA /
APE1 / NPM1 (Fig. 7B). On he con a y, a high APE1 concen a ion, he maximum
aniso opy alue a ained is lowe in he p esence o NPM1, sugges ing ha NPM1 can
compe e wi h he low-a ini y, o - a ge binding o APE1 o DNA. Howe e , APE1
a ini y o he speci ic si e on DNA (supposedly he abasic si e-mimicking THF) is
sligh ly enhanced in he p esence o NPM1 (Fig. 7C), ende ing a KD alue o 25.3 ± 8
nM, as compa ed o 39 ± 8 nM in he absence o NPM1 (p alue < 0.05).
We ha e also explo ed he o ma ion o complexes in e na y mix u es o APE1, DNA
and NPM1 h ough na i e gel elec opho esis (Fig. 8). Wi h equimola concen a ions o
APE1 and he “p oduc ” abasic DNA (Dumbbell P), a de ined band, p obably
co esponding o a 1:1 complex can be de ec ed (Fig. 8A), while a highe APE1:DNA
a ios (4:1) an he e ogeneous species, pu a i ely con aining mo e molecules o APE1
bound o o - a ge si es is ob ained (Fig. 8B). The addi ion o NPM1 pa ly compe es
wi h he he e ogeneous, o - a ge binding, bu no wi h he speci ic complex (Fig.
8A,B). On he con a y, when he di e en bands a e analyzed by densi ome y, i can
be concluded ha NPM1 a ou s he o ma ion o he APE1:DNA (1:1) complex, ei he
i i s quan i y is ollowed based on Coomassie s aining (p o ein amoun ) o GelRed
(DNA). Collec i elly, hese esul s sugges ha NPM1 may be able o a ou he
o ma ion o a speci ic, disc e e APE1:abasic DNA complex, opposing o al e na i e,
18
o - a ge APE1:DNA in e ac ions. Appa en ly, his selec ion o he speci ic,
p oduc i e binding esul s in a highe a ini y o abasic DNA (Fig. 7C) and, ul ima ely,
in he s imula ion o APE1 ac i i y (Fig. 7A).
3. Discussion
Ou esul s indica e ha , unde ce ain oxida i e s ess condi ions, i.e. KB O3 ea men ,
NPM1 is pa ly eleased om nucleoli and emains in ch oma in egions (p obably
euch oma in), which migh co espond o DNA damage epai pla o ms.
Nucleoplasmic ansloca ion o NPM1 upon a numbe o di e en s esses, including
oxida i e damage, has been p e iously epo ed (Colombo e al., 2011; Yang e al.,
2016; Sekha e al., 2014) and seems a common heme in he esponse o cellula
pe u ba ions o his and o he nucleola p o eins (An oniali e al., 2014; Sco &
Oe inge 2016). Mo eo e , NPM1 localiza ion dynamics has been p oposed o depend
on pos ansla ional modi ica ions ( he p o ein is amenable o egula ion by
phospho yla ion, ubiqui yla ion, sumoyla ion, ace yla ion, PARyla ion, e c). In his
sense, we ha e obse ed ha he oxida i e ea men induces phospho yla ion o NPM1
a leas on T199. This modi ica ion pe haps educes he a ini y o he p o ein o G-
quad uplex o ming DNA sequences in he nucleolus, p omp ing nucleola elease upon
DNA damage. Thus, phospho yla ion o NPM1 would be ela ed no only wi h DSBs
epai (Koike e al., 2010), bu also wi h epai o oxida i e damage. Recen ly,
glu a hionyla ion o a Cys esidue in he C- e minal domain o NPM1 has been
desc ibed as a mechanism unde lying nucleola oxida ion s a e sensing and elease om
nucleolus (Yang e al., 2016). The mul iple pos ansla ional modi ica ions a ec ing
NPM1 migh impac in di e en ways i s localiza ion dynamics and DDR- ela ed
ac i i ies.
The exac ole o NPM1 in BER is so a unknown. The pa icipa ion o NPM1 in his
and se e al o he DDR pa hways migh be ela ed o he his one chape oning ac i i y o
he p o ein, since ch oma in emodelling mus be a equi emen in all epai p ocesses.
On he o he hand, he nucleola en ichmen in no mal condi ions o bo h NPM1 and
APE1, and he ac ha hey a e able o bind no only DNA bu also RNA, and indeed
display enzyma ic ac i i ies on RNA, a leas in i o, has inspi ed he no ion ha hey
migh collabo a e in RNA quali y con ol ac i i ies in addi ion o DDR (An oniali e al.,
2014). Mo eo e , ou da a poin o a di ec ole o NPM1 in he BER machine y. The
19
simila i y be ween he beha iou o NPM1 and APE1 when he cells a e subjec ed o
oxida ion (dependence on KB O3 concen a ion, localiza ion wi hin ce ain nuclea
egions, and esis ance o de e gen - solubiliza ion) s ongly sugges s ha NPM1 could
be ec ui ed o he same s uc u es on ch oma in as APE1 (Campalans e al., 2013), i.e.,
pla o ms whe e ac i e BER is aking place.
I NPM1 is in ol ed in BER egula ion, he in e ac ion be ween NPM1 and APE1 may
be ele an as pha macological a ge (Pole o e al., 2015), since in e e ing wi h DNA
epai ep esen s a cance he apy s a egy. An in i o in e ac ion be ween NPM1 and
APE1 has been desc ibed implying he N- e minal, co e domain o NPM1, and he
lexible, N- e minal egion o APE1 (Vasco o e al., 2009; Pole o e al., 2013). In ou
u he cha ac e iza ion o he NPM1/APE1 associa ion, we demons a e a speci ic
binding whe e pa o he linke egion o NPM1 (bu no he egion comp ising he 106
las esidues, including he C- e minal domain) is in ol ed, explaining he abili y o he
AML leukaemia- ela ed NPM1 mu an o also bind APE1 and appa en ly seques e i
when abe an ly loca ing in he cy oplasm o AML cells (Vasco o e al., 2014, and ou
unpublished obse a ions). In addi ion, we ha e ound ha mo e han one molecules o
APE1 can bind o one NPM1 pen ame , and ha he in e ac ion is en opically d i en.
Al hough un o una ely we could no ob ain e idences o an in e ac ion be ween he
wo o e exp essed p o eins in cells, using he new “F2H” echnology ( om
Ch omoTek) (unpublished esul s), an in i o associa ion has been epo ed based on
p oximi y liga ion assay (PLA) (Pole o e al., 2013). I emains o be seen whe he
phospho yla ion o NPM1 a Th 199, which acco ding o ou da a co ela es wi h
nucleola elease, o u he pos ansla ional modi ica ions o ei he NPM1 o APE1
could modula e / enhance he associa ion.
NPM1 s imula es APE1 clea age o abasic DNA (Fig. 7A, Vasco o e al., 2009), while
in e es ingly, i has been desc ibed o inhibi APE1 endonuclease ac i i y on RNA
(Vasco o e al., 2009). The in i o NPM1 egula o y e ec s, along wi h he simila i y
be ween he localiza ion dynamics beha iou o bo h p o eins (Fig. 1) sugges ha
NPM1 migh pa icipa e in epai asks on BER pa ches. Tell and co-wo ke s ha e
p oposed ha ace yla ion o Lys esidues in APE1 would inhibi he associa ion o
NPM1 unde s ess condi ions. Ne e heless, he same au ho s ha e epo ed “p oximi y
liga ion assay” (PLA) signal also ou side he nucleoli; hus we hink ha an in e ac ion
be ween bo h p o eins (e en i he binding mode migh di e ) wi hin he BER
20
machine y is easible and could imply a modula ion o APE1 unc ionali y. The
molecula basis o such egula ion emains ne e heless elusi e.
We ha e ound ha NPM1 can compe e wi h low-a ini y, o - a ge APE1 binding o
DNA, and could selec o he speci ic, p oduc i e APE1/DNA complex. NPM1
in e ac s wi h he N- e minal lexible egion o APE1 (Vasco o e al., 2009, and his
s udy). This segmen , ca. 40 esidues long, has been ela ed o egula ion o DNA and
/o RNA binding (Fan ini e al., 2010; Pole o e al., 2013), and could be necessa y o
APE1 scanning o DNA o sea ch a ge lesions, as diso de ed ails o o he DNA-
binding p o eins (Vuzman e al., 2000). NPM1, by binding o APE1 N- e minal
segmen , could main ain APE1 in an open con o ma ion, eady o p oduc i e
associa ion o abasic si es. This migh be pe haps p omo ed unde cellula condi ions o
hea y damage. NPM1 has been also p oposed o egula e NF-B ac i i y on DNA by
keeping his p o ein in an open con o ma ion (Lin e al., 2017). The modula ion exe ed
by NPM1 esembles he ole o o he chape ones, in he sense ha hei ac i i y is only
e iden in i o a high chape one: a ge p o ein a ios (Sha ma e al., 2010), as i is he
case in ou expe imen s, whe e he e is a signi ican mola excess o NPM1.
We ha e shown ha , upon binding o an abasic p oduc -mimicking DNA, APE1 su e s
con o ma ional changes and becomes signi ican ly s abilized, hus p obably mo e
compac . In line wi h his no ion, i had been desc ibed, based on mass spec ome ic
p o ein oo p in ing, ha binding o DNA a e incision esul s in p o ec ion o some
APE1 Lys esidues (Yu e al., 2010). Acco ding o ha s udy, APE1 con o ma ion is
s ill mo e “ ensioned” when bound o he subs a e, and hen pa ly “ elaxes” once he
incision o he DNA has aken place, pe haps in o de o acili a e discha ge om DNA.
Al hough in ou expe imen s we could no demons a e NPM1 compe i ion wi h APE1
binding o he abasic p oduc , NPM1, by g asping he N- e minal egion o APE1 and
ac ing on i as a le e , migh somehow a ou he elease o APE1 om DNA once
incised. This migh be acili a ed, in i o, by subsequen ly in e ening enzymes o he
BER pa hway, so ha he clea ed in e media e is no eleased un il i can be p ope ly
handled (Almeida and Sobol, 2007; Pa sons and Diano , 2013; F euden hal e al.,
2015). Then APE1 would be eady o dismoun om DNA and scan ch oma in o
ano he abasic si e. O he p o eins pa icipa ing in he BER ou e (XRCC1, Pol) ha e
also been p oposed o s imula e APE1 ac i i y by eleasing i om p oduc inhibi ion
(Vidal e al., 2001; Masuda e al., 1998).
21
4. Conclusions
Nucleophosmin (NPM1) in e ac s wi h he epai enzyme APE1 and is able o s imula e
in i o APE1 incision o abasic DNA, sugges ing ha i could ake pa in he BER
pa hway. Ou inding ha , upon oxida i e damage o cells, pa o nucleophosmin is
eleased om nucleoli (being co ela ed wi h phospho yla ion a Th 199) and ec ui ed
o insoluble pa ches on ch oma in, esembling APE1 beha io , u he suppo s his
hypo hesis. In his s udy, we ha e u he cha ac e ized he in e ac ion be ween NPM1
and APE1, showing ha pen ame ic NPM1 can bind se e al molecules o APE1 wi h
submic omola a ini y. No only NPM1 co e domain and APE1 N- e minal egion, as
p e iously p oposed, bu also he p oximal pa o NPM1 linke egion and he globula
domain o APE1 a e equi ed o p ope binding. The ac ha he NPM1 C- e minal
106 esidues a e dispensable o he ecogni ion explains why he AML leukemia-
associa ed NPM1 mu an keeps he abili y o bind APE1, which migh lead o APE1
mislocaliza ion and pa hogenic de egula ion. We ha e also ound ha NPM1 is able o
compe e wi h APE1 o - a ge binding o DNA while a ou ing he speci ic, p oduc i e
binding o he abasic si e. This s imula o y e ec , as well as a pu a i e induc ion o
APE1 elease om he incised DNA, would posi i ely egula e APE1 epai ac i i y. In
summa y, ou esul s depic a new, ope a ional ole o NPM1 in he modula ion o BER
machine y.
Funding
This wo k was unded by he Spanish Minis y o Economy and FEDER (g an
SAF2014-57743-R), he Basque Go e nmen (g an IT709-13) and he Uni e si y o he
Basque Coun y (g an GIU18/172)
Au ho ’s con ibu ion
SB designed he esea ch. All o he au ho s pe o med he expe imen s. DJL, AdB,
IdlA, JM and SB analyzed and in e p e ed he da a. S.B. w o e he pape wi h he
con ibu ion o DJL and MAU.
Con lic o in e es
22
The au ho s decla e ha hey ha e no compe ing in e es s.
Acknowledgemen s
The au ho s hank he s a om he High Resolu ion Mic oscopy Facili y (SGIke -
UPV/EHU) o echnical suppo , D . José Ma ínez (Uni e si y o G anada) o ad ice
on in e p e a ion o ITC da a, and D . José .An onio. Rod íguez (Uni e si y o he
Basque Coun y) o sugges ions and ad ice. A.dB. and M.G. we e ecipien s o an
Ikasike ellowship (Basque Go e nmen ).
23
Re e ences
Adachi H, Yu YT (2014) Pu i ica ion o adiolabeled RNA p oduc s using dena u ing
gel elec opho esis. Cu P o oc Mol Biol 105, doi: 10.1002/0471142727.mb0420s105.
Almeida KH, Sobol RW. (2007) A uni ied iew o base excision epai : Lesion-
dependen p o ein complexes egula ed by pos - ansla ional modi ica ion. DNA Repai
6, 695-711, doi:10.1016/j.dna ep.2007.01.009
An oniali G, Li ussi L, Pole o M, Tell G. (2014) Eme ging oles o he nucleolus in
egula ing he DNA damage esponse: he noncanonical DNA epai enzyme APE1/Re -
1 as a pa adigma ical example. An ioxid Redox Signal. 2014 20(4):621-639, doi:
10.1089/a s.2013.5491
A egi I, Falces J, Olazabal-He e o A, Alonso-Ma iño M, Tane a SG, Rod íguez JA,
U baneja MA, Bañuelos S. (2015) Leukemia-Associa ed Mu a ions in Nucleophosmin
Al e Recogni ion by CRM1: Molecula Basis o Abe an T anspo . PLoS One.: 10(6),
doi: 10.1371/jou nal.pone.0130610.
Ballmaie D and Epe B (2006) DNA damage by b oma e: mechanism and
consequences. Toxicology 221, 166-171, doi: 10.1016/j. ox.2006.01.009.
Bañuelos S, Lec ez B, Tane a SG, O maza G, Alonso-Ma iño M, Calle X, U baneja
MA. (2013) Recogni ion o in e molecula G-quad uplexes by ull leng h
nucleophosmin. E ec o a leukaemia-associa ed mu a ion. FEBS Le . 587, 2254-2259,
doi: 10.1016/j. ebsle .2013.05.055.
Bois e FM, an Koningsb uggen S, Na ascués J, Lamond AI (2007) The
mul i unc ional nucleolus. Na Re Mol Cell Biol 8(7), 574-585, doi:
10.1083/jcb.200702147.
Box JK, Paque N, Adams MN, Bouche D, Bolde son E, O'By ne KJ, Richa d DJ.
(2016) Nucleophosmin: om s uc u e and unc ion o disease de elopmen . BMC Mol
Biol 17(1):19, doi: 10.1186/s12867-016-0073-9.
Campalans A, Ko ulewski T, Amou oux R, Menoni H, Ve meulen W, Radicella JP.
(2013) Dis inc spa io empo al pa e ns and PARP dependence o XRCC1 ec ui men
o single-s and b eak and base excision epai . Nucleic Acids Res 41, 3115–3129, doi:
24
10.1093/na /gk 025.
Chan PK, Chan FY. (1995) Nucleophosmin/B23 (NPM) oligome is a majo and s able
en i y in HeLacells. Biochim Biophys Ac a 1262, 37-42.
Chia ella S, De Cola A, Scaglione GL, Ca le i E, G aziano V, Ba ca oli D, Lo S e zo
C, Di Ma eo A, Di Ilio C, Falini B, A co i o A, De Lau enzi V, Fede ici L. (2013)
Nucleophosmin mu a ions al e i s nucleola localiza ion by impai ing G-quad uplex
binding a ibosomal DNA. Nucleic Acids Res. 41(5):3228-3239, doi:
10.1093/na /gk 001.
Colombo E, Ma ine JC, Dano i D, Falini B, Pelicci PG (2002) Nucleophosmin
egula es he s abili y and ansc ip ional ac i i y o p53. Na Cell Biol 4, 529–533, doi:
10.1038/ncb814.
Colombo E, Bone i, P, Denchi EL, Ma inelli P, Zamponi R e al. (2005)
Nucleophosmin is equi ed o DNA in eg i y and p19A p o ein s abili y. Mol Cell
Biol 25, 8874–8886, doi: 10.1128/MCB.25.20.8874-8886.2005.
Colombo, E., Alcalay, M. & Pelicci, P. G. (2011) Nucleophosmin and i s complex
ne wo k: a possible he apeu ic a ge in hema ological diseases. Oncogene 30, 2595-
2609, doi: 10.1038/onc.2010.646.
Emmo E and Hiscox JA (2009) Nucleola a ge ing: he hub o he ma e . EMBO Rep
10(3), 231-238, doi: 10.1038/embo .2009.
Falini B, Mecucci C, Tiacci E, Alcalay M, Rosa i R e al. (2005) Cy oplasmic
nucleophosmin in acu e myelogenous leukemia wi h a no mal ka yo ype. N Engl J Med
352: 254-266, doi: 10.1056/NEJMoa041974.
Fan ini D, Vasco o C, Ma asco D, D'Amb osio C, Romanello M, Vi agliano L, Pedone
C, Pole o M, Cesa a o L, Quad i oglio F, Scaloni A, Radicella JP, Tell G. (2010)
C i ical lysine esidues wi hin he o e looked N- e minal domain o human APE1
egula e i s biological unc ions. Nucleic Acids Res. 38(22):8239-56, doi:
10.1007/s00018-010-0486-4.
Fede ici L, Falini B (2013) Nucleophosmin mu a ions in acu e myeloid leukemia: a ale
o p o ein mis olding and mislocaliza ion. P o ein Sci 22: 545-556, doi:
25
10.1002/p o.2240.
F euden hal BD, Bea d WA, Cuneo MJ, Dy khee a NS, Wilson SH. (2015) Cap u ing
snapsho s o APE1 p ocessing DNA damage. Na S uc Mol Biol 22: 924-931, doi:
10.1038/nsmb.3105.
Gas eige E, Hoogland C, Ga ike A, Du aud S, Wilkins M.R, Appel R.D, Bai och A.
(2005) P o ein Iden i ica ion and Analysis Tools on he ExPASy Se e . In:
John M. Walke (ed): The P o eomics P o ocols Handbook, Humana P ess, 571-607.
G isendi S, Mecucci C, Falini B, Pandol i PP. (2006) Nucleophosmin and cance . Na .
Re . Cance 6, 493-505, doi: 10.1038/n c1885.
G ummi CG, Townsley FM, Johnson CM, Wa en AJ, Byc o M (2008) S uc u al
consequences o nucleophosmin mu a ions in acu e myeloid leukemia. J. Biol. Chem.
283, 23326-23332, doi: 10.1074/jbc.M801706200.
Guillonneau M, Pa is F, Du oi S, Es ephan H, Béné eau E, Huo J, Co e I (2016)
Oxida i e s ess disassembles he p38/NPM/PP2A complex, which leads o modula ion
o nucleophosmin-media ed signaling o DNA damage esponse. FASEB J. 30(8):2899-
2914, doi: 10.1096/ j.201500194R.
Hadi MZ, Coleman MA, Fidelis K, Moh enweise HW and Wilson 3 d DM (2000)
Func ional cha ac e iza ion o Ape1 a ian s iden i ied in he human popula ion.
Nucleic Acids Res. 28, 3871-3879, doi: 10.1093/na /28.20.3871.
He H, Chen Q, Geo giadis MM. (2014) High- esolu ion c ys al s uc u es e eal
plas ici y in he me al binding si e o apu inic/apu ymidinic endonuclease I
Biochemis y 53(41), 6520-6529, doi: 10.1021/bi500676p.
Jackson EB, The io CA, Cha opadhyay R, Mi a S, Izumi T. (2005) Analysis o
nuclea anspo signals in he human apu inic/apy imidinic endonuclease (APE1/Re 1)
Nucleic Acids Res. 33(10), 3303-3312, doi: 10.1093/na /gki641.
Kennedy MJ, Hughes RM, Pe eya LA, Schwa z JW, Ehle s MD, Tucke CL. (2010)
Rapid blue-ligh -media ed induc ion o p o ein in e ac ions in li ing cells. Na Me hods
7(12), 973-975, doi: 10.1038/nme h.1524.
Koike A, Nishikawa H, Wu W, Okada Y, Venki a aman AR e al. (2010) Rec ui men o
32
Figu e 5. Compa ison o APE1 binding by wild ype NPM1, NPM1Mu A,
NPM1ΔC106 and NPM1 co e. (A) Binding cu es based on na i e elec opho esis
assays wi h 0.5 M (NPM1co e) o 0.1 M pen ame ( o he es o NPM1 a ian s).
The co esponding i ings o a simple binding model a e also shown, wi h indica ion o
he es ima ed KD alues. (B) Compa ison o NPM1 binding by wild ype APE1 and
APE1N33.
Figu e 6. (A) Binding o magnesium and/o abasic DNA con o ma ionally
s abilizes APE1. The mal scans based on ellip ici y a 222 nm o 4 M apo APE1 (0.5
mM EDTA) (black), holo APE1 (5 mM MgCl2) ( ed), apo APE1 in he p esence o
equimola amoun o oligonuleo ide “Dumbbell P” (da k blue) and holo APE1 plus
DNA (g een). Scans we e co ec ed o he DNA con ibu ion. The hea ing a e was
1ºC/min. (B) Seconda y s uc u e o APE1 and APE1ΔN33 is sensi i e o DNA
binding. Amide I band o he in a ed spec a o APE1 (solid line) and APE1N33
(b oken line) in he absence (blue) and p esence ( ed) o equimola amoun s o he oligo
“Dumbbell P”. The spec a we e eco ded in D2O bu e a 20ºC. The con ibu ion o
bu e was subs ac ed and he baseline was co ec ed be ween 1700 and 1600 cm-1.
Figu e 7. NPM1 s imula es APE1 ca alyzed incision o abasic DNA, and a o s
speci ic binding o APE1 o he subs a e. (A) Incision o an abasic- mimicking
oligonucleo ide (5 M) by inc easing concen a ions o APE1 as ollowed by DNA
u ea-polyac ylamide dena u ing elec opho esis, in he absence (emp y symbols) and in
he p esence (solid symbols) o 5 M NPM1. (B) Binding o 0 – 10 M APE1 o 10 nM
Dumbbell S in he absence (emp y) o p esence (solid) o 5 M NPM1, based on
luo escence aniso opy. (C) The same binding da a o panel B shown in he ange 0-1
M APE1 and loga i hmic scale and wi h he co esponding i ed cu es (p- alue <
0.05). Da a a e a e ages o h ee expe imen s, ba s indica ing s anda d de ia ion.
Figu e 8. E ec o NPM1 on APE1/DNA complex o ma ion as ollowed by na i e
elec opho esis. (A) Na i e 4-16% PA gel o mix u es o 5 M oligo (O), 2 M APE1
(A) and 0-10 M NPM1 (N) (pen ame ), along wi h he co esponding indi idual
species and bina y mix u es as con ols. The lanes co esponding o he e na y mix u es
a e labelled acco ding o NPM1 concen a ion. Images o he gel s ained wi h
Coomassie o de ec ion o p o eins (blue) and GelRed o DNA (whi e) a e o e laid.
(B) Na i e PAGE o a mix u e o 5 M oligo and 20 M APE1 in he absence (OA) and
33
in he p esence (OAN) o 1 M NPM1. The gel was s ained as in (A). (C) Densi ome y
o he band co esponding o he APE1:DNA (1:1) complex s ained wi h Coomassie
(squa es) o GelRed ( iangles) as a unc ion o NPM1 concen a ion.
34
Fig. 1
35
Fig. 2
36
Fig. 3
37
Fig. 4.
38
Fig. 5.
39
Fig. 6
40
Fig. 7
A
🞅 - NPM1
B
🞅 - NPM1
C
🞅 - NPM1
41
Fig. 8
