HERC2 deficiency activates C-RAF/MKK3/p38 signalling pathway altering the cellular response to oxidative stress

Vol.:(0123456789)
1 3
Cellula and Molecula Li e Sciences (2022) 79:548
h ps://doi.o g/10.1007/s00018-022-04586-7
ORIGINAL ARTICLE
HERC2 de iciency ac i a es C‑RAF/MKK3/p38 signalling pa hway
al e ing hecellula esponse ooxida i e s ess
JoanSala‑Gas on1 · Leona doPed azza1· JuanmaRami ez2· A u oMa inez‑Ma inez1· Le ieE.Rawlins3,4·
EmmaL.Baple3,4· And ewH.C osby3· UgoMayo 2,5· F ancescVen u a1· JoseLuisRosa1
Recei ed: 27 Ap il 2022 / Re ised: 3 Oc obe 2022 / Accep ed: 3 Oc obe 2022 / Published online: 14 Oc obe 2022
© The Au ho (s) 2022
Abs ac
HERC2 gene encodes an E3 ubiqui in ligase in ol ed in se e al cellula p ocesses by egula ing he ubiqui yla ion o di e en
p o ein subs a es. Biallelic pa hogenic sequence a ian s in he HERC2 gene a e associa ed wi h HERC2 Angelman-like
synd ome. In pa hogenic HERC2 a ian s, comple e absence o ma ked educ ion in HERC2 p o ein le els a e obse ed.
The mos common pa hological a ian , c.1781C > T (p.P o594Leu), encodes an uns able HERC2 p o ein. A be e unde -
s anding o how pa hologic HERC2 a ian s a ec in acellula signalling may aid de ini ion o po en ial new he apies
o hese diso de s. Fo his pu pose, we s udied pa ien -de i ed cells wi h he HERC2 P o594Leu a ian . We obse ed
al e a ion o mi ogen-ac i a ed p o ein kinase signalling pa hways, e lec ed by inc eased le els o C-RAF p o ein and p38
phospho yla ion. HERC2 knockdown expe imen s ep oduced he same e ec s in o he human and mouse cells. Mo eo e ,
we demons a ed ha HERC2 and RAF p o eins o m molecula complexes, pull-down and p o eomic expe imen s showed
ha HERC2 egula es C-RAF ubiqui yla ion and we ound ou ha he p38 ac i a ion due o HERC2 deple ion occu s in a
RAF/MKK3-dependen manne . The displayed cellula esponse was ha pa ien -de i ed and o he human cells wi h HERC2
de iciency showed highe esis ance o oxida i e s ess wi h an inc ease in he mas e egula o o he an ioxidan esponse
NRF2 and i s a ge genes. This esis ance was independen o p53 and abolished by RAF o p38 inhibi o s. Al oge he , hese
indings iden i y he ac i a ion o C-RAF/MKK3/p38 signalling pa hway in HERC2 Angelman-like synd ome and highligh
he inhibi ion o RAF ac i i y as a po en ial he apeu ic op ion o indi iduals a ec ed wi h hese a e diseases.
Keywo ds Neu ode elopmen al diso de · Angelman· Ubiqui in· MAPK· Cell s ess
In oduc ion
He edi a y neu ode elopmen al diso de s a ise om al e a-
ions in cen al ne ous sys em de elopmen and mani es
pe ina ally o du ing in ancy and childhood. Despi e show-
ing wide gene ic and clinical he e ogenei y, mos sha e some
common pheno ypic ea u es, such as de elopmen al delay,
impai ed mo o unc ion and in ellec ual disabili y. The
iden i ica ion o genes esponsible o hese diso de s has
enabled gene ic diagnosis, accu a e gene ic counselling, and
be e managemen [1].
The HECT and RCC1-like domain 2 (HERC2) gene
encodes an unusually la ge p o ein wi h 4834 amino acid
esidues. The HERC2 p o ein is an E3 ubiqui in ligase ha
unc ions in ubiqui yla ion by accep ing ubiqui in om
ubiqui in-conjuga ing enzymes (E2) and ans e ing i o
a a ge p o ein [2]. Ubiqui yla ion a ec s p o eins in many
ways, a iously ma king hem o p o easome deg ada ion
Cellula andMolecula Li e Sciences
* Jose Luis Rosa
[email p o ec ed]
1 Depa men o Physiological Sciences, Bell i ge Biomedical
Resea ch Ins i u e (IDIBELL), Uni e si y o Ba celona
(UB), C/ Feixa Lla ga s/n, 08907L’Hospi ale deLlob ega ,
Spain
2 Depa men o Biochemis y andMolecula Biology, Facul y
o Science andTechnology, UPV/EHU, Leioa, Bizkaia,
Spain
3 RILD Wellcome Wol son Medical Resea ch Cen e, RD&E
(Won o d) NHS Founda ion T us , Uni e si y o Exe e
Medical School, Exe e , UK
4 Peninsula Clinical Gene ics Se ice, Royal De on & Exe e
Hospi al (Hea i ee), Exe e , UK
5 Ike basque, Basque Founda ion o Science, 48013Bilbao,
Spain
J.Sala-Gas on e al.
1 3
548 Page 2 o 20
o , a ec ing hei ac i i y, localisa ion o in e ac ions wi h
o he p o eins. The e o e, ubiqui in ligases a e key egula o s
o many cellula p ocesses, wi h hei dys egula ion being
common in nume ous cance s and neu odegene a i e dis-
eases [3]. Fo example, HERC2 mu a ions a e associa ed
wi h b eas , skin (melanoma), gas ic, colo ec al, and hae-
ma ological (leukaemia) cance s [4]. The unde lying molec-
ula mechanism could be ha HERC2 egula es BRCA1,
XPA, USP20 o RPA2 p o ein ubiqui yla ion, in ol ed in
egula ing DNA epai and genomic s abili y [5–9]. HERC2
also egula es p53 ansc ip ional p og am by p omo ing p53
e ame isa ion and subsequen ac i a ion, independen o i s
ubiqui in ligase ac i i y [10–12].
Besides, HERC2 is essen ial du ing emb yonic de elop-
men and plays an impo an ole in egula ing mo o coo di-
na ion [13]. Mo eo e , i is highly exp essed in he ne ous
sys em and has been linked wi h he edi a y neu odegen-
e a i e diso de s [14]. Biallelic HERC2 a ian s associa ed
wi h HERC2 Angelman-like synd ome include missense
and ameshi mu a ions wi h a p ema u e s op codon ha
esul in a loss o unc ion. These cases a e associa ed wi h
a comple e loss o ma kedly educed le els o HERC2 p o-
ein [15–19]. The condi ion was i s desc ibed in Amish/
Mennoni e communi ies, associa ed wi h homozygosi y o
a HERC2 (c.1781C > T, p.P o594Leu) ounde gene a ian
a inc eased equency in he popula ion (au osomal eces-
si e men al e a da ion ype 38; OMIM # 615516) [15, 16].
P o eomic s udies o pe iphe al blood-de i ed lymphoblas s
om indi iduals wi h his condi ion sugges de angemen s
o mul iple cellula pa hways p obably in ol ing dispa a e
pa hogenic mechanisms [20]. Despi e hese e o s, he
molecula mechanisms unde lying HERC2- ela ed diso de s
emain elusi e, impeding e o s o ind po en ial ea men s
o hese a e diseases. Fu he in es iga ion o hei molecu-
la basis could e eal no only he unde lying pa hology bu
also po en ial he apeu ic a ge s.
In his s udy, we analysed in acellula signalling pa h-
ways in skin ib oblas s om indi iduals wi h he pa ho-
logical a ian HERC2 P o594Leu (HERC2 P594L). They
displayed al e ed mi ogen-ac i a ed p o ein kinase (MAPK)
signalling ha a ec ed he oxida i e s ess esponse, wi h
inc eases in C-RAF p o ein le els and MAPK p38 ac i a-
ion. These e ec s we e ep oduced in o he human and
mouse cells wi h HERC2 p o ein knockdown. Fu he mo e,
we showed ha HERC2 egula es C-RAF ubiqui yla ion and
ha HERC2 de iciency igge s MKK3/p38 pa hway ac i-
a ion in a RAF-dependen manne . In line wi h his, cells
wi h he HERC2 P594L a ian had inc eased esis ance o
H2O2-induced oxida i e s ess, dependen on he ac i i ies
o RAF and p38. Finally, we discuss bo h he implica ions
o hese indings o neu ode elopmen al diso de s caused
by HERC2 a ian s and he po en ial he apeu ic use o RAF
inhibi o s.
Ma e ials andme hods
Human cell sample, cell lines andcul u e condi ions
Samples o human skin ib oblas s we e ob ained wi h
app o ed in o med consen as p e iously desc ibed else-
whe e [16].
U2OS, HEK 293T, H1299, RAW 264.7, mouse emb y-
onic ib oblas s (MEFs) and human skin ib oblas s we e cul-
u ed in Dulbecco’s modi ied Eagle’s medium supplemen ed
wi h 10% e al bo ine se um, 2mM l-glu amine, 100 U/
mL penicillin, and 0.1mg/mL s ep omycin sulpha e. Mouse
p ima y os eoblas s we e cul u ed in Minimum Essen ial
Medium α wi h 10% FBS, 2mM l-glu amine, 1mM py u-
a e, 100 U/ml penicillin, and 0.1mg/ml s ep omycin wi h
50μg/ml asco bic acid and 4mM β-glyce ophospha e. All
cells we e main ained in a humidi ied incuba o a 37°C and
5% CO2 a mosphe e.
Cell ea men s andinduc ion o cellula s ess
Cells we e ea ed wi h one o h ee inhibi o s, as indica ed:
1µM LY3009120 (Selleckchem), 1µM So a enib (San a
C uz Bio echnology) o 10µM SB203580 (Selleckchem).
Di e en cellula s ess ypes we e induced using di e en
s esso s: oxida i e s ess by 500µM o 50µM hyd ogen
pe oxide (H2O2) (Pan eac), depending on he expe imen ;
saline s ess by 100mM NaCl.
Plasmid andsiRNAs ans ec ions
Plasmid ans ec ion was pe o med using he Lipo-
ec amine LTX me hod (15338; In i ogen, Ca lsbad,
CA, USA), acco ding o he manu ac u e ’s ins uc ions.
Myc- agged agmen s o HERC2 (F1, F2, F3, F4, F5 and
F5CT) we e kindly p o ided by D . Oh a [21]. G een luo-
escen p o ein (GFP) and C-RAF usion cons uc s (CR1,
CR2, CR3 and ull-leng h) we e gene a ed, sub-cloned
and es ed elsewhe e [22]. Plasmids exp essing HERC2
ull-leng h p o ein pcDNA5 FRT/TO SF-HERC2 (ShB-R)
(Addgene plasmid # 55613; h p:// n2 . ne / addge ne: 55613;
RRID:Addgene_55613) and pcDNA5 FRT/TO SF-HERC2
C4762S (ShB-R) (Addgene plasmid # 55614; h p:// n2 . ne /
addge ne: 55614; RRID:Addgene_55614) we e a gi om
Da id Chan [23]. His-Ubiqui in cons uc s we e kindly p o-
ided by D . E azo [24]. The plasmid exp essing a bio i-
nyla able e sion o ubiqui in had been p e iously desc ibed
elsewhe e [25].
Fo gene in e e ence, siRNAs we e ans ec ed using
he calcium phospha e me hod desc ibed elsewhe e [10].
Cus om double-s anded siRNA oligonucleo ides we e
HERC2 de iciency ac i a es C‑RAF/MKK3/p38 signalling pa hway al e ing hecellula esponse…
1 3
Page 3 o 20 548
ob ained om GeneCus (Boynes, F ance). The o wa d
sequences we e as ollows: nega i e con ol (NC) = 5′-UUC
UCC GAA CGU GUC ACG UTT; HERC2 (H2.2) = GAC
UGU AGC CAG AUU GAA ATT; HERC2 (H2.4) = GGA
AAG CAC UGG AUU CGU UTT; HERC1 = CGG CAU
GGA UGA ACA AAU UTT; MKK3 = GGA AGA AGG
AUC UAC GGA UTT; C-RAF = UAG UUC AGC AGU UUG
GCU ATT; A-RAF = AAC AAC AUC UUC CUA CAU GAG
TT; B-RAF = AAA GAA UUG GAU CUG GAU CAU TT;
p53 = GAC UCC AGU GGU AAU CUA CTT.
Len i i al pa icle p oduc ion and a ge cell
in ec ion
Len i i al ec o s we e p oduced in HEK 293 T cells. Cells
we e ans ec ed wi h 7μg pMD2.G, 7μg psPAX2 (VSV-
G) and 7μg o ei he emp y pLKO.1-Pu o o pLKO.1‐
shHERC2 by he calcium phospha e me hod. Media con-
aining len i i al pa icles we e collec ed, il e ed using
poly inyl di luo ide il e s (Millex-HV il e 0.45μm,
Millipo e SLHV033RB) and s o ed in aliquo s a − 80°C.
Ta ge cells we e seeded a a con luence o 50–60% in a
6-well pla e be o e adding 300 μL o he medium con ain-
ing he len i i al ec o s o each well. F esh medium, sup-
plemen ed wi h 5μg/mL polyb ene, was added o make a
o al olume o 1mL. Media wi h len i i al ec o s we e
emo ed he nex day and a e 24h, 5μg/mL pu omycin
was added o selec ion. MISSION shRNA clone o mouse
HERC2 (TRCN0000039444) was pu chased om Sigma-
Ald ich. The plasmid ec o pLKO.1—TRC con ol was
a gi om Da id Roo (Addgene plasmid #10879; h p://
n2 . ne / addge ne: 10879; RRID:Addgene_10879) [26],
and he VSV-G en elope exp essing plasmid pMD2.G
(Addgene plasmid #12259; h p:// n2 . ne / addge ne: 12259;
RRID:Addgene_12259) and he len i i us packaging plas-
mid psPAX2 (Addgene plasmid #12260; h p:// n2 . ne / addg e
ne: 12260; RRID:Addgene_12260) we e a gi om Didie
T ono.
P o ein ex ac ion, PAGE, andimmunoblo ing
Fo p o ein ex ac ion, cells we e washed wice in ice-cold
phospha e-bu e saline a e disca ding he media. Cell
lysis was pe o med by sc apping a e adding o NP40 lysis
bu e (50mM T is–HCl, pH 7.5, 150mM NaCl, 50mM
NaF, 0.5% NP40) con aining p o ease and phospha ase
inhibi o s as p e iously desc ibed [27]. Lysa es we e main-
ained on ice unde agi a ion o 20min, and hen cen i-
uged a 13,000×g a 4°C o 10min. Supe na an s we e
collec ed be o e analysis using he T is–Ace a e PAGE sys-
em [28]. Band in ensi ies we e de ec ed using a gel docu-
men a ion sys em (LAS-3000, Fuji ilm) and quan i ied wi h
ImageJ so wa e (Rasband, W.S., ImageJ, U. S. Na ional
Ins i u es o Heal h, Be hesda, Ma yland, USA, h ps://
imagej. nih. go / ij/).
We used he ollowing an ibodies: an i-HERC2 mono-
clonal (BD Biosciences #612366); an i-C-RAF (BD Bio-
sciences #610151); an i-Cla h in Hea y Chain (TD.1)
(San a C uz Bio echnology #sc12734); an i-P-ERK1/2
(Sigma-Ald ich #M 8159); an i-p44/42 MAPK (ERK1/2)
(Cell signalling #9102); an i-phospho-p38 (Cell signalling
#9211); an i-p38 (San a C uz Bio echnology #sc-535); an i-
HERC1 (410) [10]; an i-P-MKK3 (Cell signalling #9231);
an i-MKK3 (P o ein ech #13898–1-AP); an i-A-RAF (A-5)
(San a C uz #sc-166771); an i-B-RAF (F-7) (San a C uz
Bio echnology #sc-5284); an i-HERC2 polyclonal (b g3)
[10]; an i-c-myc (clone 9E10) (Roche #1 667 149); an i-
GFP (Abcam #ab13970); an i-Flag M2 (Sigma-Ald ich #F
3165); an i-p-HSP27 (Enzo Li e Sciences #ADI-SPA-523);
an i-HSP27 (San a C uz Bio echnology #sc-1049); an i-
NRF2 (Cell signalling #12721); an i-ubiqui yla ed p o eins
(clone FK2; Biomol); and pe oxidase-conjuga ed seconda y
an ibodies (In i ogen).
Con ocal mic oscopy
We seeded U2OS cells on glass co e slips and pe o med
ixa ion by incuba ing cells a oom empe a u e o 20min
in 4% pa a o maldehyde. Then, cells we e pe meabilised
o 20min wi h 0.05% saponin in phospha e-bu e ed saline
con aining 0.5% bo ine se um albumin. The p ima y an i-
body, an i-phospho-p38 (Cell signalling #9211) (1:200), was
incuba ed a 37°C o 1h. A e washing, Alexa-Fluo 488
seconda y an ibody (In i ogen) (1:500) was incuba ed a
37°C o 45min. Ac in ilamen s we e s ained by incuba-
ion wi h phalloidin-Alexa 647 (BioP obes) (100ng/mL) o
20min a oom empe a u e. Nuclei we e s ained wi h DAPI
(Sigma-Ald ich) (1μg/mL). All images we e acqui ed using
a con ocal lase scanning mic oscope (LSM 880 spec al,
Ca l Zeiss Mic oscopy GmbH, Jena, Ge many).
Immunop ecipi a ion andpull‑downs
Fo immunop ecipi a ion, cells we e lysed wi h CHAPS
bu e (10mM T is–HCl, pH 7.5, 100mM NaCl, 0.3%
CHAPS) con aining p o ease and phospha ase inhibi o s as
desc ibed abo e. Cell lysa es (inpu ) we e incuba ed wi h
p e-immune se um o an i-HERC2 polyclonal an ibody
(b g3) o 2h a 4°C wi h gen le o a ion and immunop e-
cipi a ed wi h p o ein A-Sepha ose (GE Heal hca e) o 1h
a 4°C. Beads we e pelle ed by cen i uga ion a 2500×g,
washed i e imes wi h CHAPS bu e , and analysed by elec-
opho esis and immunoblo .
J.Sala-Gas on e al.
1 3
548 Page 4 o 20
Fo he GFP pull-downs, supe na an s we e incuba ed
wi h 2 μL GFP-T apA (Ch omoTek) o 2h a 4°C. Pelle s
we e washed i e imes wi h CHAPS bu e and analysed by
elec opho esis and immunoblo .
Fo ubiqui ome p o eomic expe imen s, bio in-pull-
downs we e pe o med in iplica es as p e iously desc ibed
[25], in o de o compa e p o eins mo e ubiqui ina ed in
Flag-HERC2 WT-o e exp essing cells, ela i e o Flag-
HERC2 C4762S-o e exp essing cells.
Ubiqui yla ion assay
HEK 293T cells we e ans ec ed wi h he indica ed plas-
mids o 48h. Be o e he ubiqui yla ion assay, he cells
we e ea ed o 4h wi h 10µM o he p o easome inhibi o
MG132 (Sigma-Ald ich/Me ck #C2211). Then, cells we e
lysed wi h dena u ing bu e #1 (6M guanidinium-HCl,
10mM T is, 100mM Na2HPO4–NaH2PO4 bu e , pH 8)
and he cells ex ac s we e incuba ed wi h he nickel beads
(Ni2+-NTA aga ose; Qiagen) o 2h a 4°C unde o a ion.
Beads we e successi ely washed as ollows: wice wi h 1ml
o dena u ing bu e #1 plus 10mM 2-me cap oe hanol;
h ee imes wi h 1ml o bu e #2 (8M u ea, 10mM T is,
10mM 2-me cap oe hanol, 100mM Na2HPO4–NaH2PO4
bu e , pH 8); wice wi h 1ml o bu e #3 (8M u ea, 10mM
T is, 100mM Na2HPO4–NaH2PO4 bu e , pH 6.3) con ain-
ing 0.2% T i on X-100; once wi h 1ml o bu e #3 con ain-
ing 0.1% T i on X-100 and 0.5M NaCl; and h ee imes wi h
1ml o bu e #3. Finally, p o eins we e elu ed by incuba -
ing he beads wi h 200mM imidazole in 5% SDS, 0.15M
T is–HCl, pH 6.7, 30% ( ol/ ol) glyce ol, 0.72M 2-me cap-
oe hanol o 1h a 37°C wi h mixing. The samples we e
analysed by immunoblo ing as indica ed abo e.
Re e se ansc ip ion andquan i a i e PCR
To al RNA was isola ed om U2OS cells using he TRIsu e
eagen acco ding o he manu ac u e ’s p o ocol (Bioline).
To al RNA (2μg) was e e se- ansc ibed using he high-
capaci y cDNA Re e se T ansc ip ion ki (Applied Bio-
sys ems). PCR ampli ica ion eac ions we e pe o med
wi h he ABI P ism 7900 HT Fas Real-Time PCR Sys em.
Applied Biosys ems’ TaqMan Gene Exp ession Assays
(The moFishe Scien i ic) we e used o quan i y he gene
exp essions o he ollowing: GUSB (Hs00939627_m1),
NFE2L2 (Hs00975960_m1), SOD1 (Hs00533490_m1),
SOD2 (Hs00167309_m1), GPX1 (Hs00829989_Gh), and
he housekeeping gene GAPDH (Hs99999905_m1), which
was used o no malise.
MTT assay o cell iabili y andcell p oli e a ion
Using 96-well pla es, U2OS cells and human skin ib oblas s
we e seeded o inal concen a ion o 10,000 cells/well o
15,000 cells/well, espec i ely. A e incuba ion a 37°C
o 24h in he cell incuba o , we ini ia ed ea men s, as
indica ed and pe o med he MTT assay (M5655; Sigma/
Me ck) acco ding o manu ac u e ’s ins uc ions. B ie ly,
we added MTT a a inal concen a ion o 0.5mg/mL o
each well, incuba ed he cells o 4h in a humidi ied incuba-
o , hen disca ded he media and solubilised he o mazan
c ys als wi h isop opanol. Finally, abso bance a a wa e-
leng h o 570nm was de e mined using a 96-well pla e
spec opho ome e .
Mi oSox s aining
To e alua e mi ochond ial eac i e oxygen species (ROS),
human skin ib oblas s we e seeded in a µ-Slide 8 well-
chambe ed co e slip a a concen a ion o 15,000 cells/well.
The nex day, cells we e s ained wi h 1µg/mL o Hoechs
33,342 (H3570, The moFishe , USA) o 30min a 37°C
and wi h 2µM Mi oSOX Red (In i ogen) o 15min a
37°C. Cells we e examined in a Zeiss LSM 880 lase scan-
ning con ocal spec al mic oscope equipped wi h an incuba-
ion con ol sys em (37°C, 5% CO2). Fluo escence in ensi y
pe cell was measu ed, quan i ied and exp essed as a bi a y
uni s (a.u). Images we e analysed using ImageJ so wa e
(Rasband, W.S., ImageJ, U.S. Na ional Ins i u es o Heal h,
Be hesda, Ma yland, USA, h ps:// imagej. nih. go / ij/).
Mi o acke s aining
Fo mi ochond ia s aining, human skin ib oblas s we e
seeded in a µ-Slide 8 well-chambe ed co e slip a a concen-
a ion o 15,000 cells/well. The nex day, cells we e s ained
wi h 1µg/mL o Hoechs 33,342 (H3570, The moFishe ,
USA) and 50nM Mi o acke Red CMXRos (M7512, The -
moFishe , USA) o 30min a 37°C. Images we e aken
using a Zeiss LSM 880 lase scanning con ocal spec al
mic oscope equipped wi h an incuba ion con ol sys em
(37°C, 5% CO2). F agmen ed mi ochond ial pe cen age was
calcula ed by coun ing sphe ical non-con iguous mi ochon-
d ial pa icles and di iding by he numbe o o al s uc u es
comp ised in he mi ochond ial ne wo k. Images we e ana-
lysed using ImageJ so wa e (Rasband, W.S., ImageJ, U.S.
Na ional Ins i u es o Heal h, Be hesda, Ma yland, USA,
h ps:// imagej. nih. go / ij/).
HERC2 de iciency ac i a es C‑RAF/MKK3/p38 signalling pa hway al e ing hecellula esponse…
1 3
Page 5 o 20 548
S a is ical analysis
The esul s indica e he means and s anda d e o o he
mean (± SEM) o , a leas , h ee independen expe imen s.
Indi idual da a poin s a e plo ed as single do s. Signi icance
was calcula ed by S uden - es and indica ed as ollows: *,
**, o *** o p alues o ≤ 0.05, ≤ 0.01, o ≤ 0.001, espec-
i ely. Figu es we e c ea ed, and s a is ical analysis was pe -
o med, using G aphPad P ism e sion 8.4.3 o Windows
(G aphPad So wa e, San Diego, Cali o nia USA), www.
g aph pad. com.
Resul s
Human HERC2 P o594Leu cells display MAPK
pa hway al e a ion
Se e al ecessi e mu a ions a ec ing he HERC2 gene cause
de elopmen al delay wi h Angelman-like ea u es [14, 19].
Knowing how pa hologic HERC2 a ian s a ec in acel-
lula signalling could e eal he unde lying pa hology and
iden i y possible he apies. The e o e, we s udied cells
om an indi idual wi h he mu an HERC2 P594L a i-
an desc ibed in mos cases. Since HERC1 had p e iously
been epo ed o egula e he ERK and p38 MAPK signal-
ling pa hways [22, 29], we wonde ed i HERC2 also had a
modula o y ole. As expec ed, cells wi h he HERC2 P594L
mu a ion showed almos unde ec able HERC2 p o ein le -
els (Fig.1A–C). In e es ingly, al hough hey showed highe
p o ein le els o C-RAF (Fig.1A), his did no co ela e
wi h he canonical ac i a ion o he ERK signalling pa hway,
assessed by ERK phospho yla ion (Fig.1B). An inc emen
in p38 phospho yla ion was also de ec ed while o al p38
p o ein le els emained s able (Fig.1C).
In o de o p o ide mo e e idence ha hese changes
in MAPK signalling pa hways a e a gene al hallma k o
disease in pa ien s wi h biallelic HERC2 mu a ions, we
analysed samples o wo mo e indi iduals ca ying he
mu an HERC2 P594L a ian . Consis en ly, pa ien s wi h
he HERC2 P594L mu a ion (P1, P2 and P3) showed lowe
HERC2 p o ein le els han he wild- ype con ols (C1, C2
and C3). In addi ion, C-RAF p o ein le els and p38 phos-
pho yla ion we e up egula ed in all h ee pa ien s, bu no
changes we e de ec ed in ERK ac i a ion (Fig.1D). These
esul s showed ha cells wi h he HERC2 P594L mu a ion
exhibi al e ed MAPK signalling pa hway ac i a ion, as
e lec ed by highe C-RAF and phospho-p38 p o ein le els.
HERC2 egula es C‑RAF p o ein le els
To del e deepe in o he molecula mechanisms in ol ed
in he al e ed MAPK signalling pa hway in HERC2 P594L
cells, we conside ed human cells wi h low le els o HERC2
p o ein sha ed his al e a ion. In knockdown expe imen s
pe o med in human U2OS cells, cells we e ans ec ed
wi h ei he a nega i e con ol (NC) small-in e e ing RNA
(siRNA), an siRNA agains HERC2, o a posi i e con ol
siRNA agains HERC1. The posi i e con ol was chosen
because p e ious wo k had shown ha HERC1 con ols
ERK and p38 signalling pa hways modula ing C-RAF
p o ein le els [22, 29]. HERC2 knockdown mimicked he
e ec obse ed in HERC2 P594L cells, wi h deple ion o
HERC2 co ela ing wi h inc eased C-RAF p o ein le els.
As expec ed, his was also obse ed a e HERC1 silenc-
ing (Fig.2A). HERC2 deple ion modi ied nei he A-RAF
no B-RAF p o ein le els (Fig.2B, C). These da a indica ed
ha RAF egula ion by HERC2 is speci ic o he C-RAF
iso o m.
Nex , we analysed he RAF MAPK signalling pa hway, in
which canonical RAF ac i a ion igge s ERK phospho yla-
ion [29]. We no ed ha C-RAF up egula ion obse ed a e
HERC1 deple ion co ela ed wi h inc eased phospho yla ed
ERK le els, while o al ERK p o ein le els emained s able.
Howe e , we de ec ed no changes in ERK phospho yla ion
in he HERC2-deple ed cells (Fig.2D). These esul s sug-
ges ed ha C-RAF up egula ion caused by HERC2 deple ion
was no signalled h ough he canonical MEK/ERK pa hway.
HERC2 egula es p38 phospho yla ion
Gi en ha HERC1 egula es he MKK3/p38 axis h ough a
RAF-dependen mechanism [29], we decided o s udy i his
mechanism was he same o HERC2. We analysed le els
o p38 phospho yla ion in U2OS cells ans ec ed wi h a
nega i e con ol siRNA, an siRNA agains HERC2, and a
posi i e con ol siRNA agains HERC1. We obse ed he
induc ion o p38 phospho yla ion in HERC2-deple ed cells,
hough wi h o al p38 p o ein le els emaining s able and
highe C-RAF p o ein le els(Fig.3A). Analogous beha -
iou was de ec ed in HERC1-deple ed cells(Fig.3A). The
same esul s o p38 phospho yla ion we e ob ained when
silencing HERC2 wi h siRNAs con aining di e en RNA
sequences (HERC2 H2.2 and HERC2 H2.4) (Fig.3B).
The phospho yla ion o p38 is associa ed wi h i s ac i-
a ion and nuclea ansloca ion. To check his, we ana-
lysed p38 subcellula localisa ion. Immuno luo escence

J.Sala-Gas on e al.
1 3
548 Page 6 o 20
expe imen s showed inc eased p38 nuclea localisa ion
in HERC2-deple ed cells (Fig.3C). This was quan i ied
assessing he nucleus:cy oplasm a io, which was highe
in HERC2-deple ed cells compa ed wi h con ol cells
(Fig.3D).
A e HERC2 silencing, p38 ac i a ion, was eplica ed in
o he human cells, such as he p53-lacking human non-small
lung ca cinoma cell line (H1299) and he non- umo igenic
human kidney 293T cell line (HEK 293T) (Fig.3E). In
addi ion, he same esul s we e ob ained in mouse cells and
Fig. 1 Pa ien -de i ed cells
wi h a homozygous mu a ion
in human HERC2 gene show
MAPK pa hway al e a ions.
A–C We analysed lysa es o
human skin ib oblas s om an
indi idual wi h he wild- ype
HERC2 (HERC2 WT) and he
p.P o594Leu mu an HERC2
a ian (HERC2 P594L) by
immunoblo , using he indi-
ca ed an ibodies. C-RAF (A),
phospho-ERK (P-ERK) (B) o
phospho-p38 (P-p38) (C) le els
we e quan i ied and no malised
based on cla h in hea y chain
(CHC), ERK o p38 p o ein le -
els, espec i ely. The esul s a e
exp essed ela i e o he con ol
condi ion. Plo s ep esen
mean ± s anda d e o o he
mean. Rep esen a i e esul s a e
shown o expe imen s epea ed
a leas h ee imes and he da a
poin s o each expe imen al
epe i ion a e plo ed as single
do s. (D) We analysed lysa es
o human skin ib oblas s
om h ee di e en con ol
indi iduals wi h he wild- ype
HERC2 (C1, C2 and C3) and
h ee di e en pa ien s wi h he
HERC2 P594L mu an a ian
(P1, P2 and P3) by immunoblo .
Le els o HERC2 and C-RAF
p o eins we e quan i ied as in
A. Le els o P-ERK and P-p38
we e quan i ied as in B, C,
espec i ely. The esul s a e
exp essed ela i e o he con ol
condi ion. Plo s ep esen
mean ± s anda d e o o he
mean. Rep esen a i e esul s a e
shown o expe imen s epea ed
a leas h ee imes and he da a
poin s o each o he indi-
iduals analysed a e plo ed as
single do s. Signi icance le els:
ns = non-signi icance; *p ≤ 0.05;
**p ≤ 0.01; ***p ≤ 0.001
HERC2 de iciency ac i a es C‑RAF/MKK3/p38 signalling pa hway al e ing hecellula esponse…
1 3
Page 7 o 20 548
when using a di e en HERC2 silencing me hod. RAW
264.7 mac ophage cell line, p ima y mouse os eoblas s and
MEFs we e in ec ed wi h len i i al pa icles ca ying ei he
an emp y ec o as a con ol (plKO) o a sho hai pin RNA
(shRNA) agains HERC2. All HERC2 knockdown cells
p esen ed highe phospho-p38 p o ein le els compa ed o
con ols, while o al p38 p o ein le els emained cons an
(Fig.3F). In conjunc ion, hese esul s demons a ed ha
HERC2 pa icipa es in egula ing p38 signalling.
HERC2 egula es heMKK3/p38 pa hway
h oughc oss alk media ed byC‑RAF
MAPK kinase (MAPKK o MKK) media es p38 ac i a ion
h ough phospho yla ion. MKK3 is he dominan iso o m
Fig. 2 HERC2 egula es C-RAF p o ein le els. A U2OS cells we e
ans ec ed wi h an siRNA nega i e con ol (NC), an siRNA agains
HERC2, o an siRNA agains HERC1. The indica ed p o ein le els
we e analysed by immunoblo . Le els o C-RAF p o eins we e quan-
i ied, no malised based on cla h in hea y chain (CHC) p o ein le -
els (loading con ol), and exp essed ela i e o he con ol condi ion.
B, C U2OS cells we e ans ec ed wi h a NC siRNA o an siRNA
agains HERC2. Le els o A-RAF (B) o B-RAF (C) we e analysed
by immunoblo , quan i ied, no malised based on CHC p o ein le els,
and exp essed ela i e o he con ol condi ion. (D) U2OS cells we e
ans ec ed as in A, and he indica ed p o ein le els we e analysed by
immunoblo . Phospho-ERK (P-ERK) le els we e quan i ied, no mal-
ised based on ERK p o ein le els and exp essed ela i e o he con-
ol condi ion. Plo s ep esen he mean ± s anda d e o o he mean.
Rep esen a i e esul s a e shown om expe imen s epea ed a leas
h ee imes and he indi idual da a poin s a e plo ed as single do s.
Signi icance le els: ns = non-signi icance; *p ≤ 0.05; **p ≤ 0.01;
***p ≤ 0.001
J.Sala-Gas on e al.
1 3
548 Page 8 o 20
in human U2OS cell lines [29], and i s ac i a ion has been
analysed by measu ing i s phospho yla ion a Se 189 [30].
Thus, we analysed MKK3 ac i a ion and i s o al p o ein
exp ession in HERC2-deple ed U2OS cells, e ealing ha
nei he MKK3 phospho yla ion a Se 189 no o al MKK3
p o ein le els we e al e ed compa ed wi h con ol cells
(Fig.4A). To con i m whe he p38 phospho yla ion ig-
ge ed by HERC2 deple ion depends on MKK3, we co-
ans ec ed U2OS wi h an MKK3 siRNA and ei he he
nega i e con ol o he HERC2 siRNA. This e ealed ha
MKK3 knockdown signi ican ly abolished he inc emen
in p38 phospho yla ion a e HERC2 deple ion (Fig.4B).
HERC2 de iciency ac i a es C‑RAF/MKK3/p38 signalling pa hway al e ing hecellula esponse…
1 3
Page 9 o 20 548
These da a sugges ed ha MKK3 ac i a ion caused he
inc ease in phospho-p38 independen o phospho yla ion
a Se 189.
Gi en he inding ha HERC2 egula es C-RAF and p38
ac i a ion, we used wo speci ic RAF kinase inhibi o s o
iden i y a po en ial c oss alk mechanism be ween he wo
pa hways. LY300912 was used o inhibi all RAF iso o ms,
and So a enib was used o inhibi only B-RAF and C-RAF.
In absence o he inhibi o s, cells showed an inc ease in p38
phospho yla ion a e HERC2 deple ion; ema kably, how-
e e , his inc ease was clea ly ab oga ed a e incuba ion
wi h LY3009120 o So a enib inhibi o s o 1h (Fig.4C).
Since RAF iso o ms in e ac by o ming di e en he e odi-
me s [31], some imes all iso o ms mus be deple ed o escue
he egula o y e ec s media ed by RAF p o eins. The e o e,
we co- ans ec ed U2OS cells wi h siRNAs agains C-RAF
o all h ee RAF iso o ms (A-RAF, B-RAF and C-RAF)
along wi h ei he he nega i e con ol siRNA o he siRNA
agains HERC2 o achie e knockdown (Fig.4D). Al hough
silencing C-RAF alone was insu icien o educe p38 phos-
pho yla ion signi ican ly, silencing all h ee iso o ms led o
a signi ican dec ease in p38 ac i a ion in HERC2-deple ed
cells (Fig.4D). Unlike pha macological inhibi ion o RAF,
iple knockdown ailed o p oduce a comple e ab oga ion
o p38 phospho yla ion a e HERC2 deple ion, which is
p obably due o he ac ha siRNA silencing did no achie e
su icien RAF iso o ms knockdown. Al oge he hese esul s
con i m he exis ence o a c oss alk be ween he RAF and
p38 signalling pa hways egula ed by HERC2.
HERC2 in e ac s wi hC‑RAF
To u he in es iga e he mechanism behind C-RAF egula-
ion by HERC2, we analysed whe he hese wo p o eins can
in e ac . In immunop ecipi a ion expe imen s in U2OS cells
wi h a speci ic an i-HERC2 an ibody (b g3), endogenous
HERC2 and C-RAF immunop ecipi a ed, while HERC1 did
no , indica ing ha he in e ac ion o HERC2 and C-RAF
was independen o HERC1 (Fig.5A). RAF he e o-dime isa-
ion be ween i s iso o ms is a well- epo ed p ocess [31], and
consis en wi h his, A-RAF and B-RAF we e also de ec ed
in HERC2 immunop ecipi a ed complexes (Fig.5B, C). The
same esul s we e ob ained in he human 293T cell line
(Fig.5D–F).
To iden i y he egion o HERC2 in e ac ing wi h C-RAF,
we co-exp essed a GFP-C-RAF usion p o ein wi h a se ies
o Myc-HERC2 usion p o eins in HEK 293T cells (Sup-
plemen a y Fig.1A), ollowed by pull-down assays wi h
GFP-binding beads. Cons uc s F4, F5, and F5CT coim-
munop ecipi a ed wi h GFP-C-RAF, indica ing ha he
HERC2 and C-RAF p o ein in e ac ion occu s mainly in
he ca boxyl- e minus o HERC2 polypep ide chain. F5CT
cons uc , which con ains he HECT domain holding he
ubiqui in ligase ac i i y, showed he highes a ini y wi h
C-RAF, sugges ing ha his is he mos ele an in e ac-
ion si e (Supplemen a y Fig.1A). HEK 293T cells we e
hen co- ans ec ed wi h a Flag-HERC2 ull-leng h usion
p o ein along wi h GFP (as a nega i e con ol) o he GFP-
C-RAF usion cons uc s (CR1, CR2, CR3 o ull-leng h)
o map he C-RAF egion in ol ed. In he GFP pull-down,
Flag-HERC2 was coimmunop ecipi a ed wi h CR1, CR3,
and he ull-leng h cons uc s (Supplemen a y Fig.1B). To
cha ac e ise his in e ac ion u he , we co-exp essed he F4
Myc-HERC2 cons uc wi h GFP-C-RAF usion cons uc s
and pe o med a GFP pull-down, which showed p e e en ial
co-immunop ecipi a ion o he F4 cons uc wi h CR3 (Sup-
plemen a y Fig.1C). In pa allel, he same expe imen was
done bu wi h he F5CT Myc-HERC2 cons uc ins ead o
F4, and his e ealed co-immunop ecipi a ion o F5CT wi h
CR1 and CR3 (Supplemen a y Fig.1D). In conjunc ion,
pull-down expe imen s con i med he in e ac ion be ween
HERC2 and C-RAF, and indica ed he possible domains
in ol ed. The HERC2 HECT domain, con ained in he
F5CT cons uc , showed he highes a ini y o C-RAF and
i s ca aly ic domain (CR3), sugges ing ha he HECT and
CR3 domains could be he mos ele an a he physiologi-
cal le el. Subsequen s uc u al s udies should con i m his
ele ance.
HERC2 egula es C‑RAF ubiqui yla ion
Ha ing shown ha he ubiqui in E3 ligase HERC2 in e -
ac s wi h C-RAF and egula es i s p o ein le els, we wan ed
Fig. 3 HERC2 egula es p38 phospho yla ion. A U2OS cells we e
ans ec ed wi h an siRNA nega i e con ol (NC), an siRNA agains
HERC2, o an siRNA agains HERC1. The indica ed p o ein le -
els we e analysed by immunoblo . Le els o phospho-p38 (P-p38)
we e quan i ied, no malised based on o al p38 p o ein le els, and
exp essed ela i e o he con ol condi ion. B U2OS cells we e ans-
ec ed wi h an siRNA nega i e con ol (NC) and wo di e en siRNA
sequences agains HERC2: H2.2 o H2.4. The indica ed p o ein le els
we e analysed by immunoblo and phospho-p38 le els we e quan i-
ied and ep esen ed as in (A). C U2OS cells ans ec ed wi h NC o
HERC2 siRNA we e analysed by immunoblo agains he indica ed
p o eins and by con ocal mic oscopy. Fixed cells we e s ained o
phospho-p38 (g een), F-ac in wi h phalloidin ( ed), and nuclei wi h
DAPI (blue) and analysed by immuno luo escence. D Fluo escence
in ensi y in he nucleus and cy oplasm pe cell was measu ed and
quan i ied. The a io nucleus/cy oplasm was calcula ed. Each da a
poin ep esen mean o a di e en ield. E HEK 293T and H1299
cells we e ans ec ed wi h a NC o HERC2 siRNA. The indica ed
p o ein le els we e analysed by immunoblo . F A RAW 264.7 mouse
mac ophage cell line, mouse p ima y os eoblas s and mouse emb y-
onic ib oblas s (MEFs) we e in ec ed wi h len i i al pa icles ca y-
ing ei he he emp y plKO ec o as a nega i e con ol (plKO) o an
shRNA agains HERC2. The indica ed p o ein le els we e analysed
by immunoblo . Plo s ep esen mean ± s anda d e o o he mean.
Rep esen a i e esul s a e shown om expe imen s epea ed a leas
h ee imes and he indi idual da a poin s a e plo ed as single do s.
Signi icance le els: ns = non-signi icance; *p ≤ 0.05; **p ≤ 0.01;
***p ≤ 0.001
◂
J.Sala-Gas on e al.
1 3
548 Page 16 o 20

HERC2 de iciency ac i a es C‑RAF/MKK3/p38 signalling pa hway al e ing hecellula esponse…
1 3
Page 17 o 20 548
HERC2-deple ed cells, con i med ha p38 ac s ups eam o
NRF2 ac i a ion. S ill, gi en he a ie y o p38 subs a es
we canno disca d ha o he ansc ip ion ac o s, apa om
NRF2, could also be in ol ed in he egula ion o he s udied
an ioxidan genes. The ansc ip ion ac o ATF-2 is ano he
impo an media o o p38 in he induc ion o SOD2 exp es-
sion upon H2O2-induced oxida i e s ess in MEFs [45]. This
possible coope a ion be ween NRF2 and ATF-2, o some
o he ansc ip ion ac o a ge ed by p38, should be s udied
u he .
O e all, ou indings may ha e bo h physiological and
clinical epe cussions. Physiologically, we e ealed a
p o-su i al unc ion o p38 ha is egula ed by HERC2.
HERC2 po en ially ine- unes he cellula esponse o oxida-
i e s ess by con olling p o ein le els o C-RAF and, he e-
o e, C-RAF/MKK3/p38 signalling o egula e an ioxidan
gene exp ession. Clinically, hese indings may be ele an
o cance , as well as indi iduals wi h HERC2 Angelman-like
synd ome due o biallelic HERC2 gene a ian s [16, 18].
Se e al HERC2 mu a ions ha e been associa ed wi h a
wide numbe o umou s [4]. In enal cance , highe HERC2
gene exp ession co ela es wi h be e pa ien p ognosis
[46], suppo ing he hypo hesis ha HERC2 may ac as a
umou supp esso [4, 46]. We p e iously demons a ed
ha HERC2, and NEURL4, egula e he ansc ip ional
ac i i y o he umou supp esso p53, acili a ing i s
oligome isa ion. HERC2 knockdown acco dingly inc eases
cell p oli e a ion due o he impai ed capabili y o a es he
cell cycle h ough p53 [10–12]. Equally, al hough he p e-
cise mechanism emains elusi e, i is well es ablished ha
he p oduc ion o eac i e oxygen species in umou cells
inc eases due o he highe me abolic a e, wi h he esul ing
excess being coun e ed by an inc eased an ioxidan cellula
esponse [47]. Suppo ing his, in mice, oncogenic alleles
o K as, B a and Myc, associa ed wi h inc eased N e2l2
exp ession. This appea o s ably enhance NRF2 an ioxi-
dan p og am and lowe in acellula eac i e oxygen species
[48]. Fu he mo e, in p53 mu a ed cance cells, he NRF2-
dependen an ioxidan esponse was selec i ely modula ed
o enhance cance cell su i al [49]. Ou da a e eal a new
mechanism by which HERC2 de iciency may con ibu e o
umou malignancy by impai ing p53 ansc ip ional ac i -
i y, and also by boos ing he cellula an ioxidan esponse
making cance cells mo e esis an o oxida i e s ess
(Fig.8). In his con ex , combina ion ea men s wi h d ugs
causing non-geno oxic ac i a ion o p53 oligome isa ion and
FDA-app o ed RAF inhibi o s, such as So a enib, ep esen
po en ial he apeu ic candida es o umou s associa ed wi h
HERC2 de iciency.
Finally, a p e ious p o eomic analysis o human HERC2
mu an s (including he p.P o594Leu a ian s udied he e)
has al eady iden i ied an en ichmen o he NRF2-media ed
oxida i e s ess esponse in HERC2 mu an s compa ed o
con ol g oup [20]. In addi ion, p o ein–p o ein in e ac-
ion ne wo ks con aining signal ansduc ion p o eins and
MAPKs we e ound o be di e en ially exp essed in HERC2
mu an s [20]. Ou esul s add o hese obse a ions and,
impo an ly, p o ide a possible molecula mechanism expla-
na ion. I would be in e es ing o u he esea ch o s udy
he implica ion o a ch onic ac i a ion o he C-RAF/MKK3/
p38 signalling pa hway in neu onal cells wi h HERC2 de i-
ciency. Al e a ions in p38 MAPK signalling in neu ons ha e
been linked o neu odegene a i e diseases including Pa kin-
son’s disease, Alzheime ’s disease and amyo ophic la e al
scle osis (ALS) [50]. The e o e, we canno discoun he pos-
sibili y ha al e a ions in his pa hway could be associa ed
wi h clinical ou comes in HERC2 Angelman-like synd ome.
Consis en wi h his, p e ious s udies ha e shown ha SOD1
o e exp ession, in which gene a ian s a e associa ed wi h
ALS and whose mRNA le els we ound o be inc eased ol-
lowing HERC2 deple ion, is associa ed wi h de ec s in he
ce ebella a chi ec u e [51, 52]. In addi ion, while exces-
si e ROS elici oxida i e s ess, hei pe sis en deple ion,
as obse ed in HERC2-de icien cells (Supplemen a y
Fig.4A), leads o an opposi e condi ion called educ i e
s ess. Pe sis en ac i a ion o an ioxidan signalling can
cause educ i e s ess and lead o pa hology. In HERC2
P594L cells, he o e ac i a ion o NRF2 signalling could
be one o he causes. In ac , NRF2 sus ained ac i a ion has
Fig. 7 HERC2 de iciency al e s cellula esis ance o H2O2-induced
oxida i e s ess. A Human skin ib oblas s de i ed om an HERC2
wild- ype indi idual (HERC2 WT) and an indi idual wi h he
p.P o594Leu HERC2 mu an a ian (HERC2 P594L) we e ea ed
wi h 500µM H2O2 o induce oxida i e s ess o he indica ed ime
poin s and p o ein le els we e analysed by immunoblo . Phospho-
p38 (P-p38) le els we e quan i ied, no malised based on o al p38
p o ein le els and exp essed ela i e o he non- ea ed con ol condi-
ion (HERC2 WT, = 0). Plo s ep esen mean ± s anda d e o o he
mean o 4 independen expe imen s (n = 4). B Human skin ib oblas s
we e ea ed as in (A) and images we e acqui ed by op ical mic os-
copy a e he indica ed ea men imes, wi h ep esen a i e images
shown om expe imen s epea ed h ee imes (n = 3). C, D HERC2
WT and HERC2 P594L human skin ib oblas s we e ea ed wi h
500µM H2O2 o 6h (C) o wi h 50µM H2O2 o 24h (D) as indi-
ca ed. Cells we e ea ed 1µM LY3009120, 1µM So a enib o 10µM
SB203580 in he speci ied condi ions 1 h be o e adding H2O2. An
MTT assay was pe o med. Da a a e p esen ed as a pe cen age ela-
i e o he con ol and un ea ed condi ion. E U2OS we e ans ec ed
wi h an siRNA nega i e con ol (NC) o an siRNA agains HERC2.
A p53 siRNA was added when indica ed (siRNA p53). Subsequen ly,
cells we e pla ed in a 96-well pla e and allowed o g ow o 48h o
e alua e cell p oli e a ion and an MTT assay was pe o med. Da a a e
p esen ed ela i e o he con ol condi ion (NC, WT p53). F U2OS
cells we e ans ec ed wi h he NC o HERC2 siRNA along wi h
p53 siRNA (NC + p53/HERC2 + p53). Cell iabili y was assessed by
MTT assay (unde he same condi ions men ioned in D). Plo s ep-
esen mean ± s anda d e o o he mean. Rep esen a i e esul s a e
shown om expe imen s epea ed a leas h ee imes and he indi-
idual da a poin s a e plo ed as single do s. Signi icance le els:
ns = non-signi icance; *p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001
◂
J.Sala-Gas on e al.
1 3
548 Page 18 o 20
al eady been linked o educ i e s ess [53]. Highligh ing he
impo ance o educ i e s ess on pa hology, mu a ions in
key componen s o he cellula educ i e s ess esponse can
cause de elopmen al diseases. Fo ins ance, FEM1B gain-
o - unc ion mu a ion, which cause a pe sis en ac i a ion
o he educ i e s ess esponse, elici de elopmen al syn-
d omes wi h some simila i ies o he HERC2 Angelman-like
synd ome [33, 34]. An example o he damage ha educ-
i e s ess can exe on cells is ha i can induce mi ochon-
d ial dys unc ion and impac on he co ec cell unc ion
[54, 55]. Acco dingly, we obse ed an inc eased numbe
o agmen ed mi ochond ia in HERC2 P594L cells, which
is a common ea u e obse ed in neu odegene a ion [56].
Howe e , mo e expe imen s a e needed o con i m hese
hypo heses and o associa e hese mechanisms wi h clinical
ou comes in HERC2 Angelman-like synd ome. All hings
conside ed, he indings in his s udy iden i y p38 and RAF
inhibi o s as po en ial he apeu ic op ions o indi iduals
who p esen wi h such a e disease.
Supplemen a y In o ma ion The online e sion con ains supplemen-
a y ma e ial a ailable a h ps:// doi. o g/ 10. 1007/ s00018- 022- 04586-7.
Acknowledgemen s We would like o hank D . Benjamín To e-
jón and D . Es he Cas año om “Cen es Cien i ics i Tecnològics
de la Uni e si a de Ba celona” (CCiT-UB), and Ms. Es he Adan-
e o o echnical assis ance. We also hank D . Oh a and D . E azo
o kindly p o iding us wi h plasmids. This esea ch was suppo ed
by he ollowing g an s: JLR (Agencia Es a al de In es igación:
PID2020-120344RB-I00/MCIN/AEI/10.13039/501100011033); and
FV (PDC2021-121776-I00 and PID2020-117278GB-I00 om MCIN/
AEI/10.13039/501100011033 and FEDER “Una mane a de hace
Eu opa” “Nex Gene a ionEU”/PRTR; and G an 202038-30 om “La
Ma a ó de TV3”). As well, his a icle is based upon wo k om COST
Ac ion P o eoCu e CA20113, suppo ed by COST (Eu opean Coope a-
ion in Science and Technology). J.S.-G. and A.M.-M. ecei ed FPU
Fellowships (FPU17/02413 and FPU18/06325, espec i ely) om he
Spanish Minis y o Uni e si ies.
Au ho con ibu ions Concei ed and designed he expe imen s: JSG,
UM, FV, and JLR. Pe o med he expe imen s: JSG, LP, JR, AMM, and
JLR. All he au ho s analysed he da a. The i s d a o he manusc ip
was w i en by JSG and JLR, and all he au ho s commen ed on p e i-
ous e sions o he manusc ip . All he au ho s ead and app o ed he
inal manusc ip .
Funding Open Access unding p o ided hanks o he CRUE-
CSIC ag eemen wi h Sp inge Na u e. This s udy was unded
by he ollowing g an s: JLR (Agencia Es a al de In es igación:
PID2020-120344RB-I00/MCIN/AEI/10.13039/501100011033) and
FV (PDC2021-121776-I00 and PID2020-117278 GB-I00 om MCIN/
Fig. 8 Wo king model o HERC2 unc ion in heal h and disease. In
p e ious s udies, we showed ha independen ly o he ubiqui in ligase
ac i i y, HERC2 along wi h NEURL4, acili a es p53 oligome isa-
ion o p omo e p53 ansc ip ional p og am ac i a ion. Fo example,
he a ge gene p21 egula es he cell cycle and p omo es cell cycle
a es . Unde condi ions o HERC2 de iciency o down- egula ion,
he ansc ip ional ac i a ion o p53 is impai ed due o he comp o-
mised p53 oligome isa ion p ocess [10–13]. Now, wi h da a p e-
sen ed in his s udy, we complemen his wo king model by adding
an impo an unc ion o HERC2 dependen on i s ubiqui in ligase
ac i i y. Unde no mal condi ions, HERC2 con ols C-RAF p o ein
le els by egula ing i s ubiqui yla ion and a ge ing i o p o easomal
deg ada ion. Hence, in HERC2-de icien cells, C-RAF p o ein le els
inc ease, which ac i a es a c oss alk be ween he C-RAF and MKK3/
p38 signalling pa hways. Once p38 is ac i a ed by phospho yla ion, i
ansloca es o he nucleus and ac i a es i s a ge ansc ip ion ac o s
(TFs). This e en ually ac i a es ansc ip ion o genes ela ed o he
oxida i e s ess esponse such as NFE2L2, SOD1, SOD2 and GPX1,
which p edisposes cells o an enhanced esis ance o oxida i e s ess.
The combina ion o hese e ec s in he p53/p21 and MKK3/p38 pa h-
ways may a ec bo h umo igenesis and neu onal cell homeos asis
HERC2 de iciency ac i a es C‑RAF/MKK3/p38 signalling pa hway al e ing hecellula esponse…
1 3
Page 19 o 20 548
AEI/10.13039/501100011033 and FEDER “Una mane a de hace
Eu opa” “Nex Gene a ionEU”/PRTR. And G an 202038-30 om
“La Ma a ó de TV3”). As well, his a icle is based upon wo k om
COST Ac ion P o eoCu e CA20113, suppo ed by COST (Eu opean
Coope a ion in Science and Technology). JSG and AMM ecei ed FPU
Fellowships (FPU17/02413 and FPU18/06325, espec i ely) om he
Spanish Minis y o Uni e si ies.
Da a a ailabili y All da a analysed du ing his s udy o e alua e he
conclusions a e included wi hin he a icle o a ailable in supplemen-
al in o ma ion. Addi ional ela ed da a need o be eques ed om he
co esponding au ho .
Decla a ions
Con lic o in e es The au ho s decla e no compe ing in e es s.
Open Access This a icle is licensed unde a C ea i e Commons A i-
bu ion 4.0 In e na ional License, which pe mi s use, sha ing, adap a-
ion, dis ibu ion and ep oduc ion in any medium o o ma , as long
as you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce,
p o ide a link o he C ea i e Commons licence, and indica e i changes
we e made. The images o o he hi d pa y ma e ial in his a icle a e
included in he a icle's C ea i e Commons licence, unless indica ed
o he wise in a c edi line o he ma e ial. I ma e ial is no included in
he a icle's C ea i e Commons licence and you in ended use is no
pe mi ed by s a u o y egula ion o exceeds he pe mi ed use, you will
need o ob ain pe mission di ec ly om he copy igh holde . To iew a
copy o his licence, isi h p:// c ea i eco mmons. o g/ licen ses/ by/4. 0/.
Re e ences
1. Pa en i I, Rabaneda LG, Schoen H, No a ino G (2020) Neu ode-
elopmen al diso de s: om gene ics o unc ional pa hways.
T ends Neu osci 43:608–621. h ps:// doi. o g/ 10. 1016/j. ins. 2020.
05. 004
2. Ga cía-Cano J, Ma inez-Ma inez A, Sala-Gas on J e al (2019)
HERCing: s uc u al and unc ional ele ance o he la ge HERC
ubiqui in ligases. F on Physiol. h ps:// doi. o g/ 10. 3389/ phys.
2019. 01014
3. Husnjak K, Dikic I (2012) Ubiqui in-binding p o eins: decod-
e s o ubiqui in-media ed cellula unc ions. Annu Re Bio-
chem 81:291–322. h ps:// doi. o g/ 10. 1146/ annu e - bioch
em- 051810- 094654
4. Sala-Gas on J, Ma inez-Ma inez A, Ped azza L e al (2020) He c
ubiqui in ligases in cance . Cance s (Basel) 12:1–13. h ps:// doi.
o g/ 10. 3390/ cance s120 61653
5. Lee TH, Pa k JM, Leem SH, Kang TH (2014) Coo dina ed egu-
la ion o XPA s abili y by ATR and HERC2 du ing nucleo ide
excision epai . Oncogene 33:19–25. h ps:// doi. o g/ 10. 1038/
onc. 2012. 539
6. Peng Y, Dai H, Wang E e al (2015) TUSC4 unc ions as a
umo supp esso by egula ing BRCA1 s abili y. Cance Res
75:378–386. h ps:// doi. o g/ 10. 1158/ 0008- 5472. CAN- 14- 2315
7. Yuan J, Luo K, Deng M e al (2014) HERC2-USP20 axis egu-
la es DNA damage checkpoin h ough Claspin. Nucleic Acids
Res 42:13110–13121. h ps:// doi. o g/ 10. 1093/ na / gku10 34
8. Zhu M, Zhao H, Liao J, Xu X (2014) HERC2/USP20 coo -
dina es CHK1 ac i a ion by modula ing CLASPIN s abili y.
Nucleic Acids Res 42:13074–13081. h ps:// doi. o g/ 10. 1093/
na / gku978
9. Wu W, Roku anda N, Takeuchi J e al (2018) HERC2 acili a es
BLM and WRN helicase complex in e ac ion wi h RPA o sup-
p ess G-quad uplex DNA. Cance Res 78:6371–6385. h ps:// doi.
o g/ 10. 1158/ 0008- 5472. CAN- 18- 1877
10. Cubillos-Rojas M, Amai -Pinedo F, Pei ó-Jo dán R e al (2014)
The E3 ubiqui in p o ein ligase HERC2 modula es he ac i i y o
umo p o ein p53 by egula ing i s oligome iza ion. J Biol Chem
289:14782–14795. h ps:// doi. o g/ 10. 1074/ jbc. M113. 527978
11. Cubillos-Rojas M, Schneide T, Ba ons R e al (2017) NEURL4
egula es he ansc ip ional ac i i y o umo supp esso p o ein
p53 by modula ing i s oligome iza ion. Onco a ge 8:61824–
61836. h ps:// doi. o g/ 10. 18632/ onco a ge . 18699
12. Ga cía-Cano J, Sánchez-Tena S, Sala-Gas on J e al (2020) Regu-
la ion o he MDM2-p53 pa hway by he ubiqui in ligase HERC2.
Mol Oncol 14:69–86. h ps:// doi. o g/ 10. 1002/ 1878- 0261. 12592
13. Cubillos-Rojas M, Schneide T, Hadjebi O e al (2016) The
HERC2 ubiqui in ligase is essen ial o emb yonic de elopmen
and egula es mo o coo dina ion. Onco a ge 7:56083–56106.
h ps:// doi. o g/ 10. 18632/ onco a ge . 11270
14. Pé ez-Villegas EM, Ruiz R, Bachille S e al (2021) The HERC
p o eins and he ne ous sys em. Semin Cell De Biol. h ps:// doi.
o g/ 10. 1016/j. semcdb. 2021. 11. 017
15. Pu enbe ge EG, Jinks RN, Wang H e al (2012) A homozygous
missense mu a ion in HERC2 associa ed wi h global de elopmen-
al delay and au ism spec um diso de . Hum Mu a 33:1639–
1646. h ps:// doi. o g/ 10. 1002/ humu. 22237
16. Ha lalka GV, Baple EL, C oss H e al (2013) Mu a ion o HERC2
causes de elopmen al delay wi h angelman-like ea u es. J Med
Gene 50:65–73. h ps:// doi. o g/ 10. 1136/ jmedg ene - 2012- 101367
17. Mo ice-Pica d F, Bena d G, Rez ani HR e al (2016) Comple e
loss o unc ion o he ubiqui in ligase HERC2 causes a se e e
neu ode elopmen al pheno ype. Eu J Hum Gene 25:52–58.
h ps:// doi. o g/ 10. 1038/ ejhg. 2016. 139
18. Elpido ou M, Bes S, Poul e JA e al (2021) No el loss-o - unc-
ion mu a ion in HERC2 is associa ed wi h se e e de elopmen al
delay and paedia ic le hali y. J Med Gene 58:334–341. h ps://
doi. o g/ 10. 1136/ jmedg ene - 2020- 106873
19. Vincen KM, Ea on A, Yassaee VR e al (2021) Delinea ing he
expanding pheno ype o HERC2- ela ed diso de s: he impac o
biallelic loss o unc ion e sus missense a ia ion. Clin Gene
100:637–640. h ps:// doi. o g/ 10. 1111/ cge. 14039
20. Ab aham JR, Ba na d J, Wang H e al (2019) P o eomic in es iga-
ions o human HERC2 mu an s: Insigh s in o he pa hobiology o
a neu ode elopmen al diso de . Biochem Biophys Res Commun
512:421–427. h ps:// doi. o g/ 10. 1016/j. bb c. 2019. 02. 149
21. Wu W, Sa o K, Koike A e al (2010) HERC2 is an E3 ligase
ha a ge s BRCA1 o deg ada ion. Cance Res 70:6384–6392.
h ps:// doi. o g/ 10. 1158/ 0008- 5472. CAN- 10- 1304
22. Schneide T, Ma inez-Ma inez A, Cubillos-Rojas M e al (2018)
The E3 ubiqui in ligase HERC1 con ols he ERK signaling pa h-
way a ge ing C-RAF o deg ada ion. Onco a ge 9:31531–31548.
h ps:// doi. o g/ 10. 18632/ onco a ge . 25847
23. Chan NC, Den Bes en W, Swe edoski MJ e al (2014) Deg ada ion
o he deubiqui ina ing enzyme USP33 is media ed by p97 and he
ubiqui in ligase HERC2. J Biol Chem 289:19789–19798. h ps://
doi. o g/ 10. 1074/ jbc. M114. 569392
24. E azo T, Mo eno A, Ruiz-Babo G e al (2013) Canonical and
kinase ac i i y-independen mechanisms o ex acellula signal-
egula ed kinase 5 (ERK5) nuclea ansloca ion equi e disso-
cia ion o Hsp90 om he ERK5-Cdc37 complex. Mol Cell Biol
33:1671–1686. h ps:// doi. o g/ 10. 1128/ mcb. 01246- 12
25. Rami ez J, P ie o G, Olazabal-He e o A e al (2021) A p o -
eomic app oach o sys ema ic mapping o subs a es o human
J.Sala-Gas on e al.
1 3
548 Page 20 o 20
deubiqui ina ing enzymes. In J Mol Sci 22:4851. h ps:// doi. o g/
10. 3390/ ijms2 20948 51
26. Mo a J, G uenebe g DA, Yang X e al (2006) A len i i al RNAi
lib a y o human and mouse genes applied o an a ayed i al
high-con en sc een. Cell 124:1283–1298. h ps:// doi. o g/ 10.
1016/j. cell. 2006. 01. 040
27. Casas-Te adellas E, Ta o I, Ba ons R e al (2008) ERK and p38
pa hways egula e amino acid signalling. Biochim Biophys Ac a
Mol Cell Res 1783:2241–2254. h ps:// doi. o g/ 10. 1016/j. bbamc .
2008. 08. 011
28. Cubillos-Rojas M, Amai -Pinedo F, Ta o I e al (2010) Simul a-
neous elec opho e ic analysis o p o eins o e y high and low
molecula mass using T is-ace a e polyac ylamide gels. Elec o-
pho esis 31:1318–1321. h ps:// doi. o g/ 10. 1002/ elps. 20090 0657
29. Ped azza L, Schneide T, Ba ons R e al (2020) The ubiqui in
ligase HERC1 egula es cell mig a ion ia RAF-dependen egula-
ion o MKK3/p38 signaling. Sci Rep 10:824. h ps:// doi. o g/ 10.
1038/ s41598- 020- 57756-7
30. Raingeaud J, Gup a S, Roge s JS e al (1995) P o-in lamma o y
cy okines and en i onmen al s ess cause p38 mi ogen- ac i a ed
p o ein kinase ac i a ion by dual phospho yla ion on y osine and
h eonine. J Biol Chem 270:7420–7426. h ps:// doi. o g/ 10. 1074/
jbc. 270. 13. 7420
31. B umme T, McInnes C (2020) RAF kinase dime iza ion: impli-
ca ions o d ug disco e y and clinical ou comes. Oncogene
39:4155–4169. h ps:// doi. o g/ 10. 1038/ s41388- 020- 1263-y
32. Bellezza I, Giambanco I, Minelli A, Dona o R (2018) N 2-Keap1
signaling in oxida i e and educ i e s ess. BBA Mol Cell Res
1865:721–733. h ps:// doi. o g/ 10. 1016/j. bbamc . 2018. 02. 010
33. Man o d AG, Rod íguez-Pé ez F, Shih KY e al (2020) A cellula
mechanism o de ec and alle ia e educ i e s ess. Cell 183:46-
61.e21. h ps:// doi. o g/ 10. 1016/j. cell. 2020. 08. 034
34. Man o d AG, Mena EL, Shih KY e al (2021) S uc u al basis and
egula ion o he educ i e s ess esponse. Cell 184:5375-5390.
e16. h ps:// doi. o g/ 10. 1016/j. cell. 2021. 09. 002
35. Galligan T, Ma inez-noe G, A nd V e al (2015) P o eomic anal-
ysis and iden i ica ion o cellula in e ac o s o he gian ubiqui in
ligase HERC2. J P o eome Res 14:953–966. h ps:// doi. o g/ 10.
1021/ p 501 005
36. Cano as B, Neb eda AR (2021) Di e si y and e sa ili y o p38
kinase signalling in heal h and disease. Na Re Mol Cell Biol
22:346–366. h ps:// doi. o g/ 10. 1038/ s41580- 020- 00322-w
37. Rojo de la Vega M, Chapman E, Zhang DD (2018) NRF2 and he
Hallma ks o cance . Cance Cell 34:21–43. h ps:// doi. o g/ 10.
1016/j. ccell. 2018. 03. 022
38. Suzuki T, Yamamo o M (2015) Molecula basis o he Keap1-
N 2 sys em. F ee Radic Biol Med 88:93–100. h ps:// doi. o g/ 10.
1016/j. ee adbio med. 2015. 06. 006
39. Kang KW, Ryu JIH, Kim SG (2000) The essen ial ole o phos-
pha idylinosi ol 3-kinase and o p38 mi ogen-ac i a ed p o ein
kinase ac i a ion in he an ioxidan esponse elemen -media ed
GSTA2 induc ion by dec eased glu a hione in H4IIE hepa oma
cells. Mol Pha macol 58:1017–1025. h ps:// doi. o g/ 10. 1124/ mol.
58.5. 1017
40. Alam J, Wicks C, S ewa D e al (2000) Mechanism o heme
oxygenase-1 gene ac i a ion by cadmium in MCF-7 mamma y
epi helial cells: ole o p38 kinase and N 2 ansc ip ion ac o . J
Biol Chem 275:27694–27702. h ps:// doi. o g/ 10. 1074/ jbc. M0047
29200
41. Zippe LM, Mulcahy RT (2000) Inhibi ion o ERK and p38 MAP
kinases inhibi s binding o N 2 and induc ion o GCS genes.
Biochem Biophys Res Commun 278:484–492. h ps:// doi. o g/ 10.
1006/ bb c. 2000. 3830
42. Yu R, Chen C, Mo YY e al (2000) Ac i a ion o mi ogen-
ac i a ed p o ein kinase pa hways induces an ioxidan esponse
elemen -media ed gene exp ession ia a N 2-dependen mecha-
nism. J Biol Chem 275:39907–39913. h ps:// doi. o g/ 10. 1074/ jbc.
M0040 37200
43. Huang HC, Nguyen T, Picke CB (2002) Phospho yla ion o
N 2 a Se -40 by p o ein kinase C egula es an ioxidan esponse
elemen -media ed ansc ip ion. J Biol Chem 277:42769–42774.
h ps:// doi. o g/ 10. 1074/ jbc. M2069 11200
44. He F, Ru X, Wen T (2020) NRF2, a ansc ip ion ac o o s ess
esponse and beyond. In J Mol Sci 21:1–23. h ps:// doi. o g/ 10.
3390/ ijms2 11347 77
45. Gu ié ez-Uzquiza Á, A echede a M, B agado P e al (2012)
p38α media es cell su i al in esponse o oxida i e s ess ia
induc ion o an ioxidan genes: e ec on he p70S6K pa hway.
J Biol Chem 287:2632–2642. h ps:// doi. o g/ 10. 1074/ jbc. M111.
323709
46. Schneide T, Ma inez-Ma inez A, Cubillos-Rojas M e al (2019)
La ge HERCS unc ion as umo supp esso s. F on Oncol 9:524.
h ps:// doi. o g/ 10. 3389/ onc. 2019. 00524
47. Pe illo B, Di DM, Pezone A e al (2020) ROS in cance he apy:
he b igh side o he moon. Exp Mol Med 52:192–203. h ps://
doi. o g/ 10. 1038/ s12276- 020- 0384-2
48. Denicola GM, Ka e h FA, Hump on TJ e al (2011) Oncogene-
induced N 2 ansc ip ionp omo es ROS de oxi ica ion and
umo igenesis. Na u e 475:106–109. h ps:// doi. o g/ 10. 1038/ na u
e10189
49. Lisek K, Campane E, Ciani Y e al (2018) Mu an p53 unes
he NRF2-dependen an ioxidan esponse o suppo su i al o
cance cells. Onco age 9:20508–20523. h ps:// doi. o g/ 10. 18632/
onco a ge . 24974
50. Co ea SAL, Eales KL (2012) The ole o p38 MAPK and i s
subs a es in neu onal plas ici y and neu odegene a i e disease.
J Signal T ansduc 2012:649079. h ps:// doi. o g/ 10. 1155/ 2012/
649079
51. Jaa sma D, Haasdijk ED, G asho n JAC e al (2000) Human Cu/
Zn supe oxide dismu ase (SOD1) o e exp ession in mice causes
mi ochond ial acuoliza ion, axonal degene a ion, and p ema u e
mo oneu on dea h and accele a es mo oneu on disease in mice
exp essing a amilial amyo ophic la e al scle osis mu . Neu obiol
Dis 7:623–643. h ps:// doi. o g/ 10. 1006/ nbdi. 2000. 0299
52. A sha P, Ash a i N, Jiao X e al (2017) O e exp ession o human
SOD1 leads o disc e e de ec s in he ce ebella a chi ec u e in
he mouse. F on Neu oana 11:22. h ps:// doi. o g/ 10. 3389/ nana.
2017. 00022
53. Rajaseka an NS, Va adha aj S, Khande ao GD e al (2011) Sus-
ained ac i a ion o nuclea e y h oid 2- ela ed ac o 2/an ioxi-
dan esponse elemen signaling p omo es educ i e s ess in he
human mu an p o ein agg ega ion ca diomyopa hy in mice. An i-
oxidan s Redox Signal 14:957–971. h ps:// doi. o g/ 10. 1089/ a s.
2010. 3587
54. Ma WX, Li CY, Tao R e al (2020) Reduc i e s ess-induced
mi ochond ial dys unc ion and ca diomyopa hy. Oxid Med Cell
Longe . h ps:// doi. o g/ 10. 1155/ 2020/ 51369 57
55. Xiao W, Loscalzo J (2020) Me abolic esponses o educ i e
s ess. An ioxidan s Redox Signal 32:1330–1347. h ps:// doi. o g/
10. 1089/ a s. 2019. 7803
56. Kno AB, Pe kins G, Schwa zenbache R, Bossy-We zel E (2008)
Mi ochond ial agmen a ion in neu odegene a ion. Na Re Neu-
osci 9:505–518. h ps:// doi. o g/ 10. 1038/ n n24 17
Publishe 's No e Sp inge Na u e emains neu al wi h ega d o
ju isdic ional claims in published maps and ins i u ional a ilia ions.