ARTICLE OPEN
SUMO2/3 modifica ion o ansc ip ion-associa ed p o eins
con ols cell iabili y in esponse o oxygen and glucose
dep i a ion-media ed s ess
F ancisco Galla do-Chamizo
1
, Román González-P ie o
1,2
, Vahid Ja a i
1
, Noelia Luna-Peláez
1
, Al ed C. O. Ve egaal
2
and
Ma io Ga cía-Domínguez
1
✉
© The Au ho (s) 2025
Because limi ed oxygen and glucose supply o issues is a se ious challenge ha cells mus p ope ly measu e o decide be ween
su i ing o igge ing cell dea h, o ganisms ha e de eloped accu a e mechanisms o sensing and signaling hese condi ions. In
ecen yea s, signaling h ough pos ansla ional modifica ion o p o eins by co alen a achmen o he Small Ubiqui in-like
Modifie (SUMO) is gaining no o ie y. Enhanced sumoyla ion in esponse o oxygen and glucose dep i a ion (OGD) cons i u es a
sa egua d mechanism o cells and a new a enue o he apeu ic in e en ion. Howe e , indisc imina e global sumoyla ion can limi
he he apeu ic po en ial ha a mo e p ecise ac ion on selec ed a ge s would ha e. To clea up his, we ha e conduc ed a
p o eomic app oach in P19 cells o iden i y specific SUMO a ge s esponding o OGD and o in es iga e he po en ial ha hese
a ge s and hei sumoyla ion ha e in p ese ing cells om dea h. P o eins unde going sumoyla ion in esponse o OGD a e mos ly
ela ed o ansc ip ion and RNA p ocessing, and he majo i y o hem a e apidly desumoyla ed when es o ing oxygen and
glucose (ROG), confi ming he high dynamics o his modifica ion. Since OGD is linked o b ain ischemia, we ha e also s udied cells
di e en ia ed in o neu ons. Howe e , no majo di e ences ha e been obse ed be ween he SUMO-p o eomes o p oli e a ing and
di e en ia ed cells. We show ha he o e exp ession o he ansc ip ion ac o SOX2 o he SUMO ligase PIAS4 has a mani es cell
p o ec i e e ec la gely depending on hei sumoyla ion, and ha main aining he sumoyla ion capaci y o he co egula o NAB2 is
also impo an o ace OGD. Con e sely, sumoyla ion o he plu ipo ency ac o OCT4, which is sumoyla ed unde OGD, and is a
a ge o he SUMO p o ease SENP7 o desumoyla ion a e ROG, seems o block i s cell su i al-p omo ing capaci y. Thus, be e
ou comes in cell p o ec ion would ely on he app op ia e combina ion o sumoyla ed and non-sumoyla ed o ms o selec ed
ac o s.
Cell Dea h Disco e y (2025) 11:230 ; h ps://doi.o g/10.1038/s41420-025-02513-w
INTRODUCTION
Limi ed physiological o pa hological supply o oxygen and
glucose o cells se iously comp omises cell iabili y. Many cell
ypes in he o ganism and cance cells, especially hose inside
solid umo s, success ully adap o hese condi ions unde a mild
limi ing supply. Howe e , se e e oxygen and glucose dep i a ion
(OGD) has d ama ic consequences o cell su i al. Fo ins ance,
ischemia, especially in he b ain, depending on he du a ion o he
p ocess and he a ec ed su ace, can cause d as ic damage and a
se e e loss o unc ionali y in he pa ien ’s b ain, p o oundly
a ec ing no mal daily asks, and cons i u ing a se ious public
heal h p oblem nowadays.
Sumoyla ion o p o eins has been epo ed o play a key ole in
p omo ing cell su i al unde OGD condi ions. Sumoyla ion consis s
o he a achmen o he Ubiqui in-like small polypep ide SUMO
(Small Ubiqui in-like MOdifie ) o Lys (K) esidues, o en embedded
in he consensus sequence ΨKXE (Ψ, la ge hyd ophobic esidue; X,
any esidue), wi hin p o eins, as a pos ansla ional modifica ion [1].
A e p o eoly ic ma u a ion by specific SUMO p o eases o se e al
C- e minal amino acids, SUMO is ac i a ed by he he e odime ic
SAE1/UBA2 E1 enzyme o be ans e ed o he SUMO E2
conjuga ing enzyme UBC9, which sumoyla es a ge p o eins,
equen ly helped by an E3 ligase. The SUMO p o eases a e also
in ol ed in ecycling SUMO om a ge s and he mos s udied a e
hose o he SENP amily [2]. SUMO ligases, h ough di e en
mechanisms, acili a e SUMO ans e om UBC9 o a ge p o eins,
and hose o he PIAS amily ha e been ex ensi ely in es iga ed [3].
P o eases and ligases a e key SUMO egula o s, which appea highly
dys egula ed in cance [4].
SUMO is essen ial in e eb a es and pa icipa es in many
physiological p ocesses; hus, i also associa es wi h mul iple
diseases [5]. Up o fi e SUMO molecules ha e been iden ified in
humans. Howe e , clea e idence o conjuga ion capaci y and
wide oles ha e been es ic ed o SUMO1, 2, and 3. SUMO2 and 3
a e highly homologous and undis inguishable by an ibodies and
unc ionally, and as such hey a e usually e e ed o as SUMO2/3.
Recei ed: 29 Oc obe 2024 Re ised: 21 Ma ch 2025 Accep ed: 27 Ma ch 2025
1
Andalusian Cen e o Molecula Biology and Regene a i e Medicine-CABIMER, CSIC-Uni e sidad de Se illa-Uni e sidad Pablo de Ola ide, Se ille, Spain.
2
Depa men o Cell and
Chemical Biology, Leiden Uni e si y Medical Cen e , Leiden, Ne he lands. ✉email: [email p o ec ed]
www.na u e.com/cddisco e y
O icial jou nal o CDDp ess
1234567890();,:
Indeed, i is SUMO2/3, he SUMO pa alog mos ly associa ed wi h
he esponse o OGD condi ions. In con as o SUMO1, SUMO2/3
is abundan in he cell in he unconjuga ed o m and apidly
a ached o p o eins in esponse o se e al s ess s imuli [6],
including ischemia [7,8]. Ano he di e ence be ween SUMO1 and
SUMO2/3 is he p esence in SUMO2/3 o a sumoyla ion consensus
si e, which acili a es he o ma ion o poly-SUMO chains.
Fi s unde physiological condi ions du ing hibe na ing o po
in squi els and subsequen ly unde pa hological condi ions by
inducing ischemia in mice by middle ce eb al a e y occlusion,
inc eased and massi e sumoyla ion o p o eins has been
co ela ed wi h neu op o ec ion ( e iewed in [9]). Indeed, di e se
epo s ha e es ablished be e ou comes in cell iabili y
associa ed wi h SUMO o UBC9 o e exp ession and wo se
ou comes associa ed wi h SUMO silencing o SENP p o eases
o e exp ession [10–14]. Howe e , hese app oaches su ely lead o
indisc imina e sumoyla ion o many p o eins, which can limi he
posi i e ou comes due o he nega i e e ec s o ce ain modified
p o eins. Thus, he iden ifica ion o specific a ge s and he p ecise
cha ac e iza ion o hei oles in he sumoyla ed o m is
undamen al o mo e p ecisely and selec i ely op imize in e en-
ion a his le el. In his con ex , a ew wo ks ha e explo ed he
iden ifica ion o SUMO a ge s in esponse o OGD, bu wi h
limi ed ou comes [15,16]. In one case, a educed numbe o
p o eins ha e been iden ified, and in gene al, hese s udies lack
unc ional analysis highligh ing he ele ance o a ge modifica-
ion in he con ex o cell iabili y unde OGD-media ed s ess.
In his s udy, we ha e in es iga ed he changes in he SUMO2
p o eome o cells subjec ed o OGD, aking ad an age o he
di e en iable P19 cell line, o compa e SUMO2 a ge s unde
no mal g ow h condi ions wi h hose in di e en ia ed cells. Ou
esul s indica e ha dozens o p o eins a e simila ly modified
unde bo h condi ions, mainly g ouping in o ansc ip ion- ela ed
unc ional ca ego ies. In addi ion, we also show how
o e exp ession o sumoyla ion mu an s o selec ed ac o s may
lead o di e en e ec s on cell iabili y. This opens he possibili y
o he apeu ically in e e ing wi h he sumoyla ion o specific
ac o s o ge be e ou comes in cell iabili y in esponse o
ischemia.
RESULTS
P o eomic analysis e eals SUMO2-modifica ion o mul iple
ansc ip ion-associa ed p o eins in esponse o oxygen and
glucose dep i a ion
As he sumoyla ion o p o eins has been associa ed wi h ole ance
o s ess condi ions, and OGD has se ious consequences unde
pa hological and physiological ci cums ances, we aimed a
iden i ying p o eins modified by SUMO in esponse o OGD o
shed ligh on mechanisms igge ed in esponse o dele e ious
condi ions and o disco e new a ge s o pu a i e he apeu ic
in e en ion. Fo his pu pose, we ha e pe o med a p o eomic
app oach o pu i y and iden i y SUMO2 conjuga es. Fo he s udy,
we ha e chosen mouse emb yonal ca cinoma P19 cells as hey
ha e he in e es o p esen ing plu ipo en cha ac e is ics, being
easily di e en iable h ough se e al well-es ablished p o ocols
[17,18].
We fi s de e mined he app op ia e iming in P19 cells o he
inc ease in sumoyla ion unde OGD and he dec ease a e
es o ing oxygen and glucose (ROG). Kine ic analysis indica ed
ha maximal modifica ion was eached a e 2.5 h o OGD and
ha 2.5 h o ROG was enough o es o e no mal sumoyla ion
le els (Fig. 1A). Changes in he le el o he hypoxia ac o HIF1α
we e eco ded as a con ol o he oxygen dep i a ion p ocess
(Fig. 1A). Thus, o p o eomic s udies, we es ablished a p o ocol o
2.5 h o OGD and 2.5 h o ROG a e 2.5 h o OGD (Fig. 1B). To
acili a e pu ifica ion o sumoyla ed p o eins, s able cell lines
exp essing a 10× His- agged (His
10
) SUMO2 cons uc [19] we e
gene a ed. Se e al an ibio ic- esis an clones we e isola ed and
es ed o exp ession o His
10
-SUMO2 using an i-His an ibodies
(Supplemen a y Fig. S1A). We selec ed clone 5 (designa ed as #A5
cells om now on) o u he expe imen s as i p esen ed simila
le els o exp ession o endogenous SUMO2/3 and His
10
-SUMO2,
as de e mined by wes e n blo wi h an i-SUMO2/3 an ibodies,
which e ealed a double band co esponding o he endogenous
p o ein and he His
10
- agged ansgenic e sion wi h lowe
mig a ion (Supplemen a y Fig. S1B).
Ex ac s om #A5 and pa en al P19 cells as a con ol, no mally
g owing (p oli e a ing) o di e en ia ed, we e p epa ed unde
dena u ing condi ions. His
10
-SUMO2-modified p o eins we e
pu ified by pull-down using Ni-NTA beads (Fig. 2), and samples
we e p epa ed o mass spec ome y-based p o eomics analysis
(MS). Hea maps confi med he ep oducibili y o he esul s in he
di e en eplica es a e p o ein pu ifica ion, and also showed
en ichmen o a subse o p o eins unde OGD in compa ison wi h
con ol and ROG condi ions in #A5 cells and no in he pa en al
P19 cell line (Supplemen a y Fig. S2). A o al o 670 p o eins we e
consis en ly iden ified by MS in a leas one condi ion in
p oli e a ing cells, and 434 could be conside ed sumoyla ion
subs a es (Supplemen a y Da ase S1). We fi s ocused on
changes in sumoyla ion be ween OGD and con ol condi ions,
conside ing only changes wi h a p- alue < 0.05 and a log
2
old
change (FC) o a leas 1. Analysis indica ed ha 136 p o eins
inc eased and 43 dec eased hei sumoyla ion le el by OGD in
p oli e a ing cells (Fig. 3A le ). We also compa ed OGD wi h ROG
and ound ha he g ea majo i y o p o eins ha inc eased hei
sumoyla ion unde OGD (87%), hen dec eased i a e ROG
(Fig. 3B le ). Gene on ology (GO) analysis o p o eins inc easing
hei sumoyla ion by OGD in p oli e a ing cells indica ed ha
p o eins g ouped in o unc ional ca ego ies mos ly ela ed o
ansc ip ion and RNA p ocessing, bu also o sumoyla ion and
ch oma in o ganiza ion (Fig. 3C uppe pa ). I we conside ed only
Fig. 1 Dynamics o p o ein sumoyla ion by SUMO2/3 unde OGD
and a e ROG. A P19 cells we e subjec ed o he indica ed imes o
OGD and o he indica ed imes o ROG a e 2.5 h o OGD, and cell
lysa es unde dena u ing condi ions we e p obed by wes e n blo
wi h an i SUMO2/3 an ibodies. Induc ion o HIF1αwas also
egis e ed as a con ol o OGD-associa ed hypoxia. α-TUBULIN was
e ealed as a loading con ol. Twen y mic og ams o o al p o ein
we e loaded pe lane. Th ee independen expe imen s we e
conduc ed and ep esen a i e wes e n blo s a e shown.
BSchema ic ep esen a ion o he wo kflow o OGD and ROG
imes in he nex expe imen s.
F. Galla do-Chamizo e al.
2
Cell Dea h Disco e y (2025) 11:230
hose p o eins ha a e inc easing hei sumoyla ion unde OGD
dec eased i by ROG (118 p o eins), GO ca ego ies we e simila
(Supplemen a y Fig. S3A). On he o he hand, om he 43 p o eins
dec easing hei sumoyla ion by OGD, only 23 (53%) eco e ed he
ini ial sumoyla ion s a e wi h ROG (Supplemen a y Fig. S3B).
In e es ingly, he 43 p o eins dec easing hei sumoyla ion by
OGD g ouped in o unc ional GO ca ego ies ela ed o cell cycle
egula ion, cell di e en ia ion/s emness, and DNA epai (Supple-
men a y Fig. S3C).
Simila p o eins a e modified in p oli e a ing and
di e en ia ed cells in esponse o OGD
Taking ad an age o he di e en ia ion abili y o P19 cells, we
aimed o compa e p o eins changing hei sumoyla ion s a e by
OGD-ROG in p oli e a ing and di e en ia ed cells. The mos
e ec i e me hod o P19 di e en ia ion is o ea hem wi h
e inoic acid (RA), which gi es ise o cells wi h a
neu oec ode mal-like pheno ype, exp essing se e al neu onal
ma ke s [20]. We compa ed p oli e a ing cells wi h cells di e -
en ia ed o 7 days. We checked o e ficien di e en ia ion by
down egula ion o he plu ipo ency ma ke OCT4 and induc ion o
he neu onal-specific ma ke ßIII-TUBULIN (Fig. 3D). A e p o ein
pu ifica ion (Fig. 2), MS and da a analysis, we de e mined ha only
56 p o eins we e inc easing hei sumoyla ion le el a e OGD (Fig.
3A igh and Supplemen a y Da ase S2). As o p oli e a ing cells,
mos o hem (73%), dec eased hei sumoyla ion by ROG (Fig. 3B
igh ), and again, simila ly o p oli e a ing cells, p o eins g ouped
o GO ca ego ies mos ly ela ed o ansc ip ion and RNA
p ocessing bu also o o eb ain de elopmen , suppo ing he
e ficiency o he di e en ia ion p o ocol (Fig. 3C lowe pa ).
Howe e , when compa ing he p o eins wi h inc eased sumoyla-
ion unde OGD in p oli e a ing and di e en ia ed cells, we
obse ed ha mos o hem we e he same p o eins, i.e., 45 o he
56 p o eins (80%) wi h inc eased sumoyla ion unde OGD in
di e en ia ed cells, also we e among he p o eins p esen ing
inc eased sumoyla ion by OGD in p oli e a ing cells (Fig. 3E). This
was also obse ed when compa ing OGD wi h ROG, 35 o 43
p o eins (81%) dec easing hei sumoyla ion by ROG in di e -
en ia ed cells we e in common wi h p oli e a ing cells (Supple-
men a y Fig. S3D). Thus, he p o eins showing al e ed sumoyla ion
by OGD-ROG in di e en ia ed cells seem o be simila o hose
iden ified in p oli e a ing cells.
SUMO2-modifica ion in esponse o OGD is confi med o
selec ed ansc ip ion ac o s and co ac o s
Since mos o he iden ified SUMO2 a ge s a e ela ed o
ansc ip ion we selec ed ansc ip ion-associa ed p o eins o
alida ion, based on medium-high log
2
FC alues and p e ious
iden ifica ion o a ge K esidues [21–27](Table1and
Supplemen a y Fig. S4A). We chose 6 di e en p o eins o
analysis, which included ansc ip ion ac o s and ansc ip ion
co egula o s. On one hand, we selec ed he ansc ip ion ac o s
OCT4 and SOX2, in ol ed in he plu ipo ency o s em cells bu
also in di e en ia ion, he sumoyla ion o which has been
al eady s udied [25–27]. On he o he hand, we selec ed a
a ie y o co egula o s, he sumoyla ion o which has been also
p e iously s udied. They included: NAB2, a co egula o o he
ansc ip ion ac o KROX20, which depends on
NAB2 sumoyla ion o i s de elopmen -associa ed oles, as we
ha e p e iously desc ibed [22]; he high mobili y g oup p o ein
HMG20B, he sumoyla ion o which we ha e p e iously
implica ed in neu onal di e en ia ion h ough he Co-REST
complex [21]; NCOR1, he ligand-independen co ep esso o
hy oid-ho mone and RA ecep o s [24]; and PIAS4, a SUMO-
specific E3 ligase bu also a co egula o o he STAT amily and
o he ansc ip ion ac o s [23,28]. Because OCT4, as expec ed,
and PIAS4 we e no de ec ed in di e en ia ed cells, we alida ed
hese p o eins, oge he wi h SOX2 and NAB2, in p oli e a ing
cells, and HMG20B and NCOR1 in di e en ia ed cells.
As obse ed in Fig. 4A, sumoyla ed OCT4 was e ficien ly pulled
down a e OGD bu no in con ol o ROG condi ions. Sumoyla ed
OCT4 was also ain ly isible in ex ac s, and in bo h cases
appea ed as mul iple bands, indica i e o se e al SUMO2
molecules modi ying OCT4 (Fig.4A). Sumoyla ed SOX2 also
appea ed exclusi ely associa ed wi h OGD, bu in con as o
OCT4 was obse ed as a single band (Fig.4A). NAB2 also appea ed
as se e al bands, bu we ha e p e iously demons a ed ha his is
due o he p esence o 2 independen sumoyla ion si es [22].
Again, sumoyla ed NAB2 was associa ed wi h OGD (Fig. 4A).
Con e sely, al hough p o eomic da a indica ed inc eased sumoy-
la ion o PIAS4 associa ed wi h OGD, we also de ec ed some
sumoyla ion unde con ol condi ions a e pull-down, which was
emo ed by ROG (Fig. 4A). In he case o HMG20B, we obse ed
Fig. 2 Pu ifica ion o His
10
-SUMO2 modified p o eins unde OGD
and a e ROG in p oli e a ing and di e en ia ed cells. P19 o #A5
cells, p oli e a ing o di e en ia ed in o neu ons o 7 days,
subjec ed o no (con ol, C) o 2.5 h o OGD, o o 2.5 h o ROG
a e 2.5 h o OGD, we e used o lysa e p epa a ion unde
dena u ing condi ions, ollowed by p o ein pu ifica ion wi h His-
bind ma ix o p ecipi a e p o eins modified by His
10
-SUMO2. To al
ex ac s, as well as pulled-down p o eins, we e p obed by wes e n
blo wi h an i-SUMO2/3 an ibodies, which in o al ex ac s de ec
p o eins modified by SUMO2/3 in P19 cells and p o eins modified by
a mix o SUMO2/3 and His
10
-SUMO2 in #A5 cells, while in pull-down
ac ions essen ially de ec His
10
-SUMO2-modified p o eins in #A5
cells. Pulled-down p o eins we e employed o mass spec ome y
analysis, in which samples om P19 cells we e used as a nega i e
con ol. Fo analysis, 3 and 4 independen eplica es we e p epa ed
o p oli e a ing and di e en ia ed cells, espec i ely. Selec ed
ep esen a i e wes e n blo s a e shown o each condi ion. Twen y
mic og ams o o al p o ein we e loaded pe lane in he case o o al
ex ac s. α-TUBULIN was egis e ed as a loading ma ke in o al
ex ac s.
F. Galla do-Chamizo e al.
3
Cell Dea h Disco e y (2025) 11:230
sumoyla ion unde all condi ions (Fig. 4B). Howe e , OGD cou sed
wi h inc eased HMG20B sumoyla ion compa ed wi h con ol o
ROG condi ions, which also co ela ed wi h dec eased le els o he
unmodified HMG20B p o ein unde OGD (Fig. 4B). Sumoyla ed
NCOR1 was again exclusi ely de ec ed unde OGD (Fig. 4B). Gi en
he high molecula weigh o NCOR1, which has been epo ed o
con ain se e al sumoyla ion si es [24], we canno dis inguish i
sumoyla ed NCOR1 co esponds o mul iple bands, coming ei he
om mul iple sumoyla ion si es as om poly-SUMO2 chains,
al hough his is p obably he case.
OCT4 is a SENP7 a ge o desumoyla ion a e es o ing
oxygen and glucose
I has been epo ed ha mu a ion o a single K esidue in OCT4
e ficien ly abolishes OCT4 sumoyla ion [26,27], and as al eady
indica ed, OCT4, in pa icula , is de ec ed in gels as se e al
mig a ion p oduc s when i is sumoyla ed unde OGD (Fig. 4A).
These obse a ions, in ag eemen wi h he abili y o SUMO2 o
o m poly-SUMO chains, s ongly sugges ha he se e al
mig a ion p oduc s obse ed co espond o di e en leng hs o
a single poly-SUMO2 chain. Because SENP7 has been specifically
F. Galla do-Chamizo e al.
4
Cell Dea h Disco e y (2025) 11:230
implica ed in edi ing poly-SUMO2/3 chains [29], and we ha e
p e iously desc ibed ha SENP7 is induced by ROG a e being
deple ed unde OGD [30], we wonde ed whe he sumoyla ed
OCT4 can be a SENP7 a ge o desumoyla ion. To add ess his
ques ion, we decided o conduc gain- and loss-o - unc ion
expe imen s. We fi s o e exp essed a Flag- agged e sion o
SENP7 in cells subjec ed o OGD, which esul ed in he
disappea ance o sumoyla ed o ms ha we e de ec ed unde
OGD wi h he emp y ec o (Fig. 5A). In cells no subjec ed o OGD
(con ol condi ions), o e exp essing SENP7 had no consequences
on OCT4 sumoyla ion, as he endogenous p o ein is exp essed
and ende s OCT4 desumoyla ed (Fig. 5A). Nex , we knocked
down he endogenous SENP7 by using a specific shRNA molecule.
As obse ed in Fig. 5B, his molecule did no a ec sumoyla ed
OCT4 unde OGD, because unde hese condi ions SENP7 is no
exp essed. Howe e , unde ROG, when SENP7 is es o ed, e ficien
OCT4 desumoyla ion is obse ed when exp essing a con ol
shRNA molecule, bu no when SENP7 is knocked down by he
specific shRNA (Fig. 5B). Thus, sumoyla ed OCT4 is e ealed as a
a ge o SENP7 unde ROG condi ions.
Al e ed sumoyla ion o selec ed ac o s has an impac on cell
iabili y ollowing ha m ul OGD
To ge insigh in o how he sumoyla ion o he iden ified ac o s
impac s he esponse o cells o OGD, we decided o e alua e cell
iabili y a e ha m ul (20 h) OGD, unde o e exp ession condi ions o
wild- ype (WT) o sumoyla ion mu an e sions o selec ed p o eins.
To his end, we chose OCT4, SOX2, NAB2, and PIAS4, and gene a ed
sumoyla ion mu an e sions by mu a ing he p e iously iden ified
a ge K esidues o R (KR) (Supplemen a y Fig. S4A). We also checked
o simila exp ession le els o WT and KR e sions a e ans ec ion
o he co esponding cons uc s (Supplemen a y Fig. S4B).
OGD-media ed cy o oxici y was de e mined by Annexin V/
p opidium iodide (PI) labeling/inco po a ion, which eached alues
o 47% and 48%, espec i ely, in con ol condi ions using he emp y
ec o , in compa ison wi h a ound 12% de ec ed a e ans ec ion
o he emp y ec o in he absence o OGD (Supplemen a y Fig.
S4C). A e subjec ing he cells o ha m ul OGD, we obse ed ha
o e exp essing an OCT4 mu an unable o be sumoyla ed esul ed
in lowe oxici y compa ed wi h con ol condi ions (emp y ec o ).
A he same ime, WT o e exp ession had no e ec (Fig. 6).
Con e sely, o e exp ession o WT SOX2 esul ed in a s ong
dec ease o cell oxici y, while mu an SOX2 sligh ly educed
oxici y, al hough no s a is ically significan (Fig. 6). Rega ding
NAB2, he WT e sion had no e ec . In con as , o e exp ession o
he sumoyla ion mu an esul ed in inc eased cy o oxici y (Fig. 6).
Finally, PIAS4 beha ed qui e simila ly o SOX2 since o e exp ession
o he WT e sion educed oxici y, while he sumoyla ion mu an
had no impac on cell iabili y (Fig. 6). The e o e, al e ed
sumoyla ion o selec ed ac o s di e en ially impac s cell iabili y
when o e exp essed. No ably, o all he p o eins analyzed, he
Fig. 3 P o eins unde going inc eased sumoyla ion unde OGD a e ela ed o ansc ip ion and RNA p ocessing. A Volcano plo s o
p o eins changing hei sumoyla ion s a e in esponse o OGD as de e mined by mass spec ome y analysis in p oli e a ing and di e en ia ed
cells. Those p o eins ou o he es ablished p- alue and old change (FC) cu o s a e ep esen ed in g ay. P o eins unde going desumoyla ion
a e ep esen ed in ed, while hose inc easing hei sumoyla ion a e ep esen ed in g een. Selec ed p o eins a e highligh ed in iole . Numbe s
ep esen he numbe o p o eins in each ca ego y. BO e lapping be ween p o eins unde going inc easing sumoyla ion in esponse o OGD
(OGD s. con ol, C l) and hose unde going desumoyla ion a e ROG (OGD s. ROG) in p oli e a ing and di e en ia ed cells has been
ep esen ed by Venn diag ams. CGene on ology (GO) analysis o p o eins inc easing sumoyla ion in esponse o OGD in p oli e a ing and
di e en ia ed cells has been ep esen ed by bubble g aphics. Bubble size ep esen s he numbe o p o eins in each ca ego y, also indica ed
nex o each bubble. p- alue cu o s o 5 × 10
−6
and 2 × 10
−3
we e es ablished o p oli e a ing and di e en ia ed cells, espec i ely. DWes e n
blo o checking o p oli e a ing (0 days) and di e en ia ed (7 days) cells by analysis o he plu ipo ency and di e en ia ion ma ke s OCT4 and
ßIII-TUBULIN, espec i ely; α-TUBULIN was egis e ed as a loading ma ke . EO e lapping be ween p o eins unde going inc easing
sumoyla ion in esponse o OGD in p oli e a ing (p ol.) and di e en ia ed (di .) cells, ep esen ed by Venn diag ams. B,ENumbe s on he op
indica e he o al numbe o p o eins in each condi ion. En ichmen (en ich.) o he o e lapping, oge he wi h he associa ed p- alue, as
de e mined by he hype geome ic es , is indica ed below.
Table 1. Top- anking sumoyla ed p o eins.
OGD*
ROG
OGD
ROG
QSER1
10,48
10,01
n.d.
n.d.
NCOR1
8,35
8,44
4,83
4,48
NCOA2
7,89
6,85
4,33
4,15
NOSIP
7,20
6,25
3,49
2,18
EMSY
7,11
6,04
n.d.
n.d.
ELMSAN1
6,45
5,60
n.d.
n.d.
HNRNPA0
6,09
6,33
n.d.
n.d.
MKL1
6,04
6,47
3,01
2,81
ATF1
6,00
7,27
n.d.
n.d.
RBM10
5,87
5,68
1,54
1,63
TCERG1
5,72
6,05
2,27
1,72
FOXK2
5,64
5,51
4,43
2,95
QRICH1
5,53
5,64
2,90
2,03
RBM14
5,52
4,78
1,93
0,73
ESF1
5,45
5,34
2,14
1,47
RBM6
5,08
3,93
2,06
1,22
HNRNPM
5,08
4,15
2,90
1,44
SOX2
5,07
5,45
4,23
3,83
HMCES
5,07
4,81
2,43
2,12
MKL2
5,05
5,69
3,81
3,33
PIAS4
5,02
4,37
n.d.
n.d.
NAB2
4,99
5,73
5,43
3,48
TNRC18
4,93
4,52
2,31
2,85
ERF
4,87
4,87
n.d.
n.d.
SP1
4,86
3,53
n.d.
n.d.
MAML1
4,84
4,62
1,48
1,48
MYEF2
4,74
4,22
3,87
2,39
TRERF1
4,74
4,76
n.d.
n.d.
CDCA2
4,71
3,98
2,06
2,89
NCOA3
4,59
4,54
n.d.
n.d.
PHC1
4,56
4,88
3,72
2,94
HMG20B
4,51
3,91
3,34
3,53
POU5F1
4,51
3,61
n.d.
n.d.
n.d. no de ec ed.
a
Log
2
FC alues o op- anking p o eins sumoyla ed unde OGD (OGD s
con ol), and desumoyla ed a e ROG (OGD s ROG), in p oli e a ing cells.
Values o di e en ia ed cells a e also included. Valida ed p o eins a e
shaded in g ay. POU5F1, OCT4.
F. Galla do-Chamizo e al.
5
Cell Dea h Disco e y (2025) 11:230
e ec s caused by he WT e sions we e significan ly di e en om
hose caused by he sumoyla ion mu an s.
DISCUSSION
Oxygen and glucose a e essen ial o all cell ypes o he di e en
issues ha make up he human body. Tissue a chi ec u e and
physiology may cause some cells o be na u ally subjec ed o
limi ed condi ions o hese nu ien s. Howe e , se e e dep i a ion
o hem has d as ic consequences on cell iabili y. Damage depends
no only on he le el o dep i a ion bu also on he du a ion o he
insul , and cells need o accu a ely measu e i o make he decision
o su i ing o inducing cell dea h o a oid comp omising he
co ec unc ioning o issues o o gans. Cance cells inside solid
umo s ha e adap ed o OGD condi ions, which gi es hem
selec i e ad an ages. In all hese scena ios, obus and coo dina ed
sensing and signaling mechanisms a e equi ed o p o ide he
app op ia e esponse o success ully cope wi h s ess condi ions.
Almos wo decades ago i was obse ed o he fi s ime a
co ela ion be ween a d as ic educ ion in blood flow and inc eased
p o ein sumoyla ion in squi els’b ains du ing hibe na ion o po
[13], sugges ing ha p o ein sumoyla ion displays a p o ec i e ole.
Since hen, se e al wo ks ha e expe imen ally co obo a ed his
hypo hesis in cul u ed cells and animal models, showing ha
enhanced p o ein sumoyla ion p o ec s cells agains ischemia
( e iewed in [9]). Mos o hese wo ks ha e shown ha massi e
and indisc imina e enhancemen o sumoyla ion esul s in sub-
s an ial cell p o ec ion. Howe e , a non-selec i e inc ease in
sumoyla ion p obably limi s be e ou comes and educes he
e ec i eness o a pu a i e he apeu ic applica ion. This highligh s
he need and ele ance o iden i ying specific SUMO a ge s playing
pa icula oles in cell su i al in esponse o OGD condi ions, o
selec i ely ac on hem o imp o e cell su i al ou comes. Indeed, i
is no only impo an o iden i y SUMO a ge s in esponse o OGD
bu also o p ecisely in es iga e he e ec s o hei inc eased
sumoyla ion since we canno exclude he possibili y ha inc eased
sumoyla ion o ce ain a ge s is de imen al o cell p o ec ion.
In his wo k, we ha e iden ified a o al o 191 di e en p o eins
changing hei sumoyla ion s a e in esponse o OGD. The
majo i y o hem (77%) expe ienced inc eased sumoyla ion, and
o hese, 85% we e subsequen ly desumoyla ed in a ew hou s
when es o ing no mal g ow h condi ions, suppo ing ha
sumoyla ion is a highly dynamic p ocess. Genes coding o hese
p o eins mainly g oup o GO ca ego ies ela ed o ansc ip ion
and RNA splicing, indica ing ha p o eins in ol ed in gene
exp ession a e immedia e SUMO a ge s in esponse o OGD.
Fig. 4 Valida ion o he sumoyla ion o selec ed a ge s by SUMO2 unde OGD and ROG. Valida ion o p o eomic esul s by selec ion o
se e al p o eins unde going inc eased sumoyla ion unde OGD and dec eased sumoyla ion a e ROG, as indica ed by mass spec ome y, o
es hem by wes e n blo wi h specific an ibodies agains he di e en p o eins, be o e (ex ac ) and a e p ecipi a ion o His
10
-SUMO2 wi h
His-bind Ma ix (pull-down), in p oli e a ing (A) and di e en ia ed (B) cells. Twen y mic og ams o o al p o ein we e loaded pe lane in he
case o ex ac s. α-TUBULIN (p esen in ex ac s bu no a e pull-down) was egis e ed as a loading ma ke . Black a owheads indica e
unmodified p o eins while whi e a owheads indica e sumoyla ed p oduc s. C con ol condi ions, O OGD, R ROG. *Unspecific band.
Fig. 5 OCT4 is desumoyla ed by SENP7 when es o ing oxygen and
glucose. A E ec o SENP7 o e exp ession on OCT4 sumoyla ion in
esponse o OGD was es ed by ans ec ion o a Flag- agged SENP7
exp ession cons uc (S7) o he emp y ec o (–), unde no mal
g ow h (con ol) o OGD condi ions. O e exp essed SENP7 was
e ealed wi h an i-Flag an ibodies. BE ec o SENP7 knockdown on
OCT4 sumoyla ion in esponse o ROG a e OGD was es ed by
ans ec ion o a shRNA cons uc agains SENP7 (shS7) o a con ol
shRNA (shC), unde OGD, o a e ROG ollowing OGD. Endogenous
SENP7 was e ealed wi h an i-SENP7 an ibodies. A,BTwen y
mic og ams o o al p o ein we e loaded pe lane. α-TUBULIN was
analyzed as a loading ma ke . High and low exposu es (expos.) o
he wes e n blo signal o OCT4 a e shown o be e app ecia ion
o unmodified and modified o ms.
F. Galla do-Chamizo e al.
6
Cell Dea h Disco e y (2025) 11:230
Al hough p e ious p o eomic s udies ela ed o OGD o hypoxia
condi ions ha e iden ified a mo e limi ed numbe o p o eins
unde going changes in sumoyla ion, and he expe imen al
app oaches employed di e in some aspec s om hose used in
ou wo k, hey ha e also iden ified SUMO a ge s in ol ed in he
con ol o gene exp ession [15,16,31]. Since SUMO has been
no ably associa ed wi h ansc ip ion ep ession [32], we can
specula e ha inc eased sumoyla ion o ansc ip ion- ela ed
p o eins in esponse o OGD is a way o ep ess o diminish
ansc ip ion in he cell unde hese condi ions. This ag ees wi h
he p e iously epo ed p edominan gene down egula ion a e
4 h o OGD in p ima y neu on cul u es [33]. Also, conce ning his,
among p o eins being sumoyla ed in esponse o OGD, we ha e
iden ified he NAB2 co ep esso , which has also been iden ified as
a SUMO a ge in esponse o OGD o hypoxia in p e ious
p o eomic s udies [16,31], and which we ha e p e iously epo ed
o depend on sumoyla ion o display ep esso ac i i y du ing
hindb ain de elopmen [22]. Addi ionally, we iden ified PIAS4 as a
SUMO2 a ge unde OGD, acco ding o a p e ious epo o
a ious PIAS p o eins inc easing hei sumoyla ion in esponse o
OGD [16]. O no e, i would be in e es ing o iden i y SUMO a ge s
in he absence o glucose alone and compa e hem wi h hose
om OGD s udies and hose p e iously iden ified unde hypoxia
[31].
Rema kably, despi e he lowe numbe o p o eins iden ified in
di e en ia ed cells, no g ea di e ences we e obse ed be ween
he OGD-associa ed SUMO2 p o eome o p oli e a ing and di e -
en ia ed cells. This indica es ha SUMO2 a ge s in esponse o OGD
a e p obably gene al ac o s egula ing common cellula aspec s
like me abolism, g ow h, and di ision, and no cell ype-specific
p ocesses. This ag ees wi h he ac ha OGD is ha m ul o all cell
ypes and issues, and i p obably igge s he same signaling
pa hways in all o hem o main ain cell iabili y and ac i a e
al e na i e ways o ob aining ene gy. On he o he hand, in
p oli e a ing cells, we ha e also obse ed a mino i y g oup o
p o eins unde going desumoyla ion in esponse o OGD. These
p o eins a e g ouped in GO ca ego ies ela ed o cell cycle, DNA
epai , and di e en ia ion/s emness. I is known ha SUMO plays
ele an oles in cell cycle p og ession and he main enance o
plu ipo ency ( e iewed in [1]), being pa icula ly equi ed o mi osis
[34]. Thus, he desumoyla ion o key ac o s in ol ed in hese
p ocesses can esul in hei in e up ion. O pa icula in e es a e
p o eins SMARCE1, SMARCC1, PHF10, PBRM1, and BRD7 since hey
o m pa o he SWI/SNF complex PBAF [35]. SWI/SNF complexes
a e well-known ch oma in emodeling complexes in ol ed in many
p ocesses, including cell g ow h, di ision, and di e en ia ion. Mo e
p ecisely, PBAF is no ably in ol ed in egula ing cell di e en ia ion
and p obably cell- ype iden i y, being also equi ed o he
main enance o genomic in eg i y du ing mi osis ( e iewed in
[36]). PBAF is especially ele an o ne ous sys em de elopmen ,
bo h o he main enance o neu al p ogeni o s and o he
di e en ia ion o pos -mi o ic cells, and he sumoyla ion o PHF10
and PBRM1 subuni s, in pa icula , has been linked o complex
unc ionali y [37,38]. I has been p oposed ha a majo ole o
SUMO is o acili a e complex assembly due o he simul aneous
p esence o sumoyla ion si es and non-co alen SUMO in e ac ing
mo i es (SIMs) in a ious subuni s o a complex, which unde go
p o ein-g oup sumoyla ion [39]. This is especially e iden o
ch oma in ep esso complexes [32], in which he absence o
sumoyla ion may des abilize subuni in e ac ions and complex
a chi ec u e. Con e sely, sumoyla ion may also des abilize he
in e ac ion o ce ain p o eins. Fo ins ance, wo o he p o eins ha
we ha e cha ac e ized he e, OCT4 and SOX2, which ope a e as a
he e odime o sus ain Nanog exp ession o plu ipo ency main-
enance in s em cells, ha e been desc ibed o disassemble when
sumoyla ed [40]. Since we ha e obse ed inc eased sumoyla ion o
bo h p o eins unde OGD, we can an icipa e he e odime
disassembly unde hese condi ions.
OCT4 and SOX2 ha e been analyzed, oge he wi h NAB2 and
PIAS4, o hei impac on cell su i al when o e exp essing WT
o sumoyla ion de ec i e (KR) e sions. Ou analysis has
demons a ed a a ie y o consequences depending on he
p o ein. In gene al, o e exp ession o WT e sions ei he has no
e ec o is beneficial o cell su i al. The beneficial e ec o
o e exp essing PIAS4 is ab oga ed i i s sumoyla ion is impeded,
indica ing ha PIAS4-media ed su i al depends on i s sumoyla-
ion. SOX2 beha es simila ly, al hough o e exp ession o i s KR
mu an has also a posi i e impac on cell su i al, indica ing ha
he beneficial e ec o SOX2 is in pa independen o
sumoyla ion. Su p isingly, OCT4 beha es di e en ly han
SOX2 since i is beneficial only when sumoyla ion is impeded,
sugges ing ha sumoyla ion blocks i s capaci y o p omo e cell
su i al and ha in e e ing wi h i could be a o able o he
cells. Sumoyla ion has been epo ed o di e en ially impac he
ansc ip ion ac i i y o OCT4 and SOX2 [26,41], which is
exemplified in he con ol o Nanog exp ession [40]. This is no
unexpec ed, since, despi e a join ole as a he e odime in s em
cells, hese ac o s also ac independen ly du ing de elopmen
[42]. Finally, o e exp ession o WT NAB2 does no a ec cell
su i al bu su p isingly impedingi ssumoyla ionisde imen al
Fig. 6 Al e ed sumoyla ion o selec ed ac o s di e en ially impac s cell su i al a e OGD. The pe cen age o cy o oxici y was es ima ed
in cells subjec ed o ha m ul OGD by p opidium iodide (PI) inco po a ion and Annexin V labeling, a e ans ec ion o he indica ed
exp ession cons uc s o he indica ed p o eins: emp y ec o (con ol, −), wild ype e sion (WT), o sumoyla ion mu an (KR). Values a e
means ± s.d. o h ee independen de e mina ions. The s a is ical significance o he di e ences in cy o oxici y was analyzed by one-way
ANOVA (p< 0.05) ollowed by Tukey’s pos - es . Di e ences wi h he co esponding con ol a e indica ed on op o he ba s. O he di e ences
be ween g oups o samples a e indica ed wi h lines: *p< 0.05, **p< 0.01, ***p< 0.001.
F. Galla do-Chamizo e al.
7
Cell Dea h Disco e y (2025) 11:230
o i , indica ing ha main aining i s sumoyla ion capaci y is
necessa y o cell p o ec ion. Then, hese esul s suppo he
a o emen ioned need o unc ionally analyze each SUMO a ge
independen ly o i s ole in cell iabili y. As commen ed,
al hough he gene al ou come o indisc imina e inc eased
sumoyla ion unde OGD is posi i e in e ms o cell su i al,
only a de ailed analysis o each sumoyla ed a ge will p o ide
in o ma ion on he bes combina ion o sumoyla ed and non-
sumoyla ed o ms o di e en p o eins o significan ly inc ease
p o ec ion in ha m ul condi ions and exploi i he apeu ically
(Fig. 7). Hence, he apies aimed a p omo ing he sumoyla ion o
PIAS4, NAB2 and SOX2 while p omo ing he desumoyla ion o
OCT4, h ough d ug-media ed inhibi ion o specific p o eases
and ligases, may esul in be e su i al ou comes. I has been
shown ha he g a in ischemic s oke o neu al s em cells
displaying enhanced sumoyla ion due o UBC9 o e exp ession
leads o inc eased su i al and di e en ia ion [43]. Simila ly, cells
o e exp essing WT PIAS4, NAB2, and SOX2, bu mu an OCT4,
could also show highe su i al a es and, he e o e, be o
he apeu ic in e es o ischemia.
As SUMO2/3 is he majo SUMO o m esponding o s ess
condi ions and can o m poly-SUMO chains [44], we hypo he-
sized and ha e demons a ed ha one o ou poly-sumoyla ed
a ge s unde OGD, OCT4, is he subs a e o desumoyla ion
a e ROG by he SUMO p o eases SENP7, specialized in edi ing
poly-SUMO2/3 chains [29]. We ha e p e iously implica ed SENP7
in p omo ing cell su i al a e ha m ul OGD in umo cells [30],
showing an opposi e e ec o he SENP3 p o ease p omo ing
cell dea h [45,46]. Besides, SENP1 has also been in ol ed in
neu op o ec ion du ing ansien b ain ischemia/ epe usion
[47]. In e es ingly, bo h SENP1 and SENP3 ha e been demon-
s a ed o be ca aly ically inac i a ed by hypoxia in a apid and
e e sible manne [48], while we ha e shown e e sible down-
egula ion o SENP7 du ing OGD [30]. All his illus a es he
impo ance o SENP p o eases in he sensing and signaling o
hypoxia o OGD condi ions, which p obably pa icipa e in
measu ing he dep h o he damage and con ibu e o making
he co ec decision be ween su i ing o inducing cell dea h.
Thus, i is p obably no he inc eased sumoyla ion o p o eins
pe se unde OGD ha associa es wi h cell p o ec ion bu he
subsequen desumoyla ion a e ROG by specific SENP p o-
eases, which pa icipa e in signaling o elabo a e he app o-
p ia e esponse o ha m ul condi ions in e ms o cell su i al.
MATERIALS AND METHODS
Cell cul u e, ans ec ion, and gene a ion o s able cell lines
P19 cells we e di ec ly ob ained om ATCC (LGC S anda ds, Ba celona,
Spain) as au hen ica ed. They we e cul u ed in Dulbecco’s modified Eagle’s
medium (DMEM) (Sigma-Ald ich, S . Louis, MO, USA) supplemen ed wi h
10% e al bo ine se um (FBS, Sigma-Ald ich), and 10 ml/l o an an ibio ic
solu ion wi h Penicillin (100 U/ml) and S ep omycin (10 mg/ml) (Sigma-
Ald ich). Cells we e es ed o he absence o mycoplasma. T ans ec ion
was pe o med wi h Lipo ec amine 2000 (In i ogen, Li e Technologies,
Paisley, UK), and cells we e ha es ed o subjec ed o u he analysis 24 h
la e . Fo knockdown wi h shRNA-exp essing plasmids, cells we e
subjec ed o u he analysis a e 72 h o ans ec ion. Fo s able cell
lines, he in eg a ion o plasmid pAV1219 [49], d i ing he exp ession o
His
10
-SUMO2, was selec ed in 4 µg/ml pu omycin o 10–14 days o he
appea ance o esis an isola ed colonies. Neu onal di e en ia ion was
induced by RA a 0.5 µM in no mal cul u e medium con aining 5% FBS
ins ead o 10% FBS o 4 days (wi h a eplacemen a day 2) in non-
adhe en pla es, ollowed by 3 addi ional days in no mal medium in
adhe en pla es in he absence o RA. OGD condi ions we e ob ained by
incuba ing he cells a he indica ed imes in o e nigh equilib a ed no mal
medium, bu based on DMEM lacking glucose (Sigma-Ald ich), in HEPA
Class 100 incuba o s (The mo Fishe Scien ific, Wal ham, MA, USA) wi h 1%
oxygen. Ha m ul OGD o cy o oxici y expe imen s was ob ained by
subjec ing cells o OGD o 20 h.
Wes e n blo
Fo wes e n blo unde dena u ing condi ions, cells we e lysed in 8 M u ea,
10 mM T is-HCl pH 8.0 bu e . Lysis bu e o non-dena u ing condi ions
consis ed o 50 mM T is-HCl pH 7.5, 150 mM NaCl, 1% T i on X-100,
supplemen ed wi h he EDTA-con aining comple e p o ease inhibi o
cock ail (Sigma-Ald ich). The B ad o d eac i e assay (Bio-Rad, He cules, CA,
USA) was used o p o ein quan ifica ion. P o eins we e sepa a ed in SDS-
con aining polyac ylamide gels and hen ans e ed o PVDF memb anes
(GE Heal hca e, Buckinghamshi e, UK) o an ibody hyb idiza ion. Mem-
b anes we e subsequen ly p ocessed wi h he chemiluminescence ECL
sys em (Bio-Rad) and examined in a ChemiDoc XRS appa a us (Bio-Rad).
An ibodies used we e as ollows: mouse an i-Flag (1:2000, M2, #F1804,
Sigma-Ald ich), mouse an i-α-TUBULIN (1:10000, DM1A, #T9026, Sigma-
Ald ich), mouse an i-SUMO2/3 (1:2000, 8A2, #ab81371, Abcam, Camb idge,
UK), mouse an i-HIF1A (1:5000, #MAB1536, R&D Sys ems, Minneapolis, MN,
USA), mouse an i-His (1:3000, #27471001, GE Heal hca e), mouse an i-NAB2
(1:1000, 1C4, #sc-23867, San a C uz B g., Dallas, TX, USA), abbi an i-OCT4
(1:1000, H-134, #sc-9081, San a C uz B g.), abbi an i-PIAS4 (1:500, #PA5-
20954, The mo Fishe Scien ific), abbi an i-SOX2 (1:1000, EPR3131,
#ab92494, Abcam), mouse an i-HMG20B (1:500, 4.21, #sc-53123, San a
C uz B g.), goa an i-NCOR1 (1:500, C-20, #sc-1609, San a C uz B g.), abbi
an i-SENP7 (1:1000, #ab187126, Abcam), goa an i-mouse HRP (1:10000,
#A4416, Sigma-Ald ich), goa an i- abbi HRP (1:10000, #A6154, Sigma-
Ald ich), donkey an i-goa HRP (1:10000, #A50-201P, Be hyl Labo a o ies
Inc., Mon gome y, TX, USA). Unc opped images o wes e n blo s a e
p o ided in a Supplemen a y File.
Pu ifica ion o His
10
-SUMO2 conjuga es and mass
spec ome y sample p epa a ion
His
10
-SUMO2 conjuga es we e pu ified as in [50]. Cell pelle s om 3
p oli e a ing and 4 di e en ia ed cells eplica es we e lysed in 10 pelle
olumes o guanidine lysis bu e (6 M guanidine-HCl, 100 mM sodium
phospha e, and 10 mM T is, bu e ed a pH 8.0) and sonica ed o
p omo e DNA agmen a ion. Nex , lysa es we e supplemen ed wi h
Fig. 7 Schema ic ep esen a ion o sumoyla ion-media ed p o ec-
ion o P19 cells. The ca oon shows inc eased sumoyla ion o
p o eins du ing OGD and subsequen desumoyla ion a e ROG in
p oli e a ing and di e en ia ed P19 cells (RA- ea ed). Key p o eins,
he sumoyla ion o which is impo an o su i al ou comes, a e
depic ed. Less SUMO a ge s we e de ec ed in di e en ia ed cells,
bu hey we e simila o hose de ec ed in p oli e a ing cells. In he
la e , he ne balance in e ms o cell p o ec ion was posi i e (+)
e en hough modifica ion o some ac o s is un a o able o su i al
(–).
F. Galla do-Chamizo e al.
8
Cell Dea h Disco e y (2025) 11:230
imidazole o 50 mM and β-me cap oe hanol o 5 mM. Twen y mic oli es
(d y olume) o p e-washed Ni-NTA aga ose beads (QIAGEN, Aus in, TX,
USA) we e added pe 1 ml lysa e and incuba ed o e nigh a 4 °C while
o a ing. Nex , beads we e washed o 10 min wi h 1 ml o he ollowing
wash bu e s: wash bu e 1 (6 M guanidine-HCl, 0.1% T i on X-100,
10 mM imidazole, 5 mM β-me cap oe hanol, 100 mM sodium phospha e,
and 10 mM T is, pH 8.0), wash bu e 2 (8 M u ea, 0.1% T i on X-100,
10 mM imidazole, 5 mM β-me cap oe hanol, 100 mM sodium phospha e,
10 mM T is, pH 8.0), wash bu e 3 (8 M u ea, 10 mM imidazole, 5 mM
β-me cap oe hanol, 100 mM sodium phospha e, 10 mM T is, pH 6.3),
wash bu e 4 (8 M u ea, 5 mM β-me cap oe hanol, 100 mM sodium
phospha e, 10 mM T is, pH 6.3), wo imes. A e washing, His
10
-SUMO2
conjuga es we e elu ed o 30 min wi h one bead olume o elu ion
bu e (7 M u ea, 500 mM imidazole, 100 mM sodium phospha e, and
10 mM T is, bu e ed a pH 7.0). The elu ion p ocedu e was epea ed
ano he wo imes, and all elua es we e pooled and passed h ough
0.45-μMfil e s (Ul a ee, Me ck (Millipo e), Da ms ad , Ge many). In he
case o he samples co esponding o di e en ia ed cells, samples we e
hea ed a 95 °C o 5 min. Nex , samples passed h ough a educ ion-
alkyla ion p ocedu e a oom empe a u e consis ing o 1 mM di hio-
h ei ol (DTT) o 30 min, 5 mM chlo oace amide o 30 min, and 6 mM
DTT o ano he 30 min. Subsequen ly, samples we e dilu ed 4 imes
wi h 50 mM ammonium bica bona e, 250 ng T ypsin (V-5111, P omega,
Madison WI, USA) added, and incuba ed o e nigh s ill and in he da k.
The esul ing pep ides we e pu ified using C-18 in-house-assembled
S ageTips [51], Speed Vac, and esuspended in 10 µl o 0.1% o mic acid
in wa e .
Mass spec ome y da a acquisi ion
Mass spec ome y analysis was pe o med on an EASY-nLC 1000 sys em
(P oxeon, Odense, Denma k) connec ed o a Q-Exac i e O bi ap
(The mo Fishe Scien ific) h ough a nano-elec osp ay ion sou ce. The
ch oma og aphy g adien was pe o med h ough e e se phase in a
15 cm in-house packed column wi h 1.9 μm C18-AQ beads (Rep osphe -
DE,Pu ,D .Manish,Amme buch-En ingen,Ge many).Theg adien
leng h was o 95 min om 2% o 30% ace oni ile in 0.1% o mic acid,
ollowed by column e-equilib a ion a a 200 nl/min flow a e. A Top10
da a-dependen acquisi ion mode was applied. The scan ange was
400–2000 m/z. Full-scan MS spec a we e acqui ed a a a ge alue o
3×10
6
, a esolu ion o 70,000, and a maximum injec ion ime o 20 ms.
The highe -collisional dissocia ion andem mass spec a (MS/MS) we e
eco ded a a a ge alue o 1 × 10
5
wi h a esolu ion o 17,500, an
isola ion window o 2.2 m/z, a no malized collision ene gy (NCE) o 25,
and a maximum injec ion ime o 60 ms. A dynamic exclusion window o
60 s was conside ed, and ions wi h cha ge 1 and >6 we e excluded om
igge ing MS2 analysis.
Mass spec ome y da a analysis
Mass spec ome y RAW da a was analyzed using MaxQuan ( . 1.5.5.1)
acco ding o [52], wi h s anda d se ings wi h he ollowing modifica-
ions: Th ee missed clea ages we e allowed o T ypsin/P in p oli e a ing,
and ou missed o A gC in di e en ia ed cells om an in-silico
p edic ed Mus musculus p o eome (Unip o 10 h Oc 2016). In
di e en ia ed cells, Ca bamyl (K) was added as a a iable modifica ion.
Label- ee quan ifica ion (LFQ) was enabled, bu no enabled, and he
maximum pep ide mass was inc eased o 5000. Ma ch-be ween- uns
was enabled. S a is ical analysis o MaxQuan ou pu was pe o med in
he Pe seus compu a ional pla o m [53]. LFQ alues we e log
2
ans o med, and Con aminan s, Re e se, and Only iden ified by si e
p o eins we e emo ed, as p o eins no consis en ly iden ified in all
eplica es in a leas one condi ion as well. Missing alues we e andomly
impu ed acco ding o s anda d se ings and 2- ailed - es s pe o med o
compa e be ween condi ions. Tables we e expo ed in MS Excel o
comp ehensi e b owsing o he da a (Supplemen a y Da ase s
S1 and S2). The mass spec ome y p o eomics da a ha e been
deposi ed o he P o eomeXchange Conso ium ia he PRIDE [54]
pa ne eposi o y wi h he da ase iden ifie PXD056069.
Plasmids and shRNA molecules
Plasmid pAV1219 was cons uc ed by cloning a His
10
-SUMO2 cDNA
agmen ob ained by PCR amplifica ion in he Ps I-XhoIsi eso pLV-
IRES-Pu o [49]. Exp ession cons uc s we e based on he pAdRSV-Sp
plasmid, and hose o SENP7, PIAS4, NAB2, and NAB2-KR ha e been
p e iously desc ibed [22,30]. Fo OCT4 and SOX2 exp ession cons uc s,
he comple e cDNAs (ATG o STOP) o Pou5 1 (Oc 4)andSox2 we e
ob ained by PCR amplifica ion om e o ansc ibed (High-Capaci y
cDNA Re e se T ansc ip ion Ki , Applied Biosys ems, Ca lsbad, CA, USA)
o al RNA isola ed om P19 cells (RNeasy ki , QIAGEN), and cloned
be ween he NheIandNsiI si es o pAdRSV-Sp wi h a Flag epi ope ag a
he N- e minus. De ec i e sumoyla ion mu an s o PIAS4, OCT4, and
SOX2 we e gene a ed by s anda d PCR echniques o subs i u e a ge K
esidues (Supplemen a y Fig. S4A) by R esidues. shRNA cons uc s we e
based on he pSupe plasmid (OligoEngine, Sea le, WA, USA), and bo h
con ol shRNA and shRNA agains Senp7 ha e been p e iously desc ibed
[55].
Cy o oxici y measu emen
Cy o oxici y in cells a e ha m ul OGD was e alua ed by de e mina ion o
Anexin V/PI labeling/inco po a ion by flow cy ome y. To his end, we used
he FITC Annexin V Apop osis De ec ion Ki wi h PI (Immunos ep,
Salamanca, Spain), acco ding o he manu ac u e ’s ins uc ions. A e
sample p ocessing, hey we e analyzed in a BD FACSCalibu flow
cy ome e appa a us (BD Biosciences, San Jose, CA, USA).
S a is ical and addi ional analyses
S a is ical analyses o cy o oxici y expe imen s we e pe o med wi h he
so wa e P ism 9.5.1 (G aphPad). G aphics show mean alues ± s.d. om 3
independen de e mina ions. We used one-way ANOVA (p< 0.05) ollowed
by he Tukey pos - es o mul iple compa isons o s a is ical analysis o
he di e en g oups (*p< 0.05, **p< 0.01, ***p< 0.001). No mal dis ibu ion
and simila a iances we e assumed. The significance o o e lapping in
Venn diag ams was analyzed h ough he hype geome ic es a h ps://
sys ems.c ump.ucla.edu/hype geome ic/index.php. Venny 2.1 a h p://
bioin ogp.cnb.csic.es/ ools/ enny/index.h ml was used o build Venn
diag ams, which we e d awn o scale a h ps://www.me a-cha .com/
enn#/display. GO unc ional ca ego ies we e analyzed using DAVID [56]a
h ps://da id.nci c .go / ools.jsp.
DATA AVAILABILITY
P o eomic da a analysis is p o ided in Supplemen a y Da ase s S1 and S2. Mass
spec ome y da a ha e been deposi ed o he P o eomeXchange Conso ium ia he
PRIDE pa ne eposi o y wi h he da ase iden ifie PXD056069 (Use name:
[email p o ec ed].uk, Passwo d: me7u OzRODq). Unc opped images o
wes e n blo s a e con ained in he Supplemen a y File unc opped wes e n blo s.
REFERENCES
1. Ve egaal ACO. Signalling mechanisms and cellula unc ions o SUMO. Na Re
Mol Cell Biol. 2022;23:715–31.
2. Claessens LA, Ve egaal ACO. SUMO p o eases: om cellula unc ions o disease.
T ends Cell Biol. 2024;34:901–12.
3. Li X, Rasul A, Sha i F, Hassan M. PIAS amily in cance : om basic mechanisms o
clinical applica ions. F on Oncol. 2024;14:1376633.
4. La a-U eña N, Ja a i V, Ga cia-Dominguez M. Cance -associa ed dys egula ion o
sumo egula o s: p o eases and ligases. In J Mol Sci. 2022;23:8012.
5. Yang Y, He Y, Wang X, Liang Z, He G, Zhang P, e al. P o ein SUMOyla ion
modifica ion and i s associa ions wi h disease. Open Biol. 2017;7:170167.
6. Tempe D, Piechaczyk M, Bossis G. SUMO unde s ess. Biochem Soc T ans.
2008;36:874–8.
7. Cima os i H, Lindbe g C, Bomhol SF, Ronn LC, Henley JM. Inc eased p o ein
SUMOyla ion ollowing ocal ce eb al ischemia. Neu opha macology. 2008;54:
280–9.
8. Yang W, Sheng H, Wa ne DS, Paschen W. T ansien global ce eb al ischemia
induces a massi e inc ease in p o ein sumoyla ion. J Ce eb Blood Flow Me ab.
2008;28:269–79.
9. Ka andika P, Ge s l JVE, Kappel AD, Won SY, Dubinski D, Ga cia-Segu a ME, e al.
SUMO he apeu ics o ischemic s oke. Pha maceu icals. 2023;16:673.
10. Cima os i H, Ashikaga E, Jaa a i N, Dea den L, Rubin P, Wilkinson KA, e al.
Enhanced SUMOyla ion and SENP-1 p o ein le els ollowing oxygen and glucose
dep i a ion in neu ones. J Ce eb Blood Flow Me ab. 2012;32:17–22.
11. Da wyle AL, La ig-Tunnemann G, Yang W, Paschen W, Lee SL, Di nagl U, e al.
SUMO2/3 conjuga ion is an endogenous neu op o ec i e mechanism. J Ce eb
Blood Flow Me ab. 2011;31:2152–9.
12. Lee YJ, Cas i P, Bemb y J, Ma ic D, Auh S, Hallenbeck JM. SUMOyla ion pa ici-
pa es in induc ion o ischemic ole ance. J Neu ochem. 2009;109:257–67.
F. Galla do-Chamizo e al.
9
Cell Dea h Disco e y (2025) 11:230
