scieee Science in your language
[en] (orig)

Role of carbonic anhydrases in skin wound healing

Read accessible full text

Role of carbonic anhydrases in skin wound healing

Author: Barker, Harlan,Aaltonen, Marleena,Pan, Peiwen,Vähätupa, Maria,Kaipiainen, Pirkka,May, Ulrike,Prince, Stuart,Uusitalo-Järvinen, Hannele,Waheed, Abdul,Pastorekova, Silvia,Sly, William S,Parkkila, Seppo,Järvinen, Tero AH
Year: 2017
Source: https://trepo.tuni.fi/bitstream/10024/101957/1/role_of_carbonic_anhydrates_2017.pdf
OPEN
ORIGINAL ARTICLE
Role o ca bonic anhyd ases in skin wound healing
Ha lan Ba ke 1,5, Ma leena Aal onen1,5, Peiwen Pan1, Ma ia Vähä upa1, Pi kka Kaipiainen1, Ul ike May1,
S ua P ince1, Hannele Uusi alo-Jä inen1,2, Abdul Waheed3, Sil ia Pas o eko á4, William S Sly3,
Seppo Pa kkila1,2 and Te o AH Jä inen1,2
Skin wound closu e occu s when ke a inocy es mig a e om he edge o he wound and e-epi helialize he epide mis. Thei
mig a ion akes place p ima ily be o e any ascula iza ion is es ablished, ha is, unde hypoxia, bu ela i ely li le is known
ega ding he ac o s ha s imula e his mig a ion. Hypoxia and an acidic en i onmen a e well-es ablished s imuli o cance cell
mig a ion. The ca bonic anhyd ases (CAs) con ibu e o umo cell mig a ion by gene a ing an acidic en i onmen h ough he
con e sion o ca bon dioxide o bica bona e and a p o on. On his basis, we explo ed he possible ole o CAs in issue
egene a ion using mouse skin wound models. We show ha he exp ession o mRNAs encoding CA iso o ms IV and IX a e
inc eased (~25 × and 4 × , espec i ely) du ing he wound hypoxic pe iod (days 2–5) and ha cells exp essing CAs o m a band-
like s uc u e benea h he mig a ing epide mis. RNA-Seq analysis sugges ed ha he CA IV-specific signal in he wound is
mainly de i ed om neu ophils. Due o he high le el o induc ion o CA IV in he wound, we ea ed skin wounds locally wi h
ecombinan human CA IV enzyme. Recombinan CA IV significan ly accele a ed wound e-epi helializa ion. Thus, CA IV could
con ibu e o wound healing by p o iding an acidic en i onmen in which he mig a ing epide mis and neu ophils can su i e
and may o e no el oppo uni ies o accele a e wound healing unde comp omised condi ions.
Expe imen al & Molecula Medicine (2017) 49, e334; doi:10.1038/emm.2017.60; published online 19 May 2017
INTRODUCTION
The healing o a human skin wound is a complex and
highly coo dina ed biological p ocess ha in ol es di e se
phenomena such as hemos asis, inflamma ion, e-epi helializa-
ion, angiogenesis, fib oplasia and finally issue emodeling.1,2
‘Re-epi helializa ion’is he la e al mig a ion o ke a inocy es
ac oss a wound bed, which when success ul, closes he
wound.1,2
Re-epi helializa ion begins wi hin hou s a e inju y. I is
belie ed ha a c i ical swi ch o he ini ia ion o ke a inocy e
mig a ion is he acu e change in oxygen ension.1Tha is,
when he skin is wounded and he de mal blood essels a e
clo ed and no longe able o deli e oxygen o he
skin, he ke a inocy es expe ience he s ess o acu e hypoxia
and ini ia e mig a ion o close he de ec . Rema kably,
he e-epi helializa ion akes place mainly unde hypoxia,
and wound closu e can be comple ed be o e any new
e- ascula iza ion occu s.1
S udies ha e shown ha in addi ion o hypoxia, acidosis
s imula es cance cell mig a ion.3,4 Ca bonic anhyd ases (CA)
a e a amily o zinc me alloenzymes ha egula e he
issue acid–base equilib ium by ca alyzing he e e sible
hyd a ion o ca bon dioxide o bica bona e ions and p o ons
(CO2+H2Oo–4HCO3
−+H+).3–5Fi een human CA iso o ms
ha e been ound, o which 12 a e ac i e and 3 inac i e.6,7 These
isoenzymes a e exp essed o some ex en in all issues and
o gans, bu pa icula ly in hose ha a e me abolically highly
ac i e such as he b ain and kidney.3,5 In e es ingly, he
exp ession o CA p o eins IX and XII is induced by hypoxia
in di e en umo s.8Th ough hei abili y o egula e pH and
gene a e an acidic en i onmen , hese enzymes endow umo
cells wi h su i al ad an ages unde hypoxia/acidosis condi-
ions and con e an inc eased abili y o mig a e.
The po ency o he CAs, IX and XII, o s imula e cell
mig a ion unde hypoxia p omp ed us o in es iga e he ole o
CAs in skin wound healing. Qui e unexpec edly, no hing is
known ega ding CAs du ing skin wound healing, al hough he
es o a ion o CA ac i i y is ela ed o duodenal ulce s. We
assumed ha he exp ession o ce ain CA amily membe s
could inc ease when a wound is exposed o hypoxia and,
1
Facul y o Medicine and Li e Sciences, Uni e si y o Tampe e, Tampe e, Finland;
2
Eye Cen e, Fimlab Labo a o ies, Depa men o O hopedics and
T auma ology, Tampe e Uni e si y Hospi al, Tampe e, Finland;
3
Edwa d A Doisy Depa men o Biochemis y and Molecula Biology, Sain Louis Uni e si y
School o Medicine, Sain Louis, MO, USA and
4
Cen e o Molecula Medicine, Ins i u e o Vi ology, Slo ak Academy o Sciences, B a isla a, Slo ak Republic
Co espondence: P o esso T Jä inen, Facul y o Medicine and Li e Sciences, Uni e si y o Tampe e, Tampe e FI-33014, Finland.
E-mail: bl eja@u a.fi
5These au ho s con ibu ed equally o his wo k.
Recei ed 23 Sep embe 2016; e ised 7 Decembe 2016; accep ed 9 Decembe 2016
Expe imen al & Molecula Medicine (2017) 49, e334; doi:10.1038/emm.2017.60
&
2017 KSBMB. All igh s ese ed 2092-6413/17
www.na u e.com/emm
heo e ically, could s imula e e-epi helializa ion by gene a ing
an acidic en i onmen o ke a inocy es. Thus, we s udied he
exp ession pa e n o he enzyma ically ac i e CAs in wounds
in mice, and based on he esul s also ea ed he wounds wi h
exogenous ecombinan CA IV enzyme.
MATERIALS AND METHODS
Gene a ion o skin wounds
Fo he quan i a i e PCR (qPCR) and immunohis ochemis y
analyses, 8-week-old male BALB/c mice (weighing 23–25 g) we e
used. Mice we e ed wi h s anda d labo a o y pelle s and wa e
ad libi um. All animal expe imen s we e pe o med in acco dance
wi h p o ocols app o ed by he Na ional Animal E hics Commi ee o
Finland.
Six-millime e diame e , ull hickness (including panniculus ca nosus
muscle) excision wounds we e c ea ed in he do sal skin unde
se oflu ane anes hesia as p e iously desc ibed.9A a ious ime poin s,
he animals we e killed, and he wounded issue was collec ed and
p ocessed o u he analyses.
qPCR analysis
To al skin wound RNA collec ed a a ious ime poin s was con e ed
o cDNA by e e se ansc ip ion using he High Capaci y
cDNA Re e se T ansc ip ion Ki o RT-qPCR (Applied Biosy ems,
Fos e Ci y, CA, USA). Duplica e qRT-PCR eac ions we e pe o med
wi h Powe SYBR SYBRG een eagen s (Applied Biosy ems) on an
ABI 7000 Real Time PCR Sys em (Applied Biosy ems). De ails o he
p ime s used a e lis ed in Supplemen a y Table 1. As nega i e con ols,
no- empla e and no- e e se ansc ip ase con ols we e also included
(which we e nega i e he ein). The da a analyses we e pe o med
acco ding o Li ak and Schmi gen.10
Exp ession and pu ifica ion o ecombinan human CA IV
cDNA o he sec e o y o m o human CA IV was cloned in pET-11d,
a bac e ial exp ession ec o , as p e iously desc ibed.11 B iefly, he
E. coli s ain Rose a(DE3)pLysS was used o enzyme p oduc ion.
The enzyme was pu ified using a CA-inhibi o a fini y column.11
A fini y-pu ified enzyme was dialyzed agains 20 mMammonium
bica bona e o se e al changes o bu e . Endo oxin emo al was
pe o med using a specialized endo oxin emo al column h ee imes
(Hyglos Blue Endo oxin Remo al Column, S a nbe g, Ge many). The
p o ein samples we e hen e-dialyzed wi h saline, fil e -s e ilized and
s o ed a 4 °C. Recombinan CA IV was analyzed on an NuPAGE 4–
12% g adien gel (The moFishe Scien ific, Wal ham, MA, USA). The
enzyma ic ac i i y o ecombinan CA IV was assayed as p e iously
desc ibed.11
Skin wound ea men ial wi h ecombinan CA IV
Fo he ea men ial, 8-week-old male li e ma e BALB/c mice we e
used. Two 6-mm, ci cula , ull- hickness wounds we e gene a ed in
each animal by biopsy punch. Donu -shaped 12-mm silicone splin s
(G ace Bio-Labs, Bend, OR, USA) we e placed a ound he wounds
and a fixed wi h glue and su u es o p e en wound con ac ion.12
Plas ic colla s we e used o p e en he mice om ea ing o he
splin s.12 The mice ha unde wen su ge y we e di ided in o h ee
g oups. The fi s g oup consis ed o mice ea ed wi h CA IV; pu ified
CA IV was mixed wi h 30% Plu onic-127 (Sigma-Ald ich) gel as
p e iously desc ibed,13 and he pH was 7.0. Fi y mic oli e s o he
mix u e was added o each wound o fill up he en i e wound ca i y.
The second g oup ecei ed a CA inhibi o , ace azolamide (1.0 mM,
O ion, Espoo, Finland), in Plu onic-127 gel. The hi d g oup se ed as
a con ol g oup and was ea ed wi h 50 μl o phospha e-bu e ed
saline mixed wi h Plu onic-127 gel. The final concen a ion o
Plu onic-127 in each solu ion was 18%. Each mouse was pho o-
g aphed a e he silicone splin s had been secu ely su u ed in place
and hen killed on day 5.
RNA-Seq analysis o wound-specific issues
RNA-Seq da a om human B-cells, CD4-posi i e cells, CD8-posi i e
cells, monocy es, neu ophils, na u al kille (NK) cells, monocy es
(A ayExp ess accession E-GEOD-60424),14 de mal fib oblas s
(E-GEOD-72589), undi e en ia ed and di e en ia ed ke a inocy es
(E-MTAB-1717),15 M1 and M2 mac ophages (E-GEOD-36952), and
skin cells (E-MTAB-2836)16 we e e ie ed om he A ay Exp ess
da abase. Reads o all samples we e pooled by issue ype and mapped
o CA genes in he human genome using he Bow ie17 and Topha 18
modules o he Topha package u ilizing supe compu e esou ces
p o ided by CSC–IT Cen e o Science o he Finnish Minis y o
Educa ion and Cul u e. Subsequen ma ches we e hen compa ed o
e e ence CA ansc ip s uc u es om he Ensembl da abase using
he Cu compa e module o he Cu flinks package.19 No malized
agmen s pe kilobase o ansc ip pe million mapped ead (FPKM)
alues we e gene a ed o each ansc ip iden ified in each issue using
he Cu no m module o he Cu flinks package o compa e exp ession
alues ac oss he pooled samples. The exp ession le els o each
CA we e compa ed ac oss issue ypes using he Cu me ge and
Cu di modules o Cu flinks. The esul s we e isualized using he
cummeRbund R lib a y.
S a is ical analysis
The da a a e p esen ed as he mean±s.d. Fo compa isons o mul iple
g oups, s a is ical analysis was conduc ed by wo-way analysis o
a iance complemen ed by he Bon e oni pos hoc es o pai wise
compa isons be ween he es g oups. P- alues o0.05 we e conside ed
significan .
Supplemen a y Me hods
Ex ac ion o RNA, he oxygen-induced e inopa hy model,
immunohis ochemis y, his ology, quan i a i e analysis o his ology
(wound healing), p omo e analysis o he p ima y CA4 (human gene
o CA IV p o ein) ansc ip and gene on ology (GO) analysis o
p o eins we e pe o med using s anda d me hods20–24 and a e
desc ibed in de ail in he Supplemen a y Da a se .
RESULTS
Exp ession o CA IV and CA IX is induced du ing he
hypoxic phase o wound healing
To in es iga e he ole o CAs in he skin wound healing
p ocess, we de e mined he exp ession pa e n o 11 ac i e
membe s o he CA amily (CA I, II, III, IV, Vb, VI, VII, IX,
XII, XIII and XV) by qPCR analysis o wounds a di e en
s ages o healing, and compa ed hei exp ession o he le els
obse ed in no mal, unwounded skin. Hypoxia pe sis ed in ou
excision wound model o 5 days a e wounding, a e which
ex ensi e angiogenesis ascula ized he wound bed.9Among
he CA amily membe s, he exp ession o CA IV mRNA was
induced mos s ongly a almos 25- old abo e no mal 2 days
a e he wounding and emained ele a ed mo e han 5- old a
5 days a e wounding (Figu e 1). CA IX was he only o he
Ca bonic anhyd ases in skin wounds
H Ba ke e al
2
Expe imen al & Molecula Medicine
enzyme ha showed ele a ed exp ession du ing wound healing;
i s exp ession peaked a 3 days o a le el mo e han ou old
ha o no mal skin and emained ele a ed a e day 5 o
healing (Figu e 1). A e he skin wound was ascula ized
( om day 5 on), none o he CAs showed any enhanced
exp ession a he mRNA le el (Figu e 1).
As CAs IV and IX showed inc eased mRNA exp ession
du ing he hypoxic phase o wound healing, we decided o
explo e hei exp ession in de ail by immunohis ochemis y.
CA IV p o ein exp ession s a ed o accumula e in ea ly
g anula ion issue a day 5 a e healing and emained posi i e
in he wound un il day 10, a e which he exp ession
disappea ed (Figu e 2). CA IV p o ein was exp essed mainly
by cells in g anula ion issue wi h especially s ong, band-like,
exp ession jus benea h he mig a ing epide mis (Figu e 2).
CA IX p o ein exp ession, in u n, was al eady obse ed in he
skin wound a day 2 and emained ele a ed h oughou he
wound healing p ocess, ha is, a e CA IV had disappea ed
om he wound issue (Figu e 2). Bo h he epide mis and
unde lying g anula ion issue exp essed CA IX, bu he
s onges CA IX p o ein exp ession was also obse ed in he
mig a ing epide mis and jus benea h he mig a ing epide mis
in he op laye o he de mis (Figu e 2).
CA ansc ip s in wound- ela ed cells
To elucida e he cell sou ces ha p oduce CAs du ing wound
healing, a ho ough analysis o human RNA-Seq da a was
pe o med. RNA-Seq analysis o 12 cell ypes and issues
e ealed ha a CA4 ansc ip was significan ly exp essed
only in skin and neu ophils (Figu e 3). In bo h issues,
only he 312-amino-acid ENST00000300900 CA4 ansc ip
was exp essed. In he skin, 12 CA genes we e exp essed
(Supplemen a y Table 2). O hese, CA4 had one o he lowes
le els o exp ession, wi h a no malized FPKM alue o 160.18.
In neu ophils, he CA4 exp ession le el was significan ly
highe han ha in skin, wi h a FPKM o 7162.48. CA9 was
significan ly exp essed in fib oblas s, skin, and mac ophages
(Figu e 3), wi h significan ly highe exp ession in fib oblas s
compa ed wi h he o he issues.
Cha ac e iza ion o he CA4 p omo e
The unexpec ed and high-le el induc ion o CA4 mRNA in he
skin wound p omp ed us o cha ac e ize he CA4 p omo e in
de ail o p o ide clues ega ding he po en ial egula ion o i s
ansc ip ion. Ou compa a i e genomics analysis o he
aligned CA4 p omo e egions o 15 mammal species e ealed
a dis inc clus e o high sco ing and well-conse ed po en ial
ansc ip ion ac o -binding si es immedia ely ups eam o
a simila ly p edic ed TATA-binding p o ein-binding si e
(Figu e 4). All 15 species p esen ed a s ong signal o
a CTCF-binding si e loca ed om −55 o −44 bp ups eam
o he TSS (Figu e 4). In 13 o he species, including humans,
his egion was an exac ma ch o p e iously expe imen ally
de e mined CTCF-binding si es. O e lapping he p obable
CTCF si e we e p edic ed binding si es o SP1, KLF4
and MZF1. The SP1 and MZF1 si es ma ched p e iously
expe imen ally de e mined si es in all 15 species, while
KLF4 had a ma ch in 14 species. The binding loca ions we e
−55 o −44 on he sense-s and (CTCF), −50 o −39 on he
an i-sense s and (SP1), −47 o −41 on he sense-s and
(MZF1), and −49 o −39 on he sense-s and (KLF4)
(Figu e 4). Impo an ly, ou esul s did no e eal any
HIF-1α-binding si es in he CA4 p omo e .
GO analysis o biological p ocesses associa ed wi h CA4 in
skin
Nex , we wan ed o unde s and he po en ial unc ion o CA4
in skin wound healing and hus pe o med GO en ichmen
analysis o biological p ocesses associa ed wi h CA4 exp ession.
GO en ichmen analysis o genes wi h a s ong exp ession
co ela ion (X0.50) wi h CA4 in skin esul ed in 40 e ms ha
we e ep esen ed wo- old o highe in ou se e sus he
expec ed le els (Supplemen a y Table 3). The e ms wi h he
highes o e - ep esen a ion in ou se we e s ongly ela ed o
immune cell ec ui men . In addi ion, e ms ela ed o
angiogenesis, endocy osis, inflamma ion and ion homeos asis
we e also o e - ep esen ed.
Exp ession o Ca 4 is no induced by hypoxia in he pu e
hypoxia-d i en angiogenesis model
To explo e whe he hypoxia induced he exp ession o Ca 4
( oden gene o CA IV p o ein) du ing wound healing,
we nex employed a pu e hypoxia-d i en angiogenesis o
oxygen-induced e inopa hy model. We could no de ec any
induc ion o Ca 4 mRNA ei he by hypoxia a P12 o by
e ascula iza ion a P17, hus elimina ing he possibili y ha
CA4 is a hypoxia inducible gene (Supplemen a y Figu e 1).
Figu e 1 mRNA exp ession o CA4 and CA9 du ing wound healing.
Skin excision wounds we e gene a ed in WT mice as desc ibed in
he Me hods sec ion. No mal skin and skin wound samples we e
collec ed om unwounded mice and om mice killed a a ious
ime poin s a e wounding. The skin samples we e p ocessed o
qPCR analysis as desc ibed in he Me hods. The esul s o all
enzyma ically ac i e CAs a e shown as he mean ±s.d. Animal
numbe s: unwounded: n=2; day 2: n=2; day 3: n=2; day 5:
n=3; day 7: n=2; day 10: n=2; day 14: n=2.
Ca bonic anhyd ases in skin wounds
H Ba ke e al
3
Expe imen al & Molecula Medicine
Exp ession and cha ac e iza ion o ecombinan CA IV
To u he explo e he unc ion o CA IV in skin wound
healing, we p oduced and pu ified ecombinan CA IV enzyme.
CA IV is especially sui able o he apeu ic applica ions because
i is an ex acellula enzyme and has one o he highes
enzyma ic ac i i y le els among he CAs. In SDS gel elec o-
pho esis, a majo band was iden ified a 30 kDa ha
ep esen ed ull-leng h ecombinan CA IV (Supplemen a y
Figu e 2). The a fini y-pu ified CA IV enzyme had a specific
ac i i y o 3000–4000 uni s pe mg o pu e enzyme.
Wound ea men wi h CA IV and ace azolamide
Nex , we examined he e ec s o ecombinan CA IV and
endogenously exp essed CAs on wound healing, ocusing on
he e-epi helializa ion ob ained by ke a inocy e mig a ion
du ing he hypoxic phase o healing. Roden s wound closu e
can occu ia wo di e en means: wound con ac ion o ue
e-epi helializa ion.12 To alle ia e wound con ac ion and o
explo e e-epi helializa ion specifically, a special skin excision
wound model was employed, in which a ound silicone splin
is su u ed in o he skin o fi mly a ach he unde lying de mis
and subcu is.12
The wounded animals we e di ided in o h ee g oups
ha ecei ed opically applied saline (con ol), ecombinan
CA IV enzyme o CA inhibi o (aze azolamide) in Plu onic-127
gels.13 The ewe enodi e encesin hewoundsizeimmedia ely
a e wounding o in he size o he scab co e ing he wound a
he end o he ea men ial (Supplemen a y Figu e 3). The
ecombinan CA IV enhanced he wound e-epi helializa ion
(Figu e 5). The epi helial ongues we e significan ly longe in he
ecombinan CA IV- ea ed wounds han in he con ol wounds
(Po0.0001) (Figu e 5). We could no de ec any di e ences in
he amoun o g anula ion issue p oduced in he wounds
among he h ee ea men g oups (Figu e 5).
DISCUSSION
Skin wound closu e is achie ed by e-epi helializa ion, ha is,
he la e al mig a ion o ke a inocy es. This p ocess occu s
mainly unde hypoxia, bu i s s imuli a e poo ly unde s ood.
The p esen s udy shows ha mRNA exp ession o wo
membe s o he CA amily, CA IV and IX, a e inc eased
du ing he ea ly phase o wound healing, and he cells
exp essing he CAs o m a band-like s uc u e jus unde nea h
he mig a ing epide mis in he healing skin wound. Fu he -
mo e, we demons a e ha exogenous applica ion o ecombi-
nan Ca 4 enzyme accele a es wound e-epi helializa ion
du ing he hypoxic phase o wound healing.
In hypoxic issues, such as as -g owing cance s and wounds,
me abolic p ocesses p oduce la ge amoun s o bo h lac ic acid
and ca bon dioxide. CAs a e enzymes ha con e ca bon
dioxide and wa e o bica bona e and p o ons.3–5Because he
in acellula pH mus emain s able, all excess acidi y is
e ficien ly expo ed om he cells, which in u n acidifies he
ex acellula en i onmen o p o ide a a o able en i onmen
o cance cell mig a ion and in asion.25 CAs ha e been linked
o his p ocess in epo s showing a CA inhibi o -induced
educ ion o cance cell in asion in con as o a acili a ion o
cance cell mig a ion and in asion by CA exp ession.25 CA IX,
in u n, is unique in he sense ha i can s imula e cell
mig a ion independen o ex acellula acidifica ion.26,27
Namely, i can enhance cell mig a ion by weakening cell
adhesions26,27 as well as by s imula ing Rho-GTPase dependen
cell mo ili y.26 In ou s udy, CA IX exp ession pe sis ed in he
Figu e 2 P o ein exp ession o Ca 4 and Ca 9 du ing wound
healing. Skin excision wounds we e gene a ed in WT mice as
desc ibed in he Me hods sec ion. No mal skin and skin wound
samples we e collec ed om unwounded mice and om mice killed
a a ious ime poin s a e wounding. The skin samples we e
p ocessed o IHC analysis, and CA IV and CA IX we e de ec ed
using specific an ibodies as desc ibed in he Supplemen a y
Me hods. The esul s ob ained a all s udied ime poin s a e shown.
CA IV p o ein exp ession s a ed o accumula e in he ea ly
g anula ion issue a day 5 o healing, and posi i e exp ession
emained in he wound un il day 10, a e which he exp ession
disappea ed. CA IV p o ein was exp essed mainly by he cells in
g anula ion issue, wi h especially s ong exp ession jus benea h
he mig a ing epide mis (a ows). CA IX p o ein exp ession, in u n,
was al eady de ec ed in he skin wound a day 2 and emained
ele a ed h oughou he wound healing p ocess. Animal numbe s:
all ime poin s: n=6 mice wi h wo wounds; o al n=12.
Ca bonic anhyd ases in skin wounds
H Ba ke e al
4
Expe imen al & Molecula Medicine
mig a ing epide mis h oughou he healing p ocess, whe eas
he exp ession o CA IV disappea ed subs an ially ea lie . The
s ong exp ession o CA IX obse ed in he mig a ing
epide mis long beyond he hypoxic phase could be ela ed o
i s capabili y o s imula e cell mig a ion independen ly o
acidifica ion o he hypoxic en i onmen , and may be s imu-
la ed by hepa ocy e g ow h ac o unde no moxic
condi ions.27
Figu e 3 Exp ession o CA4 and CA9 in wound- ela ed cells. RNA-Seq da a o mul iple immune and wound-healing- ela ed cells was
e ie ed om expe imen s pe o med using he A ayExp ess da abase and mapped o he genomic loca ions o all CA genes using Topha .
The eads we e hen me ged using Cu me ge, and di e en ial exp ession was de e mined using Cu di . Finally, he esul s we e
manipula ed using he R package cummeRbund and isualized in R. Exp ession abundance measu emen s a e ep esen ed on he yaxis
as no malized agmen s pe kilobase o ansc ip pe million mapped eads (FPKM) o CA4 (a)andCA9(b).
Figu e 4 Compa a i e genomics analysis o he CA4 p omo e . An alignmen o 15 mammalian sequences co esponding o he p omo e o
he ull-leng h human CA4 ansc ip ENST00000300900 was analyzed o pu a i e ansc ip ion ac o -binding si es by compa a i e
genomics. The 10 bes sco ing ansc ip ion ac o s we e included in he figu e, whe e heigh indica es he numbe o species suppo ing ha
p edic ion. A posi i e yaxis esul indica es a TFBS p edic ed on he sense s and, while he nega i e yaxis indica es he an i-sense s and.
Ca bonic anhyd ases in skin wounds
H Ba ke e al
5
Expe imen al & Molecula Medicine

O he wo CAs ha showed enhanced exp ession du ing
wound healing, CA IX is known o be induced by hypoxia. I
has a hypoxia- esponse elemen in i s p omo o egion and is
one o he known a ge genes o hypoxia inducible ac o -1α
(HIF-1α).3,4 S ikingly, he exp ession o CA IV was subs an-
ially highe han CA IX a he mRNA le el du ing hypoxia in
he skin wound. The exp ession o CA IV is usually educed in
umo s,28 and he p esen esul s show ha i does no possess
a classical hypoxia- esponse elemen in i s p omo e egion o
HIF-1αbinding. We u he employed a pu e hypoxia-d i en
angiogenesis model in he e ina and could no de ec any
changes in he exp ession o Ca 4 by a ying he oxygen le els.
Ou esul s, in essence, elimina e he possibili y ha hypoxia is
esponsible o he induc ion o Ca 4 exp ession.
Compa a i e genomics p edic ion o po en ial ansc ip ion
ac o -binding si es in he p omo e o he p ima y human
CA4 ansc ip ENST00000300900 e ealed some high-sco ing
candida es ha may al e he exp ession o his p o ein.
In pa icula , he e is a clus e o well-conse ed binding
si es o KLF4, SP1 and MZF1, all o which we e posi ioned
nea a simila ly conse ed CTCF-binding si e. CTCF is a well-
es ablished con ibu o o he ch oma in configu a ion and
gene ansc ip ion ac oss he genome,29 and hus i s p esence
and p oximi y p o ide g ea e weigh o he o he p oximal
p edic ions. Fu he mo e, CTCF also p omo es wound
healing.30 MZF1 is in ol ed in myeloid cell di e en ia ion31
and is a s ong enhance o cell mig a ion and in asion,32 hus
po en ially explaining he o igin o he s ong CA4 exp ession
in neu ophils. KLF4, in u n, has been shown o be exp essed
du ing inflamma ion33 and wound healing34–37, and o acil-
i a e cu aneous wound healing.36
Ou RNA-Seq analysis e ealed low le els o CA4 RNA in
no mal skin and high le els in neu ophils. This finding implies
ha he obse ed ele a ion o CA4 mRNA and p o ein in ou
expe imen s may be due o he ec ui men o neu ophils o
he wound si e, which is consis en wi h he finding ha
abundan neu ophil ex a asa ion occu s in skin wounds
apidly a e wounding.
The ew neu ophils ha pe sis in he wound a e known o
o m a band-like s uc u e immedia ely benea h he mig a ing
epide mis,38 a pa e n ha is s ikingly simila o he exp ession
o CA IV. Despi e long being conside ed de imen al o issue
egene a ion in gene al, especially du ing he ea ly phase o
healing, neu ophils a e cu en ly conside ed c ucial o skin
wound healing and issue egene a ion.39 Specifically, neu o-
phils ha su i e in he inju ed issue o a long pe iod o ime
a e c ucial o issue egene a ion.40,41 In e es ingly, neu ophils
equi e an acidic ex acellula en i onmen o be p ope ly
ac i a ed and o p olong hei unc ional li e span, ha is,
a oid apop osis.42,43 Fu he mo e, excessi e, ea ly neu ophil
apop osis (NETosis) has, in u n, been shown o e a d diabe ic
wound healing in bo h mice and humans,40 while p ope
wound healing can be escued in diabe es by he inhibi ion o
ea ly NETosis, ha is, keeping neu ophils ali e and p olonging
hei li espan.40 This phenomenon could specifically explain
why neu ophils exp ess an ex acellula membe o he
Figu e 5 Accele a ed e-epi helializa ion du ing wound healing in mice ea ed wi h ecombinan Ca 4 enzyme. Mice wi h ull- hickness
skin excision wounds we e ea ed wi h ei he ecombinan Ca 4 enzyme o CA inhibi o applied opically o he wound immedia ely a e
wounding. Sca s we e ha es ed on day 5, and he e-epi helializa ion (a) and c oss-sec ional a ea o he g anula ion issue (b)o he
wounds we e quan ified by examining wo mic oscopic sec ions om each wound. The esul s a e exp essed as he a e age o he wo
alues. The e we e fi e animals, each wi h wo wounds, in e e y ea men g oup. *Po0.05; analysis o a iance. The esul s a e exp essed
as he mean ±s.d., n=10 wounds. (c) Rep esen a i e sec ions om wounds ea ed wi h ecombinan CAIV o CA inhibi o and collec ed
on day 5 a e wounding a e shown o e-epi helializa ion. Scale ba s, 600 μm low magnifica ion, 240 μmhighmagnifica ion.
Ca bonic anhyd ases in skin wounds
H Ba ke e al
6
Expe imen al & Molecula Medicine
CAs, CA IV, which is c ucial o ex acellula acidifica ion.5
Especially s ong p oduc ion o CA IV and IX jus benea h he
mig a ing epide mis sugges s ha hese enzymes could gene a e
an acidic ‘mic o-en i onmen ’wi hin healing wound issue o
selec i ely aid no jus neu ophil su i al and unc ion bu also
ke a inocy e mig a ion, ha is, closu e o he wound.
Conce ning CA IX exp ession in he skin wound, RNA-Seq
analysis confi med he exp ession o CA9 mRNA only in
fib oblas s and skin, wi h significan ly highe exp ession in
fib oblas s. The CA IX p o ein has been shown o be
up egula ed in umo fib oblas s unde going hypoxia,44,45
while fib oblas -de i ed HIF-1αhas been shown o be c ucial
o wound healing.46
Based on he exp ession pa e n o CA amily membe s, we
conclude ha CAs IV and IX a e esponsible o main aining
he pH balance du ing skin wound healing. We we e able o
obse e he induc ion o CA IV and CA IX mRNAs du ing he
ea ly hypoxic phases o wound healing, and bo h mRNAs
e u ned o he same le el as he no mal skin in 7 days. This
finding implies ha he e is a apid induc ion o hese p o eins
in esponse o hypoxia (HIF1α-media ed o CA IX) and
inflamma ion (CA IV).
To unde s and he unc ion o CAs in issue egene a ion
in gene al, we ea ed skin wounds wi h ecombinan human
CA IV enzyme and wi h a clinically used CA inhibi o o block
he endogenous CA ac i i y. The CA IV enzyme is especially
sui able o pha maceu ical s udies because i is an ex acellula
enzyme wi h one o he highes enzyma ic ac i i y le els
among he CAs.5,47 We could de ec accele a ed wound
e-epi helializa ion wi h ecombinan CA IV, bu blocking
he ac i i y o endogenously exp essed CA had no significan
e ec on wound e-epi helializa ion. This disc epancy is mos
p obably explained by he ac ha endogenous CA accumula-
ion in he skin wound s a s a leas a ew days a e he
wounding. Thus, he CA inhibi o was unlikely o encoun e
any enhanced CA p o ein exp ession du ing he majo i y o
ou s udy pe iod (days 0–5), whe eas he addi ion o exogenous
ecombinan CA enzyme p o ided a clea benefi o e
endogenous enzyme because i was p esen in he wound
immedia ely a e wounding. This esul is highly ele an o
CA4 because ou da a indica e ha i s p ima y sou ce in he
wound bed is neu ophils, which begin o ex a asa e o he
wound wi hin minu es a e inju y and a e c ucial o wound
healing bu also equi e ex acellula acidifica ion o p olong
hei su i al, which in u n is c ucial o he p ope p og es-
sion o wound healing. Fu he mo e, closu e o he wound by
ke a inocy e mig a ion s a s immedia ely a e wounding.
Immedia e acidifica ion o he ex acellula milieu by exogen-
ous CA4 should p o ide a boos o ke a inocy e mig a ion
du ing he ea ly s ages o healing.
Fu u e s udies a e wa an ed o add ess whe he CAs can be
used o s imula e issue egene a ion du ing inju y-induced
hypoxia. CA IV may pa icipa e in p omo ing issue egene a-
ion by gene a ing an acidic ‘mic o-en i onmen ’ o
neu ophils o su i e and o he mig a ing epide mis o close
he wound du ing he hypoxic phase o wound healing.
CONFLICT OF INTEREST
The au ho s decla e no conflic o in e es .
ACKNOWLEDGEMENTS
We hank Ma ianne Ka lsbe g and Ma ja-Leena Koskinen o p ac ical
suppo and o pe o ming he his ochemical wo k and
immunohis ological s aining and M s Guille mina Ga cia
(San o d Bu nham P ebys Medical Disco e y Ins i u e, La Jolla,
CA, USA) is hanked o he echnical expe ise and help wi h
quan i a i e mic oscopy. This wo k was unded by he Sig id Juselius
Founda ion, he Academy o Finland, Päi ikki and Saka i Sohlbe g
Founda ion, Ins umen a ium Resea ch Founda ion, Finnish Medical
Founda ion, Pi kanmaa Hospi al Dis ic Resea ch Founda ion and he
Finnish Cul u al Founda ion. We also hank CSC –IT Cen e o
Science o he Finnish Minis y o Educa ion and Cul u e o
p o iding he high-pe o mance compu ing esou ces needed o
pe o m he compu a ional analyses.
Au ho con ibu ions: TJ, MA, HU-J and SPa designed he esea ch.
SPas aised he CA IX an ibody. AW and WSS p oduced he
ecombinan CA IV enzyme and aised he CA IV an ibody. HB
pe o med he compa a i e genomics, RNA-Seq and GO en ichmen
compu a ional analyses. MA, HB, PK, PP, MV, UM and SPa
pe o med he esea ch. MA, HB, PP, SPa and TJ analyzed he da a.
MA, HB, SPa and TJ w o e he manusc ip . MA, HB, PP, SPa and TJ
gene a ed he figu es. All au ho s e iewed and accep ed he ex o he
manusc ip .
1 Woodley DT, Wysong A, DeCle ck B, Chen M, Li W. Ke a inocy e mig a ion
and a hypo he ical new ole o ex acellula hea shock p o ein 90 alpha in
o ches a ing skin wound healing. Ad Wound Ca e 2015; 4:203–212.
2 Ma in P, Nunan R. Cellula and molecula mechanisms o epai in acu e
and ch onic wound healing. B J De ma ol 2015; 173:370–378.
3 Benej M, Pas o eko a S, Pas o ek J. Ca bonic anhyd ase IX: egula ion and
ole in cance . Subcell Biochem 2014; 75:199–219.
4 Pas o eko a S, Pa kkila S, Pa kkila AK, Opa sky R, Zelnik V, Saa nio J e al.
Ca bonic anhyd ase IX, MN/CA IX: analysis o s omach complemen a y DNA
sequence and exp ession in human and a alimen a y ac s. Gas oen e -
ology 1997; 112:398–408.
5 Waheed A, Sly WS. Memb ane associa ed ca bonic anhyd ase IV (CA IV): a
pe sonal and his o ical pe spec i e. Subcell Biochem 2014; 75:157–179.
6 Hil o M, Innocen i A, Mon i SM, De Simone G, Supu an CT, Pa kkila S.
Recen ad ances in esea ch on he mos no el ca bonic anhyd ases, CA
XIII and XV. Cu Pha m Des 2008; 14:672–678.
7 Aspa wa A, Tol anen ME, Pa kkila S. An upda e on ca bonic anhyd ase-
ela ed p o eins VIII, X and XI. J Enzyme Inhib Med Chem 2013; 28:
1129–1142.
8 I ano S, Liao SY, I ano a A, Danilko i ch-Miagko a A, Ta aso a N,
Wei ich G e al. Exp ession o hypoxia-inducible cell-su ace ansmem-
b ane ca bonic anhyd ases in human cance . Am J Pa hol 2001; 158:
905–919.
9 Jä inen TA, Ruoslah i E. Molecula changes in he ascula u e o inju ed
issues. Am J Pa hol 2007; 171:702–711.
10 Li ak KJ, Schmi gen TD. Analysis o ela i e gene exp ession da a using
eal- ime quan i a i e PCR and he 2(-Del a Del a C(T)) me hod. Me hods
2001; 25:402–408.
11 Waheed A, Pham T, Won M, Okuyama T, Sly WS. Human ca bonic
anhyd ase IV: in i o ac i a ion and pu ifica ion o disulfide-bonded
enzyme ollowing exp ession in Esche ichia coli.P o ein Exp Pu i 1997;
9:279–287.
12 Da idson JM, Yu F, Opalenik SR. Splin ing s a egies o o e come
con ounding wound con ac ion in expe imen al animal models. Ad Wound
Ca e 2013; 2:142–148.
13 Blanc-B ude OP, Yu J, Simosa H, Con e MS, Sessa WC, Al ie i DC. Inhibi o
o apop osis p o ein su i in egula es ascula inju y. Na Med 2002; 8:
987–994.
Ca bonic anhyd ases in skin wounds
H Ba ke e al
7
Expe imen al & Molecula Medicine
14 Linsley PS, Speake C, Whalen E, Chaussabel D. Copy numbe loss o he
in e e on gene clus e in melanomas is linked o educed T cell infil a e
and poo pa ien p ognosis. PLoS ONE 2014; 9:e109760.
15 Jones M, D y IR, F amp on D, Singh M, Kanda RK, Yee MB e al. RNA-seq
analysis o hos and i al gene exp ession highligh s in e ac ion be ween
a icella zos e i us and ke a inocy e di e en ia ion. PLoS Pa hog 2014;
10: e1003896.
16 Uhlen M, Fage be g L, Halls om BM, Lindskog C, Oks old P, Ma dinoglu A
e al. P o eomics. Tissue-based map o he human p o eome. Science
2015; 347: 1260419.
17 Langmead B, T apnell C, Pop M, Salzbe g SL. Ul a as and memo y-
e ficien alignmen o sho DNA sequences o he human genome. Genome
Biol 2009; 10:R25.
18 T apnell C, Pach e L, Salzbe g SL. TopHa : disco e ing splice junc ions
wi h RNA-Seq. Bioin o ma ics 2009; 25:1105–1111.
19 T apnell C, Williams BA, Pe ea G, Mo aza i A, Kwan G, an Ba en MJ
e al. T ansc ip assembly and quan ifica ion by RNA-Seq e eals
unanno a ed ansc ip s and iso o m swi ching du ing cell di e en ia ion.
Na Bio echnol 2010; 28:511–515.
20 May U, P ince S, Vähä upa M, Lai inen AM, Nieminen K, Uusi alo-Jä inen H
e al. Resis ance o R-Ras knockou mice o skin umou induc ion. Sci Rep
2015; 5: 11663.
21 Uusi alo-Ja inen H, Ku okawa T, Muelle BM, And ade-Go don P,
F iedlande M, Ru W. Role o p o ease ac i a ed ecep o 1 and 2 signaling
in hypoxia-induced angiogenesis. A e ioscle Th omb Vasc Biol 2007; 27:
1456–1462.
22 Mi H, Mu uganujan A, Casag ande JT, Thomas PD. La ge-scale gene
unc ion analysis wi h he PANTHER classifica ion sys em. Na P o oc
2013; 8:1551–1566.
23 Vähä upa M, P ince S, Va aja S, Me imo T, Ka aja M, Kinnunen K e al.
Lack o R-Ras leads o inc eased ascula pe meabili y in ischemic
e inopa hy. In es Oph halmol Vis Sci 2016; 57:4898–4909.
24 Jä inen TA, Ruoslah i E. Ta ge -seeking an ifib o ic compound enhances
wound healing and supp esses sca o ma ion in mice. P oc Na l Acad Sci
USA 2010; 107: 21671–21676.
25 Pas o eko a S, Pa kkila S, Za ada J. Tumo -associa ed ca bonic anhyd ases
and hei clinical significance. Ad Clin Chem 2006; 42:167–216.
26 Shin HJ, Rho SB, Jung DC, Han IO, Oh ES, Kim JY. Ca bonic anhyd ase IX
(CA9) modula es umo -associa ed cell mig a ion and in asion. J Cell Sci
2011; 124:1077–1087.
27 S as o a E, Wi a ski W, Csade o a L, Kosik I, Sk a ko a L, Huliko a A e al.
Ca bonic anhyd ase IX in e ac s wi h bica bona e anspo e s in lamelli-
podia and inc eases cell mig a ion ia i s ca aly ic domain. JBiolChem
2012; 287:3392–3402.
28 Niemela AM, Hynninen P, Mecklin JP, Kuopio T, Kokko A, Aal onen L e al.
Ca bonic anhyd ase IX is highly exp essed in he edi a y nonpolyposis
colo ec al cance . Cance Epidemiol Bioma ke s P e 2007; 16:
1760–1766.
29 Vie i Rudan M, Hadju S. Gene ic ailo s: CTCF and cohesin shape he
genome du ing e olu ion. T ends Gene 2015; 31:651–660.
30 Wang L, Wu X, Shi T, Lu L. Epide mal g ow h ac o (EGF)-induced co neal
epi helial wound healing h ough nuclea ac o kappaB sub ype- egula ed
CCCTC binding ac o (CTCF) ac i a ion. JBiolChem2013; 288:
24363–24371.
31 Hui P, Guo X, B ad o d PG. Isola ion and unc ional cha ac e iza ion o he
human gene encoding he myeloid zinc finge p o ein MZF-1. Biochemis y
1995; 34:16493–16502.
32 Eguchi T, P ince T, Wegiel B, Calde wood SK. Role and egula ion o myeloid
zinc finge p o ein 1 in cance . J Cell Biochem 2015; 116: 2146–2154.
33 Se illa LM, La o e V, Ca celle E, Boix J, Vodak D, Mills IG e al.
Glucoco icoid ecep o and Kl 4 co- egula e an i-inflamma o y genes in
ke a inocy es. Mol Cell Endoc inol 2015; 412:281–289.
34 Kaushik DK, Gup a M, Das S, Basu A. K uppel-like ac o 4, a no el
ansc ip ion ac o egula es mic oglial ac i a ion and subsequen neu oin-
flamma ion. JNeu oinflamma ion 2010; 7:68.
35 Liu J, Zhang C, Liu Z, Zhang J, Xiang Z, Sun T. Honokiol down egula es
K uppel-like ac o 4 exp ession, a enua es inflamma ion, and
educes his opa hology a e spinal co d inju y in a s. Spine 2015; 40:
363–368.
36 Ou L, Shi Y, Dong W, Liu C, Schmid TJ, Naga ka i P e al. K uppel-like
ac o KLF4 acili a es cu aneous wound healing by p omo ing fib ocy e
gene a ion om myeloid-de i ed supp esso cells. JIn es De ma ol2015;
135:1425–1434.
37 Li J, Zheng H, Wang J, Yu F, Mo is RJ, Wang TC e al. Exp ession o
K uppel-like ac o KLF4 in mouse hai ollicle s em cells con ibu es o
cu aneous wound healing. PLoS ONE 2012; 7: e39663.
38 G gu ic-Smi h LM, Lee HH, Gandhi JA, B ennan MB, DeLeon-Rod iguez CM,
Coelho C e al. Neu openia exace ba es in ec ion by Acine obac e bau-
mannii clinical isola es in a mu ine wound model. F on Mic obiol 2015; 6:
1134.
39 Pa kos CA. Neu ophil-epi helial in e ac ions: a double-edged swo d.
Am J Pa hol 2016; 186:1404–1416.
40 Fadini GP, Menegazzo L, Riga o M, Sca olini V, Poncina N, B u ocao A
e al. NETosis delays diabe ic wound healing in mice and humans. Diabe es
2016; 65:1061–1071.
41 Ho ckmans M, Ring L, Duchene J, San o i o D, Schloss MJ, D echsle M
e al. Neu ophils o ches a e pos -myoca dial in a c ion healing by pola iz-
ing mac ophages owa ds a epa a i e pheno ype. Eu Hea J 2017; 38:
187–197.
42 Ma inez D, Ve meulen M, T e ani A, Ceballos A, Saba e J, Gambe ale R
e al. Ex acellula acidosis induces neu ophil ac i a ion by a mechanism
dependen on ac i a ion o phospha idylinosi ol 3-kinase/Ak and ERK
pa hways. JImmunol2006; 176:1163–1171.
43 T e ani AS, Andonegui G, Gio dano M, Lopez DH, Gambe ale R, Minucci F
e al. Ex acellula acidifica ion induces human neu ophil ac i a ion.
JImmunol1999; 162:4849–4857.
44 San i A, Caselli A, Paoli P, Co i D, Camici G, Pie accini G e al. The e ec s
o CA IX ca alysis p oduc s wi hin umo mic oen i onmen . Cell Commun
Signal 2013; 11:81.
45 Ishii G, Sangai T, I o T, Hasebe T, Endoh Y, Sasaki H e al. In i o and
in i o cha ac e iza ion o human fib oblas s ec ui ed selec i ely in o
human cance s oma. In J Cance 2005; 117:212–220.
46 Dusche D, Maan ZN, Whi am AJ, So kin M, Hu MS, Walmsley GG e al.
Fib oblas -specific dele ion o hypoxia inducible ac o -1 c i ically impai s
mu ine cu aneous neo ascula iza ion and wound healing. Plas Recons
Su g 2015; 136:1004–1013.
47 Wande no h PM, Mannowe z N, Szczy ba J, G annemann L, Wol A,
Becke HM e al. No mal e ili y equi es he exp ession o ca bonic
anhyd ases II and IV in spe m. JBiolChem2015; 290:
29202–29216.
This wo k is licensed unde a C ea i e Commons
A ibu ion-NonComme cial-Sha eAlike 4.0 In e -
na ional License. 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 license,
unless indica ed o he wise in he c edi line; i he ma e ial is no
included unde he C ea i e Commons license, use s will need o
ob ain pe mission om he license holde o ep oduce he
ma e ial. To iew a copy o his license, isi h p://
c ea i ecommons.o g/licenses/by-nc-sa/4.0/
Supplemen a y In o ma ion accompanies he pape on Expe imen al & Molecula Medicine websi e (h p://www.na u e.com/emm)
Ca bonic anhyd ases in skin wounds
H Ba ke e al
8
Expe imen al & Molecula Medicine