Disse ação de candida u a ao g au de Mes e em
Oncologia – especialização em Oncologia Molecula
– subme ida ao Ins i u o de Ciências Biomédicas de
Abel Salaza da Uni e sidade do Po o.
O ien ado – P o esso Dou o Rui Gil da Cos a
Ca ego ia – In es igado
A iliação – LEPABE, Faculdade de Engenha ia da
Uni e sidade do Po o; G upo de Oncologia
Molecula e Pa ologia Vi al, CI-IPOP, Ins i u o
Po uguês de Oncologia do Po o
Co-O ien ado – P o esso Dou o Rui Medei os
Ca ego ia – P o esso Con idado Associado
A iliação – Ins i u o de Ciências Biomédicas de
Abel Salaza ; G upo de Oncologia Molecula e
Pa ologia Vi al, CI-IPOP, Ins i u o Po uguês de
Oncologia do Po o
Co-O ien ado – P o esso Dou o Manuel Vilano a
Ca ego ia – P o esso Associado
A iliação – Ins i u o de Ciências Biomédicas de
Abel Salaza
CARLOS EDUARDO REIS DOS SANTOS
MODULATING THE CD8
+
T CELL RESPONSE IN HUMAN
PAPILLOMAVIRUS-INDUCED LESIONS: STUDIES IN THE K14-
HPV16 TRANSGENIC MOUSE MODEL
II
III
Alguns dos esul ados incluídos nes a disse ação de mes ado o am subme idos
pa a publicação:
- San os C., Fe ei inha P., Sousa H., Ribei o J., Bas os M.M.S.M., Faus ino-Rocha
A.I., Oli ei a P.A., Medei os R., Vilano a M., Gil da Cos a R.M. (2015) Kine ics o CD8
+
T
cells in human papilloma i us-induced lesions – da a om K14-HPV16 ansgenic mice.
Submi ed o he jou nal Tumo Biology wi h he numbe TUBI-D-15-04094.
- San os C., Ne o T., Fe e inha P., Sousa H., Ribei o J., Bas os M.M.S.M., Oli ei a
P.A., Medei os R., Vilano a M., Gil da Cos a R.M. (2015) Celecoxib p omo es
deg anula ion o CD8
+
T cells in HPV-induced lesions o ansgenic mice. Submi ed o he
jou nal An i i al Resea ch wi h he numbe AVR-D-16-00002.
Alguns esul ados o am ambém publicados em a as de um cong esso in e nacional:
- San os C., Fe ei inha P., Sousa H., Ribei o J., Bas os M.M.S.M., Faus ino-Rocha
A.I., Oli ei a P.A., Medei os R., Vilano a M., Gil da Cos a R.M. (2015) Role o CD8
+
T
cells in HPV-induced skin cance : da a om a K14-HPV16 ansgenic mouse model;
P oceedings o he HPV2015 – 30
h
In e na ional Papilloma i us Con e ence & Clinical
and Public Heal h Wo kshops; 17
h
-21
s
Sep embe 2015, Lisbon, Po ugal. (Bes pos e
p esen a ion awa d)
IV
V
“When you wan o succeed as bad as you wan o b ea he hen you’ll be
success ul” – E ic Thomas

VI
VII
Ag adecimen os
Nes e momen o, que ma ca o inal de mais uma e apa da minha ida, não pode ia
deixa de ag edece às pessoas que de alguma o ma con ibuí am pa a a conclusão
des a e apa. Es e úl imo ano oi bas an e desa ian e e exigen e e odo o conhecimen o e
apoio que ecebi oi undamen al pa a chega a é aqui.
Em p imei o luga , gos a ia de ag adece à pessoa que me ecebeu e que me
acompanhou mais de pe o no desen ol imen o des e abalho, o meu o ien ado ,
P o esso Dou o Rui Gil da Cos a. Ag adeço a sua con iança em mim pa a le a a cabo
es e abalho, odos os conhecimen os que me ansmi iu, a disponibilidade pa a ajuda
no que osse p eciso, odos os conselhos e a cons an e boa disposição.
Ao P o esso Dou o Rui Medei os, meu co-o ien ado , ag adeço oda a ajuda
o necida ao longo des e ano e po udo o que me ensinou. Mesmo es ando cheio de
abalho a anja a manei a de me ecebe semp e que apa ecia no seu gabine e do IPO
pa a me pode ajuda no que eu p ecisasse e semp e bem-dispos o.
Ao P o esso Dou o Manuel Vilano a, ambém meu co-o ien ado , po udo aquilo
que con inua a ensina -me, pelo seu igo que an o ap ecio, pela disponibilidade que
a anja semp e que eu p eciso de ala consigo mesmo quando a quan idade de abalho
não o “deixa espi a ”.
A es e conjun o de o ien ado es ag adeço odas as discussões das quais su gi am
ideias de abalho, ag adeço udo o que me ensina am e, acima de udo, ag adeço a
comp eensão e o apoio que semp e demons a am pe an e a minha ida de es udan e-
a le a. Não só comp eendem como ambém alo izam o que aço e é ealmen e bom
sen i esse apoio.
Ao Ped o Fe ei inha, mais do que um co-o ien ado no passado, um amigo e
colega in es igado no p esen e. Mui o do que ap endi sob e abalha num labo a ó io
de o-o a ele. Pa a além de udo o que me ensinou e con inua a ensina es á semp e
dispos o a ajuda no que o p eciso. Não só é dedicado ao abalho como ambém é uma
pessoa óp ima pa a i oma um ca é e e uma con e sa animada.
Às es an es pessoas que abalham no Labo a ó io de Imunologia Má io A ala
Cha es (ICBAS), em especial à Enca nação Rebelo, Joana Al es e Alexand a Co eia,
ag adeço o ca inho e a boa disposição de odos os dias, a disponibilidade pa a ajuda no
que p ecisasse e as con e sas aleg es e en iquecedo as.
À P o esso a Paula Oli ei a, à Ana Faus ino-Rocha e ao Tiago Ne o ( odos da
Uni e sidade de T ás-os-Mon es e Al o Dou o) ag adeço odo o apoio que me de am
du an e a ealização dos abalhos expe imen ais.
VIII
Po úl imo, mas mui o impo an es, a amília e os amigos o mam um conjun o de
pila es que se em de apoio nos momen os mais complicados e de di e são quando a
ocasião o pe mi e. Àqueles que me acompanha am mais de pe o du an e es e ano
mesmo, em alguns casos, es ando longe (Tiago Sil a, Bea iz Sil a, Helena Pinhei o,
Ma iana Pe ei a, Inês Sousa, Hildebe o Mo ei a, Ana Rosa, Luciana Lei e, Ri a Fa ia,
Miguel Gonçal es, Cá ia Oli ei a, Joana Pin o, Rena a Lopes, Diogo Edi, Isabel Pai a,
Ca a ina Pe ei a e, cla o, aos meus colegas de equipa) ag adeço pela o ça que me
de am quando as coisas pa eciam não co e bem, a mo i ação que me passa am
quando es a me al a a, as con e sas en iquecedo as sob e o abalho, os momen os de
descon ação, ga galhada e di e são quando es es e am possí eis e, p incipalmen e, po
se em e dadei os amigos e comp eende em os sac i ícios que aço em p ol dos meus
objec i os e ambições enquan o es udan e-a le a.
Aos meus pais e i mã, ag adeço po udo aquilo que não é possí el ag adece o
su icien e. Ag adeço pela educação que me de am e po udo o que me ensina am, po
odas as condições que semp e me de am pa a o meu c escimen o, pelas chamadas de
a enção, po exigi em o melho de mim, po me apoia em em odas as decisões, po me
o ien a em. Se hoje posso dize que enho o gulho no que me o nei de o-o in ei amen e
a eles. Ob igado.
IX
Index
Ag adecimen os .............................................................................................................. VII
Index ................................................................................................................................ IX
Abb e ia ions ................................................................................................................... XI
Resumo ......................................................................................................................... XIII
Abs ac .......................................................................................................................... XV
Chap e 1: In oduc ion ..................................................................................................... 1
The Basis o HPV In ec ion and Cell T ans o ma ion ..................................................... 4
HPV In ec ion s. Hos Immune Response .................................................................... 5
HPV-associa ed Ch onic In lamma ion .......................................................................... 6
HPV-induced Ca cinogenesis: The Bes Models o Unde s and I ................................. 9
K14-HPV16 T ansgenic Mice as a Model o In lamma ion-associa ed Ca cinogenesis12
Modula o s o he immune esponse and hei e ec o e CD8
+
T cells ........................15
Chap e 2: Objec i es ......................................................................................................17
Chap e 3: Kine ics o CD8
+
T cells in human papilloma i us-induced lesions..................19
In oduc ion ..................................................................................................................21
Ma e ial & Me hods ......................................................................................................22
Animals .....................................................................................................................22
Mice geno yping........................................................................................................22
S udy Design ............................................................................................................22
His ology ...................................................................................................................22
P epa a ion o Single-Cell Suspensions ....................................................................23
Immunopheno yping .................................................................................................23
S a is ical Analysis ....................................................................................................23
Resul s .........................................................................................................................24
T ansgenic Mice Show Epide mal Hype plasia and Dysplasia ..................................24
Inc eased CD8
+
T lymphocy es numbe s and ac i a ion in HPV16
+/-
mice ................24
Discussion ....................................................................................................................26
Conclusion ...................................................................................................................27
Chap e 4: P aquiloside inhibi s umou -in il a ing CD8
+
T cells in HPV- ansgenic mice .29
In oduc ion ..................................................................................................................31
Ma e ial & Me hods ......................................................................................................32
Mice ..........................................................................................................................32
XVI

1
Chap e 1: In oduc ion
2
3
Human papilloma i us (HPV) is he pa hogen esponsible o he mos common
sexually ansmi ed in ec ion among bo h men and women in he Uni ed S a es [1]. Mos
sexually ac i e people will be in ec ed by HPV du ing hei li e. The as majo i y o HPV
in ec ions in immunocompe en indi iduals a e clea ed by he immune sys em, bu in
app oxima ely 10% o cases [2, 3] he pe sis ence o his in ec ion will cause he
de elopmen o disease and p og ession o cance . HPV has opism o skin and mucosal
epi helia especially om he anogeni al and o opha yngeal a eas. High- isk HPVs (mos ly
ypes 16, 18, 31 and 33) ac as c ucial e iologic agen s in ce ical ca cinoma [4] – which is
he six h mos equen malignancy in women wo ldwide [5] – and a e he cause o some
head and neck neoplasms whe e incidence depends on he coun y and o he associa ed
isk ac o s like obacco [6, 7].
Since 1863 [8], when Vi chow i s hypo hesised a link be ween in lamma ion and
cance , many s udies ha e been ca ied ou in o de o unde s and he unde lying
mechanisms o his ela ion and i s in e enien s. The impo ance o his ela ion was
u he ein o ced by he awa d o he Nobel P ize o Robin Wa en and Ba y Ma shall o
hei wo k in disco e ing Helicobac e pylo i as he e iologic agen o in lamma ion-
associa ed gas ic cance [9]. The umou mic oen i onmen is cha ac e ised by he
p esence o a a ie y o leucocy es which ha e been shown o be a ac ed by
chemokines p oduced by umou and s omal cells [10]. Lymphocy es a e an impo an
pa o in lamma o y in il a es in he umou mic oen i onmen [8, 11] and di e en
lymphocy ic popula ions can play di e en oles. Fo ins ance, T helpe (Th) 1, Tγδ and
cy o oxic T cells a e associa ed wi h disease- ee su i al. In con a y, Th2, Th17 and T
egula o y (T eg) cells a e allied wi h a poo e ou come [12]. Neu ophils, umou -
associa ed mac ophages (TAMs), dend i ic cells (DCs) and mas cells ep esen o he
leucocy e popula ions also equen ly ound in he umou en i onmen ha may play dual
oles du ing ca cinogenesis [8, 13]. These umou -associa ed in lamma o y cells, oge he
wi h he umou cells, p oduce and sec e e a wide se o cy okines (e.g.: umou nec osis
ac o (TNF)-α, in e leukin (IL)-1, IL-6 and monocy e colony-s imula ing ac o ) and
chemokines (e.g.: monocy e chemo ac ic p o ein-1) ha may ei he help umou
p og ession o con ain i [14-17].
In he case o HPV-associa ed ca cinogenesis, he i us in ec s ke a inocy es and
an immunological esponse a ises as cy okines (mos ly TNF-α, IL-1 and ype I and II
in e e ons – IFNs) a e eleased by he in ec ed ke a inocy es and local immune cells [18].
Howe e , HPVs may a oid he e ec o unc ions o p o-in lamma o y cy okines by
di e en mechanisms [19-22]. Thus, i he in ec ion pe sis s, he chance o neoplas ic
ans o ma ion ises [23]. Besides i s impo ance in he ea ly s ages o ca cinogenesis,
4
inna e and adap i e immune esponses a e c i ical ac o s in ol ed in malignan
p og ession.
The Basis o HPV In ec ion and Cell T ans o ma ion
Human papilloma i uses ha e hei genome di ided in h ee main egions: he
non-coding uppe egula o y egion, he ea ly egion encoding six known oncop o eins
and he la e egion, which encodes he capsid p o eins. The oncop o eins encoded in he
ea ly egion, mainly E6 and E7 bu also E5, a e key pa icipan s in he in ec ious cycle and
a e he sou ce o he gene ic al e a ions which po en ially lead o HPV-associa ed
ca cinogenesis as e iewed elsewhe e [24]. The E5 oncogene is belie ed o ha e low
impac in he main enance o he malignan pheno ype o ce ical cance cells as i
appea s o be dele ed in a ious ca cinomas [25]. Ne e heless, when p esen , i akes an
impo an pa in mul iple mechanisms a ou ing neoplas ic ans o ma ion. Some in i o
s udies ha e shown ha he HPV16 E5 oncop o ein can p omo e cell cycle p og ession
and DNA syn hesis by down- egula ing he exp ession o umou supp esso p o eins p21
and p27 [26, 27]. Mo eo e , his oncop o ein is cen al o e asion o he hos
immunosu eillance. In E5-exp essing cell lines an igen-p esen ing majo
his ocompa ibili y complex (MHC) class I molecules appea o be apped in he Golgi
appa a us which educed hei su ace exp ession [28, 29]. The down- egula ion o MHC
class I on cells exp essing E5 ul ima ely impai s hei ecogni ion and clea ance by
cy o oxic T lymphocy es (CTL). The ea ly genes E6 and E7 a e known o ha e g ea
oncogenic po en ial. They play a pi o al ole in HPV-media ed ca cinogenesis as hey
eac wi h impo an umou supp esso p o eins impai ing hei unc ions. Thus, E6 binds
o and p omo es he deg ada ion o he p53 umou supp esso p o ein [30, 31], impai ing
impo an mechanisms o de ence such as DNA epai and apop osis. Addi ionally, he E6
p o ein acili a es ca cinogenesis by educing cellula senescence (as a esul o
elome ase ac i a ion [32]), ac i a ing he NF-κB pa hway and e ading he in e e on
esponse [24]. The E7 oncop o ein binds o umou supp esso p o eins o he
e inoblas oma amily o p o eins (pRb) and induces hei inac i a ion [33, 34], he e o e
inc easing DNA syn hesis and boos ing cellula p oli e a ion. Fu he mo e, E7 is able o
p omo e genomic ins abili y, which inc eases he likelihood o malignan p og ession [24].
Toge he , he E6 and E7 oncogenes a e capable o ans o ming and induce he
immo aliza ion o ke a inocy es [35]. Un il oday, he h ee ea ly p o eins men ioned
abo e appea o be, o all HPV p o eins, he ones wi h highe oncogenic po en ial. Gi en
hei oles in e ading hos immuni y, i is impo an o unde s and how he human immune
sys em esponds o HPV in ec ion.
5
HPV In ec ion s. Hos Immune Response
Along he yea s, mammals ha e g own complex inna e and adap i e immune
mechanisms o con ol au oimmuni y and o p o ec he hos agains mic obial in ec ions.
Vi uses can be de ec ed by cellula ecep o s gene ally designa ed as pa e n ecogni ion
ecep o s (PRRs), which igge inna e immune mechanisms ha subsequen ly may be
ollowed by a mo e speci ic adap i e immune esponse. These cellula ecep o s a e
p o icien a ecognising se e al pa hogen-associa ed molecula pa e ns, like i al DNA o
RNA o bac e ial cell wall o in acellula componen s, and also in e ac ing wi h dange -
associa ed molecula pa e ns like in acellula molecules eleased om damaged cells
and cells unde going unp og ammed cell dea h [36]. HPVs a e double-s anded DNA
i uses ha a e mos likely o be ecognized by Toll-like ecep o 9 (TLR-9). This ecep o
is in acellula ly exp essed in he memb ane o endolysosomes and ecognizes
unme hyla ed CpG-DNA sequences ha bou ed by pa hogens like i uses, bac e ia and
p o ozoa [36]. Howe e , HPVs appea o somehow su pass hese PRRs. The
oncop o eins E6 and E7 may p omo e he down- egula ion o TLR-9 as has been shown
in i o o HPV16 [37]. The e o e, a he s a , he HPV in ec ion seems o occu wi hou
he awa eness o he hos immune sys em [38]. Fu he mo e, in case o a pe sis en
in ec ion, he i us impai s immune cell unc ions as well as se e al molecula pa hways
in ol ed in he immune esponse [39]. Also, o he inna e immune mechanisms, like he
IFN esponse, can be ab oga ed by he ac ions o some HPV oncop o eins, as men ioned
abo e.
HPV is ansmi ed h ough skin- o-skin con ac , usually h ough mic o wounds.
The e exis s a a ie y o immune cells di ided be ween he de mis and epide mis. In he
epide mis, he main cellula popula ion a e ke a inocy es bu he e can also be ound
Lange hans cells and T lymphocy es [39]. Ke a inocy es, he main a ge s o HPV, a e
capable o some immune unc ions. These cells can p oduce and elease cy okines,
induce he ac i a ion o memo y T cells and may also ac as non-p o essional an igen-
p esen ing cells (APCs) [40]. Hence, ke a inocy es a e o g ea impo ance in he ini ia ion
o an immune esponse agains HPV in ec ions. On he o he hand, he majo i y o he so-
called “classical” immune cells (mac ophages, DCs, T lymphocy es and o he s) can be
ound in he de mis [39]. Bo h ke a inocy es and immune cells on si e a e capable o p o-
in lamma o y cy okine p oduc ion [41] and an igen p esen a ion in esponse o se e al
pa hogens. Al hough he inna e immune esponse o HPV in ec ions emains poo ly
unde s ood, he immunological p ocesses desc ibed abo e a e hough o be he main
ones esponsible o he ini ia ion o an adap i e immune esponse, which will de e mine
he clea ance o pe sis ence o he in ec ion. In ac , only abou 10% o HPV in ec ions will

6
de elop clinical symp oms and possibly e ol e o cance [2, 3]. The e o e, wa s and low
g ade ce ical in aepi helial neoplasms (CIN1) a e in mos cases o ced o eg ess by he
cell-media ed adap i e immune esponse [42]. I has been shown ha CD4
+
and CD8
+
HPV16-speci ic T cells mig a e o he skin a e in ade mal challenge wi h HPV16
pep ides [43]. This cellula mig a ion e en appea s o be ollowed by an an ibody-
media ed esponse, as obse ed in HPV-in ec ed women [44]. S ill, an ibody i es a e low
and many women may e en no p oduce hem.
HPV-associa ed Ch onic In lamma ion
The abo e desc ibed mechanisms a e hough o explain how a human immune
sys em eac s since he beginning o a HPV in ec ion un il he momen o immunological
clea ance. Howe e , as seen ea lie , HPV oncop o eins ha e ways o e ading he hos
immune esponse. This e asion a ou s he pe sis ence o he in ec ion, which in u n may
p omo e umou igenesis.
In his con ex , he cellula con en o he in lamma o y in il a e is decisi e o
ei he neoplas ic ans o ma ion o elimina ion o in ec ed/ ans o med cells (Fig. 1). While
some cellula popula ions suppo HPV’s immune e asion and cell ans o ma ion
mechanisms, o he s ha e a ole in inhibi ing malignan p og ession o e en in po en ia ing
umou eg ession. Fo ins ance, T egula o y (T eg) cells acili a e he i al immune
e asion. These cells display he CD4
+
CD25
+
FoxP3
+
pheno ype and play an
immunosupp essi e ole impo an o p e en sel -agg ession by he immune sys em [45].
Because o hei immunosupp essi e unc ion, T eg cells a e o en associa ed o
pe sis en in ec ion and umou p og ession. Indeed, s udies ha e shown a co ela ion
be ween inc eased equencies o FoxP3
+
T cells a bo h sys emic [46] and ce ical [47]
le el and HPV pe sis en in ec ion. Fu he mo e, a s udy has shown he chemokine
CXCL12 (which is no exp essed in no mal skin) o ac as a chemoa ac an o T eg cells
[48]. This wo k shows e idence ha HPV s imula es an inc eased p oduc ion o CXCL12
which d aws a ising numbe o FoxP3
+
cells. Besides T eg cells, ano he subse o CD4
+
T cells can ha e p o- umou unc ions. Type-2 helpe T (Th2) cells can also con ibu e o
impai he p o ec i e immune esponse, p edominan ly o he Th1 ype, in he umou
mic oen i onmen as hey p oduce IL-4, -5 and -13, as well as IL-10. As seen ea lie , his
ype o immune esponse acili a es malignan p og ession [12]. Fu he mo e,
mac ophages a e hough o be a majo componen o umou in il a es [8] and a e highly
equen in HPV-associa ed umou s [49]. In he p esence o Th2- ype cy okines like IL-4
and IL-13, umou -associa ed mac ophages (TAMs) a e al e na i ely ac i a ed and
acqui e he M2-pheno ype [50]. Unlike he M1 mac ophages, hese M2 cells exp ess
7
a ious ea u es and unc ions ha a e commonly associa ed wi h immune egula ion and
umou p og ession. Fo ins ance, M2 TAMs, oge he wi h mas cells, a e an impo an
sou ce o g ow h ac o s (e.g.: ascula endo helial g ow h ac o – VEGF), cy okines (e.g.:
TNF-α, ans o ming g ow h ac o -β), p o eases (e.g.: ma ix me allop o einases) and
o he molecules which in u n p omo e angiogenesis and issue emodelling [51, 52].
These e en s acili a e he in ake o nu ien s and p o ide o he s imuli which help umou
g ow h, in asion and me as asis. On he o he hand, TAMs can be classically ac i a ed
upon mic obial s imuli o induced by p o-in lamma o y cy okines, such as IFN-γ, o play
impo an an i- umou unc ions [50]. M1- ype TAMs can, he e o e, ac as induce and
e ec o cells in Th1- ype immune esponses agains umou s. They usually display an IL-
12
high
, IL-23
high
pheno ype and p oduce high le els o p o-in lamma o y cy okines (IL-1β,
IL-6 and TNF-α) and eac i e ni ogen species [53]. Besides M1 TAMs, o he cellula
popula ions play impo an an i- umou oles among he la gely he e ogeneous umou
in lamma o y in il a es. T lymphocy es a e also impo an immune e ec o s. Type-1
esponses, media ed by Th1 cells ep esen he p o o ypic esponse adequa e o
e adica e cance cells. The di e en ia ion o his ype o CD4
+
T cells is d i en by IL-12
and hey a e cha ac e ised by he p oduc ion o p o-in lamma o y cy okines like IFN-γ. Th1
cance cells can di ec ly induce he killing o umou cells h ough TNF- ela ed apop osis-
inducing ligand (TRAIL) and/o Fas Ligand (FasL) pa hways [54] and can induce he
cy o oxic ac i i y o CD8
+
T lymphocy es and o he e ec o cells h ough he p oduc ion o
speci ic cy okines [55]. In u n, CD8
+
T cells, which may di e en ia e in o cy o oxic T
lymphocy es (CTL), a e commonly p o ec i e agains in acellula pa hogens, like i uses,
and umou cells. These lymphocy es we e obse ed in highe equency in HPV-posi i e
ca cinomas when compa ed wi h HPV-nega i e lesions [56]. Fu he , his di e ence
co ela es wi h he pa ien s’ p ognosis as a highe CD8
+
T cell equency is associa ed
wi h augmen ed o e all su i al. Mo eo e , “na u al kille ” (NK) cells and NKT cells ac
simila ly o CTL bu a e pa o he inna e immune esponse. Unlike CTL, NK cells do no
equi e he p esence o MHC I molecules on a ge cells o exe hei cy o oxic unc ion.
Thus, i HPV success ully down- egula es MHC I exp ession, he eby a oiding he ac ion
o CTL, in ec ed cells may s ill be a ge ed by NK cells. Like CTL, NK and NKT cells
exp ess IFN-γ, pe o in and g anzyme and a e hus capable e ec o s in he an i- i al
immune esponse [39] as well as being able o also pa icipa e in he killing o cance cells
[57]. Addi ionally, se e al APCs (like DCs and Lange hans cells) a e o g ea alue and
a e equen in he umou mic oen i onmen . They a e essen ial o he ac i a ion o
e ec o cells h ough an igen p esen a ion and cy okine p oduc ion [13].
8
Fig.
1
:
Schema ic ep esen a ion o he immune cell in il a e in HPV-
induced ca cinogenesis. F om op o bo om, HPV
in ec ion induces neoplas ic ans o ma ion
wi h associa ed issue emodelling e en s which lead o hos immune esponse.
This esponse could ei
he con ibu e o disease p og ession o eg ession which is e lec ed in he cellula popula ions
ec ui ed in ei he case. CD8
+
, Th1- ype CD4
+
T, NK cells and M1 mac ophages a e commonly associa ed wi h umou
eg ession while, T eg, Th2- ype CD4
+
T cells and M2 mac ophages a e usually ela ed o malignan p og ession.
9
The c oss- alk o immune cells wi h HPV-in ec ed cells is s ill no ully unde s ood.
Howe e , i is gene ally ag eed ha he in lamma ion associa ed wi h he i al in ec ion
acili a es he de elopmen o HPV-associa ed cance s. In gene al e ms, egula o y T
cells, M2 TAMs, mas cells and Th2 cells mos likely ep esen he immune cells wi h he
g ea es con ibu ion o pe sis en in ec ions and malignan p og ession. Addi ionally,
HPV- ans o med ke a inocy es we e shown o p oduce he immunosupp essi e cy okine
IL-10 in HPV-associa ed ce ical cance [58], which may be igge ed by he i us
oncop o eins as an immune e asion mechanism. On he o he hand, M1 mac ophages,
NK and NKT cells, CTL and Th1 cells a e, wi h he suppo o APCs, he majo
pa icipan s in an i- i al esponses and umou eg ession. S ill, i is impo an o
emphasize ha his whole in e ac ion in he umou milieu is g ea ly complex due o he
he e ogenei y o cellula popula ions he ein, which likely jus i ies he di icul y in
cha ac e ising he ela ionship be ween in lamma o y cells and HPV in ec ion.
HPV-induced Ca cinogenesis: The Bes Models o Unde s and I
Along he yea s, scien is s ha e been imp o ing ou unde s anding o ch onic HPV
in ec ion and i s associa ion wi h ca cinogenesis. In i o and in i o animal models a e
key elemen s owa ds he e olu ion o scien i ic knowledge (Table 1) and he in i o ones
a e al eady widely used in HPV-induced ca cinogenesis esea ch. HPV- ela ed in i o
s udies include, in mos cases, he use o single-laye cul u es o common ce ical
ca cinoma cell lines (HeLa, SiHa, CaSki and o he s). In he las wo decades, a mo e
sophis ica ed ype o in i o assay was de eloped and used as an al e na i e model.
O gano ypic (o a ) cul u es allow he p oli e a ion and di e en ia ion o epi helial cells a
an ai -liquid in e ace on a de mal-equi alen suppo [59-61]. Thus, hese a cul u es a e
use ul o s udy he e en s occu ing in human s a i ied epi helia in he cou se o HPV
in ec ion and HPV-induced ca cinogenesis.
Alongside wi h he cell cul u e echnique, in i o animal models a e ex ensi ely
exploi ed o scien i ic pu poses. On one hand, bo ine ca le, dogs and abbi s can be
used o s udy bo ine, canine o al and co on ail abbi papilloma i uses, espec i ely [62,
63]. The s udy o hese animal papilloma i uses is impo an as hey a e e iologic agen s
o diseases in a m and companion animals, p o ide in i o models o HPV esea ch and
help o disco e new subjec s o s udy on HPV. Howe e , hese la ge animals a e no
common objec s o s udy because hey a e associa ed wi h a g ea deal o inancial cos s
and e hical issues. In u n, he mouse is he mos commonly used es subjec in heal h-
associa ed s udies. Mice display a g ea simila i y o humans in e ms o ana omy,
physiology, including he immune one, and gene ics which, in addi ion o i s cos -
16
[123]. B acken oxins a e hypo hesized o d i e he malignan p og ession o
papilloma i us-induced uppe diges i e lesions in ca le and human popula ions, as
ema ked by Chang e al. [124]. P aquiloside is a chemically uns able compound ha has
been shown o induce splenic whi e pulp a ophy, neu openia, educed NK cells ac i i y
and a B-cell lymphop oli e a i e malignancy [125-128]. Howe e , un il now, he e a e no
epo s o an associa ion be ween p aquiloside and CD8
+
T lymphocy es. In u n,
celecoxib (CXB) is an immunomodula o y d ug which speci ically inhibi s cyclooxygenase-
2 (COX-2). This enzyme plays a key ole in he de elopmen o an in lamma o y esponse
[129] and is o e -exp essed in se e al ypes o cance [130-134]. By selec i ely inhibi ing
COX-2, celecoxib diminishes he p oduc ion o p os aglandins wi hou signi ican ly
impai ing he COX-1 iso o m [135]. Ac ually, a ew s udies ha e al eady enligh ened he
e ec o his selec i e COX-2 inhibi o o e cy o oxic T cells [136-140]. Howe e , whe he
celecoxib enhances o dec eases CD8
+
T lymphocy es in il a ion and ac i i y a ies
be ween s udies. S ill, none o hese s udies analysed he e ec o celecoxib o e CTL in
HPV-associa ed lesions.

17
Chap e 2: Objec i es
The pu pose o his wo k was o assess he in il a ion and ac i i y o cy o oxic
CD8
+
T lymphocy es in HPV-induced skin lesions, using he K14-HPV16 mouse model.
Thus, he wo k was ocused in some mo e speci ic aims:
- To pe o m a kine ic s udy o he CTL in il a ion and ac i a ion by compa ing
lesions om di e en -aged mice;
- To analyse he immunomodula o y e ec o p aquiloside o e CD8
+
T cell
in il a ion and ac i a ion;
- To analyse he immunomodula o y e ec o celecoxib o e CD8
+
T cell in il a ion
and ac i a ion.
18
19
Chap e 3: Kine ics o CD8
+
T cells in human
papilloma i us-induced lesions
1
1
The con en s o his chap e we e adap ed om:
- San os C., Fe ei inha P., Sousa H., Ribei o J., Bas os M.M.S.M., Faus ino-Rocha A.I.,
Oli ei a P.A., Medei os R., Vilano a M., Gil da Cos a R.M. (2015) Kine ics o CD8
+
T cells in
human papilloma i us-induced lesions – da a om K14-HPV16 ansgenic mice. Submi ed o he
jou nal Tumo Biology wi h he numbe
TUBI-D-15-04094.
20
21
In oduc ion
Human papilloma i us (HPV) is he main e iologic agen o ce ical cance [4], bu
can also o igina e anal, skin and head and neck malignancies [6, 141, 142]. Only a
pe sis en in ec ion by an oncogenic HPV ype (mainly HPV 16 and 18) can induce
malignan ans o ma ion. In app oxima ely 90% o HPV in ec ed indi iduals he hos
immune sys em is able o igh he i us and hinde i s dissemina ion. Howe e , in he
emaining 10%, HPV can e ec i ely e ade immune de enses and lead o clinical disease
[2, 3].
K14-HPV16 ansgenic mice, c ea ed nea ly wo decades ago [73], a e an use ul in
i o animal model o he s udy o HPV-induced ca cinogenesis. This model sha es a
numbe o mo phologic and molecula simila i ies o HPV- ela ed human disease [76],
hus unc ioning as an excellen eplica o he mul i-s age p ocess o ca cinogenesis.
Ta ge ing o HPV16 oncogenes o ke a inocy es by he ke a in-14 (K14)
p omo e /enhance is he key cha ac e is ic o his model [73].
HPV-induced ca cinogenesis is associa ed wi h p og essi ely in ense ch onic
in lamma ion. The e o e, a g ea di e si y o immune cells and a mul iplici y o soluble
media o s can be ound wi hin he umou mic oen i onmen [18]. The in lamma o y
in il a es p o ide p o- and an i- umou s imuli, which will a ou ei he he de elopmen o
he eg ession o he lesion [55, 143].
CD8
+
T cells a e es ic ed o majo his ocompa ibili y complex class I (MHC I)
molecules which can p esen pep ides gene a ed om in acellula i uses and/o umou
cells [117]. Upon ecognizing speci ic pep ides p esen ed on MHC I molecules on he
su ace o p o essional an igen p esen ing cells, which also display co-s imula o y
molecules, hese T cells can di e en ia e in o cy o oxic T lymphocy es (CTL), helped by
cy okine s imuli [116]. CTL a e an impo an pa o umou -speci ic immuni y. They may
elimina e a ge cells ei he di ec ly, h ough he elease o ly ic g anules con aining
se e al enzymes (such as pe o in and g anzyme) o by he engagemen o dea h
ecep o s (e.g.: FAS-FASL), and indi ec ly, ollowing an in e e on-γ-dependen
mechanism which leads o cell cycle inhibi ion, apop osis and s imula ion o mac ophage
an i- umou ac i i y [118, 119].
Thus, using he K14-HPV16 ansgenic mouse model, he aim was o examine he
kine ics o CD8
+
T cell in il a ion in HPV-induced lesions du ing mul i-s ep ca cinogenesis
a di e en ime poin s. The goal was also poin ed o unde s and whe he hese cells we e
ac i a ed by de e mining he p esence o he lysosome-associa ed memb ane p o ein 1
(LAMP1), also known as CD107a, a he cell su ace.

22
Ma e ial & Me hods
Animals
Gene a ion o K14-HPV16 mice on a FVB/n backg ound has been p e iously
epo ed [73]. K14-HPV16 ansgenic mice we e gene ously dona ed by D s. Je ey A bei
and Douglas Hanahan (Uni e si y o Cali o nia) h ough he USA Na ional Cance Ins i u e
Mouse Reposi o y. The animal expe imen s we e app o ed by he Uni e sidade de T ás-
os-Mon es e Al o Dou o e hics commi ee (10/2013) and he Po uguese Ve e ina y
Di ec o a e (0421/000/000/2014). Animals we e main ained and b ed acco ding o
Po uguese (Dec e o-Lei 113/2013, Augus 7
h
) and Eu opean (EU Di ec i e 2010/63/EU)
legisla ion, unde con olled condi ions o empe a u e (23 ± 2 ºC), ligh -da k cycle (12h
ligh /12h da k) and ela i e humidi y (50 ± 10 %), using ha dwood bedding. Food and
wa e we e p o ided ad libi um.
Mice geno yping
Animals we e geno yped a weaning, using ail ip samples as desc ibed p e iously
[113, 114]. B ie ly, nucleic acids we e ex ac ed and DNA quali y and pu i y we e
assessed. HPV16-E6 and -E2 genes we e ampli ied o con i m he p esence o HPV DNA
and mouse β-globin was used as con ol. Amplicons leng hs we e con i med by aga ose
gel elec opho esis. Only hemizygous emales we e used o he ansgenic mouse g oups
in he expe imen .
S udy Design
Ten wild- ype (WT) (HPV16
-/-
, G oup 1) and wel e ansgenic (HPV16
+/-
, G oup 2)
emale mice we e eu hanized a 24-26 weeks o age. La e , six HPV16
-/-
(G oup 3) and
six HPV16
+/-
(G oup 4) emale mice we e sac i iced when 28-30 weeks-old. The mice we e
humanely sac i iced by in ape i oneal pen oba bi al o e dose, ollowed by exsanguina ion
h ough ca diac punc u e. Ches skin samples (app oxima ely 4 cm
2
) we e collec ed o
cell isola ion and low cy ome y analysis. Ma ched samples we e collec ed o his ological
examina ion.
His ology
Skin samples we e ixa ed in 10% neu al bu e ed o malin o 48h. Samples we e
dehyd a ed h ough g aded alcohols and xylene and pa a in-embedded in an au oma ic
STP 120 p ocesso (Mic on, Boise, ID). 2 µm- hick sec ions we e s ained wi h
haema oxylin-eosin (H&E) o his ological examina ion on ligh mic oscopy. Skin samples
23
we e classi ied as no mal skin, epide mal hype plasia o epide mal dysplasia. Mino
dysplas ic oci on a hype plas ic backg ound we e also eco ded.
P epa a ion o Single-Cell Suspensions
Ches skin samples we e cleaned om excessi e blood essels and a issue and
cu in o small pieces. These agmen s we e incuba ed o 2 hou s wi h 125 U/ml ype I
collagenase (Gibco, Li e Technologies, Paisley, UK) in RPMI-1640 medium
complemen ed wi h 1% glu amine, 1% penicillin-s ep omycin-ampho e icin B, 1% HEPES
bu e (all om Sigma, S . Louis, MO) and 10% oe al bo ine se um (BioWes , Nuaillé,
F ance) a 37 ºC and 150 pm in a 3031 o bi al incuba o (GFL, Bu gwedel, Ge many).
The esul ing cell suspension was hen il e ed and cen i uged a 300  o 10 min a 4 ºC.
Cells we e hen essuspended in phospha e bu e ed saline con aining 1% bo ine se um
albumin and 20 mM sodium azide ollowed by ex acellula s aining.
Immunopheno yping
Following cell isola ion, he su ace pheno ype o he collec ed cells was assessed
by low cy ome y using speci ic monoclonal an ibodies (mAb). To p e en non-speci ic
an ibody binding, cells we e incuba ed wi h an i-mouse CD16/CD32 mAb o FcγR
blocking. This was ollowed by incuba ion wi h an i-CD8 mAb phycoe y h in-cych ome 5-
conjuga e (clone 53-6.7, BD Biosciences, San Diego, CA) and an i-CD107a mAb
phycoe y h in-conjuga e (clone eBio1d4b, eBioscience, San Diego, CA). Following
ex acellula s aining, he cells we e washed, ixed in 2% o maldehyde and washed wi h
phospha e bu e ed saline con aining 1% bo ine se um albumin and 20 mM sodium azide.
An ibody-labelled cells we e analysed in an EPICS XL low cy ome e using he
EXPO32ADC so wa e (Beckman Coul e , Miami, FL). The collec ed da a iles we e
analysed using he FlowJo so wa e 10.0.7 (FLOWJO, LLC, Ashland, OR).
S a is ical Analysis
S a is ical analyses we e execu ed in he G aphPad so wa e ( e sion 6.0,
G aphPad So wa e, Inc. La Jolla, CA). S a is ical analysis be ween g oup pai s was
pe o med using he Mann-Whi ney es .
24
Resul s
T ansgenic Mice Show Epide mal Hype plasia and Dysplasia
Analysis o skin samples showed ha he o ali y o WT mice (g oups 1 and 3)
p esen ed no mal skin his ology (Fig. 2a) whils all ansgenic mice p esen ed skin lesions.
In all 12 mice composing g oup 2 i was possible o obse e simple o papilla y, di use,
a iably se e e epide mal hype plasia and papilloma osis wi h o hoke a o ic
hype ke a osis (Fig. 2b). In lamma ion was mild, wi h a ew mac ophages, lymphocy es
and mas cells p esen in he supe icial de mis. Also in his g oup, 2 animals (16.7%)
p esen ed small epide mal dysplas ic oci. In g oup 4, 3 animals (50.0%) p esen ed di use
epide mal dysplasia (Fig. 2c); o he 3 (50.0%) showed mul i ocal epide mal dysplasia in a
hype plas ic backg ound. Sub-epide mal angiogenesis and de mal in lamma o y in il a es
we e p ominen , showing nume ous mixed mononuclea leukocy es and neu ophils.
Fig. 2:
His opa hological changes induced by HPV16 oncogenes in FVB/n mice, H&E. a – WT animal; No mal skin
his ology,  . b – 24-26 week-old HPV16
+/-
animal; Epide mal hype plasia ex ending o he ollicula
in undibulum and is hmus,  . c – 28-30 week-old HPV16
+/-
animal; Epide mal dysplasia,  . No e enhanced
sub-epide mal in lamma o y cell in il a ion.
Inc eased CD8
+
T lymphocy es numbe s and ac i a ion in HPV16
+/-
mice
In o de o assess he p esence o CD8
+
T cells in HPV-associa ed lesions,
lymphoid cells we e isola ed om ches skin issue and analysed by low cy ome y (Fig.
3a). As shown in Fig. 3b, ches skin samples om HPV16
+/-
mice (g oups 2 and 4)
p esen ed a signi ican ly highe pe cen age o CD8
+
T cells when compa ed wi h hose o
WT animals (g oups 1 and 3). Al hough he pe cen age o skin CD8
+
T cells was sligh ly
highe in g oup 4 han in g oup 2, i did no each s a is ical signi ican di e ence.
In o de o de e mine i he CD8
+
T cells ound in he skin o HPV16
+/-
mice
p esen ed e idence o cy o oxic ac i i y, he su ace exp ession o CD107a was
e alua ed. In CD8
+
T lymphocy es CD107a eaches he cell su ace when ly ic g anules
su e exocy osis, hus exposing i s memb ane p o eins [144]. The e o e, his lysosome-
associa ed memb ane p o ein is a commonly used su oga e ma ke o CTL deg anula ion
[145]. As shown in Fig. 3c, he pe cen age o CD107a
+
CD8
+
T cells in ches skin samples
25
o mice om g oups 2 and 4 we e ound signi ican ly highe when compa ed wi h
espec i e con ols o g oups 1 and 3. This indica es ha in he HPV16
+/-
mice skin
in il a ing CTL eleased cy o oxic g anules. Mo eo e , in g oup 4 mice, 100% o which
show mul i ocal o di use dysplasia, a ma kedly highe pe cen age o CD8
+
T cells
exp ess CD107a (P  0.001) han in g oup 2 mice, o which only 16.7% show ocal
dysplas ic lesions. This esul shows a posi i e co ela ion in he p opo ions o ac i a ed
CD8
+
T cells and lesion se e i y.
Fig. 3:
Pe cen age o CD8
+
and CD8
+
CD107a
+
T cells wi hin o al lymphoid-ga ed cells ob ained om ches skin
samples om WT and K14-HPV16 ansgenic mice. a – Rep esen a i e analysis o he ga ing s a egy employed.
Numbe s wi hin g aphs co espond o he pe cen age o he ga ed popula ion. b – Pe cen ages o CD8
+
T cells and
c – pe cen ages o CD8
+
CD107a
+
T cells in ga ed CD8
+
T cells we e de e mined by low cy ome y a e skin issue
collec ion and diges ion wi h collagenase. Ches skin samples we e collec ed om 24-26 weeks-old HPV
-/-
and
HPV
+/-
mice (G oups 1 and 2, espec i ely) and 28-30 weeks-old HPV
-/-
and HPV
+/-
mice (G oups 3 and 4,
espec i ely). G oup 1, n = 10; G oup 2, n = 12; G oup 3, n = 6; G oup 4, n = 6. Each do ep esen s an indi idual
animal. Ba s ep esen he mean alue in each g oup. ** P  0.01; *** P  0.001.
32
low MHC I exp ession, CD8
+
T cells play a decisi e ole in igh ing HPV-induced
o opha yngeal cance s [155].
In he p esen wo k, we hypo hesize ha p aquiloside, he b acken oxin, exe s i s
immunosupp essi e e ec by coun e ac ing he ac ion o CD8
+
T cells agains
papilloma i us-induced lesions. This would acili a e i al pe sis ence and he p og ession
o ea ly HPV- and BPV-induced lesions, posing majo isks o human and animal heal h.
In o de o es ha hypo hesis, we employed HPV16- ansgenic mice and analysed he
e ec o p aquiloside on he popula ion o skin in il a ing CD8
+
T cells.
Ma e ial & Me hods
Mice
Cons uc ion o K14/HPV16 mice on a FVB/n backg ound has been p e iously
epo ed [73]. These animals de elop cha ac e is ic mul i-s age cu aneous and u e ine
ce ical ca cinogenesis, and we e gene ously dona ed by D s. Je ey A bei and Douglas
Hanahan (Uni e si y o Cali o nia) h ough he USA Na ional Cance Ins i u e Mouse
Reposi o y. The animal expe imen s we e app o ed by he Uni e sidade de T ás-os-
Mon es e Al o Dou o e hics commi ee (10/2013) and he Po uguese Ve e ina y
Di ec o a e (0421/000/000/2014). Animals we e main ained and b ed acco ding o
Po uguese (Dec e o-Lei 113, Augus 7
h
) and Eu opean (EU Di ec i e 2010/63/EU)
legisla ion, unde con olled condi ions o empe a u e (23 ± 2 ºC), ligh -da k cycle (12h
ligh /12h da k) and ela i e humidi y (50 ± 10 %), using ha dwood bedding. Food and
wa e we e p o ided ad libi um.
Mice geno yping
Animals we e geno yped a weaning, using ail ip samples as desc ibed p e iously
[113, 114]. B ie ly, nucleic acids we e ex ac ed and DNA quali y and pu i y we e
assessed. HPV16-E6 and -E2 genes we e ampli ied o con i m he in eg a ion o HPV
DNA in o he mouse genome and a agmen o mouse β-globin was also ampli ied o
con i m he quali y o he ex ac ed DNA. Leng hs o he agmen s we e con i med by
aga ose gel elec opho esis. Only hemizygous emale mice we e used o he ansgenic
mice g oups in he expe imen .
P aquiloside isola ion
P aquiloside was isola ed om b acken as p e iously desc ibed [156] wi h mino
modi ica ions. B ie ly, 1000 g (d ied weigh ) b acken c osie s we e ha es ed a A cos de
Valde ez, Po ugal, 41º 49´ 12´´ N, 8º 24´ 11´´ W) and a sample was deposi ed a he

33
Uni e sidade de T ás-os-Mon es e Al o Dou o he ba ium ( e e ence no. 18248). B acken
was blended in dis illed wa e (10 L), s i ed a oom empe a u e o 1 hou and he
ex ac was adso bed on 3 L o XAD-2 esin (Supelco, Sigma, S . Louis, MO). The esin
was elu ed wi h me hanol (10 L) and he me hanol ex ac was concen a ed, dissol ed in
wa e (400 ml) and ex ac ed wi h bu anol (5500 ml). The bu anol ex ac was
ch oma og aphed on silica gel (Me ck, Kenilwo h, NJ). F ac ions con aining p aquiloside
we e sepa a ed wice on oc adecyl-sylane silica gel (Fujy-Silysia, Kasugai Aichi, Japan)
using me hanol-wa e mix u es o ob ain pu e p aquiloside. The compound was
dis inguished om o he closely- ela ed b acken illudane oxins on he basis o i s
cha ac e is ic
1
H and
13
C nuclea magne ic esonance (NMR) signals [157] using an
A ance III 400 MHz spec ome e (B ucke , Bille ica, MA). Aliquo s we e p epa ed o
each expe imen al week (7.5 mg), eeze-d ied and kep a -20ºC un il use.
S udy Design
Thi y ansgenic (HPV16
+/-
) and 15 wild- ype (WT, HPV16
-/-
) emale 18-20 weeks-
old mice, showing di use cu aneous c us ing and papilloma osis we e employed. The
animals we e sepa a ed in o h ee expe imen al g oups: g oup 1 (n = 15, HPV16
-/-
mice),
g oup 2 (n = 15, HPV
+/-
un ea ed mice) and g oup 3 (n = 15, HPV16
+/-
mice ea ed o ally
wi h 0.5 mg p aquiloside pe week, o 10 consecu i e weeks). The expe imen al animals
we e moni o ed daily o signs o s ess o disease. All su i ing mice we e eu hanized a
28-30 weeks o age by an in ape i oneal pen oba bi al o e dose, ollowed by ca diac
punc u e and exsanguina ion. Ches skin samples (app oxima ely 4 cm
2
) we e collec ed
o low cy ome y analysis. Ma ched samples we e collec ed o his ological examina ion.
P aquiloside adminis a ion and oxici y
Each week, a 7.5 mg p aquiloside aliquo was dissol ed in 300 µl e hanol (25
mg/mL) and he indi idual 0.5 mg dose (20 µl e hanol) was added o a s anda d whea
cookie (Viei a, Vila No a de Famalicão, Po ugal) agmen weigh ing ca. 100 mg. Each
dosed agmen was allowed o d y a oom empe a u e o 10 minu es and indi idually
adminis e ed o a g oup 3 animal, in an indi idual emp y cage; he inges ion was isually
moni o ed. In o de o con i m ha p aquiloside was indeed ac i e a he adminis e ed
dosage, we looked o a lymphoid malignancy [128]. Thus, his ological analysis o kidney
samples was used o con i m neoplas ic lymphoid cell in il a ion in his o gan.
34
Skin his ology
Skin samples we e ixa ed in 10% neu al bu e ed o malin. Samples we e
dehyd a ed h ough g aded alcohols and xylene and, pa a in embedded in an au oma ic
STP 120 p ocesso (Mic on, Boise, ID). 2 µm- hick sec ions we e s ained wi h
haema oxylin-eosin (H&E) o his ological examina ion on ligh mic oscopy. Skin samples
we e classi ied as no mal skin, epide mal hype plasia, mul i ocal epide mal dysplasia in a
hype plas ic backg ound and di use epide mal dysplasia.
Isola ion o a Single-Cell Suspension
Ches skin samples we e cleaned om excessi e blood essels and a issue and
cu in o small pieces. This agmen s we e incuba ed o 2 hou s wi h 125 U/ml o ype I
collagenase (Gibco, Li e Technologies, Paisley, UK) in RPMI-1640 medium
complemen ed wi h 1% glu amine, 1% penicillin-s ep omycin-ampho e icin B, 1% HEPES
bu e (all om Sigma) and 10% oe al bo ine se um (BioWes , Nuaillé, F ance) a 37 ºC
and 150 pm in a 3031 shaking incuba o (GFL, Bu gwedel, Ge many). The esul ing cell
suspension was hen il e ed and cen i uged a 300  o 10 min a 4 ºC. Cells we e hen
essuspended in phospha e bu e ed saline con aining 1% bo ine se um albumin and 20
mM sodium azide ollowed by ex acellula s aining.
Flow Cy ome y Analysis
Following cell isola ion, he cellula immune pheno ype was assessed by low
cy ome y using monoclonal an ibodies (mAb). Non-speci ic an ibody binding was
p e en ed by incuba ing cells wi h an i-mouse CD16/CD32 mAb o FcγR blocking. This
was ollowed by incuba ion wi h an i-CD8 phycoe y h in-cych ome 5-conjuga e (clone 53-
6.7, BD Biosciences, San Diego, CA), an i-CD107a (LAMP1) phycoe y h in-conjuga e
(clone eBio1d4b) and an i-CD44 phycoe y h in-cych ome 7-conjuga e (clone IM7) mAb
(bo h om eBioscience, San Diego, CA). Following ex acellula s aining, he cells we e
washed, ixed in 2% o maldehyde and washed wi h phospha e bu e ed saline con aining
1% bo ine se um albumin and 20 mM sodium azide. An ibody-labelled cells we e
analysed in an EPICS XL low cy ome e using he EXPO32ADC so wa e (Beckman
Coul e , Miami, FL). The collec ed da a iles we e analysed using he FlowJo so wa e
10.0.7 (FLOWJO, LLC, Ashland, OR).
S a is ical Analysis
Flow cy ome y s a is ical analyses we e execu ed in he G aphPad so wa e
( e sion 6.0, G aphPad So wa e, Inc. La Jolla, CA). In column and do g aphs, each poin
is ep esen a i e o an indi idual mouse and ba s ep esen he mean alue o he
35
espec i e g oup. Analysis be ween g oup pai s was pe o med using he Mann-Whi ney
es . Kaplan-Meie su i al analysis coupled wi h a log ank es was pe o med using he
PASW S a is ics so wa e ( e sion 18, IBM® SPSS®, Qua y Bay, Hong Kong).
Resul s
Gene al indings
P aquiloside was isola ed a a 0.01% yield om b acken, as p e iously epo ed
[128]. The compound's s uc u e (Fig. 4a) was con i med using NMR analysis and no
o he illudane glycoside was de ec ed (Fig. 4b, c). T ansgenic animals showed
cha ac e is ic di use cu aneous hype ke a osis and e y hema, oge he wi h a iably
in ense p u i us, while WT mice showed no mal skin. All mice in g oups 1 and 2 su i ed
he 10 weeks expe imen al pe iod. Ten mice (66.7%) om g oup 3 succumbed be o e he
end o he s udy (Fig. 4d). His ological analysis o kidney samples showed no mal
his ology in WT and un ea ed HPV
+/-
animals (Fig. 4e) and, mode a e o se e e,
mul i ocal, pe i ascula in il a ion o lymphoblas ic cells, showing highly pleomo phic
nuclei and up o 3 mi o ic igu es pe high-powe ield in 100% o p aquiloside- ea ed
animals (g oup 3) (Fig. 4 ).
36
Fig. 4:
P aquiloside and i s leukaemogenic e ec in K14-HPV16 mice. a – p aquiloside's s uc u al o mula. b and
c – pa ial 2D NMR-HSQC spec a (ob ained in CD3OD) o p aquiloside. No e he co ela ions be ween ca bons and
hei co esponding hyd ogens in he glucose esidue (b). No e in pa icula he co ela ions be ween ca bons 12
and 13 a 5.87 ppm and 10.86 ppm espec i ely, and hei co esponding hyd ogens in he cyclop opylidene ing
(c). d – Kaplan-Meie su i al analysis. HPV
+/-
mice ea ed wi h p aquiloside show signi ican ly educed su i al (P
 0.001) compa ed wi h un ea ed HPV
-/-
o HPV
+/-
animals. e – Un ea ed HPV
+/-
animal (g oup 2), showing no mal
kidney his ology; H&E 200 . – P aquiloside-exposed HPV
+/-
animal (g oup 3) showing ypical p aquiloside-
associa ed pe i ascula in il a ion by leukaemic lymphocy es. No e ma ked nuclea pleomo phism and mi o ic
igu es; H&E 200 .
37
Cu aneous lesions
The his ological analysis showed ha all g oup 1 mice (100.0%) p esen ed no mal
skin his ology (Fig. 5a) whils all ansgenic mice (g oups 2 and 3) p esen ed skin lesions.
In g oup 2, 3 animals (50.0%) p esen ed di use epide mal dysplasia (Fig. 5b); o he 3
(50.0%) showed epide mal hype plasia wi h mul i ocal dysplasia. Sub-epide mal
angiogenesis and de mal in lamma o y in il a es we e p ominen in dysplas ic lesions,
showing nume ous mixed mononuclea leukocy es and neu ophils. In g oup 3, he o ali y
o mice (100.0%) showed di use epide mal dysplasia (Fig. 5c).
Fig. 5:
His opa hological changes induced by HPV16 oncogenes in FVB/n mice, H&E. a – g oup 1 animal. No mal
skin his ology, 400 . b – g oup 2 animal. Epide mal hype plasia ex ending o he ollicula in undibulum, 400 . c –
g oup 3 animal. Epide mal dysplasia, 400 . No e loss o ke a inocy ic pola i y and di e en ia ion.
CD8
+
T lymphocy es a e p esen in p aquiloside- ea ed and un ea ed
animals
In o de o de e mine he p opo ions o CD8
+
T cells p esen in ches skin wi h o
wi hou HPV-associa ed lesions, samples we e collec ed and cells isola ed, ollowed by
low cy ome y analysis (Fig. 6a). As shown in Fig. 6b, g oup 2 and 3 (HPV
+/-
and
p aquiloside- ea ed HPV
+/-
, espec i ely) mice showed signi ican ly highe pe cen ages o
CD8
+
T cells compa ed wi h HPV
-/-
mice (P  0.05). The pe cen age o CD8
+
T cells in
ches skin was no signi ican ly di e en be ween p aquiloside- ea ed and un ea ed
HPV16
+/-
mice (g oups 3 and 2, espec i ely).

38
Fig. 6:
Pe cen age o CD8
+
T cells wi hin o al lymphoid-ga ed cells ob ained om ches skin samples om WT and
K14-HPV16 ansgenic mice. a – Rep esen a i e analysis o he ga ing s a egy u ilized. The numbe wi hin he
g aph co esponds o he pe cen age o he ga ed popula ion. b – Pe cen ages o CD8
+
T cells we e de e mined by
low cy ome y a e skin issue collec ion and enzyma ic diges ion. Ches skin samples we e collec ed om 28-30
weeks-old HPV
-/-
(G oup 1), HPV
+/-
mice (G oup 2) and p aquiloside- ea ed (PTQ) HPV
+/-
mice (G oup 3). G oup 1, n =
5; G oup 2, n = 5; G oup 3, n = 5. Each do ep esen s an indi idual animal. Ba s ep esen he mean alue in each
g oup. * P  0.05.
P aquiloside dec eases he numbe o CD8
+
CD107a
+
T cells in HPV16
+/-
mice
Nex , CD8
+
T cells ound in skin samples we e analysed wi h he objec i e o
de e mine i hey we e ac i ely deg anula ing, which indica es ongoing cy o oxic ac i i y.
Fo ha pu pose, low cy ome y analysis was pe o med o assess he exp ession o
su ace CD107a in hese cells (Fig. 7a). A signi ican ly highe pe cen age o
CD8
+
CD107a
+
T lymphocy es was obse ed in HPV
+/-
mice as compa ed wi h HPV
-/-
animals (P  0.01) (Fig. 7b). Addi ionally, HPV
+/-
mice exposed o p aquiloside also
p esen ed a highe pe cen age o CD8
+
CD107a
+
T lymphocy es when compa ed o WT
animals, bu his was signi ican ly educed when compa ed o un ea ed HPV
+/-
mice (P 
0.05).
39
Fig. 7:
Pe cen age o CD8
+
CD107a
+
T cells wi hin o al CD8
+
T cells ob ained om ches skin samples om WT and
K14-HPV16 ansgenic mice. a – Rep esen a i e analysis o he ga ing s a egy u ilized. Numbe s wi hin g aphs
co espond o he pe cen age o he ga ed popula ion. b – Pe cen ages o CD8
+
CD107a
+
T cells in ga ed CD8
+
T
cells we e de e mined by low cy ome y a e skin issue collec ion and enzyma ic diges ion. Ches skin samples
we e collec ed om 28-30 weeks-old HPV
-/-
(G oup 1), HPV
+/-
mice (G oup 2) and p aquiloside- ea ed (PTQ) HPV
+/-
mice (G oup 3). G oup 1, n = 5; G oup 2, n = 5; G oup 3, n = 5. Each do ep esen s an indi idual animal. Ba s
ep esen he mean alue in each g oup. *P 0.05; ** P  0.01.
Reduced numbe o CD8
+
CD44
+
T cells in p aquiloside- ea ed ansgenic
animals
Ha ing de e mined ha CD8
+
T cells p esen ed a cell su ace pheno ype
associa ed wi h deg anula ion, exp ession o CD44, a ma ke indica ing a memo y cell
pheno ype [158], was also assessed. As shown in Fig. 8 (a and b), a signi ican ly highe
pe cen age o CD8
+
CD44
+
T lymphocy es was ound in HPV
+/-
mice as compa ed wi h
HPV
-/-
animals (P  0.01). P aquiloside- ea ed HPV
+/-
mice p esen ed a signi ican ly
educed pe cen age o CD8
+
CD44
+
T lymphocy es compa ed wi h un ea ed HPV
+/-
mice
(P  0.05). In ac , g oup 3 did no show s a is ically di e en alues o g oup 1 (P  0.05).
40
Fig. 8:
Pe cen age o CD8
+
CD44
+
T cells wi hin o al CD8
+
T cells ob ained om ches skin samples om WT and
K14-HPV16 ansgenic mice. a – Rep esen a i e analysis o he ga ing s a egy u ilized. Numbe s wi hin g aphs
co espond o he pe cen age o he ga ed popula ion. b – Pe cen ages o CD8
+
CD44
+
T cells in ga ed CD8
+
T cells
we e de e mined by low cy ome y a e skin issue collec ion and enzyma ic diges ion. Ches skin samples we e
collec ed om 28-30 weeks-old HPV
-/-
(G oup 1), HPV
+/-
mice (G oup 2) and p aquiloside- ea ed (PTQ) HPV
+/-
mice
(G oup 3). G oup 1, n = 5; G oup 2, n = 5; G oup 3, n = 5. Each do ep esen s an indi idual animal. Ba s ep esen
he mean alue in each g oup. * P  0.05;** P  0.01.
Discussion
CD8
+
T lymphocy es a e impo an e ec o s o he cell-media ed immune
esponse, especially agains i us-in ec ed o umou cells [117]. Upon ac i a ion hese
cells may deg anula e, eleasing a se o ly ic enzymes capable o des oying a ge cells
[159]. CD107a is a lysosome-associa ed memb ane p o ein which can be used as
su oga e ma ke o deg anula ion [145]. In CD8
+
T lymphocy es, CD107a eaches he cell
su ace when ly ic g anules a e exocy osed, hus exposing i s memb ane p o eins [144].
As CD8
+
T cells need o be ac i a ed in o de o elease cy o oxic g anules, he p esence
o CD107a a he cell su ace is use ul o de e mine hei ac i a ion s a us [145].
Addi ionally, CD8
+
T cells may also acqui e a memo y pheno ype, allowing a quicke ecall
esponse, as memo y T cells can be mo e easily ac i a ed han nai e T cells [159]. CD44
is a cell-su ace glycop o ein and is cu en ly conside ed he bes ma ke o iden i y
memo y CD8
+
T cells [158]. Fu he mo e, CD44 is hough o ac as a egula o o he
mo ili y o CTL in he umou mic oen i onmen [160].
41
Al hough many papilloma i us (including some oncogenic HPV and BPV ypes)
down- egula e MHC I exp ession as a s a egy o e ade immune su eillance media ed by
CTL [154], hese cells s ill play a majo ole agains papilloma i us-induced lesions,
namely in HPV-induced o opha yngeal cance s [155]. E ading he hos immune esponse
is c i ical o main aining a long- e m in ec ion, allowing umou p og ession. In his
con ex , en i onmen al (e.g. die a y) immunosupp essan p oduc s such as b acken and
i s oxin, p aquiloside, may play an impo an ole.
In his wo k, as expec ed, WT mice had no HPV-associa ed lesions, whils age-
ma ched K14-HPV16 ansgenic mice showed hype plas ic lesions wi h dysplas ic oci o
di use dysplasia. I is possible ha he di use dysplas ic lesions obse ed in p aquiloside-
ea ed animals ep esen a mo e agg essi e s age in mul i-s ep ca cinogenesis compa ed
wi h he o en ocal dysplasia obse ed in un ea ed animals. Howe e , he small numbe
o animals ha su i ed he expe imen al pe iod and he absence o ob iously in asi e
lesions do no allow us o i mly conclude ha p aquiloside enhanced umou p og ession.
The inc easing agg essi eness o dysplas ic e sus hype plas ic lesions was
accompanied by an inc eased bulk o in il a ing immune cells.
These esul s indica e ha HPV-induced hype plas ic and dysplas ic skin lesions in
K14-HPV16 ansgenic mice show inc eased in il a ion o CD8
+
T cells when compa ed o
no mal skin om age-ma ched WT mice. P aquiloside educed he CD8
+
T cell in il a ion,
al hough only sligh ly (P > 0.05). Ne e heless, p aquiloside did induce a signi ican
educ ion in CD107a
+
and CD44
+
CD8
+
T cells.
Al oge he , hese da a show ha cy o oxic T cells mig a e o he si e o HPV-
induced lesions, a e ac i a ed o deg anula e and acqui e a memo y-pheno ype and, ha
p aquiloside is capable o impai ing he unc ion o hese cells, explici ly by inhibi ing
cellula ac i a ion and memo y pheno ype acquisi ion. The molecula mechanism behind
his impai men should be an allu ing opic o u u e esea ch.
G azing animals and human popula ions wo ldwide a e easily exposed o b acken
and i s illudane oxins, including p aquiloside and o he ela ed compounds [123]. B acken
is an abundan weed, especially in poo e pas u es, and p aquiloside accumula es in
bo ine and o ine milk and mea [161-163], besides con amina ing unde g ound wa e s
[164]. Mo eo e , some human popula ions in di e en coun ies consume b acken shoo s
(known as c ozie s, b o o de samambaia in Po uguese o wa abi in Japanese) as pa o
hei daily die [123]. In China, he p oduc ion o d y P e idium aquilinum a . la iusculum
o ood uses (locally known as juecai) was es ima ed o in ol e app oxima ely 1000
companies in a business wo h 300 million USD pe annum [165]. This is conside ed he
mos widely-consumed e n in China, wi h an es ima ed annual p oduc ion o 1200 ons in
he Zhouzhou coun y, p o ince o Hunan. B acken and p aquiloside a e well-known
48
molecula ly [76], making K14-HPV16 mice a pa icula ly use ul model o HPV esea ch.
Ma e ial & Me hods
Animals
Gene a ion o K14/HPV16 mice on a FVB/n backg ound has been p e iously
epo ed [73]. K14-HPV16 ansgenic mice we e kindly dona ed by D s. Je ey A bei and
Douglas Hanahan (Uni e si y o Cali o nia) h ough he USA Na ional Cance Ins i u e
Mouse Reposi o y. The animal expe imen s we e app o ed by he Uni e sidade de T ás-
os-Mon es e Al o Dou o E hics Commi ee (10/2013) and he Po uguese Gene al
Ve e ina y Di ec o a e (app o al no. 0421/000/000/2014). Animals we e main ained and
b ed acco ding o Po uguese (Dec e o-Lei 113, Augus 7
h
) and Eu opean (EU Di ec i e
2010/63/EU) legisla ion, unde con olled condi ions o empe a u e (23 ± 2 ºC), ligh -da k
cycle (12h ligh /12h da k) and ela i e humidi y (50 ± 10 %), using co ncob bedding. Food
and wa e we e p o ided ad libi um.
Mice geno yping
Animals we e geno yped a weaning, using ail ip samples as desc ibed p e iously
[113, 114]. B ie ly, nucleic acids we e ex ac ed and DNA quali y and pu i y we e
assessed. HPV16-E6 and -E2 genes we e ampli ied o con i m he p esence o HPV DNA
and a agmen o mouse β-globin was also ampli ied as endogenous con ol. Leng hs o
he amplicons we e con i med by aga ose gel elec opho esis. Only hemizygous emales
we e used o he ansgenic mouse g oups in he expe imen .
Expe imen al Design
Fi y- ou 18 o 20 weeks-old emale mice we e di ided in o ou expe imen al
g oups, acco ding o hei geno ype and aking in o conside a ion ha HPV ansgenes
and celecoxib (CXB) adminis a ion could induce some mo ali y: g oup 1 (HPV16
-/-
un ea ed animals, n = 12), g oup 2 (HPV16
+/-
un ea ed animals, n = 12), g oup 3
(HPV16
-/-
CXB- ea ed animals, n = 15) and g oup 4 (HPV16
+/-
CXB- ea ed animals, n =
15). All su i ing mice we e humanely eu hanized a 24-26 weeks o age by
in ape i oneal pen oba bi al o e dose, ollowed by exsanguina ion by ca diac punc u e.
Ches skin samples (app oxima ely 4 cm
2
) we e collec ed o low cy ome y analysis and
ma ched skin samples we e collec ed o his ological analysis.

49
Celecoxib adminis a ion
Celecoxib (P ize , New Yo k, NY) was dissol ed in d inking wa e a a
concen a ion o 0.5 mg/ml, es ima ing an a e age daily wa e in ake o 5.0ml pe mouse,
and a dose o 46.7 mg/kg/day and 2.5 mg/animal/day in an a e age mouse weigh ing 30
g. This is a well- ole a ed mode a e dose, as shown in p e ious assays [177]. Howe e ,
K14-HPV16 animals d ama ically inc eased hei wa e in ake when CXB was added
(possibly due o he highly pala able lac ose p esen in he ehicle) eaching up o 15 ml
pe animal. The CXB concen a ion was hus educed om he s a o he hi d week
onwa ds down o 0.2 mg/ml, esul ing in a dec ease in consump ion and an e ec i e dose
o 93 mg/kg/day and 2.8 mg/animal/day. The a e age dose du ing he o e all
expe imen al pe iod was hus 124 mg/kg/day and 3.72 mg/animal/day.
His ological Analysis
Skin samples we e ixa ed in 10% neu al bu e ed o malin o 48 hou s. Samples
we e dehyd a ed h ough g aded alcohols and xylene and pa a in embedded in an
au oma ic STP 120 p ocesso (Mic on, Boise, ID). 2 µm- hick sec ions we e s ained wi h
haema oxylin-eosin (H&E) o his ological e alua ion on ligh mic oscopy. Skin samples
we e classi ied as no mal skin, epide mal hype plasia and epide mal dysplasia.
P epa a ion o Single-Cell Suspensions
Ches skin samples we e cu in o small pieces a e excessi e blood essels and
a issue emo al. Skin agmen s we e incuba ed o 2 hou s wi h 125 U/ml ype I
collagenase (Gibco, Li e Technologies, Paisley, UK) in RPMI-1640 medium
complemen ed wi h 1% glu amine, 1% penicillin-s ep omycin-ampho e icin B, 1% HEPES
bu e (all om Sigma, S . Louis, MO) and 10% oe al bo ine se um (BioWes , Nuaillé,
F ance) a 37 ºC and 150 pm in a 3031 o bi al incuba o (GFL, Bu gwedel, Ge many).
Subsequen ly, he esul ing cell suspension was il e ed and cen i uged a 300  o 10
min a 4 ºC. Cells we e essuspended in phospha e bu e ed saline con aining 1% bo ine
se um albumin and 20 mM sodium azide ollowed by low cy ome y analysis.
Immunopheno yping
Following cell isola ion, he su ace pheno ype o he collec ed cells was assessed
by low cy ome y using speci ic monoclonal an ibodies (mAb). Cells we e incuba ed wi h
an i-mouse CD16/CD32 mAb o FcγR blocking, o p e en non-speci ic an ibody binding.
Nex , cells we e incuba ed wi h an i-CD8 mAb phycoe y h in-cych ome 5-conjuga e (clone
53-6.7, BD Biosciences, San Diego, CA) and an i-CD107a (LAMP1) mAb phycoe y h in-
50
conjuga e (clone eBio1d4b, eBioscience, San Diego, CA). Following ex acellula s aining,
he cells we e washed, ixed in 2% o maldehyde and washed wi h phospha e bu e ed
saline con aining 1% bo ine se um albumin and 20 mM sodium azide. An ibody-labelled
cells we e analysed in an EPICS XL low cy ome e using he EXPO32ADC so wa e
(Beckman Coul e , Miami, FL). The assembled da a iles we e analysed using he FlowJo
so wa e 10.0.7 (FLOWJO, LLC, Ashland, OR).
S a is ical Analysis
S a is ical analyses we e pe o med using he G aphPad so wa e ( e sion 6.0,
G aphPad So wa e, Inc. La Jolla, CA). In column and do g aphs each poin ep esen s
indi idual mice, while ba s ep esen he mean o he espec i e g oup. S a is ical
analysis be ween g oup pai s was done using he Mann-Whi ney es .
Resul s
Gene al indings
All K14-HPV16 mice showed ypical cu aneous changes, including di use
hype ke a osis and e y hema. Celecoxib- ea ed HPV
+/-
mice (g oup 4) showed signi ican
mo ali y: 10 ou o 15 mice (66.7%) succumbed be o e he end o he s udy. All mice in
g oups 1, 2 and 3 su i ed un il he end o he s udy.
His ological analysis
His ological analysis o skin samples (Table 2) showed ha all WT mice, g oups 1
and 3, p esen ed no mal skin his ology (Fig. 9a). In g oups 2 and 4 (un ea ed and CXB-
ea ed HPV
+/-
mice, espec i ely), he o ali y o mice showed simple o papilla y, di use,
a iably se e e epide mal hype plasia ex ending o he ollicula in undibula and
papilloma osis wi h o hoke a o ic hype ke a osis (Fig. 9b). Addi ionally, 2 animals (16.7%)
om g oup 2 p esen ed mul i ocal epide mal dysplas ic oci wi h pa ake a o ic
hype ke a osis wi hin he hype plas ic backg ound (Fig. 9c). The e we e signs o mild
in lamma ion, wi h a small amoun o mac ophages, mas cells and lymphocy es in he
supe icial de mis. Dysplas ic oci we e associa ed wi h inc eased numbe s o leukocy es
in il a ing he de mo-epide mic junc ion and wi h in ense angiogenesis.
51
Table 2:
His ological classi ica ion o HPV16-induced skin lesions om 24-26 weeks-old emale mice. Celecoxib –
CXB.
G oup Cu aneous lesions
Incidence (%)
No mal skin Epide mal hype plasia Epide mal dysplasia
1
(HPV
-
/
-
, n = 12) 12/12 (100%) 0/0 (0%) 0/0 (0%)
2
(HPV
+/
-
, n = 12) 0/0 (0%) 12/12 (100%) 2/12 (16.7%)
3
(HPV
-
/
-
+ CXB, n = 15) 15/15 (100%) 0/0 (0%) 0/0 (0%)
4
(HPV
+/
-
+ CXB, n = 5) 0/0 (0%) 5/5 (100%) 0/0 (0%)
Fig. 9:
His opa hological changes induced by HPV16 oncogenes in FVB/n mice, H&E. a – WT animal; No mal skin
his ology, 100 . b – CXB- ea ed HPV16
+/-
animal; Epide mal hype plasia ex ending o he ollicula in undibulum
and is hmus, 100 . c – Un ea ed HPV16
+/-
mouse; Epide mal dysplasia, 200 . No e ma ked pa ake a o ic
hype ke a osis, loss o cell pola i y, enhanced anisoka yosis and mi o ic ac i i y. The de mal-epide mal junc ion is
obscu ed by se e e in lamma o y cell in il a ion.
CTL in il a ion and ac i a ion in celecoxib- ea ed mice
Wi h he pu pose o s udy CD8
+
T cell in il a ion in HPV-induced lesions, lymphoid
cells om mice ches skin issue we e isola ed and a low cy ome y analysis o he
eco e ed cells was pe o med (Fig. 10a). Skin samples om un ea ed HPV
+/-
mice
showed a signi ican ly highe pe cen age o CD8
+
T cells compa ed wi h un ea ed WT
animals (P  0.01) (Fig. 10b). Also, al hough s a is ical signi icance was no achie ed,
CXB- ea ed WT mice showed dec eased CD8
+
T cell in il a ion when compa ed o CXB-
ea ed HPV
+/-
animals and he same was obse ed be ween samples om WT mice
ea ed wi h CXB (g oup 3) and un ea ed WT animals (g oup 1) (Fig. 10b). Mo eo e ,
HPV
+/-
mice ea ed wi h CXB (g oup 4) showed less CD8
+
T lymphocy es compa ed wi h
un ea ed ansgenic animals (g oup 2) (P  0.01) (Fig. 10b).
Nex , he objec i e was o s udy whe he he CD8
+
T lymphocy es ound in mouse
skin samples showed signs o cy o oxic ac i i y (Fig. 10a). Thus, exp ession o CD107a
(LAMP1) a he cell su ace was assessed by low cy ome y. CD107a eaches he su ace
o CTL when ly ic g anules a e eleased, exposing hei memb ane p o eins du ing
52
exocy osis [144]. Thus, his lysosome-associa ed memb ane p o ein is equen ly used as
an immunological ma ke o CTL deg anula ion [145]. Fig. 10c shows a signi ican ly highe
pe cen age o CD8
+
CD107a
+
cells in samples om HPV
+/-
mice ea ed wi h CXB (g oup
4) when compa ed wi h un ea ed HPV
+/-
(g oup 2) (P  0.01).
Fig. 10:
Pe cen age o CD8
+
and CD8
+
CD107a
+
T cells wi hin o al lymphoid-ga ed cells ob ained om ches skin
samples om WT and K14-HPV16 ansgenic mice. a – Rep esen a i e analysis o he ga ing s a egy employed.
Numbe s wi hin g aphs co espond o he pe cen age o he ga ed popula ion. b – Pe cen ages o CD8
+
T cells and
c – pe cen ages o CD8
+
CD107a
+
T cells in ga ed CD8
+
T cells we e de e mined by low cy ome y a e skin issue
collec ion and diges ion wi h collagenase. Ches skin samples we e collec ed om WT and HPV
+/-
mice (G oups 1
and 2, espec i ely) and CXB- ea ed (CXB) WT
and HPV
+/-
mice (G oups 3 and 4, espec i ely) a 24-26 weeks o
age. G oup 1, n = 5; G oup 2, n = 5; G oup 3, n = 5; G oup 4, n = 5. Each do ep esen s an indi idual animal. Ba s
ep esen he mean alue in each g oup. ** P  0.01.
53
Discussion
The de elopmen o HPV-associa ed malignancies depends on a pe sis en HPV
in ec ion. Ul ima ely, he abili y o he immune sys em o elimina e he i us is he key
elemen o decide whe he a HPV in ec ion is clea ed o e ol es o cance [2]. HPV
p esen s well-known mechanisms o e ade hos immuni y and delay i s elimina ion, hus
acili a ing i al pe sis ence [154]. When he i us is de ec ed, an inna e immune esponse
occu s, and leads o he de elopmen o an adap i e immune esponse. A undamen al
pa o his adap i e esponse is cell-media ed immuni y, cha ac e ized by he ac i i y o a
as numbe o CD8
+
and CD4
+
T lymphocy es [39]. In ac , in il a ion by hese cells in
HPV-associa ed lesions has al eady been shown o lead o eg ession [148, 178]. CTL
in il a ion d i es he esponse induced by an expe imen al he apeu ic accine in pa ien s
wi h ce ical in aepi helial lesions [171] and co ela es wi h a be e p ognosis in pa ien s
wi h HPV-posi i e head and neck cance [155, 172, 173].
Ch onic in lamma ion is a key ea u e associa ed wi h ca cinogenesis, namely in
he case o HPV in ec ion [8]. COX-2 is an enzyme wi h an impe a i e ole in he
me abolism o a achidonic acid, which leads o he p oduc ion o p os aglandins, which in
u n p omo e in lamma ion [129]. In ac , COX-2 is o e -exp essed in se e al
malignancies, including ce ical cance [130-134]. As in lamma ion is known o con ibu e
o cance p og ession [8], COX-2 inhibi ion is expec ed o esul in umou g ow h
inhibi ion whils educing in lamma ion. Some non-s e oidal an i-in lamma o y d ugs, like
aspi in and ibup o en, ha e al eady shown p omising esul s as an i- umou he apy in
bo h pa ien s and p e-clinical animal models [179-182]. Howe e , hese d ugs ha e e y
low speci ici y. Selec i e COX-2 inhibi o s such as CXB and o ecoxib ha e al eady been
used o p e en he de elopmen o colo ec al adenomas [183, 184]. Fu he mo e, his
COX-2 inhibi o has been showing p omising esul s in a umou model o human colon
cance when combined wi h chemo he apeu ic d ugs [185].
In his s udy, he e ec o he selec i e COX-2 inhibi o CXB o e he a icking
and ac i a ion o CD8
+
T lymphocy es was examined in he K14-HPV16 mouse model, a
p ope model o s udy HPV-induced ca cinogenesis due o i s g ea simila i ies wi h he
human clinical disease [76]. High mo ali y was obse ed among CXB- ea ed animals,
p esumably due o CXB- ela ed oxici y.
The esul s p esen ed he ein show ha CXB educes he numbe o umou -
in il a ing CD8
+
T cells when compa ed wi h un ea ed mice. S ill, despi e he dec ease in
cell numbe s, CXB- ea ed mice ha e a highe pe cen age o ac i a ed and deg anula ing
CTL compa ed wi h un ea ed animals. These indings sugges ha CXB educes he
numbe o umou -in il a ing CD8
+
T cells while enhancing hei e ec o unc ions. These

54
da a a e in ag eemen wi h a p e ious s udy using glioma-bea ing mice whe e COX2
-/-
mice had inc easing pe cen ages o umou -in il a ing CD8
+
CD107a
+
lymphocy es [175].
This may be explained by ano he s udy epo ing ha COX-2 ac i i y leads o CD8
+
T cell
senescence and his ajec o y may be opposed by COX-2 inhibi ion, as shown by
inc eased le els o CD28 and in e leukin-2 in CD8
+
T cells [174].
Fu he mo e, hype plas ic epide mal lesions we e e iden in all su i ing
ansgenic mice, whils all WT animals p esen ed no mal skin his ology, as expec ed.
Mul i ocal dysplas ic epide mal lesions we e es ic ed o un ea ed HPV
+/-
animals. These
indings sugges ha CXB blocked umou p og ession a he hype plas ic s age, bu he
small numbe o dysplas ic lesions obse ed does no allow o any de ini i e conclusions.
In ac , COX-2 inhibi ion boos ed he e icacy o a DNA accine exp essing he HPV E7
oncogene, by enhancing umou -in il a ing CD8
+
T cells and slowing umou g ow h [186].
Howe e , his s udy was pe o med in mice bea ing allog a ed TC1 lung cells
immo alized by he HPV16 E6 and E7 oncogenes and ans o med by he c-Ha- as
oncogene. Compa isons be ween his model and K14-HPV16 mice a e limi ed, because
allog a s do no ep oduce HPV-associa ed mul i-s ep ca cinogenesis, being di ec ly
implan ed in he subcu is wi h a ully malignan pheno ype. Also, ke a inocy es and no
lung cells a e he a ge s o papilloma i us in ec ion.
I emains unclea whe he CTL a e ac i a ed a egional lymph nodes o a he
lesion loca ion and his would be an in e es ing poin o add ess in he u u e. The esul s
p esen ed he ein sugges ha CXB induces augmen ed CTL deg anula ion in HPV-
induced lesions, possibly con ibu ing o p e en malignan p og ession in his animal
model.
Conclusion
The p esen da a con i m he po en ial o CXB o enhance deg anula ion by
umou -in il a ing CD8
+
T lymphocy es. Mo eo e , he e ec o CXB seems o be mo e
complex han p e iously hough , as i also educed he o e all numbe o umou -
in il a ing CD8
+
T cells. Fu u e s udies add essing he impac o COX-2 inhibi o s on he
p ognosis o pa ien s bea ing HPV-induced lesions should ake in o accoun bo h he
numbe o in il a ing CD8
+
T cells and hei ac i a ion s a us.
55
Chap e 6: Gene al Discussion and Conclusions
56
57
HPV-associa ed malignancies a e a genuine heal h p oblem wo ldwide, wi h
emphasis on he inc ease in he numbe o HPV-associa ed o opha yngeal ca cinomas
[152]. P ophylac ic accina ion emains he mos e ec i e me hod o con ol HPV
in ec ions. Cu en ly, he e a e wo e ec i e comme cialized accines: Ce a ix™
(GlaxoSmi hKline Biologicals, Rixensa , Belgium), a bi alen HPV16/18 accine; and
Ga dasil™ (Me ck Vaccines, Wes Poin , PA), a quad i alen HPV6/11/16/18 accine.
Bo h a e HPV L1 i us-like pa icles (VLP) accines and hey ac by boos ing he
p oduc ion o neu alising an ibodies di ec ed agains he L1 capsid p o ein [187]. Despi e
hei e ec i eness, hese accines p esen a ew limi a ions, as he i us- ype es ic ion.
The i s is es ic ed o wo HPV ypes (bi alen ) and he second is es ic ed o ou ypes
(quad i alen ). Looking o inc ease ype-speci ic p o ec ion, Me ck managed o c ea e he
nona alen HPV L1 VLP accine by adding he VLPs om i e oncogenic HPV ypes (31,
33, 45, 52 and 58) in addi ion o he ou ypes used in he quad i alen accine. This
esul ed in a 20% inc ease in p o ec ion, in addi ion o he 70% ob ained om he
quad i alen accine, in a o al o app oxima ely 90% p o ec ion agains ce i al cance
[188]. Howe e , despi e hese ad ances in disease p e en ion, an e ec i e he apeu ic
accine agains HPV-induced malignancies emains absen .
In o de o de elop a he apeu ic s a egy, i is necessa y o ga he a la ge amoun
o in o ma ion on he ea u es o disease p og ession as well as o he e en s occu ing in
he lesions mic oen i onmen . I is accep ed ha he ca cinogenesis p ocess induced by a
HPV in ec ion is ma ked by he exis ence o a p ominen immune esponse [11, 18]. Thus,
a g ea deal o cellula popula ions and chemical ac o s a e p esen and play a speci ic
ole in ca cinogenesis, ei he p omo ing o impai ing i . A he end, i should be he
balance be ween all p o- and an i- umou s imuli ha will decide whe he he i al in ec ion
is clea ed o is allowed o p og ess owa ds malignancy. The e o e, a he apeu ic accine
could possible enhance he ac ion o an i- umou e ec o s and/o inhibi he unc ion o
known p o- umou media o s.
Cy o oxic CD8
+
T lymphocy es, o ins ance, a e he main cells associa ed wi h he
elimina ion o i us-in ec ed and ans o med cells. These cells a e pa o he adap i e
immune esponse and depend on an igen p esen a ion by MHC I molecules and o he co-
s imula o y signals o become ac i a ed and exe hei unc ions [116, 117]. Howe e ,
du ing he cou se o HPV in ec ion, exp ession o MHC I molecules is usually down-
egula ed. This is one o he mos common immune e asion mechanisms employed by
he HPV E5 oncop o ein [154]. In il a ion o CD8
+
T cells in HPV-posi i e lesions has
al eady been associa ed wi h be e p ognosis when compa ed wi h HPV-nega i e ones
[155, 172, 173, 189]. Thus, boos ing he ac i i y o CD8
+
T lymphocy es seems a possible
he apeu ic s a egy o elimina e HPV-associa ed lesions. S ill, he HPV immune e asion
64
31. Sche ne , M., B.A. We ness, J.M. Huib eg se, A.J. Le ine, and P.M. Howley,
1990, The E6 oncop o ein encoded by human papilloma i us ypes 16 and 18 p omo es
he deg ada ion o p53. Cell, 63(6): p. 1129-36.
32. Klingelhu z, A.J., S.A. Fos e , and J.K. McDougall, 1996, Telome ase
ac i a ion by he E6 gene p oduc o human papilloma i us ype 16. Na u e, 380(6569): p.
79-82.
33. Dyson, N., P.M. Howley, K. Munge , and E. Ha low, 1989, The human
papilloma i us-16 E7 oncop o ein is able o bind o he e inoblas oma gene p oduc .
Science, 243(4893): p. 934-7.
34. Dyson, N., P. Guida, K. Munge , and E. Ha low, 1992, Homologous
sequences in adeno i us E1A and human papilloma i us E7 p o eins media e in e ac ion
wi h he same se o cellula p o eins. J Vi ol, 66(12): p. 6893-902.
35. Munge , K., W.C. Phelps, V. Bubb, P.M. Howley, and R. Schlegel, 1989, The
E6 and E7 genes o he human papilloma i us ype 16 oge he a e necessa y and
su icien o ans o ma ion o p ima y human ke a inocy es. J Vi ol, 63(10): p. 4417-21.
36. Takeuchi, O. and S. Aki a, 2010, Pa e n ecogni ion ecep o s and
in lamma ion. Cell, 140(6): p. 805-20.
37. Hasan, U.A., E. Ba es, F. Takeshi a, A. Bilia o, R. Acca di, V. Bou a d, M.
Mansou , I. Vincen , L. Gissmann, T. I ne , M. Side i, F. S uben auch, and M.
Tommasino, 2007, TLR9 exp ession and unc ion is abolished by he ce ical cance -
associa ed human papilloma i us ype 16. J Immunol, 178(5): p. 3186-97.
38. F aze , I.H., 2009, In e ac ion o human papilloma i uses wi h he hos
immune sys em: a well e ol ed ela ionship. Vi ology, 384(2): p. 410-4.
39. Hibma, M.H., 2012, The immune esponse o papilloma i us du ing in ec ion
pe sis ence and eg ession. Open Vi ol J, 6: p. 241-8.
40. Amado -Molina, A., J.F. He nandez-Valencia, E. Lamoyi, A. Con e as-
Pa edes, and M. Lizano, 2013, Role o inna e immuni y agains human papilloma i us
(HPV) in ec ions and e ec o adju an s in p omo ing speci ic immune esponse. Vi uses,
5(11): p. 2624-42.
41. Mille , L.S. and R.L. Modlin, 2007, Human ke a inocy e Toll-like ecep o s
p omo e dis inc immune esponses. J In es De ma ol, 127(2): p. 262-3.
42. S anley, M.A., 2012, Epi helial cell esponses o in ec ion wi h human
papilloma i us. Clin Mic obiol Re , 25(2): p. 215-22.
43. an den Hende, M., M.I. an Poelgees , J.M. an de Huls , J. de Jong, J.W.
D ij hou , G.J. Fleu en, A.R. Valen ijn, A.R. Wa elman, G.M. Slappendel, C.J. Melie , R.
O inga, S.H. an de Bu g, and G.G. Ken e , 2008, Skin eac ions o human

65
papilloma i us (HPV) 16 speci ic an igens in ade mally injec ed in heal hy subjec s and
pa ien s wi h ce ical neoplasia. In J Cance , 123(1): p. 146-52.
44. Ca e , J.J., L.A. Kou sky, G.C. Wip , N.D. Ch is ensen, S.K. Lee, J. Kuype s,
N. Ki ia , and D.A. Galloway, 1996, The na u al his o y o human papilloma i us ype 16
capsid an ibodies among a coho o uni e si y women. J In ec Dis, 174(5): p. 927-36.
45. Hal o sen, E.C., S.M. Mahmoud, and K.L. Bennewi h, 2014, Eme ging oles
o egula o y T cells in umou p og ession and me as asis. Cance Me as asis Re .
46. Molling, J.W., T.D. de G uijl, J. Glim, M. Mo eno, L. Rozendaal, C.J. Meije ,
A.J. an den Ee wegh, R.J. Schepe , M.E. on Blombe g, and H.J. Bon kes, 2007,
CD4(+)CD25hi egula o y T-cell equency co ela es wi h pe sis ence o human
papilloma i us ype 16 and T helpe cell esponses in pa ien s wi h ce ical in aepi helial
neoplasia. In J Cance , 121(8): p. 1749-55.
47. Sco , M.E., Y. Ma, L. Kuzmich, and A.B. Moscicki, 2009, Diminished IFN-
gamma and IL-10 and ele a ed Foxp3 mRNA exp ession in he ce ix a e associa ed wi h
CIN 2 o 3. In J Cance , 124(6): p. 1379-83.
48. Jaa a , F., E. Righi, V. Linds om, C. Lin on, M. Nohadani, S. Van Noo den, T.
Lloyd, J. Poznansky, G. S amp, R. Dina, D.V. Coleman, and M.C. Poznansky, 2009,
Co ela ion o CXCL12 exp ession and FoxP3+ cell in il a ion wi h human papilloma i us
in ec ion and clinicopa hological p og ession o ce ical cance . Am J Pa hol, 175(4): p.
1525-35.
49. Lepique, A.P., K.R. Daghas anli, I.M. Cucco ia, and L.L. Villa, 2009, HPV16
umo associa ed mac ophages supp ess an i umo T cell esponses. Clin Cance Res,
15(13): p. 4391-400.
50. Man o ani, A. and M. Loca i, 2013, Tumo -associa ed mac ophages as a
pa adigm o mac ophage plas ici y, di e si y, and pola iza ion: lessons and open
ques ions. A e ioscle Th omb Vasc Biol, 33(7): p. 1478-83.
51. Riba i, D. and E. C i ella o, 2012, Mas cells, angiogenesis, and umou
g ow h. Biochim Biophys Ac a, 1822(1): p. 2-8.
52. Condeelis, J. and J.W. Polla d, 2006, Mac ophages: obliga e pa ne s o
umo cell mig a ion, in asion, and me as asis. Cell, 124(2): p. 263-6.
53. Loca i, M., A. Man o ani, and A. Sica, 2013, Mac ophage ac i a ion and
pola iza ion as an adap i e componen o inna e immuni y. Ad Immunol, 120: p. 163-84.
54. Kayagaki, N., N. Yamaguchi, M. Nakayama, A. Kawasaki, H. Akiba, K.
Okumu a, and H. Yagi a, 1999, In ol emen o TNF- ela ed apop osis-inducing ligand in
human CD4+ T cell-media ed cy o oxici y. J Immunol, 162(5): p. 2639-47.
66
55. Ru ell, B., D.G. DeNa do, N.I. A a a, and L.M. Coussens, 2010,
Lymphocy es in cance de elopmen : pola iza ion owa ds p o- umo immuni y. Cy okine
G ow h Fac o Re , 21(1): p. 3-10.
56. Jung, A.C., S. Guiha d, S. K ugell, S. Led appie , A. B ocho , V. Dals ein, S.
Job, A. de Reynies, G. Noel, B. Wasylyk, C. Cla el, and J. Abecassis, 2013, CD8-alpha
T-cell in il a ion in human papilloma i us- ela ed o opha yngeal ca cinoma co ela es wi h
imp o ed pa ien p ognosis. In J Cance , 132(2): p. E26-36.
57. Robe son, F.C., J.A. Be zo sky, and M. Te abe, 2014, NKT cell ne wo ks in
he egula ion o umo immuni y. F on Immunol, 5: p. 543.
58. Alcoce -Gonzalez, J.M., J. Be umen, R. Tamez-Gue a, V. Be mudez-
Mo ales, O. Pe al a-Za agoza, R. He nandez-Pando, J. Mo eno, P. Ga iglio, and V.
Mad id-Ma ina, 2006, In i o exp ession o immunosupp essi e cy okines in human
papilloma i us- ans o med ce ical cance cells. Vi al Immunol, 19(3): p. 481-91.
59. Me ick, D.T., R.A. Blan on, A.M. Gown, and J.K. McDougall, 1992, Al e ed
exp ession o p oli e a ion and di e en ia ion ma ke s in human papilloma i us 16 and 18
immo alized epi helial cells g own in o gano ypic cul u e. Am J Pa hol, 140(1): p. 167-77.
60. Blan on, R.A., N. Pe ez-Reyes, D.T. Me ick, and J.K. McDougall, 1991,
Epi helial cells immo alized by human papilloma i uses ha e p emalignan cha ac e is ics
in o gano ypic cul u e. Am J Pa hol, 138(3): p. 673-85.
61. Asselineau, D., B.A. Be na d, C. Bailly, M. Da mon, and M. P unie as, 1986,
Human epide mis econs uc ed by cul u e: is i "no mal"? J In es De ma ol, 86(2): p.
181-6.
62. Campo, M.S., 2002, Animal models o papilloma i us pa hogenesis. Vi us
Res, 89(2): p. 249-61.
63. Gil da Cos a, R.M. and R. Medei os, 2014, Bo ine papilloma i us: opening
new ends o compa a i e pa hology. A ch Vi ol, 159(2): p. 191-8.
64. Nguyen, D. and T. Xu, 2008, The expanding ole o mouse gene ics o
unde s anding human biology and disease. Dis Model Mech, 1(1): p. 56-66.
65. G iep, A.E., R. He be , S. Jeon, J.K. Lohse, R.R. Dubielzig, and P.F. Lambe ,
1993, Tumo igenici y by human papilloma i us ype 16 E6 and E7 in ansgenic mice
co ela es wi h al e a ions in epi helial cell g ow h and di e en ia ion. J Vi ol, 67(3): p.
1373-84.
66. A bei , J.M., K. Munge , P.M. Howley, and D. Hanahan, 1993, Neu oepi helial
ca cinomas in mice ansgenic wi h human papilloma i us ype 16 E6/E7 ORFs. Am J
Pa hol, 142(4): p. 1187-97.
67
67. Song, S., H.C. Pi o , and P.F. Lambe , 1999, The human papilloma i us ype
16 E6 gene alone is su icien o induce ca cinomas in ansgenic animals. J Vi ol, 73(7):
p. 5887-93.
68. He be , R., A. Liem, H. Pi o , and P.F. Lambe , 1996, Squamous epi helial
hype plasia and ca cinoma in mice ansgenic o he human papilloma i us ype 16 E7
oncogene. J Vi ol, 70(3): p. 1873-81.
69. Gen he Williams, S.M., G.L. Disb ow, R. Schlegel, D. Lee, D.W. Th eadgill,
and P.F. Lambe , 2005, Requi emen o epide mal g ow h ac o ecep o o hype plasia
induced by E5, a high- isk human papilloma i us oncogene. Cance Res, 65(15): p. 6534-
42.
70. Mau o , J.P., A. Shai, H.C. Pi o , and P.F. Lambe , 2010, A ole o HPV16
E5 in ce ical ca cinogenesis. Cance Res, 70(7): p. 2924-31.
71. Jabba , S., K. S a i, M.K. Shin, H.C. Pi o , and P.F. Lambe , 2010, Human
papilloma i us ype 16 E6 and E7 oncop o eins ac syne gis ically o cause head and neck
cance in mice. Vi ology, 407(1): p. 60-7.
72. S elze , M.K., H.C. Pi o , A. Liem, J. Schweize , C. Mahoney, and P.F.
Lambe , 2010, A mouse model o human anal cance . Cance P e Res (Phila), 3(12):
p. 1534-41.
73. A bei , J.M., K. Munge , P.M. Howley, and D. Hanahan, 1994, P og essi e
squamous epi helial neoplasia in K14-human papilloma i us ype 16 ansgenic mice. J
Vi ol, 68(7): p. 4358-68.
74. Coussens, L.M., D. Hanahan, and J.M. A bei , 1996, Gene ic p edisposi ion
and pa ame e s o malignan p og ession in K14-HPV16 ansgenic mice. Am J Pa hol,
149(6): p. 1899-917.
75. A bei , J.M., P.M. Howley, and D. Hanahan, 1996, Ch onic es ogen-induced
ce ical and aginal squamous ca cinogenesis in human papilloma i us ype 16
ansgenic mice. P oc Na l Acad Sci U S A, 93(7): p. 2930-5.
76. Smi h-McCune, K., Y.H. Zhu, D. Hanahan, and J. A bei , 1997, C oss-species
compa ison o angiogenesis du ing he p emalignan s ages o squamous ca cinogenesis
in he human ce ix and K14-HPV16 ansgenic mice. Cance Res, 57(7): p. 1294-300.
77. Kimple, R.J., P.M. Ha a i, A.D. To es, R.Z. Yang, B.J. So iano, M. Yu, E.A.
A ms ong, G.C. Bli ze , M.A. Smi h, L.D. Lo enz, D. Lee, D.T. Yang, T.M. McCulloch,
G.K. Ha ig, and P.F. Lambe , 2013, De elopmen and cha ac e iza ion o HPV-posi i e
and HPV-nega i e head and neck squamous cell ca cinoma umo g a s. Clin Cance
Res, 19(4): p. 855-64.
68
78. Ahn, J., J.A. Bishop, B. Akpeng, S.I. Pai, and S.R. Bes , 2015, Xenog a
model o he apeu ic d ug es ing in ecu en espi a o y papilloma osis. Ann O ol Rhinol
La yngol, 124(2): p. 110-5.
79. Sche e , W.F., J.T. Sy e on, and G.O. Gey, 1953, S udies on he
p opaga ion in i o o poliomyeli is i uses. IV. Vi al mul iplica ion in a s able s ain o
human malignan epi helial cells (s ain HeLa) de i ed om an epide moid ca cinoma o
he ce ix. J Exp Med, 97(5): p. 695-710.
80. Capes-Da is, A., G. Theodosopoulos, I. A kin, H.G. D exle , A. Koha a, R.A.
MacLeod, J.R. Mas e s, Y. Nakamu a, Y.A. Reid, R.R. Reddel, and R.I. F eshney, 2010,
Check you cul u es! A lis o c oss-con amina ed o misiden i ied cell lines. In J Cance ,
127(1): p. 1-8.
81. Pa illo, R.A., R.O. Hussa, M.T. S o y, A.C. Rucke , M.R. Shalaby, and R.F.
Ma ingly, 1977, Tumo an igen and human cho ionic gonado opin in CaSki cells: a new
epide moid ce ical cance cell line. Science, 196(4297): p. 1456-8.
82. F iedl, F., I. Kimu a, T. Osa o, and Y. I o, 1970, S udies on a new human cell
line (SiHa) de i ed om ca cinoma o u e us. I. I s es ablishmen and mo phology. P oc
Soc Exp Biol Med, 135(2): p. 543-5.
83. Bosha , M., L. Gissmann, H. Ikenbe g, A. Kleinheinz, W. Scheu len, and H.
zu Hausen, 1984, A new ype o papilloma i us DNA, i s p esence in geni al cance
biopsies and in cell lines de i ed om ce ical cance . EMBO J, 3(5): p. 1151-7.
84. Yee, C., I. K ishnan-Hewle , C.C. Bake , R. Schlegel, and P.M. Howley, 1985,
P esence and exp ession o human papilloma i us sequences in human ce ical
ca cinoma cell lines. Am J Pa hol, 119(3): p. 361-6.
85. Nakamu a, T., L. Williams-Simons, and H. Wes phal, 1997, A human
papilloma i us ype 18 E6/E7 ansgene sensi izes mouse lens cells o human wild- ype
p53-media ed apop osis. Oncogene, 14(25): p. 2991-8.
86. Leykau , K., K. Kabsch, N. Gassle , L. Gissmann, A. Alonso, and J. Schenkel,
2008, Exp ession o he HPV11 E2 gene in ansgenic mice does no esul in al e a ions
o he pheno ypic pa e n. T ansgenic Res, 17(1): p. 1-8.
87. Schenkel, J., H. Gaisse , E.E. P o opapa, H. Weihe , L. Gissmann, and A.
Alonso, 1999, The human papilloma i us ype 11 ups eam egula o y egion igge s
hai - ollicle-speci ic gene exp ession in ansgenic mice. J In es De ma ol, 112(6): p. 893-
8.
88. Cid, A., P. Auewa akul, A. Ga cia-Ca anca, R. O seio ich, H. Gaisse , and
L. Gissmann, 1993, Cell- ype-speci ic ac i i y o he human papilloma i us ype 18
ups eam egula o y egion in ansgenic mice and i s modula ion by e adecanoyl pho bol
ace a e and glucoco icoids. J Vi ol, 67(11): p. 6742-52.
69
89. P e e le, R., G.P. Ma cuzzi, B. Akgul, H.U. Kaspe , F. Schulze, I. Haase, C.
Wickenhause , and H. P is e , 2008, The human papilloma i us ype 8 E2 p o ein induces
skin umo s in ansgenic mice. J In es De ma ol, 128(9): p. 2310-5.
90. Ghim, S., A.B. Jenson, J.A. Bubie , K.A. Sil a, R.S. Smi h, and J.P. Sundbe g,
2008, Ca a ac s in ansgenic mice caused by a human papilloma i us ype 18 E7
oncogene d i en by KRT1-14. Exp Mol Pa hol, 85(2): p. 77-82.
91. Schape , I.D., G.P. Ma cuzzi, S.J. Weissenbo n, H.U. Kaspe , V. D ies, N.
Smy h, P. Fuchs, and H. P is e , 2005, De elopmen o skin umo s in mice ansgenic o
ea ly genes o human papilloma i us ype 8. Cance Res, 65(4): p. 1394-400.
92. Daniel, D., N. Meye -Mo se, E.K. Be gsland, K. Dehne, L.M. Coussens, and
D. Hanahan, 2003, Immune enhancemen o skin ca cinogenesis by CD4+ T cells. J Exp
Med, 197(8): p. 1017-28.
93. de Visse , K.E., L.V. Ko e s, and L.M. Coussens, 2005, De no o
ca cinogenesis p omo ed by ch onic in lamma ion is B lymphocy e dependen . Cance
Cell, 7(5): p. 411-23.
94. Coussens, L.M., W.W. Raymond, G. Be ge s, M. Laig-Webs e , O.
Beh end sen, Z. We b, G.H. Caughey, and D. Hanahan, 1999, In lamma o y mas cells
up- egula e angiogenesis du ing squamous epi helial ca cinogenesis. Genes De , 13(11):
p. 1382-97.
95. Coussens, L.M., C.L. Tinkle, D. Hanahan, and Z. We b, 2000, MMP-9
supplied by bone ma ow-de i ed cells con ibu es o skin ca cinogenesis. Cell, 103(3): p.
481-90.
96. an Kempen, L.C., J.S. Rhee, K. Dehne, J. Lee, D.R. Edwa ds, and L.M.
Coussens, 2002, Epi helial ca cinogenesis: dynamic in e play be ween neoplas ic cells
and hei mic oen i onmen . Di e en ia ion, 70(9-10): p. 610-23.
97. Rhee, J.S., R. Diaz, L. Ko e s, J.G. Hodgson, and L.M. Coussens, 2004,
TIMP-1 al e s suscep ibili y o ca cinogenesis. Cance Res, 64(3): p. 952-61.
98. de Visse , K.E., L.V. Ko e s, and L.M. Coussens, 2004, Ea ly neoplas ic
p og ession is complemen independen . Neoplasia, 6(6): p. 768-76.
99. And eu, P., M. Johansson, N.I. A a a, F. Pucci, T. Tan, S. Junanka , L.
Ko e s, J. Lam, D. Taw ik, D.G. DeNa do, L. Naldini, K.E. de Visse , M. De Palma, and
L.M. Coussens, 2010, FcRgamma ac i a ion egula es in lamma ion-associa ed
squamous ca cinogenesis. Cance Cell, 17(2): p. 121-34.
100. Li, J.J., Y. Cao, M.R. Young, and N.H. Colbu n, 2000, Induced exp ession o
dominan -nega i e c-jun down egula es NFkappaB and AP-1 a ge genes and
supp esses umo pheno ype in human ke a inocy es. Mol Ca cinog, 29(3): p. 159-69.

70
101. Ma hews, C.P., A.M. Bi kholz, A.R. Bake , C.M. Pe ella, G.R. Beck, J ., M.R.
Young, and N.H. Colbu n, 2007, Dominan -nega i e ac i a o p o ein 1 (TAM67) a ge s
cyclooxygenase-2 and os eopon in unde condi ions in which i speci ically inhibi s
umo igenesis. Cance Res, 67(6): p. 2430-8.
102. Bo si, E., C. Te agna, A. B ioli, P. Tacche i, M. Ma ello, and M. Ca o,
2014, The apeu ic a ge ing o hypoxia and hypoxia-inducible ac o 1alpha in mul iple
myeloma. T ansl Res.
103. Tang, X., Q. Zhang, J. Nishi ani, J. B own, S. Shi, and A.D. Le, 2007,
O e exp ession o human papilloma i us ype 16 oncop o eins enhances hypoxia-
inducible ac o 1 alpha p o ein accumula ion and ascula endo helial g ow h ac o
exp ession in human ce ical ca cinoma cells. Clin Cance Res, 13(9): p. 2568-76.
104. Li, G., L. He, E. Zhang, J. Shi, Q. Zhang, A.D. Le, K. Zhou, and X. Tang,
2011, O e exp ession o human papilloma i us (HPV) ype 16 oncop o eins p omo es
angiogenesis ia enhancing HIF-1alpha and VEGF exp ession in non-small cell lung
cance cells. Cance Le , 311(2): p. 160-70.
105. Elson, D.A., H.E. Ryan, J.W. Snow, R. Johnson, and J.M. A bei , 2000,
Coo dina e up- egula ion o hypoxia inducible ac o (HIF)-1alpha and HIF-1 a ge genes
du ing mul i-s age epide mal ca cinogenesis and wound healing. Cance Res, 60(21): p.
6189-95.
106. Lu, Z.H., J.D. W igh , B. Bel , R.D. Ca di , and J.M. A bei , 2007, Hypoxia-
inducible ac o -1 acili a es ce ical cance p og ession in human papilloma i us ype 16
ansgenic mice. Am J Pa hol, 171(2): p. 667-81.
107. Guo, W. and F.G. Gianco i, 2004, In eg in signalling du ing umou
p og ession. Na Re Mol Cell Biol, 5(10): p. 816-26.
108. T an, T., B. Ba low, L. O'Rea , B. Ja is, Z. Li, K. Dickeson, W. Dupon , and
M. Zu e , 2011, Loss o he alpha2be a1 in eg in al e s human papilloma i us-induced
squamous ca cinoma p og ession in i o and in i o. PLoS One, 6(10): p. e26858.
109. Mohamed, M.M. and B.F. Sloane, 2006, Cys eine ca hepsins:
mul i unc ional enzymes in cance . Na Re Cance , 6(10): p. 764-75.
110. Dennema ke , J., T. Lohmulle , J. Maye le, M. Tacke, M.M. Le ch, L.M.
Coussens, C. Pe e s, and T. Reinheckel, 2010, De iciency o he cys eine p o ease
ca hepsin L p omo es umo p og ession in mouse epide mis. Oncogene, 29(11): p. 1611-
21.
111. Yu, W., J. Liu, M.A. Shi, J. Wang, M. Xiang, S. Ki amo o, B. Wang, G.K.
Sukho a, G.F. Mu phy, G. O asanu, A. G ubb, and G.P. Shi, 2010, Cys a in C de iciency
p omo es epide mal dysplasia in K14-HPV16 ansgenic mice. PLoS One, 5(11): p.
e13973.
71
112. Shah abi-Fa ahani, S., L. Wang, B.M. Zwaans, J.M. San ana, A. Shimizu, S.
Takashima, M. K eu e , L. Coul as, P.A. D'Amo e, J.M. A bei , L.A. Akslen, and D.R.
Bielenbe g, 2014, Neu opilin 1 exp ession co ela es wi h di e en ia ion s a us o
epide mal cells and cu aneous squamous cell ca cinomas. Lab In es , 94(7): p. 752-65.
113. Pai a, I., R.M. Gil da Cos a, J. Ribei o, H. Sousa, M. Bas os, A.F. Rocha,
P.A. Oli ei a, and R. Medei os, 2015, A Role o Mic oRNA-155 Exp ession in
Mic oen i onmen Associa ed o HPV-Induced Ca cinogenesis in K14-HPV16 T ansgenic
Mice. PLoS One, 10(1): p. e0116868.
114. Pai a, I., R.M. Gil da Cos a, J. Ribei o, H. Sousa, M.M. Bas os, A. Faus ino-
Rocha, C. Lopes, P.A. Oli ei a, and R. Medei os, 2015, Mic oRNA-21 exp ession and
suscep ibili y o HPV-induced ca cinogenesis - ole o mic oen i onmen in K14-HPV16
mice model. Li e Sci, 128: p. 8-14.
115. Senba, M. and N. Mo i, 2012, Mechanisms o i us immune e asion lead o
de elopmen om ch onic in lamma ion o cance o ma ion associa ed wi h human
papilloma i us in ec ion. Oncol Re , 6(2): p. e17.
116. Mesche , M.F., J.M. Cu singe , P. Aga wal, K.A. Casey, M. Ge ne , C.D.
Hamme beck, F. Popescu, and Z. Xiao, 2006, Signals equi ed o p og amming e ec o
and memo y de elopmen by CD8+ T cells. Immunol Re , 211: p. 81-92.
117. an Mon oo , N., E. an de Aa, and A.M. Wol man, 2014, Unde s anding
MHC class I p esen a ion o i al an igens by human dend i ic cells as a basis o a ional
design o he apeu ic accines. F on Immunol, 5: p. 182.
118. Dunn, G.P., L.J. Old, and R.D. Sch eibe , 2004, The h ee Es o cance
immunoedi ing. Annu Re Immunol, 22: p. 329-60.
119. T apani, J.A. and M.J. Smy h, 2002, Func ional signi icance o he
pe o in/g anzyme cell dea h pa hway. Na Re Immunol, 2(10): p. 735-47.
120. Bu e ne, M., L. Chasson, M. G ange, A. Mas, F. A noux, M. Be uzzi, P.
Naque , L. Lese man, A.M. Schmi -Ve huls , and N. Auphan-Anezin, 2015, IFNgamma
p oducing CD8 T cells modi ied o esis majo immune checkpoin s induce eg ession o
MHC class I-de icien melanomas. Oncoimmunology, 4(2): p. e974959.
121. Qiu, Z., H. Huang, J.M. G enie , O.A. Pe ez, H.M. Smilowi z, B. Adle , and
K.M. Khanna, 2015, Cy omegalo i us-Based Vaccine Exp essing a Modi ied Tumo
An igen Induces Po en Tumo -Speci ic CD8(+) T-cell Response and P o ec s Mice om
Melanoma. Cance Immunol Res, 3(5): p. 536-46.
122. Belapu ka , P., P. Goyal, and P. Tiwa i-Ba ua, 2014, Immunomodula o y
e ec s o iphala and i s indi idual cons i uen s: a e iew. Indian J Pha m Sci, 76(6): p.
467-75.
72
123. Gil da Cos a, R.M., M.M. Bas os, P.A. Oli ei a, and C. Lopes, 2012,
B acken-associa ed human and animal heal h haza ds: chemical, biological and
pa hological e idence. J Haza d Ma e , 203-204: p. 1-12.
124. Chang, F., S. Sy janen, Q. Shen, L. Wang, D. Wang, and K. Sy janen, 1992,
Human papilloma i us in ol emen in esophageal p ecance ous lesions and squamous
cell ca cinomas as e idenced by mic oscopy and di e en DNA echniques. Scand J
Gas oen e ol, 27(7): p. 553-63.
125. Hi ono, I., Y. Kono, K. Takahashi, K. Yamada, H. Niwa, M. Ojika, H. Kigoshi,
K. Niiyama, and Y. Uosaki, 1984, Rep oduc ion o acu e b acken poisoning in a cal wi h
p aquiloside, a b acken cons i uen . Ve Rec, 115(15): p. 375-8.
126. La o e, A.O., B.D. Canicei o, H.L. Wysocki, J ., M. Ha aguchi, D.R. Ga dne ,
and S.L. Go niak, 2011, Selenium e e ses P e idium aquilinum-induced immuno oxic
e ec s. Food Chem Toxicol, 49(2): p. 464-70.
127. La o e, A.O., M.S. Fu lan, M. Sakai, H. Fukumasu, I.M. Hueza, M.
Ha aguchi, and S.L. Go niak, 2009, Immunomodula o y e ec s o P e idium aquilinum on
na u al kille cell ac i i y and selec aspec s o he cellula immune esponse o mice. J
Immuno oxicol, 6(2): p. 104-14.
128. Gil da Cos a, R.M., P.A. Oli ei a, M. Vilano a, M.M. Bas os, C.C. Lopes, and
C. Lopes, 2011, P aquiloside-induced, B-cell lymphop oli e a i e and ea ly-s age
u o helial lesions in mice. Toxicon, 58(6-7): p. 543-9.
129. Ade uye, A. and K. Sales, 2012, Regula ion o in lamma o y pa hways in
cance and in ec ious disease o he ce ix. Scien i ica (Cai o), 2012: p. 548150.
130. Wol , H., K. Saukkonen, S. An ila, A. Ka jalainen, H. Vainio, and A.
Ris imaki, 1998, Exp ession o cyclooxygenase-2 in human lung ca cinoma. Cance Res,
58(22): p. 4997-5001.
131. Sano, H., Y. Kawahi o, R.L. Wilde , A. Hashi amo o, S. Mukai, K. Asai, S.
Kimu a, H. Ka o, M. Kondo, and T. Hla, 1995, Exp ession o cyclooxygenase-1 and -2 in
human colo ec al cance . Cance Res, 55(17): p. 3785-9.
132. Ris imaki, A., N. Honkanen, H. Jankala, P. Sipponen, and M. Ha konen,
1997, Exp ession o cyclooxygenase-2 in human gas ic ca cinoma. Cance Res, 57(7):
p. 1276-80.
133. Golijanin, D., J.Y. Tan, A. Kazio , E.G. Cohen, P. Russo, G. Dalbagni, K.J.
Aubo n, K. Subba amaiah, and A.J. Dannenbe g, 2004, Cyclooxygenase-2 and
mic osomal p os aglandin E syn hase-1 a e o e exp essed in squamous cell ca cinoma o
he penis. Clin Cance Res, 10(3): p. 1024-31.
73
134. Ryu, H.S., K.H. Chang, H.W. Yang, M.S. Kim, H.C. Kwon, and K.S. Oh,
2000, High cyclooxygenase-2 exp ession in s age IB ce ical cance wi h lymph node
me as asis o pa ame ial in asion. Gynecol Oncol, 76(3): p. 320-5.
135. McAdam, B.F., F. Ca ella-Lawson, I.A. Ma dini, S. Kapoo , J.A. Lawson, and
G.A. Fi zGe ald, 1999, Sys emic biosyn hesis o p os acyclin by cyclooxygenase (COX)-
2: he human pha macology o a selec i e inhibi o o COX-2. P oc Na l Acad Sci U S A,
96(1): p. 272-7.
136. Hahn, T., I. Al a ez, J.J. Kobie, L. Ramana hapu am, S. Dial, A. Ful on, D.
Besselsen, E. Walke , and E.T. Akpo iaye, 2006, Sho - e m die a y adminis a ion o
celecoxib enhances he e icacy o umo lysa e-pulsed dend i ic cell accines in ea ing
mu ine b eas cance . In J Cance , 118(9): p. 2220-31.
137. He mann, M., S. Shaw, E. Kiss, G. Camici, N. Buhle , R. Chene a d, T.F.
Lusche , H.J. G one, and F. Ruschi zka, 2005, Selec i e COX-2 inhibi o s and enal inju y
in sal -sensi i e hype ension. Hype ension, 45(2): p. 193-7.
138. Kosaka, A., T. Ohku i, and H. Okada, 2014, Combina ion o an agonis ic
an i-CD40 monoclonal an ibody and he COX-2 inhibi o celecoxib induces an i-glioma
e ec s by p omo ion o ype-1 immuni y in myeloid cells and T-cells. Cance Immunol
Immuno he , 63(8): p. 847-57.
139. K ale, D., V. O maasen, A.M. K an, C.C. Johansson, P. Auk us , E.M.
Aandahl, S.S. F oland, and K. Tasken, 2006, Immune modula o y e ec s o
cyclooxygenase ype 2 inhibi o s in HIV pa ien s on combina ion an i e o i al ea men .
AIDS, 20(6): p. 813-20.
140. Pe e sen, F.O., E.A. To heim, A.E. Dahm, I.S. Aabe ge, A. Lind, M. Holm,
E.M. Aandahl, P.M. Sandse , K. Tasken, and D. K ale, 2011, An explo a o y ial o
cyclooxygenase ype 2 inhibi o in HIV-1 in ec ion: down egula ed immune ac i a ion and
imp o ed T cell-dependen accine esponses. J Vi ol, 85(13): p. 6557-66.
141. Bzhala a, D., P. Guan, S. F anceschi, J. Dillne , and G. Cli o d, 2013, A
sys ema ic e iew o he p e alence o mucosal and cu aneous human papilloma i us
ypes. Vi ology, 445(1-2): p. 224-31.
142. Hoo s, B.E., J.M. Pale sky, J.M. Pimen a, and J.S. Smi h, 2009, Human
papilloma i us ype dis ibu ion in anal cance and anal in aepi helial lesions. In J
Cance , 124(10): p. 2375-83.
143. Balkwill, F., K.A. Cha les, and A. Man o ani, 2005, Smolde ing and
pola ized in lamma ion in he ini ia ion and p omo ion o malignan disease. Cance Cell,
7(3): p. 211-7.