Role of RKIP and Pirin in the malignant progression of cutaneous melanoma. New diagnosis and prognosis biomarkers

DOCTORAL THESIS
ROLE OF RKIP AND PIRIN IN THE
MALIGNANT PROGRESSION OF
CUTANEOUS MELANOMA. NEW
DIAGNOSIS AND PROGNOSIS
BIOMARKERS
CRISTINA PENAS LAGO
LEIOA, 2022
Supe iso s:
Ma ia Dolo es Boyano López
Ca men Ál a ez Domínguez
(cc)2022 CRISTINA PENAS LAGO (cc by-nc-nd 4.0)

II

III
Abb e ia ion
Desc ip ion
ACTB
Ac in be a
AJCC
Ame ican Join Commi ee on Cance
AJCC
Ame ican Join Commi ee on Cance
AKT
P o ein Kinase B
ALM
Ac al len igo melanoma
ANOVA
a iance analysis
BCL-3
B-cell lymphoma 3-encoded p o ein
BIC
Bayesian In o ma ion C i e ion
BPs
Biological P ocesses
BRAF
B-Ra P o o-Oncogene, Se ine/Th eonine Kinase
BSA
Bo ine Se um Albumin
Cdna
complemen a y DNA
c-MYC
Myc p o o-oncogene, bhlh ansc ip ion ac o
c-Rel,
REL P o o-Oncogene, NF-KB Subuni
CSD
Cumula i e sun damage
CT
compu ed omog aphy
DEGs
Di e en ially exp essed genes
DEPC
die hylpy oca bona e
DMEM
Dulbecco's Modi ied Eagle Medium
DMSO
dime hyl sul oxide
DNA
Deoxy ibonucleic acid
DTT
Di hio h ei ol
E2F1
E2 ansc ip ion ac o 1
EAPC
Es ima ed Annual Pe cen age Change
ECL
Enhanced chemiluminescence
ECM
Ex acellula ma ix
EDTA
E hylenediamine e aace ic acid
EGFP
Enhanced G een Fluo escen P o ein
EMT
Epi helial- o-mesenchymal ansi ion
ERK1
Mi ogen-Ac i a ed P o ein Kinase 3
ERK2
Mi ogen-Ac i a ed P o ein Kinase 1
FBS
Fe al bo ine se um
FDR
False disco e y a e
FFPE
Fo maldehyde Fixed Pa a in Embedded
GAPDH
Glyce aldehyde 3-phospha e dehyd ogenase
GEO
Gene exp ession omnibus
GFP
G een luo escence p o ein
GO
Gene On ology
GPCR
G p o ein-coupled ecep o s
GSK3β
Glycogen syn hase kinase-3 be a
H&E
hema oxylin and eosin s aining
HEMn-DP
Human epide mal melanocy es, neona al, da kly
pigmen ed
HEMn-LP
Human epide mal melanocy es, neona al, ligh ly
pigmen ed
HEMn-MP
Human epide mal melanocy es, neona al,
mode a ely pigmen ed
HMB-45
an i-melanosoma, HMB45
HRP
Ho se adish Pe oxidase
IGF-1Rs
Insulin-like g ow h ac o I ecep o
IHC
immunohis ochemis y
IV
IKK
Inhibi o O Nuclea Fac o Kappa B Kinase
JAK
Janus Kinase
JARID1B
Lysine Deme hylase 5B gene
KEGG
Kyo o Encyclopedia o Genes and Genomes
KLF4
K uppel Like Fac o 4
KRAS
KRAS P o o-Oncogene, GTPase
LM
Len igo melanoma
LMM
Len igo malignan melanoma
LUM
Lumican
MAPK
Mi ogen-ac i a ed p o ein kinase
MART-1/Melan-
A
Melanoma An igen Recognized by T Cells
MEK
Mi ogen-Ac i a ed P o ein Kinase Kinase 1
MIB1
MIB E3 Ubiqui in P o ein Ligase 1
miR-21
Mic oRNA 21
MITF
Melanocy e Inducing T ansc ip ion Fac o
MOI
Mul iplici y o In ec ion
mRNA
messenge Ribonucleic Acid
NANOG
Nanog homeobox
NCSC
neu al-c es s em-cell
NF1
neu o ib omin 1
NFKB
nuclea ac o -kappaB
NGS
Nex Gene a ion Sequencing
NM
Nodula melanoma
NRAS
NRAS P o o-Oncogene, GTPase
NTRK2
Neu o ophic Recep o Ty osine Kinase 2
OCT4
POU class 5 homeobox 1
OIS
Oncogene-induced senescence
p100/p52
Nuclea Fac o Kappa B Subuni 2
p105/p50
Nuclea Fac o Kappa B Subuni 1
PBS
Phospha e bu e saline
PCR
Polyme ase chain eac ion
PI3K
phospha idylinosi ol-3-kinases
PIR
Pi in
PKC
P o ein Kinase C
PMEL o gp-100
p emelanosome p o ein
PTEN
Phospha ase And Tensin Homolog
RAF
Ra -1 P o o-Oncogene, Se ine/Th eonine Kinase
RelA (p65)
RELA P o o-Oncogene, NF-KB Subuni
RelB
RELB P o o-Oncogene, NF-KB Subuni
RGP
adial g ow h p oli e a ion
RIN
RNA In eg i y Numbe
RIPA
Radioimmunop ecipi a ion assay bu e
RIPA
Radioimmunop ecipi a ion Assay
RKIP
Ra kinase Inhibi o p o ein
RNA
ibonucleic acid
RNU6-2
endogenous e e ence small nuclea ibonucleic
acid (snRNA) B
RPKM
Reads pe kilobase o exon model
RPMI
Roswell Pa k Memo ial Ins i u e Medium
RPS15
Ribosomal P o ein S15 gene
RPS15
Ribosomal P o ein S15
V
RTKs
ecep o y osine kinases
RT-qPCR
Real ime quan i a i e polyme ase chain eac ion
S100
S100 Calcium Binding P o ein
SDS
Sodium dodecyl sul a e
shRNA
Sho hai pin RNA
SNAIL
Snail ansc ip ional ep esso
SOX2
SRY-box ansc ip ion ac o 2
SRA
Sequence ead a chi e
SRCCA
Spea man’s ank co ela ion coe icien
SSM
Supe icial sp ead melanoma
STAT
Signal T ansduce And Ac i a o O T ansc ip ion
TAK1
Mi ogen-Ac i a ed P o ein Kinase Kinase Kinase 7
TBST
T is-bu e ed saline wi h Tween 20
TBST
T is-bu e ed saline
THY-1
THY-1 cell su ace an igen
TMM
T immed mean o M alues
TNFR1
TNF Recep o Supe amily Membe 1A
TYR
y osinase
UV
Ul a iole
VGP
e ical g ow h p oli e a ion
WHO
Wo ld Heal h O ganiza ion
XTT
2,3-bis-(2-me hoxy-4-ni o-5-sul ophenyl)-2H-
e azolium-5-ca boxanilide
ZEB
Zinc Finge E-box-binding homeobox
VI
VII

VIII
IX
Figu e 1. Tempo al incidence/mo ali y ends o Melanoma
Figu e 2. Melanoma incidence
Figu e 3. Melanoma mo ali y
Figu e 4. Skin pho o ypes
Figu e 5. C oss-sec ion o skin’s laye s
Figu e 6. Biologic e en s and molecula changes in he p og ession o melanoma
Figu e 7. Melanoma key signaling pa hways
Figu e 8. ABCDE c i e ia o he ea ly de ec ion o melanoma
Figu e 9. His ological sub ypes o melanoma and clinical-pa hological co ela ions
Figu e 10. Dis ibu ion o melanoma cell lines and p ima y cul u es o melanocy es
analysed in a wo-componen plo by Va imax o hogonal o a ion
Figu e 11. RKIP and Pi in exp ession in cell cul u e o melanocy ic cells by RT-qPCR
and Wes e n blo
Figu e 12. Immunos aining o Pi in in h ee ep esen a i e melanoma biopsies
(nega i e, low and high)
Figu e 13. Wo k low o e iew o ansduc ion p o ocol wi h len i i al pa icles and
ans ec ion p o ocol wi h lipo ec amine and o e exp ession plasmids.
Figu e 14. Wo k low o e iew o wound healing assay
Figu e 15. Wo k low o e iew o he p o ein de ec ion p ocess by wes e n blo
Figu e 16. RNA in eg i y numbe s (RINs) ob ained wi h Agilen 2100 Bioanalyze
Figu e 17. Lib a y p o ile quali ica ion by Agilen High Sensi i i y DNA Ki
Figu e 18. Mapped eads pe con ing analysed by bam ool s a s bioin o ma ics ool
Figu e 19. Dis ibu ion o he samples acco ding o diagnosis, umou loca ion and
his ological sub ype, AJCC s ages, B eslow Index exp essed in mm and e olu ion o
pa ien s diagnosed in AJCC I and II s ages
X
Figu e 20. RKIP exp ession pa e n dis ibu ion acco ding o diagnosis, his ological
ype and p ogession o melanoma pa ien s
Figu e 21. S a is ical analyses o Ra Kinase Inhibi o P o ein (RKIP) exp ession in
FFPE biopsies om pa ien s
Figu e 22. Rep esen a i e images o manually sco ed RKIP s aining ca ego ies in
FFPE biopsies om pa ien s
Figu e 23. Rep esen a i e images o Pi in s aining in FFPE biopsie om a melanoma
hema oxylin-eosin s aining and imunohis ochemis y o Pi in exp ession
Figu e 24. Pi in exp ession pa e n dis ibu ion acco ding o diagnosis, his ological
ype and p og esion o melanoma pa ien s
Figu e 25. S a is ical analyses o Pi in exp ession in FFPE biopsies om pa ien s
Figu e 26. Modula ion o RKIP exp ession in A375 and MelHO p ima y melanoma
cell lines
Figu e 27. Modula ion o RKIP exp ession in MeWO and A2058 me as a ic
melanoma cell lines
Figu e 28. RNA Sequencing da a quali y o RKIP down egula ed HEMn-LP
Figu e 29. RNA Sequencing da a and analysis a e RKIP down egula ion in HEMn-
LP
Figu e 30. RKIP as a key egula o o NANOG exp ession in melanoma
Figu e 31. Func ional assays a e Pi in up egula ion in melanoma cells
Figu e 32. RNA Sequencing da a quali y o Pi in down egula ed HEMn-LP
Figu e 33. RNA Sequencing da a analysis and alida ion a e Pi in down egula ion
in HEMn-LP
Figu e 34. PIR as a key egula o o JARID1B exp ession in melanoma
XI
XVIII
(58). La inhibición de Pi ina se ha asociado con la capacidad mig a o ia de las células de melanoma
(122) y ambién se ha p opues o como un inhibido de la senescencia celula aunque se sabe poco sob e
los mecanismos que subyacen a es e e ec o (135).
La combinación de odos es os cambios gené icos conduce a que los melanoci os no males adquie an
di e sos eno ipos malignos. Así, se ha desc i o un equilib io en e los eno ipos p oli e a i os e
in asi os en las células de melanoma, lo cual pe mi e que las células sean capaces de mos a una
plas icidad de, lo que se ha llamado, eno ipo adap a i o. De acue do con es a p opues a de plas icidad
eno ípica, los eno ipos es ables de melanoma se de inen po ansc ip omas de melanoci os
di e enciados o melanoci os de ciclo len o o ‘slow-cycling cells’ con ma cado es de célula mad e o
‘s em-like cell make s’. Las p opiedades asociadas al pe il ‘s em-like cell’ pueden explica la
pe sis encia de cie as células umo ales as el a amien o con á macos. Uno de los genes elacionados
con la egulación del man enimien o del es ado desdi e enciación es NANOG (70). Así, po ejemplo,
du an e la o mación de es e as en el melanoma, se inc emen a la exp esión de NANOG (74), y se ha
is o su implicación en la egulación de la ansición epi elio-mesenquima osa, p oceso clásicamen e
ligado al aumen o de la mo ilidad de las células cance osas a o eciendo la diseminación de la
en e medad (75). Po o o lado, el eno ipo asociado a ciclo p oli e a i o len o o ‘slow cycling’ ambién
se ha elacionado con una subpoblación de células que man iene la supe i encia de las células
umo ales (191). En e los egulado es maes os del ciclo celula e a dado se encuen a
JARID1B/KDM5B, o enzima desme ilasa de his onas especí ica de lisinas. Va ios es udios han
p opo cionado e idencia de que JARID1B es un sup eso de umo es en el melanoma maligno, ya que
sus ni eles de exp esión es án egulados a la baja e inhibe la p oli e ación celula de mane a dependien e
de Rb (82-84). Dado que la plas icidad del eno ipo es á es echamen e elacionada con el inicio, la
p og esión y la esis encia a la e apia del melanoma, es de g an in e és iden i ica cómo se gene an
es as ansiciones de eno ipo a in de iden i ica nue os en oques e apéu icos.
En o o o den de cosas, los melanomas ambién mues an una amplia a iedad a ni el his ológico,
pudiendo mos a ca ac e ís icas epi eliales, hema ológicas, mesenquima osas y neu ales, que a menudo
pueden di icul a el diagnós ico de la en e medad (96). De hecho, se ha desa ollado una a iedad de
ma cado es inmunohis oquímicos a in de acili a la labo del pe sonal clínico de los se icios de
de ma ología, como el ma cado S-100, que sigue siendo el ma cado más sensible pa a las lesiones
melanocí icas. O os ma cado es, como HMB-45, MART-1/Melan-A, i osinasa y MITF son
ela i amen e sensibles, aunque no an o como S-100. Ac ualmen e, es os bioma cado es u ilizados pa a
acili a el diagnós ico de melanoma pe mi en di e encia umo es melanocí icos de o os ipos de
umo es, pe o no se ha demos ado que alguno de ellos sea p edic i o de supe i encia pa a pacien es
con neoplasias melanocí icas. Con el desa ollo de a amien os más no edosos y especí icos, los
bioma cado es umo ales son cada día más impo an es de ca a a su u ilización en nue as es a egias
e apéu icas (106,107).

XIX
En es udios de p o eómica di e encial p e ios ealizados po nues o g upo de in es igación,
obse amos que RKIP y Pi ina e an dos p o eínas cuya exp esión di e ía signi ica i amen e en e
melanoci os de piel y células de melanoma. Po ello, conside amos que p esen aban g an po encial pa a
se u ilizados como bioma cado es del melanoma maligno.
Hipó esis y obje i os
En base a odo lo expues o an e io men e, en es a esis se ha man enido la hipó esis de que RKIP y
Pi ina son p o eínas que pueden se excelen es bioma cado es pa a el diagnós ico y p onós ico del
melanoma cu áneo además de desempeña un papel impo an e en la e iopa ogenia del melanoma
cu áneo.
Así, pa a de e mine la e acidad de es a hipó esis, nos plan eamos los siguien es obje i os especí icos:
1. Valida la exp esión de RKIP y Pi ina como ma cado es de diagnós ico y p onós ico en
melanoma.
2. E alua el papel de RKIP y Pi ina en las unciones biológicas de los melanoci os de piel.
3. De e mina en que medida RKIP y Pi ina con ibuyen a la p og ession me as ásica del
melanoma maligno.
Ma e ial y mé odos
Nues a es a egia me odológica pa a alo a el p ime obje i o se basó en un es udio e ospec i o en
una coho e pacien es con melanoma en el que se examinó la exp esión de las p o eínas RKIP y Pi ina
en co es his ológicos de biopsias de melanoma median e inmunohis oquímica (IHC). El seguimien o
clínico de los pacien es aba có de 18 meses a 5 años. Pos e io men e, se ealizó una co elación con los
da os clinicopa ológicos y la p og esión me as ásica.
Pa a e alua el segundo obje i o sob e las unciones de las p o eínas RKIP y Pi ina en la biología de
los melanoci os, ealizamos una e aluación de los cambios ansc ipcionales as el silenciamien o de
la exp esión de ambos genes de o ma independien e en células de melanoci os sanos. Los con oles y
los melanoci os RKIP o PIR silenciados se examina on median e secuenciación de ARN. Luego, los
esul ados del análisis in silico de en iquecimien o de on ología génica se alida on en líneas celula es
de melanoma a las que se les ha sob eexp esado RKIP o Pi ina, espec i amen e.
Pa a abo da el e ce obje i o, se u iliza on líneas celula es de melanoma humano p ima io y
me as ásico. Tan o la exp esión de RKIP como la de Pi ina ue on moduladas po plásmidos de o ma
XX
independien e y se ealiza on ensayos uncionales de la capacidad p oli e a i a, mig ación e in asión.
Además, se de e mina on y alida on dianas molecula es de RKIP y Pi ina median e co- ans ecciones
y de e minaciones molecula es.
Resul ados y discusión
El melanoma maligno es una o ma de cánce de piel que es ex emadamen e le al. Pa a ga an iza un
a amien o adecuado y un esul ado exi oso, es esencial un diagnós ico opo uno y p eciso del
melanoma maligno. En es e sen ido, las al e aciones molecula es en la pa ogenia del melanoma son
obje o de una in es igación muy ac i a, lo que ha lle ado a la iden i icación de oncogenes y genes
sup eso es de umo es asociados a es a en e medad pa a desa olla en oques e apéu icos que se
necesi an con u gencia. Po ello, el p ime obje i o de es e abajo ha sido es udia el alo po encial de
RKIP y Pi ina como ma cado es de melanoma, lo cual se analiza en el Capí ulo 1 de la sección de
Resul ados a a és de un es udio inmunohis oquímico en una coho e de 314 pacien es (75 ne us y 239
melanoma).
Respec o a RKIP, de mane a gene al, la exp ession in amues a de es a p o eina ue homogénea,
obse ándose un ma caje ci oplasmá ico. El análisis uni a ian e de la compa ación po g upos, mos ó
una di e encia es adís icamen e signi ica i a en e la al a exp esión que mos aban las biopsias de ne i
(94% de los casos) en e a los melanomas (51% de los casos). Además, median e un análisis de
eg esión logís ica (en el que se incluye on la edad y el sexo como co a iables) se obse ó una
asociación lineal, es deci , mayo es ni eles de p o eína se co elacionaban signi ica i amen e con una
mayo p obabilidad de que las biopsias uesen iden i icadas como ne us. Po o o lado, an o el análisis
uni a ian e como el mul i a ian e con i ma on una di e encia signi ica i a en e la exp esión de RKIP
en e las biopsies de ne i y las de melanoma diagnos icados en es adios emp anos (es adios I y II, según
AJCC 8ª Ed.). Al analiza su po encial como ma cado p onós ico, aunque no se obse a on di e encias
signi ica i as en e la exp ession de RKIP y el desa ollo de me as asis, al os ni eles de RKIP se
co elaciona on con un g oso de B eslow más bajo en mues as de odos los es adios de melanoma.
Según la bibliog a ía exis en e, a ios es udios han demos ado que los ni eles de RKIP son bajos en
una g an a iedad de cánce es y que, además, apenas se exp esa en las me as asis (115,148-150,164).
En el caso del melanoma, en conc e o, se ha obse ado una disminución de RKIP en melanoma u eal
y una baja exp ession en melanomas cu áneos, an o p ima ios como me as ásicos (52, 165). Aunque
es os es udios son in e esan es, se ealiza on con coho es pequeñas de pacien es, y, además,
compa aban los ni ele de exp esión de RKIP en ebiopsias de umo es p ima ios y biopsias omadas en
si ios me as ásicos (52,134,148). Aún así, es os esul ados mues an un cla o silenciamien o de RKIP
en elación a la malignidad en las células umo ales, aunque no in es iga on la posible u ilidad de la
exp esión de RKIPcomo ma cado p onós ico de buenaa o mala e ilución. Según nues os esul ados,
pa ece que la inción de RKIP median e inmunohis oquímica iene u ilidad como ma cado d
diagnós ico pa a pacien es de melanoma.
XXI
De o ma simila se co elacionó la exp esión de Pi ina con los da os clínicos. En p ime luga ,
menciona que la inción in amues a, a di e encia que en el caso de RKIP, ue he e ogénea,
obse ándose células con ma caje sólo nuclea , sólo ci oplasmá ico o ambos. Es e pa on he e ogéneo
no mos ó ninguna elación con el ipo his ológico, el es adio umo al o la p og esión del melanoma. Al
ealiza la compa ación po g upos, se obse ó que el 80% de los ne i mos aban una al a exp esión de
Pi ina en e al 60% de los melanomas. En elación al ipo his ológico, los y las pacien es con melanoma
de ex ension supe cial (MES) mos a on una exp esión de Pi ina simila con independencia de su
e olución. Sin emba go, en el caso del melanoma nodula (MN), quienes e en ualmen e desa olla on
en e medad me as ásica mos a on ni eles de pi ina más al os en sus biopsias p ima ias en una
p opo ción signi ica i amen e mayo que aquellos que pe manecie on lib es de en e medad. Po o o
lado, se e aluó la u ilidad de Pi ina como ma cado p onós ico emp ano, u ilizando únicamen e las
biopsias de pacien es con melanoma en es adio emp ano (es adios I y II, según AJCC 8ª Ed.), sin
emba go, no se obse ó una asociación di ec a en e la exp esión de Pi in y el hecho de pe manece
lib e de en e medad o desa olla me ás asis du an e el seguimien o. Aún así, dado que nues os da os
de melanomas p ima ios emp anos e an he e ogéneos, se pod ía espe a que o os ac o es de iesgo
po enciales pudie an es a enmasca ando la asociación en e la exp esión de Pi in y la p obabilidad de
me ás asis. Po ello, se ealizón un análisis de las di e encias en un escena io mul i a iado que incluía
la exp esión de Pi in como el e ec o de in e és, y la edad, el sexo y la p o undidad de B eslow como
posibles co a ian es. En es e caso, un ni el al o de Pi ina se asoció signi ica i amen e con una mayo
p obabilidad de me ás asis según el modelo de eg esión logís ica. Complemen a iamen e se lle ó a
cabo un análisis de Fac o es de Bayes, que esul ó se posi i o, indicando que la al a exp esión de Pi in
en una biopsia implica que sea 10 eces más p obable que se desa olle me ás asis. Finalmen e, un
análisis de Cox mos ó que pacien es con una exp esión más al a de Pi in enían más del doble de
p obabilidad de desa olla me ás asis emp ana en compa ación con aquellos que exp esaban bajos
ni eles de Pi ina. Según lo desc i o has a la echa en elación a Pi ina, pa ece es a in oluc ada en la
egulación de a ios p ocesos celula es, incluida la inhibición p o eínas quinasa, unciones
an ioxidan es y co ac o ansc ipcional (30,82,116). Además, se ha demos ado que Pi in puede
desempeña un papel en la umo igénesis a a és de su pa icipación en la egulación de la p oli e ación
celula y la p og esión maligna (182). Po o o lado, Lucciulli y colabo ado es desc ibie on una
deslocalización de Pi ina desde el núcleo has a el ci oplasma en un subconjun o de mues as de pacien es
con melanoma cuando las compa a on con las biopsias de ne i. Además, obse a on una co elación
posi i a en e los ni eles ci oplasmá icos de Pi in y la p og esión del melanoma (135). En nues o caso,
hemos podido es ablece que su de e minación median e inmunohis oquímica, jun o con el índice de
B eslow, pod ía se ú il como un indicado de p onós ico ya que pacien es con al os ni eles de exp esión
mos a on meno iempo de supe i encia lib e de en e medad.
En o o o den de cosas, hay que ene en cuen a que la incapacidad pa a comp ende los mecanismos
que subyacen a la me ás asis, la cual conduce a la mayo ía de las mue es elacionadas con el cánce ,
plan ea un p oblema impo an e pa a el desa ollo de mé odos de diagnós ico y p onós ico así como de
e apias e ec i as. Debido a es o, cen amos nues a a ención en el papel de RKIP y Pi ina en la biología
XXII
de los melanoci os no males y malignos. Pa a ello, se comenzó po el silenciamien o de ambas p o eínas
de o ma independien e en melanoci os p ima ios sanos y el pos e io análisis de su ansc ip ome
median e Secuenciación de ARN. Se comple a on los es udios con análisis molecula es y uncionales
en líneas de melanoma.
Los hallazgos elacionados con los cambios ansc ipcionales p oducidos po el silenciamien o de RKIP
en melanoci os nomales se ecogen en el Capí ulo 2 de la sección de Resul ados. Lo p ime o que llamó
nues a a ención ue que es as células mos a on modi icaciones a ni el de exp esión génica asociadas
a la i ma gené ica del cance . Conc e amen e, se obse a on al e aciones de los p ocesos celula es
ín imamen e elacionados con la ans o mación maligna de las células, como el desa ollo y la
di e enciación. Es o i ía en co condancia con la exp esión más al a de RKIP encon ada en melanoci os
di e enciados de lesiones de ne us cuando se compa a con mues as de melanoma en el Capí ulo 1. Más
especí icamen e, se encon ó que más del 70% de los genes exp esados di e encialmen e que se
incluye on en es a sección, desa ollo y di e enciación, e an dianas pu a i as de NANOG, un ac o de
ansc ipción elacionado con la oncalidad o ‘s emness’ (70). Los ensayos de co ans ección de
plásmidos pa a la sob exp esión de RKIP jun o con plásmidos del p omo o de NANOG asociado a GFP
mon a on que la p esencia de RKIP p odujo una disminución de la ac i ación del p omo o NANOG, lo
cual que apun a hacia una elación uncional en e la exp esión de RKIP y NANOG. En es e con ex o,
ambién hemos encon ado que la exp esión de miR-21, una diana aguas abajo de NANOG (78) y
elacionado con la ansición epi elio-mesénquima, ue signi ica i amen e meno en las células que
sob eexp esaban RKIP. Es as mismas células mos a on un aumen o de la capacidad mig a o ia an o
en el es de la he ida como a a és de il os con ma iz de colágeno. En línea con es os esul ados,
Lee e al. (173) no a on una g an can idad de in e e encias en e las ías eguladas po RKIP y aquellas
bajo el con ol de los p incipales ac o es de ansc ipción de allo (es deci , OCT4, KLF4, SOX2 y
NANOG) y p opusie on RKIP como un egulado del es ado de di e enciación de las células. En
conjun o, nues os esul ados sugie en que RKIP egula los es ados di e enciados en las células
melanocí icas a a és del ac o de ansc ipción NANOG.
Po o o lado, al y como se obse a en el Capí ulo 3 de la sección de Resul ados, el ansc ip oma de
los melanoci os con Pi ina silenciada e eló un en iquecimien o de genes in oluc ados en la ansición
G1/S, la o ganización de la ma iz ex acellula , la p oli e ación, la mig ación y di e enciación celula .
Uno de los egulado es del ciclo celula es JARID1B/KDM5B, una his ona desme ilasa especí ica de
Lisina. Aunque las células de melanoma que exp esan JARID1B ep esen an solo una pequeña
p opo ción de las células en las poblaciones de melanoma p ima io y me as ásico (187), nues o
conjun o de da os de RNA-seq y el análisis in silico de en iquecimien o de ac o es de ansc ipción
encon a on que JARID1B pod ía egula más de 100 de los genes di e encialmen e exp esadas as el
silenciamien o de Pi ina en los melanoci os. Así, los expe imen os de co ans ección mos a on una
disminución de la ac i ación del p omo o de JARID1B después de la sob eexp esión de Pi ina, lo que
sugie e una elación uncional en e la exp esión de Pi ina y JARID1B. Además, demos amos que la
sob eexp esión de Pi ina en las dos líneas celula es de melanoma me as ásico es udiadas condujo a una
XXIII
disminución signi ica i a en la exp esión del gen JARID1B y de sus genes diana E2F1 y c-MYC
(81,141). En elación a es o, se obse ó una bajada en la p oli e ación de las líneas de melanoma que
sob eexp esaban Pi ina, lo cual concue da con los bajos ni eles de exp esión de JARID1B, E2F1 y c-
MYC ob enidos en los ensayos de co ans ección. Así, en el con ex o de nues o es udio, hemos podido
de e mina que la capacidad p oli e a i a de las células de melanoma depende de la in e acción de Pi ina
con JARID1B, quien es á in oluc ada en la o ganogénesis, la unción de las células mad e y el
desa ollo del cance (80,81).
Con odo ello, man enemos la Tesis de que en la génesis del melanoma cu áneo pueden es a implicados
mul iples mecanismos celula es. En nues o s udio conc e o, la exp ession de RKIP y Pi ina dibujan
escena ios dis in os en un ne us en e al melanoma. La exp ession de RKIP en los melanoci os de los
ne i inhibi ía la exp ession de NANOG y sus díanas molecula es, man eniendo el es ado di e enciado
de los melanoci os. Además, la exp ession de Pi ina modula ia la p oli e ación a a és de JARID1B y
sus dianas molecula es como E2F1. Po su pa e, en el melanoma, la ausencia de RKIP induci ía un
pano ama di e en e, ya que la exp ession de NANOG a o ece ía la adquisición del eno ipo in asi o y
en es e con ex o, las células umo ales con ciclo len o inducido po JARID1B pod ía ac ua como un
es ímulo pa a en a de nue o en ciclo y a o ece la o mación de me as asis.

XXIV
Conclusiones
En nues a hipó esis inicial plan eábamos que RKIP y Pi ina e an p o eínas que desempeñaban un papel
en la e iopa ogenia del melanoma cu áneo, lo que las con e ía en excelen es bioma cado es pa a el
diagnós ico y p onós ico del melanoma cu áneo. Pa a demos a es o, in es igamos su aplicación
po encial como ma cado es de melanoma y su papel en las células melanocí icas.
En base a los esul ados p esen ados en es a esis, se pueden ex ae las siguien es conclusiones:
La de ección inmunohis oquímica de RKIP en biopsias de melanoma puede se una
he amien a ú il pa a el diagnós ico de melanoma.
La de ección inmunohis oquímica de Pi ina jun o con el índice de B eslow pod ía
usa se como ma cado p onós ico en es adios emp anos (I-II) del melanoma.
La baja exp esión de RKIP en melanoci os humanos condujo a una i ma
ansc ipcional asociada con el cánce , que incluía una des egulación de genes
elacionados con la pigmen ación y los p ocesos de desa ollo y di e enciación.
RKIP pa ece es a in oluc ado en el man enimien o del es ado de di e enciación de los
melanoci os al egula nega i amen e el ac o de ansc ipción NANOG y sus dianas
molecula es, como miR-21.
La baja exp esión de de Pi ina en melanoci os humanos condujo a un pe il
ansc ipcional ca ac e izado po una des egulación en la o ganización de la ma iz
ex acelula , la mig ación, la p oli e ación y la espues a a in e e ón ipo II.
Pi ina eje ce un e ec o an ip oli e a i o en las células de melanoma a a és de la
egulación del ac o de ansc ipción JARID1B y sus genes diana, incluidos E2F1 y c-
MYC.
A pa i de es os esul ados, en su conjun o, man enemos la Tesis de la implicación de ambas
p o eínas RKIP y Pi ina en la umo igénesis y p og esión maligna del melanoma cu áneo y pod ían
se la base del diseño de nue as es a egias e apéu icas pa a el a amien o de la en e medad
me as ásica del melanoma.
XXV
XXVI
XXVII
Abs ac
Melanoma is an ex emely le hal skin cance ha a ises as a esul o he malignan ans o ma ion o
melanocy es. The incidence o his pa hology wo ldwide has inc eased in he las 30 yea s in g ea e
p opo ion han he es o he cance , be ween 4-6% yea ly, and shows subs an ial dispa i ies be ween
popula ions. To ensu e app op ia e ea men and a success ul ou come, a imely and accu a e diagnosis
and p ognosis o malignan melanoma is essen ial. Because o his, molecula al e a ions in he
pa hogenesis o melanoma a e he subjec o mo e ac i e esea ch. In e ms o his ology, melanoma
show a wide a ie y o cha ac e is ics including epi helial, hema ological, mesenchymal, and neu al
ea u es, which can o en make he diagnosis o he disease challenging. As new bioma ke s candida es,
in his hesis we examined he exp ession o RKIP (Ra Kinase Inhibi o P o ein) and Pi in. RKIP has
been ex ensi ely epo ed as an inhibi o o key signaling pa hways in ol ed in he agg essi e umo
pheno ype and shows dec eased exp ession in se e al ypes o cance , and Pi in o iginally was
conside ed o ac as a ansc ip ional co- ac o , bu i has ecen ly been epo ed o play a ole in
umo igenesis and he malignan p og ession o many umo s. Howe e , hese s udies we e pe o med
wi h a small coho o pa ien s, so a la ge s udy is equi ed o u he e alua ion o his ma ke 's
diagnos ic o p ognos ic alue. In his con ex , his doc o al hesis’ goals we e o e alua e he po en ial
alue o RKIP and Pi in as melanoma ma ke s and hei implica ion on melanocy ic cell biology.
Rega ding RKIP, immunohis ochemis y analysis e ealed a signi ican ly highe exp ession o RKIP in
ne i compa ed wi h ea ly-s age (s age I-II, AJCC 8 h) melanoma biopsies. P oli e a ion, wound healing,
and collagen-coa ed answell assays unco e ed he implica ion o RKIP on he mo ili y bu no on he
p oli e a i e capaci y o melanoma cells as RKIP p o ein le els we e in e sely co ela ed wi h he
mig a ion capaci y o bo h p ima y and me as a ic melanoma cells bu did no al e o he pa ame e s.
As shown by RNA sequencing, endogenous RKIP knockdown in p ima y melanocy es igge ed he
de egula ion o cellula di e en ia ion- ela ed p ocesses, including genes (i.e., ZEB1, THY-1) closely
ela ed o he EMT. In e es ingly, NANOG was iden i ied as a pu a i e ansc ip ional egula o o many
o he de egula ed genes, and RKIP was able o dec ease he ac i a ion o he NANOG p omo e . In
ela ion o Pi in, he immunohis ochemis y mul i a ia e analysis e ealed ha ea ly melanoma wi h
s onge Pi in exp ession we e mo e han wice as likely o de elop me as ases du ing he ollow-up.
On he o he hand, ansc ip ome analysis o PIR down egula ed melanocy es showed a dampening o
genes in ol ed in he G1/S ansi ion, cell p oli e a ion, and cell mig a ion. In addi ion, an in silico
app oach p edic ed ha JARID1B as a po en ial ansc ip ional egula o ha lies be ween PIR and i s
downs eam modula ed genes, which was co obo a ed by co- ans ec ion expe imen s and unc ional
analysis.
To summa ize, he esul s ob ained in his hesis suppo he diagnos ic
u ili y o RKIP s aining due o he signi ican ly lowe RKIP p o ein le els
in melanoma samples, e en a ea ly s ages (I–II) o he disease, and he
use o Pi in s aining along wi h he B eslow index as a p ognos ic ma ke
a ea ly s ages (I-II) o melanoma. Mo eo e , we p opose ha RKIP could
play a ole in he main enance o he di e en ia ion s a e by nega i ely
egula ing NANOG gene exp ession and, ha Pi in could play an
impo an ole in modula ing he p oli e a i e s a e o melanoma cells by
egula ing JARID1B gene exp ession.
6

7
8
E iological ac o s o malignan melanoma
Fac o s con ibu ing o he de elopmen o cu aneous melanoma can be g ouped in o hose speci ic o
he indi idual (e.g., skin pho o ype, gene ic p edisposi ion) and hose speci ic o he en i onmen .
In e ms o g ading melanoma isks, ul a iole adia ion would be among he mos signi ican . In
addi ion o sunligh , ul a iole ligh can also be p oduced by a i icial ligh ing sys ems, such as anning
beds (18). The wa eleng hs o his ligh ange om 200 o 400nm, wi h he UVB wa eleng h being he
mos ca cinogenic o he skin (19-20). By abso bing hese wa es, he melanocy es a e comp omised in
hei abili y o epai DNA (21-22). Ca cinogenesis is consequen ly caused by he accumula ion o
gene ic al e a ions (23-24).
Ou abili y o an is one o ou na u al de enses agains sola adia ion damages, such as sunbu n. E e y
indi idual has a unique abili y o adjus o he sun, which is de e mined by hei skin pho o ype. Figu e
4 illus a es he six ypes o skin pho o ypes.
As a esul o hei ligh skin, blond o ed hai , eckles, blue eyes, and a ligh complexion, people wi h
ligh skin pho o ypes a e a g ea e isk o de eloping melanoma. (25,26). Addi ionally, i may be
impo an o conside he age a which anning began and he du a ion o he anning beha io (27)
because p olonged adia ion exposu e can cause skin bu ns ha cause he gene ic changes lis ed in he
p e ious pa ag aphs o accumula e o e ime. Acco dingly, he e is a di ec co ela ion be ween
sunbu ns and melanoma de elopmen (25, 28).
9
As desc ibed in he epidemiology sec ion, mos pheno ypic skin cha ac e is ics and sun exposu e
in ensi y a e geog aphically de e mined, implying ha he incidence o melanoma is also dependen
on geog aphical loca ion (Figu e 2). Acco dingly, he highes incidence a es ha e been epo ed in
Aus alia and New Zealand, ollowed by No he n Eu ope (29).
Addi ionally, people wi h pho o ypes I and II end o ha e pigmen ed ne i han hose wi h o he
pho o ypes. Pigmen ed ne i a e skin lesions cha ac e ized by he benign p oli e a ion o melanocy es.
App oxima ely 15% o melanoma de i es om his lesion (30). The e o e, i a pe son has mul iple
pigmen ed ne i ha a y in colo and shape, hey a e a a highe isk o melanoma.
I is c ucial o conside an indi idual's pe sonal his o y in his con ex . I is essen ial o e alua e an
indi idual's unique his o y in his con ex (31). Melanoma is one o he mos immunogenic cance , wi h
spon aneous emission occu ing in oughly 10% o 35% o cases (32). Consequen ly,
immunode iciency has also been shown o be associa ed wi h an inc eased isk o de eloping cance
(33). Addi ionally, almos 15% o melanoma occu s in pa ien s wi h a amily his o y, and a subg oup
o hese pa ien s ha e ge mline mu a ions associa ed wi h melanoma p edisposi ion (34-35). Despi e
his, he magni ude o his gene ic ac o is s ill unclea , which is why i is ecommended ha people
wi h a amily his o y main ain a ca e ul examina ion o hei skin and mo e equen consul a ions wi h
de ma ologis s (36).
Exposu e o ca cinogenic subs ances is ano he isk ac o . A s udy has indica ed ha he bicides ha e
been associa ed wi h a g ea e likelihood o de eloping ac al melanoma (palms o he hands and ee ).
10
This s udy e ealed a highe incidence o ac al melanoma in people who ha e used hese he bicides a
home is highe han in people who do no use hem (37).
His opa hology o melanoma
Melanoma de elopmen
Malignan melanoma de elops om melanocy es, melanin-p oducing cells. In addi ion o g owing in
he skin (95% o cases), melanoma can also de elop in u ea and mucous memb anes due o i s
emb yonic o igin a he neu al c es .
Skin
Among all he o gans o he body, he skin is he la ges . As a ba ie agains physical, chemical, and
biological agen s, i p o ides he i s le el o p o ec ion, p e en s wa e loss, and egula es body
empe a u e. The human skin comp ises h ee laye s: he epide mis, de mis, and subcu aneous issue
(38) (Figu e 5a).
In wha has come o be known as he epide mis, he ou e mos laye is a s a i ied epi helium made up
mainly o ke a inocy es ( ep esen ing 95% o he o al cells in his laye ), which mig a es p og essi ely
om he epide mal basemen memb ane o he skin su ace and o ms se e al well-de ined laye s along
he way. In he ollowing o de , he basal laye appea s ollowed by he spinous s a um, he g anula
s a um, and inally, he s a um co neum (Figu e 5b).
11
Ano he cell ype in his laye is Lange hans cells, which unc ion as pa o he immune sys em, looking
o an igens wi hin hei su oundings o ac i a e an immune esponse, o Me kel cells, which a e closely
associa ed wi h e minal ilamen s o cu aneous ne es and a e esponsible o sensa ion (39-40).
Besides he epide mis, melanocy es a e also p esen in ha basal laye , ha ing a a io o one melanocy e
pe en basal cells. As p e iously s a ed, he melanocy es p oduce he pigmen melanin, which is
packaged in o cellula esicles known as melanosomes, and deli e ed in o he cy oplasm o he
ke a inocy es (40) o p o ec he body om he ha m ul e ec s o ul a iole adia ion.
A laye o issue called he de mis lies benea h he epide mis. Blood essels, ne es, glands, and hai
ollicles a e ound wi hin his hick laye o connec i e issue, which is igh ly a ached o he epide mis
by he basemen laye . I is connec ed o he unde lying o gans, such as bones o muscles, ia he
subcu aneous laye o hypode mis, which consis s o loose connec i e issue ha con ains a iable
amoun s o adipose cells (cells ha s o e a ), depending upon he a ea o he body and i s nu i ional
equi emen s.
Melanoma skin cance
Acco ding o he abo e desc ip ion, skin melanoma a ises om he malignan ans o ma ion o
melanocy es in he epide mis. Among he molecula al e a ions, hese cells unde go a e mu a ions o
genes esponsible o egula ing he cell cycle, di e en ia ion, adhesion, signaling, and apop osis. I
in ol es acqui ing di e se pheno ypic ea u es by no mal melanocy es ha lead o he de elopmen o
a malignan pheno ype. In he Cla k model, one o he mos popula , he signi icance o he
his opa hological changes ela ed o he p og ession o melanoma is highligh ed (Figu e 6). The i s
pheno ypic shi in melanocy es is he de elopmen o benign ne i. Despi e dis up ing he con ol o
g ow h in hese cells, he g ow h o a ne us is limi ed, and a ely p og ess o cance (41). Oncogenes-
induced cell senescence may be esponsible o he absence o p og ession, in which oncogenes
s imula e cell g ow h. F om a molecula pe spec i e, excessi e ac i a ion o he mi ogen-ac i a ed
p o ein kinase (MAPK) signaling pa hway p omo es he de elopmen o melanoma cells. Ac i a ion o
his pa hway is caused by mu a ions in NRAS, which accoun s o app oxima ely 15 pe cen o
melanoma, o mu a ions in BRAF, which accoun o abou 50 pe cen o melanoma (42).

12
Cy ologic a ypia, which may de elop om p eexis ing benign lesions o as a new umo , seems o be
he nex s ep owa ds melanoma ollowing he Cla k model. This s age o he disease p og ession
in ol es molecula abno mali ies ha a ec cell g ow h, DNA epai , and suscep ibili y o cell dea h.
Nex is he adial-g ow h phase, which e e s o umo s ha g ow la e ally along wi h he epide mis and
do no me as a ic sp ead. The condi ion can las o yea s, and cance can be su gically emo ed wi h a
high success a e. Mani es a ions o in asi e beha io in Cla k's model occu in he e ical-g ow h
phase when melanoma cells pene a e he basemen memb ane and g ow in ade mally as expanding
nodules. E en hough his p og ession model is iewed as a linea s epwise p ocess, many melanoma
umo s may no adhe e o i in an o de ly manne . Fo example, RGP and VGP melanoma may esul
om exis ing ne i lesions o de elop spon aneously om no mal melanocy es (43).
In bo h cases, cance sp eads because o he uncon olled g ow h o malignan umo cells due o gene ic
mu a ions ha cause neoplas ic ans o ma ion and enable hem o escape inhibi o y signals. Se e al
hallma ks o cance cha ac e ize his p ocess.
13
I has been shown ha a comp ehensi e se o molecula pa hways is in ol ed in he ini ia ion,
p oli e a ion, su i al, p og ession, and in asion o a umo . In his way, MAPK, PI3K, and NFkB
signaling pa hways in e connec signi ican ly du ing melanomagenesis (Figu e 7).
Essen ially, s imula ion o G p o ein-coupled ecep o s (GPCR) esul s in he ac i a ion o PKC p o ein.
As a esul , ac i a ed PKC s imula es he MAPK pa hway. Addi ionally, he ecep o y osine kinases
(RTKs), ac i a ed h ough he binding o ex acellula g ow h ac o ligands, also media e he ac i a ion
o RAS p o ein, he op membe o he MAPK cascade. Simul aneously, RTK u ns on he PI3K pa h.
PI3K may also be ac i a ed by GPCRs, IGF-1Rs, and RAS. B ie ly, bo h MAPK and PI3K/AKT
pa hways media e cell su i al and p oli e a ion. As pa o he TNF-alpha pa hway (canonical NFkB
pa hway), binding o he cy okine o i s ecep o TNFR1 esul s in ac i a ion o TAK1. TAK1 p omo es
he agg ega ion o a downs eam kinase complex, IKK. In esponse o he phospho yla ion o IkB by
he IKK complex, NFkB is eleased. As a esul , his elemen is ansloca ed o he nucleus, leading o
he ac i a ion o genes in ol ed in cell su i al and an i-apop osis.
MAPK Pa hway
I is well known ha mi ogen-ac i a ed p o ein kinase (MAPK) is an impo an signal ansduc ion
pa hway o a ious physiological p ocesses, including cell p oli e a ion, di e en ia ion, de elopmen ,
mig a ion, apop osis, and ans o ma ion (45). I is pa icula ly ele an o he de elopmen o
melanoma. In a no mal si ua ion, he MAPK cascade is egula ed by sca olding and egula o y p o eins.
One o ha egula o s is Ra Kinase Inhibi o P o ein (RKIP), also known as phospha idyle hanolamine
binding p o ein 1 (PEBP1) (46). Ra -1 and MEK a e equi ed o he phospho yla ion o MEK and he
subsequen phospho yla ion cascade, and RKIP dis up ed he physical in e ac ion o Ra -1 and MEK
p o eins, educing he pa h ac i a ion (46). Two mechanisms de egula e he pa hway: he gain-o -
14
unc ion mu a ions, which make he RAS and RAF p o eins consis en ly p esen in he cell in hei
ac i a ed s a e i espec i e o ex e nal s imuli, and he ine ec i eness o na u al inhibi o s, such as
RKIP.
Conce ning he i s ques ion, among he mu a ions mos likely o be ound in skin cance is he
BRAFV600 mu a ion o he BRAF gene (47). In addi ion, mu a ions a ec he KRAS p o ein (mu an
KRASQ61) and he neu o ib omin 1 (NF1) p o ein. All o hese mu a ions esul in ac i a ion o he
MAPK pa hway and, consequen ly, in he p oli e a ion and su i al o cells (48).
Rega ding he second one, i has been desc ibed a loss o RKIP exp ession in melanoma. Published
da a epo he ole o RKIP in egula ing cell p oli e a ion, mig a ion, and in asion capabili y (49-
51), by mechanisms leading o Ras-ERK1/2 and NFκB pa hways inhibi ion. Ne e heless, he e is
li le known abou he pa hways egula ed by RKIP in no mal melanocy es no i s ole in he
malignan ans o ma ion o his ype o cell. Some s udies (51,52) desc ibe a g adual educ ion o
malignancy- ela ed RKIP le els in melanoma pa ien s, bu hese we e pe o med wi h a small coho
o pa ien s, so i would equi e a mo e ex ensi e s udy o es ablish he eal diagnos ic o p ognos ic
alue o his ma ke in melanoma.
PI3K-AKT Pa hway
The phospha idylinosi ol-3-kinases (PI3Ks) a e a amily o lipid kinases in ol ed in many cellula
p ocesses, including cell su i al and g ow h, di e en ia ion, p oli e a ion, ansc ip ion, and
ansla ion. Besides ansducing signals om g ow h ac o s and cy okines, he pa hway is a majo
downs eam e ec o o RTKs and G-p o ein-coupled ecep o s (GPCRs) (Figu e 9). An essen ial
componen o his signaling pa hway is AKT, which ansmi s signals by phospho yla ing di e en
downs eam e ec o a ge s, in luencing c i ical cellula p ocesses, such as apop osis, DNA epai , cell
cycle, glucose me abolism, cell g ow h, and mo ili y, in asion, and angiogenesis (53).
The e a e nega i e egula o s o p e en pe sis en and long- e m ac i a ion o PI3K-AKT signaling.
PTEN is a cen al egula o o his signaling pa hway. When PTEN is missing, AKT is cons i u i ely
ac i a ed, esul ing in malignan melanoma umo de elopmen (54,55).
NFkB Pa hway
The nuclea ac o -kappaB (NFκB) is a pleio opic ansc ip ion ac o ha egula es se e al genes
in ol ed in many c i ical pa hways ha allow o a wide ange o physiological unc ions such as
immune esponses, in lamma o y esponses, de elopmen , and cance ini ia ion and p og ession (56).
The NFkB amily comp ises i e membe s iden i iable by hei conse ed Rel homology domains, he
pa o he p o eins ha con ol DNA binding: RelA (p65), RelB, c-Rel, p100/p52, and p105/p50.
15
Regula ion o NFkB amily membe s occu s p ima ily h ough binding hem by IkB p o eins (57).
Among he membe s o he IkB amily is he p o o-oncop o ein Bcl-3, which is p edominan ly loca ed
in he nucleus. I has been shown ha Bcl-3 and Pi in, a membe o he cupin supe amily, o m a s able
complex ha enhances Bcl3-p50’s abili y o bind DNA, inducing ansc ip ion o a ious a ge genes
in ol ed in cell su i al and an iapop osis (58).
The e ec s o ul a iole i adia ion on skin cells include p omo ing in lamma o y esponses and
cy okines, many o which ha e NFkB as a downs eam a ge o e ec o . I he p esence o NFkB is
sus ained, i may esul in he augmen a ion o p o-in lamma o y media o s, which may damage issues,
leading o o gan dys unc ion and e en ually cance (59). The e a e di e en mechanisms h ough which
he ac i a ion o NFkB may occu due o ups eam de egula ion o MAPK and PI3K-AKT signaling
pa hways. In melanoma cells, hese changes inc ease p oli e a ion and esis ance o apop osis (60).
Melanoma he e ogenei y and plas ici y
Malignan melanoma is one o he mos equen ly mu a ed cance . Howe e , as desc ibed in he
p eceding sec ion, d i e mu a ions in genes such as BRAF o NRAS occu ea ly in he cou se o
melanoma de elopmen . In con as , a hie a chical pa e n has no been de ec ed be ween mu a ions
associa ed wi h me as asis, sugges ing ha ansc ip ional p og ams a e in ol ed in melanoma
p og ession (61).
Two p edominan ansc ip ional p og ams ha e been iden i ied in cul u ed melanoma cells based on
gene exp ession analysis: p oli e a i e’ pheno ype o ‘in asi e’ pheno ype. In e es ingly, he wo
pheno ypes a e no de e mined by gene ic mu a ions, and ansc ip ional ac i i y can ep og amme a
pheno ype. The e is a balance be ween bo h pheno ypes egula ed by ansc ip ional mas e egula o s,
which enable cells o be capable o adap i e pheno ype plas ici y (61).
One e idence o adap i e pheno ype plas ici y can be ound in single-cell analyses o melanoma
biopsies, which ha e de ec ed popula ions o cells wi h in asi e o p oli e a i e pheno ypic s a es and
single cells wi h ansi ional pheno ypic o ms. Based on his pheno ype plas ici y hypo hesis, s able
melanoma pheno ypes a e de ined by ansc ip omes ela ing o di e en ia ed melanocy es o slow-
cycling cells wi h neu al-c es s em-cell (NCSC) ma ke s (61). Tumo pe sis ence seems o ely on a
pa icula subse o cells wi h hese acqui ed “s em cell-like” p ope ies (62,63). Like emb yonic s em
cells (ESC), his subpopula ion o umo s em-like cells can g ow in ensi ely and in il a e local issue
(52, 62-69). The sel - enewal capabili y o emb yonic s em cells is egula ed by plu ipo ency- ela ed
ansc ip ion ac o s such as Nanog homeobox (NANOG), POU class 5 homeobox 1 (OCT4), and SRY-
box ansc ip ion ac o 2 (SOX2) (70) which a e also abe an ly exp essed in many malignan human
umo s (71–73).
22
Molecula ea u es o melanocy ic lesions
The mo phological c i e ia o de e mining a ypia a e o en disag eeable and subjec o in e -obse e
a iabili y, pa icula ly in non-con en ional diagnoses lesions. Conside ing hese challenges, he Wo ld
Heal h O ganiza ion (WHO) inco po a ed known molecula pa hways in o i s la es classi ica ion o
melanocy ic umo s, in oducing he concep o "in e media e" lesions (97). This mul idimensional
classi ica ion sugges s ha he con en ional app oach o iden i ying melanocy ic umo s as ei he
benign o malignan may no longe be adequa e. WHO 2018 iden i ies nine ca ego ies o pa hways
leading o melanoma, each d i en by gene ic ac o s (Table 3).
As a esul o hese molecula signa u es' he e ogenei y, wo c i ical implica ions eme ge: i s , i
emphasizes he need o indi idualiza ion o melanoma diagnosis, p ognosis, and ea men ; second, i
o e s a b oad ange o po en ial bioma ke s and no el pu a i e he apeu ic a ge s. In de ma ology, he
ollowing an igens/an ibodies a e commonly used o diagnose melanoma (106,107):
The equency o S100 exp ession in malignan cu aneous melanoma is app oxima ely 95%. Se e al
ac o s can in luence i s exp ession, including oo much o oo li le ixa ion ime, p e iously ozen
issue, and enzyma ic p e ea men wi h ypsin. An ibodies can be polyclonal o monoclonal, and bo h
iden i y melanocy osis in a cy oplasmic and nuclea manne . Among he many an igens de ec ed by
an i-S100, A6 is exp essed by some melanocy o ic lesions and may also be help ul when de ec ing
neu o heliomas.
The Melanoma An igen Recognized by T Cells-1 (MART1) is one o he mos impo an melanocy ic
ma ke s. Two di e en an ibodies de ec he an igen (Melan-A and A-103), exp essed by a wide ange
o benign and malignan melanocy ic lesions. Consequen ly, i is o g ea alue in de ec ing melanocy ic
di e en ia ion. On he o he hand, a diagnosis o desmoplas ic melanoma is unlikely i his ma ke is
highly exp essed in a spindle cell melanocy ic lesion.

23
Table 3. Classi ica ion o melanoma based on 2018 wo ld heal h o ganiza ion classi ica ion (97)
UV exposu e
Ca ego ies
Melanoma
sub ype
Key molecula genes
Low-CSD
melanoma
Pa hway I
Supe icial
sp eading
melanoma
BRAFV600 mu TERT mu
CDKN2A mu PTEN mu
NRAS mu TP53 mu
High-CSD
melanoma
Pa hway II
Len igo
maligna
melanoma
NRAS mu TP53 mu
KIT mu CDKN2A mu
TERT mu PTEN mu
BRAFnon-V600Emu
Pa hway III
Desmoplas ic
melanoma
NF1 mu NRAS mu
NFKBIE mu PIK3CA mu
Low o no UV
/CSD melanoma
Pa hway IV
Spi z
melanoma
ALK ea CDKN2A mu
NTRK1 ea HRAS mu
NRTK3 ea
Pa hway V
Ac al
melanoma
KIT mu ALK ea
NRTK3 ea CDKN2A mu
CCND1amp TERT mu
NRAS o BRAF mu
Pa hway VI
Mucosal
melanoma
KIT mu SF3B1 mu
CDKN2A mu CCND1 amp
CDK4 mu MDM2 amp
NRAS o BRAF mu
Pa hway VII
Melanoma
a ising in
congeni al
ne i
NRAS mu
BRAFV600E mu
Pa hway VIII
Melanoma
a ising in
blue ne i
GNA11 mu GNAQ mu
CYSLTR2 mu SF3B1 mu
BAP1 mu EIFAX mu
Pa hway IX
U eal
melanoma
GNA11 mu GNAQ mu
CYSLTR2 mu PLCB4 mu
BAP1 mu EIFAX mu
SF3B1 mu
May occu in any o mos o he
pa hways
Nodula
melanoma
Abb e ia ions: amp, ampli ica ion| CSD, cumula i e sun damage | mu , mu a ion |
ea , ea angemen
The an i-HMB45 an ibody de ec s he p emelanosome p o ein (PMEL o gp-100) in melanoma and
junc ional ne i. The exp ession o gp100 in p ima y cu aneous melanoma di e s om ha o ne i since
i is o en dis ibu ed in pa chy pa e ns h oughou he de mis. This pa e n is also ound in ne oid
melanoma. I is especially use ul in de ec ing he pa e n o ne i ma u a ion. Acco dingly, supe icial
ype Amelanocy es (epi helioid cells loca ed wi hin he epi helium o close o he epi helium and
p edominan ly pigmen ed) exp ess neu onal ma ke s and gp100. In con as , deeply loca ed ype C
melanocy es (spindle cells) exp ess schwannian ma ke s. Howe e , i has been p o en o be insensi i e
o desmoplas ic malignan melanoma.
P oli e a ion ma ke s such as MIB1 (de ec ed by an i-Ki67 an ibody) a e exp essed in p oli e a ing
cells. Simila o gp100, i s pa e n o exp ession indica es whe he o no a issue has ma u ed. A small
pe cen age o cells in bo h common and dysplas ic ne i exhibi eac i i y, which is ypically loca ed a
he de malepide mal junc ion o in mo e supe icial de mis compa men s. Melanoma, on he o he
24
hand, display a andom pa e n o immuno eac i i y and ypically ha e a p oli e a i e ac ion o o e
10%, pa icula ly nea he edge o he lesion.
The enzyme y osinase (TYR) pa icipa es in melanogenesis and he e o e is ela i ely speci ic o he
di e en ia ion o melanocy ic cells. Howe e , i also exhibi s a lowe sensi i i y o de ec desmoplas ic
melanoma.
In he manne desc ibed, he issue ma ke s S100, MART-1, and gp100/HMB45 can be used o
di e en ia e melanoma om o he ypes o cance , bu none o hese ma ke s can accu a ely dis inguish
non-malignan melano ic lesions om malignan melanoma, no a e hey able o s a i y melanoma
pa ien s by hei isk o p og ession.
25
26
27
Hypo hesis and objec i es
Backg ound
Al hough cu aneous melanoma accoun o less han 5% o all skin cance , hey a e he deadlies due o
hei p opensi y o me as asize and a lack o e ec i e ea men a ad anced s ages. Cu en ly, he
diagnosis and p ognosis o malignan melanoma a e mainly based on he e alua ion o biopsies o skin
lesions emo ed om pa ien s. Clinicians de e mine a melanoma pa ien 's p ognosis based mainly on
he B eslow and he p esence o ulce a ion and sen inel nodes (108). I has been epo ed ha 10% o
melanoma ecu ences wi hin i e yea s o ollow-up ha e occu ed in he ea ly s ages (I and II
acco ding o he 8 h edi ion o AJCC) (109). In his con ex , i has been ecen ly ound in a s udy
in ol ing 784 pa ien s ha 53.8% o pa ien s wi h me as a ic melanoma had an ini ial s age o I-II (110).
The e o e, i would appea ha con en ional isk ma ke s a e no de ec ing many ea ly melanoma wi h
he po en ial o me as asize.
The mos c i ical aspec o educing melanoma dea hs is iden i ying ma ke s ha can be used o ea ly
s a i ica ion o melanoma, pa icula ly hose wi h a wo se cou se. Rela ed o he la e g oup, he e is a
need o unde s and he biology o hin melanoma ha is ul ima ely le hal.
In p e ious wo k by ou esea ch g oup, he di e en ial p o eomic analysis ha included melanoma and
p ima y melanocy e lines highligh ed se e al no el candida e ma ke s o cu aneous melanoma (111).
In he p incipal componen analysis, whe e mRNAs and p o eins o RKIP and PIR we e analyzed
showed he highes signi icance in e ms o exp ession be ween heal hy p ima y melanocy es (HEMn-
MP, HEMn-DP, and HEMm-LP) and melanoma cells (JSG, HT144, Hs-294 , A375, Mel-HO,
WM298B, 1205lu, Mel-Juso, RMPI7951, and Colo-800) (Figu e 10).

28
In i s o iginal desc ip ion, RKIP was an inhibi o o MAPK o ERK1/2 pa hways. Howe e , la e s udies
e ealed i s addi ional ole as a egula o o o he signaling cascades, such as GPCR, GSK3b, and NF-
kB (46,112). I unc ions as a signaling swi ch in essen ial p ocesses such as di e en ia ion,
p oli e a ion, and su i al o cells. The de egula ion o his p o ein has been associa ed wi h a wide
a ie y o diseases, including cance (113-115).
As a as PIR/Pi in is conce ned, i was ini ially desc ibed as a ubiqui ously exp essed nuclea p o ein
wi h a pu a i e unc ion as a ansc ip ional co ac o (116). O e ime, se e al s udies ha e e ealed ha
his p o ein plays an essen ial ole in p ocesses such as cell cycle egula ion (117-119), in lamma o y
esponse (120,121), mig a ion egula ion, and epi helial-mesenchymal ansi ion egula ion (122-126).
Based on ou esul s quan i ying mRNA le els by RT-PCR, RKIP exp ession was gene ally educed in
melanoma cell lines compa ed o p ima y no mal melanocy es (Figu e 11a). We ound ha he p o ein
le el o RKIP was also consis en ly educed in melanoma cell lines wi h no change in p ima y cell lines
(A375, Colo-800, WM793B, Mel-Ho, and Mel-JUSO) o me as a ic cell lines (1205Lu, A2058, Hs294 ,
HT-144, MeWO, and RMPI7951) (Figu e 11b).
A lowe exp ession o he PIR gene was also obse ed in melanoma cell lines compa ed o melanocy es
(Figu e 11c). Howe e , compa ing melanoma cell lines, i was ound ha PIR mRNA le els we e mo e
he e ogeneous han in he RKIP s udy. As a esul , we obse ed a much lowe amoun o Pi in p o ein
in all melanoma cell lines han in melanocy es (Figu e 11d). Consequen ly, hese ma ke s we e licensed
unde a Eu opean Pa en (No. EP3051291. Me hod o Diagnosis and p ognosis o cu aneous melanoma).
29
Hypo hesis
Acco ding o ou p e ious esul s, we main ain he hypo hesis ha RKIP and Pi in a e p o eins ha
could play a ole in he e iopa hogenesis o cu aneous melanoma, making hem excellen bioma ke s
o diagnosis and p ognosis o cu aneous melanoma.
30
Objec i es
To achie e he hypo hesis, we se he ollowing speci ic objec i es:
31
38
he p oli e a i e a e and cell iabili y and ch omosomal and me abolic al e a ions, which can gene a e
dis u bances in he analyses ca ied ou . The e o e, cell cul u es we e es ed o mycoplasma e e y h ee
mon hs o elimina e any con amina ed cul u es. The comme cial Veno ® GeM One-S ep Tes (11-8025,
Mine a Biolabs, USA) was used, which de ec s en mycoplasma species by PCR. Fo he analysis, he
supe na an o cul u e media ha has been in cul u e o a leas 48 hou s was used.
Modula ion o gene exp ession
Gene exp ession analysis s udies how genes a e ansc ibed o syn hesize unc ional gene p oduc s:
RNA o p o eins. The s udy o he egula ion o hese pa hways p o ides in o ma ion on no mal cellula
p ocesses, such as cell di e en ia ion, o pa hological p ocesses, such as umo igenesis.
Silencing by ansduc ion wi h len i i al pa icles
Len i i al ec o s a e a high ansduc ion e icacy and sa e me hod o gene deli e y in o ha d- o-
ans e cells, such as p ima y melanocy es. Mo eo e , len i i al pa icles can be employed in s anda d
Biosa e y Le el 2 issue cul u e acili ies, as hey a e eplica ion-incompe en s.
No mal p ima y melanocy e cell line HEMn-
LP was ansduced wi h len i i al pa icles o
gene silencing ollowing he manu ac u e ’s
ins uc ion wi h mino modi ica ions (Figu e
13a). B ie ly, 24 hou s be o e i al in ec ion,
cells we e seeded in a 6-well pla e. In ou
case, we did no use polyb ene® because i
was oxic o ou p ima y cells. Polyb ene® is
a polyca ion ha neu alizes cha ge
in e ac ions o inc ease he binding be ween
he pseudo i al capsid and he cellula
memb ane.
The a io o he numbe o ansducing len i i al pa icles o he numbe o cells (Mul iplici y o
In ec ion, MOI) used in hese expe imen s was o 2 MOI o len i i al pa icles, di ec ly added o each
well and we e incuba ed o e nigh . Speci ically, he shRNA speci ied in Table 6 was used.The day
a e , he medium wi h he len i i al pa icles was eplaced, and a esh medium was added. Two days
a e ha , he cells we e selec ed wi h 5 µg/mL o Pu omicine (P8833, Sigma-Ald ich Quimica, S.A.,
Spain) o ge s able cell lines.
Table 6. Len i i al pa icles speci ic ions
Gene
Re e ence
Supplie
T ansduc ion
e iciency
con ol
copGFP Con ol
Len i i al
Pa icles
(sc-108084)
San a C uz
Bio echnology
Inc., USA
T ansduc ion
sc amble
con ol
Con ol shRNA
Len i i al
Pa icles
(sc-108080)
San a C uz
Bio echnology
Inc., USA
RKIP | PEBP1
PEBP1 shRNA
Len i i al
Pa icles
(sc-36430-V)
San a C uz
Bio echnology
Inc., USA
PIR
PIR shRNA
Len i i al
Pa icles
(sc-61359-V)
San a C uz
Bio echnology
Inc., USA

39
40
O e exp ession by an ec ion wi h plasmid
T ans ec ion plasmid is widely used o achie e e icien gene ans e in easy- o- ans e cells, such as
melanoma cells. Due o ha , A2058 and MeWO me as a ic melanoma cell lines we e ans ec ed wi h
o e exp essing plasmid o RKIP o PIR using Lipo ec amine 2000 (11668019, The mo Fishe
Scien i ic, USA) acco ding o he manu ac u e ’s ins uc ions (Figu e 13b).
B ie ly, a e seeding cells we e 70–90%
con luen he ans ec ion eagen mix was added.
This mix included Op i-MEM® Medium,
Lipo ec amine 2000 and plasmid o in e es . The
Table 7 includes he in o ma ion o he plasmid
used in his s udy. All o he ans ec ion
expe imen s we e pe o med wi h 500 ng o each
plasmid. The mix was incuba ed du ing 15-20
min and hen was ans e ed o he seeded cells.
All he expe imen al assays we e pe o med a
leas a e 24 hou s o ans ec ion and a sc amble
plasmid was used as con ol.
Func ional assays
P oli e a ion assays
Cell p oli e a ion can be used o assess no mal cell heal h, o measu e esponses o oxic insul , o as a
p ognos ic and diagnos ic ool in se e al cance . The cell iabili y was de e mined using he
s anda dized XTT ki (Roche Molecula Biochemicals, USA). This assay is based on he abili y o iable
cells o ans o m he yellow XTT e azolium sal in o o ange o mazan in he p esence o a educing
eagen . The p ocess occu s only in iable, me abolically ac i e cells. The amoun o o mazan and he
o ange dye o med is easily quan i iable using a pla e spec opho ome e by measu ing he abso bance
a 490 nm.
Fo his assay, cells we e seeded in la -bo omed 96-well pla es wi h he app op ia e densi y acco ding
o he cell line in 100 µL o cul u e medium and le o e nigh in he o en o adhe e o he well. A e
adding he educing eagen and XTT mix u e o he wells in a 1:50 a io, he pla e was incuba ed o 4
hou s in he incuba o . The abso bances we e ead a 490 nm u ilizing a pla e spec opho ome e BioTek
Table 7. Plasmids speci ica ions
Gene
Re e ence
Supplie
sc amble
pCMV-myc-
DKK
O iGene
Technologies,
Inc, USA
RKIP
PEBP1
pCMV-PEBP1-
myc-DDK
O iGene
Technologies,
Inc, USA
PIR
pCMV-PIR-
myc-DKK
O iGene
Technologies,
Inc, USA
Nanog
p omo e
PL-SIN-
Nanog-EGFP
#21321
Addgene,
USA
Ja id1B
p omo e
EWI024-FP
pLU-
JARID1B1B
p omo e -
EGFP
Ke a as Inc.,
USA
41
Syne gy HTX (Agilen Technologies, Inc, USA). Cell iabili y was calcula ed as he pe cen age o cell
iabili y wi h espec o con ol cells as ollows: (sample abso bance/con ol abso bance) * 100.
Mig a ion assays
Cell mig a ion occu s du ing c i ical physiological p ocesses and is dys egula ed in pa hological
si ua ions, such as cance me as asis and in lamma ion. Fo mig a ion capaci y assays (Figu e 14), a e
seeding he cells in 24-well pla es, he monolaye s we e incuba ed wi h 0.5 µg/mL o Mi omycin C o
2 hou s. Then hey we e sc aped wi h a s e ile plas ic mic opipe e ip. The wound closu e was obse ed
o e 48 hou s, and he pho os we e aken a 0, 24, and 48 hou s wi h a ligh mic oscope. Then, each
g oup's mig a ion pe cen age was calcula ed acco ding o he con ol.
Mo eo e , he answell ac i e mig a ion assay was pe o med using Type I-Collagen coa ed inse s
wi h 6.5-mm-diame e polyca bona e il e s (8-μm po e size). Cells (1 × 104) suspended in 200 μL o
DMEM wi hou FBS we e seeded in he op chambe s. The bo om chambe s we e illed wi h 300 μl o
DMEM con aining 10% FBS (as a chemoa ac an ). Cells we e allowed o mig a e o e nigh . The non-
mig a ed cells on he uppe su ace o he il e we e ca e ully and ho oughly emo ed wi h co on
swabs. Mig a ed cells we e ixed wi h a mix o cold 4% pa a o maldehyde plus 2% e hanol and s ained
wi h c ys al iole . Fi e images pe inse we e aken using a compound op ical mic oscope, and
mig a ed cells we e quan i ied by ImageJ so wa e. Resul s we e exp essed as he a e age numbe o
mig a ed cells pe well ob ained om h ee sepa a e expe imen s done in iplica e.
42
Da a analysis
In all quan i a i e es s pe o med (p oli e a ion and mig a ion assays), he SPSS e sion 26 p og am
was used. Fishe exac es was applied o check he no mali y o he da a. To compa e di e ences
be ween he wo g oups, S uden 's - es was pe o med. P<0.05 was conside ed s a is ically signi ican ,
P<0.01 was conside ed highly s a is ically signi ican , and P>0.05 was conside ed s a is ically
insigni ican .
Molecula analysis
P o ein analysis
P o ein ex ac ion
Melanoma cells and p ima y melanocy es we e ha es ed by ypsiniza ion, washed wi h PBS and lysed
in RIPA lysis bu e (80 mM T is-HCl pH 8, 150 mM NaCl, 1% NP 40, 0.5% sodium deoxychola e,
0.1% SDS) con aining P o ease Inhibi o Cock ail (Sigma-Ald ich Quimica S.A., Spain) o 15 minu es
on ice. Lysa es we e hen clea ed by cen i uga ion a 1500 xg o 5 minu es. To al p o ein concen a ion
was de e mined using he bicinchoninic acid assay.
P o ein quan i ica ion
P o ein quan i ica ion was pe o med by p epa ing a s aigh s anda d wi h inc easing known
concen a ions o Bo ine Se um Albumin (BSA). To p epa e he samples, each eppendo was
adequa ely named and mixed 5 L o each sample wi h 45 L o Milli-Q H2O. Then, knowing ha he
ela ion be ween bicinchoninic acid and cup ic sul a e had o be 50:1, 1mL o his mix was pou ed in o
each eppendo , and all samples we e incuba ed a 37oC in da kness o abou 30 minu es. Finally, he
abso bance was measu ed a 562 nm in a spec opho ome e .
P o ein de ec ion by wes e n blo ing
Wes e n blo is a widely used analy ical echnique o he s udy o p o eins. This me hod allows he
de ec ion o a single p o ein wi hin a biological sample, speci ically wi h an an ibody ha ecognizes a
unique epi ope on he p o ein o in e es . P o ein de ec ion using his echnique in ol es a i s s ep in
sepa a ing he p o eins based on hei molecula weigh , ans e ing hem o a memb ane, and
subsequen ly labeling hem wi h he an ibody o in e es .
43
SDS polypolyac ylamide gel elec opho esis
The SDS polypolyac ylamide gel elec opho esis echnique (SDS-PAGE) allows he sepa a ion o
dena u ed p o eins on gels wi h a polypolyac ylamide ma ix. The gels we e p epa ed by polyme iza ion
o polyac ylamide, aking in o accoun he pe cen age o his polyme concen a ion. The highe he
pe cen age o polyac ylamide, he smalle he po e size and he be e he esolu ion o low molecula
mass p o eins. On he con a y, i a p o ein wi h high molecula weigh is o be de ec ed, he pe cen age
o polyac ylamide mus be lowe ed o ob ain la ge po es ha allow a be e esolu ion o la ge p o eins.
In his s udy, 12% o polyac ylamide concen a ion was used.
Table 8. Pe cen age o polypolyac ylamide used depending on he p o ein o in e es
% polyac ylamide
15%
10%
7.5%
5%
KDa ange
12-43
16-68
36-94
57-212
The gels a e composed o wo di e en pa s (Table 9):
 S acking gel: I has la ge po es ha allow he p o eins o mig a e eely and ge s acked a he
in e ace be ween s acking and unning gel. The pu pose is ha p o eins s a mig a ing a he
same ime.
 Running gel: The esol ing pa o he gel, in which p o eins un acco ding o hei molecula
weigh .

44
Table 9. Composi ion o he s acking and unning gel
S acking gel
Running gel
4%
7.5%
10%
12%
15%
30% Ac y-bis
1 mL
30% Ac y-bi
3.8 mL
5 mL
6 mL
7.5 mL
0.5M T is-HCL
(pH6.8)
0.75 mL
2M T is-HCL
(pH8.8)
3 mL
3 mL
3 mL
3 mL
H2O miliQ
5.75 mL
H2O miliQ
9 mL
7 mL
6 mL
4.5 mL
20% SDS
37.5 mL
20%SDS
75 mL
75 mL
75 mL
75 mL
10%APS
62.5 mL
10%APS
75 mL
75 mL
75 mL
75 mL
TEMED
6.25 mL
TEMED
5 mL
5 mL
5 mL
5 mL
E e y sample con ained 2 µL o Di hio h ei ol (DTT, which b eaks down hyd ogen bonds) and 4 µL o
Laemmli 1X loading bu e (60 mM T is-HCl pH 6.8, 2% p/ SDS, 10% glyce ol, 0.002% blue
b omophenol, 1mM DTT and bidis illed wa e up o 20µL). Then, samples we e boiled a 95oC o 5
minu es and loaded wi h 7 µL o molecula weigh ladde P ecision Plus P o einTM Dual Colo s
S anda ds o BioRad and 20 µL o each sample. The assembly was co e ed wi h he elec opho esis
bu e (Table 10). Then, elec opho esis was a 100 V o 10 minu es; and hen a 180 V o a ound 45
minu es.
Table 10. Composi ion o 1L o he elec opho esis bu e (5X)
T is ( is(hyd oxyme hyl)aminome hane)
15.15 g
Glycine
72.1 g
Sodium dodecyl sulpha e
5 g
Milli-Q H2O
Up o 1 L
T ans e ence
This s ep in ol es ans e ing he p o eins om he gel o a ni ocellulose memb ane (Wha man GmbH-
GE Heal hca e, Dassel, Ge many) by applying ol age. Fo ha , an elec oblo ing casse e con aining
a small pillow, wo hin il e s, he gel wi h p o eins, he ni ocellulose memb ane, and ano he wo
il e s plus a small pad was assembled and placed on he elec odes in he blo ing uni wi h ans e
bu e (Table 11). The ans e ence condi ions we e 3 hou s a 300 mA. To e i y a success ul ans e ,
he ni ocellulose memb ane was incuba ed o 2 minu es in Ponceau ed solu ion. This s ain binds
e e sibly o he posi i ely cha ged unc ional g oups o he p o ein (amino g oup) and he non‐pola
egions. A e checking he ans e ence, he memb ane was washed ex ensi ely in wa e un il he dye
was gone.
45
Table 11. Composi ion o 1L o he ans e bu e
T is ( is(hyd oxyme hyl)aminome hane)
5.8 g
Glycine
29 g
Sodium dodecyl sulpha e
1 g
Me anol
200 mL
Milli-Q H2O
Up o 1 L
Immunos aining
The blo s we e incuba ed wi h PBS con aining 5% Bo ine Se um Albumin and 0.1% Tween-20 o 1
hou o block nonspeci ic binding and hen incuba ed wi h an app op ia e dilu ion o p ima y an ibody
a 4oC o e nigh (Table 12). The memb ane was washed wi h TBST h ee imes (10 min/each ime),
hen incuba ed wi h goa an i-mouse Ho se adish Pe oxidase (HRP) conjuga ed seconda y an ibody o
2 hou s a oom empe a u e. Finally, p o eins we e isualized by enhanced chemiluminescence using
he Supe Signal® Wes Pico Chemiluminescen Subs a e (The mo Fishe Scien i ic, USA). The
ollowing able includes he an ibody de ails used in his s udy.
Table 12. De ails o sou ces and concen a ions o an ibodies used o wes e n blo in his s udy
An ibody name
Dilu ion
Re e ence
Recombinan An i-RKIP an ibody
1:5000
Abcam, UK ab76582
An i-Pi in abbi polyclonal an ibody
1:1000
The mo Fishe Scien i c, USA
PA5-29777
An i-gamma Tubulin an ibody
1:1000
Abcam, UK ab 11321
Goa F(ab')2 An i-Mouse IgG(H+L),
Human ads-HRP
1:8000
Sou he nBio ech, USA 1032-05
Goa an i-Rabbi IgG (H+L) Seconda y
An ibody, HRP
1: 10000
Abcam, UK ab6721
Gene exp ession analysis
RNA ex ac ion and quan i ica ion
To al RNA om cul u ed cells was isola ed using he RNeasy Mini ki (Qiagen Inc, Ge many). The
en i e p ocess was ca ied ou on he ice and wi h s e ile ma e ial. A e collec ing he cells, he numbe
o cells was coun ed, and 1 x 107 cells we e sepa a ed since RNA ex ac ion was pe o med om 1 x
107 cells. They we e cen i uged a 300 xg o 5 minu es, and he PBS was emo ed. The pelle was
esuspended wi h 600 µL o a solu ion consis ing o 1 ml o RLT bu e and 10 µL o -
me cap oe hanol. The sample was passed en imes h ough a sy inge wi h a 20G needle o comple e
and homogeneous cell lysis. Then, he 70% e hanol addi ion s ep is ollowed by he ans e o 700 µL
o he sample o an RNeasy sepa a ion column. A e cen i uging i a 1500 xg o 15 seconds, he
elua e was disca ded, and he column was placed in he same collec ing ube. This s ep was epea ed as
o en as necessa y un il he en i e sample was collec ed. Then 700 µL o RW1 bu e was added o he
46
column o clean i , and i was cen i uged a 1500 xg o 15 seconds. The elua e was again disca ded,
and he ea men wi h DNAse was ca ied ou , o which 80 µL o a solu ion con aining 10 µL o
DNAse and 70 µL o RDD bu e we e added. I was incuba ed o 15 minu es a oom empe a u e (20-
30oC), and hen 700 µL o bu e RW1 was added and cen i uged o 15 seconds a 1500 xg. The elua e
was disca ded, and he column was ans e ed o a new collec ing ube. Two washes we e hen ca ied
ou wi h 500 µL o RPE bu e each, cen i uging a 1500 xg o 15 seconds in he i s wash and 2
minu es in he second. Finally, he column was ans e ed o a 1.5 mL Eppendo ube, and he RNA
was elu ed by adding 50 µL o DEPC wa e (Ambion Inc., USA) by cen i uga ion a 1500 xg o 1
minu e. Samples we e co ec ly labeled, quan i ied by NanoD op (The mo Fishe Scien i ic, USA), and
s o ed a -80oC un il use o ensu e RNA in eg i y.
Real Time quan i a i e Polime ase Chain Reac ion (RTqPCR)
Fo each sample, cDNA was syn hesized om 1 µg o al RNA using he iSc ip TM cDNA Syn hesis
ki (Bio-Rad, USA) acco ding o he manu ac u e ’s ins uc ion. The eac ion mix u e con ained 0.1 µL
cDNA om he e e se ansc ip ion eac ion, oge he wi h o wa d and e e se speci ic p ime s and
iQTM SYBR® G een Supe mix (Bio-Rad, USA) in a inal eac ion olume o 20 µL. Quan i a i e eal-
ime RT-PCR assays we e ca ied ou using an iCycle PCR pla o m (Bio-Rad, USA). The PCR
eac ion began wi h hea ing a 95oC o 10 min, ollowed by 45 cycles o desna u a ion a 95oC o 30
sec, annealing a he co esponding empe a u e o each gene (56-61oC) o 20 sec and ex ension a
72oC o 30 sec. Each assay included a nega i e con ol consis ing o he absence o cDNA. Exp ession
da a we e gene a ed om 2 ampli ica ion eac ions wi h samples and con ols un in iplica e. Op ical
da a ob ained by eal- ime PCR we e analyzed using he MyiQ Single-Colo Real-Time PCR De ec ion
Sys em So wa e .1.0 (Bio-Rad, USA). The exp ession o h ee di e en housekeeping genes (ACTB,
GAPDH, and RPS15) also was analyzed o no malize exp ession da a using he Gene Exp ession Mac o
So wa e Ve sion 1.1 (Bio-Rad Labo a o ies, He cules, CA, USA), whe e he ela i e exp ession alues
we e compu ed by he compa a i e C me hod (132,133). The sequences o p ime s used a e speci ied
in he ollowing Table 13.
47
Table 13. RTqPCR p ime s’ sequences
Gene name
Fo wa d p ime
Re e se p ime
ACTB
5'-AGATGACCCAGATCATGTTTGAG-3'
5'-GTCACCGGAGTCCATCACG-3'
c-MYC
5'-GCTCCTGGCAAAAGGTCAG-3'
5'-GTTGTGCTGATGTGTGGAGAC-3'
E2F1
5'-TGACATCACCAACGTCCTTGA-3'
5'-CTGTCGGAGGTCCTGGGTC-3'
GAPDH
5'-CCTGTTCGACAGTCAGCCG-3'
5'-CGACCAAATCCGTTGACTCC-3'
JARID1B
5'-GACTGGGACAACAGAACCT-3'
5'-TGGACTAACACCATGGAGG-3'
LUM
5'-AACTGCCCTGAAAGCTACCC-3'
5'-AGCCACTGCAGATCAGTTACA-3'
NTRK2
5′-CTCCCGGAATTGGGTTGGAG-3′
5′-GGGGCGCAGATTCCTTGTTA-3′
PIR
5'-GGAGCCTCAGTACCAGGAACT-3'
5'-CTTGGACTTTATTCCCAGGGC-3'
RKIP
5′-AATAGACCCACCAGCATTTCG-3′
5′-TGCCACTGCTGATGTCATTG-3′
RPS15
5'-CGACCAAATCCGTTGACTCC-3'
5'-CGGGCCGGCCATGCTTTACG-3'
THY-1
5'-GTTTGACCAGGAAAGCAGCG-3'
5'CTCTTGGGAGCTTGGGACAG-3'
ZEB1
5′-GTGCAGTTACACCTTTGCA-3′
5′-CACATGTCTTTGATCTCTTCCT-3′
Has-miR-21-
5p
-
5′-UAGCUUAUCAGACUGAUGUUGA-3′
Has_RNU6-2
-
5′-CGCTTCGGCAGCACATATACTA-3′
Gene exp ession le els esul s we e analysed using he SPSS e sion 26 p og am. Fishe exac es was
applied o check he no mali y o he da a. To compa e di e ences be ween wo g oups, S uden 's - es
was pe o med. P<0.05 was conside ed s a is ically signi ican , P<0.01 was conside ed highly
s a is ically signi ican , and P>0.05 was conside ed s a is ically insigni ican .
Fo de ec ion o ma u e miRNA, cDNA was p epa ed in a e e se ansc ip ion eac ion using miSc ip
HiSpec Bu e om he miSc ip II RT Ki (Qiagen Inc, Hilden, Ge many). Oncogene miR-21 was
de ec ed by RT-qPCR. The p ime s used we e Has-miR-21-5p (MS00009079, Qiagen Inc, Hilden,
Ge many) and Has-RNU6-2 (MS00033740, Qiagen Inc, Hilden, Ge many) as e e ence ma u e
miRNA. The RT-qPCR assay was conduc ed unde he ollowing condi ions: S age 1: 15 min a 95 °C;
S age 2: 60 cycles o 15 s a 94 °C, 30 s a 55 °C, 30 s a 70 °C, 1 s a 72; and S age 3: 5 s a 95 °C, 1
min a 65 °C. The eal- ime luo escence in ensi y was moni o ed a each cycle o he hi d s age. Ligh
Cycle ® 480 II Real-Time PCR Sys em (Roche, Basilea, Swi ze land) was used o pe o m he eac ion.
RNA sequencing analysis
Second gene a ion sequencing (Nex Gene a ion Sequencing o NGS) is used o analyze he p esence
o genes as well as he quan i ica ion o hei exp ession globally h oughou he en i e ansc ip ome.
The comple e p o ocol included an RNA ex ac ion s ep, ollowed by i s quan i ica ion as well as i s
pu i y and in eg i y analysis. Subsequen ly, lib a ies we e p epa ed and bioin o ma ic analysis was
ca ied ou a e sequencing.
54

55
56
57
58
59
Chap e 01
Po en ial alue o RKIP and Pi in p o eins as
melanoma diagnos ic and p ognos ic ma ke s
Cha ac e is ics o he pa ien s en olled in he s udy
Du ing his p ojec , a o al o 314 pa ien s (154 emales and 160 males) we e en olled who had been
diagnosed by pa hologis s wi h a ne us o malignan melanoma. The clinicopa hological ea u es o he
g oup a e shown in Table 19. Ou collec ion included 75 ne i and 239 melanoma. Rega dless o he
diagnosis, he incidence o ne i and melanoma was equal among men and women (Figu e 19a). Pa ien s
diagnosed wi h ne us anged in age om 24 o 78 yea s old, wi h he median age being 50 yea s old,
while hose diagnosed wi h melanoma anged in age om 23 o 87 yea s old, wi h he median age being
57 yea s old (Table 19).
Table 19. Clinical and pa hological da a om ne us and melanoma pa ien s
N (%)
N (%)
NEVUS
75
HISTOLOGICAL SUBTYPE
Age a diagnosis (yea s,
ange)
Sex
- Male
- Female
56 (24-78)
28 (37)
47 (63)
SSM
NM
ALM
LMM
LM
O he s
ND
102 (43)
53(22)
21(9)
9(4)
3 (1)
12 (5)
39 (16)
MELANOMA
239
AJCC STAGES AT DIAGNOSIS
Age a diagnosis (yea s)
Sex
- Male
- Female
57 (23-87)
131 (55)
108 (45)
In si u
IA
IB
IIA
IIB
IIC
IIIA
IIIB
IIIC
IV
34 (14)
46 (19)
54 (23)
32 (13)
15 (6)
24 (10)
9 (4)
11 (5)
6 (3)
8 (3)
LOCALIZATION
DISEASE EVOLUTION
Head and neck
T unk
Uppe limb
Lowe limb
Ac al
O he s
ND
43(18)
74(31)
24 (01)
69 (29)
21 (9)
5 (2)
3 (1)
Disease- ee
Me as asis
147 (62)
92 (38)

60
Analyzing da a acco ding o he body loca ion o he umo s, he subg oup o women had a highe
incidence o umo s associa ed wi h he lowe ex emi ies. In con as , he subse o men had mo e
melanoma a ibu ed o he head and neck and he unk (Figu e 19b).
The e we e also collec ed da a conce ning he sub ype o melanoma. As shown in Figu e 19c, he
supe icial sp eading melanoma p o ed o be he mos equen sub ype ac oss bo h sexes (43%),
ollowed by he nodula melanoma, which accoun ed o 22% o all cases. The emaining samples we e
diagnosed as len igo maligna (1%), len igo malignan melanoma (5%), and ac al melanoma (9%). Unde
he ca ego y "o he s" (12%), low- equency melanoma, such as hose o mucosal o igin, we e g ouped.
61
A no able aspec o his s udy is how many pa ien s included in i we e diagnosed a an ea ly s age o
hei disease (86%) acco ding o AJCC's s aging sys em (8 h edi ion). As shown in Figu e 19d,
melanoma we e ound in 14% o cases as in si u melanoma, 42% as s age I melanoma, and 30% as s age
II melanoma. Pa ien s diagnosed in s ages III and IV cons i u ed 14% o he o al.
In connec ion wi h ha , he samples acco ding o he B eslow index, which is used o de e mine he
hickness o umo s, was 41% o umo s ha had a hickness o less han a millime e and 30% o
umo s wi h a hickness be ween 1 millime e and 4 millime e s (Figu e 19e).
Fu he mo e, 62% o all melanoma cases analyzed in his s udy emained disease- ee. In compa ison,
38% o pa ien s we e diagnosed a an ad anced s age o de eloped me as asis du ing ollow-up ( he
inclusion c i e ia o his g oup equi ed a minimum acking pe iod o wo yea s). This s udy did no
obse e a di e ence in sex- ela ed ou comes, wi h he same numbe o men and women in he wo
g oups (Figu e 19 ).
Di e en ial exp ession o RKIP p o ein be ween ne i and
melanoma biopsies
Two independen e iewe s e alua ed RKIP s aining a nega i e, low, and high exp ession. A high
p opo ion o ne us samples exp essed RKIP, whe eas a la ge numbe o melanoma samples los his
p o ein exp ession (Figu e 20a).
Among hose wi h melanoma, mos samples analyzed co espond o he his ological sub ypes o
supe icial sp eading melanoma and nodula melanoma. As shown in Figu e 20b, he pa e n o RKIP
exp ession does no di e be ween he g oups based on his ology and malignan p og ession. When
analyzing he dis ibu ion based on he s age a diagnosis, mo e samples a e ma ked as nega i e in
pa ien s wi h s ages I and II who ha e me as asized, as well as in hose wi h s ages III and IV (Figu e
20c).
62
Rep esen a i e IHC images o mos common lesions (compound ne us, supe icial sp eading
melanoma, and nodula melanoma) a e shown in Figu e 21a, while he s a is ical analyses a e
summa ized in Figu e 21b–e. Ne i samples exhibi ed highe posi i i y o RKIP s aining compa ed wi h
he whole coho o melanoma samples (Figu e 21b); 94% o ne i samples we e posi i e o RKIP
whe eas only 51% o melanoma cases p esen ed posi i e s aining. In e es ingly enough, in si u
melanoma, cha ac e ized by an excellen p ognosis upon su gical emo al, exhibi ed a s ong posi i e
RKIP exp ession in almos 80% o cases. O no e, RKIP s aining displayed minimum in asample
a ia ion and a cy oplasmic localiza ion. Uni a ia e analysis con i med he s a is ical signi icance o
obse ed di e ences among ne i and he en i e se o melanoma samples (q < 0.001) (Figu e 21b) and
bo h, uni a ia e and mul i a ia e analysis, p o ided s a is ical e idence o a di e en le el o
exp ession o RKIP in ne i and melanoma a ea ly s ages (AJCC 8 h I/II) (Figu e 21c). Mo eo e , by
means o a logis ic eg ession analysis (Ne us = 0, Melanoma = 1) o con ol o age and sex as
co a ia es (Figu e 21c), a polynomial con as expansion in RKIP demons a es ha his associa ion is
linea (β = −2.288, q < 0.001), i.e., linea inc emen s in p o ein le els co ela e signi ican ly wi h a
la ge p obabili y o he biopsies o being iden i ied as ne us, while quad a ic e ec s end o be
mode a e and non-signi ican (β = 0.465, q = 0.218).
63
70

71
Chap e 02
RKIP egula es melanocy e
di e en ia ion by modula ing he
s emness- ela ed ansc ip ion ac o
As a signaling swi ch, RKIP plays a ole in essen ial p ocesses such as di e en ia ion, p oli e a ion, and
cell su i al. Mo eo e , de egula ion o RKIP exp ession has been associa ed wi h a wide a ie y o
diso de s such as neu ological diseases, diabe es, al e ed spe miogenesis and cance (113-115). Wi h a
gene al loss o RKIP exp ession in umo issues and i s demons a ed abili y o in luence pa hways
leading o umo cell p oli e a ion o an in asi e pheno ype (65-67, 134), a mo e comp ehensi e
analysis o RKIP-dependen pa hways on p ima y cells is necessa y o de e mine p o ile al e a ions ha
may in luence cellula ans o ma ion o p omo e an agg essi e pheno ype. Thus, in his wo k, we
examined he ole o RKIP in he biology o p ima y melanocy es and malignan melanoma cells.
In ol emen o RKIP on malignancy- ela ed p ope ies o
melanoma cells
To s udy he in ol emen o RKIP in he pa hogenesis o melanoma, we modi ied he endogenous RKIP
le els exp ession on A375 and MelHo p ima y melanoma cell lines and A2058 and MeWo me as a ic
melanoma cell lines. Down egula ion o endogenous RKIP was accomplished by RKIP shRNA
len i i al pa icles, while RKIP-o e exp essing plasmids we e used o inc ease cellula RKIP le els.
Down egula ion by shRNA led o a dec ease o up o he 70-80% on he endogenous RKIP mRNA le el
on selec ed p ima y melanoma cells lines (Figu e 26a) which was also consis en wi h a educ ion on
he p o ein pe cen age (Figu e 26b). Reduc ion o endogenous RKIP by len i i al silencing did no al e
p oli e a ion capabili y o A375 and MelHO cells (Figu e 26c). By con as , he RKIP-down egula ed
p ima y melanoma cells showed a signi ican ly inc ease in mo ili y, assessed bo h by wound healing
and collagen-coa ed answell assays (Figu e 26d-e).
72
To ein o ce ou da a ega ding he in ol emen o RKIP exp ession in melanoma cell mo ili y, MeWO
and A2058 me as a ic melanoma cell lines we e ans ec ed wi h a RKIP-o e exp essing plasmid
esul ing in a 5 and 15- old inc ease o RKIP-mRNA le el, espec i ely (Figu e 27a). In addi ion, we
de ec ed a concomi an ele a ion o in acellula RKIP-p o ein pe cen age (Figu e 27b).
Consis en wi h ou p e ious esul s in p ima y melanoma, he inc ease on cellula RKIP exp ession
le el led o a dec ease in he mig a ion capabili y o melanoma cells (Figu e 27c). Su p isingly, bo h
73
analyzed cell lines showed di e en beha io on he ac i e mig a ion assay (Figu e 27d); hus, while no
di e ences we e obse ed in MeWO cells, RKIP o e exp ession clea ly diminished he capaci y o
A2058 cells o pass h ough a collagen-based ba ie . O no e, basal collagen- h ough mig a ion ac i i y
o MeWO cells was signi ican ly lowe han ha o A2058 cells (da a no shown).
B ie ly, cellula RKIP le els we e in e sely co ela ed o he mig a ion capabili y o bo h, p ima y and
me as a ic melanoma cell lines, while no majo e ec was de ec ed on cellula p oli e a ion.
74
T ansc ip ome modula ion by RKIP down egula ion in HEMn-
LP cells
Wi h he aim o elucida ing he molecula mechanisms whe eby RKIP could modula e p ocesses ela ed
o cellula malignancy, RKIP was down egula ed in p ima y melanocy es (HEMn-LP) by he abo e
desc ibed shRNA len i i al pa icles. In ec ion esul ed in a 70-80% educ ion o RKIP mRNA and 40%
o p o ein le el (Figu e 28).
Two independen eplica es o con ol (shCTR) and RKIP knockdown (shRKIP) HEMn-LP samples
we e subjec ed o RNA sequencing. The i s pa o he analysis ocused on he iden i ica ion o a se
o di e en ially exp essed genes be ween shCTR and shRKIP HEMn-LP based on s anda d h eshold
Log2FC≥1, p- alue≤ 0.05 and False Disco e y Ra e (FDR) ≤0.05. The esul ing 224 di e en ially
exp essed genes we e used o moni o ing he unc ions and pa hways ha we e mainly a ec ed in
melanocy es due o he dec eased RKIP exp ession. The se o genes wi h modi ied exp ession we e
oughly equally di ided in o o e - (113) and unde -exp essed genes (111). The Log2FC, p- alues and
FDR o each gene a e de ailed in Annex B: Table S1.
In o de o gain insigh in o he unc ional cha ac e is ics o de ec ed changes, o e - and unde -exp essed
genes a e RKIP silencing we e subjec ed o pa hway (Kyo o Encyclopedia o Genes and Genomes,
KEGG) (Figu e 29a) and Gene On ology (GO) (Figu e 29b) en ichmen analyses. RKIP knockdown on
HEMn-LP cells displayed a ansc ip ional mis egula ion in he GO e m ‘cance gene signa u e’ (p-
alue < 0.001).
75
Mo eo e , he se o genes wi h al e ed exp ession upon endogenous RKIP educ ion showed an
en ichmen in a a ie y o essen ial p ocesses including de elopmen al pigmen a ion, p oli e a ion and
de elopmen al and cell di e en ia ion (p- alue < 0.05; Figu e 29b, Annex C: Table S3). In e es ingly,
RKIP knockdown led o he down egula ion o essen ial melanocy e-pigmen a ion genes such as PMEL
(Melanocy ic linage-speci ic an igen, 2- old dec ease, p alue 0.0003, FDR 0.04), MLANA (Melanoma
An igen ecognized by T-cells, 8- old dec ease, p alue 0.001, FDR 0.02), GPR143 (G-P o ein Coupled
Recep o 143,11- old dec ease, p alue 10-5, FDR 0.01) and TYRP1 (Ty osinase‐ ela ed p o ein 1, 5-
old dec ease, p alue 10-6, FDR 0.007). On he o he hand, only KIT (p o o-oncogene KIT,) was
up egula ed among he de egula ed genes belonging o he de elopmen al pigmen a ion g oup (2.3- old
inc ease, p alue 0.0001, FDR 0.02).

76
De elopmen and di e en ia ion showed he bes FDR alue among signi ican ly en iched biological
p ocesses. This signa u e encompassed 83 genes which ep esen ed 37% o he o al al e ed gene-se
and included HOX amily membe s, p o o-oncogene KIT, p o o-oncogene MYC, ZEB1 and Thy-1 cell
su ace an igen (THY-1), among o he s (Annex C: Table S3). We ocused on genes belonging o his
p ocess due o he s a is ical obus ness o his g oup on ou da a se as well as o he in ima e link among
his pa icula p ocess and he cellula mig a ion-capabili y. As shown in Figu e 29c-d, down egula ion
o endogenous RKIP led o an inc ease on he exp ession o selec ed genes, alida ing he RNA Seq
da a.
In e es ingly, neu o ophic ecep o y osine kinase 2 (NTRK2), ZEB1 and THY-1 a e no only
implica ed in de elopmen al p ocesses, as hey also known egula o s o cellula mig a ion. Based on
ou p e ious esul s ha implica ed RKIP on he mig a ion capabili y o melanoma cells, we made use
o RKIP o e exp ession o analyze he e ec on ZEB1, NTRK2 and THY-1 ansc ip ion. RKIP-d i en
ansc ip ional ep ession was con i med by RT-qPCR o ZEB1 and THY-1 in bo h cell lines (A2058
and MeWO) while NTRK2 e ealed a cell ype-dependen esponse (Figu e 29d). Taking oge he , RKIP
e ealed he capaci y o modula e genes in ol ed in essen ial p ocesses (e.g. De elopmen and
di e en ia ion) and o ep ess genes wi h desc ibed oles in cellula mig a ion.
NANOG as a pu a i e ansc ip ion ac o egula ed by RKIP
RKIP has no desc ibed unc ion as a di ec ansc ip ional egula o . Thus, obse ed ansc ip ional
al e a ions imply he p esence o ye unknown ansc ip ion ac o s o egula o s downs eam RKIP. We
ocused on de egula ed genes belonging o de elopmen and di e en ia ion and conduc ed an in silico
app oach in o de o de ec po en ial ansc ip ion ac o s ac ing be ween RKIP and i s downs eam
modula ed genes. Se en y-one pe cen o genes in his ca ego y we e pu a i e a ge s o NANOG
ansc ip ion ac o (Figu e 30a, Annex C: Table S4, Table S5). NANOG is a ansc ip ion ac o
in ol ed in he main enance o s emness and o en linked o cance agg essi eness (Figu e 30b).
The e o e, we wonde whe he RKIP could somehow modula e NANOG exp ession. To analyze his
poin , we made use o a cons uc encoding he NANOG p omo e a ached o he Enhanced G een
Fluo escen P o ein (EGFP) coding sequence, and cells we e co ans ec ed wi h ei he emp y plasmid
(pCTR) o RKIP-coding plasmid (pRKIP). Ac i a ion o NANOG p omo e was de e mined as he
pe cen age o cells exp essing EGFP. As shown in Figu e 30c, inc eased RKIP exp ession led o a
signi ican dec ease on NANOG p omo e ac i a ion; a simila e ec was obse ed in bo h cell lines.
The gene miR-21 is a desc ibed a ge o NANOG (78). To u he alida e he implica ion o RKIP in
NANOG egula ion, we de e mined miR-21 ansc ip ion le el upon RKIP o e exp ession. As shown
in Figu e 30d, he exp ession o miR-21 was signi ican ly lowe on RKIP-o e exp essing cells.
77
These indings poin owa ds he in ol emen o NANOG downs eam RKIP in he egula ion o gene
exp ession.
78
79
86
(Figu e 31c) we e in ag eemen wi h his molecula analysis, since bo h c-MYC and E2F1 a e p o eins
in ol ed in cell p oli e a ion and cell cycle egula ion. In ha case, he cells o e exp essing Pi in
showed a dec eased p oli e a ion compa ed o con ol cells. These da a a e consis en wi h he
down egula ion o JARID1B gene exp ession.
Toge he , hese esul s sugges ha PIR migh modula e melanoma p oli e a ion by a ge ing he slow-
cycling ansc ip ional egula o JARID1B.

87
88
89
Discussion
RKIP and Pi in as bioma ke s o melanoma diagnosis and
p ognosis
Melanoma is an ex emely le hal o m o skin cance . A imely and accu a e diagnosis o malignan
melanoma is undamen al o ensu ing app op ia e ea men and a success ul ou come. Howe e ,
his ologically, melanoma exhibi s a wide ange o ea u es, which include epi helial, hema ologic,
mesenchymal, and neu al cha ac e is ics (96). In some cases, his makes he diagnosis o he disease
di icul .
Classically, IHC-media ed ou ine iden i ica ion o melanocy ic lesions include he use o melanocy e
and melanoma ma ke s, like y osinase (TYR) and y osinase- ela ed p o eins (TYRP1 and DCT),
gp100 and Melan-A (142); none heless, he u ili y o a combined immunohis ochemical analysis
including Bcl-2, nuclea S100A4, Ki67 and MITF o imp o e he isk s a i ica ion o ea ly-s age
malignan melanoma pa ien s has been ecen ly epo ed (143).
Due o ha , he molecula al e a ions in ol ed in he pa hogenesis o melanoma ep esen a opic o
ac i e esea ch, which has enabled he iden i ica ion o disease-associa ed key, oncogenes, and umo
supp esso genes p o iding a scien i ic ounda ion o u gen ly needed he apeu ic app oaches
(144,147).
Based on ou p e ious esul s om a compa a i e p o eomic analysis be ween melanocy es and
melanoma cell lines esul s, we selec ed RKIP and Pi in as candida es o melanoma bioma ke s. As a
consequence, in his hesis, we ocused on he analysis o RKIP and Pi in exp ession by
immunohis ochemis y in melanoma pa ien biopsies.
Rega ding RKIP, se e al s udies ha e shown his p o ein exhibi s low exp ession le els in a ious
umo s and i is o en absen in me as ases (115,134, 148,149, 150-164). In ag eemen , dec eased RKIP
exp ession has been associa ed wi h me as a ic u eal melanoma while low le els o RKIP we e de ec ed
on bo h me as a ic as well as non-me as a ic cu aneous melanoma biopsies (52,165). These s udies,
al hough in e es ing, we e ca ied ou wi h small coho s o pa ien s. In addi ion, in hose s udies
claiming he associa ion among low RKIP exp ession and me as asis, dec eased RKIP exp ession was
assessed by compa ison o p ima y umo s and biopsies a me as a ic si es (52,134,148). Resul s
ob ained om he a o emen ioned wo ks e ealed a clea malignancy- ela ed silencing o RKIP on
umo cells, al hough hey did no analyze he possible p edic i e ole o RKIP. Ou s udy, including 75
ne i and 239 samples o malignan melanoma, allowed deepening on he diagnos ic and p ognos ic
90
alue o his p o ein. O no e, all melanoma biopsies we e ob ained om he p ima y lesion, which may
explain he lack o s a is ical RKIP-s aining di e ences among s age I–IV melanoma. Also, he coho
size o s age III and IV melanoma pa ien s was small (when compa ed o s age I–II pa ien s) and we
canno discha ge i s e ec when analyzing all s ages oge he .
In ela ion o Pi in exp ession, i seems o pa icipa e in he egula ion o di e en cellula p ocesses,
ac ing as a p o ein kinase inhibi o , an ioxidan o pu a i e ansc ip ional co- ac o (143,147,165-167).
Some e idence has implica ed Pi in in umo igenesis by p omo ing cell p oli e a ion and malignan
p og ession o se e al cance (168,169). As RKIP, o e alua e he impac o Pi in on melanoma
umo igenesis, we i s analyzed he exp ession o Pi in by immunohis ochemis y in he consecu i es
slides and in he same way as in RKIP s udy. In bo h ne us and melanoma g oups, s ong Pi in
exp ession was obse ed and signi ican di e ences we e de ec ed among benign and malignan lesions,
wi h homogenously s ong Pi in exp ession in benign melanocy es om ne i ela i e o he
he e ogeneous exp ession in malignan melanoma.
Looking o new me as a ic bioma ke s we ocused on ea ly s age melanoma (s age I–II acco ding o
AJCC 8 h edi ion) in o de o e alua e RKIP and Pi in use ulness o disc imina e among pa ien s wi h
good and bad e olu ion o he disease. He e, ou esul s ag ee wi h p e ious s udies on he diagnos ic
capabili y o RKIP s aining, as melanoma samples exhibi ed an o e all dec ease in s aining when
compa ed wi h benign lesions (i.e., ne i). Un o una ely, RKIP s aining was no able o dis inguish s age
I–II pa ien s wi h a a o able e olu ion o he disease om hose who e en ually de eloped me as asis.
Ne e heless, i is wo h men ioning he associa ion among s ong RKIP s aining and lowe B eslow
index ac oss all melanoma s ages (s age I–IV) sugges ing ha RKIP may no de e mine umo
malignancy bu may be ela ed o he p ima y umo posi ion o p og ess h ough he skin. In he case
o Pi in s aining, based on mul i a ia e analyses by Logis ic Reg ession and Cox models, including age
and he B eslow index as co- a ia es, s ong Pi in exp ession was signi ican ly associa ed wi h a isk o
me as asis, sugges ing i s impo ance as a p ognos ic ma ke .
To summa ize, ou s udy suppo s he diagnos ic u ili y o RKIP s aining due o he signi ican ly lowe
RKIP p o ein le els in melanoma samples, e en a ea ly s ages (I–II) o he disease. Addi ionally, Pi in
s aining along wi h he B eslow index seem o be a p ognos ic ma ke a ea ly s ages (I-II) o melanoma,
since high Pi in p o ein le els a e associa ed wi h a mo e signi ican p obabili y o me as asis, as well
as a sho e ime un il his clinical end-poin .
91
RKIP Regula es Di e en ia ion-Rela ed Fea u es in Melanocy ic
Cells
Del ing in o he s udy o RKIP p o ein as ega ds he pa hogenesis o melanoma, we ca ied ou
molecula and unc ional assays using melanocy es and melanoma cell lines. In acco dance wi h ou
his opa hological esul s and p e iously published s udies (51, 152), we ound ha bo h RKIP mRNA
and p o ein exp ession we e signi ican ly lowe in melanoma cell lines han in p ima y cul u es o
melanocy es wi h he excep ion o he Mel-HO cell line; his cell line exhibi ed RKIP mRNA le el
simila o ha obse ed on melanocy es bu a educed p o ein con en ha sugges s he in ol emen o
a pos - ansc ip ional mechanism limi ing ansla ion. O no e, RKIP has been desc ibed as a a ge o
se e al mic oRNAs able o egula e cellula p o ein le el (112).
Se e al au ho s ha e sugges ed ha RKIP may no ha e a signi ican ole in p ima y umo s bu ha
ins ead, his p o ein could play an impo an ole as a me as a ic supp esso (115,134, 148,149, 150-
164). In his sense, and despi e he desc ibed ole o RKIP in he egula ion o he MAPK/ERK
pa hway, RKIP has been implica ed on he in asi e beha io o malignan melanoma cells bu no on
hei p oli e a i e capabili y (51). Mo eo e , Schoen gen and Jonic (68) desc ibed he in ol emen o
RKIP on he co ical ac in o ganiza ion du ing he memb ane changes ha happen du ing umo cell
mig a ion. In ag eemen wi h p e ious s udies, we con i med he implica ion o RKIP on he mo ili y
o malignan melanoma cells as RKIP exp ession was in e sely co ela ed wi h he mig a ion capabili y
o bo h, p ima y and me as a ic melanoma cell lines. Ne e heless, modula ion o he cellula RKIP
le el did no show an in luence on he p oli e a i e ac i i y o melanoma cells. The e o e, conside ing
he ele ance o cellula mo ili y on umo me as asis, hese esul s suppo a ole o RKIP loss in
melanoma dissemina ion.
To de ine he cellula mechanisms egula ed by RKIP ha could explain he selec i e o ce a o ing a
dec eased p esence o his p o ein on melanoma when compa ing wi h benign lesions (i.e., ne i), RKIP
gene was silenced using len i i us in p ima y melanocy es and RNA sequencing we e pe o med o
analyze he ansc ip ome changes de i ed om RKIP modula ion. The ansc ip ome o melanocy es
a e RKIP silencing e ealed a ansc ip ional mis egula ion in cance gene signa u e. Among o he s,
his signa u e included al e ed exp ession pa e n o he
oncogenes KIT, BCL3, MAF, MYC, MYCL, HOXA9, CDC25B, and PIM1. In ou da a, all o hem
showed a wo o i e- old inc ease, suppo ing he ole o RKIP like a umo supp esso gene (170).
In e es ingly enough, down egula ion o RKIP exp ession on melanocy es esul ed in he al e a ion o
cellula p ocesses in ima ely linked o malignan ans o ma ion o cells, such as de elopmen and
di e en ia ion. Mo eo e , de elopmen al pigmen a ion, a p ocess speci ically linked o he melanocy ic

92
lineage, was also en iched. Acco ding o ou RNA-seq da a, RKIP would be a ep esso o KIT and an
induce o TYRP1, MLANA, and PMEL gene exp ession, among o he s. TRYP1, MLANA,
and PMEL a e among he bes -known ansc ip ional a ge s o he mas e melanogenic egula o
mic oph halmia-associa ed ansc ip ion ac o (MITF) (171) and i would be o in e es o u he
analyze he possible c oss alk among RKIP and MITF. In addi ion, da a indica e ha RKIP ep esen s
a b ake o he EMT p ocess, by egula ing he exp ession o genes such us ZEB1, THY-1 and NTRK2.
Scien i ic e idence demons a es ha in a he e ogeneous umo mass, hose cells esponsible o d ug-
esis ance, ecu ence and me as asis con ain cha ac e is ics o s em cells, ha is, he abili y o sel -
enew and di e en ia e in any cell ype o he umo mass (172). In his wo k, a e silencing o RKIP
in HEMn-LP melanocy es, mo e han 70% o he di e en ial exp ession genes belonging o
de elopmen and di e en ia ion we e ound o be pu a i e a ge s o NANOG. NANOG has been
iden i ied as one o he c ucial induce s o his s em cell-like s a e ype (70) and is abe an ly exp essed
in many ypes o umo s (71-73). We obse ed ha ansien o ced-inc ease o RKIP exp ession in
me as a ic melanoma cells led o he dec ease o NANOG p omo e ac i a ion poin ing owa ds a
unc ional ela ionship among RKIP and NANOG exp ession. In line wi h hese esul s, Lee e al. (173)
no iced a high amoun o c oss alks be ween pa hways egula ed by RKIP and hose unde he con ol
o main s emness ansc ip ion ac o s (i.e., OCT4, KLF4, SOX2, and NANOG) and p oposed RKIP as
a egula o o he di e en ia ion s a e o cells. This hypo hesis would be in ag eemen wi h he s onge
RKIP exp ession ound in di e en ia ed melanocy es om ne i lesions when compa ing wi h
melanoma samples.
In addi ion o he main enance o he s emness, NANOG has been also implica ed in he EMT (69,76)
and by egula ing he exp ession o ZEB1 and THY-1 among o he genes (78). In ac , EMT and
de elopmen o s emness p ope ies a e o en closely ela ed p ocesses (174). As p e iously men ioned,
hese wo genes a e among hose wi h de egula ed exp ession in ou RNA-seq s udy. ZEB1 is one o
he majo ac i a o s o he EMT p og am and inc easing e idence places ZEB1 also as an impo an
egula o o di e en ia ion, p oli e a ion, DNA damage esponse and cell su i al (175). In e es ingly,
ZEB1 is among he ansc ip ion ac o s d i ing he ea ly hyb id EMT s a e and hyb id EMT s a es (i.e.,
s a es wi h in e media e cha ac e is ics among ully epi helial and ully mesenchymal cells) ha e been
linked o collec i e cells mig a ion and highes me as a ic po en ial (174). This esul , oge he wi h he
obse ed modula ion o he cellula mig a ion capaci y d i en by RKIP, a e in line wi h he apid RKIP
diminu ion obse ed on malignan lesions, e en a ea ly s ages, as well as he associa ion among low
B eslow index and p esence o RKIP. In ac , capaci y o a umo o deepen on he skin equi es he
acquisi ion o cha ac e is ics as hose blocked by RKIP. On he o he hand, THY-1 is a p o ein
implica ed in he endo helium ans asa ion o melanoma cells du ing me as asis sp eading (139). These
esul s could be indica ing he implica ion o RKIP loss in he plas ici y equi ed o he in a- and
ex a asa ion du ing melanoma me as asis. In his con ex , we ha e also ound ha he exp ession o
miR-21 was signi ican ly lowe on RKIP-o e exp essing cells. miR-21 is a known a ge o NANOG
(78) and an impo an induce o EMT a ec ing mig a ion and in asion capabili y (176-178) sugges ing
93
a possible ole o his onco-miRNA in melanoma malignancy (78,179). These esul s poin owa ds he
in ol emen o NANOG downs eam RKIP in he egula ion o gene exp ession ela ed o malignan
pheno ype o melanoma cells.
To summa ize, we p opose ha RKIP could play a ole in he main enance o he di e en ia ion s a e
by nega i ely egula ing NANOG gene exp ession al hough u he esea ch would be equi ed o a
be e desc ip ion o he unde lying mechanism.
Pi in dampens he p oli e a ion o malignan cells by
down egula ing JARID1B/KDM5B exp ession
The cellula ac i i y o Pi in has mainly been s udied in e ms o ex acellula ma ix (ECM)
umo igenici y (82,125,180) and hus, despi e i s b oad dis ibu ion, he e is li le in o ma ion ega ding
he ole o Pi in in non- ans o med cells and issues (116,181). To be e unde s and he ole o Pi in
in a melanocy ic con ex , we i s ly s udied PIR/Pi in exp ession in p ima y melanocy es and melanoma
cell lines by RT-qPCR and in Wes e n Blo s, demons a ing ha p ima y melanocy es exhibi gene ally
s onge and mo e homogenous Pi in exp ession han melanoma cell lines, which had signi ican ly
lowe exp ession and mo e he e ogenei y among he di e en cell lines analyzed. We s udied
p oli e a ion and mig a ion o me as a ic melanoma cells in which Pi in was o e exp essed and we
ound his up egula ion did no modi y mig a ion bu a he , i did induce a signi ican dec ease in he
p oli e a ion a e o bo h he melanoma cell lines s udied. In his con ex , con o e sial esul s ha e been
ound in di e en umo s. Fo example, in DLD1 colo ec al cance cells Pi in does no a ec iabili y
o mig a ion (126), whe eas knocking down Pi in in b eas cance cells was seen o signi ican ly dampen
in i o p oli e a ion and dec ease xenog aph umo g ow h in mice (182). In melanoma, Pi in has been
ela ed o an inhibi ion o mig a ion (122), and i has been p oposed o be an inhibi o o melanocy e
senescence (135) and a malignan bioma ke (183).
In acco dance wi h he an ip oli e a i e ac i i y obse ed when Pi in is o e exp essed in melanoma
cells, he ansc ip omic analysis ollowing PIR-silencing in p ima y melanocy es he e e ealed an
en ichmen o genes in ol ed in he nega i e egula ion o cell p oli e a ion, he G1/S ansi ion and
ex acellula ma ix o ganiza ion and posi i e egula ion o cell mig a ion (Annex C: Table S6).
Fu he mo e, dele e ious mu a ions in he PIR gene we e ecen ly iden i ied in b eas cance ha could
a ec p o ein s uc u e, s abili y and unc ion (182). These esul s could explain he disc epancies ound
when s udying di e en cance o di e en umo cell lines. On he o he hand, melanoma he e ogenei y
was ecen ly p oposed o be due o he co-exis ence o di e en melanoma cell pheno ypes and adap i e
pheno ype plas ici y gi en ha ansc ip ional ep og amming could d i e melanoma p og ession (184).
94
T ansc ip ional ep og amming has been de ec ed a di e en s ages o melanoma, wi h enhanced
mesenchymal ai s in ci cula ing melanoma cells and p oli e a i e ea u es in me as a ic umo s (184).
Hence, cells wi h di e en pheno ypes may in e ac in a coope a i e manne and con ibu e o success ul
me as a ic p og ession (61,185).
In ecen yea s, a en ion is being paid o epigene ic egula ion in melanoma (85), which led o he
desc ip ion o JARID1B as an epigene ic egula o implica ed in he ansc ip ional ep og amming o
se e al umo cells and in umo he e ogenei y (186). Al hough JARID1B exp essing melanoma cells
ep esen only a small p opo ion o he cells in he p ima y and me as a ic melanoma popula ions (187),
he RNA-seq da ase and he ansc ip ion ac o en ichmen analysis ound ha JARID1B could a ge
mo e han 100 o he DEGs iden i ied. Fu he mo e, co- ans ec ion expe imen s showed a dec ease o
JARID1B p omo e ac i a ion a e Pi in o e exp ession, poin ing o a unc ional ela ionship be ween
Pi in and JARID1B exp ession. In addi ion, we demons a ed ha he o e exp ession o Pi in in bo h
he me as a ic melanoma cell lines s udied led o a signi ican dec ease in JARID1B gene exp ession,
and ha o i s a ge genes E2F1 and c-MYC (81,141). These esul s may explain he an ip oli e a i e
e ec o Pi in obse ed in melanoma cell lines. Indeed, in canine o al melanoma cell lines JARID1-
inhibi o s d i e an i-p oli e a i e ac i i y and o e came cispla in esis ance (188).
F om ou da a, we belie e ha in no mal melanocy es Pi in exp ession could egula e he a e o
p oli e a ion h ough JARID1B and he E2F1 pa hway, al hough he exp ession o o he genes a o s
melanoma umo s acqui ing an in asi e pheno ype h ough he exp ession o genes ela ed o he
epi helial-mesenchymal ansi ion (69,76,78,189). Indeed, Pi in is unc ionally associa ed wi h se e al
p o eins in ol ed in cy oskele on eo ganiza ion, such as WASF2 and NCKAP1, which could explain
he link be ween Pi in o e exp ession and malignan p og ession (122, 190). In his cell con ex , he
delay in cell cycle p og ession p oduced by JARID1B down egula ion could s imula e umo cells o
e-en e he cell cycle, inc easing p oli e a ion. Tumo cells wi h a slow-cycling pheno ype may be
me abolically ac i e and highly agg essi e, wi h inc eased po en ial o g ow and me as asize
(84,191,192).
We p opose ha Pi in could play an impo an ole in modula ing he p oli e a i e s a e o melanoma
cells by egula ing JARID1B gene exp ession. Howe e , u he esea ch will be necessa y o be e
unde s and he mechanisms unde lying his phenomenon, which could shed ligh on use ul he apeu ic
s a egies o hese umo s.
Finally, i we app oach all he esul s ob ained om his hesis as a whole, we can hink ha bo h sys ems
RKIP/NANOG and Pi in/JARID1B could be wo king oge he , so i would be acing wo scena ios. On
he one hand, benign melanocy es om ne us, in which he high exp ession RKIP main enance he
di e en ia ion s a e blocking he s emness ansc ip ion ac o NANOG. Addi ionally, high Pi in
exp ession may be egula ing he cell cycle h ough JARID1B and E2F1 exp ession. In he o he hand,
melanoma cells, in which he absen o RKIP exp ession p oduces a di e en pano ama. The exp ession
95
o NANOG a o s he acquisi ion o in asi e pheno ype h ough he exp ession o genes ela ed o he
mesenchymal epi helial ansi ion (69,76,78). In his cellula con ex , he delay in he cell cycle
p og ession p oduced o JARID1B down egula ion could ac s a s imulus o he umo cell o e-en e
in he cell cycle and me as asize. Melanoma cells could be able o adjus hei pheno ype o mee ex e nal
su i al equi emen s (86).
102

103
Re e ences
1. Sung H, Fe lay J, Siegel R.L, e al. Global
cance s a is ics 2020: GLOBOCAN
es ima es o incidence and mo ali y
wo ldwide o 36 cance s in 185
coun ies. CA. Cance J. Clin. 2021, 71:
209–249. DOI: 10.3322/caac.21660
2. Lin L, Yan L, Liu Y, e al. Incidence and
dea h in 29 cance g oups in 2017 and
end analysis om 1990 o 2017 om he
Global Bu den o Disease S udy. J
Hema ol Oncol. 2019, 12:96. DOI:
10.1186/s13045-019-0783-9
3. Pa k S.J, Oh C.M, Kim BW, e al.
Na ionwide incidence o ocula
melanoma in Sou h Ko ea by using he
na ional cance egis y da abase (1999–
2011). In es . Oph halmol. Vis. Sci.
2015, 56: 4719–24. DOI:
10.1167/io s.15-16532
4. Ka alinic A, Eisemann N, Waldmann A.
Skin cance sc eening in Ge many.
Documen ing melanoma incidence and
mo ali y om 2008 o 2013. D sch.
A z ebl. In . 2015, 112:629–34.
DOI: 10.3238/a z ebl.2015.0629
5. Pille on S, So o-Pe ez-de-Celis
E, Vigna J, e al. Es ima ed global cance
incidence in he oldes adul s in 2018 and
p ojec ions o 2050. In J Cance . 2021,
148:601-608. DOI: 10.1002/ijc.33232
6. Rahib L, Wehne M.R, Ma isian L.M, e
al. Es ima ed P ojec ion o US Cance
Incidence and Dea h o 2040. JAMA
Ne w Open. 2021, 4:e214708.
DOI: 10.1001/jamane wo kopen.2021.47
08
7. Smi enaa R, Pe e sen K.A, S ewa K, e
al. Cance incidence and mo ali y
p ojec ions in he UK un il 2035. B J
Cance . 2016, 115:1147-1155.
DOI: 10.1038/bjc.2016.304
8. Wei H.K, Thompson T.D, S ewa S.L,
e al. Cance Incidence P ojec ions in he
Uni ed S a es Be ween 2015 and 2050.
P e Ch onic Dis. 2021, 18:E59.
DOI: 10.5888/pcd18.210006
9. Singh G.K, Jemal A. Socioeconomic and
Racial/E hnic Dispa i ies in Cance
Mo ali y, Incidence, and Su i al in he
Uni ed S a es, 1950-2014: O e Six
Decades o Changing Pa e ns and
Widening Inequali ies. J En i on Public
Heal h. 2017, 2017:2819372.
DOI: 10.1155/2017/2819372
10. Saginala K, Ba souk A, Alu u J.S, e al.
Epidemiology o Melanoma. Med.
Sci.2021, 9:63.
DOI: 10.3390/medsci9040063
11. Li Z, Fang Y, Chen H, e al.
Spa io empo al ends o he global
bu den o melanoma in 204 coun ies and
e i o ies om 1990 o 2019: Resul s
om he 2019 global bu den o disease
s udy. Neoplasia. 2022, 24:12-21.
DOI: 10.1016/j.neo.2021.11.013
12. Jones O.T, Ranmu hu C.K.I, Hall P.N, e
al. Recognising Skin Cance in P ima y
Ca e. Ad The . 2020, 37:603-616.
DOI: 10.1007/s12325-019-01130-1
13. Liu-Smi h F, Fa ha A.M, A ce A, e al.
Sex di e ences in he associa ion o
cu aneous melanoma incidence a es and
geog aphic ul a iole ligh
exposu e. Jou nal o he Ame ican
Academy o De ma ology. 2017, 76:
499–505.e3.
DOI: 10.1016/j.jaad.2016.08.027
104
14. Side is E, Thomas S.J. Pa ien s' sun
p ac ices, pe cep ions o skin cance and
hei isk o skin cance in u al Aus alia.
Heal h P omo J Aus . 2020, 31:84-92.
DOI: 10.1002/hpja.253
15. W igh C.Y, Kapwa a T, Singh E, e al.
T ends in Melanoma Mo ali y in he
Popula ion G oups o Sou h A ica.
De ma ology. 2019, 235:396-399.
DOI: 10.1159/000500663
16. Mille K.D, Noguei a L, Ma io o A.B, e
al. Cance ea men and su i o ship
s a is ics, 2019. CA Cance J Clin. 2019,
69:363-385. DOI: 10.3322/caac.21565
17. Yang D.D, Salciccioli J.D, Ma shall D.C,
e al. T ends in malignan melanoma
mo ali y in 31 coun ies om 1985 o
2015. B J De ma ol. 2020, 183:1056-
1064. DOI: 10.1111/bjd.19010
18. Kozmin S.G, Slezak G, Reynaud-Angelin
A, e al. Ul a iole A adia ion is highly
mu agenic in cells ha a e unable o epai
7,8-dihyd o-8-oxoguanine in
Saccha omyces ce e isiae. P oc. Na l.
Acad. Sci. USA. 2005, 102:538–543.
DOI: 10.1073/pnas.0504497102
19. Ichihashi M, Ueda M, Budiyan o A, e al.
UV-induced skin damage. Toxicology.
2003, 189:21-39. DOI: 10.1016/s0300-
483x(03)00150-1
20. Ca S, Smi h C, We nbe g J.
Epidemiology and Risk Fac o s o
Melanoma. Su g. Clin. No h Am. 2020,
100:1–12.
DOI: 10.1016/j.suc.2019.09.005
21. Ob ado E, Liu-Smi h F, Dellinge R.W,
e al. Oxida i e s ess and an ioxidan s in
he pa hophysiology o malignan
melanoma. Biol. Chem. 2019, 400: 589–
612. DOI: 10.1515/hsz-2018-0327
22. Douki, T. Oxida i e S ess and
Geno oxici y in Melanoma Induc ion:
Impac on Repai Ra he Than Fo ma ion
o DNA Damage? Pho ochem. Pho obiol.
2020, 96, 962–972.
23. Khan, A.Q.; T a e s, J.B.; Kemp, M.G.
Roles o UVA adia ion and DNA
damage esponses in melanoma
pa hogenesis. En i on. Mol.
Mu agenesis. 2018, 59, 438–460.
24. T ucco, L.D.; Mund a, P.A.; Hogan, K.;
Ga cia-Ma inez, P.; Vi os, A.; Mandal,
A.; Macagno, N.; Gaudy-Ma ques e, C.;
Allan, D.; Baenke, F.; e al. Ul a iole
adia ion–induced DNA damage is
p ognos ic o ou come in melanoma.
Na . Med. 2019, 25, 221–224.
25. Raimondi, S.; Suppa, M.; Gandini, S.
Melanoma Epidemiology and Sun
Exposu e. Ac a De m. Vene eol. 2020,
100, 250–258.
26. Bishop, J. Len igos, Melanocy ic Nae i
and Melanoma. In Rook’s Tex book o
De ma ology, 8 h ed.; Blackwell
publishing: Ox o d, UK, 2010; Volume
54, pp. 32–56. In . J. Mol. Sci. 2021, 22,
6395 12 o 14
27. Dja id, A.R.; S onesi e , C.; Fulle on,
B.T.;Wang, S.W.; Ta a o, M.A.; Kwin a,
B.D.; G imes, J.M.; Geskin, L.J.;
Saenge , Y.M. E iologies o Melanoma
De elopmen and P e en ion Measu es:
A Re iew o he Cu en E idence.
Cance s 2021, 13, 4914.
h ps://doi.o g/10.3390/ cance s13194914
28. Hausaue , A.; Swe e , S.; Cockbu n, M.;
Cla ke, C. Inc eases in melanoma among
105
adolescen gi ls and young women in
Cali o nia: T ends by socioeconomic
s a us and UV adia ion exposu e. A ch.
De ma ol. 2011, 147, 783–789.
29. T aka elli, M.; Bylai e-Bucinskiene, M.;
Co eia, O.; Cozzio, A.; De V ies, E.;
Medenica, L.; Nago e, E.; Paoli, J.;
S a igos, A.J.; Del Ma mol, V.; e al.
Clinical assessmen o skin pho o ypes:
Wa ch you wo ds! Eu . J. De ma ol.
2017, 27, 615–619.
30. on Schuckmann LA, Hughes MCB,
Ghias and R, e al. Risk o Melanoma
Recu ence A e Diagnosis o a High-
Risk P ima y Tumo . JAMA
De ma ol. 2019;155(6):688–693.
doi:10.1001/jamade ma ol.2019.0440
31. Bo ella-Es ada R, T a es V, Requena C,
Guillen-Ba ona C, Nago e E. Co ela ion
o his ologic eg ession in p ima y
melanoma wi h sen inel node s a us.
JAMA De ma ol. 2014 Aug;150(8):828-
35. doi:
10.1001/jamade ma ol.2013.9856.
PMID: 24898614.
32. Escandell I, Ma ín JM, Jo dá E. No el
Immunologic App oaches o Melanoma
T ea men . Ac as De mosi iliog . 2017
Oc ;108(8):708-720. English, Spanish.
doi: 10.1016/j.ad.2017.01.017. Epub
2017 May 18. PMID: 28527857
33. Toussi, A., Mans, N., Welbo n, J., &
Kiu u, M. (2020). Ge mline mu a ions
p edisposing o melanoma. Jou nal o
cu aneous pa hology, 47(7), 606–616.
h ps://doi.o g/10.1111/cup.13689
34. Damsky, W.; Bosenbe g, M. Melanocy ic
ne i and melanoma: Un a eling a
complex ela ionship. Oncogene 2017,
36, 5771–5792.
35. Zocchi, L.; Lon ano, A.; Me li, M.; Dika,
E.; Nago e, E.; Quaglino, P.; Puig, S.;
Ribe o, S. Familial Melanoma and
Suscep ibili y Genes: A Re iew o he
Mos Common Clinical and De moscopic
Pheno ypic Aspec , Associa ed
Malignancies and P ac ical Tips o
Managemen . J. Clin. Med. 2021, 10,
3760. h ps://doi.o g/10.3390/
jcm10163760
36. S anganelli, I.; De Felici, M.B.; Mandel,
V.D.; Caini, S.; Raimondi, S.; Co so, F.;
Belle ba, F.; Quaglino, P.; Sanlo enzo,
M.; Ribe o,nS.; e al. The associa ion
be ween pes icide use and cu aneous
melanoma: A sys ema ic e iew and
me a-analysis. J. Eu . Acad. De ma ol.
Vene eol. 2020, 34, 691–708
37. Kani akis J. Ana omy, his ology and
immunohis ochemis y o no mal human
skin. Eu J De ma ol. 2002 Jul-
Aug;12(4):390-9; quiz 400-1. PMID:
12095893.
38. Ana omy and Physiology o he Skin,
Jou nal o he De ma ology Nu ses’
Associa ion: No embe /Decembe 2011 -
Volume 3 - Issue 6 - p 366 doi:
10.1097/JDN.0b013e31823cccbe
39. Law on S (2019) Skin 1: he s uc u e and
unc ions o he skin. Nu sing Times
[online]; 115, 12, 30-33.
40. Cla k WH J , Elde DE, Gue y D
IV,Eps ein MN, G eene MH, Van Ho n
M.A s udy o umo p og ession: he
p ecu so lesions o supe icial sp eading
and nodula melanoma. Hum Pa hol
1984;15:1147-65
106
41. Leona di GC, Falzone L, Salemi R,
Zanghì A, Spandidos DA, Mccub ey JA,
Candido S, Lib a M. Cu aneous
melanoma: F om pa hogenesis o he apy
(Re iew). In J Oncol. 2018
Ap ;52(4):1071-1080. doi:
10.3892/ijo.2018.4287. Epub 2018 Feb
27. PMID: 29532857; PMCID:
PMC5843392.
42. Mille AJ, Mihm MC. Melanoma. New
England Jou nal o Medicine.
2006;355(1):51–65.
doi:10.1056/NEJM a052166
43. Ma ghoob, Nadeem G., Liopy is,
Kons an inos and Jaimes, Na alia.
"De moscopy: A Re iew o he S uc u es
Tha Facili a e Melanoma De ec ion"
Jou nal o Os eopa hic Medicine, ol.
119, no. 6, 2019, pp. 380-390.
h ps://doi.o g/10.7556/jaoa.2019.067
44. Teixido C, Cas illo P, Ma inez-Vila C,
A ance A, Alos L. Molecula Ma ke s and
Ta ge s in Melanoma. Cells. 2021 Sep
5;10(9):2320. doi:
10.3390/cells10092320. PMID:
34571969; PMCID: PMC8469294.
45. Raman, M.; Chen, W.; Cobb, M.H.
Di e en ial egula ion and p ope ies o
MAPKs. Oncogene 2007, 26, 3100–3112.
46. Yeung K., Sei z T., Li S., Janoschk P.,
McFe ank B., Kaise C., Feek F.,
Ka sanakisk K.D., Rose D.W., Mischak
H., e al. Supp ession o Ra -1 kinase
ac i i y and MAP kinase signalling by
RKIP. Na u e. 1999;401:e173.
doi: 10.1038/43686.
47. Da ies, H.; Bignell, G.R.; Cox, C.;
S ephens, P.; Edkins, S.; Clegg, S.;
Teague, J.; Wo endin, H.; Ga ne , M.J.;
Bo omley, W.; e al. Mu a ions o he
BRAF gene in human cance . Na u e
2002, 417, 949–954.
48. Nissan, M.H.; P a ilas, C.A.; Jones, A.M.;
Rami ez, R.; Won, H.; Liu, C.; Tiwa i, S.;
Kong, L.; Han ahan, A.J.; Yao, Z.; e al.
Loss o NF1 in cu aneous melanoma is
associa ed wi h RAS ac i a ion and MEK
dependence. Cance Res. 2014, 74, 2340–
2350.
49. Beshi A.B., Ren G., Magpusao A.N.,
Ba one L.M., Yeung K.C., Fen eany G.
Ra kinase inhibi o p o ein supp esses
nuclea ac o -κB-dependen cance cell
in asion h ough nega i e egula ion o
ma ix me allop o einase
exp ession. Cance Le . 2010;299:137–
149. doi: 10.1016/j.canle .2010.08.012.
50. Jila eanu L.B., Zi o C.R., Aziz S.A.,
Con ad P.J., Schmi z J.C., Sznol M.,
Camp R.L., Rimm D.L., Kluge H.M. C-
Ra is associa ed wi h disease p og ession
and cell p oli e a ion in a subse o
melanomas. Clin. Cance
Res. 2009;15:5704–5713.
doi: 10.1158/1078-0432.CCR-09-0198.
51. Schuie e M.M., Ba aille F., Hagan S.,
Kolch W., Bosse ho A.-K. Reduc ion in
Ra Kinase Inhibi o P o ein Exp ession
Is Associa ed wi h Inc eased Ras-
Ex acellula Signal-Regula ed Kinase
Signaling in Melanoma Cell
Lines. Cance Res. 2004;64:5186.
doi: 10.1158/0008-5472.CAN-03-3861.
52. Ca dile V., Malapon e G., Lo e o C.,
Lib a M., Caggia S., T o a o F.M.,
Musumeci G. Ra kinase inhibi o p o ein
(RKIP) and phospho-RKIP exp ession in
melanomas. Ac a
107
His ochem. 2013;115:795–802.
doi: 10.1016/j.ac his.2013.03.003.
53. Scheid, M.P.; Woodge , J.R. PKB/AKT:
Func ional insigh s om gene ic models.
Na . Re . Mol. Cell Biol. 2001, 2, 760–
768.
54. S ahl, J.M.; Cheung, M.; Sha ma, A.;
T i edi, N.R.; Shanmugam, S.;
Robe son, G.P. Loss o PTEN p omo es
umou de elopmen in malignan
melanoma. Cance Res. 2003, 63, 2881–
2890.
55. Li o, P.; P a ilas, C.A.; Joseph, E.W.;
Tadi, M.; Halilo ic, E.; Zub owski, M.;
Huang, A.; Wong, W.L.; Callahan, M.K.;
Me ghoub, T.e al. Relie o p o ound
eedback inhibi ion o mi ogenic
signaling by RAF inhibi o s a enua es
hei ac i i y in BRAFV600E melanomas.
Cance Cell 2012, 22, 668–682.
56. Ben-Ne iah Y, Ka in M. In lamma ion
mee s cance , wi h NF-κB as he
ma chmake . Na Immunol. 2011 Jul
19;12(8):715-23. doi: 10.1038/ni.2060.
PMID: 21772280.
57. Baldwin AS. Con ol o oncogenesis and
cance he apy esis ance by he
ansc ip ion ac o NF-kappaB. J Clin
In es . 2001 Feb;107(3):241-6. doi:
10.1172/JCI11991. PMID: 11160144;
PMCID: PMC199203.
58. Dechend R, Hi ano F, Lehmann K,
Heissmeye V, Ansieau S, Wulczyn FG,
Scheide ei C, Leu z A. The Bcl-3
oncop o ein ac s as a b idging ac o
be ween NF-kappaB/Rel and nuclea co-
egula o s. Oncogene. 1999 Jun
3;18(22):3316-23. doi:
10.1038/sj.onc.1202717. PMID:
10362352.
59. Geo ge J, Lim JS, Jang SJ, e al.
Comp ehensi e genomic p o iles o small
cell lung cance . Na u e. 2015 Aug
6;524(7563):47-53. doi:
10.1038/na u e14664. Epub 2015 Jul 13.
PMID: 26168399; PMCID:
PMC4861069.
60. Dhawan P, Richmond A. A no el NF-
kappa B-inducing kinase-MAPK
signaling pa hway up- egula es NF-kappa
B ac i i y in melanoma cells. J Biol
Chem. 2002 Ma 8;277(10):7920-8. doi:
10.1074/jbc.M112210200. Epub 2001
Dec 28. PMID: 11773061; PMCID:
PMC2668260.
61. A oza ena I, Wellb ock C. Pheno ype
plas ici y as enable o melanoma
p og ession and he apy esis ance. Na
Re Cance . 2019 Jul;19(7):377-391. doi:
10.1038/s41568-019-0154-4. Epub 2019
Jun 17. PMID: 31209265.
62. F iedmann-Mo inski D, Ve ma IM.
Dedi e en ia ion and ep og amming:
o igins o cance s em cells. EMBO Rep.
2014 Ma ;15(3):244-53. doi:
10.1002/emb .201338254. Epub 2014
Feb 14. PMID: 24531722; PMCID:
PMC3989690.
63. Gab iela-F ei as M, Pinhei o J, Raquel-
Cunha A, Ca doso-Ca nei o D, Ma inho
O. RKIP as an In lamma o y and Immune
Sys em Modula o : Implica ions in
Cance . Biomolecules. 2019 No
22;9(12):769. doi:
10.3390/biom9120769. PMID:
31766768; PMCID: PMC6995551.

108
64. Ma inho O., G anja S., Ja aquemada T.,
Caei o C., Mi anda-Gonçal es V.,
Hona a M., Cos a P., Damasceno M.,
Rosne M.R., Lopes J.M., e al.
Down egula ion o RKIP is associa ed
wi h poo ou come and malignan
p og ession in gliomas. PLoS
ONE. 2012;7:e30769.
doi: 10.1371/jou nal.pone.0030769.
65. Beshi A.B., Ren G., Magpusao A.N.,
Ba one L.M., Yeung K.C., Fen eany G.
Ra kinase inhibi o p o ein supp esses
nuclea ac o -κB-dependen cance cell
in asion h ough nega i e egula ion o
ma ix me allop o einase
exp ession. Cance Le . 2010;299:137–
149. doi: 10.1016/j.canle .2010.08.012.
66. Jila eanu L.B., Zi o C.R., Aziz S.A.,
Con ad P.J., Schmi z J.C., Sznol M.,
Camp R.L., Rimm D.L., Kluge H.M. C-
Ra is associa ed wi h disease p og ession
and cell p oli e a ion in a subse o
melanomas. Clin. Cance
Res. 2009;15:5704–5713.
doi: 10.1158/1078-0432.CCR-09-0198.
67. Schuie e M.M., Ba aille F., Hagan S.,
Kolch W., Bosse ho A.-K. Reduc ion in
Ra Kinase Inhibi o P o ein Exp ession
Is Associa ed wi h Inc eased Ras-
Ex acellula Signal-Regula ed Kinase
Signaling in Melanoma Cell
Lines. Cance Res. 2004;64:5186.
doi: 10.1158/0008-5472.CAN-03-3861.
68. Schoen gen F, Jonic S. PEBP1/RKIP
beha io : a mi o o ac in-memb ane
o ganiza ion. Cell Mol Li e Sci. 2020
Ma ;77(5):859-874. doi: 10.1007/s00018-
020-03455-5. Epub 2020 Jan 20. PMID:
31960115.
69. Yada AK, Desai NS. Cance S em Cells:
Acquisi ion, Cha ac e is ics, The apeu ic
Implica ions, Ta ge ing S a egies and
Fu u e P ospec s. S em Cell Re Rep.
2019 Jun;15(3):331-355. doi:
10.1007/s12015-019-09887-2. PMID:
30993589.
70. Molo sky AV, Pa dal R, Mo ison SJ.
Di e se mechanisms egula e s em cell
sel - enewal. Cu Opin Cell Biol. 2004
Dec;16(6):700-7. doi:
10.1016/j.ceb.2004.09.004. PMID:
15530784.
71. Wa anabe M, Ohnishi Y, Inoue H, Wa o
M, Tanaka A, Kakudo K, Nozaki M.
NANOG exp ession co ela es wi h
di e en ia ion, me as asis and esis ance
o p eope a i e adju an he apy in o al
squamous cell ca cinoma. Oncol Le .
2014 Jan;7(1):35-40. doi:
10.3892/ol.2013.1690. Epub 2013 No
19. PMID: 24348816; PMCID:
PMC3861537.
72. Ling GQ, Chen DB, Wang BQ, Zhang LS.
Exp ession o he plu ipo ency ma ke s
Oc 3/4, Nanog and Sox2 in human b eas
cance cell lines. Oncol Le . 2012
Dec;4(6):1264-1268. doi:
10.3892/ol.2012.916. Epub 2012 Sep 13.
PMID: 23197999; PMCID:
PMC3506717.
73. Zhang Y, Wang Z, Yu J, Shi Jz, Wang C,
Fu Wh, Chen Zw, Yang J. Cance s em-
like cells con ibu e o cispla in esis ance
and p og ession in bladde cance . Cance
Le . 2012 Sep 1;322(1):70-7. doi:
10.1016/j.canle .2012.02.010. Epub 2012
Feb 15. PMID: 22343321.
109
74. Pe ego M, To o e o M, T agni G,
Ma iani L, Deho P, Ca bone A, San inami
M, Pa uzzo R, Mina PD, Villa A, P a esi
G, Cossa G, Pe ego P, Daidone MG,
Alison MR, Pa miani G, Ri ol ini L,
Cas elli C. He e ogeneous pheno ype o
human melanoma cells wi h in i o and
in i o ea u es o umo -ini ia ing cells. J
In es De ma ol. 2010 Jul;130(7):1877-
86. doi: 10.1038/jid.2010.69. Epub 2010
Ap 8. PMID: 20376064.
75. Bo ull A, Ghislin S, Deshayes F, Lau iol
J, Alcaide-Lo idan C, Middendo p S.
Nanog and Oc 4 o e exp ession inc eases
mo ili y and ansmig a ion o melanoma
cells. J Cance Res Clin Oncol. 2012
Jul;138(7):1145-54. doi: 10.1007/s00432-
012-1186-2. Epub 2012 Ma 11. PMID:
22406932.
76. Sun C, Sun L, Jiang K, Gao DM, Kang
XN, Wang C, Zhang S, Huang S, Qin X,
Li Y, Liu YK. NANOG p omo es li e
cance cell in asion by inducing
epi helial-mesenchymal ansi ion
h ough NODAL/SMAD3 signaling
pa hway. In J Biochem Cell Biol. 2013
Jun;45(6):1099-108. doi:
10.1016/j.biocel.2013.02.017. Epub 2013
Ma 7. E a um in: In J Biochem Cell
Biol. 2018 Dec;105:144. PMID:
23474366.
77. B able z T, Kallu i R, Nie o MA,
Weinbe g RA. EMT in cance . Na Re
Cance . 2018 Feb;18(2):128-134. doi:
10.1038/n c.2017.118. Epub 2018 Jan 12.
PMID: 29326430.
78. Palla A.R., Piazzolla D., Alcaza N.,
Cañame o M., G aña O., Gómez-López
G., Dominguez O., Dueñas M., Pa amio
J.M., Se ano M. The plu ipo ency ac o
NANOG p omo es he o ma ion o
squamous cell ca cinomas. Sci.
Rep. 2015;5 doi: 10.1038/s ep10205.
79. Yesilkanal AE, Rosne MR. Ta ge ing
Ra Kinase Inhibi o y P o ein Regula ion
and Func ion. Cance s (Basel). 2018 Sep
4;10(9):306. doi:
10.3390/cance s10090306. PMID:
30181452; PMCID: PMC6162369.
80. Ha meye KM, Facomp e ND, He lyn M,
Basu D. JARID1 His one Deme hylases:
Eme ging Ta ge s in Cance . T ends
Cance . 2017 Oc ;3(10):713-725. doi:
10.1016/j. ecan.2017.08.004. Epub 2017
Sep 12. PMID: 28958389; PMCID:
PMC5679451.
81. Han M, Xu W, Cheng P, Jin H, Wang X.
His one deme hylase lysine deme hylase
5B in de elopmen and cance .
Onco a ge . 2017 Jan 31;8(5):8980-8991.
doi: 10.18632/onco a ge .13858. PMID:
27974677; PMCID: PMC5352456.
82. Pe ez-Dominguez F, Ca illo-Bel án D,
Blanco R, Muñoz JP, León-C uz G,
Co alan AH, U zúa U, Cala GM,
Aguayo F. Role o Pi in, an oxida i e
s ess senso p o ein, in epi helial
ca cinogenesis. Biology (Basel). 2021 Feb
4;10(2):116. doi:
10.3390/biology10020116. PMID:
33557375; PMCID: PMC7915911.
83. Xhabija B, Kidde BL. KDM5B is a mas e
egula o o he H3K4-me hylome in s em
cells, de elopmen and cance . Semin
Cance Biol. 2019 Aug;57:79-85. doi:
10.1016/j.semcance .2018.11.001. Epub
2018 No 16. PMID: 30448242; PMCID:
PMC6522339.
110
84. Vogel FCE, Bo dag N, Zügne E,
T ajko ic-A sic M, Chau is é H,
Shannan B, Vá aljai R, Ho n S, Magnes C,
Thomas Si eke J, Schadendo D, Roesch
A. Ta ge ing he H3K4 deme hylase
KDM5B Rep og ams he me abolome and
pheno ype o melanoma cells. J In es
De ma ol. 2019 Dec;139(12):2506-
2516.e10. doi: 10.1016/j.jid.2019.06.124.
Epub 2019 Jun 21. PMID: 31229500.
85. Giun a EF, A ichiello G, Cu ie o M,
Pappala do A, Bosso D, Rosano a M,
Diana A, Gio dano P, Pe illo A, Fede ico
P, Fabozzi T, Pa ola S, Riccio V, Mucci B,
Vanella V, Fes ino L, Daniele B, Ascie o
PA, O a iano M, On Behal O Sci o
You h. Epigene ic egula ion in
melanoma: ac s and hopes. Cells. 2021
Aug 11;10(8):2048. doi:
10.3390/cells10082048. PMID:
34440824; PMCID: PMC8392422.
86. Facomp e ND, Ha meye KM, Sole X,
Kab aji S, Belden Z, Sahu V, Whelan K,
Tanaka K, Weins ein GS, Mon one KT,
Roesch A, Gimo y PA, He lyn M, Rus gi
AK, Nakagawa H, Ramaswamy S, Basu D.
JARID1B enables ansi be ween dis inc
s a es o he s em-like cell popula ion in
o al cance s. Cance Res. 2016 Sep
15;76(18):5538-49. doi: 10.1158/0008-
5472.CAN-15-3377. Epub 2016 Aug 3.
E a um in: Cance Res. 2017 Dec
15;77(24):7136. PMID: 27488530;
PMCID: PMC5026599.
87. Ma ghoob, Nadeem G., Liopy is,
Kons an inos and Jaimes, Na alia.
"De moscopy: A Re iew o he S uc u es
Tha Facili a e Melanoma De ec ion"
Jou nal o Os eopa hic Medicine, ol.
119, no. 6, 2019, pp. 380-390.
h ps://doi.o g/10.7556/jaoa.2019.067
88. F iedmanRJ, RigelDS, Kop AW. Ea ly
de ec ion o malignan melanoma: he ole
o physician examina ion and sel -
examina ion o he skin. CA Cance J
Clin. 1985;35(3):130-
151.10.3322/canjclin.35.3.130
89. Chambe lainAJ, F i schiL, KellyJW. No
dula melanoma: pa ien s’ pe cep ions o
p esen ing ea u es and implica ions o
ea lie de ec ion. J Am Acad
De ma ol. 2003;48(5):694-701.
doi:10.1067/mjd.2003.216
90. KellyJW. Nodula melanoma: how
cu en app oaches o ea ly de ec ion a e
ailing. J D ugs De ma ol. 2005;4(6):790-
793
91. Cla k WH, F om L, Be na dino EA,
Mihm MC. The His ogenesis and
Biologic Beha io o P ima y Human
Malignan Melanomas o he Skin.
Cance Res.1969;29(3):705-27.
92. B eslow A. Thickness, c oss-sec ional
a eas and dep h o in asion in he
p ognosis o cu aneous melanoma. Ann
Su g. 1970;172(5):902-8.
93. Teje a-Vaque izo A, Ribe o S, Puig S,
Boada A, Pa adela S, Mo eno-Ramí ez D,
Cañue o J, de Unamuno B, B inca A,
Descalzo-Gallego MA, Osella-Aba e S,
Cassoni P, Ca e a C, Vidal-Sica S,
Bennássa A, Rull R, Alos L, Requena C,
Boluma I, T a es V, Pla Á, Fe nández-
O land A, Jaka A, Fe nández-Figue es
MT, Hila i JM, Giménez-Xa ie P, Viei a
R, Bo ella-Es ada R, Román-Cu o C,
111
Fe ándiz L, Iglesias-Pena N, Fe ándiz
C, Mal ehy J, Quaglino P, Nago e E;
SENTIMEL g oup. Su i al analysis and
sen inel lymph node s a us in hin
cu aneous melanoma: A mul icen e
obse a ional s udy. Cance Med. 2019
Aug;8(9):4235-4244. doi:
10.1002/cam4.2358. Epub 2019 Jun 18.
PMID: 31215168; PMCID:
PMC6675713.
94. Dai Oga a, Kenji o Namikawa, Aki a
Takahashi, Naoya Yamazaki, A51 e iew
o he AJCC melanoma s aging sys em in
he TNM classi ica ion (eigh h
edi ion), Japanese Jou nal o Clinical
Oncology, Volume 51, Issue 5, May
2021, Pages 671–
674, h ps://doi.o g/10.1093/jjco/hyab02
2
95. Keung, E. Z., & Ge shenwald, J. E.
(2018). The eigh h edi ion Ame ican Join
Commi ee on Cance (AJCC) melanoma
s aging sys em: implica ions o
melanoma ea men and ca e. Expe
e iew o an icance he apy, 18(8), 775–
784.
96. Sca ena, C., Mu as, D., & Tomei, S.
(2021). Cu aneous Melanoma
Classi ica ion: The Impo ance o High-
Th oughpu Genomic
Technologies. F on ie s in oncology, 11,
635488.
97. Elde DE, Bas ian BC, C ee IA, Massi D,
Scolye RA. The 2018 Wo ld Heal h
O ganiza ion Classi ica ion o Cu aneous,
Mucosal, and U eal Melanoma: De ailed
Analysis o 9 Dis inc Sub ypes De ined
by Thei E olu iona y Pa hway. A ch
Pa hol Lab Med 2020;144:500–22.
98. Elde DE, Massi D, Scolye RA,
Willemze R, edi o s. WHO Classi ica ion
o Skin Tumou s. Fou h ed. Lyon,
F ance: IARC; 2018.
99. Busam KJ, Ge ami P, Scolye RA.
Pa hology o Melanocy ic Tumo s.
Else ie ; 2019.
100. Plaza JA, P ie o VG. Pa hology o
Pigmen ed Skin Lesions. Sp inge ; 2017.
101. Busam KJ, Ge ami P, Scolye RA.
Pa hology o Melanocy ic Tumo s.
Else ie ; 2019.
102. Massi G. Leboi . PE. His ological
Diagnosis o Ne i and Melanoma.
Sp inge ; 2014.
103. Basu o-Lozada P, Molina-Aguila C,
Cas aneda-Ga cia C, Vázquez- C uz ME,
Ga cia-Salinas OI, Ál a ez-Cano A, e al.
Ac al len iginous melanoma: basic ac s,
biological cha ac e is ics and esea ch
pe spec i es o an unde s udied disease.
Pigmen Cell Melanoma Res 2021;34:59–
71.
104. Fe nandez-Flo es A, Cassa ino DS.
His opa hological diagnosis o ac al
len iginous melanoma in ea ly s ages.
Ann Diagn Pa hol 2017;26:64–9.
105. Bobos M. His opa hologic classi ica ion
and p ognos ic ac o s o melanoma: a
2021 upda e. I al J De ma ol Vene ol.
2021 Jun;156(3):300-321. doi:
10.23736/S2784-8671.21.06958-3. Epub
2021 May 13. PMID: 33982546.
106. P ie o VG, Shea CR.
Immunohis ochemis y o melanocy ic
p oli e a ions. A ch Pa hol Lab Med.
2011 Jul;135(7):853-9. doi:
10.5858/2009-0717-RAR.1. PMID: 2173