Oncogenic D816V-KIT signaling in mas cells causes
pe sis en IL-6 p oduc ion
by A aceli Tobío, Gee hani Banda a, Denise A. Mo is, Do-Kyun Kim, Michael P. O'Connell,
Hi sh D. Koma ow, Melody C. Ca e , Daniel Sm z, Dean D. Me cal e, and Ana Oli e a
Haema ologica 2019 [Epub ahead o p in ]
Ci a ion: A aceli Tobío, Gee hani Banda a, Denise A. Mo is, Do-Kyun Kim, Michael P. O'Connell,
Hi sh D. Koma ow, Melody C. Ca e , Daniel Sm z, Dean D. Me cal e, and Ana Oli e a.
Oncogenic D816V-KIT signaling in mas cells causes pe sis en IL-6 p oduc ion.
Haema ologica. 2019; 104:xxx
doi:10.3324/haema ol.2018.212126
Publishe 's Disclaime .
E-publishing ahead o p in is inc easingly impo an o he apid dissemina ion o science.
Haema ologica is, he e o e, E-publishing PDF iles o an ea ly e sion o manusc ip s ha
ha e comple ed a egula pee e iew and ha e been accep ed o publica ion. E-publishing
o his PDF ile has been app o ed by he au ho s. A e ha ing E-published Ahead o P in ,
manusc ip s will hen unde go echnical and English edi ing, ypese ing, p oo co ec ion and
be p esen ed o he au ho s' inal app o al; he inal e sion o he manusc ip will hen
appea in p in on a egula issue o he jou nal. All legal disclaime s ha apply o he
jou nal also pe ain o his p oduc ion p ocess.
Copy igh 2019 Fe a a S o i Founda ion.
Published Ahead o P in on Ap il 4, 2019, as doi:10.3324/haema ol.2018.212126.
Tobío e al.,
1
Oncogenic D816V-KIT signaling in mas cells causes pe sis en
IL-6 p oduc ion
A aceli Tobío1, Gee hani Banda a1, Denise A. Mo is1, Do-Kyun Kim1
,
Michael P.
O’Connell2, Hi sh D. Koma ow1, Melody C. Ca e 1, Daniel Sm z1, Dean D. Me cal e1 ¶, Ana
Oli e a1¶*
1Mas Cell Biology Sec ion, Labo a o y o Alle gic Diseases, Na ional Ins i u e o Alle gy
and In ec ious Diseases, Na ional Ins i u es o Heal h, Be hesda, Ma yland, USA
2Gene ics and Pa hogenesis o Alle gy Sec ion, Labo a o y o Alle gic Diseases, Na ional
Ins i u e o Alle gy and In ec ious Diseases, Na ional Ins i u es o Heal h, Be hesda,
Ma yland, USA
¶ DDM and AO a e senio co-au ho s
*Co espondence:
Ana Oli e a
ana.oli e [email protected]
Running i le: STAT5 and IL-6 dys egula ion in mas ocy osis
Wo d Coun :
Abs ac : 250
Main ex : 4,294
Numbe o Figu es: 6
Con ains a Supplemen a y Appendix ile: Supplemen a y Me hods, Supplemen a y Tables 1
and 2, Supplemen a y Figu es S1-5 and Re e ences
Acknowledgemen s
We hank Robin Eisch in he ole as p o ocol s udy coo dina o , Linda Sco , he nu se
p ac i ione on he s udy p o ocols, Pahul Hanj a and I ina Ma ic o p o iding in o ma ion
on pa ien s and Daly Can a e o a anging o pa ien bone ma ow samples.
Tobío e al.,
2
Abs ac
Pe sis en dys egula ion o IL-6 p oduc ion and signaling ha e been implica ed in he
pa hology o a ious cance s. In sys emic mas ocy osis, inc eased se um le els o IL-6
associa e wi h disease se e i y and p og ession, al hough he mechanisms in ol ed a e no
well unde s ood. Since sys emic mas ocy osis o en associa es wi h he p esence in
hema opoie ic cells o a soma ic gain-o - unc ion a ian in KIT, D816V-KIT, we examined
i s po en ial ole in IL-6 up egula ion. Bone ma ow mononuclea cul u es om pa ien s wi h
g ea e D816V allelic bu den eleased inc eased amoun s o IL-6 which co ela ed wi h he
pe cen age o mas cells in he cul u es. In acellula IL-6 s aining by low cy ome y and
immuno luo escence was p ima ily associa ed wi h mas cells and sugges ed a highe
pe cen age o IL-6 posi i e mas cells in pa ien s wi h highe D816V allelic bu den.
Fu he mo e, mas cell lines exp essing D816V-KIT, bu no hose exp essing no mal KIT o
o he KIT a ian s, p oduced cons i u i ely high IL-6 amoun s a he message and p o ein
le els. We u he demons a e ha abe an KIT ac i i y and signaling a e c i ical o he
induc ion o IL-6 and in ol e STAT5 and PI3K pa hways bu no STAT3 o STAT4.
Ac i a ion o STAT5A and STATB downs eam o D816V-KIT was media ed by JAK2 bu
also by MEK/ERK1/2, which no only p omo ed STAT5 phospho yla ion bu also i s long-
e m ansc ip ion. Ou s udy hus suppo s a ole o mas cells and D816V-KIT ac i i y in
IL-6 dys egula ion in mas ocy osis and p o ides insigh s in o he in acellula mechanisms.
The indings con ibu e o a be e unde s anding o he physiopa hology o mas ocy osis and
sugges he impo ance o he apeu ic a ge ing o hese pa hways.
Tobío e al.,
3
In oduc ion
Mas ocy osis de ines a g oup o he e ogeneous diso de s cha ac e ized by he accumula ion
o neoplas ic/clonal mas cells in he skin, bone ma ow (BM) and o he o gans 1.
Mas ocy osis is clinically subdi ided in o sys emic (SM) and cu aneous (CM) mas ocy osis,
bo h o which a e comp ised o se e al a ian s de ined in acco dance wi h his ologic and
clinical pa ame e s and o gan in ol emen 1. Soma ic a ian s in he ecep o o s em cell
ac o (SCF), KIT, ha ende i cons i u i ely ac i e o en associa e wi h SM, pa icula ly
p.(D816V), a missense in he y osine kinase domain o KIT. D816V-KIT may be
accompanied by a ian s in o he genes ha u he con ibu e o he oncogenic expansion o
mas cells 2-4.
In e leukin-6 (IL-6) is a pleio opic cy okine p oduced by se e al cell ypes including
s omal, hema opoie ic and umo cells. In addi ion o i s in ol emen in no mal
in lamma o y p ocesses and hos immune de ense mechanisms, IL-6 may con ibu e o
malignancy in a ange o cance s including mul iple myeloma, B cell and non-B cell
leukemias and lymphomas 5, 6, by modula ing cellula de elopmen , g ow h, apop osis,
me as asis and/o cellula esis ance o chemo he apy 6. As ele a ed IL-6 le els in he se um
o pa ien s wi h such malignancies ha e been associa ed wi h poo clinical ou comes,
blocking IL-6 o i s syn hesis in hese pa ien s is iewed as a po en ial he apeu ic a enue 7, 8.
In SM, he le els o se um IL-6 a e highe in pa ien s wi h agg essi e e sus indolen
a ian s o SM and ha e been associa ed wi h ad e se clinical ea u es o mas ocy osis such
as accumula ion o mas cells in he BM, o ganomegaly, ele a ed yp ase le els 9, 10,
os eopo osis and/o bone pain 11. Al hough p og ession in o mo e agg essi e disease wi hin
pa ien s wi h indolen SM (ISM) occu s only in a subse o pa ien s, IL-6 plasma le els
signi ican ly co ela e wi h disease p og ession and lowe p og ession- ee su i al,
sugges ing ha blockade o IL-6 syn hesis o unc ion may be bene icial in cases wi h
abe an IL-6 pa hways 10. O he s udies ha e shown ha IL-6 p omo es he di e en ia ion,
g ow h and deg anula ion o no mal mas cells 12, and induces he p oduc ion o eac i e
oxygen species by malignan mas cells and hei accumula ion in issues in a model o
mas ocy osis 13. Despi e he po en ial implica ions o disease pa hology, he cell ypes and
he mechanisms ha may con ibu e o he cons i u i ely ele a ed IL-6 le els in mas ocy osis
a e no known.
In his s udy, we es he hypo hesis ha cells exp essing gain o unc ion a ian s o KIT,
pa icula ly D816V-KIT, con e he abili y o cons i u i ely p oduce IL-6. As will be shown,
ex- i o BM mas cells om pa ien s wi h SM elease IL-6 in co ela ion wi h he allelic
equency o D816V-KIT. We u he demons a e ha exp ession o D816V-KIT causes
pe sis en IL-6 induc ion by mechanisms independen o au oc ine eed- o wa d loops
in ol ing IL-6 and signal ansduce and ac i a o o ansc ip ion 3 (STAT3) desc ibed in
o he malignan cells, bu dependen on oncogenic KIT-de i ed signals. These signals include
phospha idylinosi ide 3-kinase (PI3K) pa hways and oncogenic STAT5 ac i a ion by bo h
janus kinase 2 (JAK2) and, unexpec edly, by he mi ogen-ac i a ed p o ein kinase
Tobío e al.,
4
MEK/ERK1/2 pa hways. These da a expand ou unde s anding o he po en ial mechanisms
ini ia ing enhanced IL-6 p oduc ion in mas ocy osis and emphasize a ge s o he apeu ic
in e en ion in cases o high IL-6 p o iles and suspec ed disease p og ession.
Me hods
A de ailed desc ip ion o he me hods used in his s udy can be ound in Online
Supplemen a y Appendix
Pa ien s o s udy
BM samples we e ob ained when clinically indica ed om pa ien s wi h SM classi ied
acco ding o he WHO guidelines 1, 3, 14 (Online Supplemen a y Table 1). All human samples
we e ob ained a e in o med consen , on clinical p o ocols app o ed by he Ins i u ional
Re iew Boa d o he Na ional Ins i u e o Alle gy and In ec ious Diseases (02-I-0277 and 08-
I-0184) in ag eemen wi h he decla a ion o Helsinki. D816V-KIT mu a ion analysis and i s
allele bu den in he BM (D816V-KIT equency) we e de e mined by allele-speci ic PCR
om pa ien blood genomic DNA 15.
Cell lysa es
Cell lines we e cul u ed as desc ibed in he Online Supplemen a y Appendix. To ob ain
lysa es o wes e n blo s, 3x106 cells we e pla ed in 6 well pla es and incuba ed wi h o
wi hou he indica ed inhibi o s o 2 h in se um- ee media. Cells we e lysed as desc ibed 16.
IL-6 measu emen s
Cells (3x106), we e pla ed in 6 well pla es o 2 h o o e nigh in 6 mL o se um- ee media
o exclude he possibili y ha any ex insic s imulan p esen in he se um would in luence
he esul s. IL-6 eleased in o he media was measu ed by ELISA (R&D Sys ems). Human
colo ec al ca cinoma HCT116 cells we e s imula ed wi h 20 ng/mL µM PMA plus 1 µM
ionomycin o e nigh and he supe na an s hen collec ed o IL-6 measu emen s.
Mononuclea cells in BM aspi a es om pa ien s we e cul u ed in S emP o-34 medium wi h
human ecombinan SCF (100 ng/mL) o 2 o 4 days. IL-6 eleased in o he media was
de e mined by ELISA. Al e na i ely, IL-6 exp ession in single cells was de e mined by low
cy ome y using a LSRII low cy ome e . BM cells we e incuba ed wi h B e eldin A o 4 h
and s ained wi h an an ibody cock ail con aining an i-CD3-QDOT605, an i-CD34-APC, an i-
KIT-BV605 and an i-Fc
ε
RI-FITC, o 30 min. Cells we e ixed, pe meabilized and s ained
wi h an i-IL-6-PE o 30 min. Exp ession o IL-6 in mas cells (CD3-/CD34-/KIT+/Fc
ε
RI+)
was analyzed using FlowJo so wa e.
Quan i a i e eal- ime PCR

Tobío e al.,
5
HMC-1.2 cells (3x106) we e pla ed in 6-well pla es in 6 mL and incuba ed o 2 h in se um
ee media. Cellula RNA was ex ac ed and e e se- ansc ibed in o cDNA. cDNA was hen
ampli ied using TaqMan® Gene Exp ession Mas e Mix and Taqman® Gene Exp ession
Assays o IL-6, STAT3, STAT4, STAT5A, STAT5B o GAPDH as desc ibed in he Online
Supplemen a y Appendix.
Knockdown o STAT ansc ip ion ac o s
Knockdown o STAT3 and STAT4 was pe o med by len i i al-media ed ansduc ion o
small hai pin RNA (sh-RNA) (Sigma-Ald ich, S . Louis, MO) as desc ibed 17. STAT5
mRNA was silenced by a small in e e ence-RNA (si-RNA) “ON-TARGET” pool om
Dha macon (La aye e, CO), in oduced in o cells by elec opo a ion.
S a is ical analysis
Da a we e exp essed as mean ±SEM. Values we e om a leas 3 independen expe imen s,
each pe o med a leas in duplica e. S a is ically signi ican di e ences we e calcula ed by
using he S uden - es (unpai ed). S a is ical signi icance was indica ed as ollows: *P
<0.01, **P<0.01, ***P <0.001 and ****P<0.0001.
Resul s
Release o IL-6 om pa ien ’s bone ma ow cells and i s associa ion wi h D816V-KIT
and mas cell equencies
The le els o IL-6 in se um 9 as well as he allelic equency o D816V-KIT18 co ela e
wi h he le els o yp ase, a su oga e ma ke o mas cell bu den. As mas cells o en
accumula e in he BM in SM, we es ed he abili y o BM cells o p oduce IL-6 in sho - e m
cul u es. BM mononuclea cells isola ed om pa ien s wi h SM showed a ied abili y o
elease IL-6 in o he cul u e media a e 2 o 4 days, wi h mo e elease obse ed in cells om
pa ien s wi h a highe BM D816V-KIT allelic equency (Figu e 1A). Al hough cells o he
han ma u e mas cells may exp ess D816V-KIT 18, he elease o IL-6 in o he media
co ela ed wi h he pe cen age o mas cells wi hin BM li e cells (Figu e 1B), sugges ing a
con ibu o y ole o mas cells in IL-6 p oduc ion. Addi ionally, we analyzed in acellula
IL-6 s aining in single BM mas cells by low cy ome y om h ee sepa a e pa ien s
desc ibed in he Online Supplemen a y Table 1. Pa ien 1 had idiopa hic anaphylaxis and did
no mee c i e ia o SM and hus was used as a con ol. This pa ien had no de ec able
D816V-KIT, 0.098% o BM cells we e CD3-/CD34-/KIT+/Fc
ε
RI+ (mas cells) and a mino
pe cen age o hese we e IL-6 posi i e (0.063%) (Figu e 1C, le panel). Howe e , CD3-
/CD34-/KIT+/Fc
ε
RI+ cells om pa ien s 2 and 3, wi h BM D816V equencies o 2.7% and
5.5%, we e ∼77% and 99% posi i e o IL-6, espec i ely (Figu e 1C, middle and igh
panels). In his ex- i o expe imen , whe e BM cells we e cul u ed up o 4 d in he p esence o
Tobío e al.,
6
SCF, cell lineages o he han mas cells (KIT+/Fc
ε
RI-, KIT-/Fc
ε
RI+ and KIT-/Fc
ε
RI-) also
showed posi i e in acellula IL-6 s aining (Online Supplemen a y Table 2). Howe e , he
highes IL-6 mean luo escence in ensi y (MFI) was associa ed wi h KIT+ cells. As SCF in
he media may induce IL-6 di ec ly o h ough au oc ine/pa ac ine signals in clonal and non-
clonal cells in hese cul u es, we examined he exp ession o IL-6 in BM biopsies by
immuno luo escence (IF). IF images o BM o pa ien s wi h SM indica ed ha IL-6
in acellula con en was mos ly associa ed wi h mas cells (Online Supplemen a y Figu e
S1). Thus, he combina ion o ex- i o and in si u expe imen s sugges s mas cells as he
p edominan p oduce s o IL-6, wi h a iable pa icipa ion o o he cell lineages. Da a a e
also consis en wi h an associa ion be ween inc eased IL-6 exp ession by BM mas cells and
D816V-KIT equency.
Exp ession o D816V-KIT causes IL-6 up egula ion and sec e ion
Gi en hese associa ions and ha mas cells and hei p ogeni o s o en ca y soma ic
D816V-KIT a ian s in SM, we in es iga ed he hypo hesis ha D816V-KIT exp ession
in insically p omo es IL-6 p oduc ion. Thus, we analyzed he p oduc ion o IL-6 a he
p o ein and message le els in a ious cell lines exp essing o no D816V-KIT. In ag eemen
wi h ou p e ious obse a ions, he HMC-1.2 mas ocy osis mas cell line which ha bo s KIT
wi h D816V plus ano he missense a ian in he jux a memb ane domain o KIT (V560G),
showed ma kedly highe IL-6 mRNA syn hesis13 (Figu e 2A, le panel) and IL-6 elease
han HMC-1.1 cells, which exp ess only he monoallelic V560G-KIT a ian (Figu e 2A,
igh panel).
The mas ocy oma mouse mas cell line P815 ca ying he homolog o he D816V-KIT a ian
(D814Y-KIT), also eleased signi ican ly mo e IL-6 han p ima y mouse BMMCs (Figu e
2B). Fu he mo e, exp ession o human D816V-KIT in an immo alized mouse mas cell line
ha lacks KIT 13, 19, unlike cells exp essing no mal KIT o ec o alone, eleased signi ican
amoun s o IL-6 in o he media (Figu e 2C). In agg ega e, he esul s using he a ious mas
cell lines demons a e an associa ion be ween exp ession o D816V-KIT in mas cells and
pe sis en ansc ip ion and elease o IL-6. This may no be es ic ed o mas cells, since
in oduc ion o D816V-KIT by CRISPR in he colo ec al ca cinoma cell line HCT116 also
p omo ed IL-6 ansc ip ion and elease (Figu e 2D le and igh panels, espec i ely) when
cells we e s imula ed wi h PMA and ionomycin, indica ing ha in hese cells D816V-KIT
p imes HCT116 cells o mo e obus IL-6 p oduc ion.
KIT y osine kinase ac i i y is equi ed o IL-6 p oduc ion in mas cells
Ligand-ac i a ed KIT signaling induces IL-6 p oduc ion in mas cells and enhances IgE-
ecep o -d i en IL-6 p oduc ion 20. Indeed, s imula ion o KIT by SCF in he LAD2 cell line,
which exp esses no mal KIT, induced IL-6 elease, albei he amoun s we e quan i a i ely
limi ed (Figu e 3A). HMC-1.1 showed highe p oduc ion o IL-6 han LAD2 cells,
pa icula ly when s imula ed wi h SCF (Figu e 3A). Howe e , no only was IL-6 p oduc ion
~100 old highe in uns imula ed HMC-1.2 han in LAD2 o HMC-1.1 cells, bu ligand-
Tobío e al.,
7
induced s imula ion o KIT did no cause any u he IL-6 sec e ion by HMC-1.2 cells
(Figu e 3A). These da a a e consis en wi h he conclusion ha ligand-independen signals
induced by oncogenic KIT ac i i y, pa icula ly hose om D816V-KIT, a e mo e e ec i e
in inducing IL-6 sec e ion han ligand-ac i a ed KIT signals.
We nex blocked D816V-KIT ac i i y in HMC-1.2 cells wi h dasa inib, a y osine kinase
inhibi o ha e ec i ely supp esses he ac i i y o he ac i e, open con o ma ion o D816V-
KIT. Dasa inib ma kedly educed IL-6 mRNA le els (Figu e 3B) and IL-6 sec e ion (Figu e
3C), while concen a ions o he y osine kinase inhibi o ima inib ha a e ine ec i e in
blocking D816V-KIT, did no al e IL-6 mRNA le els (Figu e 3B), u he sugges ing an
in ol emen o D816V-KIT signaling. Inhibi ion by ge i inib o he epide mal g ow h ac o
ecep o (EGFR), a y osine kinase ecep o ha can d i e IL-6 p oduc ion in ans o med
cells 21, had no signi ican e ec s on IL-6 mRNA le els in HMC-1.2 cells (Figu e 3B).
Simila o HMC-1.2 cells, dasa inib e ec i ely inhibi ed IL-6 elease in P815 mu ine
mas ocy oma cells (Figu e 3D). In con as , pe sis en IL-6 p oduc ion in HMC-1.2 cells did
no appea o be media ed ia eed- o wa d loops o ac i a ion by o he ecep o s as hose
epo ed o he IL-6R 22, sphingosine-1-phospha e ecep o s 23 o he TGF
β
ecep o 24 in
o he neoplas ic cells, since i was no al e ed by speci ic blockage o hese ecep o s (Online
Supplemen a y Figu e S2 A-C). O e all, hese da a sugges ha signals om D816V-KIT
can p omo e ligand-independen IL-6 p oduc ion wi hou in ol ing some o he mos
common au oc ine eed- o wa d loops desc ibed in malignan cells.
The cons i u i e elease o IL-6 by uns imula ed HMC-1.2 was enhanced by s imuli such as
complemen componen 5a (C5a), IL-1 β, 10% FBS and PMA/ionomycin (Online
Supplemen a y Figu e S3A, le panel) sugges ing ha he p oduc ion and sec e ion o IL-6
due o D816V-KIT can syne gize wi h o he complemen a y signals o en i onmen al cues.
O in e es , he in acellula con en o IL-6 p o ein was only ~10% o he o al IL-6 eleased
(Online Supplemen a y Figu e S3A), and al hough dasa inib did no a ec his pe cen age, i
also educed he o al in acellula con en . Thus, dasa inib inhibi ed he ansc ip ion,
in acellula con en and elease o IL-6, (Figu e 3B-C and Online Supplemen a y Figu e
S3A, igh panel), consis en wi h he conclusion ha D816V-KIT signals cause cons i u i e
de no o p oduc ion and elease o IL-6 wi hou egula ing s o age. In addi ion, we
demons a e ha IL-6 p o ein eleased by HMC-1.2 cells is biologically ac i e since
condi ioned media o hese cells caused IL-6 ecep o (IL-6R)-media ed STAT3
phospho yla ion in LAD2 cells which exp ess and espond o IL-6R ac i a ion 12 (Online
Supplemen a y Figu e S3B).
D816V-KIT-induced IL-6 p oduc ion is dependen on JAK2, ERK and PI3K pa hways
MAPKs (ERK1/2 and p38) and PI3K pa hways a e pa o he oncogenic signals de i ed
om D816V-KIT ac i i y 2, 25. Since hese pa hways may a ec IL-6 exp ession 26-28, we
in es iga ed hei po en ial oles in D816V-KIT-induced IL-6 p oduc ion. While inhibi ion o
p38 wi h SB203580 had mino e ec s on IL-6 syn hesis, inhibi ion o he ERK1/2 pa hway
using a MEK1/2 inhibi o (U0126) (Figu e 4A) o inhibi ion o he PI3K pa hway by he
Tobío e al.,
8
PI3Kα/δ/β inhibi o LY294002 (Figu e 4B), caused a 50 o 60% educ ion, espec i ely, in
IL-6 p oduc ion a he message (le panels) and p o ein le els ( igh panels).
The JAK/STAT axis is also known o be p ominen ly up egula ed by D816V-KIT ac i i y 29,
30. JAK2 is ac i a ed by SCF 31 leading o STAT phospho yla ion and ansloca ion in o he
nucleus, whe e i exe s i s ansc ip ional ac i i y 32. Inhibi ion o JAK2 by he JAK2
selec i e inhibi o ed a inib (TG101348) ma kedly blocked IL-6 cons i u i e ansc ip ion
and cy okine elease (Figu e 4C, le and igh panels, espec i ely). Ruxoli inib which
inhibi s JAK1 in addi ion o JAK2, simila ly inhibi ed IL-6 exp ession, al hough a highe
concen a ions han ed a inib (Figu e 4D). Howe e , o aci inib, a pan-JAK inhibi o
p e e en ial o JAK3 and o a lesse ex en JAK1, was less e ec i e (Figu e 4D). Simila o
HMC-1.2 cells, in mouse P815 cells inhibi ion o MEK/ERK1/2, PI3K o JAK2 pa hways
ma kedly educed IL-6 elease in mouse P815 cells (Figu e 4E). The da a hus implica e
JAK2 in D816V-KIT induced IL-6 p oduc ion.
Since JAK2 ac i a ion has also been epo ed in ce ain cells o ac i a e PI3K o ERK 32, 33,
we u he in es iga ed he po en ial in e - ela ionships be ween JAK2 and he ERK and
PI3K pa hways. While inhibi ion o KIT by dasa inib, as expec ed, blocked he
phospho yla ion o JAK2, AKT and ERK1/2 signaling pa hways downs eam o he ecep o
by 40 o 80% (Figu e 4F), inhibi o s o JAK2, PI3K/AKT o MEK1/2/ERK1/2 educed
phospho yla ion o hei espec i e a ge s, bu did no show signi ican e ec s on any o he
o he s, sugges ing ha hese signals a e ac i a ed independen ly om each o he .
D816V-KIT-induced IL-6 p oduc ion is dependen on STAT5
JAK phospho yla es STATs, and STAT amily membe s such as STAT3 34, 35, STAT4 36 and
STAT5 37 ha e been implica ed in he egula ion o IL-6 ansc ip ion in a ious cells and
condi ions. As he le els o exp ession o phospho yla ion o STAT3, STAT4 and STAT5
(Online Supplemen a y Figu es S4 A-C) a e inc eased in HMC-1.2 and o he cells wi h
D816V-KIT 30; and STAT4 and STAT5 a e up egula ed in BM mas cells om pa ien s wi h
SM 29, 38, 39, we in es iga ed hei possible in ol emen in he induc ion o IL-6 ansc ip ion
by silencing STAT3-5 exp ession using sh-RNA o si-RNA.
Sh-RNA-media ed STAT3 silencing esul ed in >75% educ ion in STAT3 a he messenge
and p o ein le els (Figu e 5A) bu did no a ec IL-6 p oduc ion by HMC-1.2 (Figu e 5A,
ed ba ). Simila ly, a selec i e small inhibi o molecule o STAT3, C188-9, a
concen a ions ha caused >80% educ ion in STAT3 phospho yla ion (Online
Supplemen a y Figu e S5A, uppe panel) did no al e cons i u i e IL-6 p oduc ion by hese
cells (Online Supplemen a y Figu e S5A, lowe panel). Neu alizing an ibodies o he IL-
6R, which signals h ough STAT3, we e also ine ec i e on IL-6 p oduc ion (Online
Supplemen a y Figu e S2A). Reduc ion o STAT4 message and p o ein by >50% using sh-
RNA knockdown was also inconsequen ial o IL-6 pe sis en p oduc ion by hese cells
(Figu e 5B).
Tobío e al.,
15
D816V mu a ion bu den in adul s wi h indolen sys emic mas ocy osis. Eu J Haema ol.
2013;91(2):106-111.
19. Sm z D, Banda a G, Zhang S, e al. A no el KIT-de icien mouse mas cell model
o he examina ion o human KIT-media ed ac i a ion esponses. J Immunol Me hods.
2013;390(1-2):52-62.
20. I o T, Sm z D, Jung MY, e al. S em cell ac o p og ams he mas cell ac i a ion
pheno ype. J Immunol. 2012;188(11):5428-5437.
21. Gao SP, Ma k KG, Leslie K, e al. Mu a ions in he EGFR kinase domain media e
STAT3 ac i a ion ia IL-6 p oduc ion in human lung adenoca cinomas. J Clin In es .
2007;117(12):3846-3856.
22. G i enniko S, Ka in M. Au oc ine IL-6 signaling: a key e en in umo igenesis?
Cance Cell. 2008;13(1):7-9.
23. Lee H, Deng J, Kujawski M, e al. STAT3-induced S1PR1 exp ession is c ucial o
pe sis en STAT3 ac i a ion in umo s. Na Med. 2010;16(12):1421-1428.
24. Yao Z, Fenoglio S, Gao DC, e al. TGF-be a IL-6 axis media es selec i e and
adap i e mechanisms o esis ance o molecula a ge ed he apy in lung cance . P oc
Na l Acad Sci U S A. 2010;107(35):15535-15540.
25. Ha i N, Boudo C, F iedbichle K, e al. Oncogenic Ki con ols neoplas ic mas
cell g ow h h ough a S a 5/PI3-kinase signaling cascade. Blood. 2008;112(6):2463-
2473.
26. Guo X, Ge l RE, Sch ade JW. De ining he in ol emen o p38alpha MAPK in he
p oduc ion o an i- and p oin lamma o y cy okines using an SB 203580- esis an o m
o he kinase. J Biol Chem. 2003;278(25):22237-22242.
27. Kopan ze Y, Helle M, S wamina han N, Rudiko S. IL-6 media ed ac i a ion o
STAT3 bypasses Janus kinases in e minally di e en ia ed B lineage cells. Oncogene.
2002;21(44):6791-6800.
28. Li J, Lan T, Zhang C, e al. Recip ocal ac i a ion be ween IL-6/STAT3 and
NOX4/Ak signalings p omo es p oli e a ion and su i al o non-small cell lung cance
cells. Onco a ge . 2015;6(2):1031-1048.
29. G imwade LF, Happe ield L, T is am C, e al. Phospho-STAT5 and phospho-Ak
exp ession in ch onic myelop oli e a i e neoplasms. B J Haema ol. 2009;147(4):495-
506.
30. Lasho T, Te e i A, Pa danani A. Inhibi ion o JAK-STAT signaling by TG101348: a
no el mechanism o inhibi ion o KITD816V-dependen g ow h in mas cell leukemia
cells. Leukemia. 2010;24(7):1378-1380.
31. Mo ales JK, Falanga YT, Depc ynski A, Fe nando J, Ryan JJ. Mas cell homeos asis
and he JAK-STAT pa hway. Genes Immun. 2010;11(8):599-608.
32. Rane SG, Reddy EP. Janus kinases: componen s o mul iple signaling pa hways.
Oncogene. 2000;19(49):5662-5679.
33. Jain N, Zhang T, Fong SL, Lim CP, Cao X. Rep ession o S a 3 ac i i y by ac i a ion
o mi ogen-ac i a ed p o ein kinase (MAPK). Oncogene. 1998;17(24):3157-3167.
34. Chang Q, Daly L, B ombe g J. The IL-6 eed- o wa d loop: a d i e o
umo igenesis. Semin Immunol. 2014;26(1):48-53.
35. Yoon S, Woo SU, Kang JH, e al. NF-kappaB and STAT3 coope a i ely induce IL6
in s a ed cance cells. Oncogene. 2012;31(29):3467-3481.
36. Nguyen HN, Noss EH, Mizoguchi F, e al. Au oc ine Loop In ol ing IL-6 Family
Membe LIF, LIF Recep o , and STAT4 D i es Sus ained Fib oblas P oduc ion o
In lamma o y Media o s. Immuni y. 2017;46(2):220-232.

Tobío e al.,
16
37. Ba ns ein BO, Li G, Wang Z, e al. S a 5 exp ession is equi ed o IgE-media ed
mas cell unc ion. J Immunol. 2006;177(5):3421-3426.
38. Teodosio C, Ga cia-Mon e o AC, Ja a-Ace edo M, e al. Gene exp ession p o ile o
highly pu i ied bone ma ow mas cells in sys emic mas ocy osis. J Alle gy Clin
Immunol. 2013;131(4):1213-1224, 1
224 e1-4.
39. Baumga ne C, Ce ny-Rei e e S, Sonneck K, e al. Exp ession o ac i a ed STAT5
in neoplas ic mas cells in sys emic mas ocy osis: subcellula dis ibu ion and ole o
he ans o ming oncop o ein KIT D816V. Am J Pa hol. 2009;175(6):2416-2429.
40. Decke T, Ko a ik P. Se ine phospho yla ion o STATs. Oncogene.
2000;19(21):2628-2637.
41. B ombe g J, Wang TC. I n lamma ion and cance : IL-6 and STAT3 comple e he
link. Cance Cell. 2009;15(2):79-80.
42. Kawashima T, Mu a a K, Aki a S, e al. STAT5 induces mac ophage
di e en ia ion o M1 leukemia cells h ough ac i a ion o IL-6 p oduc ion media ed by
NF-kappaB p65. J Immunol. 2001;167(7):3652-3660.
43. Xu R, Nelson CM, Muschle JL, Veiseh M, Vonde haa BK, Bissell MJ. Sus ained
ac i a ion o STAT5 is essen ial o ch oma in emodeling and main enance o
mamma y-speci ic unc ion. J Cell Biol. 2009;184(1):57-66.
44. Bibi S, A slanhan MD, Langen eld F, e al. Co-ope a ing STAT5 and AKT signaling
pa hways in ch onic myeloid leukemia and mas ocy osis: possible new a ge s o
he apy. Haema ologica. 2014;99(3):417-429.
45. Shelbu ne CP, McCoy ME, Pieko z R, e al. S a 5 exp ession is c i ical o mas cell
de elopmen and su i al. Blood. 2003;102(4):1290-1297.
46. Nie e gall E, Reynolds J, Kok CH, e al. TGF-alpha and IL-6 plasma le els
selec i ely iden i y CML pa ien s who ail o achie e an ea ly molecula esponse o
p og ess in he i s yea o he apy. Leukemia. 2016;30(6):1263-1272.
47. Welne RS, Amabile G, Ba a ia D, e al. T ea men o ch onic myelogenous
leukemia by blocking cy okine al e a ions ound in no mal s em and p ogeni o cells.
Cance Cell. 2015;27(5):671-681.
48. Nelson EA, Walke SR, Weisbe g E, e al. The STAT5 inhibi o pimozide dec eases
su i al o ch onic myelogenous leukemia cells esis an o kinase inhibi o s. Blood.
2011;117(12):3421-3429.
49. F iedbichle K, Ke enyi MA, Ko acic B, e al. S a 5a se ine 725 and 779
phospho yla ion is a p e equisi e o hema opoie ic ans o ma ion. Blood.
2010;116(9):1548-1558.
50. Pi che TJ, Pe e sen H, Gus a sson JA, Haldosen LA. Ex acellula signal- egula ed
kinase (ERK) in e ac s wi h signal ansduce and ac i a o o ansc ip ion (STAT) 5a.
Mol Endoc inol. 1999;13(4):555-565.
Tobío e al.,
17
Figu e legends
Figu e 1. P oduc ion o IL-6 by bone ma ow cells and mas cells is enhanced in
pa ien s wi h SM in associa ion wi h he D816V-KIT allelic bu den. (A) IL-6 elease
om BM mononuclea cells isola ed om i e pa ien s wi h SM wi h D816V-KIT bone
ma ow allelic equencies o <5 (pa ien s 4 and 5 in Supplemen a y Table 1) o >5 (pa ien s
6, 7, and 10). BM aspi a es we e cul u ed o 2-4 days and IL-6 eleased in o he cul u e
media was measu ed by ELISA. Da a a e he mean ± SEM. (B) Co ela ion be ween IL-6
eleased in o he media by BM cells and he pe cen age o mas cells in hose cul u es which
was de e mined by low cy ome y. (C) Flow cy ome y his og ams showing in acellula IL-
6 s aining in BM mas cells. Mas cells we e ga ed as CD3-/CD34-/KIT+/Fc
ε
RI+ wi hin he
BM cells o h ee pa ien s (pa ien s 1-3 in Supplemen a y Table 1), wi h D816V-KIT allelic
equencies in he bone ma ow also indica ed in he igu e. Pa ien 1had idiopa hic
anaphylaxis and did no mee c i e ia o SM bu was used as a con ol. The pe cen age o IL-
6 posi i e cells wi hin he mas cell popula ion is indica ed in he his og ams.
Figu e 2. Cells wi h D816V-KIT cons i u i ely exp ess and elease IL-6. (A) IL-6 mRNA
exp ession (le ) and IL-6 eleased in o he media ( igh ) by he mas ocy osis cell lines HMC-
1.1 (wi h V560G) and HMC-1.2 (wi h V560G and D816V) a e 2 h in se um ee media. (B-
C) Compa ison o IL-6 eleased in o he media by he mouse P815 mas ocy oma mas cell
line (wi h D814Y-KIT) compa ed o no mal mu ine BMMCs (B), and by he mu ine mas
cell line MCBS-1 (which lacks c-Ki ) ans ec ed wi h human KIT o D816V-KIT compa ed
o MCBS-1 ans ec ed wi h ec o alone (C). (D) Compa ison o IL-6 mRNA exp ession
(le ) and IL-6 eleased in o he media ( igh ) by he human colo ec al ca cinoma cell line
HCT116 exp essing o no D816V-KIT. HCT116 cells we e s imula ed wi h PMA and
ionomycin o e nigh . IL-6 mRNA exp ession was de e mined by q-RT-PCR and ela i e
exp ession was calcula ed in ela ionship o he exp ession o GAPDH using he ΔC me hod
and exp essed as old change compa ed o HMC-1.1 (A, le panel), o he HCT116 pa en al
cell line (D, le panel). All da a (A-D) a e he mean ± SEM o h ee independen expe imen s
done in iplica es.
Figu e 3. D816V-KIT oncogenic ac i i y d i es ligand-independen IL-6 induc ion. (A)
IL-6 eleased o he ex acellula media was measu ed in LAD2, HMC-1.1 and HMC-1.2
cells incuba ed o 48 h (37C, 5%CO2) in se um ee medium in he p esence o absence o
100 ng/mL SCF. (B) IL-6 mRNA le els we e measu ed in HMC-1.2 a e 2 h incuba ion in
se um ee medium in he p esence o absence o he KIT inhibi o s, dasa inib and ima inib,
o he EGFR inhibi o ge i inib a he indica ed concen a ions. Rela i e exp ession o IL-6
mRNA was ob ained by no malizing o he exp ession o GAPDH using he ΔC me hod and
he esul s a e exp essed as old change compa ed o un ea ed cells. The e ec o he KIT
inhibi o dasa inib (0.5 µM) on he sec e ion o IL-6 by HMC-1.2 (C) o by P815 mas cells
(D) was de e mined a e 6 h incuba ion in se um ee medium. All da a a e he mean ± SEM
o h ee independen expe imen s done in iplica es.
Tobío e al.,
18
Figu e 4. MEK/ERK-, PI3K- and JAK2-media ed pa hways a e independen ly
ac i a ed by D816V-KIT and con ibu e o ligand-independen IL-6 induc ion. (A-B)
E ec o inhibi ion o MEK/ERK1/2 (U0126), p38 (SB203580) (A) o PI3K pa hways
(LY294002) (B) on he exp ession o IL-6 mRNA (le panels) a e ea men o 2 h and he
elease o IL-6 in o he media ( igh panels) by HMC-1.2 cells o 16 h. (C) E ec o he
JAK2 inhibi o ed a inib a he indica ed concen a ions on he exp ession o IL-6 mRNA
(le ) and he elease o IL-6 in o he media ( igh ) by HMC-1.2 cells. (D) E ec o a ious
concen a ions o inhibi o s o JAK1/2 ( uxoli inib) and JAK3 ( o aci inib) on he exp ession
o IL-6 mRNA by HMC-1.2 cells. E) E ec o inhibi ion o MEK/ERK1/2, PI3K o JAK2 on
he p oduc ion o IL-6 by mouse P815 cells a e 6 h o incuba ion. (F) E ec o inhibi o s
o KIT (KIT-I: dasa inib; 0.5 µM), JAK2 (JAK2-I: ed a inib; 1 µM), PI3K (PI3K-I:
LY294002; 10 µM) and MEK/ERK1/2 (ERK-I: U0126; 10 µM) on he phospho yla ion o
KIT, JAK2, AKT and ERK1/2 in he indica ed amino acid esidues. In F, inhibi o s we e
incuba ed o 2 h in se um ee medium. Numbe s unde each phospho yla ed band a e mean
± SEM o a leas h ee expe imen s and ep esen old changes in he ela i e band
luo escence (no malized o he co esponding o al exp ession) compa ed o un ea ed cells.
Rela i e exp ession o IL-6 mRNA was ob ained by compa ing o he exp ession o GAPDH
using he ΔC me hod and he esul s we e exp essed as old change compa ed o un ea ed
cells. Da a ep esen he mean ± SEM and a e om h ee independen expe imen s.
Figu e 5. STAT5A and B a e key in media ing D816V-KIT-induced pe sis en IL-6
p oduc ion. E ec o STAT3 (A) and STAT4 (B) knockdown by len i i al sh-RNA on hei
espec i e a ge s and on he exp ession o IL-6 mRNA by HMC-1.2 cells. Fo sh-STAT4,
he e ec o wo di e en cons uc s a e shown (#1 and #3). Wes e n blo gels unde nea h he
ba g aphs ep esen he e ec o sh-STAT3 (A) o sh-STAT4 (B) cons uc s on he p o ein
le els o STAT3 and STAT4, espec i ely. Lysa es om duplica e samples a e shown. The
numbe s unde each pai a e mean ± SEM o a leas 3 sepa a e expe imen s and ep esen
old change in he ela i e band luo escence compa ed o he sh-RNA non- a ge (con ol).
Ac in con en was used as loading con ol. (C, D) E ec o silencing STAT5A, STAT5B o
he combina ion o bo h STAT5A&B by si-RNA on he exp ession o STAT5A and STAT5B
mRNA (C) and o IL-6 mRNA (D) in HMC-1.2. The blo s unde he ba g aph in C show he
e ec o STAT5 silencing in he p o ein le els o STAT5A/B and he numbe s unde he
blo , he old change as in A and B. O no e, he an ibody used ecognizes bo h STAT5A and
B.
Figu e 6. MEK/ERK signaling pa hway coope a es wi h JAK2 in he egula ion o
STAT5 o induce IL-6 while PI3K egula es IL-6 independen ly o STAT5. (A) E ec o
inhibi o s o KIT (KIT-I: dasa inib; 0.5 µM), JAK2 (JAK2-I: ed a inib; 1 µM), PI3K
(PI3K-I: LY294002; 10 µM) MEK/ERK1/2 (ERK-I: U0126; 10 µM), STAT5 (STAT5-I:
CAS285986-31-4; 50 µM), o hei combina ion on he p o ein and phospho yla ion le els o
STAT5 in HMC-1.2 cells. (B) E ec o inhibi o s o JAK2 and ERK1/2 alone o in
combina ion on STAT5 phospho yla ion in HMC-1.2 cells. In A and B, lysa es we e ob ained
a e 2 h incuba ion wi h he indica ed inhibi o s in se um ee medium. Da a unde he blo s
a e he mean ± SEM o h ee independen expe imen s done in duplica es and ep esen old
Tobío e al.,
19
change in he ela i e band luo escence compa ed o he un ea ed. STAT5 con en was used
o no malize he da a. (C) Inhibi ion o MEK/ERK1/2 educes STAT5A and STAT5B
exp ession (uppe and lowe panels, espec i ely) a e 16 h incuba ion. Rela i e exp ession
o STAT5 mRNA was ob ained by compa ing o he exp ession o GAPDH using he ΔC
me hod and he esul s a e exp essed as old change compa ed o un ea ed cells a each ime
(2 h o 16 h). (D, E) Inhibi o s o STAT5, PI3K and MEK/ERK1/2 addi i ely p e en IL-6
mRNA induc ion (D, le panel) and IL-6 elease in HMC-1.2 cells (D, igh panel) and P815
cells (E). The elease o IL-6 in o he media was de e mined a e 6 h. (F) Inhibi o s o
JAK2, PI3K and MEK/ERK1/2 addi i ely p e en IL-6 mRNA induc ion in HMC-1.2 cells.
All da a ep esen mean ± SEM o a leas h ee independen expe imen s.

Tobío e al., 6
RNA “ON-TARGET” pool o human STAT5A (L-005169-00-0005), STAT5B (L-
010539-00-0005), o a non- a ge ing si-RNA pool (D-001810-10-05) om Dha macon
(La aye e, CO). T ans ec ed cells we e kep in cul u e a 37oC in 5% CO2 o 72 h. Cells
we e hen washed, incuba ed in se um ee cul u e media o 2 h and hei cellula p o ein
o RNA con en s ex ac ed.

Tobío e al., 7
Supplemen a y Table 1- D816V-KIT allelic equency and yp ase le els in he
pa ien s used in his s udy
Pa ien
numbe
Diagnosis
Sex
BM D816V-KIT
F equency (%)
Se um T yp ase
(ng/mL)
1
IA*
F
Nega i e
4
2
ISM
M
2.74
35
3
ISM
M
5.50
182
4
ASM
F
1.41
479
5
ISM
F
0.35
105
6
ISM
F
31.08
293
7
ISM
M
13.44
291
8
ISM
M
0.06
14
9
ISM
M
0.03
10
10
SSM
M
44.66
590
*BM aspi a es we e ob ained om his pa ien wi h idiopa hic anaphylaxis (IA) du ing an e alua ion o
possible SM. SM was no diagnosed in his pa ien and hus indings we e used as a con ol in his s udy.
ISM: indolen sys emic mas ocy osis; SSM: smoulde ing sys emic mas ocy osis. Pa ien s we e classi ied
acco ding o he WHO c i e ia 8-11
Tobío e al., 8
Supplemen a y Table 2- In acellula IL-6 s aining o bone ma ow cells in pa ien s
wi h SM
Pa ien /
D816V-KIT
F equency
(%)
KIT+/Fc
e
RI+
KIT+/Fc
e
RI-
KIT-/Fc
e
RI+
KIT-/Fc
e
RI-
IL-6
MFI
%
IL6+cells
IL-6
MFI
%
IL6+cells
IL-6
MFI
%
IL6+cells
IL-6
MFI
%
IL6+cells
1/ (0.0%)
-
21
-
17
-
20
-
17
2/ (2.7%)
336
74
167
56
254
61
117
47
3/ (5.5%)
3371
98
768
88
524
86
522
73
BM cells we e cul u ed o 3-4 d and in acellula IL-6 s aining in B e eldin A- ea ed cells was analyzed by
FACS. CD3-/CD34- cells we e ga ed and he mean luo escence in ensi y (MFI) o in acellula IL-6 and he
pe cen age o cells wi h IL-6+ s aining de e mined wi hin each popula ion (KIT+/FcεRI+, KIT+/FcεRI-, KIT-
/FcεRI+ and KIT-/FcεRI-). Pa ien s 1-3 a e he same as shown in Figu e 1C and as de ined in Supplemen a y
Table 1. Pa ien 1 had idiopa hic anaphylaxis and did no mee c i e ia o SM bu was used as a con ol.
Tobío e al., 9
Supplemen a y Figu e S1. In acellula IL-6 in bone ma ow specimens om
pa ien s wi h SM is mainly associa ed wi h mas cells. Bone ma ow biopsies om
pa ien s # 7 (A) and #10 (C) de ined in Supplemen a y Table 1 we e p epa ed o
immuno luo escence s aining wi h an i-mas cell yp ase ( ed) and an i-IL-6 (g een).
Nuclei we e iden i ied by DAPI s aining. In he o e lay, p oximal co-localiza ion is shown
in yellow o whi e depending on he in ensi y o he indi idual an i- yp ase and an i-IL-6
s aining. Shown a e a eas wi h dispe se mas cell in il a ion o be e isualiza ion. Scale
ba s a e 50 µM (le panels) and 20 µM ( igh panels). Righ panels a e he magni ied a eas
indica ed by whi e squa es in he le panels.
An i-IL-6 An i- yp ase O e lay+DAPI
Pa ien #7
Pa ien #10
A
B
20 µM
20 µM
Tobío e al., 10
Supplemen a y Figu e S2- Au oc ine ac i a ion o IL-6, S1P and TGF-β ecep o s
does no media e he cons i u i e exp ession o IL-6 in HMC-1.2 cells. E ec o an i-
IL-6 ecep o (IL-6R) an ibody ( ocilizumab; 100 µg/mL) (A, le panel), an i-gp130
an ibody (1 µg/mL) (A, igh panel), S1PR1,3, S1PR2 and S1PR4 an agonis s (VPC 23019;
1 µM, JTE 013; 1 µM and CYM 50358; 1 µM) (B) and a TGFβ-R an agonis (SD 208; 5
µM) (C), on IL-6 sec e ion by HMC-1.2 cells. The an ibodies and an agonis s we e
incuba ed o 24 h in se um ee media. The da a ep esen mean ± SEM o wo expe imen s
pe o med in duplica e.
Tobío e al., 11
Supplemen a y Figu e S3. Cons i u i e sec e ion o IL-6 by HMC-1.2 is enhanced by
a ious s imuli; and sec e ed IL-6 is biologically ac i e. (A) To al IL-6 eleased in o he
media by HMC-1.2 in compa ison wi h o al in acellula IL-6. Cells we e incuba ed a
37oC o e nigh in se um ee media wi h o wi hou C48/80 (500 ng/mL), C5a (500
ng/mL), PMA (20 ng/mL) and Ionomycin (Io) (1 µM), LPS (10 µg/mL), IL-1b (100
ng/mL), o ull media con aining 10% FBS. In he igh panel, he e ec s o he KIT
inhibi o dasa inib (0.5 µM) on eleased IL-6 and in acellula IL-6 a e shown. The
pe cen ages o in acellula compa ed o eleased IL-6 we e 10% wi h o wi hou dasa inib,
o in he p esence o C48/80; 6% a e s imula ion wi h C5a, LPS and PMA/Io; and 2%
a e incuba ion wi h IL-1b o 10% FBS. (B) IL-6 eleased by HMC-1.2 induces STAT3
phospho yla ion in LAD2 cells. LAD2 cul u es we e ea ed o 30 min wi h condi ioned
media om HMC-1.2 (HMC-1.2-C.M.) in he p esence o absence o he neu alizing IL-
6R an ibody ocilizumab (100 µg/mL) o incuba ed wi h se um- ee media wi h o wi hou
50 ng/mL ecombinan IL-6. Cell lysa es we e ob ained and phospho yla ion o STAT3
de e mined by Wes e n blo ing.

Tobío e al., 12
Supplemen a y Figu e S4. Baseline exp ession le els and phospho yla ion s a e o
STAT amily membe s in HMC-1.2 compa ed o HMC-1.1 cells. (A) Exp ession le els
o STAT3 mRNA (uppe panel) and p o ein (lowe panel) in HMC-1.1 and HMC-1.2 cells.
(B) Exp ession le els o STAT4 mRNA (uppe panel) and p o ein (lowe panel) in HMC-
1.1 and HMC-1.2 cells. (C) Exp ession le els o STAT5 mRNA (le panel, STAT5A;
middle panel, STAT5B) and p o ein ( igh panel) in HMC-1.1 and HMC-1.2 cells. Bo h
phospho yla ed and o al STAT amily membe s a e shown. Rela i e exp ession o STAT
mRNA was ob ained by compa ing o he exp ession o GAPDH using he DC me hod and
he esul s we e exp essed as old change compa ed o HMC-1.1 cells. Cells we e incuba ed
o 2 h in se um- ee media be o e ob aining he cell lysa es o RNA. The da a ep esen
mean ± SEM o h ee expe imen s pe o med in duplica e.
Tobío e al., 13
Supplemen a y Figu e S5. Inhibi ion o STAT5 bu no STAT3 educes cons i u i e
exp ession o IL-6 by HMC-1.2 cells. (A) E ec o he small STAT3 inhibi o , C188-9,
a he indica ed concen a ions, on STAT3 phospho yla ion (uppe panel) and he
exp ession o IL-6 mRNA (lowe panel). (B-C) E ec o he STAT5 inhibi o CAS285986-
31-4, a he indica ed concen a ions, on STAT5 phospho yla ion (B, uppe panel), he
exp ession o IL-6 mRNA (B, lowe panel) and IL-6 p o ein (C) in HMC-1.2 cells. The
exp ession o IL-6 p o ein in C was e alua ed by in acellula s aining and FACS analysis
as desc ibed in Me hods. (D) E ec o he STAT5 inhibi o on IL-6 sec e ion by P815 cells.
In A-D, cells we e incuba ed in se um- ee media wi h he co esponding inhibi o s o 2 h
(IL-6 mRNA exp ession) o 6 h (IL-6 elease). (E) Combined e ec o knocking down bo h
STAT5A and B by si-RNA and incuba ion wi h he STAT5 inhibi o o 2 h, a he
indica ed concen a ions, on he cons i u i e exp ession o IL-6 in HMC-1.2. Rela i e
exp ession o IL-6 mRNA was ob ained by compa ing o he exp ession o GAPDH using
he DC me hod and he esul s we e exp essed as old change compa ed o un ea ed o
cell ea ed wi h si-non- a ge con ol. The da a ep esen mean ± SEM o a leas wo
indi idual expe imen s.
0.0
0.5
1.0
1.5
Rela i e IL-6 mRNA
010 15
STAT3-I (µM)
(C188-9)
020 50
0.0
0.5
1.0
1.5
Rela i e IL-6 mRNA
****
STAT5-I (µM)
(CAS 285986-31-4)
0
50
100
150
IL-6 elease (pg/mL)
***
050
STAT5-I (µM)
(CAS 285986-31-4)
P815 cells
A B
0.0
0.5
1.0
1.5
Rela i e IL-6 mRNA
STAT5A&B siRNA
STAT5-I (µM)
(CAS 285986-31-4)
si-Con ol
0 20 50
HMC-1.2
C D E
HMC-1.2
MFI
Uns ained 74
Vehicle 666
STAT5-I 442
10.3±0.1
HMC-1.2
0 10 15 (µM)
pSTAT3
(Y705)
STAT3
STAT3-I :
KDa
86
86
0.1±0.1
β Ac in46
pSTAT5
(Y694)
STAT5
β Ac in
1 0.6±0.04
HMC-1.2
0.5±0.1
46
90
90
0 10 15 (µM)
STAT5-I :
KDa
Tobío e al., 14
Supplemen a y Re e ences
1. Bu e ield JH, Weile D, Dewald G, Gleich GJ. Es ablishmen o an imma u e mas
cell line om a pa ien wi h mas cell leukemia. Leuk Res. 1988;12(4):345-355.
2. Tobio A, Al onso A, Bo ana LM. C oss- alks be ween c-Ki and PKC iso o ms in
HMC-1560 and HMC-1560,816 cells. Di e en ole o PKCδ in each cellula line. Cellula
Immunology. 2015;293(104-112.
3. Sunds om M, Vliago is H, Ka lbe g P, e al. Func ional and pheno ypic s udies o wo
a ian s o a human mas cell line wi h a dis inc se o mu a ions in he c-ki p o o-oncogene.
Immunology. 2003;108(1):89-97.
4. Sm z D, Banda a G, Zhang S, e al. A no el KIT-de icien mouse mas cell model o
he examina ion o human KIT-media ed ac i a ion esponses. J Immunol Me hods.
2013;390(1-2):52-62.
5. Kuehn HS, Bea en MA, Ma HT, Kim MS, Me cal e DD, Gil illan AM. Syne gis ic
ac i a ion o phospholipases Cgamma and Cbe a: a no el mechanism o PI3K-independen
enhancemen o FcepsilonRI-induced mas cell media o elease. Cell Signal. 2008;20(4):625-
636.
6. Ki shenbaum AS, Akin C, Wu Y, e al. Cha ac e iza ion o no el s em cell ac o
esponsi e human mas cell lines LAD 1 and 2 es ablished om a pa ien wi h mas cell
sa coma/leukemia; ac i a ion ollowing agg ega ion o FcepsilonRI o FcgammaRI. Leuk Res.
2003;27(8):677-682.
7. Kim DK, Bea en MA, Me cal e DD, Oli e a A. In e ac ion o DJ-1 wi h Lyn is
essen ial o IgE-media ed s imula ion o human mas cells. J Alle gy Clin Immunol.
2018;142(1):195-206 e198.
8. Ho ny HP, Me cal e DD, Akin C, e al. Mas ocy osis. In: Swe dlow SH, Campo E,
Ha is NL, Ja ee ES, Pile i SA, S ein H, e al., eds. WHO classi ica ion o umou s o
haema opoie ic and lymphoid issues. 2017 ed. Lyon, F ance: IARC P ess, 2017:62-69.
9. Valen P, Akin C, Ha mann K, e al. Ad ances in he Classi ica ion and T ea men o
Mas ocy osis: Cu en S a us and Ou look owa d he Fu u e. Cance Res. 2017;
10. Valen P, Akin C, Me cal e DD. Mas ocy osis: 2016 upda ed WHO classi ica ion and
no el eme ging ea men concep s. Blood. 2017;129(11):1420-1427.
11. Valen P, Ho ny HP, Esc ibano L, e al. Diagnos ic c i e ia and classi ica ion o
mas ocy osis: a consensus p oposal. Leuk Res. 2001;25(7):603-625.