1
Vol.:(0123456789)
Scien i ic Repo s | (2021) 11:9873 | h ps://doi.o g/10.1038/s41598-021-87968-4
www.na u e.com/scien i ic epo s
Edel osine nanoemulsions inhibi
umo g ow h o iple nega i e
b eas cance in zeb a ish
xenog a model
So ia M. Sa ai a1,3, Ca lha Gu ié ez‑Lo e a2, Jeanne e Ma ínez‑Val2, Sainza Lo es1,
Belén L. Bouzo1, Sand a Díez‑Villa es1,3, Sand a Alijas1, Alba Pensado‑López2,4,
Abi Judi Vázquez‑Ríos1,3, Lau a Sánchez2 & Ma ía de la Fuen e1,3*
T iple nega i e b eas cance (TNBC) is known o being e y agg essi e, he e ogeneous and highly
me as a ic. The s anda d o ca e ea men is s ill chemo he apy, wi h adjacen oxici y and low
e icacy, highligh ing he need o al e na i e and mo e e ec i e he apeu ic s a egies. Edel osine, an
alkyl‑lysophospholipid, has p o ed o be a p omising he apy o se e al cance ypes, upon deli e y
in lipid nanopa icles. The e o e, he objec i e o his wo k was o explo e he po en ial o edel osine
o he ea men o TNBC. Edel osine nanoemulsions (ET‑NEs) composed by edel osine, Miglyol 812
and phospha idylcholine as excipien s, due o hei good sa e y p o ile, p esen ed an a e age size
o abou 120 nm and a neu al ze a po en ial, and we e s able in bio ele an media. The abili y o
ET‑NEs o in e up umo g ow h in TNBC was demons a ed bo h in i o, using a highly agg essi e
and in asi e TNBC cell line, and in i o, using zeb a ish emb yos. Impo an ly, ET‑NEs we e able o
pene a e h ough he skin ba ie o MDA‑MB 231 xenog a ed zeb a ish emb yos, in o he yolk
sac, leading o an e ec i e dec ease o highly agg essi e and in asi e umo al cells’ p oli e a ion.
Al oge he he esul s demons a e he po en ial o ET‑NEs o he de elopmen o new he apeu ic
app oaches o TNBC.
Cance is one o he majo heal h p oblems wo ldwide due o i s high a e o mo bidi y and mo ali y. Among
he di e en ypes o cance , b eas cance is he second mos p e alen , wi h an es ima ion o wo million new
cases in 2018, and is he leading cause o dea h in women wo ldwide1. I ep esen s a he e ogeneous g oup o
umo s ha is cu en ly classi ied in i e di e en sub ypes acco ding o hei his ological and molecula pa e ns2.
Among hese sub ypes, TNBC accoun s o 15–20% o he cases. TNBC lacks o he exp ession o es ogen and
p oges e one ecep o s and human epide mal g ow h ac o 2 ecep o (HER2). This sub ype is known o i s
he e ogenei y, immunogenici y and agg essi e biological beha io , high me as a ic a e, and complex me as a ic
pa e ns, al oge he leading o a high equency o elapse and low su i al a es2,3.
Conside ing he lack o e icien a ge ed he apies, chemo he apy is s ill he mos commonly used ea -
men o TNBC2,3. Howe e , chemo he apeu ic d ugs a e well known o being esponsible o se e al ad e se
e ec s in pa ien s due o hei poo accumula ion in he umo si e and me aboliza ion be o e eaching i , which
augmen hei in insic high oxic p o ile4. O he he apeu ics as DNA damage agen s, angiogenesis inhibi o s,
an i-and ogens and immune checkpoin inhibi o s a e unde clinical e alua ion5,6.
Edel osine (1‐O‐oc adecyl‐2‐O‐me hyl‐sn‐glyce o‐3‐phosphocholine o ET) is a syn he ic lipid membe
o alkyl-lysophospholipids amily ha unlike o he chemo he apeu ic d ugs does no ac on he le el o DNA.
I s high apop o ic ac ion on umo cells is in pa ela ed o i s accumula ion in hei plasma memb ane and
ac i a ion o lipid a s7,8. Edel osine was es ed in phase I9 and phase II10 ials in pa ien s wi h acu e leukemia,
OPEN
1Nano-Oncology and T ansla ional The apeu ics Uni , Heal h Resea ch Ins i u e o San iago de Compos ela
(IDIS), Clinical Uni e si y Hospi al o San iago de Compos ela (CHUS), SERGAS, San iago de Compos ela,
Spain. 2Depa men o Zoology, Gene ics and Physical An h opology, Campus o Lugo, Uni e si y o San iago
de Compos ela, Lugo, Spain. 3Cance Ne wo k Resea ch (CIBERONC), Mad id, Spain. 4Cen e o Resea ch
in Molecula Medicine and Ch onic Diseases (CiMUS), Uni e si y o San iago de Compos ela, San iago de
Compos ela, Spain. 5These au ho s con ibu ed equally: So ia Mendes Sa ai a and Ca lha Gu ié ez-Lo e a. *email:
[email p o ec ed]
2
Vol:.(1234567890)
Scien i ic Repo s | (2021) 11:9873 | h ps://doi.o g/10.1038/s41598-021-87968-4
www.na u e.com/scien i ic epo s/
by in a enous and o al adminis a ion, showing p omising esul s in pu ging bone ma ow o ansplan a ion.
Edel osine was also es ed in a phase II s udy in pa ien s su e ing om b ain umo s (inope able o p e iously
ea ed wi h o he he apies showing no posi i e esul s), by o al adminis a ion. The esul s showed ha ET has
he capaci y o s op b ain umo g ow h and imp o e pa ien ’s quali y o li e11. S ill, when adminis e ed o al o
in a enously, his d ug leads o gas oin es inal and hemoly ic oxici y, espec i ely12–14.
Nano echnology has al eady demons a ed i s capaci y o imp o e he access o chemo he apeu ic d ugs o
he si e o ac ion while dec easing hei seconda y side e ec s. In his sense, di e en au ho s ha e p oposed
ET encapsula ion in lipidic nanosys ems in o de o dec ease i s ela ed oxici y issues while imp o ing i s bio-
a ailabili y, and he e o e i s e ec i eness on he ea men o di e en cance s, such as lymphoma15, leukemia16,
os eosa coma17, b eas cance 18, glioma19 as well as ela ed me as asis20, among o he s.
Rega ding b eas cance applica ion, Azna e al. obse ed a s ong inhibi ion o MCF7 cells (human b eas
cance cell line wi h es ogen and p oges e one ecep o s, i. e. no TNBC-de i ed) p oli e a ion and a no ably
dec ease in he cell iabili y upon he ea men wi h ET lipid nanopa icles in compa ison o he ee d ug.
Despi e he encou aging esul s, he pe o mance o his o mula ion was no s udied in i o. In he case o
TNBC, conside ing i s high me as a ic a e in b ain a ea lie s ages, Ren e al. showed ha mice injec ed wi h
b ain me as a ic cells de i ed om TNBC pa ien s, p esen ed a signi ican inhibi ion o b ain me as a ic umo
g ow h as well as o he o ma ion o mac o-me as ases upon he ea men wi h ee ET21.
Taking his in o conside a ion, in his wo k we p opose he p epa a ion o nanome ic emulsions comp ising
edel osine (ET-NEs) o he managemen o TNBC. Unlike he p e iously men ioned s udies ha used me hods
as high shea homogeniza ion and ul asonica ion o he p epa a ion o he lipidic nanopa icles, he ein we used
he simple and mild me hodology o e hanol injec ion, and na u al cell componen s as phospha idylcholine (PC)
and iglyce ides as excipien s, he eby a oiding he use o su ac an s.
ET-NEs’ oxici y was es ed bo h in i o, using a highly agg essi e and in asi e TNBC cell line (MDA-
MB-231) and in i o using zeb a ish (Danio e io) as animal model. Zeb a ish model was selec ed since i p o-
ides he complexi y o in i o condi ions ha he in i o cell assays canno and p esen s a ele an s uc u al and
unc ional homology o humans wi h mo e han 70% o o hologue human genes22. Fu he mo e, in compa ison
o commonly used oden animal models i is a mo e cos -e ec i e and less ime-consuming model due o i s
as e de elopmen and small physical size. Fo ins ance, basic de elopmen is nea ly comple ed wi hin 24h and
sexual ma u a ion is eached in 3–5mon hs. None heless, one o i s main a ac i e ea u es is he anspa en
body o zeb a ish emb yos and la ae, which allows a eal- ime acking o he adminis e ed o injec ed luo es-
cen ly labeled d ugs/nanoca ie s and cells o in e es 23.
Rega ding he cance ield, di e en zeb a ish cance models ha e been de eloped by ansgene exp ession
and xeno ansplan a ion o human umo cell lines o p ima y pa ien -de i ed cells, among o he me hods. The e
a e some unique ea u es ha make zeb a ish an ideal cance model, being especially ele an i s anspa ency,
which as men ioned abo e allows o moni o umo de elopmen and me as ases o ma ion and sp eading, as
well as hei esponse o ea men s24–26. Addi ionally, he delay in he adap i e immune sys em de elopmen
(10–14days) poses a clea ad an age o e oden models, as he immunosupp ession o animals is no equi ed27.
I is also wo h no ing ha neo ascula iza ion can be also s udied due o he simila i y be ween zeb a ish and
humans’ ascula u e28. Jus as impo an ly, acco ding o he Eu opean Food Sa e y Adminis a ion, zeb a ish
up o 5days pos - e iliza ion (dp ) a e less p one o expe ience any pain he eby complying wi h e hical con-
side a ions on animal expe imen a ion (3R p inciple)29. Fo hese easons, zeb a ish cance models a e gaining
ele ance and di e en esea ch g oups a e using hem o he e alua ion o a numbe o d ugs30–32 as well as
nanomedicines33–35, o en as complemen a y models o he mu ine ones. On he o he hand, Nada e al., used
zeb a ish MDA-MB-231 xenog a model o es ing hei pla inum/hyd oxyapa i e nanopa icles-based he apy
agains bone cance , conside ing he high a e o his ype o b eas o me as asize in o bone36.
Following his line, in he p esen wo k we decided o de e mine he e icacy o he de eloped nanosys em
in i o and in i o in a zeb a ish TNBC model o s udy he an i umo al po en ial o he p oposed edel osine
nanosys ems.
Resul s and discussion
De elopmen and cha ac e iza ion o he NEs. ET-NEs, and hei con ol o mula ion (C-NEs) we e
p epa ed by adap ing he e hanol injec ion me hod p e iously op imized by ou g oup o he p epa a ion o
nanome ic emulsions on a single s ep37,38, allowing he s aigh o wa d o ma ion o he nanosys ems, as ep-
esen ed in Fig.1. Edel osine is composed by a long ca bon chain, and phospha e and qua e na y amine g oups.
Conside ing i s lipophilici y, i is expec ed ha i can o m emulsions upon combina ion wi h an oil (Miglyol)
Figu e1. Schema ic ep esen a ion o con ol (C-NEs) and edel osine (ET-NEs) nanosys ems.
3
Vol.:(0123456789)
Scien i ic Repo s | (2021) 11:9873 | h ps://doi.o g/10.1038/s41598-021-87968-4
www.na u e.com/scien i ic epo s/
and an addi ional phospholipid(PC). While he ca bon chain can be inco po a ed in o he oily co e o he s uc-
u e, he phospha e and qua e na y amine g oups can ge exposed on i s su ace, simila ly o PC. Indeed, ET
was e icien ly o mula ed ollowing his p ocess. ET-NEs composed o 85, 10.7 and 4.3% o Miglyol, ET and PC,
espec i ely, p esen ed a small pa icle size, a monodispe se popula ion, and a neu al ze a po en ial (Table1).
The con ol o mula ion (C-NEs) was also p epa ed by eplacing ET o PC, p esen ing a inal composi ion o
85 and 15% o he excipien s Miglyol and PC, espec i ely, which esul ed in a simila neu al ze a po en ial in
compa ison o ET-NEs and a sligh ly smalle a e age size o 100nm (Table1).
The nanoca ie s p ope ies a e in luenced by he nanoca ie composi ion, encapsula ed d ug, as well as he
o mula ion echnique and used sol en s. Fo ins ance, using a di e en p opo ion o leci hin (a ype o PC)/
Miglyol o indome hacin encapsula ion, ace one as sol en , as well as he pou ing me hod, led o he o ma-
ion o pa icles o abou 220 nm39. Eskanda e al. also p epa ed leci hin/Miglyol pa icles o abou 213nm
( o ans- e inol deli e y) by using high p essu e homogeniza ion wi hou any o ganic sol en 40. On he o he
hand, he addi ion o ca ionic su ac an s as hexadecyl ime hylammonium b omide (CTAB) o Miglyol-leci hin
by Teijei o e al. led o he o ma ion o nanoca ie s wi h simila size and polydispe si y o ou nanosys em,
bu wi h a posi i e ne cha ge ins ead o neu al/nega i e one, which in ha wo k was equi ed o he coa ing
wi h hyalu onic acid34. The esul ing physicochemical cha ac e is ics ob ained o he ET o mula ion he ein
de eloped a e o pa icula in e es when in a enous adminis a ion is en isioned, since pa icles o his size
(< 200nm) a e able o passi ely en e in o he umo by he enhanced pe meabili y and e en ion (EPR) e ec 41
and a e less p one o accumula e in he li e and spleen, imp o ing hei accumula ion in he umo 42. On he
o he hand, he neu al ze a po en ial makes hese nanoca ie s less p one o opsoniza ion43 main aining hei
physicochemical p ope ies while ci cula ing in he blood and ex ending hei ci cula ion ime in he body. In
addi ion, i makes he nanoca ie s mo e biocompa ible in compa ison o posi i ely cha ged ones which in e ac
indisc imina ely wi h cells by elec os a ic in e ac ions44,45.
Aside om he physicochemical p ope ies o he de eloped nanoca ie , i s simple and sa e composi ion,
based on GRAS (Gene ally Recognized as Sa e by he Uni ed S a es FDA) ma e ials (Miglyol) and cell compo-
nen s (PC) makes i sui able o he in ended applica ion. In addi ion, he simple and s aigh o wa d me hodol-
ogy used o i s p epa a ion makes i mo e ad an ageous in e ms o a pha maceu ical indus y poin o iew,
han o he lipid nanoca ie s epo ed on li e a u e o ET deli e y, which we e p epa ed by ho and high shea
homogeniza ion ollowed by ul asonica ion17–19. Wi h espec o he concen a ion o ET in he suspension, we
epo ed abou 119µg ET/mg o mula ion ( heo e ical concen a ion o ET, calcula ed as µg ET/(mg Miglyol + mg
PC)) while o he wo ks ha e epo ed concen a ions o 13 o 33µg ET/mg o mula ion15,16,18,19.
In o de o de e mine he easibili y o hese o mula ions o hei es ing in i o, ET-NEs as well as he
con ol o mula ion o e e ence (C-NEs), we e incuba ed in cell cul u e media (DMEM wi h 1% FBS) a 37
ºC, and hei pa icle size was de e mined o a pe iod o 4h(Fig.2). As o mula ions showed a good s abili y
and main ained hei size, we p oceeded wi h hei cy o oxic p o ile e alua ion in i o in MDA-MB-231 cells.
In i o s udies in TNBC cell line. Taking in o accoun ET ela ed ad e se e ec s and i s lipophilici y, di -
e en au ho s ha e p oposed he use o lipid-based nanopa icles o ET deli e y15–17,19. Blanco-P ie o g oup has
showed ha lipid nanosys ems composed o Comp i ol o P eci ol and polyso ba e 80, p epa ed by ho and high
shea homogeniza ion combined wi h ul asonica ion, a e able o dec ease ET hema opoie ic oxici y, and a e
also esponsible o he imp o ed d ug an i umo al e icacy ei he in i o and in i o in glioma, lymphoma and
leukemia mice umo models15,16,19. Mo e ecen ly his g oup has combined ET wi h doxo ubicin17,20, and wi h
gemci abine-squalenic acid conjuga es46 o os eosa coma and ela ed me as asis. This g oup has also es ed he
e ec o ET lipid pa icles in i o in he b eas cance cell line MCF718. None heless, o he bes o ou knowl-
edge, ET has no been es ed o TNBC. Conside ing he agg essi eness o his ype o umo and he an i umo al
e icacy o ET in o he cance s, we decided o s udy i s e ec on TNBC by combining i wi h ou nanosys em
p epa ed by using he simple injec ion me hod, as p e iously men ioned.
In he p esen s udy, ET-NEs and C-NEs we e incuba ed wi h MDA-MB-231 cells o 24h, a inc easing
concen a ions o ET (1.3 o 210µg/mL co esponding o 12.5 o 2000µg/mL o NEs). As i possible o obse e
in Fig.3, and as expec ed, he con ol o mula ion C-NEs do no show cy o oxic e ec s a he es ed ange. This is
due o he ca e ul selec ion o he excipien s, Miglyol, a GRAS medium chain iglyce ide commonly used in sel -
emulsi ying sys ems, and PC, a majo cons i uen o cell memb anes ha is p esen in di e en nano o mula ion
composi ions. Wi h espec o ou o mula ion ET-NEs, i s an i umo al e icacy was dose-dependen , p esen ing
a hal maximal inhibi o y concen a ion (IC50) o 6.9µg/mL (13.2µM) a e 24h o incuba ion, while he IC50
o ee ET was an o de o magni ude highe (13.9µg/mL, 26.5µM).
As p esen ed in Table2, we obse ed he same endency in a lung adenoca cinoma (H460) cell line as well
as in panc ea ic adenoca cinoma cell line (MIA PaCa-2) and panc ea ic cells ep esen ing li e me as asis (L3.
pl6). These cance cell lines we e selec ed o being ep esen a i e o p e alen and di icul o ea umo s. As
Table 1. Physicochemical p ope ies o edel osine nanoemulsions (ET-NEs) and he con ol o mula ions
(C-NEs).
Nanosys em Size (nm) PDI Ze a po en ial (mV)
C-NEs 131 ± 3 0.1 − 2 ± 0
ET-NEs 123 ± 13 0.1 − 1 ± 0
4
Vol:.(1234567890)
Scien i ic Repo s | (2021) 11:9873 | h ps://doi.o g/10.1038/s41598-021-87968-4
www.na u e.com/scien i ic epo s/
expec ed, he one p esen ing a highe esis ance o ea men , equi ing a highe ET-NEs dose, was ou main
a ge cell line MDA-MB-231 (Fig.3).
A dose- and ime-dependen e ec was obse ed by o he au ho s in human b eas cance (MCF7) and leu-
kemia cell lines wi h P eci ol-polyso ba e 80 lipid nanopa icles16,18. Ou esul s showed a highe po en ial o
ET o killing a esis an b eas cance cell line in ela ion o he esul s epo ed by Azna e al. in non- esis an
MCF7 b eas cance cells (IC50 12.9µg/mL a e a longe incuba ion ime o 72h)18. Resul s ela ed o leukemia
cell lines epo ed IC50 o 20µM in he esis an cell line K562, a e 72h ea men , and IC50 o 3.4µM o he
sensi i e cell line MOLM-1316. Al oge he , and conside ing he a iabili y due o cell lines es ed, we can conclude
ET can be e icien ly deli e ed o cance cells upon o mula ion as ET-NEs, p o iding highe e iciencies han
he d ug in solu ion (Table2), and a supe io beha io in ela ion o o he nanosys ems ha ha e been es ed
o longe pe iods o ime16,18.
Con ocal expe imen s addi ionally con i med ha ET-NEs we e e icien ly and highly in e nalized by he
MDA-MB-231, when compa ed o he C-NEs. Figu e4 p esen s a galle y showing se e al sec ions o show ha
di e ences a e e iden all h ough he sample (Fig.4a) as well as a ep esen a i e single plan o all he samples
(Fig.4b). The g een s aining ha is obse ed in cells ea ed wi h ET-NEs co esponds o he labelled o mula-
ion (ET-NEs include TopFluo -PC in a simila amoun as in he con ol o mula ion). The highe capaci y o
ET-NEs (p < 0.0001) o be in e nalized (Fig.4c) migh be explained by he well-desc ibed cha ac e is ic o ET o
accumula e in cells’ plasma memb ane7,8, which he ein p o ided ET-NEs he capaci y o be highly in e nalized.
In i o s udies in zeb a ish emb yos. Being awa e ha he in i o scena io does no e lec he com-
plexi y ound in i o, nex s ep in ou esea ch aimed o de e mine he oxici y, biodis ibu ion, and an i umo al
e icacy, o he ET-NEs in i o, using o ha pu pose zeb a ish emb yos. This animal model ep esen s a s ep
Figu e2. S abili y o con ol NEs (C-NEs) and edel osine NEs (ET-NEs) upon incuba ion wi h zeb a ish
medium (ZFM, s e ile dechlo ina ed ap wa e ) and cell cul u e medium (CCM, DMEM supplemen ed wi h 1%
FBS) o 0, 1 and 4h, a 37°C.
Figu e3. MDA-MB-231 cell iabili y upon he ea men wi h inc easing concen a ions o ee edel osine (ET)
edel osine nanoemulsions (ET-NEs) and con ol nanoemulsions (C-NEs) du ing 24h a 37°C.
Table 2. Hal maximal inhibi o y concen a ion (IC50) o ET-NEs and ee ET upon incuba ion wi h di e en
cance cell lines (lung (H460), panc ea ic (MIA PaCa-2), panc ea ic li e me as asis (L3.pl6)) o 24h a 37 °C.
Nanosys em
(µg/ml) MDA-MB-231 H460 MIA PaCa-2 L3.pl6
ET-NEs 6.9 2.2 1.2 2.5
F ee ET 13.9 6.1 3.0 5.9
5
Vol.:(0123456789)
Scien i ic Repo s | (2021) 11:9873 | h ps://doi.o g/10.1038/s41598-021-87968-4
www.na u e.com/scien i ic epo s/
u he o unde s and umo he e ogenei y, umo cells beha io and di e en mechanis ic aspec s o cance .
Due o i s unique cha ac e is ics, i is conside ed a p omising ool in he de elopmen o no el he apeu ic
s a egies30,47.
P io o es hese o mula ions in i o, ET-NEs and C-NEs we e incuba ed in SDT wa e a 28 ºC, and
hei pa icle size de e mined o a pe iod o 4h, in o de o ensu e ha he e we e no changes on NEs’ phys-
icochemical p ope ies ha could comp omise he assay and lead o a misin e p e a ion o he esul s. Bo h
o mula ions we e able o main ain hei size (Fig.2), he e o e we p oceeded wi h hei in i o e alua ion in
zeb a ish emb yos.
Toxici y assays
Toxici y o nanosys ems upon incuba ion wi h 0–4 hp and 72 hp zeb a ish. Zeb a ish emb yos
we e incuba ed wi h inc easing concen a ions (up o 1000µg/mL) o NEs wi h and wi hou edel osine. Abo e
his concen a ion he media became whi e and he emb yos could no be clea ly obse ed h ough he mic o-
scope, he e o e we did no es highe concen a ions. In acco dance o wha was p e iously obse ed in i o,
a he es ed concen a ions, C-NEs did no lead o he dea h o 0 and 72 hp zeb a ish emb yos, 96h pos -
ea men , i espec i ely o he incuba ion empe a u e.
In he case o he ET-NEs, he LC50 o he zeb a ish main ained a 28 ºC was 12.89µg/mL (0 hp ) and
8.6µg/m (72 hp ), while he ones main ained a 34 ºC p esen ed a LC50 o 11.4µg/mL (0 hp ) and 3.2µg/mL
(72 hp ). The a ia ions in oxici y da a seem o be due o he incuba ion empe a u es used in he expe imen ,
as empe a u e is a ele an and highly a iable abio ic ac o in na u e ha plays an impo an ole du ing ish
emb yonic de elopmen 48. Fu he mo e, as empe a u e inc eases, i can become a physical s esso and aise
he aqua ic o ganisms’ ene gy me abolism and in u n, he bioa ailabili y o oxican s49.
Aside om LC50, o he oxici y indexes as LC10, LOEC and NOEC a e p esen ed in Table3. All he equi e-
men s o FET es we e accomplished: he mo ali y in nega i e con ol emb yos was ≤ 10%, he ha ching a e
was ≥ 80% o he nega i e con ol, whe eas he posi i e con ol was 100% dea hs (minimum equi ed is 30%).
The e we e no signi ican mo phological changes o ha ching delay in compa ison wi h he con ol g oup,
and ollowing he OECD guidelines, abno mal o non-ha ched emb yos a 96 hp we e excluded om he assay.
In gene al, he esul s show ha ET-NEs a e clea ly mo e oxic in con as o he con ol o mula ion (C-NEs),
a ac ha was expec ed due o he ca e ul selec ion o he con ol o mula ion composi ion (Miglyol and PC).
In speci ic, Miglyol, a GRAS medium chain iglyce ide (used in sel -emulsi ying sys ems), and PC, a majo
cons i uen o cell memb anes (also p esen in di e en nano o mula ions). These esul s a e in he same line
as he in i o assays, showing a high compa ibili y o he C-NEs and a dose-dependen oxici y o he ET-NEs.
Taking in o conside a ion he esul s o he oxici y assay (Table3), ET-NEs a 1.5µg/mL (LC10) was selec ed
o u he s udying he an i umo al e icacy o his nanosys em.
Figu e4. In e naliza ion o NEs in MDA-MB 231 cells. (a) Galle y o con ocal mic oscopy images o MDA-MB
231 cells incuba ed wi h 150µg/mL o C-NEs and ET-NEs labeled wi h TopFluo -PC (g een channel) o
4h a 37 ºC. (b) Rep esen a i e con ocal mic oscopy single-plan image o con ol s C-NEs and ET-NEs
in e naliza ion by MDA-MB 231 cells. (c) Fluo escence in ensi y (AU, a bi a y uni s) o wo maximum
p ojec ion images ( esul an om he combina ion o all he sec ions shown in (a)) was de e mined using
ImageJ so wa e. S a is ical analysis was pe o med using es . P alue ****p < 0.0001. Cell nuclei was s ained
wi h Hoechs (blue channel). Scale ba s co espond o 25µm.

6
Vol:.(1234567890)
Scien i ic Repo s | (2021) 11:9873 | h ps://doi.o g/10.1038/s41598-021-87968-4
www.na u e.com/scien i ic epo s/
Toxici y o nanosys ems upon injec ion in 48 hp zeb a ish. Zeb a ish emb yos o 48 hp we e
injec ed wi h 5μg/ml o C-NEs and ET-NEs in he yolk sac o caudal ein and incuba ed a 28 ºC. As expec ed,
no dea h was obse ed 96h a e ea men wi h any o he es ed o mula ions. These esul s assu e ha he e
would be no dea hs as a esul o he injec ion i sel combined wi h he es ed nanosys ems concen a ion.
Biodis ibu ion s udies. One o he g ea es ad an ages o zeb a ish model o ganism is ha i o e s he
possibili y o s udying and as acking he dis ibu ion o nanopa icles h oughou o he o ganism due o i s
anspa ency. In addi ion, i allows a di ec obse a ion o nanopa icles ci cula ion and hei in e ac ion wi h
cells33–35. In his kind o s udy, choosing he igh dye and he app op ia e con ols is o ex eme impo ance in
o de o be su e ha we a e acking he nanosys ems indeed and a oid he misin e p e a ion o biodis ibu ion
da a. On one hand we selec ed DiR, which is commonly used in biodis ibu ion assays and is easily encapsula ed
in he nanosys em’s oil co e. None heless, i can be eleased om he oil co e o nanosys ems causing an appa -
en bu alse cellula up ake. The e o e, we also used TopFluo , co alen ly linked o he PC, and he ein ancho ed
o he memb ane o he NEs, o help us ensu ing ha he ob ained signal was due o he p esence o he NEs by
DiR/TopFluo -PC co-localiza ion, and no due o he p esence o ee luo opho es.
Emb yos o 72 hp we e incuba ed wi h 500µg/mL o DiR and TopFluo -PC labeled C-NEs o s udy i s
in e naliza ion and dis ibu ion. This high concen a ion was necessa y o be able o obse e he NEs wi h
enough amoun o luo escen labeling, unde he con ocal mic oscope. As men ioned be o e, C-NEs we e highly
compa ible e en a he maximum es ed concen a ion o 1000µg/mL (LC50 could no be assessed), whe eas
he de e mined LC50 o ET-NEs was 8.6µg/mL. Hence, a he necessa y concen a ion (500µg/mL) o he
biodis ibu ion assay, ET-NEs would be le hal o he emb yos, which lead us o pe o m he s udy only wi h he
con ol blank o mula ion (C-NEs).
Con ocal mic oscope analysis showed ha NEs we e e icien ly in e nalized by he exposed emb yos, espe-
cially in o he yolk sac (Fig.5), which demons a es he capaci y o he C-NEs o c oss biological ba ie s upon
di ec con ac wi h he skin o zeb a ish emb yos (wi hou cho ion). F ee DiR and TopFluo -PC (Fig.5a con ol)
we e jus included o e e ence and no o compa a i e pu poses. Zeb a ish o 48 hp a e p o ec ed by he cho-
ion. A e his pe iod o ime, he zeb a ish loses his laye (ha ching) and he skin becomes he main biological
ba ie p o ec ing he emb yo om he ex e nal en i onmen . As indica ed, in his s udy we incuba ed he NEs
wi h 72 hp emb yo meaning ha he NEs we e in di ec con ac wi h he emb yo skin. Teijei o-Valiño e al.
ound impo an di e ences be ween he zeb a ish biological ba ie s (cho ion s skin) in e ms o he pe me-
abili y and oxici y o posi i ely cha ged NEs and nega i ely cha ged nanocapsules (NCs)34. The NCs con aining
an ex e nal shell o hyalu onic acid/p o amine had he capaci y o pe mea e h ough he cho ion and skin laye s,
unlike he NCs con aining only a hyalu onic shell ha emained associa ed o he ex e nal laye o hese ba ie s
being unable o pene a e. Acco ding o he au ho s he pene a ion abili y o NCs con aining p o amine could
be due o p o amine, which aside om i s ne posi i e cha ge is known as a cell pene a ing pep ide, bu could
be also due o he p esence o PEG-s ea a e.
He ein he biodis ibu ion o he ET-NEs, unde he es ablished condi ions i. e. he equi ed concen a ion
o con ocal mic oscopy analysis (500µg/mL), was no s udied conside ing hei highe oxici y han he C-NEs.
Unlike nega i ely cha ged nanoca ie s de eloped by Teijei o-Valiño e al., ou nanosys ems p esen a neu al
su ace cha ge and he e o e we hypo hesized ha C-NEs capaci y o pene a e h ough he zeb a ish skin migh
be due o he p esence o PC, which is a majo cons i uen o cell memb anes. In addi ion, aking in o conside a-
ion he in i o in e naliza ion esul s in MDA-MB 321 cells, a simila and e en g ea e pene a ion capaci y
would be expec ed o ET-NEs.
An i umo al E icacy o ET‑NEs in zeb a ish emb yos xeno ansplan ed wi h TNBC cells. Aside
om zeb a ish anspa ency, he possibili y o ansplan human cance cells in o his animal model is also use-
ul o de e mining he e icacy o an an i-cance he apy. In his sense, se e al models ha e been de eloped,
leading o a be e unde s anding o di e en c i ical aspec s o cance such as p oli e a ion and in asion, umo
o ma ion, angiogenesis, me as asis o immune cell esponse50–55. Fu he mo e, he umo mic oen i onmen
is mo e accu a ely ep esen ed han in in i o models, and he in e ac ion o umo cells wi h he hos can be
in e ed, o ins ance, om he immune cells’ beha io , due o he conse ed cell in e communica ion be ween
humans and zeb a ish56. These ea u es, oge he wi h hose p e iously men ioned, make he zeb a ish a highly
aluable pla o m no only o un a el umo beha io , bu also o es di e en po en ial an i-cance d ugs and
o pe o m high- h oughpu sc eenings o no el he apeu ic compounds. O e all, his b ings esea che s close
o unde s anding pa ien s’ esponse o ea men and, in u n, o pe sonalized medicine57.
Table 3. Toxici y o 0 and 72 hp emb yos exposed o ET-NEs o 96h. hp : hou pos - e iliza ion; LC10,
LC50, NOEC and LOEC a e ep esen ed in µg/mL.
hp T (ºC) LC10 LC50 NOEC LOEC
028 9.8 12.89 10 5
34 8.6 11.4 10 5
72 28 4.4 8.6 5 1
34 1.5 3.2 5 1
7
Vol.:(0123456789)
Scien i ic Repo s | (2021) 11:9873 | h ps://doi.o g/10.1038/s41598-021-87968-4
www.na u e.com/scien i ic epo s/
In his wo k, ou objec i e was o de e mine he po en ial o an i- umo ac i i y o ET-NEs on zeb a ish
emb yos ansplan ed wi h MDA-MB-231 cells (exp essing GFP). The yolk sac model is a widely accep ed model
o d ug e icacy s udies58,59 and he ein i was selec ed in o de o a oid modi ica ion o he inhe i ed pheno ype
o he injec ed cells52,60,61 (MDA-MB-231) and o educe space limi a ions o cell p oli e a ion62.
S ole o e al. ansplan ed zeb a ish emb yos wi h di e en b eas cance cell lines and showed ha MDA-
MB-231 and MDA-MB-435 cells p esen a high a e o umo o ma ion, essel densi y and me as a ic beha io
in con as wi h low agg essi e b eas cance s as BT-47463. In he same line, Asokan e al. obse ed ha a e
MDA-MB-231 cells injec ion in o he zeb a ish blood ci cula ion, he cells p oli e a ed mainly in he adjacen
a eas bu abou 30% p esen ed a me as a ic p o ile64. Simila ly, Me ca ali e al. showed he dissemina ion and
coloniza ion o di e en zeb a ish issues a e MDA-MB-231 xenog a bu con e sely, when he non-in asi e
cell line MCF7 was injec ed, a mig a o y pheno ype was no obse ed. Addi ionally, a e injec ing b eas cance
pa ien -de i ed bone me as a ic p ima y cells hey obse ed bone ma ow opism, as cells we e able o mig a e
Figu e5. Con ocal mic oscopy images o 72 hp zeb a ish emb yos (a) incuba ed wi h 500µg/mL o C-NEs
labeled wi h DiR and TopFluo -PC, o 4h a 34 °C. Con ol e e s o non- ea ed zeb a ish, which p esen
au o- luo escence. Scale ba s co espond o 250µm. (b) Z-s ack images. Red channel: DiR. G een channel:
TopFluo -PC.
8
Vol:.(1234567890)
Scien i ic Repo s | (2021) 11:9873 | h ps://doi.o g/10.1038/s41598-021-87968-4
www.na u e.com/scien i ic epo s/
o he caudal hema opoie ic issue, esembling he pa ien ´s clinical p o ile65. These s udies p o ided e idence
ha his model would be eliable and sui able o es he he apeu ic e iciency o he de eloped o mula ion.
In his sense, a 0 and 48h pos - ea men o he xeno anspla ed emb yos we analyzed he umo g ow h by
con ocal mic oscopy. As i is possible o obse e in Fig.6a, a 48 hp he e was a educ ion o he luo escence in
he yolk sac o he emb yos ea ed wi h ET-NEs, meaning ha he e was a educ ion in he numbe o cance
cells, a ac ha alida ed he po en ial o ou app oach o he ea men o iple nega i e b eas cance . As
expec ed, in he case o he con ol and C-NEs ea ed emb yos, he e was an inc ease o he luo escen a ea,
indica ing umo g ow h.
In addi ion, he analysis pe o med wi h Quan iFish p og am66 o quan i y he luo escence in ensi y and
de e mine he cell g ow h and sp ead a he umo si e, co obo a es he success ul and signi ican (p < 0.0001)
educ ion o cell p oli e a ion in xenog a ed emb yos ea ed wi h he ET-NEs in compa ison wi h C-NEs
ea ed emb yos and he con ol (Fig.6b). O he au ho s ha e also been able o e alua e di e en nanoca i-
e s in zeb a ish emb yos xenog a s wi h impo an posi i e esul s, ei he by di ec injec ion o incuba ion
acco ding o he size, physicochemical cha ac e is ics and p ope ies o he nanopa icles. E ensen e al. used
polye hylene glycol-coa ed nanopa icles in p os a e cance xenog a ed emb yos and demons a ed a low up ake
by mac ophages and a speci ic a achmen o umo -like s uc u es35. In e es ingly, Yang and colleagues de el-
oped b ain endo helial cell-de i ed exosomes and used hem as ehicle o deli e VEGF siRNAs in a zeb a ish
xeno ansplan ed b ain umo model. These na u al nanoca ie s we e able o c oss de blood–b ain ba ie ,
inhibi VEGF and dec ease he luo escence in ensi y o cance cells67. Liu e al., used iphenylphosphonium/
hyalu onic acid-based nanopa icles bea ing doxo ubicin in xeno ansplan ed emb yos and obse ed he inhibi-
ion o b eas cance cells p oli e a ion wi hou signi ican side e ec s68.
Wi h ega d o TNBC, so a only Nada and colleagues ha e used zeb a ish MDA-MB-231 xenog a model
in o de o de e mine he an i- umo al e icacy o pla inum-loaded hyd oxyapa i e nanopa icles. The au ho s
used his animal model since his ype o b eas umo equen ly me as asize in o he bones, which was he main
a ge o he de eloped he apy. Fo ha , he nanopa icles we e co-injec ed wi h MDA-MB-231 in o he zeb a ish
blood ci cula ion, a signi ican (p < 0.01) educ ion o abou 1.9 old in he numbe o cance cells was obse ed 48
hpi, in compa ison o he un ea ed con ol. Ou nanosys em, composed o he d ug edel osine, wi h a di e en
mechanism o ac ion (does no ac on he DNA le el), and o mula ed as a nanoemulsions making use o GRAS
ma e ials, also led o a signi ican educ ion (p < 0.001) o umo p oli e a ion a e o abou 2.3 old in compa i-
son o he un ea ed con ol. To he bes o ou knowledge, i is he i s ime ha he an i umo al e iciency o
edel osine, o mula ed as an emulsion ollowing a simple me hodology and a sa e-by-design app oach, is s udied
in TNBC using zeb a ish xenog a as animal model. We p o ed ha ET-NEs can be e iciency in e nalized by
cance cells, leading o a he apeu ic e ec , in i o and in i o (Fig.6).
Conclusions
ET-NEs we e success ully o mula ed by e hanol injec ion, p esen ing sui able physicochemical p ope ies. The
highe oxici y o ET-NEs compa ed o ee ET p o ides e idence ha he ET an i umo al capaci y is imp o ed
in a dose-dependen manne , in i o. In addi ion, in i o esul s p o ed ha he nanosys ems can pene a e
h ough biological ba ie s, especially he ones con aining edel osine, which signi ican ly dec eased he p oli -
e a ion o umo al cells in zeb a ish emb yos bea ing MDA-MB-231 xenog a umo , leading o i s eg ession.
O e all, he ob ained esul s indica e ha he de eloped o mula ion, ET-NEs, is a p omising he apy o he
ea men o TNBC.
Figu e6. Con ocal analysis o umo cell p oli e a ion in zeb a ish (a) be o e (0 hp ) and a e 48 hp . Zeb a ish
o 48 hp wi hou cho ion we e injec ed wi h MDA-MB 231 cells exp essing GFP and subsequen ly incuba ed
wi h ET-NEs and C-NEs a 34 °C. Un ea ed xenog a ed emb yos we e used as con ol. (b) Quan iFish analysis
p og am was used o quan i y he luo escence in ensi y and de e mine umo p oli e a ion. S a is ical analysis
was pe o med using One-way ANOVA ollowed by a Tukey es . P alues ***p < 0.001, ****p < 0.0001, ns, no
signi ican .
9
Vol.:(0123456789)
Scien i ic Repo s | (2021) 11:9873 | h ps://doi.o g/10.1038/s41598-021-87968-4
www.na u e.com/scien i ic epo s/
Me hods
Ma e ials. Miglyol 812 (CAS 37332–31-3) and phospha idylcholine (PC, CAS 8002–43-5) we e pu chased
om Lipoid Ludwigsha en, Ge many. Edel osine (CAS 77286–66-9) was acqui ed om San a C uz Bio echnol-
ogy. DiR lipophilic cyanine dye (CAS 100068–60-8) was supplied om The mo Fishe Scien i ic, and TopFluo -
PC (CAS 1246355–63-4) om A an i Pola Lipids. E hanol (high pu i y) was ob ained om PanReac Appli-
Chem. Dulbecco’s Modi ied Eagle Medium (DMEM) was ob ained Sigma, RPMI 1640 Medium, Fe al Bo ine
Se um (FBS) and penicillin–s ep omycin om The mo Fishe Scien i ic.
P epa a ion and cha ac e iza ion o nanoemulsions. Edel osine nanoemulsions (ET-NEs) com-
posed by Miglyol 812, PC and ET we e o mula ed by adap ing he e hanol injec ion me hod37,69. In b ie , 4mg
o Miglyol, 0.2mg o PC and 0.5mg o ET we e dissol ed in 100 µL o e hanol. ET-NEs we e ins an aneously
o med by injec ing he o ganic phase in o 1mL o ul apu e wa e unde magne ic s i ing, a oom empe a-
u e (RT). ET-NEs we e le unde magne ic s i ing o 10min o ensu e hey we e comple ely o med. Con ol
NEs (C-NEs) we e p epa ed by he same me hod wi h 4mg o Miglyol and 0.7mg o PC.
The pa icle size, polydispe si y index (PDI), and ze a po en ial o ET-NEs and C-NEs we e measu ed by
Dynamic Ligh Sca e ing and Lase Dopple Anemome y, using a Ze asize NanoZS (Mal e n Ins umen s
L d., UK). Samples we e dilu ed o 1:10 wi h ul apu e wa e and he measu emen s we e pe o med a 25 ºC
wi h a de ec ion angle o 173º. All da a is exp essed as a mean alue ± s anda d de ia ion (SD).
S abili y in ele an biological media. The colloidal s abili y o ET-NEs and C-NEs was e alua ed in
ele an biological media, o ensu e ha he o mula ions main ained hei p ope ies du ing he in i o and
in i o es ing. The o mula ions wi h/wi hou ET we e incuba ed wi h DMEM supplemen ed wi h 1% o FBS
and wi h s e ile dechlo ina ed ap (SDT) wa e a 37 and 28 ºC, espec i ely, unde cons an ho izon al shaking.
Pa icle size was measu ed up o 4h. In all cases, o mula ions we e dilu ed 1:10 ( / ).
P epa a ion o DiR and TopFluo ‑PC labelled NEs. Fluo escen labelled NE we e p epa ed as p e i-
ously desc ibed in sec ion “P epa a ion and cha ac e iza ion o nanoemulsions”. B ie ly, o he in i o assays
TopFluo -PC (2µg) was mixed wi h he o he compounds o he o ganic phase ( inal olume o 100 µL), while
o he in i o assays, 5 and 10µg o DiR and TopFluo -PC, espec i ely, we e used. The o ganic phase was hen
injec ed in 1mL o milli-Q wa e and kep unde magne ic s i ing a oom empe a u e o en minu es.
In i o s udies. T iple nega i e b eas adenoca cinoma cells MDA-MB-231 (ATCC HTB-26), and pan-
c ea ic ca cinoma MIA PaCa-2 (ATCC CRL-1420), L3.pl6 (CVCL_0384) we e cul u ed in DMEM, while lung
cance cell line H460 (ATCC HTB-177) was cul u ed in RPMI a 37 ºC in a humidi ied a mosphe e o 95% ai
and 5% CO2. All cell cul u e medium (CCM) we e supplemen ed wi h 10% FBS and 1% penicillin–s ep omycin.
All cell lines we e passed be o e eaching 80% con luence, 2–3 imes a week, and he cul u e media was eplaced
e e y second day.
Twen y housand MDA-MB-231 cells and 10,000 cells o MIA PaCa-2, L3.pl6 o H460 (cells/well we e seeded
in 96-well pla es and allowed o adhe e and g own o e nigh a 37 ºC and 5% CO2. A e wa ds, media was
eplaced wi h 100 µL o CCM and 25 µL o ee ET (s ock solu ion in e hanol and u he dilu ed in wa e ), ET-
NEs and C-NEs (p e iously dilu ed in wa e ) a inc easing concen a ions (1.3 o 210µg/mL o ET co espond-
ing o 12.5 o 2000µg/mL o NEs) wi h 6 wells/condi ion and le incuba ing a 37°C o 24h. Addi ionally, 25
µL o PBS wi h 1% T i on X-100 we e added as posi i e con ol and 25µL o milli-Q wa e as nega i e con ol.
A e wa ds, cells we e washed wi h PBS, and cell iabili y was de e mined by incuba ing he cells wi h 110mL o
a il e ed MTT solu ion (0.5mg/mL in non-supplemen ed DMEM) a 37 ºC. A e 3h, he solu ion was emo ed
and he o mazan c ys als we e solubilized wi h 110mL o DMSO (15min a 37°C, p o ec ing om ligh ). MTT
educ ion was de e mined by measu ing he ligh abso bance a 570nm in a mic opla e spec opho ome e (DTX
880 Mul imode De ec o , Beckman Coul e ). Cell iabili y was calcula ed in pe cen age ela ed o un ea ed
con ol wells, sub ac ing backg ound alues.
In e naliza ion s udies. C-NEs and ET-NEs we e labeled wi h Top-Fluo -PC as p e iously desc ibed. In
o de o e alua e hei capaci y o be in e nalized by MDA-MB 231 cells, 80,000 cells we e seeded o e a glass
co e slip in a 24-well pla e. A e 24h a 37 ºC, he cells we e washed wi h PBS and a e wa ds incuba ed wi h
he nanosys ems a a concen a ion o 150µg/mL o 4h a 37 ºC. A e , cells we e washed wice wi h PBS, ixed
o 15min a RT wi h 4% (w/ ) pa a o maldehyde and washed wice again wi h PBS. Cell nuclei was s ained
wi h Hoechs o 5min and hen cells we e washed h ee imes wi h PBS. The co e slips whe e hen moun ed
o e mic oscope slides and le o d y o e nigh , p o ec ed om ligh . Fluo escence in ensi y o TopFluo -PC
(p esen in C-NEs and ET-NEs) on maximum p ojec ion images was de e mined using ImageJ so wa e. Fluo-
escence in ensi y o he con ol was no de e mined since i only p esen s luo escen cell nuclei (Hoechs ) and
no TopFluo -PC died cy oplasm.
In i o s udies in zeb a ish. One-yea -old adul wild- ype zeb a ish we e main ained in a con olled
aqua ic acili y wi h pu i ied and dechlo ina ed wa e by a e e se osmosis sys em, wi h he ollowing condi-
ions: 27 ºC (± 1 ºC), pH 7 (± 0.5), 14/10h ligh /da k pho ope iod and conduc i i y 650 µS/cm in 30 L aqua ia a
a a e o one ish pe li e o wa e . Zeb a ish emb yos we e ob ained om ma ing adul s acco ding o p e iously
desc ibed p ocedu es70. The emb yos we e collec ed and washed wi h osmosis wa e in Pe i dishes and 0–4h
pos - e iliza ion (hp ) emb yos we e selec ed wi h an in e ed op ical mic oscope (Nikon TMS). All p ocedu es