Mic ochannel Acous opho esis does no Impac Su i al
o Func ion o Mic oglia, Leukocy es o Tumo Cells
Miguel A. Bu guillos
1,5.
, Cecilia Magnusson
6.
, Ma ia No din
2.
, And eas Lensho
2
, Pe Augus sson
2
,
Magnus J. Hansson
7
, Eskil Elme
´
7
, Hans Lilja
6,8,9,10
, Pa ik B undin
1
, Thomas Lau ell
2,4
,
Tomas Deie bo g
3
*
1Neu onal Su i al Uni , Depa men o Expe imen al Medical Science, Wallenbe g Neu oscience Cen e , Lund Uni e si y, Lund, Sweden, 2Depa men o Measu emen
Technology and Indus ial Elec ical Enginee ing, Lund Uni e si y, Lund, Sweden, 3Expe imen al Neu oin lamma ion Labo a o y, Depa men o Expe imen al Medical
Science, Wallenbe g Neu oscience Cen e , Lund Uni e si y, Lund, Sweden, 4Depa men o Biomedical Enginee ing, Dongguk Uni e si y, Seoul, Sou h Ko ea,
5Depa men o Oncology-Pa hology, Cance Cen um Ka olinska, Ka olinska Ins i u e , S ockholm, Sweden, 6Depa men o Labo a o y Medicine, Lund Uni e si y, Ska
˚ne
Uni e si y Hospi al, Malmo
¨, Sweden, 7Mi ochond ial Pa hophysiology Uni , Depa men o Clinical Sciences, Lund Uni e si y, Lund, Sweden, 8Depa men s o Su ge y
(U ology) and Labo a o y Medicine, Memo ial Sloan-Ke e ing Cance Cen e , New Yo k, Uni ed S a es o Ame ica, 9Nu ield Depa men o Su gical Sciences, Uni e si y o
Ox o d, Ox o d, Uni ed Kingdom, 10 Ins i u e o Biomedical Technology, Uni e si y o Tampe e, Tampe e, Finland
Abs ac
Backg ound:
The use o acous ic o ces o manipula e pa icles o cells a he mic o luidic scale (i.e. acous opho esis),
enables non-con ac , label- ee sepa a ion based on in insic cell p ope ies such as size, densi y and comp essibili y.
Acous opho esis holds g ea p omise as a cell sepa a ion echnique in se e al esea ch and clinical a eas. Howe e , i has
been sugges ed ha he o ce ac ing upon cells unde going acous opho esis may impac cell iabili y, p oli e a ion o cell
unc ion ia sub le pheno ypic changes. I his we e he case, i would sugges ha he acous opho esis me hod would be a
less use ul ool o many cell analysis applica ions as well as o cell he apy.
Me hods:
We in es iga e, o he i s ime, se e al key aspec s o cellula changes ollowing acous opho e ic p ocessing. We
used wo se ings o ul asonic ac ua ion, one ha is used o cell so ing (10 V
pp
ope a ing ol age) and one ha is close o
he maximum o wha he sys em can gene a e (20 V
pp
). We used mic oglial cells and assessed cell iabili y and
p oli e a ion, as well as he in lamma o y esponse ha is indica i e o mo e sub le changes in cellula pheno ype.
Fu he mo e, we adap ed a simila me hodology o moni o he esponse o human p os a e cance cells o
acous opho e ic p ocessing. Las ly, we analyzed he espi a o y p ope ies o human leukocy es and h ombocy es o
explo e i acous opho e ic p ocessing has ad e se e ec s.
Resul s:
BV2 mic oglia we e unal e ed a e acous opho e ic p ocessing as measu ed by apop osis and cell u no e assays
as well as in lamma o y cy okine esponse up o 48 h ollowing acous opho esis. Simila ly, we ound ha acous opho e ic
p ocessing nei he a ec ed he cell iabili y o p os a e cance cells no al e ed hei p os a e-speci ic an igen sec e ion
ollowing and ogen ecep o ac i a ion. Finally, human h ombocy es and leukocy es displayed unal e ed mi ochond ial
espi a o y unc ion and in eg i y a e acous opho e ic p ocessing.
Conclusion:
We conclude ha mic ochannel acous opho esis can be used o e ec i e con inuous low-based cell
sepa a ion wi hou a ec ing cell iabili y, p oli e a ion, mi ochond ial espi a ion o in lamma o y s a us.
Ci a ion: Bu guillos MA, Magnusson C, No din M, Lensho A, Augus sson P, e al. (2013) Mic ochannel Acous opho esis does no Impac Su i al o Func ion o
Mic oglia, Leukocy es o Tumo Cells. PLoS ONE 8(5): e64233. doi:10.1371/jou nal.pone.0064233
Edi o : Sal a o e V. Pizzo, Duke Uni e si y Medical Cen e , Uni ed S a es o Ame ica
Recei ed Decembe 3, 2012; Accep ed Ap il 12, 2013; Published May 27, 2013
Copy igh : ß2013 Bu guillos e al. This is an open-access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License, which pe mi s
un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal au ho and sou ce a e c edi ed.
Funding: This wo k was suppo ed by g an s om he Swedish go e nmen al agency o inno a ion sys ems, VINNOVA, CellCARE, g an No. 2009-00236, he
Swedish Resea ch Council g an no. 2012-2229 and 2010-4389, by he Gyllens ie nska K appe up, The Royal Physiog aphic Socie y, C a oo d, A.E. Be ge , Wibe g,
Be g all, G&J Kock, Ca l T ygge ounda ions, he SSF S a egic Resea ch Cen e (C ea e Heal h), Swedish Cance Socie y [11-0624], Finnish Funding Agency o
Technology and Inno a ion (TEKES), Na ional Cance Ins i u e [R33 CA127768-03, R01CA160816 and P50-CA92629], Sidney Kimmel Cen e o P os a e and
U ologic Cance s, and Da id H. Koch h ough he P os a e Cance Founda ion, Na ional Ins i u e o Heal h Resea ch (NIHR) Ox o d Biomedical Resea ch Cen e
P og am. The unde s had no ole in s udy design, da a collec ion and analysis, decision o publish, o p epa a ion o he manusc ip .
Compe ing In e es s: Lau ell is a boa d membe , ounde and sha eholde o AcouSo AB ha de elops acous opho esis echnology. Lau ell and Lensho a e
in en o s o acous opho esis pa en s owned by E ysa e AB o which nei he has any engagemen . Lau ell holds esea ch g an s in he ield o acous opho esis
om se e al Swedish unding Agencies; Swedish Science Council, FORMAS, Vinno a, Knu and Alice Wallenbe g Founda ion. Lilja and B undin a e ounde s and
sha eholde s o AcouSo AB, which de elops acous opho esis echnology. Lau ell, Augus sson, Magnusson a e in en o s o a pa en applica ion in he
acous opho esis ield, owned by Acouso AB. D . Lilja holds pa en s o licensed and comme cialized ee PSA, in ac PSA, and hK2 assays. The au ho s he eby
con i m ha he decla ed pa en along wi h any o he ele an decla a ions ela ing o employmen , consul ancy, pa en s, p oduc s in de elopmen o modi ied
p oduc s do no al e hei adhe ence o all he PLOS ONE policies on sha ing da a and ma e ials.
* E-mail: omas.deie bo [email protected]
.These au ho s con ibu ed equally o his wo k.
PLOS ONE | www.plosone.o g 1 May 2013 | Volume 8 | Issue 5 | e64233
In oduc ion
The use o acous ic o ces o handle pa icles and cells in
mic o luidic sys ems (i.e. mic ochannel acous opho esis) is gaining
inc eased a en ion [1]. The applica ion in which he acous o-
pho esis me hod can be used include pa icle manipula ion [2,3],
deple ion [4], washing [5,6,7], ac iona ion [8], a e e en so ing
[9,10], concen a ion [11] and cell cycle synch oniza ion [12].
This no el cell manipula ion echnique is label- ee and enables
sepa a ion by unique cell p ope ies, e.g. comp essibili y. In iew o
i s high ep oducibili y, eliabili y and he ac ha his echnology
can be applied o mos cell ypes, acous opho esis holds g ea
p omise as a cell manipula ion echnique in se e al esea ch and
clinical se ings [13].
While acous opho esis is eme ging as a new echnology in
se e al esea ch a eas, he e a e doub s o whe he he induced
acous ic o ces and luid handling a e ha m ul o he cells.
Ques ions ha a e ele an o his echnology i acous opho e ic
applica ions a e o be used wi h clinical se ing. Ea lie s udies on
he impac o acous ic esonan sys ems on cells ha e been ecen ly
e iewed by Wiklund (2012) [14]. Mo eo e , Ryll and coau ho s
s udied Chinese hams e o a y cells in a pe used mac oscale
acous ic cell e en ion de ice o 50 days and concluded ha no
ha m was obse ed o his cell ype [15]. In ano he s udy, Wang
and collabo a o s s udied mouse hyb idoma cells, which we e
acous ically apped in a high po osi y polyes e mesh wi h a low
in ensi y, esonan acous ic ield [16], concluded ha he acous ic
ield p oduced a negligible e ec on cell iabili y in a sho - e m
exposu e. Simila ly, Hul s o¨m and colleagues [17] as well as
E ande e al. [18], using cos-7 cells om e al monkey kidney and
a neu al s em cells espec i ely, s udied cell iabili y in an
acous ic ap. Howe e , bo h s udies concluded ha cell iabili y
was no a ec ed. Mo eo e , E ande e al. success ully g ew yeas
cells wi hin he ap o demons a e ha cell p oli e a ion was no
a ec ed [18].
Al hough acous opho e ic echnology shows g ea p omise,
acous opho e ic manipula ion o cells in a clinical se ing mus be
s udied in de ail. Bazou and colleagues s udied human li e
ca cinoma cells (HepG2) in an acous ic ap and de e mined ha
cell iabili y and p oli e a ion we e no a ec ed [19]. Using a
con inuous low sys em, Jo¨nsson and coau ho s sepa a ed e y h-
ocy es om lipid pa icles and concluded ha he e had been no
inc ease o hemolysis o e y h ocy es a e passing h ough an
acous opho e ic de ice [20]. Recen ly Dykes e al. emo ed
pla ele s om pe iphe al blood p ogeni o cell p oduc s by
acous opho esis and cell iabili y and colony- o ming abili ies o
he p ogeni o cells was s udied. Fu he mo e, mo phological
s udies as well as pla ele ac i a ion assays concluded ha he cells
we e no ha med by he acous opho e ic ea men [21]. Howe e ,
he li e a u e s ill lacks a ho ough examina ion on he e ec o
mic ochannel acous opho esis using sho - e m acous ic exposu e
imes wi h long- e m iabili y and pheno ypic cha ac e iza ion.
Especially cha ac e iza ions o impo an long- e m unc ional
biological pa ame e s such as in lamma o y esponse, cell ac i a-
ion esponse and espi a ion ha e no been s udied in de ail.
I he acous opho esis echnology is used in he clinical se ing,
he impac on cell su i al and he sub le pheno ypic changes ha
may be induced mus be in es iga ed in de ail. Hence, in his s udy
we examine se e al key cellula changes ollowing acous opho esis
and use mic oglial cells, a cell ype known o eadily eac o
en i onmen al cues o es cell iabili y and p oli e a ion. We also
moni o mo e sub le changes in cellula pheno ype by analyzing
in lamma o y s a us a e acous opho e ic p ocessing. Finally, we
use simila me hodology o examine he e ec on human p os a e
cance cells and s udy he di e en p ope ies o cell espi a ion
using human leukocy es and h ombocy es. Ou s udies indica e
ha mic ochannel acous opho esis can be uned o e ec i e cell
handling wi hou ha ing any e ec o cell iabili y, p oli e a ion,
espi a ion o in lamma o y s a us.
Ma e ials and Me hods
E hics S a emen
The s udy was app o ed by he egional e hical e iew boa d o
Lund, Sweden (pe mi numbe 113/2008). Blood samples we e
collec ed om heal hy blood dono s a he blood dono cen al,
Ska˚ne Uni e si y Hospi al, Lund. Samples we e ob ained a e
w i en in o med consen was acqui ed.
Acous opho esis De ice and Se up
The acous opho esis chip design is illus a ed in igu e 1. The
acous opho esis chip was ab ica ed in ,100.-silicon h ough
double sided pho oli hog aphy and aniso opic we e ching using
KOH. A bo o loa glass lid o seal he low channel was anodically
bonded o he chip. The chip s uc u e is comp ised o i u ca ion
inle ollowed by a 20 mm long acous ic ocusing channel (squa e
c oss-sec ion; 375 mm6150 mm) co esponding o a e en ion ime
o abou 1 second in he acous ic ield o a low a e o 100 mL
min
21
and a i u ca ion ou le egion. Fo he iabili y
expe imen s only one inle and ou le was used. The cen e ou le
low a e was con olled by a sy inge pump (WPI sp210iwz, Wo ld
P ecision Ins umen s Inc., Sa aso a, FL, USA) and se o 100 mL
min
21
and he cell suspension was d awn in o he cen e inle om
an Eppendo ube a he same speed. To a oid sedimen a ion in
he ube du ing he cou se o he expe imen s he cells we e mixed
gen ly by pipe ing. Samples we e collec ed ei he by a 6-po 2-
way sample loop (V-451, Upchu ch Scien i ic, Oak Ha bo , WA,
USA) wi h a olume o 100 mL connec ed in se ies wi h he ou le
o di ec ly in he sy inge ( o he human blood cell espi a ion
s udies). To ac ua e he chip a piezoce amic ansduce (PZ26,
esonan a 2 MHz, Fe ope m, K is gaa d, Denma k) was used.
The piezoce amic was d i en by a wa e o m gene a o (Agilen ,
33250A) se a a equency o 1.94 MHz, ampli ied by an in-
house-buil ampli ie . The ol age o e he ansduce was
measu ed using an oscilloscope (TDS 1002, Tek onix UK L d.,
B acknell, UK). The empe a u e o he chip was con olled using
a Pel ie elemen egula ed by a Pel ie egula o (TC2812-RS232,
CoolT onic GmbH, Beinwill am See, Swi ze land) and se o 37uC
o limi e ec s emana ing om ele a ed empe a u e exposu e
caused by powe dissipa ion in he piezo ce amic a ele a ed
ope a ing ol age ha could a ec he iabili y o he cells. The
cells we e p ocessed h ough he chip and exposed o ul asound a
ei he 10 o 20 V
pp
ope a ing ol age o un h ough he chip
wi hou ul asound exposu e o e eal any in luence o he
mic o luidic sys em. Unp ocessed cells we e used as a con ol.
The acous ic ene gies he cells we e exposed o we e highe han
would no mally be needed o ocus cells in his chip o he
pa icula se ings s a ed abo e. In his se up, h ombocy es we e
less p one o ocus in he middle o he mic ochannel due o hei
small sizes.
Es ima ion o he Acous ic Ene gy Densi y
To es ima e he acous ic ene gy densi y (E
ac
) inside he
acous opho e ic mic ochannel polys y ene mic opa icles
(,0.02%
w
,Ø7mm) (Sigma-Ald ich. Co., S . Louis, MO, USA)
suspended in T i on X-100 (0.01%) in PBS we e p ocessed in he
de ice. To es ima e E
ac
he mic opa icles we e d awn in o he
chip om he side inle s a low a e o 50 mL min
21
and a clean
Impac o Acous opho esis on Cells
PLOS ONE | www.plosone.o g 2 May 2013 | Volume 8 | Issue 5 | e64233
bu e was pumped in o he cen e inle a a low a e o 450 mL
min
21
. The beads we e hen ocused in he cen e o he channel
and collec ed om he cen e ou le a a low a e o 250 mL min
21
using wo 6-po 2-way sample loops (V-451, Upchu ch Scien i ic,
Oak Ha bo , WA, USA), each wi h a olume o 100 mL. Excess
luid was disca ded om he side ou le s also a a low a e o
250 mL min
21
.
Ma e ial. The luo escen ma ke o DNA 7-Aminoac ino-
mycin D (7-AAD), he syn he ic and ogen agonis me hyl ieno-
lone (R1881), 2,3-Bis(2-me hoxy-4-ni o-5-sul ophenyl)-2H- e a-
zolium-5-ca boxanilide(XTT), Giemsa s ain, 29-(4-
hyd oxyphenyl)-5-(4-me hyl-1-pipe azinyl)-2,59-bi-1H-benzimid-
azole ihyd ochlo ide hyd a e, bisBenzimide (Hoechs ), and
Lipopolysacha ide (LPS om Esche ichia coli, se o ype 026:B6)
we e pu chased om Sigma Ald ich (S ockholm, Sweden).
Te ame hyl hodamine, e hyl es e , pe chlo a e (TMRE) was
pu chased om In i ogen (S ockholm, Sweden). Cy okines we e
de ec ed using a Mul iplex ki om Meso Scale Disco e y
(Gai he sbu g, USA).
Cell cul u e. The mu ine mic oglial BV2 cells we e a kind
gi o D . Sand a Cecca elli and we e es ablished and i s
cha ac e ized 1992 by Bocchini e al. [22]. All Cell lines we e
cul u e in DMEM (GIBCO) and supplemen ed wi h 10% e al
bo ine se um (FBS) (Sigma-Ald ich), 55 IU ml
21
penicillin and
55 mgml
21
s ep omycin (Sigma-Ald ich). Fo he expe imen s
pe o med wi h he BV2 cells and LPS, he media was
supplemen ed wi h 5% FBS. All cells we e cul u ed a 37uCina
humidi ied a mosphe e con aining 5% CO
2
. The human p os a e
cance cell lines DU145, LNCaP, PC3 and VCaP we e all
ob ained om he Ame ican Type Cul u e Collec ion (ATCC)
and g own acco ding o ATCC ecommenda ions.
Cell iabili y in BV2 cells. The BV2 mic oglial cells we e
suspended (4 million cells mL
21
) in cul u e medium supplemen ed
wi h 2% se um o acili a e pheno ypic/in lamma o y changes o
he cells. A e he acous opho e ic p ocessing possible induc ion o
acu e cell dea h was examined by dye exclusion using ypan blue.
XTT assay. 5000 BV2 cells pe well we e seeded in 96-well
pla es a e mic ochannel acous opho e ic p ocessing and cul u ed
o 24 o 48 h. The XTT assay was pe o med ollowing he
manu ac u e’s ecommenda ion and he da a a e p esen ed as
pe cen age in espec o he cells exposed o mic o luidic handling
wi hou ul asound.
TMRE s aining. 20,000 BV2 cells we e seeded in 12-well
pla es o measu e hei Mi ochond ial po en ial (DY) a e
acous opho e ic p ocessing. A inal concen a ion o 25 nM o
TMRE s aining was added o he medium and he luo escence
was measu ed using low cy ome y (FACS Can oII, FACSDi a
so wa e, BD Biosciences, San Jose, USA).
Cell mig a ion. BV2 cells we e seeded in 6-well pla es a e
acous opho e ic p ocessing. Twen y- ou hou s a e seeding, a
yellow 100-ml-pipe e ip was used o d aw a line h ough he
con luen cul u e dish. A e addi ional 24 h o cul u e, he wid h
o he line, i.e. how much cells had mig a ed, was analyzed.
Hoechs s aining. BV2 cells we e seeded in 12-well pla es
and ixed in 4% PFA du ing 20 minu es a 37uC. The pla es we e
hen washed wice wi h PBS and Hoechs was added a a
concen a ion o 1 mgmL
21
in PBS o 10 min a oom
empe a u e. A e wo mo e washes wi h PBS, moun ing media
was added o he wells and he numbe o cells ha p esen ed
condensed nuclei was quan i ied using an in e ed mic oscope.
Fo each well a leas 300 cells we e coun ed.
Clonogenic assay. 1000 BV2 cells we e seeded in duplica es
in pe i dishes and 10 mL o medium was added o each dish.
A e h ee days, 5 mL o medium was exchanged wi h esh
media. A e se en days he media was ca e ully emo ed and he
cells we e ixed wi h 70% e hanol du ing 10 min a oom
empe a u e. A e he ixa ion he cells we e s ained wi h 10% ( /
) GIEMSA s aining o 15 min a oom empe a u e. The pla es
we e hen washed wi h MilliQ wa e and d ied a e which hey
we e scanned and analyzed using ImageJ 1.43 so wa e.
Immunoblo . A e acous opho e ic p ocessing, BV2 cells
we e seeded in 6-well pla es and a e one day exposed o LPS o
24 h. The cells we e hen washed wice in PBS and lysed on ice
using loading bu e . Then, he cells we e sonica ed 3 imes du ing
5 seconds each ime while kep on ice. All samples we e boiled o
4 min, esol ed on 10% SDS polyac ylamide gels a 100 V and
ansblo ed on o ni o-cellulose memb anes o 2 h a 30 V. The
memb anes we e blocked o a leas 1 h in a bu e (5% non- a
milk powde and 0.1% NaN3 in PBS, pH 7.4) and p obed wi h
p ima y an ibodies (dilu ed in 5% bo ine se um albumin and
0.1% NaN3 in PBS, pH 7.4) agains iNOS (San a C uz, sc-650;
1:1000) and b-ac in (Sigma, A5316 1:4000) o e nigh . The
memb anes we e hen washed in PBS con aining 0.1% Tween-
20 and incuba ed wi h he seconda y an ibody (dilu ed 1:5000 in
2.5% milk in PBS) o 1 h and washed 3 imes in PBS. The p o ein
bands we e isualized by ECL (Ame sham Biosciences) acco ding
o he manu ac u e ’s ins uc ions.
Cy okine analysis. A e acous opho e ic p ocessing, BV2
cells we e seeded in 96-well pla es. A e 24 h, cells we e exposed
Figu e 1. Pic u e and illus a ion o he se up. (a) Pho o o he
acous opho esis mic o luidic sys em i s p esen ed by Augus sson
e al. [27]. (b) C oss sec ional iew o cell dis ibu ion in he
mic ochannel wi hou ul asound (le ) and wi h ul asound o ming
an ul asound s anding wa e ( igh ). (c) Illus a ion o acouscous opho -
esis chip. Fo he p esen s udy, only one o he channel segmen s was
used allowing cells o be exposed o ul asound.
doi:10.1371/jou nal.pone.0064233.g001
Impac o Acous opho esis on Cells
PLOS ONE | www.plosone.o g 3 May 2013 | Volume 8 | Issue 5 | e64233
o LPS o 24 h and he supe na an s we e collec ed. The cy okine
p oduc ion was measu e using he Meso Scale Disco e y
elec ochemiluminescence ki (Th1/Th2 cy okines, Ul a-Sensi i e
ki , K15013C-1) using he manu ac u e ’s ins uc ions.
Cell iabili y in p os a e cance cell lines. DU145, PC3,
LNCaP and VCaP cells we e de ached by ypsin/EDTA and
washed wice in PBS and e-suspended in comple e cell g ow h
medium. Immedia ely a e acous opho e ic p ocessing (0 h)
ypan blue exclusion was used o es ima e he pe cen age o
iable cells. The cells we e coun ed in a Bu¨ ke chambe and
a e wa d seeded in 6-well pla es, (200,000 cells/well) and g own
o 24 o 48 h. Cells we e ha es ed wi h ypsin/EDTA, washed
wi h PBS and subsequen ly s ained wi h 7-AAD o 20 minu es.
Cell iabili y was es ima ed o 10,000 cells by low cy ome y
(FACS Can oII, FACSDi a so wa e, BD Biosciences), 7-AAD-
nega i e cells wi h high o wa d sca e s we e selec ed as iable.
PSA sec e ion. Cell lines (LNCaP and VCaP) exp essing he
and ogen ecep o (AR) we e ha es ed and subjec ed o he
acous opho e ic p ocessing. Unp ocessed cells we e used as a
con ol g oup. Cells we e seeded in 96-well pla es (10,000 cells/
well), and g own in comple e g ow h medium o 24 h in he
p esence o ei he R1881 (1 nM), o ehicle (PBS/DMSO). 80 mL
g ow h medium was emo ed om each well o PSA analysis.
The samples we e s o ed a 220uC un il analysis. PSA concen-
a ions we e measu ed using he DELFIA P os a us F ee/To al
PSA assay om Pe kinElme , Tu ku, Finland [23].
Human blood cell espi a ion. Th ombocy es and leuko-
cy es om h ee heal hy blood dono s we e subjec ed o
acous opho e ic p ocessing and mi ochond ial unc ion was
analyzed by high- esolu ion espi ome y as desc ibed in Sjo¨ all
e al. [24].
S a is ics. We used analysis o a iance (ANOVA) o
s a is ical compa isons be ween expe imen al g oups. Expe imen s
we e pe o med in iplica es o mo e. Da a a e exp essed as mean
6SD and P,0.05 was conside ed as signi ican ly di e en .
Resul s
Acous ic Ene gy Es ima ions
To es ima e he acous ic ene gy (E
ac
) a he di e en d i ing
ol ages, we used 7 mm diame e mic opa icles. Mic opa icles
we e used ins ead o cells do o hei homogenous size dis ibu ion,
which allows o a be e measu emen . The d i ing ol age
squa ed (U
pp
)
2
has a linea ela ion o he E
ac
in he ollowing
exp ession;
U2
pp
E
U2
~Eac~
p2
ako
4
whe e U
is he ansi ion ol age applied when he ansi ion om
pa icles exi ing h ough he side ou le s o exi ing om he cen e
ou le akes place, E
is he co esponding acous ic ansi ion
ene gy, p
a
is he acous ic p essu e ampli ude and k
o
is he
comp essibili y o he suspending liquid. Fo his pa icula chip,
using he 7 mm polys y ene mic opa icles and a o al low a e o
500 mL min
21
,U
2
was measu ed o be 147 V
pp
2
and E
was hen
es ima ed om simula ed pa icles ajec o ies o be 69 J/m
3
.
Th ough his exp ession E
ac
o he ol ages o 10 V
pp
and 20 V
pp
was calcula ed o be 47 J/m
3
and 188 J/m
3
, which co espond o
acous ic p essu e ampli udes o 0.639 MPa and 1.28 MPa,
espec i ely. These ene gy le els a e compa able o o he
acous opho esis mic ochip p ocessing de ices, which commonly
ope a e a acous ic ene gy densi y le els o 10–100 J/m
3
[25,26].
See Augus sson e al. 2012 o a mo e de ailed desc ip ion o he
acous ic ene gy densi y es ima ion [27].
Acous ic Cell P ocessing
Mic ochannel acous opho e ically p ocessed cells may be
damaged due o hyd odynamic shea , ul asonic exposu e o high
empe a u es. To a oid cell damage due o high empe a u es, we
e alua ed he e ec o acous opho esis on cell iabili y in a
con olled empe a u e en i onmen (37uC). Hyd odynamic shea
damage was also excluded by using cells p ocessed h ough he
chip wi hou ul asound exposu e as con ols.
In he acous opho esis mic ochannel, cells we e ocused in he
cen e when ope a ed a a hal wa eleng h esonance a 2 MHz.
A ampli ude o 10 V
pp
, all cells we e ocused in he cen e o he
mic ochannel. Ampli ude o 20 V
pp
was also employed o gene a e
a s onge acous ic ield han needed o ocus cells, and he eby
enable he s udy o possible ha m ul e ec s on he cells when
exposed o ele a ed acous ic adia ion o ces. The se up allowed
maximum 25 V
pp
, howe e using his high ol age gene a ed
empe a u e luc ua ions abo e he empe a u e c i e ia o 37
˚C.
We es ed he e ec 25 V
pp
on immedia e cell dea h ( ypan blue
exclusion) using he BV2 mic oglial cells wi hou de ec ing any
ad e se e ec (da a no shown). As a con ol, cells we e p ocessed
h ough he chip wi hou an ac i e ul asound ( unc ion gene a o
se o 0 V
pp
). In o de o s udy he e ec o acous opho esis on cell
su i al and elucida e possible ad e se e ec s when using
acous opho esis o cell so ing, a s anda d acous opho esis chip
was used ( ig. 1, desc ibed in Ma e ial and Me hods). Cells we e
passed h ough he chip a a low a e o 100 mL min
21
co esponding o a e en ion ime o abou 1 second in he
acous ic ield, a e en ion ime ele an o mos acous opho e ic
p ocessing.
Acous ic Cell P ocessing in a Mic ochannel does no Al e
Cell Dea h and Viabili y o Mic oglial Cells
Viabili y and cell dea h analysis. We se ou o in es iga e
possible dele e ious e ec s o acous opho esis by using he
mic oglia cell line BV2, an immo alized mouse mic oglial cell
used widely and success ully in medical esea ch [28]. Using
ypan blue exclusion, no ad e se e ec s on he acu e cell dea h
induced by he acous opho esis handling could be de ec ed (C l,
cells no en e ing he acous opho esis chip 9.765.7% cell dea h; 0
V
pp
, 11.468.8% cell dea h; 10 V
pp
, 8.967.0% cell dea h; 20 V
pp
,
11.666.9% cell dea h).
Nex , we s udied possible long- e m cellula al e a ions beyond
he acu e e ec o acous opho esis. A e going h ough he
acous opho esis chip, BV2 cells we e seeded again o 24 h o 48 h
o in es iga e possible long- e m e ec s. Fi s , we used XTT assay
(mi ochond ial dehyd ogenase ac i i y) o s udy a possible
al e a ion in he iabili y/p oli e a ion o acous opho esis. How-
e e , no di e ences in iabili y o he cells exposed o he di e en
condi ions could be de ec ed in he XTT assay a ei he 10 o 20
V
pp
o acous ic ocusing a any ime poin , 24 o 48 h, a e
seeding ( ig. 2a). To u he elucida e he e ec o acous opho esis,
BV2 cells we e PFA- ixed hen Hoechs s ained a 24 and 48 h
and he numbe o cells unde going cell dea h, con aining
apop o ic bodies, we e quan i ied. Despi e a p og essi e cell dea h
om 24 o 48 h, we did no ind any changes in numbe o cells
unde going apop osis due o he acous opho e ic p ocessing
( ig. 2b). Fu he mo e, we did no de ec any changes in he
mi ochond ial po en ial (indi ec ly measu emen o cells a isk o
unde going apop osis) by TMRE due o acous opho esis ( ig. 2c).
Simila ly, he cell mig a ion speed as desc ibed p e iously [29],
was unal e ed a e acous opho e ic p ocessing a 24 o 48 h a e
Impac o Acous opho esis on Cells
PLOS ONE | www.plosone.o g 4 May 2013 | Volume 8 | Issue 5 | e64233
cul u ing (da a no shown). The cellula changes o e a longe
ime pe iod (7 days) was also assessed o s udy a possible al e a ion
in cell u no e using a clonogenic assay howe e , no ad e se
e ec we e de ec ed ( ig. 2d) simila o ou ea lie ime poin s (24
o 48 h).
In lamma o y esponse. Al hough we could no de ec any
acous opho esis-induced al e a ions on cell dea h o cell iabili y,
mino cellula al e a ions and pheno ypic changes canno be uled
ou . In his aspec , in lamma o y mic oglial cells a e sui able cells
o de ec sub le di e ences in hei in lamma o y esponse due o
he exposu e o acous ic o ces. We he e o e explo ed he
possibili y ha acous opho e ic p ocessing could al e he in lam-
ma o y esponse igge ed by a p oin lamma o y esponse.
Lipopolyssacha ide (LPS) is a lipoglycan ound in he ou e
memb ane o G am-nega i e bac e ia and a e y po en Toll-like
ecep o 4 (TLR4) ligand and induce o in lamma ion. LPS-
Figu e 2. Unal e ed iabili y o BV2 mic oglial cell line ollowing acous opho e ic p ocessing (10V
pp
and 20V
pp
). BV2 cells we e passed
h ough he acous opho esis chip wi h he unc ion gene a o se a 0, 10 and 20 V
pp
. A e going h ough he chip, he BV2 cells we e seeded again
o 24 and 48 h. Cell iabili y was measu ed by XTT (a), apop o ic nuclei appea ance (b) and dec ease o mi ochond ial po en ial -Ym- (c), showing no
di e ence be ween expe imen al g oups. Simila , no di e ence was de ec ed by clonogenic assay (d) used o s udy su i al and p oli e a ion a 7
days ollowing acous opho e ic p ocessing. The g aphs show he esul s om a leas h ee sepa a e expe imen s and he da a a e shown as means
6SD. Signi icance alue P,0.05, ns deno es non-signi ican .
doi:10.1371/jou nal.pone.0064233.g002
Impac o Acous opho esis on Cells
PLOS ONE | www.plosone.o g 5 May 2013 | Volume 8 | Issue 5 | e64233

challenged BV2 cells we e analyzed o he p oduc ion o
inducible ni ic oxide syn hase, iNOS, ( ig. 3a,b) ha is highly
induced unde in lamma o y s imulus (such as LPS). The medium
was also collec ed and used o measu e he mic oglial elease o
p oin lamma o y cy okines (IL-1b, IL-12 and TNF-a; ig. 3c,d,e)
and an i-in lamma o y molecules (IL-10; ig. 3 ). We ound ha ,
bo h he p o ein exp ession o p oin lamma o y iNOS and he
elease o p o- and an i-in lamma o y cy okines upon LPS-
challenge we e unal e ed a e exposu e o acous opho e ic
p ocessing ( ig. 3).
In luence o Acous opho esis on Viabili y o Sec e o y
Func ion o P os a e Cance Cells
Acous opho esis-based mic o luidic de ices a e cu en ly being
explo ed as a mean o en ich and disc imina e iable om non-
iable umo cells in pe iphe al blood [27]. The e o e, we ini ia ed
a de ailed cha ac e iza ion o he iabili y and sec e o y unc ion
o umo cells subsequen o subjec ing hem o acous opho e ic
p ocessing. We i s cha ac e ized he p os a e cance cell lines
DU145, PC3, LNCaP in e e ence o cell iabili y/p oli e a ion
using he same XTT assay ha was used o he iabili y es o he
BV2 mic oglial cells. We did no ind any changes in cell iabili y
a ei he 24 o 48 h a e acous opho e ic p ocessing (da a no
shown). Subsequen ly, we de e mined he ex en o cell dea h
using ypan blue exclusion di ec ly ollowing acous opho e ic
p ocessing wi hou obse ing any change in cell dea h o he
di e en p os a e cance cell lines, DU145, PC3, LNCaP o
VCaP. Finally, we examined cell dea h a 24 and 48 h using
lo escence ac i a ed cell so ing (FACS) and ga ed cells o 7-
AAD-up ake. Again, simila o ou esul s abo e, no ad e se e ec
ollowing acous opho e ic p ocessing when compa ed o unp o-
cessed cells was iden i ied ( ig. 4a–d).
We wen on o de e mine whe he acous opho esis al e s he
unc ional p ope ies o iable cells p ocessed h ough he
mic ochip. PSA sec e ion is a sensi i e indica o o and ogen
ecep o (AR) signaling ac i i y in AR-dependen umo cells. We
used he p os a e cance cell lines LNCaP and VCaP and s udied
bo h endogenous PSA sec e ion as wells as sec e ion ollowing
adminis a ion o he syn he ic and ogen R1881. We did no ind
any signi ican di e ences in PSA sec e ion in umo cells
subjec ed o acous opho e ic p ocessing compa ed o unp ocessed
cells, nei he a e adminis a ion o R1881 no o uns imula ed
cells. The e we e no measu able di e ences in e e ence o cells
p ocessed h ough he mic ochip in absence o ul asound o when
hey we e exposed o ul asound a inc easing ol age ( ig. 5a–b).
Mi ochond ial Respi a o y Func ion o Human Blood
Cells is no Al e ed Following Acous opho esis
Acous opho e ic p ocessing o human blood is an eme ging
esea ch ield o acous ic cell manipula ion in mic ochannels. To
u he elucida e possible cellula al e a ion due o acous opho -
esis, we exposed h ombocy es and leukocy es isola ed om
human blood o he same acous ic ea men used o he BV2
mic oglial cells and p os a e cance cells desc ibed abo e. High-
esolu ion espi ome y was pe o med o e alua e mi ochond ial
espi a o y unc ion and in eg i y. Fu he mo e, endogenous
oxygen consump ion was e alua ed in in ac cells (da a no
shown), and espi a ion du ing maximal oxida i e phospho yla ion
(Oxphos) and maximal lux h ough he elec on anspo sys em
was assessed using a mul iple subs a e/inhibi o i a ion p o ocol
in digi onin-pe meabilized cells.
We s udied he maximal espi a ion du ing Oxphos using bo h
complex I- and complex II-linked subs a es o h ombocy es
( ig. 6a) and leukocy es ( ig. 6b) and ound no e ec ollowing
acous opho e ic p ocessing. We u he in es iga ed he emaining
espi a o y ac i i y ollowing inhibi ion o ATP syn hase wi h
oligomycin, so called LEAK o s a e 4 espi a ion, and ound no
di e ences in espi a o y ac i i y ollowing passage h ough he
acous opho esis mic ochip. We also checked o di e ences in
o he espi a o y s a es, wi hou de ec ing any al e a ions due o
acous opho e ic p ocessing (da a no shown). Any p ocess in e -
e ing wi h me abolic pa hways leading o espi a o y inhibi ion
would be eadily de ec ed a maximal Oxphos espi a ion.
Simila ly, any p ocess in e e ing wi h inne memb ane in eg i y
o inc eased u iliza ion o p o on mo i e o ce o o he pu poses
han ADP phospho yla ion would be eadily de ec ed a LEAK/
s a e 4 espi a ion. Thus, he e is no indica ion o any nega i e
mi ochond ial e ec o an inc ease in ene gy demand in he cells
by acous opho e ic cell so ing.
Discussion
The acous opho e ic echnology allows non-con ac cell han-
dling based on in insic cell p ope ies ha includes size, densi y
and comp essibili y which enables he de elopmen o no el label-
ee cell so ing s a egies [27]. Howe e , ex ensi e analysis on he
e ec s o so ing me hodology a e necessa y no only o a
echnique o be es ablished in clinical se ings bu also o assu e
ha esea ch esul s do no ge ain ed due o a ha sh so ing
echnology. Fo a cell handling o so ing echnologies o be alid
o clinical o esea ch pu poses, hey need be non-pe u bing o
he cells p ocessed. Hence, in his s udy we ha e pe o med a
comp ehensi e s udy wi h di e en cells using a ious echniques
aimed a de e mining any cellula al e a ions ha acous opho e ic
p ocessing could induce. Ou conclusions in his s udy a e ele an
o acous opho e ic so ing o cells in mic o luidic sys ems a low
a e in he ange up o 100 ml/min (channel c oss-sec ions
375*150 mm) and wi h e en ion imes in he acous ic s anding
wa e o abou 1 second, a p essu es up o 1.28 MPa, a an
ac ua ion equency o 2 MHz.
The e ec o acous opho e ic p ocessing on mic oglial cells was
i s elucida ed. These cells a e he in lamma o y su eillance cells
o he b ain simila o mac ophages in o he issues [30]. Mic oglia
espond easily o sub le changes in he mic oen i onmen ,
including exposu e o adia ion [31], nanopa icles [32] and
mino empe a u e changes [33]. The mu ine mic oglial cell line
BV2 is an in lamma o y cell line widely used in medical esea ch
[28]. We i s used his cell line and ound no changes in su i al/
apop osis p oli e a ion o mig a ion a 24 h, 48 h o 7 days a e
acous opho e ic p ocessing using bo h manual cy ological analysis
and low cy ome y (TMRE). A e demons a ing ha acous o-
pho esis did no al e aspec s o iabili y, we u he de e mined i
acous opho esis could induce any changes in he in lamma o y
s a e o he cells. Mic oglia exposed o acous opho e ic p ocessing
we e also s imula ed wi h LPS ac i a ing he pa e n ecogni ion
ecep o TLR4. This ecep o is known o ac i a e he MyD88 o
TRAM/TRIF pa hways. S imula ion o hese pa hways will
igge in acellula pa hways ac i a ing in lamma o y ansc ip-
ion ac o s (IRF/NFkB/CREB/AP-1) ha will lead o iNOS
ac i a ion and cy okine elease. Ou da a show no al e a ion in
iNOS and cy okine esponse ollowing acous opho e ic p ocessing
sugges ing ha hese in lamma o y pa hways a e no a ec ed by
acous opho esis. In iew o he suscep ible ac i a ion o mic oglial
cells by a ious s imulus, ou da a sugges ha acous opho esis is a
mild cell-handling echnology.
P os a e cance is he mos commonly diagnosed cance in men
and has he second highes dea h incidence [34]. Howe e , he
Impac o Acous opho esis on Cells
PLOS ONE | www.plosone.o g 6 May 2013 | Volume 8 | Issue 5 | e64233
Figu e 3. In lamma o y esponse o BV2 cells upon LPS challenge ollowing acous opho esis is no changed. A e acous opho e ic
p ocessing, BV2 cells we e seeded and s imula ed he nex day wi h LPS o 24 h. We obse ed no al e a ion due o acous opho e ic p ocessing in he
exp ession o iNOS (inducible ni ic oxide syn hase)(a,b), he elease o p oin lamma o y cy okines IL-1b(x), IL-12 (d), TNF-a(e) o he an i-
Impac o Acous opho esis on Cells
PLOS ONE | www.plosone.o g 7 May 2013 | Volume 8 | Issue 5 | e64233
molecula backg ound o p os a e umo p og ession is cu en ly
no comple ely unde s ood. Mos p os a e cance cells equi e
and ogen s imula ion o g ow h and p oli e a ion and and ogen
dep i a ion he apy is he i s line o ea men o mos p os a e
cance pa ien s wi h dissemina ing disease. Al hough e ec i e in
educing umo g ow h and size, he ea men selec s o
and ogen independen cance cells, esul ing in umo s esis an
o and ogen he apy and p og ession in o cas a ion- esis an
p os a e cance [35]. Molecula in e oga ion o isola ed ci cula -
ing umo cells would p o ide aluable in o ma ion on umo
p og ession and me as asis. I is c ucial o u u e cell e alua ion,
ha he sepa a ion echnology’s in insic p ope ies do no al e
in lamma o y cy okine IL-10 ( ). The g aphs show he esul s om a leas h ee sepa a e expe imen s and he da a a e shown as means 6SD.
Signi icance alue P,0.05, ns deno es non-signi ican .
doi:10.1371/jou nal.pone.0064233.g003
Figu e 4. P os a e cance cell iabili y is no a ec ed by acous opho esis. Cell iabili y was de e mined a e acous opho esis (0, 10 and 20
V
pp
), o ou p os a e cance cell lines: DU 145 (a), PC3 (b), LNCaP (c) and VCaP (d). Le panels show he pe cen age o cell dea h di ec ly a e
acous opho esis measu ed by ypan blue exclusion. Righ panels show he cell dea h quan i ied a 24 and 48 h a e acous opho esis by low
cy ome y. Cells nega i e o he DNA binding luo och ome 7-Aminoac inmycin (7-AAD) we e de ined as iable. A leas 10,000 cells we e coun ed
and he pe cen age o dead cells was de e mined. Cells no subjec ed o acous opho esis we e used as con ol cells. The g aphs show he esul s
om a leas h ee sepa a e expe imen s and he da a a e shown as means 6SD. Signi icance alue P,0.05, ns deno es non-signi ican .
doi:10.1371/jou nal.pone.0064233.g004
Impac o Acous opho esis on Cells
PLOS ONE | www.plosone.o g 8 May 2013 | Volume 8 | Issue 5 | e64233
Figu e 5. Acous ic cell sepa a ion in a mic ochannel does no al e PSA sec e ion by p os a e cance cells. The and ogen ecep o (AR)
exp essing cell lines LNCaP and VCaP we e used o e alua e he impac o acous opho esis on PSA sec e ion. A e acous opho esis un a 0, 10 and
20 V
pp
, he sec e ion o PSA was measu ed in he absence o p esence o he AR ligand R1881 (1 nM o 24 h) in he LNCaP cell line (a) and in he
VCaP cell line (b). Cells no p ocessed h ough he chip we e used as con ol cells. The g aphs show he esul s om h ee sepa a e expe imen s and
he da a a e shown as means 6SD. Signi icance alue P,0.05, ns deno es non-signi ican .
doi:10.1371/jou nal.pone.0064233.g005
Figu e 6. Mi ochond ial espi a o y unc ion in human leukocy es and h ombocy es a e no al e ed ollowing acous opho esis.
Leukocy es and h ombocy es we e passed h ough he acous opho esis chip un a 0, 10 and 20 V
pp
. Maximal espi a ion du ing using bo h complex
I- and complex II-linked subs a es o h ombocy es (a) and leuko cy es (b) was unal e ed a e acous opho esis, suppo ing ha acous opho esis
does no a ec me abolic pa hways impo an o espi a ion. The emaining espi a o y ac i i y ollowing inhibi ion o ATP syn hase wi h
oligomycin, so called Leak o s a e 4 espi a ion, was also una ec ed by acous opho esis, in h ombocy es (c) and leukocy es (d). This da a con i m no
e ec o acous opho esis on inne mi ochond ial memb ane in eg i y o changed u iliza ion o p o on mo i e o ce o o he pu poses han ADP
phospho yla ion. Unp ocessed cells we e used as con ol cells. The g aphs show he esul s om h ee sepa a e expe imen s and he da a a e shown
as means 6SD. Signi icance alue P,0.05, ns deno es non-signi ican .
doi:10.1371/jou nal.pone.0064233.g006
Impac o Acous opho esis on Cells
PLOS ONE | www.plosone.o g 9 May 2013 | Volume 8 | Issue 5 | e64233