A NONINVASIVE METHOD FOR COUNTING HUMAN PLURIPOTENT STEM CELL NUMBERS BY LIVE CELL IMAGING Mika Suga 1, Hiroaki Kii 2, Takayuki Uozumi 2, Yasujiro Kiyota 2 and Miho K Furue 1

1 Dept. of Disease Bioresources Research, Laboratory of Stem Cell Cultures, National Inst. of Biomedical Innovation, Osaka, Japan 2 Instruments company, NIKON CORPORATION, Yokohama, Japan

Background Celldensity isacritical factorforcontrollingbothgrowthanddifferentiationof

humanpluripotentstem(hPS)cellsincludingembryonicstem(ES)cellsandinduced

pluripotentstem (iPS)cells.Despite the factof theevolutionofhPScellculture

techniques,countingcellnumbersisstillproblematic.Therefore,wehavedeveloped

anoninvasivecellcountingmethodforhPScellsthroughanalyzinglivecellimages.

Outline AJapaneseadulthumanskin fibroblast (JCRB0534:TIG-114) -derived iPS-

TIG114-4f1human iPScells (JCRB1437,JCRBCellBank,NIBIO,Osaka,Japan),

whichwasestablishedbyYamanaka’sgroup (CiRA,KyotoUniversity)1)2) and

H9humanEScells (WA09,WISCBank,WiCellResearch Institute,Madison,

WI,USA)3)4)wereseededon inactivatedmouseembryonic fibroblasts (MEF)as

describedpreviously 4), and thenstainedwithacell-permeantSYTO24green

fluorescentnucleicacidstain(Invitrogen).Phasecontrastandfluorescentimagesof

thecellsobtainedintermittentlyinacellincubatorobservationsystem,BioStation CT (NikonCorporation)wereanalyzedbyautomated imageanalysissoftware

CL-Quant(NikonCorporation).AreaofES/iPScellcolonycoveragewasmeasuredfromphasecontrast images.Numberofstainednucleiwascounted fromgreen

fluorescent images. Immediatelyafter imaging, theconventionalcellcountingby

hemocytometerwasperformedtocomparewiththenumbersoffluorescentnuclei.

Materials and Methods Cell culture H9and iPS-TIG114-4f1weremaintainedon irradiatedmouseembryo

fibroblastfeedercells(MEF,MilliporeCo.) inanKSR-basedmediumsupplemented

with5ng/mlhumanrecombinantFGF-2(KatayamaKagakuKogyoLTD.).H9were

passagedwith1mg/mlDispase(Roche)andaplasticscraper (SumitomoBakelite

Co.LTD.).iPS-TIG114-4f1weremechanicallypassagedwithEZPassage(Invitrogen).

Thecellsweresplitataratioof1:5–1:8every6days.HumanEScellswereused

followingtheGuidelinesforutilizationofhumanembryonicstemcellsoftheMinistry

ofEducation,Culture,Sports,ScienceandTechnologyofJapan.Thisresearchusing

hES/iPScellswasapprovedby the institutionalethical reviewboardatNational

InstituteofBiomedicalInnovation.

Live cell imaging Thecellsseededona6-wellplatewereculturedintheBioStation

CTat37°C/5%CO2,andmonitoredby time lapse live imaging.Phasecontrast

imageswerecapturedevery12hoursautomaticallyatamagnificationof4x.

Todetectnucleiof livingcells, thecellswerestainedwith1µMSYTO24green

fluorescentdye (Invitrogen).Fluorescent imagesof thestainednucleiandphase

contrastimageswereobtainedbyBioStationCT,andanalyzedbyCL-Quantwiththe

custommaderecipe.Thisrecipeenablestodetectindividualcoloniesandmeasure

theareacoveredbythesecoloniesfromthephasecontrastimageandmeasurethe

numberofnucleifromthefluorescentimage.

Conventional cell counting Thecellsweredissociatedusing0.025%trypsinand

0.01%EDTA inPhosphateBufferedSaline (Gibco) andcountedbydisposal

hemocytometer,OneCellCounter(BioMedicalScience,LTD.).

Results Thefluorescentnucleuscountingwithimagesandtheconventionalcellcountingbyhemocytometershowedsimilarresults.Inthecaseoftotalcellnumbersabove

1x104,cellnumbersbynucleuscountingweresimilarandreproduciblewiththosebyconventionalcellcounting.(seeFigure1.)

Therewasasignificantcorrelationbetweenthecolonycoverageareaandthenucleus/cellcounting.Thecorrelationcurveswerespecificforeachcellline.(seeFigure2.)

Figure 1. Daily cell growth of hPS cells was quantified by imaging.

(A)Thegreenfluorescentandphasecontrast imagesof livinghPScellsstained

withSYTO24. (Left)Merged imageofSYTO24fluorescenceandphasecontrast.

(Right)DetectionofhPScoloniesfromphasecontrastimage.NotethatMEFswere

separatedfromhPScolonies.(B)Eachnucleus(right,markedasgreenpoints)was

detectedandsegmentedfromtheSYTO24fluorescentimage(left)usingCL-Quant.

Allobjectswithinthewellweremeasuredtocountcellnumbers. (C)Comparison

ofcellcountingbetweenconventionalmethod (hemocytometer)and the image

analysis(nucleistaining).(D)Dailyincreaseofcolonycoverageareawasquantified

byimageanalysis.

Figure 2. Colony coverage area was analyzed from phase contrast images.

(A)Colonycoverageareawasanalyzedfromphasecontrastimages.Yellowregion

isthedetectedhPScolonies.MEFsweresuccessfullyremovedbyourhPScolony

extractingprocedure. (B)Correlationbetweencellcountbynucleistainingand

colonycoveragearea.Centerarea(7680x7820pixel;240.23mm2)ofeachwellwas

analyzed.NotethatthecorrelationcurvesaredifferentbetweeniPS-TIG114-4f1and

H9.

Conclusions Hereweshowedasignificantcorrelationbetweenthecellnumberscountedbynucleifluorescentstainingandthecolonycoverageareameasuredfromphasecontrastimaging.Theseresultsshow

thatnumbersofhumanPScellscanbeestimatedfromthetotalcolonycoverageareathroughphasecontrastimaging.Thuswehavedevelopedanewnoninvasivecellcountingmethod.Furthermore,

obtainingtime-lapsephasecontrastimagesenablesustomonitorcolonymorphologicalchangesandtocalculategrowthrateduringhumanPScellculture.Foraccurateestimationofcellnumbers,itis

neededtodeterminethecorrelationcurveofthecellnumbersandthecolonycoverageareaforeachhumanPScelllines.ThiscorrelationratiomightcharacterizeasthepropertyofeachhPScelllines.

Ournoninvasivetechniquecouldbeusefulforqualitycontrolorhigh-throughputscreeninganalysiswithoutwastingordamagingthecells.

References1)Takahashi,et al.(2007)Cell30;131(5):861-72.2)ISCI(2011)Nat Biotechnol. 27;29(12):1132-44.3)Thomson,et al.(1998)Science282:1145-1147.4)Amit,et al.(2000) Developmental Biology227:271-278.

Acknowledgement ThisresearchissupportedbyNEDO(NewEnergy

andIndustrialTechnologyDevelopmentOrganization)

ontheproject“FundamentalTechnologyforPromoting

theIndustrialApplicationofHumanStemCells”.

1000000

100000

10000

10000 20 40 60 80 100 120 140

y = 20.754x2 + 794.39xR2 = 0.9748

y = 14.293x2 + 438.19xR2 = 0.9777 iPS-TIG114-4f1

H9(per

240

.23

mm

2 )C

ell c

ount

Colony coverage area (mm2)

iPS-TIG114-4f1

H9

(A)(A)

(B)

(B)

(D)

(C)

1×107

1×106

1×105

1×104

1 2 3 4 5 6 7 8 9

5×108

5×107

5×106

5×105

1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 Day

Day1 2 3 4 5 6 7 8 9

hemocytometer

nuclei staining

cell

coun

t p

er w

ell

colo

ny c

over

age

area

(pix

el p

er w

ell)

iPS-TIG114-4f1 H9

iPS-TIG114-4f1 H9