Automated Long-term 3D Cell-Based Toxicity Studies Using a Flowchip System

Evan F Cromwell1; Michelle Leung1; Anthony Thai1; Cathy Olsen2; and Oksana Sirenko2

1Protein Fluidics, Inc. 875 Cowan Rd, Suite B, Burlingame CA 94010, 2Molecular Devices, LLC, 3860 N First Street San Jose, CA 95134

INTRODUCTIONThere is an increasing interest in using three-dimensional (3D) cell structures for modelingtumors, organs, and tissue to accelerate translation research.1 Significant progress has beenmade in formation of such structures to recapitulate the in vivo environment but performingcomplex assays with them can be challenging. For example, manual treatment, staining, andprocessing of spheroids and organoids is typically labor-intensive and prone to disruption orloss of samples. Here we report on use of a novel microfluidic-based system (Pu·MA System®)to perform automated assays with spheroids, organoids, and microtissues. It is ideal forapplications for long-term toxicity, oncology therapeutics, single organoid secretion, andmetabolite sampling.

Pu·MA System flowchip: This flowchip has sample chambers connected to multiple reservoirsto enable media and buffer exchange, supernatant sampling, and in situ lysing. A key aspectto the device is a protective chamber that holds 3D cell models and allows reagent exchangeswithout disrupting the structures.

Spheroid Toxicity & Imaging: Capabilities of the system were demonstrated by assayingspheroid toxicity over a 48-hour period with HCT116 spheroids (colon cancer cell line).Spheroids were incubated in Pu·MA System flowchips along with media or “media +compound”. After incubation, the spheroids were analyzed by ImageXpressTM Micro ConfocalImaging System for Live/Dead cells and by CellTiter-Glo luminescence using a SpectraMax® iD5Plate Reader for ATP activity.

Single Spheroid Metabolomics: Ability to perform single spheroid metabolomics was alsodemonstrated with HCT116 spheroids. Spheroids were incubated 2 hours with compoundsand then lysed in situ. Spheroid lysate was analyzed for a panel of metabolites using massspectrometry (Children’s Medical Center Research Institute Metabolomics Facility at UTSouthwestern) showing significant differences between Treated and Untreated spheroids.

CELL VIABILITY ASSAY

CONCLUSIONS• We have demonstrated capabilities of a novel automated 3D Cell-based assay system that

performs complex protocols with spheroids in an incubator environment.

• Long term toxicity was assayed using high resolution confocal imaging of spheroidsincubated with various compounds and Live/Dead staining and analysis.

• The Pu·MA System has also been shown to perform in situ lysing of spheroids to providehighly sensitive metabolomic profiling with minimal perturbation.

• The ability to analyze spheroids and organoids in situ in order to capture toxicityinformation and metabolomic profiles shows great promise for oncology research.

Figure 1. Live Dead Analysis of HCT 116 Spheroids incubated for 24 hours with and without 100 nM Staurosporine. Cells were stained with a mixture of DAPI, Calcein AM, and EthD.

INSTRUMENTATION

AUTOMATED REAGENT EXCHANGEThe Pu·MA System can automatically perform complex assay steps using proprietarymicrofluidics. All reagents are loaded into flowchips and then incubation, media exchanges,cell secretion sampling and other steps are executed by the system program.• 3D Cell models remain in the protected in bottom chamber• Can exchange up to 95% of media without drying cells• No direct fluid flow over cells to disrupt cell structures

1Three-Dimensional in Vitro Cell Culture Models in Drug Discovery and Drug Repositioning, Langhans, S.A. Frontiers in Pharma. 2018, 9, 1.2High-Content Assays for Characterizing the Viability and Morphology of 3D Cancer Spheroid Cultures, Sirenko, O. et al. Assay and Drug Dev Tech. 2015, 13, 402.3Metabolomics and Metabolic Diseases: Where DoWe Stand? Newgard, C.B. Cell Metabolism 2017, 25, 4380.4Cancer cell specific inhibition of Wnt/β-catenin signaling by forced intracellular acidification, Melnki, S. et al., Cell Discovery, 2018, 4, 37

References:

Figure 7. Relative amount of ATP in Control and Treated HCT116 spheroids as determined by luminescence signal. Error bars = +/- 1 StDev (n=3)

Figure 6. Top: Metabolite signals from HCT116 spheroids. Error bars represent +/- 1 SD (n=3). Bottom: Confocal images of representative spheroids and Principle Component Analysis of results.

• HCT116 cells were dispensed 2,000 cells per well (384-well ULA round bottom plate,Corning) and incubated for 48 hours until they formed tight spheroids.2

• The spheroids were transferred to a Pu·MA System flowchip in either media (Control) ormedia + compound (Treated).

• Flowchips were placed into a Pu·MA System in an incubator (37°C & 5% CO2 and thenautomatically processed. At the end of the compound incubation, spheroids were stainedand washed prior to imaging.

• Imaging was done with an ImageXpress Micro Confocal Imaging system and analysisperformed using the Custom Module Editor with MetaXpress imaging and analysis software.

• Incubations can be performed for 1 to 5 days with up to 3 media exchanges with the currentformat.

SPHEROID TOXICITY & IMAGING

The Pu·MA System and flowchips includes:

• Automated media exchanges occur with cells in protected chamber

• Supernatants can be collected to monitor cell secretion

• Cells can be lysed in situ for sensitive metabolomic profiling

• Spheroids can be imaged in the flowchip, or samples removed for immunoassay or metabolomics analysis

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

Control Staurosporine (100 nM)

Perc

en

t N

um

ber

of

Cel

ls

HCT116 Spheroids Live/Dead Analysis

Live Dead

• 20 µl fluid transfers in 8 lanes for 3 differentflowchips (Source Well to Sample Well)

• Fluid was transferred to a Sample Well,incubated for 30 min, and then transferredback to the Source well.

• Finale volume in the Source well wasmeasured gravimetrically

• Final Volume CV = 3%

Cell viability was measured for HCT116spheroids incubated with compounds for 48hours in Pu·MA System flowchips. (Compounds:C59 30 mM; Mitomycin 30 mM; Etoposide 100mM).

Amount of ATP present in each spheroid samplewas measured using the CellTiter-Glo assay(Promega). 10 µl of CellTiter-Glo reagent wasautomatically transferred into the sample wellsand incubated for 10 min. Luminescence signalwas measured using a SpectraMax iD5 PlateReader (Molecular Devices).

SINGLE SPHEROID METABOLOMICS

Control

C59 30µM

1.E-07

1.E-06

1.E-05

1.E-04

1.E-03

1.E-02

Ad

eno

sin

e

Ad

eno

sin

e m

on

op

ho

sph

ate

S-A

de

no

sylm

eth

ion

ine

Ad

eno

sin

e d

iph

osp

hat

e (A

DP

)

Ace

tyl-

L-am

ino

adip

ate

L-G

luta

thio

ne

oxi

diz

ed

NA

D+

UD

P-N

-ace

tyl-

D-h

exo

sam

ine

Ad

eno

sin

e tr

iph

osp

hat

e (

ATP

)

P-D

MEA

2'_

3'-

Cyc

licC

MP

Gu

anin

e

UD

P-H

exo

se

Cyt

idin

e m

on

op

ho

sph

ate

Cyc

licA

DP

-rib

ose

Gu

ano

sin

e 5

'-d

iph

osp

hat

e (G

DP

)

Uri

din

e5'-

dip

ho

sph

ate(

UD

P)

Ino

sin

e

Gu

ano

sin

e

Gly

cera

lde

hyd

e 3

-Ph

osp

hat

e

Glu

-Asp

Gu

ano

sin

e m

on

op

ho

sph

ate

(GM

P)

Ace

tyl-

hex

osa

min

e-p

ho

sph

ate

Cyt

idin

e 5

'-d

iph

osp

ho

cho

line

UTP

Oro

tic

acid

CD

P-e

than

ola

min

e

Ino

sin

e-5

'-m

on

op

ho

sph

ate

(IM

P)

NA

DP

+

Gu

ano

sin

etri

ph

osp

hat

e(G

TP)

Deo

xyp

ento

se-5

-ph

osp

hat

e-1

Dih

ydro

xyac

eto

ne

ph

osp

hat

e

2-(

a-D

-Man

no

syl)

-3-p

ho

sph

ogl

ycer

ate

Fru

cto

se/G

luco

se 1

,6-b

isp

ho

sph

ate

Cys

tein

ylgl

ycin

e(C

ys-G

ly)

Pen

tose

-ph

osp

hat

es

Ad

enyl

succ

inic

aci

d

D-S

edo

he

ptu

lose

7-p

ho

sph

ate

Dip

tham

ide

Met

hyl

guan

ine

Ph

osp

ho

en

olp

yru

vic

acid

2-A

ce-N

-β-a

sp-2

-deo

xyh

exo

pyr

ano

syla

min

e

3-H

ydro

xyca

rnit

ine

(1

8:1

)

Thym

idin

e 5

'-m

on

op

ho

sph

ate

Deo

xy-h

eptu

lose

-7-p

ho

sph

ate

O-P

ho

sph

ory

leth

ano

lam

ine

Tyro

sylp

hen

ylal

anin

e

Ad

eno

sin

e d

iph

osp

hat

e ri

bo

se

CM

P-2

-am

ino

eth

ylp

ho

sph

on

ate

Car

nit

ine

(1

6:0

)

Car

nit

ine

(1

6:1

)

Ph

osp

ho

glyc

eric

aci

d

Untreated C59 Wnt Inhibitor Staurosporine

-0.8

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

Spheroids Metabolomic Results with in situ Lysing - PCA

C59

Control

Stauro

Colon cancer spheroids (2K cells) were incubated at 37°Cwith compounds (C59 Wnt inhibitor4 and Staurosporinekinase inhibitor) for 2 hours and then automatically lysedby the Pu·MA System with 80% acetonitrile solution.Over 50 metabolites were measured in the lysates withexcellent signal/noise. Significant differences inmetabolomic profiles were observed between untreated(Control) and treated spheroids.

Use of Pu·MA System for in situ lysing of spheroids for metabolomic analysis was successfullydemonstrated. The ability to lyse organoids in situ in order to capture metabolomic profileswith minimal perturbation shows great promise for oncology research.3

Figure 5. Cluster analysis of metabolite signals (CN=Control; C59=C59 60 µM; ST=Staurosporine 1 µM). Signals are Blank subtracted

Figure 2. Schematic showing typical protocol steps available with the Pu·MA System. All steps are performed automatically inside an incubator.

Figure 3. Fluid transfer demonstration results. Error bars represent +/- 1 SD (n=8)

Figure 4. Schematic of fluid transfer in flowchip sample well. A spheroid in a well is shown on Left.

Fluid transfer repeatability wasdemonstrated by exchange of multiplebuffers back and forth between Pu·MASystem flowchip wells.

Pu·MA System Flowchips are designed with chambers and reservoirs arranged in a convenientmulti-well plate format (384-well spacings) and provide up to 32 tests per plate.Once spheroids and reagents are loaded into the flowchips, the plate is placed into the Pu·MASystem and reagent exchanges are done automatically through microfluidic channelsconnected to the protected sample chamber.

Multiple reagent exchanges are performed with assay protocols using built-in program. Assayprotocols are pre-loaded into the system and run using an intuitive touch-screen interface.The whole Pu·MA System can be placed in an incubator to run assays at 37°C and 5% CO2. Thesystem architecture and use of pneumatics to move fluids provides gas exchange to the samplechambers.

MICROFLUIDIC ASSAY BACKGROUND

The ImageXpressTM Micro Confocal Automated Imaging System (IXM-C) includes:

• Four colors + transmitted light

• Environmental control

• Automated data analysis

The system is controlled by MetaXpressTM Automated Imaging Acquisition and Analysis Software

0

5

10

15

20

25

Transfer 1 Transfer 2 Transfer 3 Transfer 4

Fin

al V

olu

me

(u

L)

Flowchip 1

Flowchip 2

Flowchip 3

60

70

80

90

100

110

120

130

C59 MytomycinC Etoposide

Lum

i Sig

nal

(x1

00

0)

HCT116 Spheroid 48 Hour Cell Viability

Control Treated

Row Min Row Max

In Out

Place Flowchips into Pu·MA System

Run Assay inside incubator

Load Spheroids into Flowchips

Fill Reagents into Flowchips