CTC isolation & analysis in an integrated microsystem · • HRP label & substrate optimized for...

© IMEC 2013

CTC isolation & analysis in an integrated microsystem

Chengxun Liuimec

[email protected]

© IMEC 2013

Challenges for CTC isolation

2

General• Low abundance: down to 1 CTC/mLamong billions of normal cells

• CTCs are probably heterogeneous -within one patient, and among patient subgroups

Cellular level• Size/morphological overlap between CTC and normal cells

Molecular level• No generally present carcinoma-specific markers are identified

• Normal blood cells may express similar markers (e.g. EpCAM)

WORLD CTC BERLIN 2013

© IMEC 2013 3

When microtechnology meets bio : size matters!

Small means:

� Single cell resolution

� Higher concentration per volume

� High reliability and efficiency

Also means:

• Micro/nano sensors & actuators

• Micro/nano fluidic handling

• Macro-to-micro interfacing

1 m1 mm1 µm

Microarray MRINeuron chip

1 nm

Cell VirusDNAMolecule

This is where we come into play.

imec


© IMEC 2013

The consortium

4

NanoelectronicsMolecular biology

Clinical validation Instrumentation

• Project type: FP7 integrated project

• Project number: 257743

• Project duration: 01.09.2010 – 31.08.2014

• Total cost: 10 million euro

• Project coordinator: Prof. Liesbet Lagae (imec)


© IMEC 2013

Technology overview

5

Probe immobilization

DNA hybridization

HRP labeling and detection

Multiplexed gene amplification (MLPA)

Active sieve for CTC identification & counting

Immunomagneticcell isolation

Multiplexed gene detection


© IMEC 2013

300 nm beads for immunomagnetic cell isolations

20 spiked MCF-7 cells in 5ml medium

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

9.0

10.0

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

110%

An

tib

od

y C

ou

pli

ng

[µ

g/m

g]

Re

cov

ery

Ra

te

Bead Name

result 1

result 2

result 3

Antibody µg/mg

� 300 nm: balanced cell recovery & cell-bead size contrast� Antibody: EpCAM (first test shown in graph) + stem markers

(in progress)WORLD CTC BERLIN 2013

© IMEC 2013

Isolating individual viable CTCs?

� How to identify CTCs?

� How to isolate CTCS from WBCs?

7WORLD CTC BERLIN 2013

© IMEC 2013

Technology overview

8


DNA hybridization







© IMEC 2013

An active micro sieve for CTC isolation

9

50 µm

3 µm

CTC

> 10 µm

WBC

5-10 µmMP (< 3 µm)

Active sieve

SiMOSInterconnect

EIS electrodeSiO2Trapping structure

Flow

� CTC identification by electrical impedance measurement (EIS)

� CTC positioning by dielectrophoresis (DEP)

� Electrical CTC lysis

Active sieve = a transistor chip with through-silicon pores

(US20120064567 A1)


© IMEC 2013

An active micro sieve for CTC isolation

10

Inlet

Outlet

Flow

12 mm

14

mm

� 10,000 single cell measurement pores

� On-chip microfluidic structure

� Front-end: TSMC 0.18 µm (Taiwan)

� Back-end & packaging: imec

Chip layout

Cell


© IMEC 2013

Cell positioning by dielectrophoresis

11

Cell manipulation by DEP

▸ Position the cell close enough to the EIS electrode (in-plane)

▸ Reduce the cell-electrode gap (out-of-plane)

▸ Cell lysis after identification

DI water

(1 .2e-4 S/m)

100

102

104

106

108

1010

-1

-0.5

0

0.5

1

Frequency (Hz)

Cla

usiu

s-M

osso

tti fa

cto

r (R

ea

l)

8.5% sucrose

(2.9e-4 S/m)

1 mM NaCl/PBS

(0.01 S/m)

DMEM (1.37 S/m) &

1× PBS (1.47 S/m)

1 M NaCl

(7.6 S/m)

• Wide pDEP freq. spectrum

• EIS & DEP by the same pair of electrodes

pDEP

nDEP


© IMEC 2013

Electrical cell positioning and lysis

12

Tumor cell capture & WBC repulsion by DEP

Calcein / esterase leaked out after cell lyzed by electric pulses


© IMEC 2013

Cell impedance measurement

13

� Different CTC membrane impedance for from normal blood cells

� Cell membrane impedance dominates the middle frequency range [1KHz, 1 MHz]

� Very tight cell-electrode contact is required to ensure high seal resistance (Rseal)

Csol

Rsol

Cline, IN Cline, OUT

CPEcont CPEcont

Rseal Rseal

RICM

RmembCmemb RmembCmemb

CPE Cell membrane impedance

Cytoplasm

(a)


© IMEC 2013

Cell membrane capacitance

14

� Cell membrane capacitance characterized by DEP measurements. Min. 50 individual cells measured for every cell line. Cell size measured and normalized.

� Tumor cells exhibit higher membrane capacitance than normal blood cells. � Cell capacitance can be a good measurement for TC, or CTC, identification from WBCs!

0.E+00

1.E+01

2.E+01

3.E+01

4.E+01

5.E+01

C_m

em

(fF

/µm

2)


© IMEC 2013

Technology overview

15


DNA hybridization







© IMEC 2013 16

Primer sequence Y(19bp)

LHO (24bp)

Pre-amplification productSpecific Reverse (RT) primer

Specific Forward primer

Primer sequence X(23bp)

RHO (20-30bp)

• RT primer is being used for reverse transcription

• RT primer and specific forward primer are used in following pre-amplification

• LHO and RHO are allowed to hybridize next to each other to denatured pre-amplification product

• LHO and RHO carry the universal primer sequences, which are used for amplification after ligation

• Only perfect hybridized LHOs and RHOs are connected by ligation.

Multiplex ligation-dependent probe amplification (MLPA)

Allowing both CE & microarray detection


© IMEC 2013

The full MIRACLE MLPA gene panels

� Breast cancer (30 genes)

17

AKT2ALDH1A1BMI1CD24CD44CDH1CDH2CEACAM3EGFRERBB2ESR1FN1HUEW1KRT19MKI67

mTORMUC1MYCPGRPIK3CAPLAURPROM1PTENPTPRCTACSTD1TERTTOP2ATWIST1VEGFAVIM

� Prostate cancer (16 genes)

ALDH1AMACRAREGFTEPCAMERBB2FOLH1 (PSMA)KLKB1KRT19LDHAPCA3PTPRCTMPRSS2-ERGbTMPRSS2-ETV1bHUWE1OAZ1

(Core gene panel for initial testing)


© IMEC 2013

MLPA sensitivity

18

NegCtrlPosCtrl

Single Cells

• MCF7 cells

• No RNA purificationMLPA product lengths: 115-170 bp


© IMEC 2013

MLPA sensitivity

19

Correlated results between electrophoresis & microarray

(1) (2)

(3) (4)

0

10000

20000

30000

40000

50000

60000

70000

[FU]

Single MCF7 cells

1 cell Oslo (1)

1 cell Oslo (2)

1 cell Oslo (3)

1 cell Oslo (4)

0

20000

40000

60000

80000

[FU]

MCF7 Negative & Positive control

NegCtrl H2O

NegCtrl H2O

PosCtrl RNA

PosCtrl RNA

MLPA product lengths: 115-170 bp


© IMEC 2013

MLPA for TCs spiked in WBC

20

� MLPA product lengths: 115-170bp� 20 TC for every sample � Still pretty high TC specific signal at 800:1.

TC WBC

800:1100:150:1

0

20000

40000

60000

80000

MNC:TC 800:1

MLPA product lengths: 115-170 bp


© IMEC 2013

MLPA robustness to RNA sample degradation

21

RIN = 6 �� Blood+TC

RIN = 1.3 �� Blood+TC

RIN = 6 �� MLPA

RIN = 1.3 �� MLPA

Intact RNA

Degraded RNA

MLPATotal RNA

NegCtrlSpecific MLPA signal even from seriously degraded RNA!


© IMEC 2013


23

Pneumatic interface

DNA PCB

Turning valve

Light barrier

Turning valve motor

Upper Peltier

Lower Peltier

DNA heater


© IMEC 2013

Principle of DNA detection

� Electrochemical DNA detection with enzymatic redox labels

• Well known working principle

• Good balance between sensitivity, specificity & reliability for PCR amplicons

• Electrical readout = multiplexing, compact, fast, cheap

26


DNA hybridization



© IMEC 2013

T= 0.5 sec

T= 208 sec

T= 833 sec

T= 3333 sec

Challenges for multiplex DNA detection

� The redox product can diffuse between electrodes

• HRP label & substrate optimized for reaction-limited kinetics rather than diffusion-limited

• Optimal electrode design against molecule diffusion

• Low-noise readout allowing rapid measurement before diffusion occurs

- MIRACLE: 0.5 seconds for all 64 electrodes

- Autolab®: several minutes

27

XWORLD CTC BERLIN 2013

© IMEC 2013

DNA sensor chips

28

Made of standard printed circuit board (PCB) technology, modified with IC-level gold finishing.

The 64 electrode sensor chip

The packaged sensor chip (before fluidic integration)

The integrated amplification-detection module


© IMEC 2013

Multi-gene detection for single cell MLPA amplicons

� Wide dynamic range for single cell amplicon sensing

29

0

5

10

15

20

25

30

35

40

45

CD24 ERBB2 KRT19 HUEW1 Controls

Cu

rre

nt

/ -n

A

AUTOLAB MIRACLE

0

10

20

30

40

50

60

70

80

90

100

CD24 ERBB2 KRT19 HUEW1 Controls

Cu

rre

nt

/ -n

A

HUWE1 HUWE1

0

100

200

300

400

0 1 2 3 4 5 6

Cu

rre

nt

/ n

A

Concentra on / nM

KRT19

Range of single cell amplicons

� Individual MLPA products prepared from single MCF7 cell

� Similar sensitivity & specificity as commercial instrument with much shorter signal integration time


© IMEC 2013

Conclusions

� Tumor cells exhibit different membrane capacitance from blood cells

30

Cell � MLPA is sensitive, specific and robust for TC gene analysis

� Sensitive multiplex gene detection demonstrated

Next: from functional integration to clinical tests


CTC isolation & analysis in an integrated microsystem · • HRP label & substrate optimized for...

Documents

Transcript of CTC isolation & analysis in an integrated microsystem · • HRP label & substrate optimized for...