A REAL-TIME CELL PROLIFERATION AND MOTILITY

MONITORING SYSTEM

Nicola Moscelli, Sander van den Driesche, Michael J. Vellekoop

Institute of Sensors and Actuator Systems, Vienna University of Technology, Gusshausstrasse 27-29/E366, Vienna, Austria

Wojciech Witarski

Institute of Virology, Slovak Academy of Sciences, Dubravska Cesta 9, Bratislava, Slovak Republic

Keywords: Optical detection, CCD sensor, Real-time monitoring, Epithelial cells.

Abstract: In this contribution we present a compact imaging system to monitor the proliferation and the motility of

cells in real-time. Our monitoring system is compatible with standard multi-well plates and operates in CO

temperature and humidity controlled cell-culture incubators. Adherent grown epithelial cells in a multi-well

plate well, positioned on top of a custom made holder, have been monitored in real-time with a fixed CCD

image sensor. As light source an LED is placed above the plate holder. A field of view of 3.3 × 2.5 mm

was achieved by using a 4.6 × 4.0 mm

image sensor and mini lens system. The image sensor has a

resolution of 640 × 480 pixels. Consequently, the obtained sensing resolution of the imaging system is about

5 μm. The cell monitoring system has first been validated by visualizing micro-beads of known dimensions.

Then, our system has been successfully tested, tracking the migration paths and proliferation of respectively

adherent grown MDCK (Madin-Darby canine kidney) and A549 (human lung carcinoma) epithelial cells.

1 INTRODUCTION

The study of cell motility is of importance to

understand the mechanisms related to many

physiological and pathological processes such as

tumour progression and immunological responses

(Suresh, 2007). A commonly used technique to track

and quantify cell motility is by conducting gold

colloidal phagokinetic assays (Niinaka, 2001):

because of their movement, the adherently grown

cells displace gold monodisperse nanoparticles

deposited on the substrate, tracing clearly visible

paths. This technique is attractive for time-lapse

measurements, where the preservation of optimal

cultivation conditions such as fixed temperature,

humidity, oxygen and CO

levels for biological

culture growth is required. In order not to influence

the cultivation conditions during cell proliferation

and motility experiments, lab microscopes are

normally equipped with additional expensive stages

to control temperature, humidity, and CO

levels

(Poujade et al., 2007). Our proposed real-time

imaging system is a compact and low-cost

alternative to such an optical setup: it can be

operated in standard lab incubators and is also

compatible with unmodified lab disposables. Gabriel

et al. have presented a real-time cell monitoring

solution for incubator comprising a contact imaging

device. These devices are based on an image sensor

array directly in contact with the sample to be

investigated (Gabriel et al., 2009). However, this

proposed imaging device needs ad hoc modified lab

disposables and an elaborate cleaning step when

serial experiments are desired. Moreover, unlike

conventional optical imaging systems, where the

resolution depends on the pixel number and the lens

magnification, contact imaging devices have a

resolution which depends on the pixel size together

with the distance between the object and the sensor

surface (Ji et al., 2007). This limits the suitability of

such devices, because the sensor needs to be

decoupled by the biological sample to preserve its

functionality.

In the following section, we provide a

description of our monitoring system setup. In

Section 3 we present and discuss the validation tests

and the cell monitoring experiments.

230

Moscelli N., van den Driesche S., Vellekoop M. and Witarski W..

A REAL-TIME CELL PROLIFERATION AND MOTILITY MONITORING SYSTEM.

DOI: 10.5220/0003167002300233

In Proceedings of the International Conference on Biomedical Electronics and Devices (BIODEVICES-2011), pages 230-233

ISBN: 978-989-8425-37-9

 2011 SCITEPRESS (Science and Technology Publications, Lda.)

Figure 1: (a) Schematic of the monitoring system setup; (b) Photograph of the system with a 24-well plate in the incubator.

To withstand the high humidity condition inside the incubator, the sensor and the printed circuit board have been

hermetically sealed with Parafilm and Duct tape.

2 SENSOR SYSTEM

Our monitoring system operates with standard

disposable multi-well plates. A multi-well plate is

positioned on top of a custom made Teflon holder. A

CCD sensor (Sony ICX-098BQ with Fire-i Digital

Remote Camera Board) is placed underneath the

multi-well plate, aligned and fixed in close

proximity to the well containing the biological

sample. The image sensor features a resolution of

640 × 480 pixels over a surface of 4.6 × 4.0 mm

Together with a mini lens, the sensor is capable to

focus over an area of 3.3 × 2.5 mm

, yielding to an

overall optical resolution of approximately 5 µm in

both dimensions. The CCD sensor is capable of

progressive scanning, and thus full frame non-

interlaced acquisition. Typical image acquisition

settings (frame rate, shutter, gain, gamma, white

balance, black level) are fully adjustable in order to

obtain the finest image conditions. Moreover, the

sensor is placed on a remote unit which allows

minimal area encumbrance under the multi-well

plate. A schematic representation and a photo of the

optical imaging system setup are depicted in Fig. 1.

In order to prevent condensation on the electronics

due to high humidity conditions in the incubator, we

sealed the sensor and the PCB with several layers of

polymeric films and tapes (Parafilm and Duct tape),

as depicted in Fig. 1b. A 10° angle focused

illumination is provided by a LED light source

(Luxeon Star/O) mounted orthogonally at 20 cm

distance over the multi-well plate. Vertical

illumination from the top of the biological sample is

attractive to minimize light scattering through the

radiation path (the light radiation has to cross the

cover lid of the multi-well, the cultivation medium,

the cells, and the bottom substrate of the well). The

cells are almost transparent and therefore hard to be

optically detected, if high light scattering occurs.

The supplied power for the LED was about 36 mW;

during our tests in incubator, no noticeable

overheating has been experienced.

3 RESULTS AND DISCUSSION

3.1 System Validation

In order to characterize our imaging system, we have

visualized particles of known shape and size. For

this scope, we have selected monodisperse non

transparent polystyrene spherical beads with a

diameter of 12 µm (Sigma Aldrich Fluka 885110).

In Fig. 2 the sedimented beads imaged by our

system are shown. The pixel distribution has been

determined by visualizing four individual beads and

two beads in close proximity. The comparable

intensity distributions of the four individual beads

show that each bead corresponds approximately to a

2 × 2 pixel matrix. This result is consistent with the

calculated optical system resolution of 5 × 5 µm

per

pixel. Also, the image shows the feasibility to

clearly discriminate between two beads in close

proximity, as long as the distance between them

measures more than one pixel.

(

)

(

)

A REAL-TIME CELL PROLIFERATION AND MOTILITY MONITORING SYSTEM

231

Figure 2: On top, individual polystyrene spherical beads of

12 µm diameter observed with our imaging system; below,

magnified pixel distributions of four different single beads

(1 to 4) and two beads in close proximity (5).

3.2 Cell Monitoring

We have tested our optical system with two different

epithelial cell lines: human carcinomic alveolar

basal (A549, ATCC CCL-185) and Madin-Darby

canine kidney cells (MDCK, ATCC CCL-34). The

cultivation media consisted of DMEM (Dulbecco's

Modified Eagle Medium) with 4.5 g/L glucose, 10%

fetal calf serum, 2 mM L-glutamine, and antibiotics.

The cells have been incubated at 37°C and with 5%

concentration (a NAPCO CO

1000 incubator

was used).

In a first experiment, we have tracked the

motility of adherently grown individual MDCK

cells. In order to promote cell movement, HGF

(Hepatocyte Growth Factor) in a concentration of 50

ng/ml has been added to the cell sample under

investigation. The frame-capture rate of the sensor

system has been set to one frame each 3 minutes,

which was sufficient enough to detect changes of

cell position within the parameters of this

experiment. After the cells have sedimented and

attached to the bottom of the well, their activity has

been recorded for 2.5 hours. In Fig. 3a-d, four

frames highlight the movement of two cells. The

pictures illustrate only a section of the detected area.

Also the migration paths covered by the two cells

during the 2.5 hours experiment time frame are

plotted (see Fig. 4). The trajectories of the cells have

been obtained with MATLAB software (MathWorks

Inc.) by extracting the coordinates of the cells per

each recorded frame. During the 2.5 hours

observation, both cells have covered a distance of

about 220 µm, which corresponds to a velocity of

about 1.5 µm/min. The detected migration paths

shown demonstrate that the achieved pixel

resolution of our system is adequate for epithelial

cell motility studies.

Figure 3: (a) ‒ (d) Frames showing sparse MDCK cells in

movement: cell positions as well as morphological

changes during motion are clearly detectable.

Figure 4: Migration trajectories of the two selected cells

during the 2.5 hours experiment; in that period both cells

covered a distance of about 220 µm, which corresponds to

a velocity of about 1.5 µm/min.

In a second experiment, we have observed the

proliferation of adherently grown A549 epithelial

cells in sparse concentration. Identical sensor

recording settings were kept for this second

measurement. In Fig. 5 a 2 hours time lapse of

dividing A549 cells is depicted in 8 frames. The

pictures clearly illustrate cell division in all its

BIODEVICES 2011 - International Conference on Biomedical Electronics and Devices

232

Figure 5: An A549 cell before, during and after the division process. (a) The cell is visible with its protrusions; (b) The cell

assumes a round shape; (c) The cell (brighter) partially detaches from the substrate and starts the division process; (d) ‒ (f)

The cell is in the cytokinetic phase; (g) End of the mitosis; (h) The two daughter cells get separated. The whole division

process (from (c) to (g)) takes place in about 50 minutes.

phases and prove that epithelial cell events related to

morphological cell changes can be well monitored in

time by our system.

In addition, from both experiments other

biologically relevant parameters such as the amount

of cells and their position in real-time can be

obtained. The shown frames have not been

processed with any software image manipulation

technique.

4 CONCLUSIONS

The successful results of the shown experiments

prove the suitability of our optical monitoring

system for real-time adherent grown cell

observation. Furthermore, by utilizing multiple

sensors the simultaneous recording and tracking of

different cell samples in the same multi-well plate is

feasible. This would allow real-time observation of

multiple cells exposed to diverse analytes (e.g.:

growth factors, motility inhibitors, and toxins).

Finally, the compatibility of our system with

unmodified lab disposables together with standard

incubators makes it an attractive and versatile

analysis tool for biomedical applications.

ACKNOWLEDGEMENTS

This project is part of the EU Marie Curie Research

Training Network “On-Chip Cell Handling and

Analysis” CellCheck (MRTN-CT-2006-035854).

The authors gratefully acknowledge Filippo Iuliano

of the Institute of Virology of the Slovak Academy

of Sciences in Bratislava for his valuable help in the

early stage of the device development.

REFERENCES

Suresh, S., 2007. Biomechanics and Biophysics of Cancer

Cells. Acta Materialia, Vol. 55, Elsevier.

Niinaka, Y., Haga, A., Raz, A., 2001. Quantification of

Cell Motility: Gold Colloidal Phagokinetic Track

Assay and Wound Healing Assay. Methods in

Molecular Medicine, Vol. 58, Springer.

Poujade, M., Grasland-Mongrain, E., Hertzog, A.,

Jouanneau, J., Chavrier, P., Ladoux, B., Buguin, A.,

Silberzan, P., 2007. Collective Migration of an

Epithelial Monolayer in Response to a Model Wound.

PNAS, Vol. 104, No. 41.

Gabriel, M., Picollet-D'hahan, N., Block, M., Haguet, V.,

Monitoring Adherent Cells by Contact Imaging, 2009.

In Proc. MicroTAS 2009.

Ji, H., Sander, D., Haas, A., Abshire, P. A., Contact

Imaging: Simulation and Experiment, 2007. IEEE

Trans Circ Sys, Vol. 54, No. 8.

A REAL-TIME CELL PROLIFERATION AND MOTILITY MONITORING SYSTEM

233