Printing of biochemically-patterned slides OK of biochemically-patterned slides1.pdfimmobilized on...
Transcript of Printing of biochemically-patterned slides OK of biochemically-patterned slides1.pdfimmobilized on...
INNOPSYS Parc Activestre – 31390 CARBONNE – France
Tel : +33 561 971 974 / Fax : +33 561 971 975 / www.innopsys.com
Printingofbiochemically-patternedslideswiththeInnoStamp40®fordeterministiccellimmobilization.
Jean‐christophe CAU1 1 Innopsys, Parc activestre, Carbonne – France – [email protected]
www.innopsys.com
www.biosoftlab.com
Introduction
Mastering the adhesion landscape of living
cells on a surface is emerging as a new tool for
investigating many fundamental mechanisms
of cell biology such as shape control,
differentiation, division, polarity or motility [1‐
12]. The common point between all these
studies is the production of micro‐patterns of
various shapes and dimensions along which
adherent cells are immobilized in a
deterministic way. The control of cell adhesion
through this technology allows the biologists to
design specific scenari and record how various
living cell types (neural, epithelial, tumoral and
stem cells) respond to them. Because this
technology allows reproducing identical
precise patterns along well arranged periodical
arrays, these experimental observations can be
made systematically over large population of
cells thus reaching a high level of
representativeness. The micro‐contact printing
of proteins of the Extra Cellular Matrix (ECM)
on conventional glass slides or coverslips
turned out to be an efficient method for
fabricating these micro‐patterned cell culture
substrates. However, when this technological
process is made manually, it fails to generate
reproducible patterns over large surfaces thus
limiting the actual potential of this new
approach. In order to bring this process to a
standard level in cell biology, we propose to
automate the printing process of ECM proteins
on cell culture substrates. As a matter of fact, a
fine control of the stamping parameters is a
sine qua none condition to be able to print in
one step reproducible patterns of different
shape, size and pitch.
In this note, we present a dedicated study in
which we screen different pattern shapes and
pattern sizes for the deterministic
immobilization of cancer cells. With the
perspective in mind of remote production of
these patterned cell‐culture supports, we have
also investigated the conservation potential
over time of the printed slides before cell
seeding.
We have used a patterned stamp designed
with four different classes of features (disc,
line, square, and triangle) of different sizes
(from 10 to 50µm) and pitches (10 to 100µm).
The samples were seeded with Prostate Cancer
3 (PC3) cells. Then, we have quantified the
spreading of the immobilized cells on these
different patterns and the selectivity of the
immobilization on the adhesive patterns
compared to the antifouling background.
1‐Cellular growth on patternedsamples fabricated with a fullyautomatedmicrocontact printer: theInnoStamp40.
The workflow of this study is based on three
steps (figure 1). The first one is the fabrication
of ECM (Extra Cellular Matrix) protein micro
INNOPSYS Parc Activestre – 31390 CARBONNE – France
Tel : +33 561 971 974 / Fax : +33 561 971 975 / www.innopsys.com
patterns on glass slides. For that, we used the
automatic microcontact printer named
InnoStamp40. It is based on magnetic‐field
assisted microcontact printing [13]. This
technology allows for:
An automation of the full process of
microcontact printing,
A fine control of the pressure applied
during the printing step,
An alignment of the patterns on the
sample,
A high reproducibility of the process.
Three elements are handled in the
InnoStamp40® (www.innopsys.com):
A PDMS stamp with the micropatterns,
A sample (in this study: a glass slide),
The ink (In this study, it is a 100µg/mL
fibronectin solution with PBS 1X).
This equipment automatically generates
micropatterns of fibronectin on the glass slide
according to the layout of figure 2. Four
different shapes of patterns with five different
sizes (from 10µm to 50µm) and four different
gaps (from 10µm to 100µm) are printed and
arranged into periodic arrays. Each shape array
is repeated at least one time on a stamp.
The second step is the backfilling of the
fibronectin micro patterns with an antifouling
layer (figure 1). For that, we incubated the
printed sample in a 100µg/mL solution of Pll‐g‐
PEG during 1h. After drying, the space between
fibronectin patterns is filled with Pll‐g‐PEG.
The third step is the PC3‐GFP cells seeding
(figure 1). The PC3‐GFP cells are seeded during
3h at 37°C with a concentration of 42.000
cells/cm². Then, they are washed with PBS and
dehydrated with 3 baths of 50%, 75% and 100%
of ethanol.
Figure 1: the workflow of Cellular growth on
patterned samples
The images are acquired with a standard
fluorescence microscope.
INNOPSYS Parc Activestre – 31390 CARBONNE – France
Tel : +33 561 971 974 / Fax : +33 561 971 975 / www.innopsys.com
Figure 2: layout of the micropatterns. There are 4
different cells with different shapes: disc, line,
square, and triangle. Their size varies between
10µm to 50µm and the gap between the patterns
varies between 10 to 100µm.
2‐Differentshapesofthebiochemicalpatterns.
The cells are selectively captured on the micro‐
patterned slides. We can observe on figure 3
that the PC3‐GFP cells in adhesion with the
surface of the sample decorate the printed
fibronectin patterns. The PC3‐GFP cells which
are selectively immobilized on the molecular
patterns spread on and fill them while the very
few cells immobilized outside the patterns are
circular and do not spread. We quantify the
selectivity of the immobilization as the number
of cells counted inside the printed patterns
divided by the number of cells counted in the
PLL‐g‐PEG background. Table 1 displays the
selectivity observed for the four different
shapes. We find a good selectivity for all the
shapes (higher than 75%). With triangles, the
selectivity is lower than the line patterns. In a
more general manner, we observe that the
higher the spatial confinement produced by
the patterns is, the lower the selectivity is.
Selectivity (inside/outside
%)
Triangle 75 Square 85 Round 91 Line 99
Figure 3: fluorescence images of PC3‐GFP cells
immobilized on ECM micro patterns of various
shapes (the patterns are depicted by dashed lines).
3‐Quantification of the area of a celltrappedontoabiochemicalpattern.
After shape investigation, we studied the
minimal surface needed to trap a single PC3‐
GFP cell and the area occupied by captured
cells. To achieve that, we selected a region near
the interface between two areas (figure 4). In
order to distinguish and well identify the
different areas (one with 50µm and the other
one with 40µm triangle patterns), we have also
printed a dashed‐line pattern composed of
20µm*10µm rectangles (violet arrow).
We first estimated the mean area of the cells
trapped into triangles by dividing the triangle
INNOPSYS Parc Activestre – 31390 CARBONNE – France
Tel : +33 561 971 974 / Fax : +33 561 971 975 / www.innopsys.com
area by the number of trapped cell. We found
a mean area of 420µm².
Figure 4: A selected region at the interface between
two areas. The upper part is made of 50µm triangles
and the lower part is made of 40µm triangles. At the
interface, rectangles of 20µm*10µm are printed
(yellow dashed line).
We can observe in figure 4 that single cells
decorate the rectangular patterns of the
printed dashed line. However, they do not
follow exactly the contours of the rectangle but
instead they overfill the patterns. The area of
these patterns is 200µm². Compared to the cell
trapped into triangles, we assume that a
200µm² rectangular area is enough to trap
single cells but not enough to make the
contours of the cell match the contours of the
printed features. This is why the PC3‐GFP cells
on this kind of patterns exhibit circular shapes
and not rectangular ones. The interplay
between PC3‐GFP cells and 200µm² rectangles
of fibronectin enables to achieve a
deterministic immobilization of single cells on
the patterns but do not conform the cell shape
to the contours of the pattern. The circular
shape of these cells could also be
demonstrated by the interplay of the
protruding parts of the cells (outside of the
patterns) with the antifouling layer.
Single cells trapped on the 30µm triangles
(figure 5) seem to be shaped by the patterns.
The area of these triangles is 450µm², which is
close to the 420µm² area found previously.
Several cells can sometimes be observed on
these patterns because the pattern area is a bit
larger than 420µm² and also due to cell size
heterogeneity. Put together, these results
indicate that, for patterns of area close to
200µm2, single cells can be deterministically
immobilized but their shape is not constricted
by the shape of the patterns, while for patterns
of area close to 400 µm2, single cells are
selectively immobilized and shaped by the
printed patterns.
Figure 5: fluorescence image of PC3‐GFP cells
immobilized on 30µm printed triangles (scale bar
40µm).
4‐Aggregates of cells near themolecularpatterns.
We have also investigated the interplay
between cells according to their positions into
the patterns.
We observe on figure 6 the cell spreading on
lines of different widths: 10, 20 and 40µm. On
10µm lines (figure 6.A), the PC3‐GFP cells
spread on the entire length of the lines and
create a pattern similar to a cell “rosary”. If
there are more cells across, PC3‐GFP cells
reduce their spreading in order to “cohabit”
within the line. Cells found outside the patterns
exhibit circular shapes and they hang on to the
spread ones. For 20µm and 40µm‐wide lines
(figure 6.B and C), the cells interplay is similar.
However, due to a lower spatial confinement,
PC3‐GFP cells are less stressed and, thus, can
shape the lines with two or three cells across.
On figure 7, PC3‐GFP cells shape discs of 40µm
diameter but some additional cells hang on to
INNOPSYS Parc Activestre – 31390 CARBONNE – France
Tel : +33 561 971 974 / Fax : +33 561 971 975 / www.innopsys.com
them. We observe the same shape difference
between cells inside (spread) and outside
(circular) the patterns. This effect being visible
for different shape and size patterns, we
assume that it is due to some cell aggregates
present in the seeding solution.
Figure 6: fluorescence images of PC3‐GFP cells
trapped immobilized on 10µm (A), 20µm (B) and
40µm (C) wide printed lines. Scale bar: 40µm.
Figure 7: fluorescence images of PC3‐GFP cells
immobilized on 40µm circular patterns. Scale bar:
40µm.
5‐Study of the patterned slides shelftime
To conclude this study, we have investigated if
these printed slides could be used days after
the manufacturing process. Indeed, in order to
evaluate this technology, it could be interesting
to send to researchers some printed slides,
raising the question: “what is the shelf time of
the printed slide?”. To answer this question,
we compared cell culture on a fresh printed
slide and cell culture operated 12 days after
printing in different conditions of conservation.
The temperature in the conservation phase
was changed (room temperature, 4°C or ‐20°C)
and two media of conservation were tested
(ambient air or PBS1X). We could not notice
any difference across these conditions. This is
why we present the results after 12‐day
conservation at room temperature, ambient
air. We are now using these parameters to
send some samples to interested customers.
C B
A
INNOPSYS Parc Activestre – 31390 CARBONNE – France
Tel : +33 561 971 974 / Fax : +33 561 971 975 / www.innopsys.com
Figure 8: comparison between a fresh printed slide
(A) and a 12‐day stored patterned glass slide (B).
On figure 8, we can observe no significant
differences between freshly printed slides and
conserved ones. The selectivity is better after
12 days but the number of cells per pattern is
higher. This trend could be explained by a small
variation in the cell concentration of the
seeding solution.
Conclusion
Patterned slides with ECM proteins can be
routinely and reproducibly produced using the
InnoStamp40 microcontact printer. Adherent
cells can be immobilized deterministically on
these patterns. Single‐cell arrays can be
generated. The shape of these individual cells
can be dictated by the shape of the patterns, if
the area of each pattern is smaller than the
mean area of a cell in adhesion on a non‐
spatially confined zone. To complete the study,
we have characterized the conservation of the
patterned slides before cell seeding and have
proven that functional micro‐patterned slides
are compatible with standard shipping.
Acknowledgment
This work was done within the framework of a
joint laboratory named BIOSOFT
(www.biosoftlab.com) between LAAS‐CNRS
and INNOPSYS. Author wants to acknowledge
Julie FONCY, Charline BLATCHE, Emmanuelle
TREVISIOL and Christophe VIEU.
References
[1] McBeath R et al. Cell shape, cytoskeletal
tension, and RhoA regulate stem cell lineage
commitment. Dev Cell. 2004 Apr;6(4):483‐95.
[2] Tseng Q et al. Spatial organization of the
extracellular matrix regulates cell–cell junction
positioning. Proc Natl Acad Sci U S A. 2012 Jan
31;109(5):1506‐11
[3] Gao L et al. Stem cell shape regulates a
chondrogenic versus myogenic fate through
Rac1 and N‐cadherin. Stem Cells. 2010 Mar
31;28(3):564‐72.
[4] Mrksich M et al. Using microcontact printing
to pattern the attachment of mammalian cells
to self‐assembled monolayers of
alkanethiolates on transparent films of gold
and silver. Exp Cell Res. 1997 Sep
15;235(2):305‐13.
[5] Huang S et al. Shape‐dependent control of
cell growth, differentiation, and apoptosis:
switching between attractors in cell regulatory
networks. Exp Cell Res. 2000 Nov 25;261(1):91‐
103.
A
B
INNOPSYS Parc Activestre – 31390 CARBONNE – France
Tel : +33 561 971 974 / Fax : +33 561 971 975 / www.innopsys.com
[6] Théry M et al. The extracellular matrix
guides the orientation of the cell division axis.
Nat Cell Biol. 2005 Oct;7(10):947‐53.
[7] Théry M. Micropatterning as a tool to
decipher cell morphogenesis and functions. J
Cell Sci. 2010 Dec 15;123(Pt 24):4201‐13.
[8] James J et al. Subcellular curvature at the
perimeter of micropatterned cells influences
lamellipodial distribution and cell polarity. Cell
Motil Cytoskeleton. 2008 Nov;65(11):841‐52.
[9] Dupin I et al. Classical cadherins control
nucleus and centrosome position and cell
polarity. J Cell Biol 185(5):779‐786
[10] Desai AR et al. Cell polarity triggered by
cell‐cell adhesion via E‐cadherin. J Cell Sci. 2009
Apr 1;122(Pt 7):905‐11.
[11] Théry M et al. Cell shape and cell division.
Curr Opin Cell Biol. 2006 Dec;18(6):648‐57.
[12] Parker KK et al. Directional control of
lamellipodia extension by constraining cell
shape and orienting cell tractional forces.
FASEB J. 2002 Aug;16(10):1195‐204.
[13] Brock A et al. Geometric Determinants of
Directional Cell Motility Revealed Using
Microcontact Printing. Langmuir, 2003, 19 (5),
pp 1611–1617
[14] CAU JC et al. Magnetic field assisted
microcontact printing: A new concept of fully
automated and calibrated process.
Microelectronic Engineering, volume 110,
October 2013, Pages 207–214