WO2014118311A1

WO2014118311A1 - Device and method for controlling the spatial distribution of cells and their progeny

Info

Publication number: WO2014118311A1
Application number: PCT/EP2014/051873
Authority: WO
Inventors: Yoran MARGARON; Sébastien DEGOT; Violaine CHAPUIS-PERROT
Original assignee: Cytoo
Priority date: 2013-02-01
Filing date: 2014-01-31
Publication date: 2014-08-07

Abstract

The invention relates to a method for controlling the spatial distribution of adherent cells and their progeny, comprising the steps of: - providing a device comprising a surface with a plurality of cell-adhesive patterns on said surface, said cell-adhesive patterns being disposed so as to form at least one network, wherein: (i) the distance between adjacent cell-adhesive patterns of said network is chosen so as to allow a cell seeded on a pattern to move to at least one adjacent pattern, and (ii) the shape of the cell-adhesive patterns and distance between adjacent patterns of said network is selected so as to promote conformation of one such cell to a pattern and migration of daughter cells to adjacent patterns of said network; - seeding at least one cell onto at least one pattern; - culturing said cell such that it divides and progeny moves to other patterns of said network, at least part of said cultured cells being conformed to one respective cell-adhesive pattern of the network over time during several cell generations.

Description

DEVICE AND METHOD FOR CONTROLLING THE SPATIAL DISTRIBUTION OF CELLS AND THEIR PROGENY

FIELD OF THE INVENTION

The invention relates to a device and a method for controlling the spatial distribution of adherent cells and their progeny.

BACKGROUND OF THE INVENTION

Most mammalian cells derived from solid tissues are adherent.

They can be cultured on unlimited adhesive surfaces, such as tissue culture plastic or microcarriers, which may be coated with extracellular matrix components to increase adhesion properties and provide signaling pathways supervising cell growth and differentiation.

When grown in an appropriate cell medium in a cell incubator with controlled temperature and gas mixture, cells can divide and daughter cells are randomly disposed in the neighboring area.

When needed, cells can be detached by known techniques such as a mixture of trypsin-EDTA and a small amount of these cells can be recycled to seed a new culture.

However, when cells are grown on two-dimensional (2D) surfaces, they adopt random shapes, present different sizes and have random movements.

Such variability in the cell population distribution induces a high heterogeneity of their behavior.

Moreover, when cells are attached together to form colonies, their location in the colony can induce a broad variability of individual signaling pathways regulation.

For example, cells at the periphery of the colony are fully proliferating compared to cells located at the center that are physically constrained by surrounding cells and usually poorly proliferating.

In addition, analysis of daughter cells shapes is difficult because they are not often well individualized after mitosis.

Another type of adherent culture is an organotypic culture, which involves growing cells in a three-dimensional (3D) environment as opposed to the classical two-dimensional culture dishes.

This 3D culture system is biochemically and physiologically more relevant to in vivo tissue, but is technically challenging to maintain.

In order to promote adhesion of cells in a specific and determined position, document EP 1 664 266 proposes a culture device comprising an adhesive surface having a determined shape onto which only one cell can adhere. Once cells are attached to such surfaces, they use the adhesive clues provided to adopt the overall shape of the surface while cell's organelles are disposed in a controlled organization.

Cells may then be normalized from their spatial configuration that adopts the geometry of the adhesive surface.

However, pattern shape and size influence cell homeostasis.

Since adapted micropattern parameters vary for each cell type and culture condition, a specific device has to be developed for each specific experiment to be carried out.

Another drawback of this technique is that detailed analysis of organized cell compartments must be performed before the first cell division (i.e. within about 18 to 30 hours from seeding, depending on the cell line). Indeed, once two cells (daughter cells from the initial mother cell) are on the same pattern, the stability of each cell shape cannot be controlled, except for a few specific motifs.

A culture device may comprise several patterns that are separated by a cytophobic surface, the distance between adjacent patterns being large enough to prevent spreading of a cell over two patterns or more.

Document US 5,776,748 discloses a device for adhering cells on a plurality of rectangular patterns arranged according to a grid array. The cytophobic regions separating the patterns are sufficiently wide so as to confine cell spreading to the respective pattern and prevent any contact between cells adhering on adjacent patterns. As an artifact, it has been observed a few cells bridging between adjacent patterns. For said few cells, because cell spreading is not restricted to a confined area, the cell shape is not controlled. Besides, for the cells that remain confined to one respective pattern, the control of cell shape is lost as soon as the first cell division occurs, since two or more progeny cells are constrained to adhere to the same pattern.

Document EP 2 180 042 discloses a device to constrain multicellular arrangements in stable and reproducible spatial configuration.

Such a device comprises a plane surface onto which a set of two or more adhesive patterns are formed, wherein adjacent patterns are separated from each other by an essentially non-adhesive surface, the distance between said patterns being large enough to prevent a cell on a pattern to reach another pattern. On the other hand, the area covered by the set, i.e. the patterns and the essentially non-adhesive surface, is chosen so as to adhere a determined number of at least two cells.

In such a device, the patterns belonging to the same set may be connected by an adhesive zone.

The device may comprise several sets of patterns, each set being spatially isolated from the others by a cytophobic surrounding surface, such that cells seeded on a set cannot reach another set. However, such pattern designs are mainly organized for interacting cell lines such as epithelial cells or cells building tight cell-cell junctions.

Besides, the set geometries are designed to accommodate a defined number of cells, preventing the spatial stabilization to occur before the exact number of cells is reached.

And therefore giving a limited timeframe for analysis before another division occurs and number of cells on the set of patterns exceeds this ideal number.

Indeed, as shown on Figures 1A and 1 B, areas available to receive daughter cells are restricted.

Only cells dividing at the periphery can access to free adhesive area.

To the contrary, cells positioned at the center are surrounded by cells. As they cannot access to free adhesive area, such cells do not receive structural clues and do not stabilize properly.

Before division (Figure 1A), cells C are stably spread (in light grey) on "T" motifs (referred to as T).

Mitosis (Figure 1 B) induces two types of cell population:

- the ones located at the border of the structure which can divide and recover a stable shape (cells Cb), and

- the ones enclosed at the center where a control of cell location is not provided by the structure (cells Cc);

A goal of the invention is to provide device and method for controlling the spatial distribution of a cell population from their initial seeding to several generations of daughter cells.

SUMMARY OF THE INVENTION

According to a first aspect, the invention provides a method for controlling the spatial distribution of adherent cells and their progeny over time, comprising the steps of:

- providing a device comprising a surface with a plurality of cell-adhesive patterns on said surface, said cell-adhesive patterns being disposed so as to form at least one network consisting of a plurality of recurring elementary arrangements, each elementary arrangement comprising at least one cell-adhesive pattern, wherein:

(i) the distance between adjacent cell-adhesive patterns of said network is chosen so as to allow a cell seeded on a pattern to move to at least one adjacent pattern, and

(ii) the shape of the cell-adhesive patterns and distance between adjacent patterns of said network is selected so as to promote conformation of one such cell to a pattern and migration of daughter cells to adjacent patterns of said network;

- seeding at least one cell onto at least one pattern of said network; - culturing said at least one cell such that it divides and progeny are able to move to other patterns of said network, at least part of said cultured cells being conformed to one respective cell-adhesive pattern of the network over time during several cell generations.

In the present text, a pattern means a cell-adhesive surface whose dimension is sufficient to anchor a single and having a determined shape enabling controlling the spatial organization of said cell.

Such a pattern may be continuous, i.e. formed of a single adhesive surface, or discontinuous, i.e. formed of a plurality of discrete adhesive surfaces that together define the shape and size of the pattern.

In the present text, a network means a plurality of patterns that are disposed according to recurring elementary arrangements and with a distance between patterns that allows spreading or migration of a cell from a pattern to an adjacent pattern.

An elementary arrangement may consist of a single pattern or of at least two patterns. In the latter case, the patterns may be identical or have different shapes.

For example, said recurring elementary arrangements may be disposed in parallel rows and/or columns, in staggered rows, in fractal figures, or in any regular geometric disposition, etc.

In the present text, the distance between adjacent patterns is defined as the minimal distance between adjacent patterns. If adjacent patterns are in contact, said distance is null.

In this regard, it is to be noted that arrangement of several patterns (such as in documents EP 1 664 266 or US 5,776,748) or several sets of patterns (such as in document EP 2 180 042) on a same device, wherein said patterns or set of patterns are separated by a cytophobic surface that prevents spreading of cells to another pattern or set of patterns respectively, do not form a network as defined above. Indeed, in these devices, the patterns or sets of patterns are independent from each other in terms of cell control.

According to an advantageous embodiment, the disposition of the cell-adhesive patterns and/or of the elementary arrangements within a network is selected so as to control the flow of cell division on said network.

The size and/or shape of the cell-adhesive patterns is preferably selected so as to allow stabilization of the shape of a single cell seeded on said pattern.

Adjacent cell-adhesive patterns of said network may be at least partially separated by a cytophobic surface.

For example, the shape of said at least one cell-adhesive pattern may be approximately a disc, a crossbow, an L, a Y, a rectangle, or any combination thereof.

Each elementary arrangement may consist of a single pattern. Alternatively, each elementary arrangement may consist of at least two identical patterns.

Otherwise, each elementary arrangement may consist of at least two patterns having different shapes and/or sizes and/or orientations, wherein at least a one pattern, called normalizing pattern, has a shape chosen so as to promote conformation of one cell to said pattern and at least one other pattern, called intermediate pattern, has a shape chosen so as to promote cell displacement from one normalizing pattern to another normalizing pattern.

Said recurring elementary arrangements may be disposed in parallel rows and/or columns, in staggered rows, according to a regular geometric array or according to a fractal figure.

Advantageously, the rate of cells conformed to one respective cell-adhesive pattern of the network is greater than or equal to 20%, in particular comprised between 20% and 35%.

According to an embodiment, the distance between adjacent patterns is comprised between 4 μηη and 30 μηη.

Said method advantageously further comprises acquiring images of the cells during said division and migration and determining cell shape and/or motility parameters from said acquired images.

According to an embodiment, the method comprises a preliminary step of manufacturing said device by the following steps:

- preparing a master template with at least one network of cell-adhesive patterns,

- preparing a stamp from said master template,

- inking said stamp with molecules that promote cell attachment,

- contacting the inked stamp with the surface of the device.

According to another aspect, the invention relates to a cell culture system comprising a device having a surface with a plurality of cell-adhesive patterns on said surface and at least one cell normalized on at least one of said cell-adhesive pattern, said cell-adhesive patterns being disposed so as to form at least one network consisting of a plurality of recurring elementary arrangements, each elementary arrangement comprising at least one cell-adhesive pattern, wherein:

- the distance between adjacent cell-adhesive patterns of said network is chosen so as to allow a cell seeded on a pattern to move to at least one adjacent pattern, and

- the shape of the cell-adhesive patterns and distance between adjacent patterns of said network is selected so as to promote conformation of one such cell to a pattern and migration of daughter cells to adjacent patterns of said network. Preferably, said cell culture system comprises at least one cell and progeny, wherein said cell and progeny are normalized on adjacent cell-adhesive patterns of the network.

Another aspect of the invention is a device for carrying out the method described above.

Said device comprises a surface and a plurality of cell-adhesive patterns on said surface, wherein said cell-adhesive patterns are disposed so as to form at least one network, each network consisting of a plurality of recurring elementary arrangements, each elementary arrangement comprising at least one cell-adhesive pattern, wherein the distance between adjacent cell-adhesive patterns of said network is chosen so as to allow a cell seeded on a pattern to move to at least one adjacent pattern.

According to an advantageous embodiment, said device comprises at least two networks that differ by at least one of the size, shape of the cell-adhesive patterns and the distance between adjacent cell-adhesive patterns.

Another object of the invention is a cell culture kit comprising such a device and instructions for controlling cells and their progeny according to the present invention.

This invention can be useful in target validation, drug screening, in vitro toxicology, diagnostics, biochips, and cellular or gene therapy. BRIEF DESCRIPTION OF THE DRAWINGS

Further features and advantages of the invention will be apparent from the following detailed description, referring to the appended drawings wherein:

- Figures 1A and 1 B schematically show effect of cell invasion within a set of adhesive patterns according to document EP 2 180 042:

- Figures 2A and 2B show respectively a network according to an embodiment of the invention and the corresponding elementary arrangement of cell-adhesive patterns;

- Figure 3 shows a network according to another embodiment of the invention;

- Figure 4 shows a network according to another embodiment of the invention; - Figures 5A to 5D show networks consisting of similar elementary arrangements of patterns, but with increasing distances between patterns;

- Figures 6A to 6C show networks based on identical patterns, but wherein the recurring elementary arrangements of said patterns (each elementary arrangement consisting in this case of a single pattern) are disposed respectively in rows and columns, in staggered rows and in fractal figures;

- Figures 7A to 7C show different examples of networks wherein the elementary arrangements consist of patterns having at least two different shapes;

- Figure 8 illustrates a top view of a 96-well plate according to the invention; Figures 9A to 9C relate to the characterization of the most suitable pattern(s) for normalizing HeLa cell shape and polarity: Figure 9A is an image after 72 hrs incubation on a honeycomb network consisting of Y patterns having a surface of 1600 μ² with 12 μηη distance between adjacent patterns; Figure 9B shows the amount of cell shape normalization on one pattern and Figure 9C is a cartography of the normalization of HeLa cells on a plate as shown on Figure 8; Figure 10 shows the mitochondria status of different cell types after 72 hrs incubation on a network according to the invention;

Figure 1 1 shows the quantification of cells with a spindle orientation axis normalized within a cell population cultured on a network according to the invention;

Figures 12A to 12G show respectively a "stairs" network, one elementary arrangement that forms this network and images of the cells cultured on said network acquired at different times after seeding;

Figures 13A to 13G show respectively a "honeycomb" network, one elementary arrangement that forms this network and images of the HeLa Kyoto cells cultured on said network acquired at different times after seeding;

Figures 14A to 14F show respectively a "honeycomb" network, one elementary arrangement that forms this network and images of the A549 cells cultured on said network acquired at different times after seeding;

Figures 15A to 15D show results of cell tracking using H2B-mCherry HeLa Kyoto cell fluorescence for different networks.

DETAILED DESCRIPTION OF EMBODIMENTS OF THE INVENTION

The method uses a network of cell-adhesive patterns that, allows cell shape normalization while preserving the cell physiology, said normalization being also applicable to progeny.

Network

Figure 2A shows an example of a network 100 formed of a plurality of Y-shaped cell- adhesive patterns 1 .

Said network 100 is in fact formed of a plurality of recurring elementary arrangements 10 of said cell-adhesive patterns 1 .

Figure 2B shows the elementary arrangement 10 of Y-shaped cell-adhesive patterns 1 that form said network.

In this configuration, the elementary arrangement 10 comprises six identical cell- adhesive patterns 1 that are disposed with defined orientations with respect to one another, in order to form a snowflake shape. The distance between adjacent cell-adhesive patterns 1 is chosen so as to allow a cell seeded on a pattern to move to at least one adjacent pattern

In this example, which is of course not limitative, the distances a, b and c are of 14.5 μηι, 36.2 μηη and 21 .5 μηι, respectively.

The surface 2 that separates adjacent patterns 1 is preferably a cytophobic surface.

When cells are seeded on such a network, they are able to move to one or several patterns and adopt a stable shape before the next cell division.

In some networks, cells that adhered to and adopted the shape of a pattern can move to adjacent pattern(s) and re-adopt their shape before the next cell division.

When cell division occurs, daughter cells can move to adjacent patterns and adopt an equivalent stable morphology.

Hence, according to the size of the network, this cell shape stability can be maintained over several cell divisions, thus allowing both short and long term experiment.

Cell spreading properties are specific to each cell line and depend on many parameters, including cell size and its ability to contract on matrix substratum.

Due to the variety of cell lines and assays possibly combined with the current invention, adhesives patterns with a large range of sizes and geometries may be considered.

Typically, adhesive patterns named "Disc", "Y", "L", and "Crossbow" offered by CYTOO may be used, which induce cells area of 700 μηι², 1 ,100 μηι², 1 ,600 μηι², 2,200 μηι² and 3,000 μηι².

Figures 3 and 4 shows networks made of different cell-adhesive patterns.

In the network of Figure 3, the cell-adhesive patterns are of "crossbow" type and arranged in staggered rows.

In the network of Figure.4, the cell-adhesive patterns are of "L" type and arranged in staggered rows.

Because different cell types have distinct migration properties, minimal distances between adjacent cell-adhesive patterns may be variable to adapt to cell capabilities. In particular, the minimal distance may depend on the contractility of the cells.

An array of minimal distances were defined as 0 μηη, 2 μηη, 4 μηη, 6 μηη, 7.5 μηη, 8 μηι, 12 μηι, 14.5 μηη 15 μηι, 20 μηι, 30 μηι, 40 μηη and 50 μηι.

Advantageously, the minimal distance between adjacent patterns may be comprised between 4 μηη (for poorly contractile cells, such as MDA MB231 ) and 30 μηη (for contractile cells, such as A549 or fibroblasts NIH3T3).

Usually 80 μηη is sufficient to abrogate cell movement between two adjacent patterns or networks.

Figures 5A-5D displays a variety of minimal distances between patterns within a network type. In these examples, the cell-adhesive patterns are of "Y" type and are disposed as snowflakes networks as explained above.

The only difference between these networks is the distance between adjacent patterns (corresponding to distance a shown on Figure 2).

In Figure 5A, the distance between the patterns is of 0 μηη, meaning that the patterns are in contact with one another; in Figures 5B-5D, the distance is of 4 μηη, 8 μηη and 12 μηι, respectively.

The respective orientation of the cell-adhesive patterns may also be defined in order to control the flow of cell division on the network.

Because of the infinite potentials of networks arrangement, the angular range between adjacent patterns is from 0° to 360°.

As seen in figure 6A-6C, "crossbows" patterns can be disposed as networks so as to form parallel rows and columns, staggered rows or and fractal figure such as Cayley's trees, respectively, depending on the angle between adjacent patterns.

In addition, an elementary arrangement may be formed of several identical patterns.

Alternatively, an elementary arrangement may be formed of several different patterns with variable sizes and shapes. Among said patterns, some of them (so-called normalizing patterns) may have a shape and surface sufficient for allowing conformation and stabilization of a cell thereon, whereas other patterns (so-called intermediate patterns) may have a smaller surface and are only intended for promoting spreading of a cell from one normalizing pattern to another normalizing pattern. The shape and orientation of the intermediate patterns with respect to the normalizing patterns are chosen so as to help cell displacement.

Figure 7A shows an example of a network where normalizing "crossbow" patterns are combined with intermediate rectangle patterns arranged between the "crossbow" patterns to help cell displacement.

Figure 7B shows an example of an elementary arrangement where a "disc" pattern is combined with "Y" patterns arranged around the disc. In such case, the disc may be used as a cell reservoir while the Y patterns may be normalizing patterns.

Figure 7C shows an example of a network combining rectangular patterns of different shapes arranged alternately so as to form dashed lines. The largest rectangular patterns may be normalizing patterns wherein the smallest ones may be intermediate patterns.

Device

The above-described network may be formed on a surface of any device usually intended for cell culture.

Indeed, the device may be a glass, plastic or silicon oxide substrate, a cell culture plate or dish (this list not being intended to be limitative). A device comprising such a network can be made by known technologies.

In particular, a method for manufacturing a device comprising at least one network may comprise the following steps:

- preparing a master template with at least one network of cell-adhesive patterns, - preparing a stamp from said master template,

- inking said stamp with molecules that promote cell attachment,

- contacting the inked stamp with the surface of the device,

- making cytophobic the non-printed surface of the device.

Preferably, the master template is prepared from a silicon wafer coated with a photoresist layer illuminated with UV through a mask on which the adhesive patterns have been designed.

The stamp is preferably made of PDMS (poly(dimethylsiloxane)) or of another siloxane-based polymer.

Preferably, the non-printed surface of the device is made cytophobic by incubation with an inert material such as polyethyleneglycol.

For more details about such a method, one can refer to EP 1 664 266.

In an advantageous embodiment, the device may comprise at least two networks, wherein the distance between adjacent networks is chosen so as to prevent any movement of cells from a network to another one.

In a preferred embodiment, the device is a multi-well cell culture plate, e.g. a 96-well plate such as the CYTOOplate™ plates offered by CYTOO.

In this case, a network of cell-adhesive patterns is formed in each well of the plate.

The networks can be identical (in terms of shape, size of the cell-adhesive patterns and of distance between adjacent patents) in each well.

Advantageously, at least two wells comprise different networks, said networks differing by at least one of the cell-adhesive pattern shape, size, or distance between adjacent patterns.

This configuration allows testing simultaneously, on a same plate, a plurality of network geometries in order to determine the most suitable design(s) that may be used to define the behavior of a new cell line.

A device comprising at least two different networks may be a useful tool to design a suitable pattern to long-lasting study of a determined type of cell.

To improve the designs and iterative modifications of these networks, mathematical models that predict cell organization on extended networks have been developed.

To that end, dynamic cell shape model can be determined from individual eel- adhesive patterns.

This can be done by an analysis of time lapse images of a cell spreading on a given cell-adhesive pattern, which allows defining a predictive model of the spreading. Then, it is possible to simulate cell behavior on a network formed of several cell- adhesive patterns disposed according to a given layout.

With these preliminary simulations, it is thus possible to optimize the geometries of the networks that are formed on the device.

Then, the selected networks are formed on the device and cells are cultured on each of said networks to study their behavior.

These experiments thus allow confirming the most suitable design(s) for a given cell type.

An embodiment of a 96-well plate comprising a plurality of networks is shown in Figure 8.

The wells (which are here represented as discs) are disposed according to 8 rows (R1 to R8) and 12 columns (C1 to C12).

In the 4 upper rows (R1 to R4), the cell-adhesive patterns have a "Y" shape, with a surface of the pattern varying from 700 μ² to 2200 μηη² from the first to the fourth row.

The cell-adhesive patterns of these networks are arranged according to a honeycomb layout, as shown in Figure 13A.

In the 4 lower rows (R4 to R8), the cell-adhesive patterns have a "L" shape, with a surface of the pattern varying from 700 μ² to 2200 μηη² from the first to the fourth row.

The cell-adhesive patterns of these networks are arranged according to a stairs layout, as shown in Figure 12A.

For each pair of columns (C1 -C2 to C1 1 -C12), the distance between the cell- adhesive patterns varies from 0 μηη to 20 μηη.

It is to be noted that the wells comprised in both rows R1 -R4 and columns C1 -C2 are devoid of any network: the surface of the well is formed of a uniform cell-adhesive surface like in conventional plates.

Cell culture and deposition

Cells are cultured in their classical growing media (Gibco) supplemented with 10% FBS (Sigma) and 0.5% penicilin and streptomycine (Gibco).

Cells are washed with PBS (Gibco) and dissociated using trypsine 0.05% (Gibco) for 2 minutes at 37°C.

Cells are centrifuged 4 minutes at 300 g and the pellet is resuspended in fresh culture media.

Cells are counted and diluted in culture media supplemented with 25 mM Hepes pH=7.4 (Gibco) to obtain a solution at 40 000 cells/ml.

500 μΙ are seeded on a CYTOOchip presenting a panel of networks and mounted into a CYTOOchamber.

After a 2 hrs incubation in the incubator, 2.5 ml of culture media supplemented with 25 mM Hepes pH=7.4 is added in the CYTOOchamber before setting up live cell imaging. Imaging

Living cells are imaged every 30 minutes during 72 hours using a videomicroscope (Ti Eclipse, Nikon) with controlled environment (humidified atmosphere, 7% C02, 37°C).

Cell behavior is recorded using transmitted light during the entire experiment.

Epifluorescence channels can be used to image fluorescent probes, exogenous proteins with fluorescent tags as well as fluorescent networks.

The latter are usually imaged only once at the end of the experimentation.

Video processing and analysis

Films are automatically created from raw images using macros developed at CYTOO on ImageJ software (NIH).

During this process, a grayscale merge of the transmission and fluorescent patterns channels can be realized.

Cell shape and motility parameters are quantified.

Cell motility can be analyzed using the tracking function of Metamorph software (Molecular Devices).

Characterization of cell shape normalization

A device as described above - in particular a multiwall plate as shown on Figure 8 - allows characterizing the normalization of cell shape and polarity of a specific cell line.

Figure 9A shows the characterization of normalization of HeLa cells on a plate as shown in Figure 8.

To that end, HeLa cells were deposited and incubated in each well of the plate during 72 hours.

After 72 hours incubation, the cells were stained for nucleus/actin/focal adhesions (left picture) or nucleus/golgi/centrosome (right picture).

On the right picture, the dashed line represents the nucleus/golgi axis.

Matching (shown with M circles) and non-matching (shown with NM circles) nucleus- golgi axis with respect to pattern branches could be determined.

Figure 9B shows the amount of cell shape normalization on one pattern of each of the 96 wells.

The wells can thus be classified with different levels of cell shape normalization, e.g.:

1 : more than 80% of cells normalized with a good polarity;

2: more than 60% and less than 80% of cells normalized with a good polarity;

3: less than 60% of cells normalized with good polarity;

4: no network.

Figure 9C shows four relative levels of normalization that have been attributed to each well based on the characterization of the normalization of HeLa cells: high (1 ), good ("Γ), medium (2), low (3). As mentioned above, the wells belonging to both rows R1 -R4 and columns C1 -C2 do not comprise any network (level 4) and are thus not considered in the evaluation of normalization.

This experiment shows that network geometries (either "Y" patterns arranged according to a honeycomb layout or "L" patterns arranged according to a stairs layout) with a 12 μηη distance between adjacent patterns are optimal to normalize the shape and polarity of more than 30% of HeLa cells (wells corresponding to levels 1 and 1 '). This rate of 30% is similar to the normalization rate of cells on individual cell-adhesive patterns, which shows that the network according to the invention provides optimal conformation of the cells, whatever the time and/or cell generation. In other words, the successive cell divisions do not result in a loss of the conformation rate of the cell populations. Indeed since the cells are allowed to spread over a large surface of the device, they do not encounter steric effects as with known devices.

The inventors have also shown that the above-described networks respect cell physiology.

The table below shows the comparison of cell doubling time of different cell lines described in the literature (the cells being incubated on a uniform cell-adhesive surface in this case) and measured on networks according to the invention in similar incubation conditions.

This comparison shows that the cell doubling time is substantially similar between the literature and the measurements, which means that the cell physiology is preserved by the networks.

Figure 10 shows the evaluation of cell health after 72 hours incubation on a network by using mitochondria staining (Mitotracker Green).

For each cell type, more than 85% cells are intact (I).

This means that the networks do not bring stress to the cells and thus respect their physiology.

Figure 1 1 shows the quantification of cells with a spindle orientation axis normalized within a cell population cultured on a network.

In this case, the network is either a honeycomb network formed of "Y" patterns with different surfaces or a stairs network formed of linked "L" patterns with different surfaces. In both cases the distance between adjacent patterns is of 12 μηη.

It can be seen on this figure that for selected networks it is possible to obtain more than 70% of the cells having a normalized cell division angle of 15°. Cell behavior has been assessed on different types of networks and the results are described below.

Experimental example n°1

In this first experiment, HeLa kyoto cells were seeded on a network composed of continuous "L" patterns coated with fibronectin (adhesion protein).

Figure 12A shows said network.

Figure 12B shows an elementary arrangement of said network, consisting of two "L" patterns that are linked so as to form a "stairs" shape.

The distance a was in this case of 42.8 μηη.

Cells were imaged by videomicroscopy during 72 hours, with an acquisition each 30 minutes.

Figures 12C-12G show images acquired at different times, i.e. 0 hour (cell deposition); 25.5 hours; 30.5 hours; 43 hours and 66 hours, respectively.

As can be seen on these successive images, cells adhered, spread on a single "L" pattern and adopted a triangular shape.

Cells were able to migrate on different patterns and divide.

Daughter cells moved to the neighboring pattern, thus reproducing the stairs shape of the network.

This method thus induces the control of the cell spatial distribution.

Experimental example n°2

In this second experiment, HeLa kyoto cells were seeded on a "honeycomb" network (see Figure 13A) composed of "Y" patterns coated with fibronectin (adhesion protein) that were separated by a cytophobic area of at least 7.5 μηη (distance b) (see Figure 13B).

The length of each branch of the "Y" patterns was of 18 μηη (distance a) and the height of a honeycomb was of 76 μηη (distance c).

Cells were imaged by videomicroscopy during 72 hours, with an acquisition every 30 minutes.

Figures 13C-13G show images acquired at different successive times.

As can be seen on these successive images, cells adhered, spread on a single "Y" pattern and adopted a triangular shape.

In contrast to Example n°1 , a cytophobic area separates two adjacent patterns and stabilizes cells on their initial position.

However, when cells divide, daughter cells start moving to an adjacent pattern and spread over two neighboring patterns.

Cells were normalized on one or two adhesive patterns for a given timeframe. Experimental example n°3

A549 cells are seeded on a "honeycomb" network (see Figure 14A) composed of "Y" fibronectin patterns that are separated by a cytophobic area of at least 30 μηη (distance b) (see Figure 14B).

The length of each branch of the "Y" patterns was of 18 μηη (distance a) and the height of a honeycomb was of 128 μηη (distance c).

Cells were imaged by videomicroscopy during 72 hours, with 1 acquisition every 30 minutes.

Figures 14C-14F show images acquired at different successive times.

In contrast to Example 1 but similarly to Example 2, the cytophobic area that separates two adjacent patterns maintains the cells on their initial pattern in the network.

However, in contrast to Example 2, when cells divide, the daughter cells can migrate entirely to the adjacent pattern where they completely individualize from the mother cells and adopt a stable triangular shape.

This type of networks allows therefore a tight control of cell distribution and shape normalization over time.

Experimental example n°4

This example is intended to characterize cell behavior by tracking of cell movements using H2B-mCherry cell fluorescence of HeLa Kyoto cells.

Analysis was realized from movies realized as in Example 3 using Metamorph software cell tracking tool.

The tracking results show that HeLa Kyoto cells are mobile on "Y'-based honeycomb network with a 7.5 μηη distance between adjacent patterns (Figure 15A).

This mobility is reduced when the minimal distance between adjacent patterns is increased to 15 μηη (Figure 15B).

Note also that a different orientation of the patterns does not substantially modify the cell behavior (see Figure 15C that shows a "snowflakes" network of "Y" patterns with a distance of 14.5 μηη between adjacent patterns).

As expected, a reduction of the distance between adjacent patterns increases cell mobility, as shown by a "stairs" network composed of "L"-linked patterns (i.e. with a distance between adjacent patterns of 0 μιτι) (Figure 15D).

In conclusion, the network of cell-adhesive patterns as described above have multiple advantages in view of controlling the spatial distribution of cells and of their progeny. First, cells shapes are stabilized on the adhesive pattern after the first cell division, for at least three cell divisions and potentially extendable without time and surface constraint.

In addition, there is no limitation on the cell type used; it is indeed possible to define a network layout by adjusting the shape, size and spatial orientation of the patterns and the distance between adjacent patterns in order to obtain a design specifically suited to the cells of interest.

The network is thus fully customizable depending on the cell to be tested.

As mentioned above, the possibility to test simultaneously on a same multi-well plate various networks layout allows determining quickly the most suitable design(s) for a specific cell type.

Besides, the network is suited to provide a spatial and temporal homogeneity of the cell population distribution.

Hence, the network allows optimization of the cell population distribution, increasing the number of measurements for each condition.

In addition, the network allows a better individualization of daughter cells shapes from one mitosis and direct comparison between them, as well as a comparison of cell shape and migratory behavior within and between cells generations.

Besides, the analysis is possible for cells spread on multiple adhesive patterns, decreasing patterns shape constraint and increasing the compatibility with a broad spectrum of cell types.

At last, the network provides an increase of the SCORE, i.e. the rate of single cell on an adhesive pattern, which is an important parameter in High Content Analysis (HCA).

As already mentioned, the cell types and the shapes, sizes and arrangement of the cell-adhesive patterns described above are only exemplary embodiments and are not intended to limit the scope of the invention.

REFERENCES

EP 1 664 266

US 5,776,748

EP 2 180 042

Claims

1. Method for controlling the spatial distribution of adherent cells and their progeny over time, comprising the steps of:

- providing a device comprising a surface with a plurality of cell-adhesive patterns

(1 ) on said surface, said cell-adhesive patterns being disposed so as to form at least one network (100) consisting of a plurality of recurring elementary arrangements

(10) , each elementary arrangement comprising at least one cell-adhesive pattern (1 ), wherein:

(11) the shape of the cell-adhesive patterns and distance between adjacent patterns of said network is selected so as to promote conformation of one such cell to a pattern and migration of daughter cells to adjacent patterns of said network;

- seeding at least one cell onto at least one cell-adhesive pattern (1 ) of said network;

- culturing said at least one cell such that it divides and progeny moves to other patterns of said network (100), at least part of said cultured cells being conformed to one respective cell-adhesive pattern of the network over time during several cell generations.

2. Method according to claim 1 , characterized in that the disposition of the cell- adhesive patterns and/or of the elementary arrangements within a network is selected so as to control the flow of cell division on said network.

3. Method according to any of claims 1 or 2, characterized in that the size and/or shape of the cell-adhesive patterns is selected so as to allow stabilization of the shape of a single cell seeded on said pattern.

4. Method according to one of claims 1 to 3, characterized in that adjacent cell- adhesive patterns (1 ) of said network are separated at least partially by a cytophobic surface (2).

5. Method according to any of claims 1 to 4, characterized in that the shape of said at least one cell-adhesive pattern is approximately a disc, a crossbow, an L, a Y, a rectangle, or any combination thereof.

6. Method according to any of claims 1 to 4, characterized in that each elementary arrangement consists of a single pattern.

7. Method according to any of claims 1 to 4, characterized in that each elementary arrangement consists of at least two identical patterns.

8. Method according to any of claims 1 to 4, characterized in that each elementary arrangement consists of at least two patterns having different shapes and/or sizes and/or orientations, wherein at least a one pattern, called normalizing pattern, has a shape chosen so as to promote conformation of one cell to said pattern and at least one other pattern, called intermediate pattern, has a shape chosen so as to promote cell displacement from one normalizing pattern to another normalizing pattern.

9. Method according to any of claims 1 to 8, characterized in that said recurring elementary arrangements are disposed in parallel rows and/or columns or in staggered rows.

10. Method according to any of claims 1 to 9, characterized in that said recurring elementary arrangements are arranged according to a regular geometric array or to a fractal figure.

11. Method according to any of claims 1 to 10, wherein the rate of cells conformed to one respective cell-adhesive pattern of the network is greater than or equal to 20%.

12. Method according to claim 1 1 , wherein the rate of cells conformed to one respective cell-adhesive pattern of the network is comprised between 20% and 35%.

13. Method according to any of claims 1 to 12, wherein the distance between adjacent patterns is comprised between 4 μηη and 30 μηη.

14. Method according to any of claims 1 to 13, characterized in that it further comprises acquiring images of the cells during said division and migration and determining cell shape and/or motility parameters from said acquired images.

15. Method according to any of claims 1 to 14, characterized in that it comprises manufacturing said device by the following steps:

- preparing a stamp from said master template,

- inking said stamp with molecules that promote cell attachment,

- contacting the inked stamp with the surface of the device.

16. Cell culture system comprising a device having a surface with a plurality of cell- adhesive patterns on said surface and at least one cell normalized on at least one of said cell-adhesive patterns, said cell-adhesive patterns being disposed so as to form at least one network consisting of a plurality of recurring elementary arrangements, each elementary arrangement comprising at least one cell-adhesive pattern, wherein:

- the shape of the cell-adhesive patterns and distance between adjacent patterns of said network is selected so as to promote conformation of one such cell to a pattern and migration of daughter cells to adjacent patterns of said network.

17. Cell culture system according to claim 16, characterized in that at least one cell and progeny are normalized on adjacent cell-adhesive patterns.

18. Device for carrying out a method according to one of claims 1 to 15, comprising a surface and a plurality of cell-adhesive patterns on said surface, wherein said cell-adhesive patterns are disposed so as to form at least one network, each network consisting of a plurality of recurring elementary arrangements, each elementary arrangement comprising at least one cell-adhesive pattern, wherein the distance between adjacent cell-adhesive patterns of said network is chosen so as to allow a cell seeded on a pattern to move to at least one adjacent pattern.

19. Device according to claim 18, comprising at least two networks that differ by at least one of the size, shape of the cell-adhesive patterns and the distance between adjacent cell-adhesive patterns.