FIELD OF THE INVENTION

The present invention is related to the field of cell separation.

BACKGROUND TO INVENTION Stem cell cultures consist of a specific cell type. Such pure cultures can become heterogeneous cell mixtures when parts of the cell population change their phenotype. However, purifying the culture by selection or separation of a desired cell type is currently expensive and time consuming. We describe here a novel method and device for separating similar cells by size; this is particularly useful when considering separation of haploid cells from diploid cells of the same type, as we have determined that haploid cells are generally smaller than equivalent diploid cells.

WO 96/39979 (Morsiani et al.) discloses a device for isolating cells from a tissue sample using a screening filter comprising pores 10-1000 μηη in diameter. Fluid is passed over a tissue sample, releasing individual and clumps of cells. The cell suspension is passed through a series of screening filters by force generated from a mechanical shaker before being collected in a sterile collection bag. Morsiani does not disclose separation of cell types from a heterogeneous cell population and relies on fluid flow through the screening filters.

US 2002/0192805 (Harris et al.) describes a device for separating cells from degraded tissue or cell dispersions. The device comprises a filter that captures cells from the digested tissue or cell dispersion as it is forced through the collection chamber of the device. Undigested tissue is captured by a first filter or gauze, whilst cells within the digested tissue suspension or cell dispersion are trapped by a second filter comprising pores ranging from 0.1 to 1000 μηη in diameter. The force applied to the cell suspension is described as being centrifugal, pressure, gravitational, vacuum, electromagnetic or centripetal. Harris does not disclose separation of cell types from a heterogeneous cell population and relies on fluid flow through a filter in order to isolate the cells. US 2017/0095772 (Kim et al) describes a filtration device for biological particles such as DNA, viruses or bacteria using a PEG-functionalised nanoporous membrane, which serves to improve viral retention and recovery for detection or diagnosis. In this device, particles are driven through the membrane by osmosis or electric potential. Kim also describes that a polymer brush can be applied to the surface of the membrane to enhance filtering and prevent fouling. Kim does not discuss separation of different cell types, nor does it describe the viability of cells after being filtered through this device. WO 2014/194403 (Mai) describes a collection tube for cells that are smaller than filter pores and pass through the filter, however there is no demonstration of the viability of the cells passing through the filter. Mai also describes a device wherein cells are forced through pores of a membrane. Importantly, no size separation data has been shown for these filtration devices, and these are likely to cause selective cell killing of large cells, meaning the final purity and yield of viable cells is affected. It has been shown that cell viability is poor when cells are forced through pores of a membrane (Freimann and Wutz, 2017)

Desitter et al. (Anticancer Res. Vol 31 , pp427-441 , 201 1 ) describes the use of the device of Mai, separating cells of various sizes from serum or medium and other particles that are considerably smaller than cells. Cells are harvested on a filter membrane, therefore selecting cells that are larger than the pore size which is not useful for selecting haploid cells as these are smaller than diploid cells. In addition, the device is suited to low cell numbers suspended in a large volume such as blood. It can be expected that large numbers of cells will cover the membrane, blocking passage of cells from mixed cell populations; accordingly Desitter teaches that red blood cells should be lysed prior to using the filtration device. Desitter does not allow for separation of cell types and instead involves additional biochemical or biophysical treatment for removal of unwanted cells.

Indeed, research published after the priority date of this application confirms this prediction, showing that separation can be obtained by using a 5 micrometer membrane with a filtration device; however selectivity appears to be largely caused by killing cells as a dramatic cell loss is observed, hence sacrificing cell viability (Qu et al. 2018, Haploid embryonic stem cells can be enriched and maintained by simple filtration. The Journal of Biological Chemistry 293, 5230- 5235).

US 81 14289 (Zheng et al) describes a microfiltration device to capture chosen cell types from biological samples, with a particular focus on the separation of circulating tumour cells from blood. The device comprises top and bottom porous membranes, wherein the two membranes are arranged such that the pores in the top membrane are offset from the pores in the bottom membrane. When a liquid sample passes through these membranes, the cell of choice (i.e. circulating tumour cells) are retained on the membrane due to their larger size, whilst improving viability compared to conventional filtration methods. Zheng describes that preferably samples will be filtrated under some pressure to facilitate the flow of small cells through the filtration apparatus whilst allowing for the capture of larger size tumours cells. Zheng does not discuss separation of haploid and diploid cells.

US 5,026,649 (Lyman) describes a device for growing cells in vitro on a membrane support (Transwell™) within a well of a conventional tissue culture plate. The device comprises a permeable membrane that supports growth of a cellular monolayer on its upper surface. The cells receive nutrients by diffusion through the membrane from the nutrient bath provided in the well. Transwells may also be used for migration assays, wherein a proportion of cells in a heterogeneous population are "pulled" through the pores of the permeable membrane by a chemoattractant gradient. In a migration assay it is essential to choose a pore diameter which allows an active transmigration i.e. smaller than the cell diameter to avoid unspecific dropping of the cells (Kramer et al., 2013). The purpose of the migration assay is therefore to separate cells according to their ability to respond to particular chemoattractants.

Methods for growing haploid embryonic stem cells (ESCs) are described in WO 2012/1 17254 (Wutz et al.), incorporated herein by reference. The haploid ESC lines described therein are karyotypically stable and proliferate in culture. The karyotype of the cell line is maintained by enriching the cell population for haploid cells. WO 2017/000302 (Li et al) describes the production of androgenetic haploid embryonic stem cells which can be used for developing reconstructed embryos and later for producing "semi-cloned" animals. Li further describes that the knockout of two differentially methylated regions, H19 and IG, allows for increased birth rate of such viable animals.

Many methods of purifying and selecting cell types rely on phase separation. However, phase separation methods do not result in viable haploid cells. Therefore, haploid cell enrichment is achieved in WO 2012/1 17254 using fluorescence activated cell sorting (FACS), in particular haploid (1 n) DNA content HOECHST 33342 stained living cells are sorted using a DAKO MoFlo high speed sorter. Similarly, EP 2 599 859 (Elling et a!) describes the generation of stable haploid cell lines and subsequent use of said cell lines in forward and reverse genetics, with a focus on toxin resistance. Generation of haploid cell lines is also by FACS. Such flow cytometric cell sorting methods are often complex and costly. There therefore exists a need to improve methods of cell separation. The present invention addresses this need.

SUMMARY OF INVENTION

The invention relates to a method for purifying cell population mixtures solely based on cell size and the inherent migratory properties of cells themselves. The invention is based on precisely defined micrometer sized pores in a membrane. The pore size is selected to allow a desired cell type to pass through the membrane, while preventing other cell types from doing so. Importantly, however, cells which pass through the membrane do so purely as a consequence of their migratory properties; little or (preferably) no external forces (other than ambient forces, e.g. gravity, atmospheric pressure, fluid column pressure) are applied to the cells, and little or (preferably) no fluid flow through the membrane is used. The pore size is selected for obtaining an optimal cut-off level for the separation of desired cell sizes from all the others present in the input mixture. The pore size is determined by the size and deformability of the cell type (as will be appreciated, different cell types will have different sizes and deformabilities, as will the same cell type from different species, e.g. mammalian species). Generally, this means that pore sizes are selected to be smaller than the smallest fraction of the mixture that should pass through the membrane. Specific preferred pore sizes may be determined experimentally applying a range of pore sizes around an estimated threshold pore size obtained by cell diameter measurements.

The membrane may comprise a track-etched polycarbonate membrane, weaved mesh or similar design of homogenous pore size in the micrometer range. Genetically engineered clonal colonies descending from one haploid cell consist of a large fraction of readily diploidized cells, as self-diploidized cells outgrow haploid cells after a few passages (Kaufman et al., 2001 ). In one aspect the invention relates to a method for purifying phenotypically smaller haploid cells to maintain haploid phenotype in cell culture. In another aspect, the invention relates to a method for purifying clonal haploid cells for further manipulation or study. The method can be used for purification of specific cell cycle states as the different states differ in cell size.

The method of the invention enables cell mixtures to be separated natively by a passive biophysical procedure without requiring staining of cells, chemical treatment, or elaborate technical instrumentation. An important aspect of the method is that damaging of the cell structure is mitigated by applying no or minimal external force and minimal pressure. This means that the cell viability of cells that pass through the separation device is high and allows further applications that rely on living cells. We show herein by comparative data that the method of the invention is far superior to simple filtration, resulting in higher purity and cell viability.

The invention can be tailored to the need of a range of specific applications. For example, it can be adapted to accommodate a wide range of cell numbers (down to a few hundred for small starting populations) or use in a single or multi-well format for parallelized workflows. DETAILED DESCRIPTION

Brief description of the Figures

Figure 1 Schematic representation of the operation of the purification unit

(A) A cell suspension 1 is loaded carefully into the separation unit, the pressure differential Δρ across the porous membrane 3 is kept small and selected cells with a small size start passing through when making contact with culture media 5 placed in containment unit 7. (B) Tapping agitates cells and stimulates movement and prevents clogging; Ap is kept to a minimum. C: Additional culture media is loaded into the separation unit 11 for collecting residual small cells, Ap is kept to a minimum and repeating adding media and tapping for improving yield of cells of interest in the containment unit 7. Figure 2 DNA profile analysis as measured by flow cytometry

(A) DNA profile after staining with HOECHST 33342 of the initial cell mixtures of the HATX3 cell line used for the separation experiment. Depicted is the histogram and the fitted density curve of the DNA content distribution of the cell mixture. The cell cycle phases (G1 , S, G2) for haploid (h) and diploid (d) cells are annotated, respectively. (B) DNA profiles of the HATX3 cell line after 8 passages using the separation unit (dark grey bars, dotted line) and not separating the cells (light grey bars, solid line). (C) Development of the haploidy state of the cell mixtures for the 4 cell lines (DM1 , DM1 .2, DM1 .3, HATX3) over several passages. Points represent the relative contribution of the hG1 and hS phase to the total DNA profile at a specific passage and are a measurement for the overall haploidy state in the cell mixture. Grey scales indicate whether the separation unit was applied between passages or not. Whiskers depict standard deviations (n=3). (D) Jitter plot of karyotype integrity assessed by metaphase spreads after 6 passages of HATX3 using the separation unit (n=1 1 5, mean=1 9.94). Standard deviation and mean are depicted. (E) Metaphase spread of HATX3 cells after 6 passages using the separation unit.

Figure 3 Analysis of cell populations before and after purification or filtration

(A) Green fluorescent haploid mouse ES cells and red fluorescent diploid mouse ES cells were mixed and analyzed by flow cytometry. DNA content was assessed by staining with Hoechst 33342 and the intensity distribution indicates clear ^" I n, 2n, and 4n peaks (white area and outline left to right). Gray and black shading represents the DNA profile of green and red cells or haploid and diploid ES cells, respectively. DNA profiles of red and green cell profiles are separated in the middle and right panel for clarity. The left panel shows an overlay of DNA profiles of the distinguished cell populations (merge). These cells were used as input cell population for testing the separation performance or different purification strategies. (B) Analysis of the cells purified by passing through a Separation Device as detailed herein and in Freimann and Wutz, 2017 (published after the priority date of the current application). The data are presented as in panel A and show a clear enrichment for 1 n DNA content and green cells, with a minor fraction of contaminating diploid (red) cells (C) Analysis of cells filtered through a 8 Mm pore membrane that is identical to the one used in panel B. Note the large 2n peak and substantial contamination by red diploid cells. Approximately half of the cells show no clear fluorescence signal (red or green), which might be indicative of damage. Compared to the input population (panel A) no material enrichment of green haploid cells can be observed over red diploid cells. (D) Analysis of cells after filtration through a 5 Mm pore membrane. A large fraction of cells show no clear fluorescence signal, which might be indicative of damage. The proportion of green haploid to red diploid cells provides no indication for an enrichment of haploid cells.

Figure 4 Development of ES cell cultures at day 2 after purification or filtration

(A) Fluorescence microscopy analysis of cultures that were plated after purification or filtration and grown for two days in a tissue culture incubator. Green and red ES cell colonies are derived from the haploid and diploid ES cells of the input mixture. Note that very few colonies are obtained from the cells after filtration through a 5 Mm pore membrane, whereas a large fraction of red diploid colonies is observed after filtration through a 8 Mm pore membrane. The Separation Device described herein yielded a highly pure culture of green haploid ES cells. (B) Cell numbers for each of the cultures in A were measured by counting using a Hemocytometer. Cultures obtained by purification using the Separation Device described herein contained a 2- fold and 5-fold greater number of cells compared to the cultures obtained after filtration through 8 Mm and 5 Mm pore membranes, respectively.

Definitions

Cell type is defined as a group of cells that can be distinguished from other cells according to cell size (e.g. cell diameter) and at least one other cellular characteristic, including but not limited to ploidy, gene expression (number and/or type of gene), protein expression (number and/or type of proteins), deformability, surface molecule expression, cell cycle state, infection, cell death responses, ability to proliferate and/or capacity for differentiation. The size of a cell can be determined using simple microcopy and cell imaging techniques. The deformability or viscoelastic properties of a cell can be measured by techniques known in the art, including micropipette aspiration, atomic force microscopy, microrheology or microfluidic approaches. Diploid cells are those wherein genetic information is present in two copies.

External forces are forces caused by an external agent outside of the system and exclude the internal forces applied within the system, such as the hydrostatic pressure applied by the cell suspension on the porous filter and the gravitation weight and resistance of the system components. External forces also exclude atmospheric pressure and fluid column pressure. External forces that are not applied to the cell suspension applied to the porous filter include external pressure or vacuum, strong gravitational or pseudo-gravitational forces (such as might result from centrifugation), large hydrostatic pressure and a chemo-attractant gradient that pulls cells across a filter or membrane. Haploid cells are those wherein genetic information is present in only one copy.

HATX3 is a haploid mouse embryonic stem cell line representative for mouse haploid embryonic stem cells (Monfort et al., 2015).

Description Methods for separating cells

The present invention relate to a method for separating cells of a first cell type from cells of a second given cell type, wherein the cells are provided in a single cell suspension, the method comprising: a. applying the cell suspension to a porous filter, wherein the pores of the filter are smaller than the diameter of cells of the first cell type but large enough to allow passage of cells of the second cell type,

b. collecting the cells that pass through the porous filter,

wherein no external force is applied to force the cells through the filter and wherein the collected cells are enriched for the second cell type relative to the cell suspension. The cells of the first cell type are preferably larger in diameter than the cells of the second cell type and/or the cells of the second cell type are more deformable than the cells of the first cell type.

Cells for use in the present method do not comprise a cell wall.

The collected cells may comprise viable cells, wherein preferably at least 50% of the collected cells are viable, more preferably at least 60%, more preferably at least 70%, most preferably at least 90%. The second cell type may constitutes at least 50% of the collected cells, preferably at least 60%, more preferably at least 70%, most preferably at least 95%.

In one embodiment of the invention the hydrostatic pressure across the porous filter is 30 Nm ^"2 or lower in excess of ambient atmospheric pressure. The porous filter used in the method of the invention may form the base of a vessel and the cell suspension is applied to the porous filter by placing the cell suspension in the vessel. The cells may pass through the porous filter into cell culture media contained in a collection dish. The pores of the filter may be 5, 6, 7, 8, 9, 10, 1 1 , 1 2, 13, 14 or 15 μηη in diameter, preferably 5-10, most preferably 8 μηη in diameter. The size of the pores are dependent on the physical characteristics of the first and second cell type. An example of how to determine the appropriate pore size is provided below.

The method of the invention may additional comprise the step of agitating the porous filter. This is not a necessary step but may help to increase the number of collected cells. Note that agitation does not represent an external force applied to force the cells through the filter, within the meaning of the present invention. During this additional step the hydrostatic pressure across the porous filter may be 30 Nm ^"2 or lower in excess of ambient atmospheric pressure. The method of the invention may also or alternatively comprise the additional step of applying media to the porous filter after applying the cell suspension to the filter. The method of the invention is applicable to a method for separating cells of a first cell and a second cell, wherein the first and second cell are provided in a cell suspension.

In embodiments of the method, the first cell type comprises diploid cells and the second cell type comprises haploid cells, preferably wherein the pores of the filter are 8 μηη in diameter. Both or either of the first and second cell types may be stem cells or mouse embryonic stem cells. The cells could also be human embryonic stem cells and/or induced pluripotent stem cells. The method of the invention may also be used to separate triploid, or tetraploid cells from cells of a different ploidy.

The first cell type may comprise feeder cells and the second cell type comprise stem cells. Alternatively, the first cell type may comprise cells at a difference stage of the cell cycle to the cells comprising the second cell type.

Specific cell cycle states can be of interest in the field of Omics. The methods and devices of the present invention can be used to enrich cell cultures for cells in a specific cell cycle state, if the different states differ in cell size. These cells can readily be used for downstream analysis. Normally, this is done by chemical agents arresting cells in a cell cycle phase or by FACS sorting using specific staining techniques.

In one aspect the present invention relates to the method of the invention wherein the first cell type comprises cells in the G ₀ phase and the second cell type comprises cells in the Gi , S, G ₂ and/or M phase.

Genetically engineered clonal colonies descending from one haploid cell consist of a large fraction of readily diploidized cells. To isolate clonal haploid cell lines it is necessary to separate haploid from the diploidized cells as soon as possible and before they are grown to large numbers. Such a purification of low numbers of haploid cells can be performed by the implemented purification unit.

In another aspect the invention relates to a method of separating diploid cells from a cell suspension comprising haploid and diploid cells, comprising:

applying the cell suspension to a porous filter, wherein the pores of the filter are smaller than the diameter of the diploid cells and larger than the diameter of the haploid cells, collecting the cells that passes through the porous filter,

wherein no external force is applied to force the cells through the filter and wherein the collected cells are enriched for haploid cells relative to the cell suspension.

Devices Also provided is a device for use in the method of the invention, wherein the device comprises a vessel, a porous filter and a collection dish, wherein the porous filter forms the base of the vessel.

In another aspect the invention relate to a device for use in the methods of the invention, wherein the device comprises, a vessel, a porous filter and a collection dish, wherein the porous filter forms the base of the vessel. The pores of the filter may be 5, 6, 7, 8, 9, 1 0, 1 1 , 1 2, 13, 14 or 1 5 μηη in diameter, preferably 5-1 0, most preferably 8 μηη in diameter. The size of the pores is dependent on the physical characteristics of the first and second cell type. An example of how to determine the appropriate pore size is provided in the Examples below. The filter may comprise a polycarbonate membrane or weaved mesh. Kits

In another aspect the invention relates to a kit comprising a device according to the invention together with a collection vessel for collecting the cells.

In another aspect the invention relates to a kit comprising a device according to the invention together with a cell suspension comprising cells of a first cell type from cells of a second given cell type, wherein the cells of the first cell type are larger than the cells of the second cell type. The kit may additional comprise media for culturing the cells of the first cell type and/or second cell type. The cells of the first cell type and/or second cell type in the kit may be cryopreserved. In one embodiment, the cells of the first cell type are diploid cells and the cells of the second cell type are haploid cells. The cells of the first and/or second cell type may be stem cells, particularly, mouse embryonic stem cells.

Methods for producing haploid embryonic stem cells

Haploid stem cells are widely used in forward genetics. One of the restrictions to work with haploid stem cells is their intrinsic self-diploidization tendency (duplication of the genetic information). This makes it necessary to select haploid cell cultures after a certain period of time or number of passages. Generally, this is achieved by fluorescence activate cell sorting (FACS) during the passaging of the cells, which requires a sophisticated instrument setup, and training for handling and maintenance. Dyes for staining cells can become toxic to the cells over prolonged exposure and impact on quality or viability. Similarly, microfluidic devices require specific and complex architectures that depend on pumps, piezometers, valves, microscopes etc. When size differences of cells to be separated are small, specificity of sorting or filtering cells of different sizes is often not guaranteed. Centrifugal elutriation has also been used for cell separation based on sedimentation properties. Elutriation centrifuges require expert knowledge to operate and generally cannot be multiplexed or adopted to small cell numbers and small volumes making them less suitable for mammalian cell cultures.

The cell separation device of the invention can be used during routine passaging of cell cultures, when the purification unit removes a large part of the larger diploidized cells, and enriches the newly seeded cultures for haploid cells. In particular, in one aspect the present invention relates to methods for producing a mammalian haploid embryonic stem cell, comprising: a. activating isolated oocytes in vitro to produce haploid embryos;

b. culturing the activated embryos to the 8-cell, morula or blastocyst stage; c. removing the zona pellucida;

d. isolating the inner cell mass of the activated embryos;

e. further culturing the inner cell mass

f. suspending the cultured cells to produce a cell suspension of haploid and diploid cells,

g. enriching the culture for haploid cells according to the methods of the invention for separating cells.

EXAMPLE I: Implementation of a separation device

In a particular non-limiting embodiment the invention comprises a purification/separation unit 11 comprising a membrane 3 of the appropriate pore size attached to the bottom of an open vessel. Cells are suspended in culture medium and placed within the vessel. As cells need to be viable after passing the pores, no external pressure or vacuum or strong gravitational forces (such as might result from centrifugation) or large hydrostatic pressure is applied to the device (i.e. force-free separation). The pressure should be no higher than 30 Nm ^"2 in excess of ambient atmospheric pressure (equivalent to a fill height of approximately 30 mm). Excessive pressure will lead to cell loss due to damage. The separation effect is ideally due to the cells sinking through the membrane pores without any net flow of fluid. For this, a minimum higher fluid level above the membrane is tolerable above the fluid level in the collecting vessel as long as the net flow of fluid is low enough. Practically, the medium should not be filled higher than to a level of approximately 0.7 mm. In an alternative embodiment, the medium should not be filled higher than to a level of approximately 3mm.

Separation is started by wetting the lower surface of the unit in the collection unit culture medium. The medium and cells of interest within the separation device start passing through the pores of the membrane and can be collected and used for further cultivation or downstream applications. Figure 1 shows a schematic representation of the operation of the purification unit. Hydrostatic pressure differential Δρ across the porous membrane 3 within the separation unit 11 is kept small to minimize force on cells as excess force will break mammalian cells on passage of a micrometer sized pore (Figure 1A). The separation device 11 is constructed with a pore size that allows viable passage of haploid but not diploid cells under these conditions. Gentle tapping of the separation unit for agitation of the cells can obtain a high cell yield in the collection well (Figure 1 B). Clogging of pores is prevented by gently tapping the vessel at the top. Some more medium can be added to purification unit until most cells of interest are transferred into the collection unit (Figure 1 C). The hydrostatic pressure should be kept to a minimum and must not exceed a level where cells viability is compromised.

For haploid mouse embryonic stem cells (tested as a mixture of haploid and diploid cells) the optimal pore size was determined to be 8 μηη. At 5 μηη pore size, virtually no cells migrate through the filter.

Other cell types that could be enriched using the methods would include different cell cycle stages when cell size differences could be exploited for separation (cells increase in size before cell division and thereafter separate into two smaller daughter cells), any situation where ploidy differences are of interest (this could be applied to higher ploidy levels either naturally occurring or after cell fusions, eg. triploid, or tetraploid), and separation of stem cells form mostly larger support (or feeder) cells. In one embodiment the separation unit has a 8μηη pore size polycarbonate membranes. The separation unit is constructed from a standard 5 ml tube that is 75 mm high and has a diameter of 12 mm. The bottom of the tube has been cut off and a membrane was attached to the other end of the tube with a commercial plastic adhesive.

EXAMPLE II: Characterisation of the separation device Experiments to characterize the invention and adopt it in routine cell culture use have also been carried out. For measuring yield and purity two cell lines were used: A haploid ES cell line carrying a green fluorescence marker expression transgene and one diploid mouse ES cell line carrying a red fluorescence expression. Mixing green fluorescent haploid ES cells and red fluorescent diploid ES cells facilitates the analysis of purification performance of our invention and comparison to traditional filtration through filter membranes. Indeed, the idea of using membranes with pores of defined sizes has been considered in the past but has not led to adoption for size separation of mammalian cells. We argue that this is due to the process of filtration being inadequate to leverage the size separation principle of membranes with micrometer sized pores to mammalian cells that possess no cell wall and are mechanically fragile. Our experiments below confirm this notion and demonstrate the dramatic difference between our new implementation and workflow and traditional filtration approaches.

In our experiments we consider our separation device using a 8 micrometer pore size and filtration cartridges using 8 micrometer and 5 micrometer pore sizes. The latter 5 micrometer pore sizes are much more narrow than the expected cell diameter and are tested for the reason that a study published after the priority date of the current application has appeared that documents enrichment of haploid ES cells specifically with 5 micrometer pores (Qu et al., 2018).

Figure 3 provides an analysis of cell populations before and after purification using the different cell separation setups. Whereas our method achieves excellent enrichment of haploid ES cells with less than 10% percent diploid G1 cells (red) contamination, no enrichment of haploid over diploid cells can be measured after filtration through a 8 micrometer pore membrane. We note in addition that cells appear to have lost their fluorescence quite dramatically, which we suggest is a consequence of rupturing of the cell. Even when using 5 micrometer pore sizes filtration does not result in a measurable selection of haploid ES cells over diploid ES cells, but cell damage appears drastically enhanced. These data demonstrate that a traditional filtration workflow using micrometer pore membranes does not result in cell size separation. We do observe severe cell damage inferred through the observed weakening or loss of fluorescence of the cells that might provide some selection effect through killing large cells more efficiently than small cells.

To further address the practical use case of how cultures develop after enriching with the three different setups, we plated the purified cells and analyzed the resulting cultures after two days of cell growth.

Figure 4 provides fluorescence microscopy images, and cell counts of the resulting cultures. Our method obtains approximately 5 fold higher cell numbers than filtration through 5 micrometer pore membranes showing a superior performance in cell viability. Cultures that were filtered through 8 micrometer pores show a high level of contaminating (red) cells that shows that a large number of diploid cells have passed through the filter preventing a meaningful enrichment for haploid ES cells over diploid ES cells. Notably, even though the overall purity of these cultures is much reduced, overall cell numbers remain 2- and 5-fold reduced when compared to cultures obtained by purification through our method (Separation Device). Only the latter yields highly enriched haploid ES cell cultures. For this we calculate the total yield of haploid ES cells to 5 times higher in absolute terms by using our method compared to filtration through 5 micrometer pores. This demonstrates that irrespective of the micrometer pore size filtration strongly reduces the number of haploid ES cells compared to our method and therefore makes expansion of haploid ES cells through filtration difficult to attain. In the case of filtration through 8 micrometer membranes the purity of the cultures is further judged as insufficient to be useful for many technical applications including mutagenesis for screening or isolation of new haploid ES cell lines.

In conclusion, our data demonstrate that the specific implementation of our method can effectively enrich for cell size and provides at least 5 fold higher yields of haploid ES cells, with minimal contamination of diploid ES cells. This is crucial for effective expansion of haploid ES cell cultures and facilitates the frequent use of our separation device for passaging haploid ES cell and maintaining highly pure haploid cultures. A similar enrichment has not been observed by using a traditional filtration setup and workflow irrespective of the choice of pore diameter REFERNCES

Kaufman, D.S., Hanson, E.T., Lewis, R.L., Auerbach, R., and Thomson, J.A. (2001 ). Hematopoietic colony-forming cells derived from human embryonic stem cells. Proc. Natl. Acad. Sci. U. S. A. 98, 10716-10721 .

Kramer, N., Walzl, A., Linger, C, Rosner, M., Krupitza, G., Hengstschlager, M., and Dolznig, H. (2013). In vitro cell migration and invasion assays. Mutat. Res. 752, 10-24.

Monfort, A., Di Minin, G., Postlmayr, A., Freimann, R., Arieti, F., Thore, S., and Wutz, A. (2015). Identification of Spen as a Crucial Factor for Xist Function through Forward Genetic Screening in Haploid Embryonic Stem Cells. Cell Rep. 12, 554-561 . Desitter, I., Guerrouahen, B.S., Benali-Furet, N., Wechsler, J., Janne, P. A., Kuang, Y., Yanagita, M., Wang, L., Berkowitz, J.A., Distel, R.J., et al. (201 1 ). A new device for rapid isolation by size and characterization of rare circulating tumor cells. Anticancer research 31, 427-441 .

Freimann, R., and Wutz, A. (2017). A fast and efficient size separation method for haploid embryonic stem cells. Biomicrofluidics 11, 0541 17.

Qu, C, Yan, M., Yang, S., Wang, L, Yin, Q., Liu, Y., Chen, Y., and Li, J. (2018). Haploid embryonic stem cells can be enriched and maintained by simple filtration. The Journal of biological chemistry 293, 5230-5235.