WO2003014337A2

WO2003014337A2 - Fusion of cells

Info

Publication number: WO2003014337A2
Application number: PCT/GB2002/003570
Authority: WO
Inventors: Peter Walter Andrews; Andrew Ferguson Shering; Marzena Agnieszka Flasza
Original assignee: Intercytex Limited
Priority date: 2001-08-03
Filing date: 2002-08-02
Publication date: 2003-02-20
Also published as: GB0118984D0; WO2003014337A3

Abstract

The invention provides in one aspect a method of producing a fused cell, comprising the steps of providing a porous filter, allowing parent cells to attach to either side of the porous filter, and causing fusion of the cell membranes through the pores of the porous filter so that the cytoplasms of the cells are contiguous through the porous filter whilst the chromosomes or nuclei of the parent cells remain separated by the porous filter. The invention also provides cells derived from fused cells, and compositions and tissues comprising cells derived from fused cells.

Description

Fusion of Cells

The invention relates to methods of producing fused cells, cells derived from fused cells, and compositions and tissues comprising cells derived from fused cells.

Heterokaryons are cells that contain two or more nuclei from different cell types and may or may not contain mixed cytoplasm from different cell types. Heterokaryon formation is an important concept in cell biology as the resulting cells display behaviours that were unique to each parent e.g. in the formation of hybridomas for monoclonal antibody production. In this scenario, a mortal antibody-producing B lymphocyte is fused with an immortal myeloma cell resulting in a hybridoma, which is an immortal cell line capable of producing a monoclonal antibody. Similarly, it may be advantageous to fuse any two cells in order to combine the characteristics > of both cells in a single hybrid cell. In the example of hybridoma cells, the B lymphocyte is a target cell enhanced by gaining the characteristic of immortality through fusion with a myeloma cell.

In addition, fusion is an important means for reprogramming of one cell by another. In this case, the nucleus of a target cell to be reprogrammed may be exposed to components, including any cellular proteins or other macromolecules, of another cell (a "reprogramming cell") by fusion of the target cell and the reprogramming cell to bring about nuclear reprogramming by the cytoplasm or cytoplasmic extract from the reprogramming cell. For this purpose, the cytoplasm or cytoplasmic extract could include any of the components within the cell whether originating in the body or the nucleus of the reprogramming cell. Reprogramming originates in the nucleus of the target cell where changes in chromatin structure lead to a modification of the profile of genes expressed in that cell.

Proteins that can effect reprogramming in the target cell are nuclear factors since they must enter the nucleus of the target cell to effect reprogramming. Nuclear factors are proteins (or RNAs) normally bound within the nuclear membrane (except during M-phase of mitosis in somatic cells and meiosis in germ cells) . Nuclear factors may include heteronuclear RNA ("hnRNA", i.e. messenger RNA prior to processing and export) . The hnRNA may encode reprogramming factors. The nuclear factors may include DNA binding proteins bound in chromatin to the chromosomes, for example histones, transcription factors and other ancillary factors that may affect gene expression (either directly or indirectly) .

After the fusion event a hybrid cell may enter M-phase

(mitosis in somatic cells, meiosis in eggs and oocytes) during which the nuclear membrane is broken down and proteins normally sequestered in the nucleus become free in the cytoplasm and able to act on the other cell's nucleus. Therefore, a cytoplasm may contain factors that would normally be located in the nucleus. Similarly, in eggs and oocytes the nuclear membrane is disassembled in the process of entering meiosis, a state equivalent to mitosis in somatic cells. The nucleus during M-phase ceases to exist as a discrete organelle. The egg/oocyte remains arrested in this state until fertilisation takes place. During M-phase (mitosis or meiosis), nuclear and cytoplasm components are found within the same soluble cytosol at physiological concentrations and stoichiometry. The term reprogramming is applied to nuclei/nuclear factors that are brought into the environment of a reprogramming cell, rather than to the target cell per se . For example, Dolly resulted from the implantation of the nucleus of a mammary cell from a 6-year-old ewe into an enucleated egg from another ewe . The nucleus would originally have been expressing genes specific to mammary gland function but, placed into the context of the egg, this pattern of expression was reset to reflect the new cellular environment such that expression of egg-specific genes would have been induced. This reprogramming occurred to the extent that the reprogrammed nucleus was capable of supporting the full development of the cloned sheep known as Dolly (see ilmut, I., A. E. Schnieke, et al . (1997) "Viable offspring derived from fetal and adult mammalian cells [see comments] [published erratum appears in Nature 1997 Mar 13;386 (6621): 200]" Nature 385(6619): 810-3).

Another example of nuclear reprogramming occurs when sperm enter egg cytoplasm; the sperm head is reconfigured to become a pronucleus and demonstrable changes in chromatin structure and capabilities occur (see Clarke, H. J. and Y. Masui (1986) . "Transformation of sperm nuclei to metaphase chromosomes in the cytoplasm of maturing oocytes of the mouse" J Cell Biol 102(3): 1039-46; Harrouk, . and H. J. Clarke (1993) "Sperm chromatin acquires an activity that induces microtubule assembly during residence in the cytoplasm of metaphase oocytes of the mouse" Chromosoma 102 (4) : 279-86) .

Furthermore, Kikyo et al . have demonstrated specific chromatin changes occurring in permeabilised target Xenopus somatic nuclei treated with Xenopus egg extracts (see Kikyo. N. , P. A. Wade, et al . (2000) "Active remodelling of somatic nuclei in egg cytoplasm by the nucleosomal ATPase ISWI." Science 289(5488): 2360-2.) In the specific example of Kikyo et al . (2000), the ingress of nuclear factors from the egg cytoplasm into the target nucleus was demonstrated. Previously, Gurdon et al .

(1979, "Reprogramming of transplanted nuclei in amphibia." Int. Rev. Cytol . Suppl . 9: 161-178) described the exchange of proteins between Xenopus egg cytoplasm and a target nucleus .

In the examples of reprogramming described above, the target nucleus was freely accessible to the mixture of nuclear and cytoplasmic factors in the reprogramming cell (egg or oocyte) . Similarly, during fusion of somatic cells, if the reprogramming cell were in the mitotic state at the time of fusion, the target nucleus would be brought into direct contact with nuclear factors from the reprogramming cell.

A further example of reprogramming using somatic cells is presented in Tada et al . (Tada, M., T. Tada, et al . (1997) "Embryonic germ cells induce epigenetic reprogramming of somatic nucleus in hybrid cells." Embo J 16(21): 6510-20) which describes altered patterns of methylations and gene expression in thymic lymphocytes fused to embryonal germ (EG) cells. EG cells are, like embryonal stem cells (ES) and embryonal carcinoma (EC) cells, pluripotent. Similarly, EC cell reprogramming capabilities are suggested by the recent demonstration of expression of the embryonal marker, OCT4 , in mouse thymocytes fused to human EC cells (WO 00/49138)

There are several technologies available for the fusion of two cell types resulting in the formation of heterokaryons . These technologies include Sendai virus mediated fusion; mechano-chemical methods e.g. polyethylene glycol (PEG) mediated fusion, cytochalasin/centrifugation and micromanipulation and electrofusion.

After successful heterokaryon formation, the cell may undergo cell division. This may result in unequal distribution of the genetic material from the nuclei of the two parent cells in the daughter cells; e.g. human chromosomes are preferentially lost in human/mouse cell fusion. However, several rounds of transcription from the programming cell nucleus may be required as well as endogenous factors in the cytoplasm to maintain the appropriate transcription factors needed for the correct remodelling of the genome in the target nucleus to be reprogrammed. Thus, while it may be necessary that the two cells' contents be in contact for an extended period in order to effect reprogramming or phenotypic alteration, it could be disadvantageous in the long term (due to genomic instability of the hybrid cell) for the two nuclei to inhabit the same cell body beyond the time required to effect reprogramming or other desirable phenotypic change.

A method that would allow fusion of two cells such that phenotypic changes could be effected but that would permit separation of the two partners after that change had taken place would therefore be advantageous . Such a method would allow phenotypic changes to take place in the target cell but would preclude genotypic change by physically separating the nucleus (or at least the chromosomal component of the nucleus) of the modifier cell from that of the target cell.

In a first aspect, the invention provides a method of producing a fused cell, comprising the steps of: (i) providing a porous filter; (ii) allowing a first parent cell to attach to one side of the porous filter and a second parent cell to attach to the other side of the porous filter; and

(iii) causing fusion of the cell membranes through the pores of the porous filter so that the cell cytoplasms are contiguous through the porous filter whilst the chromosomes of the parent cells remain separated by the porous filter.

The term chromosome (s) in the present application is intended to include not only intact chromosome (s) but also genomic genetic information in general.

In one embodiment, the nuclei of the parent cells remain separated by the porous filter.

The invention advantageously allows cell fusion across a porous membrane in which the nuclei or chromosomes of the parent cells are kept apart by the porous membrane.

In another embodiment, the method of the present invention includes providing one or more additional porous filters, allowing fusion of the cell cytoplasms of further parent cells whilst the nuclei of the parent cells remain separated by the porous filter.

Any cells could be used as the parent cells, provided they may be positioned onto the porous filter. In one embodiment of the invention, chemotactic agents can be used to encourage formation of pseudopodia from the cell membranes through the pores of the porous filter, to encourage fusion of the parent cells.

The invention provides a method further comprising the step of culturing the fused cell under suitable conditions to enable the parent cells which have been fused to remain viable whilst keeping the nuclei of the parent cells apart .

In a preferred aspect of the present invention, one cell to be fused is a reprogramming cell and another is a target cell to be reprogrammed. In a further aspect of the invention, the reprogramming cell has a desirable phenotypic characteristic and the target cell is a cell in which it would be advantageous to gain that phenotypic character of the reprogramming cell .

The invention provides a method further comprising culturing the fused cell under suitable conditions to maintaining the fused cell for a period of time to allow the reprogramming cell to produce factors to reprogram the target cell nucleus.

In another aspect of the present invention there is provided a further step of causing separation of the fused cell into a first parent -derived cell and a second parent- derived cell, each of which has maintained the integrity of their nuclei and/or the segregation of their chromosomes. This may enable regeneration of stable diploid cells from the parent-derived cells. The method may comprise a further step of isolating the first parent- derived cell and/or the second parent-derived cell. The method may also comprise a further step of isolating a reprogrammed parent-derived cell.

When a fused cell according to the invention is split into two, the parent cell nuclei generally remain associated with their respective outer cell membranes, and it is only the cytoplasmic material that may have been exchanged and/or mixed whilst the cells were fused. However, some mixing of the cells' outer membranes may occur and the resulting nuclei may not be uniquely and exclusively associated with their original cell membrane.

The porous membrane may be of variable thickness depending on the ability of the reprogramming and target cells to extend filopodia (i.e. pseudopodia) through the pores. Similarly, the density of the pores can be altered to maximise the probability of a cell fusion event. The pores of the porous filter may be about 0.2-10 μm, preferably 0.4-4 μm and even more preferably 1 μm in diameter. A pore size of lOμm seems a reasonable maximum since it would appear to be the largest size pore that would exclude nuclei from traversing the barrier between two cells. The minimum pore size may be about 0.2-0.4 μm as pores smaller than this may prevent filopodial (i.e. pseudopodial) growth.

A desirable pore size depends on the cell type used. For example, fibroblasts have large nuclei in excess of lOμm whereas lymphocytes have nuclear diameters of less than lOμm.

The porous filter may be 5-100 μm thick, and preferably 5- 25 μm thick. The thickness of membrane should be such as to allow cell-to-cell contact through the pores.

As used herein the term "porous filter" is to be understood to define a barrier having pores therethrough, which pores are of a predetermined size so as to exclude nuclei and or chromosomes from traversing the barrier between two cells separated by the barrier. The porous filter should be permeable to cytoplasm, cytoplasmic extract or cytoplasmic components, including for example nuclear factors such as nuclear proteins and/or hnRNA. Fusion of the parent cells is preferably induced by an electrical field. Optimal electrofusion parameters may be determined for each application and will depend, amongst other things, on the cells to be fused. An electrical pulse with a voltage of 100V-500V, for example 260V, may be used. An electrical pulse with a capacitance of 400μF -

900μF, for example 750μF, may be used.

In one embodiment of the invention, cells on opposite sides of the porous membrane are induced to fuse by the application of chemical agents such as digitonin (and/or other detergents) , streptolysin O (and/or other pore forming microbial agents) and/or polyethylene glycol. Other mechanisms may also cause cell fusion such as mechanical vibration or sonication.

In a preferred embodiment, the parent cells are plated in such a density so that at the point of fusion there exists a confluent monolayer of cells on both sides of the porous membrane. This may allow the probability of pseudopodia (or filopodia) of two parent cells meeting each other in the pores of the membrane to be maximised.

The invention also provides a composition or pharmaceutical composition comprising or consisting of a cell that has been fused according to the invention, and then divided. This composition may include a pharmaceutically acceptable carrier, excipient or diluent.

The parent cells divided from the fused cells produced according to this method may be useful in a composition or pharmaceutical composition, particularly where the nucleus of one of the parent cells has been reprogrammed.

In one embodiment, fused cells (i.e. cells with two nuclei) would form only as an intermediate composition and the final composition would utilise only separated cells

(i.e. cells with a single nucleus). However, a resultant single-nucleus cell may contain part of the plasma membranes and soluble material, including organelles such as mitochondria, from both parent cells employed in order to form that single cell. Thus, the final composition may comprise elements, excluding the nucleus but not excluding nuclear proteins, that were previously uniquely associated with the other cell .

Further, a cell produced by fusion of parent cells and subsequent separation, i.e. a parent-derived single cell, may be used in the production of tissue. Examples of such tissue types may be selected from but are not limited to the list comprising neural, smooth muscle, striated muscle, cardiac muscle, bone, cartilage, liver, kidney, respiratory epithelium, haematopoietic cells, spleen, skin, stomach, pancreas and intestine.

A tissue produced from parent-derived single cells made according to the invention may be used for transplantation .

The invention also provides a fused cell produced by the method of the invention. The invention further provides a stem cell or cell line or a progenitor cell or cell line produced by the method of the invention.

The invention further provides a method for generating a population of phenotypically modified cells.

Also provided is a pluripotent or multipotent stem cell or cell line or a progenitor cell or cell line produced by the method of the invention using ES cells or EC cells as one of the parent cells. The invention may be used in a method for determining efficiency of cell fusion. The method may comprise examining the extent to which a fluorescent dye present in the cytoplasm of a cell on one side of the porous membrane moves to the cytoplasm of a cell on the opposite side of the porous membrane.

The invention may be used in a method for assessing the extent of reprogramming of a target cell by a reprogramming cell. This method may be accomplished by examination of the morphology of the target cell. The target cell may be expected to acquire some of the morphological appearances of the reprogramming cell, for example as viewed by microscopy and/or by the use of specific antibody markers.

Also, the presence of fluorescent cell tracker dyes may be examined by fluorescence activated cell sorting (FACS) .

The invention may be used in a method for assessing reprogramming of a target cell. It is expected that the reprogrammed target cell will express new mRNAs and proteins because of the reprogramming event . These new molecules may be detected by removal of the target cell from the membrane and analysis by polymerase chain reaction (PCR) and immunocytochemistry. Gene expression may be examined by PCR analysis of pluripotent cell- specific markers (for example Oct3/4, Sox2 , Fgf4, Rexl, PEA3 , Utfl) and/or differentiated cell-specific markers (for example laminin Bl for parietal endoderm, NeuroD for neural cells) .

Using these above methods to determine, for example, cell number, cell morphology, mRNA expression and/or protein expression, the efficiency of the reprogramming event may be determined for each type of cell partner used and for steps of the fusion process, for example electrical parameters and/or buffer constituency.

Fused cells may be separated from the parent cells after reprogramming. A separated fused cell should not contain reprogramming cell genomic DNA. Quality control tests for the resulting fusion products are described herein. This may be done by karyotyping of the fusion to identify the chromosomes and confirming, for example, that a diploid cell has remains diploid. The fusion process can also be quality controlled by maintaining a normal Alu sequence PCR profile for human cells.

The invention allows for the reprogramming of a differentiated target cell by fusion with a differentiated reprogramming cell i.e. from one differentiated phenotype to another (trans-differentiation) to give a reprogrammed differentiated cell.

In another aspect, the differentiated target cell may be fused with a pluripotent undifferentiated cell to give a deprogrammed target cell with the same genetic constituency as the original target cell (deprogramming) .

One aim of the method of the invention may be to effect a phenotypic change in a target cell without a corresponding genetic change. This method could be of use to any technique in which the effects of an intracellular substance produced from one cell is tested on the intracellular compartment of another without having the nuclei of the cells merging. For instance, one example might be fusing a transfected cell that is over expressing or expressing ectopically a particular protein with a second cell type, and monitoring the effect. In this respect this method of cell fusion uses a living cell as a means to deliver substances to a target cell.

A related alternative aspect of the invention is to allow the fusion of two cells across a membrane between a target cell and a lethally irradiated cell where nuclear components can freely diffuse into the target cell but the DNA of the irradiated cell has been destroyed.

Various embodiments of the invention will now be described by way of example with reference to the figures, of which:

Fig. 1A is a schematic diagram of an electrofusion method showing plating of cells onto a porous membrane;

Fig. IB is a schematic diagram showing a cross-section of a cell-laden insert within one well of a six-well plate and a magnified area of the cell-laden insert;

Fig. 2A is a micrograph showing high plating density of PCC4 EC cells on a porous membrane;

Fig. 2B is a micrograph showing low plating density of PCC4 EC cells on a porous membrane;

Fig. 3A is a schematic diagram showing top layer (TL) and bottom later (BL) plated on each side of a porous membrane (PM) ;

Fig. 3B is a composite micrograph of top layer PCC4 EC cells plated onto the porous membrane of Fig. 3A;

Fig. 3C is a composite micrograph showing the middle section of the porous membrane of Fig. 3A; Fig. 3D is a composite micrograph of bottom layer PCC4 EC cells plated onto the porous membrane of Fig. 3A;

Fig. 4A is a graph showing effect of a single electrical pulse at voltages from 0-300V on the survival of CEM cells;

Fig. 4B is a graph showing effect of two electrical pulses at voltages from 0-300V each on the survival of CEM cells;

Fig. 5A is a photograph showing PCC4 EC cells subjected to a single electric pulse at voltages from 0-300V;

Fig. 5B is a photograph showing PCC4 EC cells subjected to two electric pulses at voltages from 0-300V each;

Fig. 6A is a photograph showing PCC4 EC cells subjected to a single electric pulse of 0-50 μS;

Fig. 6B is a photograph showing PCC4 EC cells subjected to a single electric pulse of 60-100 μS;

Fig. 6C is a photograph showing PCC4 EC cells subjected to two electric pulses of 0-50 μS each;

Fig. 6D is a photograph showing PCC4 EC cells subjected to two electric pulses of 60-100 μS each;

Fig. 7A is a dot plot from a flow cytometric analysis of CEM and PCC4 EC cells admixed in CDMEM medium and not subjected to electrofusion; Fig. 7B is a dot plot from a flow cytometric analysis of CEM and PCC4 EC cells admixed in CDMEM medium and electrofused at 180V;

Fig. 7C is a dot plot from a flow cytometric analysis of CEM and PCC4 EC cells admixed in mannitol buffer and electrofused at 180V;

Fig. 8A is a dot plot from a flow cytometric analysis of CEM and PCC4 EC cells admixed in mannitol buffer and not subjected to electrofusion;

Fig. 8B is a dot plot from a flow cytometric analysis of CEM and PCC4 EC cells admixed in DMEM medium and electrofused at 100V;

Fig. 8C is a dot plot from a flow cytometric analysis of CEM and PCC4 EC cells admixed in mannitol buffer and electrofused at 400V;

Fig. 9A is a micrograph showing DAPI staining of a PCC4 EC cell electrofused to a CEM cell;

Fig. 9B is a micrograph showing CMTFDA staining of the PCC4 EC cell and the CEM cell of Fig. 9A;

Fig. 9C is a micrograph showing CMTMR staining of the PCC4 EC cell and the CEM cell of Fig. 9A;

Fig. 9D is a composite micrograph combining the images of Fig. 9B and Fig. 9C;

Fig. 10A is a micrograph showing a cross-sectional reconstruction of a porous membrane with an upper and lower layer of cells in which the lower layer of cells has been CMFDA-stained and no electrofusion has taken place; Fig. 10B is a micrograph showing a cross-sectional reconstruction of the porous membrane and cell layers of Fig. 10A after electrofusion at 260V;

Fig. 11A is a schematic diagram showing a method of detaching upper and lower parent cells from a porous membrane ;

Fig. 11B is a schematic diagram showing a method of detaching upper parent cells from a porous membrane; and

Fig. 11C is a schematic diagram showing a method of detaching lower parent cells from a porous membrane.

Fig. 1A is a flow diagram indicating a method by which cells are adhered to both sides of the porous membrane. The steps may comprise:

1. Seeding a first population of cells onto the outer surface of membrane (for example, in a 500μl drop) ;

2. Allowing cells to adhere for 4h-24h in a sterile environment;

3. Seeding a second population of cells onto the inner surface of the membrane ; 4. Placing an insert into a well of a six-well plate and allowing cells to grow; and

5. When the cell-laden insert is ready, removing it and placing it into the electroporator .

Alternatively, the steps may comprise:

1. Seeding SNARF-1 labelled cells onto base of an insert ;

2. Allowing cells to attach overnight in incubator, then inverting into a well of a six-well plate; 3. While in six-well plate, seeding CMFDA labelled cells onto inside base of insert;

4. Allowing the second cell type to attach and leaving in incubator for 2-3 days for cells to grow and extend processes; and

5. After 2-3 days, removing the insert from six-well plate and placing into an electrofusion device.

Replacing insert back in six-well plate after electrofusion.

Fig. IB shows in cross-section a view of labelled cells attached to both sides of a porous membrane in low magnification and high magnification.

Figure legend: (1) Insert;

(2) Well of six-well plate;

(3) Tissue culture medium;

( 4 ) Porous membrane ;

(5) Labeled cells attached to the membrane; and (6) Pseudopodium

The lower diagram in Fig. IB illustrates how cells that have adhered to the porous membrane (indicated by thick black dashed line) produce pseudopodia (6) enter the pores and meet the pseudopodia of cells from the other surface, whereas nuclei are too big to pass through the pores.

Nuclear segregation of the fused cell may be maintained by keeping the parent cells separated by a PET lμm track- etched porous filter (Falcon Cat. No.3102) . Cells attach to the surface of one side of the porous filter and extend pseudopodia through pores in the porous filter to make surface contact with pseudopodia from cells on the surface of the other side. An electric field or other fusogenic agents as detailed herein are used to destabilise the opposing cell membranes, inducing fusion of the opposing cells through the intervening PET porous filter. Fusion allows free mixing of the cytoplasm of both cell types whilst the lμm pore maintains separation, integrity and continued transcription of the nuclei of the two cell types . This results in the formation of a fused cell that remains viable in culture whilst maintaining the nuclear integrity of the parent cells.

In an example in which the fused cell comprises one parent cell that is a reprogramming cell and another parent cell that is a cell to be reprogrammed, the reprogramming cell continues to produce reprogramming factors without loss of genetic material. Furthermore the resulting fused cell can be separated, avoiding the mixing of the nuclear DNA of the cells.

It is envisaged that the cells fused to form a fused cell may be separated by a porous filter but juxtaposed i.e. the cells may be directly opposed each other on either side of the porous filter. However, the cells may also be spaced laterally apart on either side of the porous filter such that the pseudopodia from one cell would have to extend through the porous filter but then also extend along the other side of the porous filter before contacting the other cell.

In addition, it is envisaged that there may be more than a single cell on each side of the porous filter. Fusion would occur between pairs separated by the porous filter but directly opposed. However, it may also be envisaged that one side of the porous filter has a number of cells immobilised on its surface, or possibly grown until a confluent layer is formed and the other side of the porous filter having only a discrete cell and/or cells immobilised. There are other ways of implementing the method such as using different porous filter pore sizes. A range of pore sizes of from about 0.2μm to about lOμm may be used.

Various porous filters may be used, examples of which are track-etched PET, hydrophilised PTFE and Cyclopore, which are proprietary inserts made by Corning Falcon and Costar.

Different electrofusion parameters may also be used such as different potential differences, shunt resistance and capacitance .

Alteration of the pore size of the membrane can be employed to perform experiments which allow cell fusion across the membrane but restrict the passage of various organelles depending on the pore size. Specifically, pore sizes greater than lOμm will allow the exchange of the nucleus but not the cytoskeleton, pore sizes from lμm to lOμm will exlude the nucleus but allow mitochondria to pass through. Pore sizes of less than lμm will allow only small organelles such as vesicles and molecules to pass through .

Experimental

The method of the invention involves fusing two cells growing on opposite faces of the same porous membrane. The method may be improved by examining several contingent aspects :

1) Methods of growing cells on opposite sides of an appropriate porous membrane;

2) Methods of fusion to allow the cellular components of cells to joined (admixed) ;

3) Separation of fused cells from each other; and

4) Detection of phenotypic change after fusion. The order of steps 3) and 4) may be reversed.

Examples of several of these contingent steps are illustrated below.

1) Growing cells on a porous membrane

Electrofusion across the porous membrane involves selection of a suitable porous membrane. Porous membranes should comprise an appropriate number and size of pores to allow cell-to-cell contact across and through the membrane. The membranes should contain a high density of pores, preferably at least lxlO⁸ pore/cm². The size of pores should not be greater then the size of the nucleus or mitotic spindle, so the genetic material of one cell cannot be transferred across the membrane to the other cell during/after fusion. A pore size between l-4μm could be used. The membrane should be strong, suitable for staining, fixing, sustaining electrical pulses and other procedures of the invention. Both sides of the membrane should be tissue culture treated and be suitable for cell growth. The thickness of membrane should be such as to allow cell-to-cell contact through the pores.

Materials and Methods

A suspension of human EC cell line 2102Ep.4D3 (Andrews et al . 1982. Int. J. Cancer 29: 523-531) at 2xl0⁷ cells.ml^"1 was prepared by trypsinisation followed by resuspension in serum containing medium (DMEM (Gibco) , 10% FCS (Gibco) and 2mM L-glutamine) . The cells were centrifuged and resuspended in serum free medium three times before being resuspended in 5ml of serum-free medium. 20μM SNARF-1 (a red fluorescent dye, Molecular Probes) dissolved in pluronic buffer was added to the cells, which were mixed well and incubated at 37°C for 8-10min. Following incubation, the cells were centrifuged and resuspended three times in serum-free medium before being counted. Cup-shaped inserts having a base formed of PET lμm track- etched porous filter (Falcon Cat. No.3102) were inverted aseptically into a large Petri dish and 500μl of cells added to a final density of 2xl0⁵ cm^"1. The Petri dish containing the inserts and cells was then incubated overnight .

The following day the medium was removed and the insert comprising porous filter inverted into a well of a 6 -well plate. Complete medium was then added into the insert and between the insert and the surrounding well walls.

A second population of 2102Ep.4D3 was prepared in an identical fashion except that the green fluorescent dye CMDFA (Molecular Probes) replaced SNARF-1. These cells were placed onto the bottom of the insert and the 6-well plates containing the inserts were placed in a cell culture incubator and incubated at 37°C until the cells attach to the porous filter.

After 2-3 days the cells were ready for electrofusion. It should be noted however that different cell types will have different growth kinetics and therefore these conditions will vary from cell partners to cell partners. The inserts were placed inside the EasyjecT Optima electroporator (EquiBio Limited) . Complete medium was added to the base electrode (3ml) and the cell-laden inserts were added. Medium was added into the insert (1ml) and the upper electrode added. The cells were pulsed with a range of voltages from 100V-500V but more specifically 260V with a range of capacitances from 400μF to 900μF but specifically a capacitance of 750μF and a range of shunt resistances but specifically an infinite shunt resistance. The medium was immediately removed and the inserts placed in a six-well plate with fresh medium and returned to the incubator.

Plating and growing cells on both sides of the porous membrane Cells should be plated on both sides of the membrane to allow maximum contact of cells across the membrane. Normally the plating density should be in the range of 1- 10xl0⁴ cells/cm² to allow for cell growth.

Results

Plating density of cells (a) High plating density- Fig. 2A illustrates an example of a suitable plating density of cells on a porous membrane. The PCC4 cells were plated onto 0.4 μm pore size 6 -well format cell culture insert (Falcon) at a density of 5xl0⁴ cells/cm² in complete DMEM and left overnight to attach. The picture was taken using a digital camera attached to a transmission light microscope. This density will allow reasonably high cell- to-cell contact across or through the porous membrane, as cells plated at this density provided a monolayer covering almost the entire surface of the membrane after 24 hours growth at 37°C humidified incubator supplied with 5% C0₂.

(b) Low plating density

Fig. 2B illustrates an example of an unsuitable plating density of cells on porous membrane. The PCC4 cells were plated onto 0.4 μm pore size 6-well format cell culture insert (Falcon) at a density 0.5xl0⁴ cells/cm² in complete DMEM and left overnight to attach. The picture was taken using a digital camera under transmission light microscopy. This cell density may be too low to allow optimal cell-to-cell contact across or through the membrane as most of the membrane remained uncovered by attached cells. Thickness of the porous membrane

To allow production of contiguous cell membrane/cytoplasm bridges between two layers of cells plated on opposite sides of the porous membrane and allow sufficient flow of cytoplasmic and nuclear components between them during/after electrofusion, the membrane should to be chosen in such a way that cell-to-cell contact through pores is allowed. The thickness of the membrane is one consideration to exercise the method. Preferably the porous membrane should be in the range of 5-100 μm in thickness and more preferably in the range of 5-25 μm thick, dependent on the cell types to be used in fusion.

Fig. 3A is a schematic diagram representing two layers of cells plated on each side of a porous membrane (PM) . Fig.

3A is a schematic representation of the data presented in

Fig. 3B, 3C and 3D. Using an automated microscope, z- sections of lμm thickness were taken across the layers of cells and the porous membrane to assess the thickness of porous membrane and the spacing between the top layer (TL) and bottom layer of cells (BL) growing on it. The arrow shows the direction in which the z-sections were taken.

The distance between the cell nuclei on each side of the porous membrane and the middle of the porous membrane is indicated to the right of the schematic diagram. The distance shown is sufficient to allow cell-to-cell contact across the membrane.

Results

In the example shown in Fig. 3B-D, PCC4 cells were plated on both sides of a Falcon cell culture insert porous filter (Falcon Cat. No.3092), high density porous membrane containing pores of size 3μm. The porous membrane was excised from the support cup of the 6-well format cell culture insert (Falcon) , fixed in 3% paraformaldehyde in PBS and mounted with Vectashield mounting medium containing 1.5μg/ml DAPI (Vector Laboratories). Using an automated fluorescence microscope, lμm z sections across layers of cells and the porous membrane were taken. Fig. 3B shows a DAPI staining of nuclei of cells plated on the top (TL in Fig. 3A) of the porous membrane. The picture is a composition of 3 sections each of lμm thickness.

Fig. 3C is a composition of 3 z sections of lμm thick each and shows the middle section of the porous membrane (PM in Fig. 3A) . The distance of this section to the nuclei of the top layer of cells was estimated to be about 6 to lOμm.

Fig. 3D shows a DAPI staining of nuclei of PCC4 cells plated onto the bottom side (BL in Fig. 3A) of the same porous membrane as in Fig. 3B. The picture is a composition of 3 sections of lμm thick each.

2) Methods of fusion for admixing of cellular components

Optimisation of electrical parameters for electrofusion.

An aim of the invention is to fuse two cells growing on opposite faces of the same porous membrane. We have determined that electrofusion is one way to effect fusion of cells on the porous membrane. It would be advantageous to determine the parameters for electrofusion yielding maximal cell survival and maximal fusion of cells. The method of electrofusion may be varied in several ways including: the types of cells to be fused, the medium in which electrofusion takes place, voltage, number and duration of electrical pulses. Optimisation of these parameters can increase the efficiency of the invention. Other mechanisms for fusing cells across a porous membrane

The method of the invention allows for processes other than electrofusion to bring about cell fusion across the membrane, for example application of detergents (e.g. saponin, digitonin) or pore forming substances (e.g. streptolysin 0) or polymers (e.g. polyethylene glycol).

Permeabilisation as a method of delivery to target somatic cells

Materials and Methods

Permeabilisation with Saponin

Adherent cultures of subconfluent Chinese Hamster Ovary (CHO) EM9 cells were washed free of growth medium with PBS. The cultures were treated with 0-50 μg/mL Saponin, a non-ionic detergent, in physiological buffer (PB; 100 mM potassium acetate, 30 mM KCl, 10 mM Na₂HP0₄, 1 mM MgCl₂, 1 mM disodium ATP, 1 mM DTT, 0.2 mM PMSF, pH 7.4). The cells were treated by applying the saponin for 1-2 minutes, immediately removing the detergent solution, washing twice with PBS, and twice with αMEM, and finally applying growth medium.

Cell permeability was monitored immediately after saponin treatment and again 24 hours later. A 2% solution of Trypan Blue was mixed with an equal volume of PBS. Medium was washed from the cell cultures and the 1% Trypan Blue solution was applied. Cells were viewed immediately by light microscopy.

Permeabilisation with Streptolysin O

Streptolysin 0 (SLO) is a bacterial toxin purified from Streptococcus pyogenes, that permeabilises the outer cellular membrane and permits uptake of large or charged molecules, including proteins into the cell cytoplasm. The pores formed can be resealed by addition of FCS or calcium to the incubation media. CHO EM9 were washed with PBS and the cells were permeabilised with 5 to 20 units/10⁶ cells of activated streptolysin 0 in serum-free medium for 10 minutes at 37°C. To reseal plasma membranes, 10% serum containing media was added and cells were incubated for a further 30 minutes at 37°C and 5% C0₂.

Using this procedure some cells were permeabilised reversibly, some cells were irreversibly permeabilized (i.e. killed), while others remained unpermeabilized. Illustrative examples are detailed below. To assess the permeabilisation efficiency cells were stained with lOμM fluorescein diacetate (FDA) , a marker for viable cells, which was added during incubation in serum free medium, and lOμg/ml of propidium iodide (PI) , a marker for dead cells, which was added after plasma membranes had been resealed. Cells were analysed by flow cytometry: green only cells were permeabilised and resealed, red only or red and green cells were dead, while colourless cells were non permeabilised.

Permeablisation with Digitonin

The outer cellular membrane of target cells can also be permeabilised using digitonin. Cells were washed with PBS and released from the growing surface using Trypsin-EDTA then centrifuged to pellet. The cell pellet was washed in KHM buffer [110 mM KOAc, 2 mM MgOAC, 20 mM HEPES (pH 7.2)] The pellet was resuspended in ice-cold KHM buffer to which digitonin was added to a final concentration of 40 μg/ml and incubated on ice for 5 minutes. Enough ice-cold KHM buffer was then added to double the volume and the whole centrifuged to pellet. The liquid was removed and the cells were resuspended in ice-cold HEPES buffer [KOAc 50 mM, HEPES 90 mM (pH 7.2)] then placed on ice for 10 minutes. Permeabilisation of the cells can be monitored by staining with trypan blue as described (above) . Optimisation for cell survival

Materials and methods

Cells growing adherently or in suspension were collected by centrifugation, washed three times in PBS, and resuspended as a single cell suspension, in 0.3M mannitol buffer (0.3M mannitol, 0. ImM MgS0₄, 50μM CaC12 and 3% BSA; pH 7.2-7.4) at a density greater then lxl0⁷cell/ml . Cells were treated with voltages in the range of 0-300 Volts (V) , 0-2 pulses, and pulse lengths in the range of 0-100 μseconds (μS) with an Electro Cell Manipulator 830 (BTX) carrying a 1-mm electrode gap. After treatment, cells were replaced into tissue culture in complete medium, allowed to recover for 24h and cell survival was assessed. Cell number was calculated for suspension cultures (number xl0^s/ml) . Adherent cultures were washed once with PBS to remove non-adherent cells, cells were fixed in situ in 70% ethanol, stained with 1% methylene blue in ethanol, and cell survival was estimated visually. Methylene blue stains cells that remain adherent to the vessel surface. Only cells surviving the treatments would remain adherent while cells killed by treatment would lift off the growing surface and be washed away.

Results

Effect of one pulse on cell survival

Fig. 4A shows surviving cell number (xl0⁶/ml) of CEM cells treated with one electric pulse of voltage in the range of 0-300 volts for 50 μs over a 1mm electrode gap. Cell survival was measured 24h after treatment. An equal number of cells at an equal cell density were treated at each voltage. Cell survival at each treatment voltage can be readily compared to the control over which no voltage was applied (DC-) . Cell survival at 100 and 150 volts was equal to that when no voltage was applied. Voltages above 150 volts (200, 250, and 300 volts) produced a marked decrease in cell survival, however, greater than 25% of cells survived all voltages.

Effect of two pulses on cell survival Fig. 4B shows survival of CEM cells 24h after applying two equal electric pulses. Voltages were applied in the range of 0-300 volts over a 1 mm electrode gap, each pulse lasting 50 μs, to cultures of equal starting cell number at identical cell density. Cell survival for each treatment can readily be compared to the control culture in which no voltage was applied (DC-) . Cell survival similar to control was observed in cultures treated with two pulses of 100, 150, and 200 volts. In contrast, cell survival was markedly decreased in cultures treated with two pulses at 250 or 300 volts. However, greater than 25% of cells survived all voltages.

Effect of one pulse in the range of 0-300 V on cell survival Fig. 5A shows methylene blue staining (dark areas in each of a-f) of adherent PCC4 EC cell cultures 24h after applying one electric pulse lasting 50μs, voltage in the range of 0V-300V, over a 1mm electrode gap to an equal number of cells at identical cell density. The figure illustrates cells treated with (a) 0 volts (no electric pulse) , (b) 100V, (c) 150V, (d) 200V, (e) 250V, (f) 300V. Cell survival was assessed visually by the extent of methylene blue staining remaining in each well (dark areas in each of a-f) as cells that don't survive treatment lose adherence to the growing surface and so would not be present for staining by methylene blue. It is clear that compared to the control (0 volts) , treatment of these cells with 100 and 150 volts had little detrimental effect on cell survival whereas treatment with 200, 250, and 300 volts had a significant negative effect on cell survival. However, some cells survived all treatments. Effect of two pulses in the range of 0-300 V on cell survival

Fig. 5B shows methylene blue staining of adherent PCC4 EC cell cultures 24h after applying two identical electric pulses lasting 50μs, voltage in the range of 0V-300V, over a 1mm electrode gap to an equal number of cells at identical cell density. The figure illustrates cells treated with (a) 0 volts (no electric pulse) , (b) 100V, (c) 150V, (d) 200V, (e) 250V, (f) 300V. Cell survival was assessed visually by the extent of methylene blue staining remaining in each well (dark areas in each of a-f) .Cells that don' t survive the treatment lose adherence to the growing surface and so would not be present for staining by methylene blue. It is clear that compared to the control (0 volts) , treatment of these cells with 100 and 150 volts had little detrimental effect on cell survival whereas treatment with 200, 250, and 300 volts had a significant negative effect on cell survival. Treatment with two pulses of 300 volts appears to be 100% lethal to these cells since there were no surviving cells visible.

Effect of one pulse in the range of 0-50 μS on cell survival Fig. 6A shows cell survival, measured by methylene blue staining, of PCC4 cells 24h after application of one electric pulse of 200V, over a 1mm electrode gap, on the same number of cells at identical cell density, with electric pulse length in the range of 0-50μS. The examples illustrate cell survival after pulses of (a) 0 μS (control), (b) lOμs, (c) 20μs, (d) 30μs, (e) 40μs, (f) 50μs applied as indicated. Compared with the control culture in (a) , to which no pulse was applied, pulses of 10 μS-50 μS yielded similar numbers of surviving cells. Effect of one pulse in the range of 60-100 μS on cell survival

Fig. 6B shows cell survival, measured by methylene staining, of PCC4 cells 24 h after application to equal cell number at equal cell densities of one electric pulse of 200V over an electrode gap of 1mm with electric pulse length of (a) 60μs, (b) 70μs, (c) 80μs, (d) 90μs, (e) lOOμs, (f) 0 μs (control) . Compared with (f) in which no pulse was applied, pulse lengths of 70-100 μs yielded a similar number of surviving cells.

Effect of two pulses in the range of 0-50 μS on cell survival .

Fig. 6C shows PCC4 cell survival, measured by methylene blue staining, 24h after application of two identical electric pulses of 250V across a 1 mm electrode gap with duration of electric pulse in the range of 0-50 μS, to equal numbers and densities of cells. Comparing (a) 0 μS (control), with (b) lOμs, (c) 20μs, (d) 30μs, (e) 40μs, (f) 50μs pulses, it is evident that two pulses of any duration in the range specified resulted in a visible decrease in cell survival and furthermore that pulses of 40 and 50 μs yielded almost complete lethality in these cells .

Effect of two pulses in the range of 60-100 μS on cell survival

Fig. 6D shows PCC4 cell survival, measured by methylene blue staining, 24h after application of two identical electric pulses of 250V across a 1 mm electrode gap with duration of electric pulse in the range of 0-50 μS, to equal numbers and densities of cells. Comparing (f) Oμs (control), with (a) 60 μS, (b) 70μs, (c) 80μs, (d) 90μs, (e) lOOμs, pulses, it is evident that two pulses of any duration in the range specified resulted in a visible decrease in cell survival such that no surviving cell were visible.

Optimisation for cell fusion

Effect of Electrofusion Buffer

Materials and methods .

Fusion using electric pulses was performed between adherent cultures of mouse Embryonal Carcinoma (EC) cell line, PCC4 , and suspension cultures of human T-cell line, CEM. Both cultures were grown in appropriate complete medium (DMEM or RPMI, respectively) supplemented with 10% foetal calf serum (FCS) and 2mM L-glutamine in an humidified environment at 37°C and 5% C0₂, until confluent. Cells were then collected by trypsinisation or centrifugation, respectively, and incubated for 30 minutes at 37°C in serum free medium containing 0.5μM-lμM Cell Tracker dyes (Molecular Probes) . The CMFDA Cell Tracker Green and CMTMR Cell Tracker Red were used, to specifically label PCC4 and CEM cells, respectively. After staining, cells were washed three times in complete medium and incubated further for 30-120 minutes in the same conditions. Stained cells were admixed in complete medium or in 0.3M mannitol buffer (see materials and methods 2A) in 1 : 1 ratio at density greater then lxlO⁷ cells/ml. A single pulse of 180 volts, 300μF capacitance, was applied using EasyjecT Plus electroporator apparatus (Equibio) .

Double staining cells in the absence of electrofusion (control)

Fig. 7A shows a representative dot plot of flow cytometric analysis of a population (10⁴ cells) of admixed CEM and PCC4 cells in CDMEM buffer and to which no electrical pulse has been applied (control) . Two distinct populations of singly stained cells were observed: CMFDA-stained PCC4 cells visible in the bottom right quadrant of the plot and CMTMR-stained CEM cells are visible in the top left hand quadrant of the plot. The minor population visible in the top right quadrant displays double (red-plus-green) fluorescence. This population most likely represents singly stained cells adhering to cells stained with the other dye (e.g. an individual PCC4 and individual CEM cell adhered to each other but not fused through the cell membranes, would be both red and green staining). The data of Fig. 7A allows the percentage of background red- and-green doubly stained entities in the population in the absence of electrofusion to be established. This population represents 0.86% of the total number of cells used in the experiment .

Double staining cells after electrofusion in CDMEM medium.

Fig. 7B shows a representative dot plot of flow cytometric analysis of cells fused in 10% CDMEM. In addition to the two distinct populations of CMTMR- and CMFDA- singly stained cells, the presence of a third population of cells, which displays double fluorescence was observed in the top right quadrant. This population represents 1.74% of the total number of cells used in the experiment. The percentage of double-staining cells is higher than the background seen in the previous figure (Fig. 7A) and we conclude that approximately 0.88% of double red-plus- green-stained cells resulted from fusion between CMTMR and CMFDA stained cells.

Double staining cells after electrofusion in mannitol buffer.

Fig. 7C shows a representative dot plot of the flow cytometric analysis of cells fused in 0.3M mannitol. The two populations of singly stained with CMTMR and CMFDA dyes cells are visible in the bottom right and in the top left quadrant. There is also a third small population of cells placed in the top right quadrant, which display double fluorescence. This population represents only 0.66% of total number of cells used in experiment. This percentage is similar to the percentage of doubly stained cells observed in mock control (Fig. 7A) . The result suggests that, under these experimental conditions mannitol buffer is not as efficient a buffer for cell fusion as is DMEM.

Effect of voltage Materials and methods

Fusion using electric pulses was performed between adherent cultures of mouse Embryonal Carcinoma (EC) cell line, PCC4, and suspension cultures of human T-cell line, CEM. Both cultures were grown and stained with Cell Tracker dyes (Molecular Probes) as described for the "Effect of Electrofusion Buffer" experiment.

Cells (PCC4 and CEM) were admixed in 0.3M mannitol buffer (see materials and methods 2A) in 1:1 ratio at density greater than lxlO⁷ cells/ml. Cells were treated with voltage in the range of 0-400 volts, 0-2 pulses, and pulse lengths in the range of 0-100 μseconds with an ElectroCell Manipulator 830 carrying a 1-mm electrode gap (BTX) . The samples were then prepared for analysis by flow cytometry.

Double staining cells in the absence of electrofusion

(control)

Fig. 8A is a representative dot plot of flow cytometric analysis of stained cells (10⁶) admixed in 0.3M mannitol fusion medium without application of electric pulse

(control conditions) . There are two main populations visible in the image: CEM cells stained with CMTMR are visible as a dense pattern in the top left quadrant of the figure while PCC4 cells stained with CMFDA are visible as a dense pattern in the bottom right quadrant of the figure. A minor population of cells occupying the top right quadrant represent a doubly stained (CMFDA and

CMTMR) population. These represent singly stained cells adhering to a cell stained with the other dye (eg. a PCC4 adhered, but not fused, to a CEM cell would generate a red-plus-green profile of staining) . Fig. 8A illustrates the percentage of background red-and-green doubly stained entities in the population in the absence of electrofusion. This population represents 0.37% of the total number of cells used in the experiment.

Double staining after electrofusion at 100 V Fig. 8B is a dot plot of flow cytometric analysis of cells subjected to one electrical pulse of 100V lasting 50 μseconds. A distinct population of doubly red-and-green stained cells is visible in the top right quadrant of the plot. This population represents 1.12% of the total number of cells used in the experiment. Since the percentage of doubly red-and-green staining cells is higher than in the control example (Fig. 8A) , we conclude that some of the double-stained cells resulted from electrofusion of cells stained individually red or green.

Double staining after electrofusion at 400 V

Fig. 8C is a dot plot of the flow cytometric analysis of cells subjected to one electrical pulse of 400V lasting 50 μseconds. The doubly stained population of cells, visible in the top right quadrant of the plot represents 0.77% of the total number of cells used in the experiment. Since the percentage of doubly red-and-green staining cells is higher than in the control (Fig. 8A) example we conclude that some of the double-stained cells resulted from electrofusion of cells stained individually red or green.

Admixing of cellular components by electrofusion Prior to establishing that cells can be grown and fused across the porous membrane, we established that cells could be fused in the absence of the porous membrane. In order to detect fusion of two cells we stained two different cell populations each with a dye emitting fluorescence in the red or green spectrum when illuminated by ultraviolet light. Successful fusion of cells from each differentially stained population would result in red and green dyes being contiguous within the fused hybrid cell. Merging the individual red and green fluorescent images of the hybrid cell will yield a third image in which the fused hybrid appears yellow.

Material and methods

Fusion using electrical pulses was performed between adherent cultures of the mouse Embryonal Carcinoma (EC) cell line, PCC4, and suspension cultures of a human T- lymphocytic cell line, CEM. Both cultures were grown in appropriate complete medium (DMEM or RPMI respectively) supplemented with 10% foetal calf serum (FCS) and 2mM L- glutamine in a humidified incubator at 37°C and 5% C0₂ until confluent. Cells were then collected by trypsinisation or centrifugation, respectively and incubated for 30 minutes at 37 °C in serum free medium containing 0.5μM to lμM final concentration of Cell Tracker dyes (Molecular Probes) . The CMFDA Cell Tracker Green and CMTMR Cell Tracker Red were used, for PCC4 and CEM, respectively. Cells were washed 3 times in complete medium and incubated further for 30-120 minutes at the same conditions .

Stained cells were admixed in complete medium in a 1:1 ratio at density greater then lxlO⁷ cells/ml and 180V, 300μF electric pulse was applied using EasyjecT Plus electroporator (Equibio) . Cells were then transferred into complete DMEM medium, collected by centrifugation, fixed with 3% paraformaldehyde and mounted with mounting medium containing 1.5μg/ml DAPI, that stains the DNA specifically (Vectashield, Vector Labotatories) and viewed by fluorescence microscopy.

Results DAPI staining

Fig. 9A shows DAPI staining of the chromosomes in the nuclei of two individual cells. DAPI is a non-cell specific dye that binds to chromosomal DNA and was used here in order that all cells would be visible at this wavelength of fluorescent illumination.

CMFDA staining

Fig. 9B shows a cell that displays fluorescent emission at 520nm. That this cell is stained with CMFDA Cell Tracker Green indicates that it consists of at least a PCC4 cell. The second cell, visible in Figure 9A by DAPI staining, is invisible at this emission since it has not been stained with CMFDA and can be assumed to be a CEM cell.

CMTMR staining

Fig. 9C shows two cells that display fluorescent emission at 650nm indicating that both cells are stained with CMTMR Cell Tracker Red. These are the same two cells visible by DAPI staining in Fig. 9A and the top-most cell is visible also in Fig. 9B due to its staining with CMFDA. Thus the top-most cell is stained with both red and green tracker dyes. The bottom-most cell is stained only with CMTMR and can thus be identified unequivocally as a CEM cell. Since the top-most cell is emitting light in red and green fluorescence, it must contain components from both CEM and PCC4 cells and can therefore be identified as a fused hybrid of both types of cell.

Merged image of CMFDA and CMTMR staining Fig. 9D shows the two cells visible in Figs 9A and 9C. This is a merged image of the two individual images captured at 520nm (Fig. 9B) and 650nm (Fig. 9C) . The merged image confirms that this cell is a product of fusion between a PCC4 cell stained with CMFDA and a CEM cell stained with CMTMR, and in the original colour image, this cell is yellow. The bottom-most cell displays fluorescence at 650nm only and this indicates that this cell has not undergone fusion and is a single CEM cell.

Fusion across the porous membrane and admixing of cellular components

Materials and Methods

Two populations of EC cell line 2102Ep.4D3 were treated as described for the "Effect of Electrofusion Buffer" experiment except that the population loaded onto the bottom of the insert were stained with CMFDA and the cell population loaded into the. well of the insert was left unstained. The inserts were placed inside the EasyjecT Optima electroporator (EquiBio Limited) . Complete medium was added to the base electrode (3ml) and the cell-laden inserts were added. Medium was added into the insert (lml) and the upper electrode added. The cells were pulsed with a range of voltages from 100V-500V but more specifically 260V with a range of capacitances from 400μF to 900μF but specifically a capacitance of 750μF and a range of shunt resistances but specifically an infinite shunt resistance. The medium was immediately removed and the inserts placed in a six-well plate with fresh medium and returned to the incubator. After 2-3 days incubation at 37°C, the inserts were washed in PBS and fixed in 2% paraformaldehyde in PBS for 30 minutes before washing in PBS again. The porous filter was then removed from the insert and mounted on a glass microscope slide with an anti-fade solution and coverslipped. The efficiency of fusion was determined by measuring the degree of dye transfer across the porous filter as determined with a confocal microscope.

Results Control image

Fig. 10 shows XZ reconstructions from a stack of images captured using a confocal microscope and ultraviolet illumination. The images demonstrate cells present on two sides of the porous membrane after electrofusion as described above. The cells adhering to the bottom surface of the porous membrane were previously stained with CMFDA whereas the cells adhering to the top surface of the porous membrane were left unstained. Unlabeled cells seeded on to the upper membrane surface are not visible in this picture. No electrical current was applied to these cells. There is not evidence of transfer of the CMFDA dye from the lower population of cells to the upper population of cells. In the figure, the following features are indicated: 1. Upper layer of cells

2. Porous membrane

3. Lower layer of cells

Fusion across the porous membrane and admixing of cellular components

Unlabeled cells were seeded on to the upper membrane surface so are not visible in the image shown in Fig. 10B. Cells labelled with cell tracker green (CMFDA) were seeded onto the lower membrane surface. An electrical field of 260V was applied to these cells and fusion was achieved. Fusion of cells across/through the porous membrane and admixing of cellular components was confirmed by the presence of CMFDA in a group of cells on the upper side of the membrane. In the figure, the following features are indicated:

4. Upper layer 5. Porous membrane 6. Lower layer of cells

3) Removal of fused cells from the porous membrane Materials and methods

Once cell fusion across the membrane has occurred and reprogramming has been achieved, the target cells can be removed from the porous membrane. There are three possible conditions. Where it is desirable to retain both parent cells, both sides of the membrane can be trypsinised/removed at once. If trypsin is only added to the inside of the insert then only those cells will be removed. Thirdly, if trypsin is only added to the six-well plate then the cells on the outside of the insert will only be removed .

(a) Removal of both parent cells from the porous membrane

As shown in Fig. 11A, a trypsin solution is applied to both sides of the cell -laden insert and both parent cells become detached from the porous membrane. The fine, fused, pseudopodia between the cells are broken by the mechanical forces as the cells lift off. The separated cells are in suspension but remain apart from the other parent cells by the insert and membrane. A solution of EDTA may also be used to remove the cells or a mixture of trypsin and EDTA. The cells may also be removed by physically scraping the cells off the porous membrane using a cell scraper or rubber policeman. Both cells may be removed and sub- cultured separately. In the situation where the reprogrammed cells are maintained on the porous membrane they can then be co-cultured with a third cell type seeded onto the vacated side of the membrane to assess the ability of the reprogrammed cell to affect the phenotype of this third cell type. Fig. 11A legend: 1. Insert 2. Wall of six-well plate 3. Parent cell on inside of insert 4. Parent cell on outside of insert 5. Detaching solution

(b) Removal of one parent cell from inside the cell-laden insert (upper layer) only

As show in in Fig. 11B, a trypsin solution is applied to the inside of the cell-laden insert only and the cells become detached. These can then be removed by aspiration and sub-cultured. The cells remaining on the insert can be used again for further fusions or they can be processed for immunocytochemistry and examined by microscopy.

(c) Removal of one parent cell from inside the well plate (lower layer) only

A trypsin solution is applied to the inside of the six- well plate on the outside of the cell-laden insert only. The cells become detached and these can then be removed by aspiration and sub-cultured. The cells remaining on the insert can be used again for further fusions or they can be processed for immunocytochemistry and examined by microscopy.

4) Detection of phenotypic change after fusion

Assaying specific gene expression

One method of determining that phenotypic change has taken place after fusion and admixing of cellular components is to assay specifically for the expression of genes that would be expressed if the target cell had been reprogrammed in the expected way. In the specific example of reprogramming of a somatic target cell through fusion with an EC cell (WO 00/49138) , reprogrammed somatic cells ould be expected to express markers of pluripotency . Such markers are characteristically expressed in pluripotent cells such as EC and ES (embryonal stem) cells and include, but are not restricted to, Oct 3/4 and Sox2.

The phenotypic changes to cells produced by the methods described above may include pluripotent properties that closely resemble those of embryonic stem cells, so that the cells may be able to differentiate and initiate differentiation pathways that result in the formation of any cell type that may be found in the adult, embryo or in extra-embryonic tissues, given appropriate culture conditions. The maintenance of an embryonic stem cell state can be monitored by assay of various markers that include the cell surface antigens SSEA3, SSEA4, TRA-1-60, TRA-1-81, by their expression of alkaline phosphatase or by expression of Oct 3/4 (as above) . Alternatively, the reprogrammed cell may, under specific conditions, be encouraged to differentiate and may then express markers of specific differentiation that diverge from the differentiation profile of the original cell. For example, a thymocyte, once reprogrammed, may express markers of endodermal differentiation including but not restricted to, laminin Bl (Chen, A.C. & Gudas , L.J. (1996) "An analysis of retinoic acid-induced gene expression and metabolism in AB1 embryonic stem cells". J. Biol. Chem. 275 (21) -14971-14980.)

Material and Methods

In order to analyse gene expression, RNA was prepared from cells that had potentially been reprogrammed and from target cells that had not been subjected to reprogramming. An aliquot of cells or RNA (not exceeding 1 μl, equivalent to 10⁴-10⁶ cells or 10-1000 ng RNA) was subjected to reverse transcription and PCR amplification as described in (Brady, G. and Iscove,N.N. (1993) . "Construction of cDNA libraries from single cells." Methods Enzymol . 225:611- 23.) using the primer NotldT (5' CAT CTC GAG CGG CCG CTT _{TTT TTT TTT TTT TTT TTT} -pηπrr, _T 3, [g_EQ J_{D NQ}. _]_] ) _to produce polyA cDNA. The polyA cDNA was subjected to "TaqMan" real-time PCR using an ABI Prism 7700 System and Universal Master Mix (Applied Biosystems Inc.) with primers and probes designed using Primer Express (ABI) and according to the manufacturers instructions. Primers and probes were tested for their unique recognition of the desired gene/cDNA sequence using NCBI BLAST analyses.

In order to detect reaction products by TaqMan real-time PCR, probes were modified by inclusion of FAM and TAMRA fluorescent labels. FAM is tagged on the 5' end and TAMRA on the 3' end of the probe. Whilst both are bound to the probe TAMRA quenches the fluorescent signal from FAM. During the PCR reaction, FAM is displaced and cleaved from the probe and, having been displaced, it is no longer quenched by TAMRA. The output of the reaction is a fluorescent signal generated by cleavage of FAM from the probe during the PCR reaction. The amount of FAM cleaved from the probe during PCR is directly proportional to amount of starting template for the gene under investigation.

The standard TaqMan reaction is 40 cycles as indicated in the manufacturer's instructions. The first cycle at which

FAM can be detected is called the "threshold" cycle (Ct) for the gene under investigation. Since there is no template to amplify and thus no FAM can be cleaved from the probe, when no polyA cDNA was included in the reaction mix, the Ct value was 40. A reaction that produced no FAM signal upon completion of 40 cycles would be equivalent to a product in which no template for PCR had been included.

A Ct value less than 40 indicates that the primers/probe recognised a target template and that FAM was cleaved as the PCR reaction displaced it from the template. The primers and probes used for detection of murine and human Oct 3/4, Sox2, GAPDH, and laminin Bl are detailed below:

PCR primers for human and mouse genes

[SEQ ID NO: 19]

Forward: AACTCGGCCCCCAACACT [SEQ ID NO: 20] Reverse: CCTAGGCCCCTCCTGTTATTATG [SEQ ID NO: 21]

GAPDH Probe: CATCTCCCTCACAATTTCCATCCCAGAC [SEQ ID NO: 22]

Forward : GGCTCGGTGACCAAGGTAAA [SEQ ID NO: 23]

Reverse : TCCATACAAAAGTAGGTGGTTAAAAACA [SEQ ID NO:

Laminin

24] Bl

Probe : ACCGAGGCAGTCATCTACAAATAACCCATCA [SEQ ID

NO : 25 ]

General Discussion

In order to exercise the method of the invention it is advantageous to determine the optimal electric parameters in order to obtain maximal cell survival and fusion of cells across the membrane. The efficiency of electrofusion depends on several parameters including: the types of cells to be fused, medium in which electrofusion takes place, voltage, number and duration of electrical pulses. In the specific embodiments presented above, the effect of increased voltage on cell survival and fusion was measured. It appeared that voltage between 100 and 400V is suitable for cell fusion, however increased voltage values result in decreasing cell survival . Additionally it has been observed that increased number of electric pulses and their duration could cause extensive cell death. It appears that pulse duration of between 0 and 100 μseconds was the most suitable for sustaining cell viability. No more then one or two pulses were applied in these experiments .

The type of electrofusion medium used is also a factor to consider in order to optimise cell fusion. In the illustrated examples two different electrofusion media were tested: Dulbecco's modification of Eagles medium

(DMEM) supplemented with 10% foetal bovine serum and 0.3M mannitol buffer. It appeared in these examples that complete medium is more suitable as a fusion medium, and it could provide protection for cells. This medium could also allow for better recovery of cells after fusion.

However, electrofusion in 0.3 M mannitol yielded fusion products at high initial cell number (Figs 7 and 8) and may be a suitable alternative to DMEM.

In addition, low conductivity buffers including sucrose or glucose containing media, phosphate buffers, and other cell media may be used for electrofusion where appropriate (Sukharev, S.I. et al . (1990) "Electrofusion of fibroblasts on the porous membrane" Biochimica et Biophysica Acta 1034: 125-131.).

Cells are fused in order to effect admixing of cellular components. We have demonstrated in the absence (Fig. 9) and in the presence (Fig. 10) of the porous membrane, that admixing of cell components does occur after fusion.

Claims

1. A method of producing a fused cell, comprising the steps of : (i) providing a porous filter,*

(ii) allowing a first parent cell to attach to one side of the porous filter and a second parent cell to attach to the other side of the porous filter; and

2. The method according to claim 1, wherein the nuclei of the parent cells remain separated by the porous filter.

3. The method according to either of claim 1 or claim 2, wherein the method includes providing one or more additional porous filters and providing additional parent cells to allow fusion of the cell cytoplasms of further parent cells whilst the chromosomes or nuclei of the parent cells remain separated by the porous filter.

4. The method according to any of claim 1 to 3 , wherein chemotactic agents are used to encourage formation of pseudopodia from the cell membranes through the pores of the porous filter, to encourage fusion of the parent cells .

5. The method according to any preceding claim, further comprising the step of culturing the fused cell under suitable conditions to enable the parent cells which have been fused to remain viable whilst keeping the chromosomes or nuclei of the parent cells apart.

6. The method according to any preceding claim, wherein one cell to be fused is a reprogramming cell and another is a target cell to be reprogrammed.

7. The method according to claim 6, further comprising culturing the fused cell under suitable conditions to maintain the fused cell for a period of time to allow the reprogramming cell to produce factors to reprogram the target cell nucleus.

8. The method according to any preceding claim, wherein fusion of the parent cells is induced by an electrical field.

9. The method according to claim 8, wherein an electrical pulse with a voltage of 100V-500V, for example 260V, is used.

10. The method according to either of claim 8 or claim 9, wherein an electrical pulse with a capacitance of 400μF -

900μF, for example 750μF, is used.

11. The method according to any preceding claim, wherein fusion of the parent cells is induced by a fusogenic agent, for example a detergent, microbial toxin and/or polymer.

12. The method according to any preceding claim, wherein the porous filter has pores with a pore size of about 0.2- 10 μm, preferably about 0.4-4 μm, for example about 1 μm.

13. The method according to any preceding claim, wherein the porous filter has at least lxlO⁸ pores cm^"2.

14. The method according to any preceding claim, wherein the porous filter is 5-100 μm thick, preferably 5-25 μm thick.

15. The method according to any preceding claim, further comprising the step of causing separation of the fused cell into a first parent-derived cell and a second parent- derived cell, each of which has maintained the integrity of their nuclei and/or the segregation of their chromosomes .

16. The method according to claim 15, comprising the step of isolating the first parent-derived cell and/or the second parent-derived cell.

17. The method according to claim 16, comprising the step of isolating a reprogrammed parent-derived cell.

18. A composition or pharmaceutical composition comprising or consisting of a parent-derived cell produced according to claim 153 or isolated according to claim 16 or 17.

19. A composition or pharmaceutical composition according to claim 18 further comprising a pharmaceutically acceptable carrier, excipient or diluent.

20. A tissue comprising parent-derived cells produced according to claim 15 or isolated according to claim 16 or claim 17.

21. The tissue of claim 20 in which the tissue is selected from neural, smooth muscle, striated muscle, cardiac muscle, bone, cartilage, liver, kidney, respiratory epithelium, haematopoietic cells, spleen, skin, stomach, pancreas and intestine.

22. The tissue for use in transplantation containing at least one parent-derived cell produced according to claim 15 or isolated according to claim 16 or claim 17.

23. A fused cell produced according to the method of claims 1 to 17.