WO2018150188A1

WO2018150188A1 - Screening method

Info

Publication number: WO2018150188A1
Application number: PCT/GB2018/050411
Authority: WO
Inventors: David Carter; Susan Brooks; Ellie BEAMAN
Original assignee: Oxford Brookes University
Priority date: 2017-02-17
Filing date: 2018-02-16
Publication date: 2018-08-23
Also published as: GB201702582D0

Abstract

The present invention relates to a method of screening for a candidate agent for use in either (i) modulating the transition of a population of substantially epithelial cells to a population comprising a mixture of epithelial cells and mesenchymal cells, or for use in (ii) modulating the transition of a population of substantially mesenchymal cells to a population comprising a mixture of mesenchymal cells and epithelial cells. To be accompanied, when published, by Figure6of the drawings.

Description

SCREENING METHOD

The present invention relates to a method for identifying candidate agents for use in modulating the transition of cells from one cell state to another cell state, in particular for use in modulating the transition of cells from an epithelial state to a mesenchymal state or vice versa, or from an HPA positive to an HPA negative state or vice versa. The candidate agent may be for use in preventing cancer metastasis. The invention may also provide a method for identifying genes, proteins or other biological molecules involved in such cell state transitions.

The ability of cells to transition between different states is believed to be important in cancer metastasis. Cancer metastasis is a major cause of death for cancer patients. It is a complex and multi-step process in which a cell leaves a primary tumour site, moves through the body to a secondary site where it forms a secondary tumour in another tissue or organ discrete from the primary cancer. Cancer cells are known to be able to transition between different states. In each state the cells express different genes and thus have different functional capabilities. An example of a cell state transition is the epithelial-mesenchymal transition (EMT), this is a process by which polarised epithelial cells that are connected to an underlying tissue matrix via adhesion lose this functional capability and acquire migratory and invasive properties characteristic of a mesenchymal cell and vice versa (the so called mesenchymal to epithelial transition MET) . It is believed that this ability to switch between these two states is important for an epithelial cell to metastasise. The pathway proposed involves cancerous epithelial cells which adhere to a local tissue matrix, transitioning to another cell type, more mesenchymal in nature, losing adherence from the primary site, gaining the ability to migrate through the primary site, move into the blood stream and migrate to a new site, once at a new site the cells (re)gain abilities to adhere to blood vessel cell walls, move through the cell wall, through the underlying tissue matrix and (re)gain the ability to bind to the underlying tissue and thus form a new secondary cancer.

An aim of the present invention is provide a screen for effective agents for use in modulating the transition of cells from one state to another, in particular for use in modulating the transition of cells from an epithelial state to a mesenchymal state or vice versa, or from an HPA positive to an HPA negative state or vice versa. The candidate agent may be for use in preventing cancer metastasis. Agents identified in the screen may be used to prevent cancer metastasis. Modulation of cellular transition may include reducing, preventing or enhancing the transition of cells from one state to another. Agents able to modulate this transition may be useful for therapeutic treatments including, without limitation, prevention of metastasis, control of tissue regeneration, alteration of tissues and organ development.

The present invention provides a method of screening for a candidate agent for use in modulating the transition of a population of cells from a first state to a second state, wherein the method comprises the steps of:

a) providing a population of cells;

b) contacting the candidate agent with the population of cells provided in a); c) screening the population of cells after contact with the candidate agent to see if the state of cells in the population has changed.

In step c) the change being screened for may be selected from:

• an increase in cells in the population in a first or second state compared to the population in a);

• a decrease in cells in the population in a first or second state compared to the population in a); and

• no change in the state of cells in the population compared to the population in a).

The cells in the first state may be in a mesenchymal state, and the cells in the second state may be in an epithelial state, or vice versa.

The cells in the first state may be HPA positive, and the cells in the second state may be HPA negative, or vice versa.

The candidate agent may be a therapeutic or prophylactic agent.

The method of the invention may also be used to screen for genes or proteins, or other agents, involved in the control of cell state transition.

The population of cells may be a population of cance In step a) the population may comprise cells which are substantially all in the same state .

In a further aspect the invention may provide a method of screening for an agent for use in modulating the ratio of cells in a first state in a population of cells to cells in a second state in the population of cells, wherein the population of cells comprises a mixture of cells some in the first state and some in the second state and wherein the method comprises the steps of:

a) providing a population of cells with a known ratio of cells in a first state to cells in a second state;

b) contacting the population of cells provided in a) with the agent;

c) screening the population of cells after contact with the agent to see if the ratio of cells in a first state to cells in a second state changes. The cells in the first state may be in a mesenchymal state, and the cells in the second state may be in an epithelial state, or vice versa.

In another aspect the invention provides a method of screening for an agent for use in either (i) modulating the transition of a population of substantially epithelial cells to a population comprising a mixture of epithelial cells and mesenchymal cells, or for use in (ii) modulating the transition of a population of substantially mesenchymal cells to a population comprising a mixture of mesenchymal cells and epithelial cells, wherein the method comprises the steps of:

a) providing a population of cells in which substantially all the cells are either mesenchymal or substantially all the cells are epithelial;

b) contacting the population of cells provided in a) with the agent;

c) screening the population of cells after contact with the agent to see if the population of cells changes from a population wherein substantially all the cells are epithelial or substantially all the cells are mesenchymal to a mixed population of cells with epithelial cells and mesenchymal cells. Preferably the agent is a candidate therapeutic or prophylactic agent. Preferably the population of cells is a population of cancer cells.

Preferably the agent is for use in preventing cancer metastasis. In a yet further aspect the present invention provides a method of screening for an agent for use in either (i) modulating the transition of a population of substantially HPA negative cells to a population comprising a mixture of HPA negative cells and HPA positive cells, or for use in (ii) modulating the transition of a population of substantially HPA positive cells to a population comprising a mixture of HPA positive cells and HPA negative cells, wherein the method comprises the steps of:

a) providing a population of cells in which substantially all the cells are either HPA positive or substantially all the cells are HPA negative;

b) contacting the candidate therapeutic or prophylactic agent with the population of cells provided in a);

c) screening the population of cells after contact with the candidate therapeutic or prophylactic agent to see if the HPA status of the population of cells changes from a population wherein substantially all the cells are HPA positive or substantially all the cells are HPA negative to a mixed population of cells with some cells HPA positive and some cells HPA negative.

Preferably the agent is a candidate therapeutic or prophylactic agent.

Preferably the population of cells is a population of cancer cells. Preferably the agent is for use in preventing cancer metastasis.

HPA or Helix pomatia agglutinin is a lectin derived from Helix pomatia which can be used to detect glycans that have a terminal N-acetylgalactosamine (GalNAc). For the purposes of this invention a HPA positive cell is defined as a cell that displays significant levels of GalNAc on its surface and is therefore recognised by, and binds high levels of HPA. An HPA negative cell is defined as a cell that does not display or displays low levels of GalNAc on its surface, and is therefore not recognised by, and does not bind to, HPA or binds only low levels of HPA. The HPA binding capability of cells in a cancer cell population is mixed (i.e. there is a mixture of HPA-positive and HPA negative cells in a normal cancer cell population); as shown herein it is possible to separate the HPA-positive and HPA negative cells, for example by using magnetic beads. Typically those cells which bind significant amounts of HPA (HPA-positive) have a more epithelial phenotype and those cells which bind lesser amounts of HPA (HPA-negative) have a more mesenchymal phenotype. It is known that certain cancers and cancer cell lines can bind HPA and that cancer and cancer cell lines that bind high levels of HPA are more metastatic than cancers that do not.

The present invention is supported by the data presented herein which demonstrates for the first time that a population of cancer cells comprises cells with different surface glycosylation, in particular that a population comprises both HPA positive and HPA negative cells, and that if the HPA positive cells are isolated from the population and observed over a period of up to 72 hours then the population will revert back to a population containing both HPA positive and HPA negative cells. Similarly, if the HPA negative cells are isolated from the population and observed over a period of up to 72 hours then the population will revert back to a population containing both HPA positive and HPA negative cells. This is a novel, previously unrecognised property of HPA-positive and HPA-negative cells, which is used in a method of the invention to measure the transition of cells between difference states. This plasticity is believed to play an important role in cancer metastasis and by disrupting it novel therapeutics for use in reducing or preventing cancer metastasis can be identified. An agent according to the invention may be a synthetic or naturally occurring molecule/biomolecule, for example the agent may be an antibody or fragment thereof, a protein, a peptide, a lectin, a small molecule drug (preferably less than < 900 daltons), an aptamer, an oligo, an shRNA, a RNAi, siRNA or any other suitable agent.

The cancer metastasis which may be prevented using a therapeutic or prophylactic agent identified by the method of the invention may be derived from any cancer, including one or more of the following cancer types : breast cancer; lung cancer; prostate cancer; renal cancer; multiple myeloma; thyroid cancer; cancers of the head and neck; cancers of the digestive tract including stomach cancer, colon cancer, colorectal cancer, and liver cancer; malignant tumours of the female reproductive organs including ovarian cancer, endometrial cancer, and cervical cancer; bladder cancer; brain tumours including neuroblastoma; sarcoma; osteosarcoma; and skin cancers such as melanoma. Preferably the cancer is prone to metastasis. The cancer may be an epithelial cancer, such as breast cancer, prostate cancer, colorectal cancer, thyroid cancer, lung cancer or a melanoma.

The population of cells for use in a method of the invention may be obtained from a tissue culture cell line, or they may be obtained from a sample from a human or animal subject.

The population of cells may be a population of cancer cells. If the cell are cancer cells the sample may be obtained from a cancer tissue, for example by way of biopsy, or may be a sample of blood, lymph, effusion, ascites or other bodily fluid from a subject with cancer. The tissue may be breast cancer tissue .

The population of cells when isolated may contain a mixture of cells in different cell states, for example HPA positive and HPA negative cells, and/or may it contain a mixture of cells in a mesenchymal state and an epithelial state .

The population of cells may be a population of cancer cells from any cancer, including one or more of the following cancer types : breast cancer; lung cancer; prostate cancer; renal cancer; multiple myeloma; thyroid cancer; cancers of the head and neck; cancers of the digestive tract including stomach cancer, colon cancer, colorectal cancer, and liver cancer; malignant tumours of the female reproductive organs including ovarian cancer, endometrial cancer, and cervical cancer; bladder cancer; brain tumours including neuroblastoma; sarcoma; osteosarcoma; and skin cancers such as melanoma. The cancer cells may be a breast cancer cells, prostate cancer cells, colorectal cancer cells, thyroid cancer cells, lung cancer cells or melanoma cancer cells. Preferably the cancer cells are from a cancer prone to metastasis. The cancer cells may be breast cancer cells.

The invention may further include the step of producing a population of cells which are substantially all in one state, for example, a population that are substantially all HPA positive or a population of cells which are substantially all HPA negative. To obtain a population of cells that is substantially HPA positive or substantially HPA negative these cells must be selected and isolated from a mixed population. This may be done by using HPA as a capture agent to recover HPA positive cells, or to remove HPA positive cells leaving a population of HPA negative cells. In order to facilitate capture of HPA positive cells, HPA may be labelled, directly or indirectly, to allow it to be used to separate bound cells from unbound cells. In an embodiment HPA may be labelled with a fluorescent marker, allowing HPA positive cells to be isolated based on their fluorescence, for example by using FACS. Alternatively, HPA may be attached to or immobilised on a solid support - this attachment or immobilisation may be direct or indirect - for example, by using biotin and streptavidin where one is attached to the solid support and the other is attached to HPA. The solid support may be planar, a column or beads. HPA positive cells may be immobilised on the solid support and thus can be easily separated from the HPA negative cells.

HPA positive cells may be in an epithelial state . They may be rounded (that is, where the ratio of the longest to the shortest diameter is close to 1 ), and may be prone to adhere to other cells. HPA negative cells may be in a mesenchymal state. They may be elongated and may be able to invade surrounding tissue . Invasion typically requires the cell to release various enzymes to break down the ECM surrounding the cells and then to 'crawl' into the circulation or tissue. Cells in a population of cells are considered to be substantially all in one state if 95% or more of the cells, preferably more than 96%, 97%, 98% or 99% of the cells are in one state. Such populations may be described as substantially pure populations.

Cells in a population of cells, for example a population of cancer cells, are considered to be substantially all HPA positive if 95% or more of the cells, preferably more than 96%, 97%, 98% or 99% of the cells are HPA positive . Similarly, cells in a population of cells are considered to be substantially all HPA negative if 95% or more of the cells, preferably more than 96%, 97%, 98% or 99% of the cells, are HPA negative. Such populations may be described as substantially pure populations. Cells in a population of cells are considered mixed if at least 6% of the cells are in a first state, and at least 6% of cells are in a second state. Preferably in a mixed population at least 10% of the cells are in a first state and at least 10% of the cells are in a second state.

Cells in a population of cells, for example a population of cancer cells, are considered mixed if at least 6% of the cells are HPA positive. Preferably in a mixed population at least 10% of the cells are HPA positive and at least 10% of the cells are HPA negative.

The term contacting in the context of the present invention means any interaction between a candidate agent and the population of cells. This may be achieved by adding the candidate agent to the growth medium in which the cells are growing. The agent may be free in solution, or may be in a suspension, or may be immobilised on a solid phase, for example, on a planar surface, a column or on a bead.

The cell population may be screened as described in step c) within at least 12 hours of contact with the candidate agent; preferably the cells are screened within at least about 24, about 48 or about 72 hours of contact with the candidate agent. Preferably the cell population is screened between about 24 and about 72 hours after contact with the candidate agent.

A method of the invention may comprise a further step, step d), of identifying an agent as a candidate therapeutic or prophylactic agent, for example for use in preventing cancer metastasis, if the population of cells remains either substantially all HPA positive or substantially all HPA negative after contact with the agent.

The method may further comprise the step of concluding that if an agent is able to prevent a population of substantially pure HPA positive or substantially pure HPA negative cells from reverting to the mixed population seen in the originating population, for example if the cells were isolated from a population with 50% HPA positive and 50% HPA negative cells but the agent means a substantially pure population reverts to only 20% positive and 80% negative cells, or vice versa, then the agent may also be identified as a potential therapeutic agent for preventing cancer metastasis. Similarly, the agent may be identified as a candidate therapeutic or prophylactic agent, for example for use in preventing cancer metastasis, if the population of cells retains the same ratio of cell types (such as HPA negative to HPA positive) after contact with the agent.

The method may further comprise the step of identifying an agent that blocks the binding of HPA positive or HPA negative cells to a secondary site such as another cell for example but not limited to binding to endothelial cells, tissue matrix, matrix substitute, or biological molecules such as collagen or other protein, sugar or lipid and thus in doing so inhibits or promotes the transition from a HPA negative to a HPA positive state or vice versa.

According to a still further aspect the invention provides an agent identified by the method of the invention for use in the prevention of cancer metastasis. The agent may be intended to be used alone or in combination with a chemotherapeutic agent.

The invention may further provide a pharmaceutical composition comprising an agent according to the invention and a pharmaceutically acceptable carrier. The pharmaceutically acceptable carrier may be, but is not limited to, sucrose, starch, and other such excipients; cellulose, methyl cellulose, and other such binders; starch, carboxymethyl cellulose, and other such disintegrating agents; magnesium stearate, and other such lubricants; citric acid, menthol, and other such flavoring agents; sodium benzoate, sodium bisulfite, and other such preservatives; citric acid, sodium citrate, and other such stabilizers; methyl cellulose, polyvinyl pyrrolidone, and other such suspending agents; surfactants and other such dispersing agents; water, physiological saline, and other such diluents; beeswax, and the like.

A pharmaceutical composition according to the present invention may be intended to be administered by oral or non-oral, e.g., intramuscular, intraperitoneal, intravenous, intraci sternal injection or infusion, subcutaneous injection, transdermal or transmucosal routes. In some embodiments, compounds or pharmaceutical compositions in accordance with this invention or for use in this invention may be administered by means of a medical device or appliance such as an implant, graft, prosthesis, stent, etc. Implants may be devised which are intended to contain and release such compounds or compositions. An example would be an implant made of a polymeric material adapted to release the compound over a period of time . The compounds may be administered alone or as a mixture with a pharmaceutically acceptable carrier e.g., as solid formulations such as tablets, capsules, granules, powders, etc.; liquid formulations such as syrups, injections, etc.; injections, drops, suppositories, pessaries. In some embodiments, compounds or pharmaceutical compositions in accordance with this invention or for use in this invention may be administered by inhalation spray, nasal, vaginal, rectal, sublingual, or topical routes and may be formulated, alone or together, in suitable dosage unit formulations containing conventional non-toxic pharmaceutically acceptable carriers, adjuvants and vehicles appropriate for each route of administration.

The compounds of the invention may be used to as therapeutic or prophylactic agent, for example to prevent cancer metastasis, in human or non-human animals. Non- human animals may include mice, rats, horses, cattle, sheep, dogs, cats, and monkeys.

The skilled person will appreciate that all preferred or optional features of the invention may be applied to all aspects of the invention. Embodiments of the present invention will now be described, purely by way of example, with reference to the accompanying drawings, in which:

Figure 1 - shows ZR75 1 cells separated into populations of either substantially all HPA positive cells - referred to as the P population (the cells are epithelial in appearance) or substantially all HPA negative cells - referred to as the N population (the cells are mesenchymal in appearance). The HPA positivity is indicated by brown labelling of the cells. HPA negative cells appear blue from the presence of a non-specific nuclear counterstain, hematoxylin. At time Ohrs, both P and N populations of cells appear rounded because at this stage they are yet to settle and adhere to the glass coverslip that they have been introduced onto. The data presented shows populations at 0 hours and then cultured over 3 days post-separation. Images were captured with a Zeiss Axioplan microscope fitted with a JENOPTIK ProgResC3 colour camera. Scale bar = 10 μιη. Figure 2 - shows ZR75 1 cells separated into populations of either, at time Ohrs, substantially all HPA positive cells (these appear to be epithelial cells and have a rounded morphology, and are referred to as the P population) or substantially all HPA negative cells (these appear to be mesenchymal cells and have an elongated morphology, and are referred to as the N population). The data presented shows populations at 0 hours and then cultured over 3 days post-separation. The separated cells have been labelled for the binding of the lectin HPA, which is indicated by a brown colouration and a counterstain, haematoxylin, which labels the nuclei of all cells blue . Thus, at time Ohrs, substantially all P population cells are HPA positive (brown coloured) and all

N population cells are HPA negative (blue coloured). Over a period of 72hrs, both N and P populations revert to a mixed population containing a mixture of HPA positive (brown and rounded) and HPA negative (blue and elongated) cells. Images were captured with a NanoZoomer 2.0-RS digital slide scanner (Hamamatsu Photonics, Welwyn, UK). Scale bar = 25 μιη.

Figure 3 - shows in graphical form information about the cells depicted in Figures 1 and 2. More specifically, Figure 3A shows the percentage of cells exhibiting different levels of HPA positivity scored as - (completely negative), + (substantially negative), ++ (slightly positive), +++ (very positive) over

72hrs. 0 hours depicts the P and N populations of ZR75 1 immediately after isolation of the pure P and N populations. The data at 24, 48 and 72 hours shows how makeup of the population changes over time to become more mixed with respect to HPA binding, and coming to resemble that of the original, unseparated parental population. Figure 3B details HPA levels in a population of ZR75 1 cells that have never been separated, the data showing the percentage of HPA positivity and negativity in the population. The parental ZR75 1 cell population is approximately 50% HPA positive (judged as ++ or +++) and 50% HPA negative (judged as - or +).

Figure 4 - shows substantially HPA positive (P) and substantially HPA negative (N) populations separated from a mixed population of MCF7 cells. Cells are shown at 0 hours (immediately after separation) and 72 hours after separation. As in Figure 1 , initially substantially HPA positive and negative populations revert to a mixed population over the time frame . Images captured with a Zeiss Axioplmicroscope fitted with a JENOPTIK ProgResC3 colour camera. Scale bar = Ι Ομιη.

Figure 5 - shows substantially HPA positive (P) and substantially HPA negative (N) populations separated from a mixed population of BT474 cells.

Cells are shown at 0 hours (immediately after separation) and 72 hours after separation. As in Figure 1 , initially substantially HPA positive and negative populations revert to a mixed population over the time frame . Images captured with a Zeiss Axioplan microscope fitted with a JENOPTIK ProgResC3 colour camera. Scale bar = Ι Ομιη.

Figure 6 - shows in graphical form information about the cells depicted in Figure 4. Figure 6A shows the percentage HPA positivity at 0 and 72hrs post- separation for MCF7 isolated P and N populations. Figure 6B shows percentage HPA positivity in unseparated MCF7 cells. The data indicate how makeup of the population has changed over time to become more mixed with respect to HPA binding, and coming to resemble that of the original, unseparated parental population. The parental MCF7 cell population is approximately 70% HPA ++ or +++ positive and 30% HPA - or +.

Figure 7 - shows in graphical form information about the cells depicted in Figure 5. Figure 7A shows the percentage HPA positivity at 0 and 72hrs post- separation for BT474 isolated P and N populations. Figure 7B shows percentage HPA positivity in unseparated BT474 cells. The data indicate how makeup of the population has changed over time to become more mixed with respect to HPA binding, and coming to resemble that of the original, unseparated parental population. The parental BT474 cell population is approximately 20% HPA ++ or +++ positive and 80% HPA - or +. Figure 8 - shows the morphology of ZR75 1 cells separated and imaged at

24hrs post-separation. Figure 8A shows an HPA-positive cell - in the epithelial state, indicated by a rounded morphology. The circular structure near the centre of the cell is an HPA-labelled Dynabead and its presence confirms that this is an HPA-positive cell. Figure 8B shows an HPA-negative cell - in the mesenchymal state indicated by an elongated morphology and numerous finger-like projections (pseudopodia). Images were captured with SEM S3400 (Hitachi). Scale bar = 20μιη.

Figure 9 - shows a comparison of the percentage of cells with pseudopodia in HPA-positive and HPA-negative cells across MCF7, ZR75 1 and BT474 cell lines indicating that HPA negative cells have significantly more pseudopodia than HPA positive cells.

Figure 10 - is an SEM image of MCF7 cells, highlighting Dynabead position in opposition to pseudopodia formation indicating that pseudopodia are not formed in areas of the cell surface where HPA-binding glycosylation is present. Images were captured with SEM S3400 (Hitachi) . Scale bar = 50μιη.

Figure 11 - shows a comparison of the invasive ability of unseparated MCF7, ZR75 1 and BT474 cells, separated HPA-positive cells and separated HPA- negative cells indicating that HPA-negative cells are significantly more invasive than either HPA-positive or the unseparated parental cell line. The invasive ability was determined by the Corning® BioCoat™ Matrigel® invasion assay. Error bars indicate standard error of the mean, and the p values are calculated from a 2-tailed t-test.

Figure 12 - illustrates the results of static adhesion assays for separated, HPA positive and HPA negative cell populations for MCF7 cells (Figure 12A), for ZR75 1 cells (Figure 12B) and for BT474 cells (Figure 12C) showing that overall HPA-positive cells are more able to adhere to endothelial monolayers than either HPA-negative cells or the unseparated parental cell line

The data presented herein demonstrates for the first time the plasticity of cancer cells, and their ability to move between an HPA positive and an HPA negative state. The results are the first to demonstrate that if a population of substantially all HPA positive cancer cells are isolated from a population of mixed HPA positive and HPA negative cells (the normal state of a cancer cell population), then the population will within 72 hours revert to a mixed cell state population again. Similarly, if a population of substantially all HPA negative cancer cells are isolated from a population of mixed HPA positive and HPA negative cells, then the population will within 72 hours revert to a mixed cell state population again. The time for reversion may be much quicker, it may be less than 12 hours, less than 24 hours, less than 36 hours, or less than 48 hours. It is believed that by changing state cancer cells are able to move and cancer is able to metastasise. For example, cells HPA positive cells appear to epithelial in nature, and tend to adhere to other cells, to have a rounded morphology and to be not very motile . Whereas HPA negative cells, appear more mesenchymal in nature, and do not tend to adhere to other cells as readily, they have a more stretched and spindly appearance and are more motile. The results herein demonstrate that cancer cells can make the transition from HPA positive to HPA negative, and vice versa, and that in different states the cells have different properties.

The invention is concerned with using this knowledge in a method to screen to identify agents that will stop or reduce this reversion thus allowing identification of potential candidate therapeutic and prophylactic agents for use in the prevention of cancer metastasis and other conditions arising because of the ability of cells to transition from one state to another. The method of the invention may also be used as a tool to identify genes, proteins and other biological molecules involved in the transition of cells between different states, for example between HPA positive and HPA negative, or between epithelial and mesenchymal states. These biological molecules may be used as a target for a therapeutic agent or as a diagnostic agent.

Materials and Methods Breast cancer cell lines

The breast cancer cell lines used in this study (see Table 1) were chosen to represent a range of breast cancer phenotypes from primary, non-metastatic cancer (BT474) to more biologically aggressive and metastatic cancers (ZR75 1 and MCF7). The cell lines all stably synthesise a heterogeneous profile of HPA-binding glycoproteins identical to those synthesized by clinical tumour samples, therefore validating their use as a model in these investigations. All breast-cancer cell-lines were obtained from European Collection of Authenticated Cell Cultures (ECACC) . Table 1

BT474 ZR751 MCF7

Derived from Invasive ductal Metastatic, Metastatic,

cancer. malignant ascites malignant pleural from invasive effusion from primary ductal invasive primary breast cancer. ductal breast cancer.

Age of patient 60 63 69

(years)

Evidence of Poorly Tumourigenic in Highly

metastasis in SCID tumourigenic in SCID mice (Yin et tumourigenic in mice SCID mice al., 2003; SCID mice

(Valentiner et al., Valentiner et al., (Schumacher et al.,

2005 ; Iorns et al., 2005). 1997; Valentiner et

2012). al., 2005).

ER status (Engel and Young, + (Engel et al., + (Brooks et al.,

1978) 1978) 1973)

HPA lectin cells predominantly cells cells predominantly labelling HPA negative or approximately half HPA positive characteristics as weakly positive HPA positive and

reported by half negative or

Brooks et al. weakly positive

(2001)

Derived by Lasfargues et al. Engel et al. ( 1978) Soule et al. ( 1973)

( 1978) Cell separation

Streptavidin-coupled M280 Dynabeads® (2.8 μιη beads containing 17% iron to make them magnetic, with 650-900 pm of free streptavidin/mg of beads) were incubated with biotinylated-HPA (from Sigma) for 30 minutes at room temperature on an end- over-end mixer. 1 μg of biotinylated HPA to 0. 1 μg of streptavidin-coupled M280 Dynabeads® was used for the separation of one million cells. To minimise stress to the cells the HPA-bead conjugate was incubated with the cells in suspension for 5 minutes at room temperature also on an end-over-end mixer. By limiting the amount of time the cells are at less than optimal temperature and conditions, minimal stress is incurred by the cells. Cells which bound to the HPA-bead conjugate were pulled to the side of a microcentrifuge tube using a magnet (as the beads are magnetic) for 2 minutes and the HPA-negative (un-bound) cells collected at the bottom of the microcentrifuge tube and were aspirated out leaving only the bound HPA-positive cells. The HPA-negative cells were placed into a new microcentrifuge tube and the magnet was removed from the HPA-positive cells which can then be re-suspended.

ZR75 1 cells were used to optimise the procedure as it has approximately equal populations of both HPA-positive and HPA-negative cells. MCF7 and BT474 cells were unsuitable for optimisation experiments as they had a small population of either HPA-positive (BT474) or HPA-negative (MCF7) cells. This was important as both populations needed to be viable post-separation, and by using ZR75 1 cells a smaller number of initial cells could be used than would have been needed with either MCF7 or BT474 cells. Three biological replicates of the ZR75 1 cells were used for the cell- separations. The positive and negative populations were grown on separately and then lectin labelled (HPA) and imaged over 3 days post-separation, at: 0, 24, 48 and 72 hours post-separation. This was then repeated with MCF7 and BT474 cells - also with 3 biological replicates each - but with cells lectin only labelled and imaged at: 0 and 72 hours post-separation, to account for the smaller HPA-positive or HPA-negative population, respectively. To ensure that the Dynabeads® used had no effect on the cell sample, a control was put in place where cells were subjected to an identical sham procedure but without the addition of biotinylated HPA.

SEM analysis of isolated populations MCF7, ZR75 1 and BT474 cells were separated based on their HPA binding status, as described above . Three biological replicates were prepared per cell line . 13mm diameter glass coverslips were sterilised in 70% industrial methylated spirit (IMS), placed in a cell culture 24-well plate and left to air-dry in a class II cell culture hood. A suspension of 40,000 cells/ml was prepared and 500μ1 of cell suspension was added per well. Plates were gently swirled once and then left to grow at 37°C with 5% C0₂ atmosphere for 24 hours. After, media was aspirated from the wells and the cells fixed with 2% v/v glutaraldehyde in PIPES buffer pH 7.4 for 30min at room temperature. Cells were then dehydrated through an ethanol series ( 10%, 20%, 30%, 50%, 70%, 90%, 95%, 100%) before 3x in absolute ethanol. The slides were critical point dried and mounted on 13mm diameter stubs using carbon tabs and then imaged using an Hitachi S3400 scanning electron microscope.

Images were captured of individual cells, so pseudopodia could be discerned. Pseudopodia are involved in cell motility and this analysis allowed determination of whether there was any relationship between HPA positivity/glycosylation and motility. If a dynabead was present, the cells were termed HPA-positive (because the dynabeads are linked to HPA). Pseudopodia were classified as projections from the main cell body. The HPA lectin labelling profile (HPA-positive or HPA-negative) and presence of pseudopodia were recorded for each cell captured per cell line.

Matrigel assay

A Corning® BioCoat™ Matrigel® plate was removed from storage at -20°C and allowed to equilibrate to room temperature (~30min) . After this time the Matrigel® matrix was rehydrated by adding 500μ1 of warmed (37°C) serum free (SF) tissue culture medium to the insert as well as 500μ1 to the well and incubated for 2hr in a humidified tissue culture incubator at 37°C with 5 % C0₂ atmosphere . Once rehydrated the SF-medium was aspirated from both the well and the inserts and 750μ1 of complete culture medium was added to the well only.

24 hours before the assay ZR75 1 cells were fed with SF-medium. On the day of the assay ZR75 1 cells were cell separated and magnetic beads removed. A cell suspension of 200,000 cells/ml in SF-medium was prepared of each of the HPA-positive and HPA-negative isolated cell populations. 250μ1 of each cell suspension was added per insert and gently swirled once to ensure an equal distribution of cells. The plate was incubated for 24 hours in a humidified tissue culture incubator at 37°C with 5% C0₂ atmosphere.

Matrigel® matrix and non-invaded cells were removed by lightly scrubbing the upper surface of the insert with a cotton-tipped swab and then repeating the process with a cotton-tipped swab lightly moistened in SF-medium. The upper surface of the insert was then washed twice with SF-medium. The inserts were then placed into a 24-well cell culture plate prepared with lml/well of ice cold methanol and fixed for 30min at 4°C. The inserts were washed twice in phosphate buffered saline (PBS) pH 7.4 and then permeabilised in 0. 1 % Triton X- 100 in PBS pH 7.4 for l Omin at room temperature . The inserts were washed twice in PBS pH 7.4 and then stained in Harris' hematoxylin for 3min at room temperature. The inserts were "blued" in running tap water and then dehydrated though an alcohol series, cleared in xylene and mounted using Depex.

Four experiments were performed each consisting of four biological replicates each of: unseparated, HPA-positive and HPA-negative populations. The membranes were observed using a Zeiss Axioplan microscope fitted with a JENOPTIK ProgResC3 colour camera. Images at low power resolution of the complete membrane were captured so that all of the invaded cells could be counted. This was achieved by imaging the membrane in a grid formation starting on the top left hand corner and working left-to-right across the membrane, then down a row and across again. If a cell appeared in a previous image, then it was not counted in the next. Static adhesion assay

13mm diameter glass coverslips were briefly sterilised in 70% v/v IMS in tap water and gently placed into the wells of 24-well cell culture plates. The culture plate lids were left off and the coverslips allowed to air dry in a class II cell culture hood. After this time 500μ1 of 0.2% w/v bovine gelatine in PBS pH 7.4 was added per well and then incubated for 30min in a humidified tissue culture incubator at 37°C with 5 % C0₂ atmosphere to set. Excess gelatine was aspirated and 100,000 human umbilical vein endothelial cells (HUVEC) were seeded per well. The culture medium was changed daily until the cells had achieved a confluent monolayer. Prior to the assay the endothelial cells were pre-stimulated with l Ong/ml of TNF-a (Sigma) in EBM-2 cell culture medium for 2hr in a humidified tissue culture incubator at 37°C with 5 % C0₂ atmosphere.

T75 flasks of cancer cells were incubated for 2hr in a humidified tissue culture incubator at 37°C with 5% C0₂ atmosphere with a warmed (37°C) solution of the fluorescent dye 8-hydroxypyrenetrisulphonic acid (PTS) (Sigma) at a concentration of l Omg/ml in appropriate cell culture medium. After this time the PTS solution was aspirated and the cells washed 5x with warmed (37°C) PBS, pH 7.4 to removed excess PTS. 10ml of preferred cell culture medium was dispensed into the culture flask and the cells were cell scraped. The cell suspension was aspirated and pelleted by centrifugation at 1 , 100 x g for 3 min, the supernatant discarded and the cell pellet either used as an unseparated population or for separation into HPA-positive and HPA-negative populations using the cell separation method. The TNF-a solution was aspirated from the endothelial cells and 200,000 cancer cells were seeded per well in 1ml of preferred cell culture medium. The plate was placed in a humidified tissue culture incubator at 37°C with 5% C0₂ atmosphere. Cancer cells were allowed to adhere for a range of times ( lmin, 5min, l Omin, 20min, 40min and lhr) before the cell suspension was aspirated and the well washed gently 3x with warmed (37°C) PBS .

Cells were fixed by adding lml/well of ice cold 25% v/v acetic acid in methanol for l Omin at 4°C. After this time the coverslips were carefully removed from the 24-well cell culture plate and mounted onto glass slides with a small amount of fluoromount anti-fade mountant. Slides were observed within 48hr and stored at 4°C in the dark to preserve fluorescence . An inverted fluorescence microscope (Zeiss Primvert fitted with AxioCamlCM l) was used to observe slides, all adhered cells were counted per coverslip. Counting was performed using the microscope in a checkerboard formation. Counts were normalised to unseparated populations within the assay and averaged to give fold change.

Results

Cell separation

To test whether the cell separation method worked ZR75 1 cells were initially used because in this cell line about 50% of the cells in the population are HPA positive. Cells were bound to HPA-biotin and then sorted using streptavidin conjugated to magnetic beads. The sorted cells which bound to the magnetic beads were referred to as the "P population" (these were positive cells) and the unbound cells were referred to as the "N population" (the negative cells). To ensure these populations were pure the cells were stained immediately after separation, and then at different time points following separation. Figures 1 and 2 illustrate the P and N ZR75 1 populations at: Ohrs, 24hrs, 48hrs and 72hrs post-separation (using different microscopes at lower and higher magnification, respectively). Initially the ZR75 1 P population is 100% HPA +++, whilst the N population is 100% -, suggesting that separation of cells has been successful. Interestingly, over time negative cells began to appear in the P population and positive cells appeared in the N population, suggesting that they are reverting back to a mixed population. This observation is quantified in Figure 3A; indeed, by 48-72 hours the proportion of negative and positive cells in both the P and N populations closely reflects the pattern of glycosylation in the cells prior to separation (Figure 3B). Taken together these results show that the cells can be successfully separated based on their surface glycosylation patterns and that the glycosylation status within a single cell is not fixed but can change over time. To test whether this was a phenomenon restricted to the ZR75 1 or potentially a feature of breast cancer cells generally, the same experiments were performed on MCF7 and BT474 cells. Figures 4 and 5 illustrate 0 and 72 hours post-separation of MCF7 and BT474 cells respectively. At 0 hours post-separation both MCF7 and BT474 P populations were nearly 100% HPA +++ and after 72 hours post-separation P populations were almost the same as unseparated proportions (as quantified in Figures 6A/B and 7A/B). Similarly, both MCF7 and BT474 N populations were - 100% HPA - and then after 72 hours post-separation were varied in their HPA-labelling characteristics and almost the same as unseparated proportions (as illustrated in Figures 6A/B and 7A/B) . These results support that the glycosylation pattern of cancer cells, and in particular breast cancer cells, is plastic and can alter over time.

SEM analysis of isolated populations

SEM analysis of the isolated populations allowed the morphology of the different cell states to be analysed. Positive and negative cells were analysed by scanning electron microscopy (SEM). Figure 8 illustrates representative examples of a HPA-positive and a HPA-negative ZR75 1 cell. The HPA-positive cell is round and the HPA-negative cell is elongated and has many pseudopodia. As illustrated in Figure 9, HPA-negative cells have significantly more pseudopodia - thus were more elongated and possibly more motile - than the HPA-positive cells. When HPA-positive cells had pseudopodia the Dynabead (bound to the HPA) was often located opposite to the projection, as illustrated in Figure 10, consistent with HPA binding being associated with a less motile/less elongated phenotype. These results suggest that there are morphological and phenotypic differences in the HPA positive and HPA negative cells.

Phenotypic differences in HPA positive and HPA negative cells

To test whether HPA positive cells (which appear more epithelial in nature) are more invasive a matrigel invasion assay was used to assess how well the cells are able to break down extracellular matrix and migrate through a pore. The HPA positive cells do not show any increased ability to invade through matrigel, whereas the HPA negative cells (which appear more mesenchymal in nature) show a greatly increased capacity to do so, suggesting that HPA negative cells are more invasive than HPA positive cells (Figure 1 1).

These results were unexpected in the light of previous work showing a correlation between positivity and metastatic ability, they do fit with the observations above that negative cells are more mesenchymal in appearance. A static adhesion assay was used to show that populations of cells with a higher HPA positive status were able to adhere to endothelial cells. This assay allowed a cells ability to adhere to a layer of endothelial cells to be assessed. This ability to bind to vascular endothelial cells is important in the step of escaping from the primary organ and entering the circulation; in addition cells need to bind to the vasculature of a secondary organ in order to successfully leave the circulation and establish a second site of tumour growth. In the static adhesion assay the tumour cells are labelled with a dye, added to a layer of endothelial cells (HUVECs) and allowed to settle for a short period of time (in this study they were tested from 30 seconds to 20 minutes); unbound cells are then removed and the number of cells that adhered to the HUVECs were counted. For all three cell lines and all times of incubation an increase in adherence of HPA positive cells to the endothelial cell layer (Figure 12) was observed.

Together the static adhesion assay and the matrigel invasion assay were intended to model different aspects of the metastatic process. In order to metastasise cancer cells need to be able to both invade and adhere to endothelium. The data presented here shows that cancer cells can switch between HPA positive cells which are more epithelial in nature and adhesive to endothelial cells, to HPA negative cells which are more mesenchymal in nature which are able to crawl and invade tissue, and back again, - therefore demonstrating that plasticity of the cells is needed to make them metastatically successful.

Conclusion The results presented here demonstrate that the presence of the glycosylation marker recognised by HPA is plastic, and can dynamically appear or disappear within cells of a population over short time scales (of up to 72 hr). In particular, the results show that for each cell line there is a characteristic proportion of HPA positive and HPA negative cells, and that when separated into different populations (HPA positive or HPA negative) the cells will revert to that characteristic proportion within about 72 hours.

The data also shows that the HPA negative and HPA positive cells have unique traits that are needed for different parts of the metastatic cascade . Specifically, HPA positive cells are better able to adhere to the vascular endothelium whilst HPA negative cells have a greater ability to invade and migrate through extracellular matrix. This plasticity between these two phenotypes is key for cancer cells to access the full range of properties required to leave the primary organ and establish a secondary tumour at a new organ. Finding drugs that block this transition by using the method of the invention will lead to new ways of inhibiting metastasis in cancer.

Claims

1. A method of screening for a candidate agent for use in either (i) modulating the transition of a population of substantially epithelial cells to a population comprising a mixture of epithelial cells and mesenchymal cells, or for use in (ii) modulating the transition of a population of substantially mesenchymal cells to a population comprising a mixture of mesenchymal cells and epithelial cells, wherein the method comprises the steps of:

b) contacting the population of cells provided in a) with the agent;

c) screening the population of cells after contact with the agent to see if the population of cells changes from a population wherein substantially all the cells are epithelial or substantially all the cells are mesenchymal to a mixed population of cells with epithelial cells and mesenchymal cells.

2. A method of screening for a candidate agent for use in either (i) modulating the transition of a population of substantially HPA negative cells to a population comprising a mixture of HPA negative cells and HPA positive cells, or for use in (ii) modulating the transition of a population of substantially HPA positive cells to a population comprising a mixture of HPA positive cells and HPA negative cells, wherein the method comprises the steps of:

c) screening the population of cells after contact with the candidate therapeutic or prophylactic agent to see if the HPA status of the population of cells changes from a population wherein substantially all the cells are HPA positive or substantially all the cells are HPA negative to a mixed population of cells with some cells HPA positive and some cells HPA negative .

3. A method of screening for a candidate agent for use in modulating the ratio of cells in a first state in a population of cells to cells in a second state in the population of cells, wherein the population of cells comprises a mixture of cells some in the first state and some in the second state;

wherein in the first state cells are in an epithelial state and in the second state cells are in a mesenchymal state, or in the first state cells are in a mesenchymal state and in the second state cells are in an epithelial state, or in the first state cells are HPA positive and in the second state cells are HPA negative, or in the first state cells are HPA negative and in the second state cells are HPA positive;

wherein the method comprises the steps of:

b) contacting the population of cells provided in a) with the agent;

c) screening the population of cells after contact with the agent to see if the ratio of cells in a first state to cells in a second state changes.

4. The method of any of claim 1 to 3 wherein the candidate agent is a candidate therapeutic or prophylactic agent.

5. The method of claim 4 wherein the candidate agent is for use in preventing cancer metastasis.

6. The method of any of claims 1 to 3 wherein the candidate agent is intended to identify genes or proteins, or other agents, involved in the control of cell state transition.

7. The method of any preceding claim wherein the population of cells is a population of cancer cells.

8. The method of any preceding claim wherein step c) is performed with 72 hours of contact with the candidate agent.

9. An agent identified by the method of any preceding claim for use in the prevention of cancer metastasis .

10. A pharmaceutical composition comprising an agent identified by any of claims 1 to 8.