Methods for labelling proteins of interest

Abstract

Methods for identifying proteins of interest A method for identifying cell surface proteins in a population of cells is provided, the method comprising the provision of a population of cells, and the biotinylation of more than one functional group of cell surface proteins in the population of cells. Kits for performing the method, and applications of the method are also provided.

Description

[0001] Methods for labelling proteins of interest

[0002] Technical Field

[0003] The present invention relates to the field of proteomics, and particularly, although not exclusively, to methods for the labelling of cell surface proteins.

[0004] Background

[0005] Approximately 22% of all proteins coded by the human genome are found at the cell surface, but such fraction of the proteome is underrepresented in the literature and in high throughput proteomic analyses (5). Addressing this knowledge gap is expected to assist in the elucidation of aberrant alterations in surface protein expression, modification and interactions directly related to human diseases (6). Thereby, a more comprehensive and thorough understanding of the surfaceome will assist in identifying surface proteins for the development of novel targeted immunotherapies.

[0006] Mass spectrometry (MS)-based proteomics has become an increasingly valuable approach for studying expression patterns of plasma membrane and cell surface proteins (7-9), making this a powerful technology to globally and systematically characterise cell surface proteins for identifying cancer biomarkers and drug targets. Therefore, although comprehensive characterization of the surfaceome remains extraordinarily challenging, proteomic approaches are starting to reveal the surfaceome of different tumour types (10-12).

[0007] To specifically and comprehensively characterize proteins only located on the cell surface, the extraction, enrichment and purification of integral proteins are prerequisites for LC-MS / MS based identification and quantification, and this still represent a challenge in the process of defining the plasma membrane proteome (13, 14). This is because of the complexity of the biochemical, topographical and structural properties of the plasma membrane proteins, including (1) the hydrophobicity of the domains that transverse the membrane (2) the generally low abundant nature of these proteins compared with the other cellular proteins and (3) and problems associated with the separation of PM proteins from those of other organelles (15).

[0008] The difficulty is heightened by the heterogeneous character of these proteins and their specific physicalchemical properties (1). In the case of transmembrane proteins, plasma membrane proteins vary in the number of membrane-spanning helices (from one to 15), the ratio of membrane hydrophobic embedded domains to soluble intra- and extracellular portions and the number of modifications, such as glycosylation’s, that contribute to protein solubility (16). This represents an obstacle in the identification of integral membrane proteins and cell surface proteins with highly hydrophobic character. A further obstacle is the limitation in the number of cells required, particularly for analysis of primary samples. To help overcome these obstacles in characterising the complexity of cell surface proteins, a number of MS-based methods for plasma membrane protein and surfaceome enrichment have been developed (17, 18). A family of such methods are based on the chemical modifications of cell surface proteins at sites of glycosylation for cell surface capture (CSC) (19). The CSC method, which is based on the chemicalderivatization of glycated residues with a biotin moiety ( / .e. biotinylation), has been used to profile the surfaceome of a range of cell lines, primary cells and tissues, resulting in a mass spectrometry derived Cell Surface Protein Atlas (CSPA) with 1492 human cell surface glycoproteins (20). As an alternative to CSC, Weekes et a / 2012 developed an aminooxy-biotinylation procedure for capturing sialylated plasma membrane glycoproteins, resulting in superior protein sequence coverage and improved plasma membrane protein enrichment (18). Nonetheless, some disadvantages of glycan- and sialyl-based CSC include a requirement for a relatively large amount of starting material (nearly 108cells or 200 mg - 1 g of tissue), as well as the potential loss of glycan structures by oxidation, leading to low sensitivity.

[0009] As an alternative to glycan-based CSC, other chemical biotinylation methods have been successfully used to capture cell surface proteins irrespective of their posttranslational modification status. In such approaches, biotin reagents are derivatised with reagents that label primary-amines (-NH2), thus allowing covalent attachment of proteomes and subsequently enrichment using avidin-bound resins or beads by affinity purification. Primary amines (-NH2) are favourable targets for biotin derivatisation because they are the most common moieties in proteins, with lysines accounting for approximately 6% of all amino acid residues and being abundantly present on the surfaces of native protein tertiary structures, making them easily accessible for biotin labelling (17). Furthermore, primary amines are the most nucleophilic of all the protein functional groups, making them a favourable target for biotin derivatisation. Additionally, the amine group at the protein N-terminus can also be labelled. To date, a variety of N-hydroxysuccinimide (NHS)-ester derivatives have been developed for cell surface protein enrichment, many of which are commercially available. NHS-ester reagents can vary in length, solubility, cell permeability, and cleavability. Regarding cell permeability, sulfo-NHS-SS-biotin with a negatively charged sulfate group is reported to be membrane impermeable, a favourable characteristic that prevents the unwanted labelling of intracellular proteins. Notably, alternative biotinylation reagent NHS-PEG4 carries a potential risk of contamination from intracellular proteins due to its hydrophobic long chain. Furthermore, intracellular proteins leaked from dead cells may also be biotinylated with NHS esters, thus confounding surfaceomic data with background proteins that are not PMPs or located on the cell surface.

[0010] A potential limitation of all NHS-ester reagents that label free amines is that surface proteins with few or no exposed amines may be overlooked in surfaceomic experiments based on such chemical labels. As an alternative to such reagents, carboxyl reactive conjugation agents, on the other hand, label surface exposed carboxyl groups on aspartic acid (Asp), glutamic acid (Glu), and protein C-termini. One advantage of carboxyl labelling is the frequencies of Asp and Glu residues on the extra-cellular Transmembrane Proteins (TMPS), which some studies have found to be larger than that of Lys residues (21). However, despite the potential for carboxyl-based biotinylation for surfaceome enrichment, these reagents have not been fully explored for this purpose.

[0011] Surfaceomic studies are starting to unravel the complexity of the cell surface and plasma membrane proteome in cancer cells (22-24). Despite these advances, however, current methods for surfaceome analysis have limitations in terms of selectivity and sensitivity. Indeed, most published approaches require the use of several million cells as input material, limiting their utility to the analysis of primary tumours for which large number of cells can be obtained. There is also a need to systematically characterise theperformance of surfaceomic methodologies carried out with low number of cells as input and explore the potential to combine different biotinylation reagents to broaden the physiochemical properties of proteins that can be characterised with these techniques.

[0012] The present invention has been devised in light of the above considerations.

[0013] Summary of the Invention

[0014] Broadly, the inventors have developed a biotinylation approach based on “multiplexing” unique biotinylation agents. A formulation of “multiplexed” biotinylating reagents, containing multiple functional group reactivities, enables a more efficient and selective labelling of cell surface proteins, thus broadening the coverage of the cell surface proteome.

[0015] In a first aspect, the present disclosure provides a method for labelling cell surface proteins in a population of cells, the method comprising:

[0016] (i) providing a population of cells;

[0017] (ii) biotinylating primary amine groups on cell surface proteins within the population of cells;

[0018] and

[0019] (iii) biotinylating carboxyl groups on cell surface proteins within the population of cells.

[0020] In some embodiments, step (ii) comprises incubating the population of cells with an amine-reactive biotinylation reagent, wherein the amine-reactive biotinylation reagent comprises a N-oxysuccinimidyl moiety. In other words, in some embodiments, the amine-reactive biotinylation reagent comprises an N-oxypyrrolidin-2,5-dionyl moiety (NHS).

[0021] An “N-oxysuccinimidyl” moiety may be according to Formula (I):

[0022]

[0023] Formula (I)

[0024] In Formula (I), R1may be selected from the group consisting of H, SO3H, and SOsNa. In preferred embodiments, R1is H or SOsNa. Although not shown in Formula (I), amine-reactive biotinylation reagents according to Formula (I) also comprise a biotin moiety.

[0025] In some embodiments, the amine-reactive biotinylation reagent may be according to Formula (II):

[0026]

[0027] Formula (II)In Formula (II), L is a linker and R1may be selected from the group consisting of H, SO3H, and SOsNa. In preferred embodiments, R1is H or SOsNa.

[0028] L may comprise or consist of a bond, or -NHXC(O)-, wherein X is a bond or spacer, as defined herein.

[0029] In some embodiments L may be cleavable. In some embodiments, L may comprise a disulfide moiety, e.g. when X comprises a spacer comprising or consisting of -(CH2)n-S-S-(CH2)n-, wherein each n is independently 1 , 2, 3, 4, or 5.

[0030] In some embodiments, L comprises a spacer. Spacer moieties include oxyalkylene groups (e.g. PEG) and alkylene groups. In some embodiments, a spacer comprises or consists of an oxyalkylene or alkylene group. In some embodiments, the spacer comprises or consists of a group -(CH2CH2O)m- or -(CH2)m-, wherein m > 1. In some embodiments, m = 1 to 24, 1 to 12, 2 to 8, 2 to 6, or 2 to 4. In some embodiments, m = 1 , m = 2, m = 3, m = 4, m = 5, m = 6, m = 7, m = 8, m = 9, m = 10, m = 11 , or m = 12. That is, in some embodiments, L comprises or consists of -NH(CH2CH2O)m(CH2)nC(O)-, wherein m > 1, and n is 1 , 2, 3, 4 or 5, preferably wherein n is 2. In some embodiments, L comprises or consists of -NH(CH2)nC(O)-, wherein n is > 1 , preferably wherein n = 1 , 2, 3, 4, or 5.

[0031] In some embodiments, step (ii) comprises incubating the population of cells with a second amine-reactive biotinylation reagent.

[0032] In some embodiments, the second amine-reactive biotinylation reagent comprises a N-oxysuccinimidyl moiety. In some embodiments, the N-oxysuccinimidyl moiety may be according to Formula (I), described herein. In some embodiments, the first amine-reactive biotinylation reagent is according to Formula (I), described herein, wherein R1is SOsNa; and the second amine-reactive biotinylation reagent is according to Formula (I), described herein, wherein R1is H.

[0033] In some embodiments, the second amine-reactive biotinylation reagent may be according to Formula (II), described herein. In some embodiments, the first amine-reactive biotinylation reagent is according to Formula (II), described herein, wherein R1is SOsNa; and the second amine-reactive biotinylation reagent is according to Formula (II), described herein, wherein R1is H.

[0034] In some embodiments, the first and second amine-reactive biotinylation reagents are respectively sulfo-NHS-SS-biotin (CAS: 325143-98-4) and NHS-PEG4-biotin (CAS: 459426-22-3).

[0035] In some embodiments, step (iii) comprises activation of carboxyl groups using a carbodiimide.

[0036] In some embodiments, step (iii) further comprises stabilisation of the carbodiimide-activated carboxyl group using N-hydroxysuccinimide (NHS) or sulfo-NHS.In some embodiments, step (iii) further comprises incubating the population of cells with a carboxylreactive biotinylation reagent following carbodiimide activation, wherein the carboxyl-reactive biotinylation reagent comprises a free amine group. That is, the carboxyl-reactive biotinylation reagent is an aminobiotin or biotinylamine (e.g. amine biotin). In some embodiments, the carboxyl-reactive biotinylation reagent may be according to Formula (III):

[0037]

[0038] Formula (III)

[0039] wherein L' is a linker.

[0040] L' may comprise or consist of a bond, or -NHX'(CH2)2-, wherein X' is a bond or spacer, as defined herein.

[0041] In some embodiments L' may be cleavable. In some embodiments, L' may comprise a disulfide moiety, e.g. when X' comprises a spacer comprising or consisting of -(CH2)n-S-S-(CH2)n- wherein each n is independently 0, 1, 2, 3, 4, or 5.

[0042] In some embodiments, L' comprises a spacer, as described herein. That is, in some embodiments, L' comprises or consists of -NH(CH2CH2O)m(CH2)2-, wherein m > 1. In some embodiments, L' comprises or consists of -NH(CH2)n(CH2)2-, wherein n is > 1 , preferably wherein n = 1 , 2, 3, 4, or 5. In some embodiments, the carboxyl-reactive biotinylation reagent is amine-PEG2-biotin.

[0043] In some embodiments, the pH of the population is lowered following step (ii), prior to step (iii). In some embodiments, the pH of the population is lowered from a pH that is between 7 and 8 to a pH that is between 5 and 6.

[0044] In some embodiments, the method further comprises the step of:

[0045] (iv) isolating the biotinylated proteins from the population of cells to produce a cell surface protein-enriched sample

[0046] In some embodiments, step (iv) comprises performing affinity capture with streptavidin-containing magnetic beads.

[0047] In some embodiments, the method further comprises performing the step of:

[0048] (v) characterising the proteins in the cell surface protein-enriched sample.

[0049] In some embodiments, step (v) comprises performing mass spectrometry analysis.In some embodiments, step (v) comprises performing LC / MS analysis.

[0050] In some embodiments, the method further comprises analysing a control sample to correct for background.

[0051] In some embodiments, the method comprises analysing a control sample to correct for non-biotinylated proteins remaining in the cell surface protein-enriched sample. In some embodiments, the mock-labelled sample may be derived from and / or have a common origin with said population of cells.

[0052] In some embodiments, the method further comprises a step of cell dissociation on a sample to produce the population of cells of step (i), optionally wherein the cell dissociation comprises the use of enzymatic dissociation methods.

[0053] In some embodiments, the population of cells has a density of at least 15 million cells / mL.

[0054] In some embodiments, the population of cells has a density of approximately 20 million cells / mL.

[0055] In some embodiments, the population of cells contains fewer than 5 million cells. In some embodiments, the population of cells contains approximately 1.2 million cells. In some embodiments, the population of cells contains approximately 0.5 million cells.

[0056] In some embodiments, the population of cells comprises cryopreserved cells.

[0057] In some embodiments, the population of cells comprises cells isolated from a human subject.

[0058] In some embodiments, the population of cells comprises cancer cells. In some embodiments, the population of cells comprises cells derived from a solid tumor. In some embodiments, the population of cells comprises AML cells, colorectal adenocarcinoma cells, breast cancer cells, or hepatocellular carcinoma cells.

[0059] In some embodiments, the population of cells comprises acute myeloid leukaemia (AML) cells. In some embodiments, the population of cells comprises colorectal adenocarcinoma cells, breast cancer cells, or hepatocellular carcinoma cells.

[0060] In a second aspect, the present invention provides a kit comprising:

[0061] one or more amine-reactive biotinylation reagents; and

[0062] a carboxyl-reactive biotinylation reagent.

[0063] In a third aspect, the present invention provides a method for identifying one or more diagnostic markers or therapeutic targets, the method comprising performing the method of labelling cell surface proteins, according to a first aspect.In some embodiments of the third aspect, the one or more therapeutic targets are at least 20% higher abundance in disease samples compared to control samples and are elevated compared to control in at least 50% of the disease samples.

[0064] In some embodiments of the third aspect, the therapeutic target is identified as a target for a candidate antibody drug conjugate and / or CAR-T cell.

[0065] The invention includes the combination of the aspects and preferred features described except where such a combination is clearly impermissible or expressly avoided.

[0066] Brief Description of the Figures

[0067] Embodiments and experiments illustrating the principles of the invention will now be discussed with reference to the accompanying figures in which:

[0068] Figure 1. Design of Surfaceome Capture by Multiplex (SUCAM) biotinylation for surfaceome exploration. SUCAM is based on the use of a biotinylating reagent formulation to label amine and carboxyl reactive groups on proteins followed by comparative analysis of labelled and mock-treated cells by mass spectrometry to identify specific and nonspecific binders to streptavidin magnetic beads.

[0069] Figure 2. Workflow for developing a standardised surface protein capture quantification method.

[0070] Figure 3. Impact of cell density and cell numbers on performance of amine labelling for surfaceome enrichment. A. Gene Ontology (GO) Enrichment Analysis of cellular component (i) and ratio to cytoplasmic proteins (ii) for cell surface labelling of a pool of AML suspension cultures seeded at the densities shown in a volume of 1 mL. B. GO enrichment (i) and ratio to cytoplasmic proteins (ii) for biotinylation carried out using 1.2 M, 2.5 M, 5 M and 20 M cells at a density of 20 million cells / mL.

[0071] Ontologies were obtained from Uniprot and from Bausch-Fluck et al (27).

[0072] Figure 4. Evaluating the efficiency of cell surface enrichment for two amine-reactive biotin conjugates both individually and in-combination. A. Number of proteins isolated by Sulfo-NHS-SS-Biotin and NHS-PEG4-Biotin in each of the named ontologies and from Bausch-Fluck etal (27). (Surfaceome by ML) across three AML cell lines. B. Overlap of cell surface proteins captured by Sulfo-NHS-SS-Biotin or NHS-PEG4-Biotin. C. Overlap of cell surface proteins captured by Sulfo-NHS-SS-Biotin, NHS-PEG4-Biotin or double labelling.

[0073] Figure 5. Enrichment efficiency of cell surface proteins using carboxyl and single amine labelling. A. GO Enrichment Analysis shows a superior efficiency in the enrichment of the cell surface using carboxyl labelling than that of amine reactive labelling with NHS-SS-Biotin and NHS-PEG4-Biotin. Carboxyl labelling using stringent immunoaffinity purification conditions further enhanced the efficacy of cell surface and external side of plasma membrane GO term enrichment. B. The cell surface to cytoplasm enrichment ratio is higher for carboxyl labelling than that of amine labelling and is further enhanced using stringent affinity purification conditions. C. Overlap of cell surface and cytoplasmic proteins for NH2-labelling andCOOH-labelling shows some unique cell surface proteins for each labelling method, with less cytoplasmic contamination by carboxyl labelling compared to amine labelling. Ontologies were obtained from Uniprot and from Bausch-Fluck etal (27).

[0074] Figure 6. Enrichment efficiencies for the Plasma Membrane-Associated proteome using SUCAM and other biotin conjugation strategies. A, B, GO Enrichment Analysis restricted to cellular component (A) and ratios of cell surface-associated proteins to cytoplasmic of proteins (B) obtained by each of the named biotinylation approaches. Ontologies were obtained from Uniprot and from Bausch-Fluck etal (27).

[0075] Figure 7. Absolute numbers of Cell Surface, Plasma Membrane and Surfaceome by ML proteome obtained by SUCAM and other biotin conjugation strategies. Total number of proteins identified by the named methods are shown. The proposed SUCAM (a.k.a. double amine plus carboxyl) approach was compared to other methods for the isolation of proteins annotated to be present in the plasma membrane, the cell surface and external side of plasma membrane. Ontologies were obtained from Uniprot and from Bausch-Fluck et al (27).

[0076] Figure 8. Overlap of the cell surface proteins between five biotinylation methods. A. Overlaps of cell surface proteins isolated by the named methods and present in the surfaceome defined in Bausch-Fluck etal (27) (Surfaceome by ML, left panel) or in the named ontology (right panel). B. Similarity in the identification of cell surface proteins between SUCAM (double amine plus carboxyl) and the other four biotinylation methods.

[0077] Figure 9. Evaluation of cell surface- associated protein abundances across five biotin conjugation methods. Mean normalised protein abundances for proteins associated to the named ontologies are shown for SUCAM (a.k.a. double amine plus carboxyl) and other biotinylation methods. Ontologies were obtained from Uniprot and from Bausch-Fluck et al (27).

[0078] Figure 10. Evaluating the quantitative nature of SUCAM across AML cell lines. A. Distribution and median Coefficient of Variation (CV’s) for proteins across replicates for individual cell lines. B.

[0079] Quantification of cell surface proteins by normalisation of the label-free signals to the molecular weight of the protein.

[0080] Figure 11. A. Ontology enrichment analysis of proteins obtained from for 20 million cells / mL starting density depicts the specificity and sensitivity for the enrichment of Cell surface-associated terms, such as Cell Surface and External side of plasma membrane and that of other subcellular compartments in the biotinylated proteome relative to the unlabelled control. B. GO network obtained as in (A).

[0081] Figure 12. A. Absolute numbers of cell surface - associated proteins obtained by different biotin conjugation strategies showing higher number of proteins annotated to the Plasma Membrane Region, Cell-Substrate Junction, Cell-Cell junction, Plasma Membrane protein complex for the SUCAM (double amine plus carboxyl) method. B. Overlaps of cell surface-associated proteins for the five biotin conditions.

[0082] C. Percentages and overlaps of unique proteins obtained by the different conjugation methods.Figure 13. A. z-score heatmap depicts changes in the abundance of labelled cell surface proteins across five conditions identified relative to mock unlabelled control. B. Absolute intensities of cell surface and surface associated proteins across the named conditions. Kruskal-Wallis test shows a significantly (p<2.2E-16) higher expression of cell surface associated proteins in the double amine plus carboxyl compared to the other labelling methods.

[0083] Figure 14: A. Cell surface and associated proteins obtained by various biotin conjugation strategies showing higher numbers of proteins identified via SUCAM method than any other strategy. B. Analysis of individual cell-surface protein expression as measured using various biotin conjugation strategies. C. Comparison of intensities of proteins with surfaceome annotation between different biotin conjugation strategies. SUCAM produced higher mass spectrometric intensities for surfaceome protein class than other methods.

[0084] Figure 15: Impact of dissociation methods on the efficiency of SUCAM. Double amine plus carboxyl (a.k.a SUCAM) was applied to single-cell suspensions of colorectal cell lines obtained using various enzymatic and non-enzymatic dissociation conditions. A. GO enrichment analysis for single-cell suspensions biotinylated using SUCAM shows superiority for both enzymatic dissociating conditions, namely Accutase and non-enzymatic conditions. B. Total number of cell surface associated proteins isolated by SUCAM was highest for Accutase followed by Collagenase and lowest for non-enzymatic conditions. C. Normalised intensities of proteins for each ontology were similar for Accutase, Trypsin and collagenase and generally lowest for non-enzymatic dissociation conditions.

[0085] Figure 16: Comparative analysis of Amine-reactive biotinylation in cell monolayer and single cell suspension. The GO Enrichment analysis / absolute numbers of proteins isolated from biotinylation of cell lines growing in monolayers and cells dissociated into single-cell suspension before biotin labelling.

[0086] Figure 17: A comparative analysis of SUCAM and other biotin conjugation methods for enriching and identifying plasma membrane associated proteins for Accutase dissociated Colorectal cell line. A (i). GO Enrichment analysis (cellular component) shows double amine plus carboxyl and double amine to be superior, with a higher enrichment of cell surface, plasma membrane and external side of plasma membrane. A (ii). ratios of cell surface-associated proteins to cytoplasmic of proteins obtained by double amine plus carboxyl and double amine are higher than for single amine biotinylating agents. B (i).

[0087] The absolute number of proteins across the named ontologies was higher for double amine plus carboxyl.

[0088] B (ii). For Surfaceome by ML, 120 proteins were isolated by SUCAM in comparison to 95 for

[0089] Sulfo. NHS. Biotin and ~80 for NHS.PEG4. Biotin, Double amine and Carboxyl. B (iii). When assessing the numbers and overlaps of cell surface proteins across the tested methods, just 24% overlap of all cell surface associated proteins across all the five methods was found, with the highest percentage (10%) of unique proteins captured by SUCAM for Surfaceome by ML. C. The mean normalised abundances of proteins annotated to a number of cell surface-associated GO terms including proteins annotated to the Cell Surface G0;0009986, External side of plasma membrane G0;0009897 and surfaceome by ML were significantly higher (p=0.018, p=0.0012 and p=0.0013, respectively by Kruskal-Wallis test) for SUCAM.Figure 18: A comparative analysis of SUCAM and other biotin conjugation methods for enriching and identifying plasma membrane associated proteins for Collagenase dissociated Colorectal cell line. A, panel i. GO Enrichment analysis (cellular component) shows double amine plus carboxyl and double amine to be superior, with a higher enrichment of cell surface, plasma membrane and external side of plasma membrane. A (ii). A network plot in GO Enrichment Analysis, shows the relationship between PM-associated GO terms enriched in Biotin vs Control across five biotin conjugation methods.

[0090] B, panel i. The absolute number of proteins across the named ontologies was higher for double amine plus carboxyl and double amine. B (ii). When assessing the numbers and overlaps of “Surfaceome by ML” proteins across the tested methods, the highest percentage (19%) of unique proteins was captured by SUCAM. C. The mean normalised abundances of proteins annotated to a number of cell surface-associated GO terms including proteins annotated to the Cell Surface G0;0009986, External side of plasma membrane G0;0009897 and surfaceome by ML were significantly higher for SUCAM.

[0091] Figure 19. Bioinformatic pipeline for quantification of proteins by LC-MS / MS in absolute units (copy numbers). A. Calibration curve for quantification of proteins by LC-MS / MS in absolute units (copy numbers). Normalized LC-MS / MS signals of the UP2 protein standard (sigma) were plotted against the number of molecules analysed (converted from molarity). B. Absolute quantification of top 30 proteins across five Colorectal cell lines.

[0092] Figure 20: A comparative analysis of SUCAM and other biotin conjugation methods for enriching and identifying plasma membrane associated proteins for Collagenase dissociated Hepatocellular carcinoma cell line. A. GO Enrichment analysis shows that SUCAM has superior efficiency at enriching for the cell surface and plasma membrane than the other cross-linking conditions. B. (i). Double amine plus carboxyl shows superiority at broadening the cell surface repertoire by isolating a wider cohort of cell surface associated proteins, particularly within cell surface, surfaceome by ML and extracellular region and external side of plasma membrane ontologies, (ii) isolation of surfaceome by ML ontology using different biotin conjugation methods (iii) When assessing the numbers and overlaps of cell surface proteins across the tested methods, just 6% overlap of all cell surface associated proteins across all the five methods was found, with the highest percentage (5%) of unique proteins captured by SUCAM for Surfaceome by ML. C. The normalised abundances of a number of cell surface associated GO:ontologies that are related to HCC patho-physiology are highest for double amine plus carboxyl.

[0093] Figure 21: A comparative analysis of SUCAM and other biotin conjugation methods for enriching and identifying plasma membrane associated proteins in Collagenase dissociated Breast cancer cell line. A (i). GO Enrichment analysis (cellular component) shows double amine plus carboxyl and double amine to be superior, with a higher enrichment of cell surface, plasma membrane and external side of plasma membrane. A (ii). ratios of cell surface-associated proteins to cytoplasmic of proteins obtained by double amine plus carboxyl are higher than for carboxyl, double amine and single amine biotinylating agents. B (i). The absolute number of proteins for ontologies such as cell surface, plasma membrane, external side of plasma membrane was higher for double amine plus carboxyl, (ii) When assessing the numbers and overlaps of cell surface proteins across the tested methods, just 45 %overlap of all cell surface associated proteins across all the five methods was found, with the highest percentage (4%) of unique proteins captured by SUCAM for all named ontologies. C. The mean normalised abundances of proteins annotated to a number of cell surface-associated GO terms including proteins annotated to the Cell Surface G0;0009986, External side of plasma membrane G0;0009897 and surfaceome by ML were significantly higher (p=0.093, p=0.081 and p=0.081, respectively by Kruskal-Wallis test) for double amine plus carboxyl.

[0094] Figure 22: A comparative analysis of SUCAM and Alkoxyamine mediated biotinylation of glycoproteins for enriching and identifying plasma membrane associated proteins. A GO Enrichment analysis shows that Alkoxyamine and SUCAM have similar efficiency at enriching for the cell surface and plasma membrane than the other cross-linking conditions, (i) dotplot visualisation GO enrichment analysis (ii) network visualisation of GO enrichment analysis B. The mean normalised abundances of proteins annotated to a number of cell surface-associated GO terms including proteins annotated to the Cell Surface G0;0009986, External side of plasma membrane G0;0009897 and surfaceome by ML were significantly higher (p=0.13, p=0.033 and p=0.069, respectively by Kruskal-Wallis test) for double amine plus carboxyl. C. The GO Enrichment analysis / absolute numbers of proteins isolated using SUCAM vs alkoxyamine mediated biotinylation.

[0095] Figure 23: GO enrichment analysis (cellular component) result of DEGs with |logFC| >0.5: A. Dotplot B. network and C. heatmap visualization of GO enrichment analysis of differentially expressed proteins in biotin labelled vs unlabelled primary breast cancer cells.

[0096] Fig 24. The identification of differentially expressed proteins in Breast Cancer primary vs Healthy tissue. A. Heatmap showing expression levels (z scores of median Iog2 absolute quantities) of cell surface proteins for primary Breast Cancer derived cells and Healthy Tissue (Normal Epithelium, Myoepithelium and fibroblasts). Protein profiles were hierarchically clustered using Pearson correlation distance and average agglomeration. Colour corresponds to the row Z-score of Iog2 transformed absolute quantifications. B. PCA on breast cancer primary specimens, grouped by breast cancer prognostic subgroup, labelling conditions and tissue type. The PCA is based on the mean expression levels of the 400 proteins most differentially expressed across tissue and conditions. PC1 and PC2 explained 19.2% and 8.5% of the variance, respectively (n = 39). C. Volcano map with differential expression of surfaceomic proteins in Breast Cancer primary vs Healthy Tissue controls. The red represents 187 up-regulated proteins and the blue represents down-regulated proteins. Protein labels intentionally blurred.

[0097] Figure 25: Stratifying primary BC specific targets to distinct clinical BC sub-types. A. Differential protein expression (Log2FC) in primary BC vs Healthy controls stratified across different clinical BC subtypes determines targets’ sub-type specificity. Protein labels intentionally blurred. B.

[0098] LoglOAbsolute. Quantity of candidate surfaceomic targets across primary BC and Healthy Tissue stratified to distinct BC sub-types.

[0099] Figure 26: Bioinformatic-based pipeline for selecting pan-proteins for ADC in BC. A (i). Dot-plot visualisation of abundance vs absolute amounts of the surfaceomic proteins highlights candidate proteinshigh in >50% of BC primary cases with a high copy number and high differential expression in primary BC vs Healthy tissue. A (ii). Selection of candidate surfaceomic proteins ranked by % of BC cases with higher expression than the mean. absolute. quantity, high absolute quantity and significantly higher in BC primary vs Healthy tissue. Protein labels intentionally blurred. B. A list of top ranked candidates ordered by absolute quantification of the tumour. Protein labels intentionally obscurred.

[0100] Fig 27. “Multi-plex” amine reactive biotinylation of single-cell suspensions obtained from dis-aggregated cryo-preserved PDAC mouse tissue returns 78 up-regulated proteins in Biotin labelled vs unlabelled cells.

[0101] Detailed Description

[0102] The method of the present invention utilizes multiplexing of unique biotinylation reagents, allowing labelling of multiple different protein functional groups in order to achieve efficient and selective labelling of proteins of interest. In particular, the present invention allows efficient and selective labelling of cellsurface proteins, thereby broadening the coverage of the cell surface proteome.

[0103] As used herein, “proteins of interest” refers to a population of proteins that are being targeted in the labelling method of the present invention. “Proteins of interest” may refer to labelled proteins (e.g. biotinylated proteins). In preferred embodiments, “proteins of interest” refers to cell-surface proteins.

[0104] As used herein, “cell surface proteins” refers to proteins that are located on the outer membrane ( / .e. the external surface of the plasma membrane) of a cell. Cell surface proteins are part of (e.g. embedded in or associated with) the plasma membrane. Cell surface proteins are sometimes referred to as “plasma membrane proteins”. Cell surface proteins as used herein may include integral membrane proteins (e.g. transmembrane proteins), peripheral membrane proteins, lipid-anchored proteins (e.g. glycosylphosphatidylinositol-anchored proteins), amphitropic proteins and tail-anchored proteins.

[0105] Biotinylation

[0106] As used herein, “biotinylation” (also referred to as “crosslinking”) refers to the process of covalently attaching biotin to a molecule. Biotin binds to streptavidin and avidin with high-affinity, allowing efficient isolation of biotinylated molecules. Proteins can be biotinylated chemically (e.g. using reactive groups linked to biotin to conjugate various functional groups in molecules) or enzymatically (e.g. biotinylation of specific lysine residues within certain sequences by bacterial biotin ligase). Proteins that are biotinylated may also be referred to as proteins that are biotin-labelled. The use of biotinylation in the study of cell surface proteins is discussed in detail in Giuliano et al. Proteomics. (2008). 8(19):4012-24.

[0107] As used herein, biotinylation of proteins encompasses covalent addition of biotin to a reactive functional group on a protein. For example, biotinylation of proteins may include the covalent addition of biotin to primary amine groups, carboxyl groups, sulfhydryl (thiol) groups, or carbohydrate groups ( / .e. on glycoproteins). In some embodiments, biotinylation of proteins according to the present invention includes the covalent addition of biotin to primary amine groups. In some embodiments, biotinylation of proteins according to the present invention includes the covalent addition of biotin to carboxyl groups. Conditionsfor performing biotinylation are not particularly limited and may be adjusted by the skilled person. For example, biotinylation (a biotinylation reaction) may be controlled in terms of duration, time, concentration, pH (of either a reaction solution comprising biotinylation reagents, or of a cell population pH on which the biotinylation reaction is performed), temperature, and in terms of additives or reagents to optimise reaction efficiency.

[0108] In some embodiments, the method of the present invention combines the biotinylation of multiple functional groups on proteins of interest in a single population of cells. That is, according to the method of the present invention, the same proteins in a population of cells are exposed to more than one (e.g. two) different biotinylation reagents for targeting different functional groups of a cell surface protein. For example, the same proteins in a population of cells may be exposed to biotinylation reagents for primary amine biotinylation and to biotinylation reagents for carboxyl biotinylation. In some embodiments, the method of the present invention comprises biotinylation of two different functional groups of proteins of interest. In some embodiments, biotinylation of proteins according to the present invention includes the covalent addition of biotin to primary amine groups and the covalent addition of biotin to carboxyl groups of proteins of interest in a single population of cells. For example, the method of the present invention may result in at least a portion of proteins of the proteins of interest being labelled at multiple (different) functional groups (e.g. double biotinylated, e.g. biotinylated both at amine and carboxyl groups).

[0109] “Biotinylation” as used herein encompasses the use of biotin analogues, including iminobiotin and desthiobiotin. Such biotin analogues are described, for example, in Hirsch et al. Anal. Biochem. (2002).

[0110] 308(2) :343-357.

[0111] In preferred embodiments, biotinylation according to the present invention refers to chemical biotinylation. Chemical biotinylation involves the use of chemical biotinylation reagents. Biotinylation reagents comprise a reactive moiety / group (such as a succinimidyl moiety), and biotin. The reactive moiety / group reacts with a particular functional group on the molecule(s) of interest in order to crosslink biotin to the molecule(s) of interest. Biotinylation reagents may further comprise a linker between the reactive moiety and the biotin, wherein the linker may comprise a spacer (also known as a “spacer arm”, or “bridge”).

[0112] The appropriate choice of biotinylation reagent will depend on the specific application. Factors including solubility, reversibility, spacer length and functional group on the molecule(s) of interest must be considered and optimised. The selection of cross-linking reagents is reviewed, for example in Mattson et al. Molecular Biology Reports (1993) 17:167-183. The skilled person is readily able to select appropriate biotinylation reagents to suit the conditions of a specific application of the method of the present invention.

[0113] In some embodiments, the biotinylation reagents according to the present disclosure are water-soluble ( / .e. hydrophilic). That is, in some embodiments, the biotinylation reagents according to the present disclosure carry a charge. Such water-soluble reagents do not readily cross the plasma membrane. In some embodiments, biotinylation reagents according to the present disclosure have a partition coefficient of <1.Biotinylation reagents according to the present disclosure may be cleavable or non-cleavable. In some embodiments, a biotinylation reagent according to the present disclosure is cleavable. For example, biotinylation reagents according to the present disclosure may be photocleavable or disulfide cleavable ( / .e. may comprise a disulfide bond that can be cleaved using reducing agents). In some embodiments, a biotinylation reagent according to the present disclosure comprises a linker comprising a disulfide bond. In some embodiments, a biotinylation reagent according to the present disclosure may be non-cleavable.

[0114] In some embodiments, a biotinylation reagent according to the present disclosure may comprise a spacer between the reactive moiety and the biotin group ( / .e. as part of a linker, described herein). Spacers may be used to reduce steric hindrance in the binding of biotin to avidin / streptavidin. Spacers may be selected to modify the solubility of a biotinylation reagent. The spacer may be of any suitable length. Commonly employed spacers include polyethylene glycol (PEG) chains (e.g. PEG2, PEG4, PEG6, PEG12 or PEG24), hydrocarbon chains (e.g. alkyl chains) and amino-alkyls (e.g. 6-aminohexanoic acid), described e.g. in Berg and Fishman, Cold Spring Harb Protoc. (2020) 1:099259.

[0115] In some embodiments, the spacer is between 1 and 100 A, for example between 4 and 50 A, for example between 10 and 30 A. Each biotinylation reagent employed in a method according to the present disclosure may have the same length spacer, or different length spacers.

[0116] In some embodiments, the spacer comprises or consists of PEG. In some embodiments, the spacer comprises or consists of a chain of PEG units, having the structure (OCH2CH2)n, wherein n > 1. In particular, n may be between from 1 to 24, from 1 to 12, or from 1 to 4. In some embodiments, the spacer may comprise or consists of PEG2, PEG4, PEG12 or PEG24. To elaborate, when the spacer consists of PEG24, it may have the structure -(OCH2CH2)24- In some embodiments, the spacer comprises an aminohexanoate spacer arm. In some embodiments, a spacer is selected to decrease the hydrophobicity of the biotinylation reagent ( / .e. increase the hydrophilicity of the biotinylation reagent).

[0117] Biotinylation reagents according to the present disclosure may comprise reactive moieties that react with any suitable functional group on a molecule(s) of interest. For example, biotinylation reagents according to the present disclosure may comprise reactive moieties that react with (e.g. crosslink biotin to) primary amine groups (e.g. located at the N-terminus of each polypeptide chain, and the side chain of lysine residues), carboxyl groups (e.g. located at the C-terminus of each polypeptide chain, and in the side chains of aspartic acid and glutamic acid), sulfhydryl groups (e.g. located in the side chain of cysteine), or carbohydrate groups ( / .e. on glycoproteins).

[0118] In some embodiments, biotinylation reagents according to the present disclosure comprise primary amine-reactive moieties. The biotinylation of primary amine groups may be performed using any suitable amine-reactive biotinylation reagent. For example, amine-reactive biotinylation reagents may comprise a N-oxysuccinimidyl moiety or a tetrafluophenyl (TFP) moiety. An N-oxysuccinimidyl moiety according to the present disclosure is derived from NHS.The N-oxysuccinimidyl moiety may be according to Formula (I):

[0119]

[0120] Formula (I)

[0121] In Formula (I), R1may be selected from the group consisting of H, SO3H, and SOsNa. In preferred embodiments, R1is H or SOsNa.

[0122] In some embodiments, the amine-reactive biotinylation reagent comprises an N-oxysuccinimidyl moiety. In some embodiments, the amine-reactive biotinylation reagent is an NHS-ester. Such embodiments may comprise a moiety according to Formula (I) wherein R1is H. In some embodiments, the amine-reactive biotinylation reagent is a sulfo-NHS-ester. In such embodiments, the amine-reactive biotinylation reagent may comprise a moiety according to Formula (I) wherein R1is SOsNa.

[0123] Although not shown in Formula (I), amine-reactive biotinylation reagents according to Formula (I) also comprise a biotin moiety.

[0124] In some embodiments, the amine-reactive biotinylation reagent comprises a linker. That is, in some embodiments, the amine-reactive biotinylation reagent may be according to Formula (II):

[0125]

[0126] Formula (II)

[0127] In Formula (II), L is a linker and R1may be selected from the group consisting of H, SO3H, and SOsNa. In preferred embodiments, R1is H or SOsNa.

[0128] L may comprise or consist of a bond, -NHC(O)-, or -NHXC(O)-, wherein X is a bond or a spacer, as described herein.

[0129] In some embodiments L may be cleavable. In some embodiments, L may comprise a disulfide moiety, e.g. when X comprises a spacer comprising or consisting of -(CH2)n-S-S-(CH2)n-, wherein each n is independently 1 , 2, 3, 4, or 5. That is, the linker may comprise or consist of -NH(CH2)n-S-S-(CH2)n-C(O)-, wherein each n is independently 1 , 2, 3, 4, or 5. Linkers comprising disulfide moieties are not particularly limited and may be selected by the skilled person.

[0130] Preferably, when the linker of a biotinylation reagent comprises a disulfide moiety, R1= SOsNa. As such, in some embodiments, L comprises or consists of - NH(CH2)n-S-S-(CH2)n-C(O)- and R1= SOsNa, wherein each n is independently 1 , 2, 3, 4, or 5, and preferably each n is independently 1 , 2, or 3. Each nmay be the same or may be different. Therefore, in some embodiments both n may be 1 , 2, or 3, and in such embodiments both n are preferably 2.

[0131] In some embodiments, L comprises a spacer. Spacer moieties include oxyalkylene groups (e.g. PEG) and alkylene groups. In some embodiments, a spacer comprises or consists of an oxyalkylene or alkylene group. In some embodiments, the spacer comprises or consists of a group -(CH2CH2O)m- or -(CH2)m-, wherein m > 1. In some embodiments, m = 1 to 24, 1 to 12, 2 to 8, 2 to 6, or 2 to 4. In some embodiments, m = 1 , m = 2, m = 3, m = 4, m = 5, m = 6, m = 7, m = 8, m = 9, m = 10, m = 11 , or m = 12. That is, in some embodiments, L comprises or consists of -NH(CH2CH2O)m(CH2)nC(O)-, wherein m > 1, and n is 1 , 2, 3, 4 or 5. In some embodiments, L comprises or consists of -NH(CH2)nC(O)-, wherein n is 1 , 2, 3, 4, or 5.

[0132] Spacers comprising PEG moieties are not particularly limited and may be selected by the skilled person. Examples of spacers comprising PEG moieties include -(CH2CH2O)m- wherein m is 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10. In some embodiments, m may alternatively be larger than 10 for adjusting solubility and possibly other properties of the spacer. For example, m may be between from 11 to 24, for example 12 or 24, such that the spacer comprises a “PEG12” or “PEG24” moiety. Preferably, when the spacer comprises a PEG moiety, R1= H. As such, in some embodiments L comprises or consists of-NH(CH2CH2O)m(CH2)nC(O)-and R1= H, wherein n is 1 , 2, 3, 4, or 5 and m is 1 , 2, 3, 4, 5, 6, 7, 8, 9, or 10. Preferably, n is 1 , 2 or 3 and m is 1 , 2, 3, 4, 5, or 6. In some embodiments, n is 2 and m is 4 or 5, and in such embodiments m is preferably 4.

[0133] In some embodiments, a spacer may comprise an alkyl moiety. Spacers comprising alkyl moieties are not particularly limited and may be selected by the skilled person. Examples of spacers comprising alkyl moieties include — (Ci-Ck-alkyl) — , wherein k is between from 1 to 30. The value of k is not particularly limited and may be for example between from 1 to 25, for example from 1 to 20, for example from 1 to 15, for example 1 to 10, for example any of 1 , 2, 3, 4, 5, 6, 7, 8, 9, or 10. In preferred embodiments, the alkyl moiety is a linear alkyl moiety. In some embodiments, the alkyl moiety may be optionally substituted, for example with one or more of any selected from the group consisting of F, Cl, CF3, OCH3, OCH2CH3, or nitrile (CN).

[0134] In some embodiments, the amine reactive biotinylation reagent is disulphide-cleavable sulfo-NHS biotin ( / .e. sulfo-NHS-SS-biotin (CAS: 325143-98-4)). In some embodiments, the amine reactive biotinylation reagents is pegylated NHS biotin ( / .e. NHS-PEG4-biotin (CAS: 459426-22-3)).

[0135] In some embodiments of the method of the present invention, step (ii) comprises incubating the population of cells with a second amine-reactive biotinylation reagent. That is, in some embodiments, step (ii) comprises incubating the population of cells with a first and a second amine-reactive biotinylation reagent.

[0136] In some embodiments, the second amine-reactive biotinylation reagent may be according to Formula (I) or (II) as described herein. In some embodiments, the first and second amine-reactive biotinylationreagents are both according to Formula (I) or (II) as described herein. In some embodiments, the first amine-reactive biotinylation reagent is according to Formula (I), described herein, wherein R1is SOsNa; and the second amine-reactive biotinylation reagent is according to Formula (I), described herein, wherein R1is H. In some embodiments, the first amine-reactive biotinylation reagent is according to Formula (II), described herein, wherein R1is SOsNa; and the second amine-reactive biotinylation reagent is according to Formula (II), described herein, wherein R1is H.

[0137] In some embodiments, the two different amine-reactive biotinylation reagents according to the present disclosure are an NHS-ester and a sulfo-NHS-ester. In some embodiments, the two different amine reactive biotinylation reagents are disulphide-cleavable sulfo-NHS biotin (e.g. sulfo-NHS-SS-biotin (CAS: 325143-98-4)) and pegylated NHS biotin (e.g. NHS-PEG4-biotin (CAS: 459426-22-3)).

[0138] In some embodiments, biotinylation reagents according to the present disclosure comprise carboxylreactive moieties. The biotinylation of carboxyl groups may be performed using any suitable carboxylreactive biotinylation reagent. The biotinylation of carboxyl groups may be performed using any suitable carboxyl-reactive biotinylation method. In some embodiments, the method of the present invention comprises activation of carboxyl groups prior to crosslinking carboxyl groups with biotin. “Activation” of functional groups (e.g. carboxyl groups) refers to the process of making the functional group more reactive. Activation of carboxyl groups, as described herein, refers to the process of making carboxyl groups more reactive to facilitate the formation of amide bonds with amine groups on biotinylation reagents. Activation of carboxyl groups may be performed using any suitable reagent. For example, imidazoles, acid anhydrides, acyl chlorides, or carbodiimides. In some embodiments, the method of the present invention comprises activation of carboxyl groups on a protein of interest ( / .e. a cell surface protein) using a carbodiimide prior to crosslinking with biotin.

[0139] In some embodiments, the carbodiimide reacts with carboxyl groups to form an O-acylisourea intermediate (which may be referred to as a “carbodiimide-activated carboxyl group). The O-acylisourea intermediate may then react with a nucleophilic group (e.g. an amine group) in a suitable biotinylation reagent in order to crosslink biotin to the molecule(s) of interest.

[0140] In some embodiments, a carbodiimide according to the present disclosure is 1 -ethy l-3-(3-dimethylaminopropyl)carbodiimide hydrochloride (EDC) or dicyclohexylcarbodiimide (DCC). In some embodiments, a carbodiimide according to the present disclosure is EDC.

[0141] In some embodiments, the O-acylisourea intermediate ( / .e. the carbodiimide-activated carboxyl group) is stabilised following carbodiimide activation and prior to biotin crosslinking. The stabilisation of the O-acylisourea intermediate may form a stable and reactive group that may react with a suitable biotinylation reagent in order to crosslink biotin to the molecule(s) of interest. The O-acylisourea may be stabilised using NHS or sulfo-NHS. In some embodiments, the carbodiimide-activated carboxyl group is stabilised using sulfo-NHS.In some embodiments, the biotinylation of carboxyl groups occurs following activation of the carboxyl groups. In some embodiments, the biotinylation of carboxyl groups on the molecule(s) of interest occurs following stabilisation of the activated carboxyl group. In some embodiments, the carboxyl-reactive biotinylation reagent comprises hydrazide (e.g. hydrazide biotin). In some embodiments, the carboxylreactive biotinylation reagent comprises a free amine group (e.g. amine biotin).

[0142] In some embodiments, biotinylation of carboxyl groups comprises carbodiimide activation of carboxyl groups on the molecule(s) of interest, followed by reaction of hydrazide-biotin with the carbodiimide-activated carboxyl group. In some embodiments, biotinylation of carboxyl groups comprises carbodiimide activation of carboxyl groups on the molecule(s) of interest, followed by stabilisation of the carbodiimide-activated carboxyl group with NHS or sulfo-NHS, followed by reaction of amine biotin with the stabilised carbodiimide-activated carboxyl group.

[0143] In some embodiments, the carboxyl-reactive biotinylation reagent is amine-biotin. In some embodiments, the carboxyl-reactive biotinylation reagent may be according to Formula (III):

[0144]

[0145] Formula (III)

[0146] wherein L' is a linker.

[0147] L' may comprise or consist of a bond, -NH(CH2)2-, or -NHX'(CH2)2-, wherein X' is a bond or a spacer, as described herein.

[0148] In some embodiments L' may be cleavable. In some embodiments, L' may comprise a disulfide moiety, e.g. when X' comprises a spacer comprising or consisting of -(CH2)n-S-S-(CH2)n- wherein each n is independently 0, 1, 2, 3, 4, or 5. That is, the linker may comprise or consist of -NH(CH2)n-S-S-(CH2)n-wherein each n is independently 1 , 2, 3, 4, or 5.

[0149] In some embodiments, L' comprises a spacer, as described herein. In some embodiments, the spacer comprises or consists of a group -(CH2CH2O)m- wherein m > 1. In some embodiments, m = 1 to 24, 1 to 12, 2 to 8, 2 to 6, or 2 to 4. In some embodiments, m = 2, m = 3, m = 4, m = 5, m = 6. That is, L' may comprise or consist of -NH(CH2CH2O)2(CH2)2-

[0150] In some embodiments, the amine-biotin is amine-PEG2-biotin.

[0151] In some embodiments, biotinylation reagents according to the present disclosure comprise carbonylreactive moieties. The biotinylation of carbonyl groups may be performed using any suitable carbonyl-reactive biotinylation reagent. The biotinylation of carbonyl groups may be performed using any suitable carbonyl-reactive biotinylation method. In some embodiments, the method of the present invention comprises oxidation of carbohydrate residues on molecule(s) of interest to form carbonyl groups prior to crosslinking with biotin. In some embodiments, the method of the present disclosure comprises using sodium periodate to oxidise carbohydrate residues to form carbonyl groups on the molecule(s) of interest. Following the formation of carbonyl groups on the molecule(s) of interest, a carbonyl-reactive biotinylation reagent may be used to crosslink the molecule(s) of interest with biotin. In some embodiments, the carbonyl-reactive biotinylation reagent comprises a hydrazide, alkoxyamine or boronic acid. In some embodiments, carbonyl-reactive biotinylation reagents according to the present disclosure comprise hydrazide. For example, a carbonyl-reactive biotinylation reagent according to the present disclosure may be hydrazide-biotin or sulfo-hydrazide biotin.

[0152] In some embodiments, biotinylation reagents according to the present disclosure comprise sulfhydrylreactive ( / .e. thiol-reactive) moieties. The biotinylation of sulfhydryl groups may be performed using any suitable sulfhydryl-reactive biotinylation reagent. For example, sulfhydryl-reactive biotinylation reagents may comprise maleimide, haloacetyle (e.g. iodoacetyl, bromoacetyl or chloroacetyl), pyridyl disulfide, thiosulfonate or vinyl sulfone. In some embodiments, sulfhydryl-reactive biotinylation reagents according to the present disclosure comprise maleimide. For example, a sulfhydryl-reactive biotinylation reagent according to the present disclosure may be BMCC-biotin ( / .e. biotinylated maleimide cyclohexane carboxylate), maleimide-biotin, or sulfo-maleimide biotin.

[0153] As used herein the “coverage” refers to the extent to which the proteins of interest ( / .e. cell surface proteins) present in a population of cells is represented / captured by a given method. Higher coverage refers to a higher amount of proteins of interest present being identified by a given method.

[0154] In some embodiments, methods comprising the biotinylation of more than one (e.g. two) different functional groups increases the coverage of cell surface proteins compared to methods comprising biotinylation of a single functional group, in equivalent samples. In some embodiments, methods comprising the biotinylation of more than one (e.g. two) different functional groups on cell surface proteins increases the coverage of cell surface proteins by greater than 1 times, e.g. one of >1.01 times, >1.02 times, >1.03 times, >1.04 times, >1.05 times, >1.06 times, >1.07 times, >1.08 times, >1.09 times, >1.10 times, >1.11 times, >1.15 times, >1.20 times, >1.25 times, >1.3 times, >1.4 times, >1.5 times, >1.6 times, >1.7 times, >1.8 times, >1.9 times, >2 times, >3 times, >4 times, >5 times compared to the coverage of cell surface proteins achieved with methods comprising biotinylation of a single functional group, in equivalent samples.

[0155] Isolation of biotinylated proteins

[0156] In some embodiments, the method of the present invention comprises a step (iv) of isolating the labelled ( / .e. biotinylated) proteins from the population of cells. In some embodiments, the method of the present invention is a method for isolating cell surface proteins in a population of cells, the method comprising:

[0157] (i) providing a population of cells;(ii) biotinylating primary amine groups on cell surface proteins within the population of cells; (iii) biotinylating carboxyl groups on cell surface proteins within the population of cells; and (iv) isolating the biotinylated proteins from the population of cells to produce a cell surface protein-enriched sample.

[0158] As used herein, “isolation” of proteins refers to the separation of proteins of interest from a complex sample. “Isolation” of proteins of interest from a sample may also be referred to as “enrichment” of such proteins. Both terms refer to the process of increasing the proportion / concentration of proteins of interest within a sample relative to other molecule(s). “Isolation” of proteins of interest may refer to the complete isolation of proteins of interest from all other molecule(s) in a complex sample or may refer to the enrichment of proteins of interest in a sample. Commonly, isolation techniques encounter non-specific binding of molecule(s) other than the protein of interest ( / .e. background). Therefore, “isolation” of proteins of interest encompasses methods wherein the concentration of proteins of interest are increased relative to other molecule(s) in a complex sample.

[0159] The isolation or enrichment of proteins of interest according to the present disclosure results in a sample enriched in proteins of interest, i.e. a protein of interest-enriched sample (e.g. a cell surface protein-enriched sample). This protein of interest-enriched sample comprises a higher proportion of proteins of interest relative to other molecule(s) compared to the sample prior to the isolation or enrichment. In some embodiments, the method of the present invention comprises a step (iv) of isolating the biotinylated proteins from the population of cells to produce a cell surface protein-enriched sample.

[0160] The method for labelling cell surface proteins in a population of cells according to the present invention exploits the biological separation of intracellular and cell surface proteins via the plasma membrane. The labelling of cell surface proteins comprises exposing the population of intact cells to labelling reagents (e.g. biotinylation reagents). As the cells are intact, this exposure results in the labelling of cell surface proteins, without substantially labelling intracellular proteins. In some embodiments, the method according to the present disclosure comprises lysing the cells in the population of cells.

[0161] The isolation or enrichment of proteins of interest (e.g. of cell surface proteins) according to the present disclosure utilises affinity purification of biotin-labelled (i.e. biotinylated) proteins of interest. Biotin binds with high affinity to biotin-binding proteins (e.g. avidin, streptavidin, neutravidin). Therefore, following biotinylation of proteins of interest, these proteins may be isolated from a complex sample using biotinbinding protein-containing reagents (e.g. streptavidin beads). The isolation of biotinylated proteins using biotin-binding proteins is described, for example, in Rybak etal. Proteomics (2004) 4:2296-2299.

[0162] A “biotin-binding protein” as used herein refers to a protein that specifically binds to biotin. “Biotin-binding proteins” include streptavidin (e.g. UniProt #P22629, v161), avidin (e.g. UniProt #P02701 , v186), NeutrAvidin™ (e.g. ThermoFisher #31000), monomeric streptavidin (described, for example, in Lim et al. Biotechnol Bioeng. (2012) 110(1):57-67) and other avidin-like biotin-binding proteins (e.g. tamavidin or bradavidin, as described, for example, in Takakura etal. FEBS J (2009) 276(5):1383-1397 and Leppiniemi etal. Protein Sci. (2013) 22(7):980-94)).In some embodiments, a “biotin-binding protein” according to the present disclosure is streptavidin or avidin. In some embodiments, a “biotin-binding protein” according to the present disclosure is streptavidin.

[0163] “Affinity purification” (or “affinity capture”) as used herein refers to the process of purifying (e.g. isolating or enriching) a molecule of interest from a complex sample based on specific binding affinity between the molecule of interest and a binding partner. Molecules of interest may be labelled to enable purification. For example, according to the present disclosure, “affinity purification” refers to the isolation of biotinylated proteins from a complex sample via specific binding between biotin and biotin-binding proteins (e.g. avidin, streptavidin, neutravidin).

[0164] Affinity purification of biotinylated proteins using biotin-binding proteins may be carried out using any suitable method. Most commonly, biotin-binding protein-coated solid supports (e.g. beads, resins, membranes) are used to capture biotinylated proteins. Affinity purification methods are described, for example, in Urh et al. Methods in Enzymology. (2009) 463:417-43, or Kurien et al. Protein Electrophoresis: Methods and Protocols, Methods in Molecular Biology. Humana Totowa, NJ https: / / doi.org / 10.1007 / 978-1-61779-821-4). In some embodiments, affinity purification of biotinylated proteins comprises affinity chromatography. In some embodiments, isolation of biotinylated proteins according to the present invention comprises the use of biotin-binding protein-coated beads. In some embodiments, isolation of biotinylated proteins according to the present invention comprises biotinbinding protein-coated magnetic beads. In some embodiments, isolation of biotinylated proteins according to the present invention comprises streptavidin-coated magnetic beads.

[0165] In some embodiments, the isolation of proteins according to the present disclosure further comprises a step of lysing the cells. The method of the present invention involves the biotinylation of cell surface proteins in a cell population of intact cells. In some embodiments, to release the cell surface proteins into solution to enable downstream isolation, the cells in the cell population are lysed. Following cell lysis, the labelled proteins can be purified using affinity purification as described herein.

[0166] Reaction conditions

[0167] In some embodiments, the ordering of biotinylation of different functional groups ( / .e. the sequence of biotinylation of functional groups on a protein, which may be determined according to the order of addition of biotinylation reagents reactive with different functional groups) in the method of the present invention is optimised to minimise disruption of cells, in particular to minimise disruption of the plasma membrane. For example, biotinylation of different functional groups requires the biotinylation reactions to be performed at different pHs. In some embodiments, the ordering of the biotinylation of different functional groups is selected to minimise exposure of the population of cells to non-neutral pH conditions.

[0168] In some embodiments, the method of the present invention comprises a step of adjusting the pH of the population of cells following biotinylation of the first functional group on proteins of interest, prior tobiotinylation of the second functional group on proteins of interest. In some embodiments, the pH is adjusted from a first pH that is between 6.5 and 8, to a second pH that is below 6.5 or above 8. In some embodiments, the pH adjustment is lowering. That is, in some embodiments, the method of the present invention comprises a step of lowering the pH of the population of cells following biotinylation of the first functional group on proteins of interest, prior to biotinylation of the second functional group on proteins of interest. In some embodiments, the pH of the population of cells is lowered from a pH that is between 6.5 and 8, to a pH that is below 6.5. In some embodiments, the pH of the population of cells is lowered from between 7 and 8, to between 5 and 6.

[0169] In some embodiments, the method of the present invention comprises the biotinylation of a first and second functional group on proteins of interest, wherein the first functional group is amine. In some embodiments, the method of the present invention comprises the biotinylation of a first and second functional group on proteins of interest, wherein the second functional group is carboxyl. In some embodiments, the method of the present invention comprises the biotinylation of a first and second functional group on proteins of interest, wherein the first functional group is amine, and wherein the second function group is carboxyl. That is, in some embodiments of the method of the present invention, amine-reactive biotinylation occurs before carboxyl-reactive biotinylation.

[0170] In some embodiments, the method of the present invention comprises pH modulating steps. The modulation of the pH of a sample can be performed using any suitable method, as will be clear to the skilled person. In some embodiments, the pH of the population of cells is modulated between biotinylation of different functional groups. In some embodiments, the pH of the population of cells is modulated / shifted from a first pH that is between 6.5 and 8, to a second pH that is below 6.5, preferably that is between 5 and 6.

[0171] In some embodiments, the method of the present invention comprises a wash step following the isolation of proteins of interest from the population of cells, to reduce capture of non-specifically bound molecule(s) to the biotin-binding protein. In some embodiments, the wash step comprises use of a wash buffer comprising NaCI. In some embodiments, the concentration of NaCI in the wash buffer is at least 150 mM, e.g. > 200 mM, > 300 mM, > 400 mM, > 500 mM, > 750 mM, > 1 M, > 1.5 M, > 2 M. In some embodiments, the concentration of NaCI in the wash buffer is 2 M.

[0172] Dissociation

[0173] In some embodiments, the method of the present invention comprises a cell dissociation step. In some embodiments, the cell dissociation step produces the population of cells that is used in subsequent steps of the method of the invention. In some embodiments, a cell-dissociation step is included in the method when the population of cells is derived from a solid tumor and / or adherent cells.

[0174] As used herein, “cell dissociation” (or “dissociation”) refers to methods for obtaining populations of cells comprising single-cell suspensions from a sample. Cell dissociation entails breaking down extracellular matrix and cell-cell junctions that hold the cells in the sample together. For example, the sample maycomprise a solid tumor sample, cells derived from solid tumor and / or adherent cells. Cell dissociation is traditionally carried out via enzymatic and mechanical dissociation of the tissue using standard laboratory techniques. Cell dissociation methods are described, for example, in Jankelow et al. (2025) Stem Cells Transl Med. 14(11):szaf055, PMID: 41248140 and Lai etal. (2022) Sci Rep 12, 5713, PMID: 35383242.

[0175] In some embodiments, the method of the present invention comprises a cell dissociation step prior to the biotinylation steps of the method. In some embodiments, the cell dissociation step produces the population of cells that is used in subsequent steps of the method ( / .e. in the biotinylation steps). In some embodiments, the step of “providing a population of cells” in the method of the present invention comprises a cell dissociation step / method, e.g. as described herein.

[0176] In some embodiments, cell dissociation comprises enzymatic or non-enzymatic methods.

[0177] Non-enzymatic methods of cell dissociation include: mechanical dissociation (e.g. mechanical disruption of samples via agitation), or chemical dissociation. Chemical dissociation utilises chelating agents to disrupt intercellular bonds. Examples of chelating agents for use in dissociation methods include, but are not limited to, ethylenediaminetetraacetic acid (EDTA), diethylenetriaminepentaacetic acid (DTP A); nitrilotriacetic acid (NTA); Dimercaprol; Metallothionein; and ethylene glycol bis(p-aminoethyl ether)-N,N,N',N'- tetracetate (EGTA). In some embodiments, cell dissociation comprises the use of chemical methods. In some embodiments, cell dissociation comprises the use of chelating agents. The chelating agent is preferably EDTA.

[0178] Enzymatic methods of cell dissociation utilise enzymes to enzymatically break down samples into singlecell suspensions. Enzymes that may be used in such enzymatic methods of cell dissociation include: collagenase, dispase, trypsin, papain, hyaluronidase, or commercial cell detachment solutions such as Accutase® or Accumax™.

[0179] In preferred embodiments, cell dissociation comprises the use of enzymatic methods. In some embodiments, cell dissociation comprises the use of Accutase®, collagenase and / or trypsin. That is, in some embodiments, the step of cell dissociation is carried out using Accutase®, collagenase and / or trypsin. In some embodiments, cell dissociation comprises the use of Accutase®. That is, in some embodiments, the step of cell dissociation is carried out using Accutase®. In some embodiments, the method of the present invention comprises a cell dissociation method, wherein the cell dissociation comprises the use of Accutase®, collagenase and / or trypsin.

[0180] In some embodiments, the method of the present invention comprises:

[0181] (i) Performing a method of dissociation on a sample in order to produce a population of cells;

[0182] (ii) biotinylating primary amine groups on cell surface proteins within the population of cells; (iii) biotinylating carboxyl groups on cell surface proteins within the population of cells; (iv) isolating the biotinylated proteins from the population of cells to produce a cell surface protein-enriched sample; and(v) characterising the proteins in the cell surface protein-enriched sample.

[0183] methods

[0184] In some embodiments, the method of the present invention comprises a step (v) of characterising the proteins in the cell surface protein-enriched sample, described herein. In some embodiments, the method of the present invention is a method for characterising (e.g. identifying) cell surface proteins in a population of cells, the method comprising:

[0185] (i) providing a population of cells;

[0186] (ii) biotinylating primary amine groups on cell surface proteins within the population of cells; (iii) biotinylating carboxyl groups on cell surface proteins within the population of cells;

[0187] (iv) isolating the biotinylated proteins from the population of cells to produce a cell surface protein-enriched sample; and

[0188] (v) characterising the proteins in the cell surface protein-enriched sample.

[0189] Characterisation of proteins according to the present disclosure may be by any suitable method, for example, immunoassays or analytical techniques (e.g. mass spectrometry). Such methods are well known to the skilled person, and are described, for example, in Kohl et al. Cold Spring Harb Protoc. (2017) 7:pdb.top093690 and Shuken etal. J Proteome Res. (2023) 22:2151-2171.

[0190] In some embodiments, characterisation of proteins according to the present disclosure comprises the use of immunoassays. Immunoassays may include, for example, ELISA, immuno blot (e.g. western blot). In some embodiments, characterisation of proteins according to the present disclosure comprises the use of analytical techniques, e.g. spectrometric or spectroscopic techniques. Analytical techniques include, for example, mass spectrometry.

[0191] In some embodiments, characterisation of proteins according to the present disclosure comprises mass spectrometry. Mass spectrometry methods for identification and characterisation of proteins are described, for example, in Shuken et al. J Proteome Res. (2023) 22:2151-2171.

[0192] In some embodiments, characterisation of proteins according to the present disclosure further comprises the use of chromatographic techniques such as liquid chromatography (e.g. high-performance liquid chromatography). Liquid chromatography includes but is not limited to reversed-phase liquid chromatography, hydrophilic interaction liquid chromatography, size exclusion chromatography, and ion exchange chromatography. In some embodiments, characterisation of proteins according to the present disclosure further comprises the use of reversed-phase liquid chromatography. In some embodiments, characterisation of proteins according to the present disclosure comprises the use of liquid chromatography-mass spectrometry ( / .e. LC-MS).

[0193] In some embodiments, the method of the present disclosure further comprises a “mock-labelling” control condition. As used herein a “mock-labelling” condition refers to an condition used to account for non-specific binding to a label’s cognate ligand (e.g. non-specific binding of molecules to biotin-binding proteins such as streptavidin). In the mock-labelling condition, the labelling reagent ( / .e. the biotincontaining reagent) is replaced with a control reagent ( / .e. a reagent not containing biotin), while all other experimental steps and conditions of the method are identical. The mock-labelling condition allows for the determination of non-specific isolation of non-biotinylated molecules ( / .e. molecules that are not proteins of interest, e.g. molecules that are not cell surface proteins), by comparison between the proteins characterised in the experimental condition (e.g. the biotinylation condition) and proteins characterised in the mock-labelling condition.

[0194] In some embodiments, the method of the present invention comprises a mock-labelling control condition performed in parallel to the experimental condition (that is, the mock-labelling condition may be performed alongside the biotinylation method of the present invention). In some embodiments, the mock-labelling condition produces a sample representing non-specific protein binding to biotin-binding proteins.

[0195] As used herein, “a sample representing non-specific protein binding to biotin binding proteins” may also be referred to as a “non-specific biotin-bound protein sample”. A non-specific biotin-bound protein sample is produced by incubating a sample comprising proteins with mock-labelling reagents (e.g. reagents not containing biotin), and then incubating the mock-labelled protein sample with biotin-binding proteins. As the proteins in the mock-labelling sample have not been labelled with biotin, any binding of mock-labelled proteins to biotin-binding proteins is non-specific and represents background or “noise”. By comparing experimental samples ( / .e. biotinylated proteins isolated using biotin-binding proteins) against such control samples, non-specific binding / background can be eliminated from the experimental sample, in order to more accurately characterise the biotinylated proteins.

[0196] In some embodiments, the method of the present invention comprises a step of comparing the cell surface protein-enriched sample from step (iv) with a non-specific biotin-bound protein sample, in order to distinguish background from signal.

[0197] In some embodiments, the method of the present invention further comprises a mock-labelling condition, wherein the mock-labelling condition comprises the steps of:

[0198] a) providing a population of cells;

[0199] b) incubating the population of cells with a mock-labelling reagent as described herein ( / .e. a control reagent not containing biotin);

[0200] c) incubating the population of cells with biotin-binding proteins as described herein, to produce a sample representing non-specific protein binding to biotin-binding proteins; wherein, the cell surface protein-enriched sample from step (iv) is compared with the sample representing non-specific protein binding to biotin-binding proteins from step (c), in order to distinguish background from signal.

[0201] In some embodiments, the method of the present invention further comprises the mock-labelling condition, wherein the steps of the mock-labelling condition are performed in parallel with the experimental condition and / or wherein the mock-labelling condition employs a starting population of cellsderived from the same source as the experimental condition. By making use of a common source of cells for the experimental and mock-labelling conditions, the method of the invention may minimize differences between the number, abundance and / or mix of surface proteins between the experimental and mocklabelling conditions and thereby improve the accuracy of the above-mentioned background noise reduction, which may in turn lead to improved performance of the method of the present invention.

[0202] In some embodiments, the characterisation of proteins according to the present disclosure further comprises the comparative analysis between proteins characterised in the protein of interest-enriched sample of the experimental condition and proteins characterised in the protein of interest-enriched sample of the mock-labelling condition. This comparative analysis enables removal of molecules from the protein of interest-enriched sample that were non-specifically isolated. In some embodiments, the control reagent comprises phosphate-buffered saline (PBS).

[0203] Populations of cells

[0204] A population of cells in accordance with the present invention may comprise any kind of eukaryotic cell (e.g. any plant or animal cell).

[0205] In preferred embodiments, a population of cells in accordance with the present invention comprises animal cells. In some embodiments, the population of cells comprises mammalian cells. In some embodiments, the population of cells comprises human cells.

[0206] In some embodiments, the population of cells comprises cells from a cell line ( / .e. immortalised cells line). In some embodiments, the population of cells comprises primary cells e.g. cells isolated / obtained directly from a subject (e.g. from a biopsy). The population of cells may have been isolated / obtained from or derived from cells isolated / obtained from a subject of interest, e.g. a subject / patient as described herein.

[0207] The population of cells may be used immediately after obtaining from a subject of interest or may be stored for a period of time before being used in the method of the invention. That is, the population of cells may have been frozen. Frozen cells can be thawed ( / .e. the population of cells may be thawed cells). Cells can be frozen in an appropriate buffer / solution used to preserve cells at freezing temperatures (e.g. buffers comprising DMSO). The population of cells according to the present invention are intact ( / .e. whole cells).

[0208] In some embodiments, the population of cells comprises freshly dissociated cells (e.g. produced using a cell dissociation method as described herein). In some embodiments, the population of cells comprises cryo-preserved cells. In some embodiments, the population of cells comprises cryo-preserved single-cell suspensions. Such cryo-preserved cells may have been previously disaggregated from fresh tissue. In some embodiments, the population of cells comprises cells derived from cryo-preserved tissues.

[0209] In some embodiments, the population of cells comprises cells affected by a disease / disorder, wherein the identification of cell surface proteins in cells affected by said disease / disorder may have diagnostic or therapeutic benefit. For example, the population of cells may comprise cells affected by adisease / disorder in which identification of cell surface proteins in such cells may provide diagnostic markers for future diagnosis of the disease / disorder, and / or therapeutic targets for future treatment of the disease / disorder. Diagnostic or therapeutic benefit may refer to benefit to a subject from whom the cells are isolated, or to benefit to the broader medical community.

[0210] In some embodiments, the disease or disorder is cancer. In some embodiments, population of cells comprises cancer cells.

[0211] A cancer in accordance with the present disclosure may be any unwanted proliferation of cells, neoplasm or tumor. The cancer may be a cancer that is benign or malignant. The cancer may be primary or secondary (e.g. metastatic). A neoplasm or tumor may be any abnormal growth or proliferation of cells and may be located in (and / or derived from cells of) any organ / tissue.

[0212] A cancer may be derived from e.g. the adrenal gland, adrenal medulla, anus, appendix, bladder, blood, bone, bone marrow, brain, breast, cecum, central nervous system (including or excluding the brain) cerebellum, cervix, colon, duodenum, endometrium, epithelial cells (e.g. renal epithelia), gallbladder, oesophagus, glial cells, heart, ileum,jejunum, kidney, lacrimal glad, larynx, liver, lung, lymph, lymph node, lymphoblast, maxilla, mediastinum, mesentery, myometrium, nasopharynx, omentum, oral cavity, ovary, pancreas, parotid gland, peripheral nervous system, peritoneum, pleura, prostate, salivary gland, sigmoid colon, skin, small intestine, soft tissues, spleen, stomach, testis, thymus, thyroid gland, tongue, tonsil, trachea, uterus, vulva, and / or white blood cells.

[0213] “Cancer” as used herein may refer to any kind of cancer, e.g. a solid tumor (for example, a colorectal tumor, a fibrosarcoma, a gastric tumor, a glioblastoma, a renal tumor, a hepatic tumor, a pulmonary tumor, a melanoma, a nasopharyngeal tumor, an oral tumor, an osteosarcoma, an ovarian tumor, a pancreatic tumor, a brain tumor, or a prostatic tumor) or a blood cancer / haematological cancer, (for example, lymphoma, a leukaemia, or a myeloma). In some embodiments, the cancer is leukaemia. For example, acute myeloid leukaemia, acute lymphoblastic leukaemia, chromic lymphocytic leukaemia or chromic myeloid leukaemia. In some embodiments, the cancer is acute myeloid leukaemia.

[0214] “Cancer cells” as used herein may refer to cells from any kind of cancer, e.g. a solid tumor (for example, a colorectal tumor, a fibrosarcoma, a gastric tumor, a glioblastoma, a renal tumor, a hepatic tumor, a pulmonary tumor, a melanoma, a nasopharyngeal tumor, an oral tumor, an osteosarcoma, an ovarian tumor, a pancreatic tumor, a brain tumor, or a prostatic tumor) or a blood cancer / haematological cancer, (for example, lymphoma, a leukaemia, or a myeloma). In some embodiments, the population of cells comprises leukaemia cells. For example, acute myeloid leukaemia, acute lymphoblastic leukaemia, chromic lymphocytic leukaemia or chromic myeloid leukaemia. In some embodiments, the population of cells comprises acute myeloid leukaemia cells.

[0215] In some embodiments, the population of cells comprises cells derived from a solid tumor. In some embodiments, the population of cells comprises cells derived from colorectal cancer, hepatocellular carcinoma, or breast cancer.In some embodiments, the population of cells according to the present invention has a density of at least 2 million cells / mL, e.g. one of > 3 million cells / mL, > 4 million cells / mL, > 5 million cells / mL, > 6 million cells / mL, > 7 million cells / mL, > 8 million cells / mL, > 9 million cells / mL, > 10 million cells / mL, > 11 million cells / mL, > 12 million cells / mL, > 13 million cells / mL, > 14 million cells / mL, > 15 million cells / mL, > 16 million cells / mL, > 17 million cells / mL, > 18 million cells / mL, > 19 million cells / mL, > 20 million cells / mL. In some embodiments, the population of cells according to the present invention has a density of approximately 20 million cells / mL.

[0216] In some embodiments, the population of cells according to the present invention contains fewer than 10 million cells, e.g. one of < 9 million cells, < 8 million cells, < 7 million cells, < 6 million cells, < 5 million cells, < 4 million cells, < 3.5 million cells, < 3 million cells, < 2.5 million cells, < 2 million cells, < 1.5 million cells, < 1.4 million cells, < 1.3 million cells, < 1.2 million cells, < 1 million cells, < 900,000 cells, < 750,000 cells, < 500,000 cells.

[0217] In some embodiments, the population of cells according to the present invention contains at least 0.1 million cells, e.g. one of > 0.2 million cells, > 0.3 million cells, > 0.4 million cells, > 0.5 million cells > 0.6 million cells, > 0.7 million cells, > 0.8 million cells, > 0.9 million cells, > 1 million cells, > 1.1 million cells, > 1.2 million cells.

[0218] In some embodiments, the population of cells according to the present invention contains fewer than 5 million cells. In some embodiments, the population of cells according to the present invention contains approximately 1.2 million cells. In some embodiments, the population of cells according to the present invention contains approximately 500,000 cells.

[0219] A subject in accordance with the present invention may be any animal or human. The subject is preferably mammalian, more preferably human. The subject may be a non-human mammal but is more preferably human. The subject may be male or female. The subject may be a patient.

[0220] The subject may have (e.g. may have been diagnosed with), may be suspected of having, or may be at risk of developing, a disease / disorder wherein the identification of cell surface proteins in cells affected by said disease / disorder may have diagnostic or therapeutic benefit. “Diagnostic or therapeutic benefit” may refer to benefit for the subject, or benefit to the broader medical community.

[0221] The subject may have (e.g. may have been diagnosed with), may be suspected of having, or may be at risk of developing, a cancer. The cancer may be any cancer as described herein.

[0222] Kits

[0223] In a further aspect, the present invention also provides a kit comprising:

[0224] one or more amine-reactive biotinylation reagents; and

[0225] a carboxyl-reactive biotinylation reagent.In some embodiments, a kit according to the present disclosure comprises (i) an amine-reactive biotinylation reagent, as described herein, and (ii) a carboxyl-reactive biotinylation reagent, as described herein.

[0226] In some embodiments, a kit according to the present disclosure comprises (i) two amine-reactive biotinylation reagents, as described herein, and (ii) a carboxyl-reactive biotinylation reagent, as described herein.

[0227] Kits according to the present disclosure may comprise a predetermined quantity of articles according to (i) and / or (ii), as described in the preceding paragraph. In some embodiments, articles according to (i) and / or (ii) are provided in containers (e.g. in vials or bottles). In some embodiments, the kit may comprise (i) two amine-reactive biotinylation reagents, as described herein, each provided in a separate container (e.g. in vials or bottles), and (ii) a carboxyl-reactive biotinylation reagent, as described herein, provided in a container (e.g. in vials or bottles). In some embodiments, the kit may comprise (i) two amine-reactive biotinylation reagents, as described herein, each provided in a predetermined quantity in separate containers (e.g. in vials or bottles), and (ii) a carboxyl-reactive biotinylation reagent, as described herein, provided in a predetermined quantity in a container (e.g. in vials or bottles). The kit may provide articles according to (i) and / or (ii) together with instructions (e.g. a protocol) as to howto employ them in accordance with a method described herein.

[0228] In some embodiments, the kit further comprises reagents, buffers and / or standards required for execution of a method according to the present disclosure. Kits according to the present disclosure may include instructions for use, e.g. in the form of an instruction booklet or leaflet. The instructions may include a protocol for performing any one or more of the methods described herein.

[0229] In some embodiments, the components of a kit according to the present invention are arranged / packaged in such a way suitable for use in a method as described herein. A kit according to the present invention may also be referred to as a ‘SUCAM-kit’, e.g. a kit that is suitable for performing the ‘SUCAM’ method of the present invention.

[0230] The present invention also provides the use of a kit as described herein in any of the methods described herein. That is, the present invention provides the use of a kit as described herein in a method of labelling, identifying and / or characterising cell surface proteins in a population of cells, as described herein. The present invention also provides the use of a kit as described herein in a method for identifying one or more diagnostic markers or therapeutic targets, as described herein.

[0231] The manufacture of kits according to the present disclosure preferably follows standard procedures which are known to the person skilled in the art.Applications of the methods

[0232] In a further aspect, the present invention provides methods for identifying one or more diagnostic markers using the labelling methods described herein. In some embodiments, the present invention provides a method for identifying one or more diagnostic markers, the method comprising:

[0233] (i) providing a population of cells;

[0234] (ii) biotinylating primary amine groups on cell surface proteins within the population of cells; (iii) biotinylating carboxyl groups on cell surface proteins within the population of cells.

[0235] In preferred embodiments, the population of cells comprises cells affected by a disease or disorder (e.g. a disease or disorder as described herein). That is, the method for identifying one or more diagnostic markers is a method for identifying one or more diagnostic markers for diagnosing a disease or disorder, by using the labelling method described herein to label cell surface proteins on cells affected by said disease or disorder.

[0236] Diagnostic markers refer to any biological parameter that may help identify or confirm the presence of a disease or disorder (as described herein). A diagnostic marker according to certain aspects of the present invention may comprise a protein (e.g. a cell surface protein). Methods for identifying diagnostic markers according to the present invention may comprise identifying cell surface proteins in a population of cells affected by a disease / disorder. Cell surface proteins can act as useful diagnostic markers to identify the presence of cells affected by certain diseases or disorders (e.g. cancer cells), as they are located on the exterior of cells (and are therefore accessible for binding for detection by, for example, antibodies or probes). The methods of the present invention may be used to characterise the cell surface proteome of a population of cells affected by a disease / disorder, in order to identify cell surface proteins within such a population that may be useful as diagnostic markers. Such diagnostic markers identified using the method of the present invention may be used to stratify clinical sub-types of a disease.

[0237] In a further aspect, the present invention provides methods for identifying one or more therapeutic targets using the labelling methods described herein. In some embodiments, the present invention provides a method for identifying one or more therapeutic targets, the method comprising:

[0238] (i) providing a population of cells;

[0239] (ii) biotinylating primary amine groups on cell surface proteins within the population of cells; (iii) biotinylating carboxyl groups on cell surface proteins within the population of cells.

[0240] In preferred embodiments, the population of cells comprises cells affected by a disease or disorder (e.g. a disease or disorder as described herein). That is, the method for identifying one or more therapeutic targets is a method for identifying one or more therapeutic targets for treating or preventing a disease or disorder, by using the labelling method described herein to label cell surface proteins on cells affected by said disease or disorder.

[0241] Therapeutic targets refer to specific molecules that can be targeted in order to treat or prevent diseases or disorders (as described herein). Therapeutic targets may be molecules that distinguish cell populations affected by a disease or disorder from healthy cell populations, in order to enable specific targeting of cells of the disease or disorder. Therapeutic targets may be molecules that play a role in thedevelopment, progression or maintenance of a disease or disorder. A therapeutic target according to certain aspects of the present invention may comprise a protein (e.g. a cell surface protein). Methods for identifying therapeutic targets according to the present invention may comprise identifying cell surface proteins in a population of cells affected by a disease / disorder. Cell surface proteins can act as useful therapeutic targets, to specifically target cells affected by certain diseases or disorders (e.g. cancer cells), rather than healthy cells, as they are located on the exterior of cells (and are therefore accessible for binding for detection by, for example, antibodies or probes). The methods of the present invention may be used to characterise the cell surface proteome of a population of cells affected by a disease / disorder, in order to identify cell surface proteins within such a population that may be useful as therapeutic targets.

[0242] In some embodiments, a therapeutic or diagnostic target is expressed at higher levels in diseased samples compared to control samples. That is, a therapeutic or diagnostic target has increased abundance in diseased samples compared to control ( / .e. healthy) samples. In some embodiments, the therapeutic or diagnostic target is expressed in cells derived from diseased samples at greater than 1 times e.g. one of >1.01 times, >1.02 times, >1.03 times, >1.04 times, >1.05 times, >1.06 times, >1.07 times, >1.08 times, >1.09 times, >1.10 times, >1.11 times, >1.15 times, >1.20 times, >1.25 times, >1.3 times, >1.4 times, >1.5 times, >1.6 times, >1.7 times, >1.8 times, >1.9 times, >2 times, >3 times, >4 times, >5 times compared to the expression in equivalent cells derived from healthy samples. In some embodiments, a therapeutic or diagnostic target has greater than 1-times e.g. one of >1.01 times, >1.02 times, >1.03 times, >1.04 times, >1.05 times, >1.06 times, >1.07 times, >1.08 times, >1.09 times, >1.10 times, >1.11 times, >1.15 times, >1.20 times, >1.25 times, >1.3 times, >1.4 times, >1.5 times, >1.6 times, >1.7 times, >1.8 times, >1.9 times, >2 times, >3 times, >4 times, >5 times abundance in diseased samples compared to equivalent healthy samples. In some embodiments, a therapeutic or diagnostic target has at least 10% higher abundance, e.g. one of >20%, >30%, >40%, >50%, >60%, >70%, >80%, >90%, >100%, >200%, >300%, >400%, >500% higher abundance in diseased samples compared to equivalent healthy samples ( / .e. control samples).

[0243] In some embodiments, a therapeutic or diagnostic target is elevated compared to equivalent cells from a healthy sample in cells from a diseased sample. In some embodiments, a therapeutic or diagnostic target is elevated compared to equivalent cells from a healthy sample in at least 10%, e.g. one of >20%, >30%, >40%, >50%, >60%, >70%, >80%, >90%, >99%, or 100% of cells from a diseased sample. In some embodiments, a therapeutic or diagnostic target is elevated compared to equivalent healthy samples in at least 10%, e.g. one of >20%, >30%, >40%, >50%, >60%, >70%, >80%, >90%, >99%, or 100% of diseased samples.

[0244] In some embodiments, a therapeutic or diagnostic target has at least 10% higher abundance, e.g. one of >20%, >30%, >40%, >50%, >60%, >70%, >80%, >90%, >100%, >200%, >300%, >400%, >500% higher abundance in diseased samples compared to equivalent healthy samples ( / .e. control samples), and is elevated compared to control samples in at least 10%, e.g. one of >20%, >30%, >40%, >50%, >60%, >70%, >80%, >90%, >99%, or 100% of diseased samples.In some embodiments, the therapeutic target is identified as a candidate target for an antibody drug conjugate therapy. In some embodiments, the therapeutic target is identified as a candidate target for a CAR-T cell therapy.

[0245] ***

[0246] Aspects and embodiments of the present invention will now be discussed with reference to the accompanying figures. Further aspects and embodiments will be apparent to those skilled in the art. All documents mentioned in this text are incorporated herein by reference.

[0247] The features disclosed in the foregoing description, or in the following claims, or in the accompanying drawings, expressed in their specific forms or in terms of a means for performing the disclosed function, or a method or process for obtaining the disclosed results, as appropriate, may, separately, or in any combination of such features, be utilised for realising the invention in diverse forms thereof.

[0248] While the invention has been described in conjunction with the exemplary embodiments described above, many equivalent modifications and variations will be apparent to those skilled in the art when given this disclosure. Accordingly, the exemplary embodiments of the invention set forth above are considered to be illustrative and not limiting. Various changes to the described embodiments may be made without departing from the spirit and scope of the invention.

[0249] For the avoidance of any doubt, any theoretical explanations provided herein are provided for the purposes of improving the understanding of a reader. The inventors do not wish to be bound by any of these theoretical explanations.

[0250] Any section headings used herein are for organizational purposes only and are not to be construed as limiting the subject matter described.

[0251] Throughout this specification, including the claims which follow, unless the context requires otherwise, the word “comprise” and “include”, and variations such as “comprises”, “comprising”, and “including” will be understood to imply the inclusion of a stated integer or step or group of integers or steps but not the exclusion of any other integer or step or group of integers or steps.

[0252] It must be noted that, as used in the specification and the appended claims, the singular forms “a,” “an,” and “the” include plural referents unless the context clearly dictates otherwise. Ranges may be expressed herein as from “about” one particular value, and / or to “about” another particular value. When such a range is expressed, another embodiment includes from the one particular value and / or to the other particular value. Similarly, when values are expressed as approximations, by the use of the antecedent “about,” it will be understood that the particular value forms another embodiment. The term “about” in relation to a numerical value is optional and means for example + / - 10%.Examples

[0253] Example 1 : Materials and Methods

[0254] Biotinylation of cell surface using NH2 Labelling and COOH labelling sequentially

[0255] Cells are washed twice with PBS+ (PBS supplemented with 0.1 mM CaCL, 1 mM MgCh) pH 7.4 and once with PBS+ pH 8.0 and then incubated with 2 mM Sulfo-NHS-SS-biotin (21331 ; Pierce) and 2 mM EZ link NHS-PEG4 biotin (21330; Pierce) in PBS+ pH 8.0 for 30 min at 4°C with gentle shaking. As a mock pull-down, cells are initially treated the same except that the 2 mM treatment with biotin 1 and biotin 2 was replaced with vehicle (PBS). Cells were washed once with PBS pH 7.4, and twice with in PBS pH 8.0 (to wash off residual biotin).

[0256] To proceed with carboxyl labelling, cells were washed with PBS+ pH 5.7 and then preincubated with 2 mM EDC and 5 mM NHS-Sulfo in PBS pH 5.7 for 15 min at4°C with gentle shaking. After preincubation, cells were washed with PBS+ pH 8.0 and subsequently incubated with 2 mM carboxyl reactive biotin (EX-Link-Amine_PEG2 Biotin (21346; Thermo) in PBS pH 8.0 for 40 min at 4°C with gentle shaking. As a mock pull-down, cells are treated the same except the 2 mM treatment with Biotin is replaced with PBS. After incubation with Amine PEG2 biotin, residual biotin was quenched with 100 mM glycine in PBS pH 8.0 and then washed twice with PBS pH 7.4.

[0257] Isolation of Biotinylated Proteins

[0258] Cells were lysed in modified RIPA buffer (50 mM Tris-HCI, 150 mM NaCI, 1 mM EDTA, 10 mM iodoacetamide, 0.2% SDS (v / w), 0.5% sodium deoxycholate, 1% Triton X-100, pH 8.0) containing protease inhibitors. Cell lysates were centrifuged at 15,000 x g for 5 min at 4 °C, the nuclear pellet was discarded, and supernatants were incubated with preconditioned Streptavidin Magnetic Beads (8817; Pierce) for 3 hr to overnight at 4°C. Unbound samples were collected, and beads were washed as follows; x3 RIPA lysis buffer, x3 in Urea buffer (8 M urea in 0.1 M Tris / HCI pH 8.5), x2 in 2M NaCI, x3 with 50 mM Ammonium bicarbonate (AMBIC), x1 with 25 mM AMBIC. Finally, the biotinylated proteins were digested on-beads at 37°C with 1 pg sequencing grade trypsin (Pierce) overnight in 25 mM ammonium bicarbonate. The eluted digested peptides were harvested (clear supernatant without the beads). The beads were then rinsed with 50 pl of 25 mM ammonium bicarbonate, and this second tryptic fraction was pooled with the first one.

[0259] Sample preparation for Proteomics

[0260] The samples were acidified to 1% TFA and peptide desalting was performed by reverse-phase liquid chromatography on an Agilent Bravo Automated Liquid Handling Platform, using the peptide clean-up v3.0 protocol. Briefly, reverse phase S cartridges 12 (Agilent, 5 pL bed volume) were primed with 250 pL 99.9% acetonitrile (ACN) with 0.1%TFA 13 and equilibrated with 250 pL of 0.1% TFA at a flow rate of 10 pL / min. The samples were loaded (770 pL) at 20 pL / min, followed by an internal cartridge wash with 250 pL of 0.1% 15 TFA at a flow rate of 10 pL / min. Peptides were then eluted with 105 pL of 1M glycolic acid with 50% ACN, 5% TFA. After desalting, 150pl of the elute is transferred to Lo-bind tubes and subjected to SpeedVac used to concentrate the samples overnight and then stored at -80°C.Mass Spectrometry

[0261] Peptides were re-suspended in 20 pL of reconstitution (97% H2O, 3% ACN, 0.1% TFA, 850fmol / pl-1 enolase peptide digest) and sonicated for 5 minutes at RT. Following a brief centrifugation, 4 pl was loaded onto a LC-MS / MS system. The LC-MS / MS platform consisted of a Dionex UltiMate 3000 RSLC coupled to Q Ex active™ Plus Orbitrap Mass Spectrometer (Thermo Fisher Scientific) through an EASY-Spray source operated as described before (25, 26).

[0262] Protein identification and quantification

[0263] For protein identification the MS data were searched using Mascot Server via Mascot Daemon (v2.8.0.1 , in-house server) which matched the MS / MS data to the Swissprot Database for Homo sapiens 2021_02.fasta (17080 sequences, 181677051 residues). The FDR was set to ~1 % (peptide matches above homology or identity threshold). Searches had the following parameters: 2 trypsin missed cleavages, mass tolerance of ±10 ppm for the MS scans and ±25 mmu for the MS / MS scans, carbamidomethyl Cys as a fixed modification and oxidation of Met, PyroGlu on N-terminal Gin and deamidated (NQ) as variable modifications.

[0264] Pescal (Peak Statistic Calculator) was used to calculate m / z and retention time (tR) values for each identified peptide ion to generate extracted ion chromatograms (XICs). The software constructed XICs for all the peptides identified in at least one of the LC-MS / MS runs across all samples. XIC mass and retention time windows were ±7 ppm and ± 1.5 min, respectively. Algorithms in Pescal then calculate area under the peak of the XICs in the generated arrays and return these values to Excel file for further analysis. The relative quantity of a peptide was calculated relative to the mean of peptide normalized ion intensities of this peptide across the samples to be compared. For the reported differentially expressed proteins at least one unique peptide assigned to them was also required. The returned protein identifications were accepted when Mascot scores were above the statistically significant threshold (expectancy < 0.05 for individual peptides) and at least two peptides matched the identified protein.

[0265] Bioinform atic analysis

[0266] A bioinformatics pipeline that incorporates R algorithms for pathway analysis and interactive visualisations (https: / / github.com / CutillasLab / protools2 / ) was used for the quantitative processing and GO enrichment analysis of the proteomic data against the location subset of ontologies or the in silico surfaceome obtained from Bausch-Fluck etal (27). Peak area normalisations were performed and then Iog2 scaled and cantered. Statistical differences between comparisons were calculated using the LIMMA function, and then adjusted for FDR using the Benjamini-Hochberg method. Proteins were considered significant if present in labelled samples with Log2fold difference > 0.5 and p-value < 0.05 compared to the mock labelled control.

[0267] Example 2: Overall method design and analytical strategy

[0268] Surfaceomic approaches, based on biotinylation-based enrichment coupled with mass spectrometry proteomics, are now routine in many laboratories. However, there is a scarcity of published systematic characterization or comparative analysis of the performance of such techniques. Paramount tosurfaceomic methods is that an effective enrichment requires specific isolation of proteins on the cell surface whilst minimising non-specific biotinylation, enabling a precise assessment of the surfaceome.

[0269] Analytical factors such as cell density, nature of functional groups in reagents and their concentration and washing conditions contribute to the performance of affinity chromatography and thus on the efficacy of surfaceomic methods based on biotinylation enrichment. Thus, in this work, as aim is to advance existing cell-surface protein enrichment methods by developing a complementarily based biotin derivatisation strategy based on multiple-functional group labelling coupled with MS-based proteomics. The proposed approach - named Surfaceome Capture by Multiplex (SUCAM) biotinylation, schematically shown in Figure 1 - consists of simultaneous and / or sequential labelling of protein functional groups with biotin derivatives, with the aim of circumventing some of the existing limitations of techniques based on single biotin derivatisation of the cell surface proteome.

[0270] To support the development of SUCAM, a label-free single-shot LC-MS / MS based analytical platform was employed which overcomes the issue of missing data points in standard data-dependent analysis (DDA) proteomic strategies (28, 29). This analytical platform included a software for label-free quantification of the peptide ion intensities recorded in the MS data collected with DDA scanning. The MS1 signal for a given peptide is integrated for each peptide across all samples being compared and quantification achieved by considering the area under the curve of the extracted ion chromatogram. Relative peptide quantifications can be directly compared across different samples to obtain relative peptide abundances and thus measure differences between samples.

[0271] An R script was used to automate the statistical analysis by adding the intensities of individual peptides that belong to a given protein. The mean abundances between biotinylated and mock labelled samples were then compared, the mock labelled samples being cells treated with vehicle (instead of the respective biotin reagent) and represented proteins bound un-specifically to the magnetic streptavidin beads.

[0272] Proteins that show significant enrichment (adjusted p-value <0.05 and fold difference >0.5) in the experimental group were then further evaluated by Gene Ontology enrichment analysis (GOEA), focusing on subcellular locations relevant to the cell surface and a set of proteins found by machine learning to be likely located on the cell surface (obtained from Bausch-Fluck etal (27)).

[0273] Example 3: Impact of cell density on surfaceome enrichment

[0274] The impact of cell densities on assay performance and nature of the isolated surfaceome was first investigated. N-Hydroxysuccinimide (NHS) esters, employed for this optimisation were chosen as useful labelling reagents due to their relatively fast labelling reaction under physiological conditions, and have been employed before in several studies (30). Labelling of cell surface proteins was carried out at densities ranging from 2 to 20 million cells / mL (Figure 3). It was found that a cell density of 20 million cells / mL resulted in optimal labelling and enhanced the efficacy of cell surface and membrane protein enrichment. This is evidenced in the greater enrichment of cell surface ontology using GOEA (Fig 3A, panel i), as well increased ratio in the number of cell surface-associated proteins relative to cytoplasmic proteins (Fig 3A, panel ii). For a cell density of 20 cells / mL, the ontologies related to “External side ofplasma membrane” and “cell surface” were 2 out of 5 top-enriched GO terms. This is further illustrated in a GO network with corresponding proteins that are increased in the biotin fraction relative to mock pulldown (Figure 11).

[0275] Example 4: Impact of cell numbers on surfaceome enrichment

[0276] Having established that a cell density of 20 million cells / mL is optimal for the isolation of cell surface protein by biotinylation, it was next investigated whether it would be possible to increase the sensitivity of the assay by lowering the number of cells but maintaining their density. This is a critical question in order to develop a sensitive assay applicable to real world clinical samples where tissue availability is limited. The efficacy of membrane proteome enrichment at different volumes of cell suspension was therefore assessed, ranging from 1 mL to 0.06 mL whist maintaining the optimum density, i.e. 20 million cells / mL (Fig 3B). Cell surface and membrane proteome analysis on 0.06 mL cell suspension at 20 million cells / mL cell density, corresponding to 1.2 M cells, produced an approximately similar cell surface to cytoplasmic ratio as the proteome analysis based on a 2 M, 5 M and 10 M cells (Fig 3B, panel ii). It was therefore concluded that 1.2 M cells at a density of 20 million cells / mL provide adequate conditions, with the added advantage of being able to scale down the assay to a manageable number of cells, useful for the analysis of primary material.

[0277] Example 5: Efficiency of cell surfaceome enrichment by different amine-reactive biotinylating reagents Despite the efficient enrichment of the cell surface at 1.2 M cells, several unspecific intracellular proteins were detected and identified in biotinylated samples (Figure 11), irrespective of the analytical method applied. Although non-specific interactions of some proteins with streptavidin-beads, resistant to the stringent washings performed cannot be ruled out, control experiments performed omitting the biotinylation step indicate that these interactions alone cannot account for the detection of many intracellular proteins in the biotinylated samples. Other possible explanations can be considered: (i) intracellular proteins are released by damaged cells prior or during cell surface biotinylation and can be labeled during the biotinylation reaction; (ii) despite being negatively charged, the biotinylation reagent can penetrate into the cells either by diffusion or by means of a specific transporter (as for instance the sodium dependent multivitamin transport system (8); (iii) cytoplasmic proteins are copurified by virtue of a strong interaction with membrane and cytoskeletal proteins, which hinder identification of low abundant cell surface proteins.

[0278] To investigate if different biotinylating reagents have differential labelling properties, the reactivity of the two most common amine-reactive biotinylation reagents was evaluated: NHS-PEG4-Biotin and the Sulfo-N-hydroxysuccinimide (NHS) ester Sulfo-NHS-Biotin. It was found that the use of the Sulfo-NHS-Biotin reagent returned a larger number of cell surface proteins compared to NHS-PEG4-Biotin (Fig 4A). For p31 / Fuji cells, despite a significant percentage (40%) of identifiable proteins over-lapping between the two agents, a large percentage (46%) of unique proteins were captured with Sulfo-NHS-Biotin, whilst the percentage of unique proteins captured with NHS-PEG4-Biotin was only 14% (Fig 4B). As the aim was to increase the density of the biotin label per proteins, it was reasoned that by combining the two agents, thecoverage of the identified cell surface proteome could potentially be expanded. To test this notion, experiments were carried out using a combination of both Sulfo-NHS-Biotin and NHS-PEG4-Biotin regents. In an initial experiment, this double labelling approach returned 7 unique proteins not detected by single labelling approaches, whereas just 0 and 3 proteins were uniquely identified when using single reagents (Fig 4C). These results suggested that a combination of biotinylating reagents would increase the repertoire of identifiable cell surface proteins.

[0279] Example 6: Efficiency of cell surfaceome enrichment by amine- and carboxyl- biotinylating reagents The observation that different biotinylating reagents return different repertoire of cell surface proteins led to the hypothesis that labelling additional protein reactive groups would increase this coverage further. A carboxyl-reactive biotinylation strategy was therefore tested as an orthogonal approach for amine-reactive biotinylation (Fig 5). Furthermore, it was also investigated whether more stringent washing of beads with 2M NaCI after labelling would eliminate non-specific proteins adsorbed to the beads. When comparing the efficacy of carboxyl, Sulfo-NHS-biotin and NHS-PEG4-Biotin at isolating cell surface associated terms, it was observed that carboxyl labelling was more efficient at isolating the cell surface than Sulfo-NHS-Biotin (as judged by GO enrichment analysis, Fig. 5A), and there was an observable improvement in isolation of the cell surface proteome using stringent immunoaffinity conditions. In support of these data, the cell surface to cytoplasmic ratio was highest for carboxyl labelling using stringent affinity purifications conditions and a much lower ratio was observed for amine labelling (Fig 5B).

[0280] Example 7: Carboxyl and amine labelling capture different surfaceomes

[0281] To test if amine and carboxyl labelling methods captured different sets of surfaceome-proteomes, the proteins identified with each technique were compared. As Fig 5C shows, 45% of identifiable cell surface proteins were unique to Sulfo-NHS-Biotin and 12% unique to the Carboxyl labelling, with an overlap of 44% proteins captured by both amine and carboxyl labelling. The percentage of cytoplasmic proteins unique to Sulfo-NHS biotin was 52%, whilst contamination with cytoplasmic proteins was significantly lower for carboxyl labelling. This suggests that a combination of unique cross-linker reactive biotin agents may enable a more efficient isolation of cell surface proteins.

[0282] Example 8: Development of a surfaceome enrichment method based on sequential carboxyl and amine labelling capture

[0283] Following the finding that a combination of biotinylating reagents isolated a unique repertoire of proteins, coupled with differing efficiencies for cell surface enrichment, it was reasoned that adding a combination of double amine and carboxyl labelling compound would increase the coverage of the identified surfaceome further. The logic for this approach is that increasing the variety of functional groups that are labelled on the cell surface would increase the density of labels per protein, increasing cell surface enrichment whilst widening the physiochemical properties of the identified proteins.

[0284] To test the validity of these assumptions, an additional comparison of the performance of five different cell surface biotinylation methods was carried out using single amine reactive biotin derivatives, combinationof two amine-reactive conjugates, carboxyl reactive conjugation and a newly developed “multiplexed” procedure combining a formulation of double amine and carboxyl reactive reagents in a single reaction. The latter approach was named SUCAM (Surfaceome Capture by Multiplex biotinylation). These modified cell surface labelling strategies were compared using a reduced quantity of starting material (1.2 M cells per reaction) in 4 cell lines and pooled AML cell lines.

[0285] For pool AML cells, consisting of P31-FUJ, HL60 and NB4, significant enrichment for “Cell Surface G0;0009986” proteins (p-value < 2.7 x 10-6) was observed for SUCAM (Fig. 6A). Both carboxyl and SUCAM led to improved cell surface-associated protein enrichment, enabling a precise assessment of the subcellular proteome associated with the plasma membrane. Consistently, the ratio of proteins belonging to ontologies associated to the cell surface (Cell periphery, Cell Surface, Plasma Membrane and Surfaceome) relative to cytoplasmic proteins were also found to be higher for labelling strategies using either just carboxyl or SUCAM ( / .e., double amine plus carboxyl, Fig 6B).

[0286] However, while enrichment of proteins with ontologies associated to the cell surface was similar with either just carboxyl or SUCAM (Fig 6B), the absolute number of proteins in such ontologies was higher for double amine plus carboxyl (Fig 7). In addition, when assessing the numbers and overlaps of cell surface proteins across the tested methods, just 27% overlap of all cell surface associated proteins across all the five methods was found (Fig. 8A), with the highest percentage (7%) of unique proteins captured by SUCAM. Sample distance heatmap shows the strongest dissimilarity between SUCAM and other four labelling methods whereas double amine, Sulfo-NHS and PEG4-biotin method had more significant correlations (Fig. 8B).

[0287] Analysis of plasma membrane-associated GO terms annotated to cell adhesion, cell-cell junction, cellsubstrate junction led to the identification of 811 proteins for the SUCAM method (Figure 12), which represented 45% of the global proteome analysis in the biotinylated vs control fraction and was superior to single amine, double amine and carboxyl.

[0288] Example 9: SUCAM produces the highest yield of the surfaceome out of all methods tested

[0289] So far, the performance of the different methods has been evaluated by counting how many proteins annotated with ontologies related to cell surface localization were identified by each of the labelling strategies. As an additional readout to assess the efficacy of the different labelling methods tested, the abundances of proteins that belong to each such subcellular location was also evaluated (measured as the sum of XICs of the different peptides that belong to a given protein). It was found that the mean normalised abundances of proteins annotated to a number of cell surface-associated GO terms including proteins annotated to the Cell Surface G0;0009986, External side of plasma membrane G0;0009897 and surfaceome by ML were significantly higher (p=0.018, p=0.0012 and p=0.0013, respectively by Kruskal-Wallis test) across four AML cell lines for SUCAM (Fig. 9). Further support for the superiority of this method is shown in Figure 13A, which depicts z-score normalised heatmap for POOL AML. Similarly, Figure 13B shows higher intensity of proteins isolated by SUCAM at enriching and isolating cell surface-associated proteins for POOL AML, FUJI and HL60 cell lines, with a higher expression and a wider coverage of cell surface proteins.

[0290] Example 10: Qualitative and quantitative nature of SUCAM across AML cell lines

[0291] To investigate the performance of the final SUCAM method, it was applied to the detection and quantification of cell surface proteomes in seven AML cell lines. After differential expression analysis, 1434 proteins were identified with ontologies relating to the cell surface or plasma membrane.

[0292] To examine its quantitative nature, SUCAM experiments were carried out in biological replicates (four independent replicates for the P31 / Fuj, HL-60, MOLM13, NOMO1, and NB4 and duplicates for Kasumi-1 and OCI-AML2). Assessment of replicates allowed calculation of the reproducibility (precision) of the quantitative method, which was evaluated by calculating coefficient of variations (CVs, i.e., standard deviation divided by mean) for individual proteins across replicates. It was found that median CVs were below 20% for 5 out of 7 cell lines analysed, with an overall median precision of 17% (range between 15% and 30%, Figure 10A).

[0293] To explore the nature of the most abundant cell surface proteins in these cell models, the quantitative values (sum of chromatographic peak areas of peptides belonging to a given protein) were divided by the molecular weight of the respective protein. This allows normalizing quantitative values for differences in the number of peptides derived from proteins of different length. The top 30 more abundant proteins in these models, shown in Figure 10B, include well known cell surface proteins in AML - such as CD43 (LEUK), CD50 (ICAM3), CD44, CD33 - but also other less well-known polypeptides with potential unexplored roles in this disease, including BASI (CD147, Basigin), AT1B3 (CD298) and RPN1 (Ribophorin-1).

[0294] Overall, these data show that SUCAM enriches and quantifies the surfaceome with quantitative precision and with a breath that allows discovering potential new drug targets.

[0295] Example 11 : Application of SUCAM to the analysis of solid tumour cell lines

[0296] To test the performance of SUCAM for the identification of cell surface proteins from cells other than those derived from leukaemia, the SUCAM method was compared to other approaches using the LS174T cell line, which is derived from colorectal adenocarcinoma and has an adherent epithelial-like phenotype.

[0297] Cells were detached using collagenase followed by surfaceome biotinylation using the different approaches outlined above, purification using streptavidin and label-free mass spectrometry for protein identification and quantification. The SUCAM method (a.k.a., double amine plus carboxyl) identified 440 proteins with ontologies associated to locations on the cell surface (Figure 14A), whereas enrichment with double amine identified 425 proteins with these ontologies. Single labelling methods with single biotinylation reagents - namely, the carboxyl labelling EX-Link-Amine PEG2, or the amine labelling NHS-PEG2 orsulfo-NHS - identified 252, 377 and 368 proteins, respectively (Figure 14A).There was variability in the performance of the different methods for individual proteins. For certain cell surface proteins, such as ABCC1 , ENPP4, SLC1 A5, SLC33A1 , SCL3A2 and SLC7A1 , to name some examples, SUCAM produced higher yields than other methods (Figure 14B). For other proteins, such as CD44, SUCAM was not superior to the double amine method, although the difference in intensities between the two approaches was small.

[0298] However, overall, comparing the distribution of intensities of proteins with surfaceome annotation revealed that SUCAM produced higher mass spectrometric intensities for this protein class than when using other methods (Figure 14C).

[0299] Example 12: Materials and methods for Examples 13 to 20

[0300] Enzyme (Collagenase) Dissociation from adherent cells and / or solid tumours into single-cell suspension Buffers and Reagents

[0301] • 10X Collagenase / Hyaluronidase in Dulbecco’s Modified Eagle’s Medium (DMEM) for the enzymatic dissociation of tissue (Catalog #07912)

[0302] Contains:

[0303] 3000 U / mL Collagenase

[0304] 1000 U / mL Hyaluronidase

[0305] DMEM (1000 mg Dglucose / L)

[0306] • DNAse I solution (1mg / ml, Catalog #07900)

[0307] • RPMI 1640 Medium (Catalog #36750)

[0308] • Ammonium Chloride Solution (Catalog #07800)

[0309] • PBS (free of Mg and Calcium) + 2% FBS + 1 mM EDTA

[0310] Immediately before use, prepare digestion medium

[0311] • Prepare 20 ml of digestion medium by combing the following;

[0312] 1 ml of Collagenase

[0313] 150 ul DNAse I solution (prevents clumping by degrading free DNA released from damaged cells)

[0314] 8 ml RPMI 1640

[0315] Mix thoroughly and warm to room temperature (15-25°C)

[0316] Dissociation Procedure:

[0317] 1. Remove old media and wash the cells with a balanced salt solution that lacks calcium and magnesium (like HBSS without Ca / Mg) or use an EDTA solution. Gently rock the flask for a minute or two to rinse the cells, then discard the wash solution

[0318] 2. Add 5 ml of digestion medium to a corning flask of adherent cells.

[0319] 3. Incubate at 37 C for ~5-10 minutes on a shaking platform (periodically check the cells under a microscope to ensure they have detached and appear rounded but are not damaged. Overincubation can harm the cells).

[0320] 4. Transfer the single cell suspension into a cell strainer.

[0321] 5. Rinse the strainer with PBS (free of Ca and Mg) containing 2%FBS and 1 mM EDTA6. Top up the tube to 50 ml with PBS + 25 FBS + 1 mM EDTA

[0322] 7. Centrifuge at 300 xg for 10 mins at RT with the break on low. Remove and discard supernatant.

[0323] 8. For Tissue dissociation only; Add 10 ml of ammoinium chloride solution to the pellet. Incubate at RT for 5 minutes.

[0324] 9. Top up to 50 ml with PBS + 2% FBS + 1 mM EDTA.

[0325] 10. Centrifuge at 300 xg for 10 mins at RT. Remove and discard the SN

[0326] 11. Resuspend the cells at 20 x 10A6 cells / ml in PBS pH 7.4

[0327] Additional methods

[0328] Combination biotinylation of amino groups and carboxyl groups of surface-exposed proteins, isolation of biotinylated proteins, sample preparation for proteomics, mass spectrometry, protein identification and quantification, and bioinformatic analysis was carried out as described in Example 1.

[0329] Example 13: Development of SUCAM strategy for enhancing cell surface protein identification in solid tumours: Overall method design

[0330] Surfaceome Capture by Multiplex (SUCAM) biotinylation aims to circumvent some of the existing limitations of techniques based on single biotin derivatisation of the cell surface proteome. This approach was initially developed and implemented in haematological cancers and adapted to target discovery in low-input (fewer than 1 million cells) clinical sample and real-world sample types, including fresh, viably cryopreserved, and dissociated solid tissues whose analysis remain limited by technical constraints, including membrane integrity, inefficient labelling and high background.

[0331] In the developmental strategy, SUCAM was compared and benchmarked to five complementary surface enrichment strategies using solid tumour (Colorectal, HCC and breast cancer) derived cell lines to assess quantitative performance, sensitivity and selectivity for cell surface enrichment from a low input amount. To extend these findings to tissue-derived samples, dissociation conditions were optimized in adherent cell lines and in PDAC mouse tissue. Enzymatic dissociation enabled accurate cell surface proteome (CSP) profiling from fewer than 1 to 2 million cells. Ultimately the novel “multiplexing” method was applied to cryopreserved dissociated breast cancer cells from 27 donors.

[0332] The proposed approach is schematically shown in Figure 1.

[0333] As discussed, novel aspects of the proposed SUCAM method include:

[0334] • Multiplexing specific biotin derivatives for multiple functional group conjugation pulls down diverse cel surface proteins whilst improving enrichment efficiency

[0335] • Uses low cell numbers (1.5 million cells)

[0336] • Is automatable (using magnetic beads)

[0337] • Can be used to compare as many samples as desired (not tandem mass tag labelling)

[0338] • Provides absolute quantification in absolute units (copy numbers per cell) in an untargeted manner to aid target prioritization.Example 14: Impact of dissociation methods on surface proteomics

[0339] The most optimum dissociation method to obtain suspension cells from solid tumour cell lines for successful implementation of the approach in primary tumours was investigated. The effect that enzyme digestion of cells may have on the repertoire of cell surface proteins available for analysis by SUCAM was observed. A fundamental consideration is that this method relies on the cellular membrane remaining intact during dissociation, hence the choice of dissociation agent is important to minimize degradation of surface markers during dissociation.

[0340] Dissociating agents that were applied to adherent colorectal cell line included Trypsin, Accutase, Collagenase, Dispase, Accumax, and non-enzymatic agents such as EDTA. The most optimum dissociation conditions at enriching and isolating the cell surface proved to be enzymatic methods, namely Accutase and Collagenase, followed by less effective proteolytic enzymatic Trypsin. Non-enzymatic conditions were least effective (Figure 15A and B). These reagents work by chelating calcium to prevent cadherins from attaching to the surface or one another, releasing cells from these surfaces during the culturing process.

[0341] Accutase, the most optimum at enriching and isolating the cell surface is an enzyme mixture with proteolytic and collagenolytic activities. Accutase is generally less damaging agent than trypsin, preserves cell surface epitopes, minimizing degradation of surface markers during dissociation., critical for downstream proteomic applications for complex solid tissue and requires no inactivation step that can lead to clumping.

[0342] To evaluate the feasibility of cell dissociation with accutase, the analysis of labelled proteins directly on cell lines growing in monolayers was compared with those that are first enzyme dissociated before labelling, Figure 16. Biotinylation of monolayer of cells with Sulfo_NHS_B iotin and NHS_PEG4_Biotin returned a lower significance of enrichment and isolated lower absolute number of proteins in the named ontologies.

[0343] Example 15: Application of SUCAM to the analysis of colorectal cell lines

[0344] The efficacy of Surfaceome Capture by “multiplex” biotinylation strategy was compared with single biotin derivatization strategies for capturing a larger, more abundant and diverse set of surface proteins. To test the performance of SUCAM for the identification of cell surface proteins in solid tumours, the method was compared to other approaches using the HCT116 cell line, which is derived from colorectal adenocarcinoma and has an adherent epithelial-like phenotype. Cells were detached using Accutase followed by surfaceome biotinylation using the different approaches labelling strategies were compared using a reduced quantity of starting material (1.2 x106cells per reaction) outlined above, purification using streptavidin and label-free mass spectrometry for protein identification and quantification.

[0345] Using GO enrichment analysis, SUCAM returned a greater significance of enrichment for “Cell Surface G0;0009986” proteins and plasma membrane than the other methods tested (Figure 17). Consistently,the ratio of proteins belonging to ontologies associated to the cell surface (Cell periphery, Cell Surface, Plasma Membrane and Surfaceome) relative to cytoplasmic proteins were higher for labelling strategies using either just SUCAM or double amine. A comparison of the five final biotin conjugation methods shows a higher number of plasma membrane, intrinsic component of plasma membrane, cell surface and cell surface by ML for double amine plus carboxyl. A Venn diagram showing the numbers and overlaps of the cell surface proteins demonstrates a superiority in Double amine plus carboxyl as the method isolates more unique proteins (10% of cell surface are unique). Normalised protein abundances for cell surface proteins relevant to CRC pathology show a higher abundance for double amine plus carboxyl as compared to the other biotin conjugation conditions.

[0346] As a proof-of-concept for the feasibility of SUCAM using collagenase dissociating agent, the five crosslinking biotinylation methods were compared in colorectal cell line HCT116. The most significant enrichment for the “cell surface”, plasma membrane and external side of plasma protein was observed for double amine plus carboxyl (SUCAM), Figure 18A. The SUCAM method (a.k.a., double amine plus carboxyl) identified 1000 proteins with ontologies associated to locations on the cell surface (Figure 18B (i)), whereas enrichment with double amine identified 600 proteins with these ontologies. Single labelling methods with single biotinylation reagents - namely, the carboxyl labelling EX-Link-Amine_PEG2, or the amine labelling NHS-PEG2 or sulfo-NHS - identified 800, 750 and 600 proteins, respectively (Figure 18B (i)). For the Surfaceome by ML, the number of proteins with unique repertoire in SUCAM was 38, representing the highest percentage (19%) of unique proteins among the five biotinylation methods (Figure 18B (ii))

[0347] Overall, comparing the distribution of intensities of proteins with surfaceome annotation revealed that SUCAM produced higher mass spectrometric intensities for this protein class than when using other methods (Figure 18C).

[0348] A bioinformatic pipeline was developed to perform quantification of protein by LC / MS / MS in absolute units, Figure 19A. To convert label-free LC-MS / MS signals into number of protein molecules an external standard method was used, where the signals of proteins of known molarity is used to construct calibration curves. Molarity can be converted to number of molecules using the Avogadro’s number after protein signals are normalized by their molecular weight. A computer program was developed in Python based on linear regression to quantify proteins in copy numbers per cell based on protein signals and calibration curves. Preliminary experiments using the Universal Dynamic range Standard mix (Sigma) showed that this method produces calibration curves with high correlation accuracy (Figure 19, R2>0.96).

[0349] Example 16: Application of SUCAM to the analysis of hepatocellular HCC cell lines

[0350] As shown in Figure 20, comparing the performance of five biotin conjugation strategies shows double amine plus carboxyl strategy to be superior at enriching for the cell surface and plasma membrane. A comparison of the five final biotin conjugation methods shows a higher number of plasma membrane, intrinsic component of plasma membrane, cell surface and cell surface by ML double amine pluscarboxyl. A Venn diagram showing the numbers and overlaps of proteins in the Surfaceome by ML ontology demonstrates a superiority in Double amine plus carboxyl as the method isolates more unique proteins (5% of cell surface are unique).

[0351]

[0352] of Breast Cancer cell lines

[0353] The performance of cell surface biotinylation methods was compared using single amine reactive biotin derivatives, combination of two amine-reactive conjugates, carboxyl reactive conjugation and a newly developed “multiplexed” procedure combining double amine and carboxyl reactive reagents in a single reaction. These modified cell surface labelling strategies were compared using a reduced quantity of starting material (1.2M cells per reaction) in one breast cancer cell line, MCF-7.

[0354] Significant enrichment for the “cell surface” was observed for both double amine plus carboxyl and double amine alone, (Figure 21 A (i)). The top-most enriched GO terms for the ‘multi-plexed’ biotin derivatising strategy show a superiority for double amine plus carboxyl at enriching for the Plasma Membrane, external side of plasma membrane and Integral component of plasma membrane. Double amine plus carboxyl significantly reduced non-specific biotinylation events, including minimisation of cytoplasmic proteins, as exemplified by a higher ratio of cell surface proteins to cytoplasmic proteins (Figure 21 A (ii)). This leads to improved PMP enrichment and enabling a precise assessment of the subcellular proteome associated with the plasma membrane.

[0355] A comparison of the five final biotin conjugation methods shows a higher number of proteins for the plasma membrane, external side of plasma membrane, cell surface, Surfaceome by ML double amine plus carboxyl, Figure 21B (i). A Venn diagram showing the numbers and overlaps of the cell surface proteins demonstrates a superiority in Double amine plus carboxyl as the method isolates more unique proteins (4% of cell surface are unique), Figure 21 B (ii).

[0356] Label-free quantification (LFQ) abundances of biotin labelled cell surface proteins were compared across five conditions respective to the mock, unlabelled control. The normalised abundances of a number of cell surface proteins that are related to AML patho-physiology are highest for double amine plus carboxyl, Figure 21 C. Therefore, double amine plus carboxyl shows superiority in the expression of the biotinylated cell surface proteins, with a higher expression and a wider coverage of cell surface proteins quantifiable by LC / MS / MS.

[0357]

[0358] Weekes etal 2012 developed an aminooxy-biotinylation procedure for capturing sialylated PM glycoproteins, resulting in superior protein sequence coverage and improved PMP enrichment (18). Nonetheless, some disadvantages of glycan- and sialyl-based CSC include a requirement for a relatively large amount of starting material (nearly 108cells or 200 mg - 1 g of tissue), as well as the potential loss of glycan structures by oxidation, leading to low sensitivity.SUCAM based co-targeting of amines and carboxyl functional group and aminooxy-biotinylation for capturing glycoproteins were compared and benchmarked using solid tumour (HCC) cell lines, albeit starting from an input of 2.5 million cells and 20 million cells, respectively to assess quantitative performance, selectivity, and sensitivity. These comparative analyses aimed to assess the degree of cell surface protein coverage and selectivity for cell surface enrichment. Despite the different starting in-put of cells, SUCAM labelling and glycoprotein labelling returned similar enrichment of the cell surface, external side of plasma membrane, integral component of plasma membrane, plasma membrane region, Figure 22A (i).

[0359] It was found that the mean normalised abundances of proteins annotated to a number of cell surface-associated GO terms including proteins annotated to the Cell Surface G0;0009986, External side of plasma membrane G0;0009897 and surfaceome by ML were significantly higher (p=0.018, p=0.0012 and p=0.0013, respectively by Kruskal-Wallis test) for SUCAM (double amine plus carboxyl) than for alkoxyamine-biotin labelling of glycoprotein, (Figure 22C). The absolute number of proteins in such ontologies was higher, albeit not significantly for double amine plus carboxyl. The limitation of this observation is the difference in the input of starting material between the two methodologies, with alkoxyamine biotinylation requiring substantially larger input. Nonetheless, the advantage of SUCAM is that enables surfaceome characterization in translational settings where sample quantity and preservation methods are often limiting.

[0360] Example 19: Surface proteomic profiling of cryo-preserved single-cell suspensions from Breast Cancer tissue

[0361] Having demonstrated that SUCAM works on freshly dissociated cells from adherent cell lines, the performance of the methodology was tested in cryo-preserved breast cancer cells previously disaggregated from fresh tissue. To evaluate the performance of the new optimized workflow under clinically relevant constraints, 0.5 x 106cryopreserved cells isolated from breast cancer tissue were used. GO Enrichment Analysis was conducted on differentially expressed cell surface proteins in biotin labelled vs mock labelled single-cells dissociated from breast cancer tissue. The results establish the feasibility of robust surfaceome profiling from limited clinical material (Figure 23).

[0362] To systematically and comprehensively categorize the surfaceome of primary breast cancer, protein abundances for all primary breast cancer and healthy samples were subjected to z score normalisation. Figure 24A is a heatmap representing the Z-score of log-transformed absolute quantifications of surface proteins in breast cancer specimens from 27 breast cancer (BC) donors and Healthy Tissue (5 Epithelial, 5 Myoepithelial and 2 healthy fibroblast donors). Across all samples, surface proteomic analysis yielded high CSP identification rates (typically >400) of which 187 are significantly increased in Breast Cancer primary vs Healthy Tissue (Figure 24C). Principal component analysis further showed clear separation of biotin labelled vs unlabelled surface proteins and clear donor-specific separation emphasizing both the heterogeneity of breast cancer and sensitivity of the method to underlying biological variation (Figure 24B).To further delineate the heterogeneity of the surfaceome in breast cancer primary samples, reflecting intracellular differences and inter-sample heterogeneity, surfaceomic proteins differentially expressed in primary BC vs Healthy tissue were stratified into clinical BC subtypes classified according to ER, HER and PR status (Figure 25A). Using this approach, the surfaceomic targets’ subtype specificity was assessed.

[0363] Two promising therapeutic strategies in oncology are chimeric antigen receptor-T cell (CAR-T) therapies and antibody drug conjugates (ADCs). To be effective and safe, these immunotherapies require surface antigens to be sufficiently expressed in tumors and less or not expressed in normal tissues. The present aim was to identify new targets for ADCs and CAR-T specifically targeting breast cancer (BC) molecular and pathology-based subtypes, using the SUCAM methodology.

[0364] To maximise the utility of publicly available tissue sets for understanding tumour-normal differential and hence streamline individual target prioritisation and validation, surfaceome profiling in breast cancer will also include normal tissue, namely myoepithelial, epithelial tissue and fibroblasts.

[0365] Identification of specific TSAs that are sufficiently expressed in tumor cells and less or not expressed in normal tissue cells is of utmost importance to avoid ‘on-target, off-tumor’ effects, which can lead to severe toxicities. As a simple target selection process, a bioinformatic-based pipeline to select for pan-proteins favourable for ADC and CAR-T cells in breast cancer was proposed (Figure 26). A key criterion for an ADC target is high surface abundance and differential expression over normal tissue. A dotplot visualisation of the Log1 Omean absolute. quantification of the target vs proportion of BC primary patients whose expression of the target is above the mean. Absolute quantity across all differentially expressed surface proteins (primary BC vs Healthy tissue). Candidate proteins are expressed in >50% of BC samples, have a high copy number and have a high differential expression between primary BC and healthy tissue. Ranking this set of candidates % of cases high relative to control returned a set of targets that require triaging and validation.

[0366] Example 20: Cell surface proteomics for Tissue samples

[0367] Surfaceomic discovery methodologies compatible with single cells dissociated from solid tissue are increasingly important given that clinical and translational research largely depends on archived samples. However, unique challenges and limitations exist for proteomic analysis of cryopreserved specimens derived from patient tissues. Tissue dissociation can disrupt membrane integrity, reduce surface labelling efficiency, or introduce degradation artifacts. In addition, the limited quantities of these samples requires a methodology that is compatible with low input.

[0368] While single-cell and multi-omic studies have shown that disaggregation of cryo-preserved tissue might sustain tissue integrity, cell viability and surface marker expression, this has not been successfully implemented for analysis of surface proteomics that rely on membrane and efficient labelling for accurate quantification.To enable surface proteomic profiling of clinical lesions with limited input material, a dissociation workflow was optimized using cryo-preserved PDAC mouse tissue as a model system to ensure compatibility with downstream labelling and enrichment. A protocol was established that maximizes post-thaw viability and surface labelling compatibility while working within input constraints typically encountered in this clinical context.

[0369] PDAC mouse tissue was embedded in OCT medium. OCT (Optimal Cutting Temperature) embedding medium is a cryo-preservative that supports and stabilizes fresh tissue samples for cryostat sectioning. OCT is composed of a blend of glycols and resins that freezes rapidly, minimizes ice crystal formation, and provides a supportive matrix for creating thin, consistent sections with a cryostat. OCT is designed to freeze rapidly with minimal damage from crystal formation, which is crucial for preserving tissue morphology. Frozen OCT embedded PDAC tissue was cut into 15 x 15 um sections, and the sections were slowly thawed at 4 °C and then 10-15 °C. Protocol 2.1. Collagenase Dissociation of solid tumours into single-cell suspension was applied to these sections. Single cell suspensions were subjected to “multiplex” amine reactive biotinylation, purified on streptavidin magnetic beads, digested and LC / MS / MS based quantification returned 78 cell surface proteins increased in the biotinylated fraction vs mock control (Figure 27).

[0370] Conclusions

[0371] This work addresses two key gaps in the field of surfaceomic discover technology:

[0372] First, by providing a direct, head-to-head systematic comparison of low-input-compatible surface enrichment strategies including single amine, double amine, carboxyl, “multiplexing” double amine plus carboxyl (SUCAM) and alkoxyamine biotinylation in solid tumour derived cell lines the superior efficacy of SUCAM for capturing a larger, more abundant and diverse set of surface proteins was confirmed.

[0373] Second, the performance of SUCAM on cryopreserved and low-input specimens was validated and its applicability to clinically relevant tissues demonstrated.

[0374] Together, these advances broaden the applicability of “SUCAM” to clinically relevant, low-input samples. The optimized workflows enable robust surfaceome characterization in clinical settings where low sample quantities and dissociation methods pose significant challenges to proteomic analysis, opening new opportunities for immunotherapeutic discovery.

[0375] References

[0376] A number of publications are cited above in order to more fully describe and disclose the invention and the state of the art to which the invention pertains. Full citations for these references are provided below. The entirety of each of these references is incorporated herein.

[0377] 1. Tan, S., Tan, H. T., andChung, M. C. (2008) Membrane proteins and membrane proteomics Proteomics 8, 3924-3932 10.1002 / pmic.200800597

[0378] 2. Marijana Rucevic, D. H., Djuro Josie (2011) Mammalian plasma membrane proteins as potential biomarkers and drug targets Electrophoresis 32, 1549-1564,3. Sasso, J. M., Tenchov, R., Bird, R., Iyer, K. A., Ralhan, K., Rodriguez, Y. etal. (2023) The Evolving Landscape of Antibody-Drug Conjugates: In Depth Analysis of Recent Research Progress Bioconjug Chem 34, 1951-2000 10.1021 / acs.bioconjchem.3c00374

[0379] 4. Zhang, X., Zhang, H., Lan, H., Wu, J., andXiao, Y. (2023) CAR-T cell therapy in multiple myeloma: Current limitations and potential strategies Front Immunol 14, 1101495 10.3389 / fimmu.2023.1101495

[0380] 5. Overington JP, A.-L. B., Hopkins AL (2006) How many drug targets are there? Nat Rev Drug Discov 5, 993-996,

[0381] 6. Li, Y, Wang, Y, Mao, J., Yao, Y, Wang, K., Qiao, Q. et al. (2019) Sensitive profiling of cell surface proteome by using an optimized biotinylation method J Proteomics 196, 33-41

[0382] 10.1016 / j.jprot.2019.01.015

[0383] 7. Pornparn Kongpracha, P. W., Noriyoshi Isozumi, Satomi Moriyama, Yoshikatsu Kanai,, andNagamori, a. S. (2022) Simple But Efficacious Enrichment of Integral Membrane Proteins and Their Interactions for In-Depth Membrane Proteomics Mol Cell Proteomics 21,

[0384] 8. Leth-Larsen, R., Lund, R. R., andDitzel, H. J. (2010) Plasma membrane proteomics and its application in clinical cancer biomarker discovery Mol Cell Proteomics 9, 1369-1382

[0385] 10.1074 / mcp.R900006-MCP200

[0386] 9. Vilen, Z., Reeves, A. E., O'Leary, T. R., Joeh, E., Kamasawa, N., andHuang, M. L. (2023) Cell Surface Engineering Enables Surfaceome Profiling ACS Chem Biol 18, 701-710

[0387] 10.1021 / acschembio.1c00865

[0388] 10. Ravenhill, B. J., Soday, L., Houghton, J., Antrobus, R., andWeekes, M. P. (2020) Comprehensive cell surface proteomics defines markers of classical, intermediate and non-classical monocytes Sci Rep 10, 4560 10.1038 / s41598-020-61356-w

[0389] 11. Goswami, M. (2017) Novel Antigen Targets for Immunotherapy of Acute Myeloid Leukemia, 12. Kohnke, T. (2022) Integrated multiomic approach for identification of novel immunotherapeutic targets in AML,

[0390] 13. Sahin, C., Reid, D. J., Marty, M. T., andLandreh, M. (2020) Scratching the surface: native mass spectrometry of peripheral membrane protein complexes Biochem Soc Trans 48, 547-558 10.1042 / BST20190787

[0391] 14. Bender, J., andSchmidt, C. (2019) Mass spectrometry of membrane protein complexes Biol Chem 400, 813-82910.1515 / hsz-2018-0443

[0392] 15. Bledi, Y, Inberg, A., andLinial, M. (2003) PROCEED: A proteomic method for analysing plasma membrane proteins in living mammalian cells Brief Funct Genomic Proteomic 2, 254-265 10.1093 / bfgp / 2.3.254

[0393] 16. Peirce, M. J., andWait, R. (2009) Membrane proteomics : methods and protocols, Humana Press, New York, NY

[0394] 17. Li, Y, Qin, H., andYe, M. (2020) An overview on enrichment methods for cell surface proteome profiling J Sep Sci 43, 292-312 10.1002 / jssc.201900700

[0395] 18. Weekes, M. R, Antrobus, R., Lili, J. R., Duncan, L. M., Hor, S., andLehner, P. J. (2010) Comparative analysis of techniques to purify plasma membrane proteins J Biomol Tech 21 , IOS- 115, https: / / www.ncbi.nlm.nih.gov / pubmed / 20808639

[0396] 19. Wollscheid, B., Bausch-Fluck, D., Henderson, C., O'Brien, R., Bibel, M., Schiess, R. et al. (2009) Mass-spectrometric identification and relative quantification of N-linked cell surface glycoproteins Nat Biotechnol 27, 378-386 10.1038 / nbt.1532

[0397] 20. Bausch-Fluck, D., Hofmann, A., Bock, T., Frei, A. P., Cerciello, F., Jacobs, A. etal. (2015) Amass spectrometric-derived cell surface protein atlas PLoS One 10, e0121314 10.1371 / journal.pone.0121314

[0398] 21. Muller, A. (2019) Covalently modifed carboxyl side chains on cell surface leads to a novel method toward topology analysis of transmembrane proteins Scientific Reports 9,

[0399] 22. Patino-Escobar, B., Ferguson, I. D., andWiita, A. P. (2022) Unraveling the surface proteomic profile of multiple myeloma to reveal new immunotherapeutic targets and markers of drug resistance Cell Stress 6, 89-92 10.15698 / cst2022.11.273

[0400] 23. Ferguson, I. D., Patino-Escobar, B., Tuomivaara, S. T., Lin, Y. T., Nix, M. A., Leung, K. K. etal.

[0401] (2022) The surfaceome of multiple myeloma cells suggests potential immunotherapeutic strategies and protein markers of drug resistance Nat Commun 13, 4121 10.1038 / s41467-022- 31810-6

[0402] 24. Xu, X., Yin, K., Xu, S., Wang, Z., andWu, R. (2024) Mass spectrometry-based methods for investigating the dynamics and organization of the surfaceome: exploring potential clinical implications Expert Rev Proteomics 21, 99-11310.1080 / 14789450.2024.2314148

[0403] 25. Gerdes, H., Casado, P., Dokal,A., Hijazi, M., Akhtar, N., Osuntola, R. etal. (2021) Drug ranking using machine learning systematically predicts the efficacy of anti-cancer drugs Nat Commun 12, 1850 10.1038 / S41467-021-22170-826. Casado, P., Rio-Machin, A., Miettinen, J. J., Bewicke-Copley, F., Rouault-Pierre, K., Krizsan, S. et al. (2023) Integrative phosphoproteomics defines two biologically distinct groups of KMT2A rearranged acute myeloid leukaemia with different drug response phenotypes Signal Transduct Target Ther 8, 80 10.1038 / s41392-022-01288-1

[0404] 27. Bausch-Fluck, D., Goldmann, U., Muller, S., van Oostrum, M., Muller, M., Schubert, O. T. et al.

[0405] (2018) The in silico human surfaceome Proc Natl Acad Sci U SA 115, E10988-E10997

[0406] 10.1073 / pnas.1808790115

[0407] 28. Cutillas, P. R., andVanhaesebroeck, B. (2007) Quantitative profile of five murine core proteomes using label-free functional proteomics Mol Cell Proteomics 6, 1560-1573 10.1074 / mcp.M700037- MCP200

[0408] 29. Casado, P., andCutillas, P. R. (2011) A self-validating quantitative mass spectrometry method for assessing the accuracy of high-content phosphoproteomic experiments Mol Cell Proteomics 10, M11000307910.1074 / mcp.M110.003079

[0409] 30. Jiang, H., D'Agostino, G. D., Cole, P. A., andDempsey, D. R. (2020) Selective protein N-terminal labeling with N-hydroxysuccinimide esters Methods Enzymol 639, 333-353 10.1016 / bs.mie.2020.04.018

[0410] 31. Hijazi, M., Smith, R., Rajeeve, V., Bessant, C., andCutillas, P. R. (2020) Reconstructing kinase network topologies from phosphoproteomics data reveals cancer-associated rewiring Nat Biotechnol 38, 493-502 10.1038 / s41587-019-0391 -9

[0411] 32. Hijazi, M., Casado, P., Akhtar, N., Alvarez-Teijeiro, S., Rajeeve, V., andCutillas, P. R. (2022) eEF2K Activity Determines Synergy to Cotreatment of Cancer Cells With PI3K and MEK Inhibitors Mol Cell Proteomics 21, 10024010.1016 / j.mcpro.2022.100240

[0412] 33. Koniev, O., andWagner, A. (2015) Developments and recent advancements in the field of endogenous amino acid selective bond forming reactions for bioconjugation Chem Soc Rev 44, 5495-5551 10.1039 / c5cs00048c

[0413] 34. Ozkan Kucuk, N. E., Sanai, E., Tan, E., Mitchison, T., andOzlu, N. (2018) Labeling Carboxyl Groups of Surface-Exposed Proteins Provides an Orthogonal Approach for Cell Surface Isolation J Proteome Res 17, 1784-1793 10.1021 / acs.jproteome.7b00825

[0414] For standard molecular biology techniques, see Sambrook, J., Russel, D.W. Molecular Cloning, A Laboratory Manual. 3 ed. 2001, Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press

Claims

Claims:

1. A method for labelling cell surface proteins in a population of cells, the method comprising:i) providing a population of cells;ii) biotinylating primary amine groups on cell surface proteins within the population of cells;andiii) biotinylating carboxyl groups on cell surface proteins within the population of cells.

2. The method according to claim 1, wherein step (ii) comprises incubating the population of cells with an amine-reactive biotinylation reagent, wherein the amine-reactive biotinylation reagent comprises a N- oxysuccinimidyl moiety according to Formula (I)wherein R1may be selected from the group consisting of H, SO3H, and SOsNa.

3. The method according to claim 1 or claim 2, wherein the amine-reactive biotinylation reagent may be according to Formula (II):wherein L is a linker and R1may be selected from the group consisting of H, SO3H, and SOsNa.

4. The method according to any one of claims 1 to 3, wherein step (ii) comprises incubating the population of cells with a second amine-reactive biotinylation reagent.

5. The method according to claim 4, wherein the first amine-reactive biotinylation reagent is according to Formula (II), wherein R1is SOsNa; and the second amine-reactive biotinylation reagent is according to Formula (II), wherein R1is H.

6. The method of any one of claims 1 to 5, wherein step (iii) comprises activation of carboxyl groups using a carbodiimide.

7. The method of any one of claims 1 to 6, wherein step (iii) further comprises stabilisation of the carbodiimide-activated carboxyl group using N-hydroxysuccinimide (NHS) orsulfo-NHS.

8. The method of claim 6 or claim 7, wherein step (iii) further comprises incubating the population of cells with a carboxyl-reactive biotinylation reagent following carbodiimide activation, wherein the carboxylreactive biotinylation reagent comprises a free amine group.

9. The method according to claim 8, wherein the carboxyl-reactive biotinylation reagent may be according to Formula (III):wherein L' is a linker.

10. The method according to any one of claims 3 to 9, wherein a linker comprises a spacer, optionally wherein a spacer comprises or consists of an alkylene or oxyalkylene group.

11. The method according to any one of claim 3 to 10, wherein a linker is cleavable, optionally wherein a linker comprises a disulphide moiety.

12. The method of any one of claims 1 to 11 , wherein the pH of the population is lowered following step (ii), prior to step (iii), optionally wherein the pH of the population is lowered from a pH that is between 7 and 8 to a pH that is between 5 and 6.

13. The method of any one of claims 1 to 12, further comprising the step of:(iv) isolating the biotinylated proteins from the population of cells to produce a cell surface protein-enriched sample14. The method of claim 13, wherein step (iv) comprises performing affinity capture with streptavidin- containing magnetic beads.

15. The method of claim 13 or 14, further comprising the step of:(v) characterising the proteins in the cell surface protein-enriched sample.

16. The method of claim 15, wherein step (v) comprises performing mass spectrometry analysis, optionally wherein step (v) comprises performing LC / MS analysis.

17. The method of claim 15 or claim 16, wherein the method further comprises analysing a control sample to correct for background.

18. The method of claim 17, wherein the method comprises analysing a mock-labelled sample to correct for non-biotinylated proteins remaining in the cell surface protein-enriched sample, optionally wherein the mock-labelled sample is derived from or has a common origin with said population of cells.

19. The method of any one of claims 1 to 18, wherein the method further comprises a step of cell dissociation on a sample to produce the population of cells of step (i), optionally wherein the cell dissociation comprises the use of enzymatic dissociation methods.

20. The method of any one of claims 1 to 19, wherein the population of cells has a density of at least 15 million cells / mL, optionally wherein the population of cells has a density of approximately 20 million cells / mL.

21. The method of any one of claims 1 to 20, wherein the population of cells contains fewer than 5 million cells, optionally wherein the population of cells contains approximately 1.2 million cells or approximately 0.5 million cells.

22. The method of any one of claims 1 to 21 , wherein the population of cells comprises cryopreserved cells.

23. The method of any one of claims 1 to 22, wherein the population of cells comprises cells isolated from a human subject.

24. The method of any one of claims 1 to 23, wherein the population of cells comprises cancer cells.

25. The method of any one of claims 1 to 24, wherein the population of cells comprises cells derived from a solid tumor.

26. The method of any one of claims 1 to 24, wherein the population of cells comprises AML cells, colorectal adenocarcinoma cells, breast cancer cells, or hepatocellular carcinoma cells.

27. A kit comprising:one or more amine-reactive biotinylation reagents; anda carboxyl-reactive biotinylation reagent.

28. A method for identifying one or more diagnostic markers or therapeutic targets, the method comprising performing the method of any one of claims 1 to 26, optionally wherein the method is performed on each of a plurality of samples of known or suspected diseased tissue (“cases”) and is performed on each of a plurality of control samples (“controls”), further optionally wherein comparative analysis is performed to identify absolute and / or relative differences in the surface proteome of cases versus controls.

29. The method of claim 28, wherein the one or more therapeutic targets are at least 20% higher abundance in disease samples compared to control samples and are elevated compared to control in at least 50% of the disease samples.

30. The method of claim 28 or 29, wherein the therapeutic target is identified as a candidate target for an antibody drug conjugate and / or a CAR-T cell.

Images