Yeast surface display technology and barcoding in a multiplexed method for the quantification of antibodies against an antigen of interest

Abstract

The present invention relates to a multiplexed method for the quantification of antibodies in a plurality of samples, said antibodies being capable of binding to an antigen of interest in a plurality of samples, wherein each of the plurality of samples is in a separate vessel or well or a multi-well plate comprising (a) contacting each of the plurality of samples with yeast cells carrying a yeast-surface display expression cassette in its genome, wherein the yeast-surface display expression cassette encodes a fusion protein comprising the antigen of interest being fused to a cell wall or plasma membrane anchored protein of yeast, and wherein the cells for each sample of the plurality of samples can be distinguished from all other cells from all other of the plurality of samples by a unique DNA barcode being present in the cells for each sample, (b) separating the yeast cells from the sample, preferably by centrifugation, or filtration or magnetic beads that bind to the surface of the yeast cells, (c) contacting the yeast cells with a fluorescently labelled antibody binding to antibodies bound to the antigen of interest, (d) pooling the yeast cells after step (b) or (c), preferably after step (c), (e) sorting the yeast cells into different sub-pools based on the fluorescent intensity of each fluorescently-labeled yeast cell, wherein the fluorescent intensity of each cell is proportional to the number of antibodies being bound to the antigen of interest on the surface of the cell and optionally growing the sub-pools of cells,, (f) isolating the DNA from the yeast cells in each sub-pool, (g) subjecting the isolated DNA of each sub-pool to sequencing of each of the unique DNA barcodes, thereby quantifying the relative frequency of individual barcodes in each sub-pool and thereby identifying the samples being present in each of the subpools, and (h) quantifying the abundance of antibodies bound to the antigen of interest in each of the plurality of samples based on the mean fluorescent intensity measured for each yeast cell population expressing the antigen of interest and carrying a unique barcode, according to the distribution of the barcodes across the different fluorescent intensity bins for each sample, wherein the mean fluorescent intensity is proportional to the abundance of antigen-specific antibodies being present in each sample.

Description

[0001] Yeast surface display technology and barcoding in a multiplexed method for the quantification of antibodies against an antigen of interest

[0002] The present invention relates to a multiplexed method for the quantification of antibodies in a plurality of samples, said antibodies being capable of binding to an antigen of interest in a plurality of samples, wherein each of the plurality of samples is in a separate vessel or well of a multi-well plate comprising (a) contacting each of the plurality of samples with yeast cells of carrying a yeast-surface display expression cassette in its genome, wherein the yeast-surface display expression cassette encodes a fusion protein comprising the antigen of interest being fused to a cell wall or plasma membrane anchored protein of yeast, and wherein the cells for each sample of the plurality of samples can be distinguished from all other cells from all other of the plurality of samples by a unique DNA barcode being present in the cells to be contacted with each sample, (b) separating the yeast cells from the sample, preferably by centrifugation, or filtration or magnetic beads that bind to the surface of the yeast cells, (c) contacting the yeast cells with a fluorescently labelled antibody binding to antibodies bound to the antigen of interest, (d) pooling the yeast cells after step (b) or (c), preferably after step (c), (e) sorting the yeast cells into different sub-pools based on the fluorescent intensity of each fluorescently-labeled yeast cell, wherein the fluorescent intensity of each cell is proportional to the number of antibodies being bound to the antigen of interest on the surface of the cell and optionally growing the sub-pools of cells, (f) isolating the DNA from the yeast cells in each sub-pool, (g) subjecting the isolated DNA of each sub-pool to sequencing of each of the unique DNA barcodes, thereby quantifying the relative frequency of individual barcodes in each sub-pool and thereby identifying the samples being present in each of the subpools, and (h) quantifying the abundance of antibodies bound to the antigen of interest in each of the plurality of samples based on the mean fluorescent intensity measured for each yeast cell population expressing the antigen of interest and carrying a unique barcode, according to the distribution of the barcodes across the different fluorescent intensity bins for each sample, wherein the mean fluorescent intensity is proportional to the abundance of antigen-specific antibodies being present in each sample.

[0003] In this specification, several documents including patent applications and manufacturer’s manuals are cited. The disclosure of these documents, while not considered relevant for the patentability of this invention, is herewith incorporated by reference in its entirety. More specifically, all referenced documents are incorporated by reference to the same extent as if each individual document was specifically and individually indicated to be incorporated by reference.

[0004] Antibodies are an essential component of the immune system that play a critical role in fighting off infections and diseases. Antibody quantification in serum, plasma, or other body fluids has become increasingly important in the diagnosis, monitoring, and treatment of diseases. One of the most used techniques for antibody quantification is the Enzyme-Linked Immunosorbent Assay (ELISA). ELISA is a highly sensitive method, most frequently a colorimetric-based assay, in which antibodies present in the sample bind to specific antigens that are immobilized to a solid surface. The bound primary antibody is then detected using a secondary antibody conjugated to an enzyme, which converts a chromogenic substrate into a detectable signal that is proportional to the concentration of the antibody in the sample. Alternative immunoassays used in the clinic, claiming frequently higher sensitivity over ELISA, include the Chemiluminescent immunoassays (CLIA) and the Electro-chemiluminescent immunoassays (ECLIA) that have both as readout the generation of chemiluminescence. The main difference among them, is the type of reaction triggered by the antibody-antigen binding, either a chemical- or electrochemical-reaction, that results in chemiluminescence.

[0005] A recent breakthrough in antibody quantification includes the bead-based immunoassays that allows the simultaneous quantification of multiple antibodies in a single sample. Among them, the Luminex xMAP platform (Multi-Analyte Profiling), uses color-coded microspheres coated with specific antigens to capture antigen-binding antibodies present in the sample. The sample is incubated with a mixture of microspheres, and the captured antibodies are detected using a biotinylated secondary antibody and a fluorescently labeled streptavidin. The color / fluorescence intensity of each microsphere is measured using a flow cytometer, allowing for the simultaneous detection of hundreds of different antibodies in a single sample. However, one of the main limitations of this technology is that it requires the production, purification, and immobilization of each one of the antigens on microspheres. This process is not trivial and frequently results in batch-to-batch bead variations and lack of reproducibility. Additional disadvantages are high cost (in terms of both equipment and reagents) and the potential for crossreactivity between the different microspheres, which can lead to inaccurate results if it is not properly controlled.

[0006] Therefore, the simultaneous quantification of different antibodies in multiple samples remains difficult and, thus, nowadays samples are generally processed sequentially, and each type of antibody is analyzed independently. The present invention therefore provides of a novel method for the simultaneous quantification of antibodies in multiple samples.

[0007] Accordingly, the present invention relates to a multiplexed method for the quantification of antibodies in a plurality of samples, said antibodies being capable of binding to an antigen of interest in a plurality of samples, wherein each of the plurality of samples is in a separate vessel or well of a multi-well plate comprising (a) contacting each of the plurality of samples with yeast cells carrying a yeast-surface display expression cassette in its genome, wherein the yeast-surface display expression cassette encodes a fusion protein comprising the antigen of interest being fused to a cell wall or plasma membrane anchored protein of yeast, and wherein the cells for each sample of the plurality of samples can be distinguished from all other cells from all other of the plurality of samples by a unique DNA barcode being present in the cells for each sample, (b) separating the yeast cells from the sample, preferably by centrifugation, or filtration or magnetic beads that bind to the surface of the yeast cells, (c) contacting the yeast cells with a fluorescently labelled antibody binding to antibodies bound to the antigen of interest, (d) pooling the yeast cells after step (b) or (c), preferably after step (c), (e) sorting the yeast cells into different sub-pools based on the fluorescent intensity of each fluorescently-labeled yeast cell, wherein the fluorescent intensity of each cell is proportional to the number of antibodies being bound to the antigen of interest on the surface of the cell and optionally growing the sub-pools of cells, (f) isolating the DNA from the yeast cells in each sub-pool, (g) subjecting the isolated DNA of each subpool to sequencing of each of the unique DNA barcodes, thereby quantifying the relative frequency of individual barcodes in each sub-pool and thereby identifying the samples being present in each of the subpools, and (h) quantifying the abundance of antibodies bound to the antigen of interest in each of the plurality of samples based on the mean fluorescent intensity measured for each yeast cell population expressing an antigen of interest and carrying a unique barcode, according to the distribution of the barcodes across the different fluorescent intensity bins for each sample, wherein the mean fluorescent intensity is proportional to the abundance of antigen-specific antibodies being present in each sample.

[0008] A multiplexed method for the quantification of antibodies against an antigen of interest in a plurality of samples as used herein refers to a method wherein the antibodies against an antigen can be quantified simultaneously in a plurality of samples.

[0009] The plurality of samples refers with increasing preference to at least (for each number) 2, 3, 4, 5, 6, 7, 8, 9,10, 11 , 12, 24, 48, 96, 192, 384, and 1536 samples (or more, preferably a multiple of 96). The nature and examples of samples will be further detailed here below.

[0010] The method of the invention is a multiplexed method because antibodies being capable of binding to an antigen of interest can be quantified in parallel in a plurality of samples. It is also to be understood that the method of the invention can be further multiplexed at the end of the antigen of interest, i.e., the method for the quantification of antibodies against an antigen of interest in a plurality of samples can also be used for the quantification of antibodies against a plurality of antigens of interest in a plurality of samples by running a plurality of methods of the invention in parallel or together. For instance, the DNA barcodes as described herein may also be used to differentiate different antigens of interest. In this regard a plurality of antigens of interest is with increasing preference at least 2, at least 3, at least 5, at least 10, at least 15, at least 25, at least 50, at least 100, at least 200, at least 300, at least 400, at least 500 or at least 1000 different antigens of interest.

[0011] In this connection it is to be understood that it is an essential feature of the invention that the cells for each sample of the plurality of samples can be distinguished from all other cells from all other of the plurality of samples by a unique DNA barcode being present in the cells for each sample. Within one sample, the unique barcodes can optionally be further used to differentiate between different antigens of interest. For instance, in the case three different antigens of interest that are to be distinguished in one sample the sample is contacted with (1) yeast cells carrying a first yeast-surface display expression cassette in its genome, wherein the yeast-surface display expression cassette encodes a first fusion protein comprising the first antigen of interest being fused to a cell wall or plasma membrane anchored protein of yeast, (2) yeast cells carrying a second yeast-surface display expression cassette in its genome, wherein the yeast-surface display expression cassette encodes a second fusion protein comprising the second antigen of interest being fused to a cell wall or plasma membrane anchored protein of yeast, and (3) yeast cells carrying a third yeast-surface display expression cassette in its genome, wherein the yeast-surface display expression cassette encodes a third fusion protein comprising the third antigen of interest being fused to a cell wall or plasma membrane anchored protein of yeast. In this case unique barcodes are used that not only distinguish this sample from all other samples of the plurality of sample but also distinguish the three kinds of yeast cells with the three kinds of expressed antigens from each other. Thereby the method of the invention optionally allows to the quantification of antibodies in a plurality of samples as well as for each of the samples the separate quantification of the three kinds of antibodies binding to the three different kinds of antigens of interest.

[0012] While the above examples have been provided examples have been provided for three different antigens the examples apply mutatis mutandis to any number of plurality of antigens, such as only 2 or up to 1000.

[0013] An antigen is generally any molecule (e.g., a protein or peptide) that can bind to a specific antibody or T-cell receptor. The antigen of interest is not particularly limited and can be any antigen for which it is desired to know how many antibodies binding said antigen are present in a sample. Of particular interest are antigens of pathogens, such as antigens of viral or bacterial proteins, but also of human proteins that can be targeted during autoimmune diseases or cancer (neoepitopes, cancer associated antigens). For example, the level of antibodies against an antigen of a pathogen in a sample obtained from a subject, can provide information about the immunity status of the subject against the pathogen. Similarly, the presence of antibodies reacting with antigens of human proteins, is frequently associated to autoimmune disorders (e.g., lupus erythematosus, multiple sclerosis, diabetes type 1 , etc.).

[0014] Various suitable vessels, wells and multi-well plates for samples are available in the prior art. Nonlimiting but preferred examples of vessels are tubes, such 1 .5 mL or 2.0 mL Eppendorf tubes. Nonlimiting but preferred samples of multi-well plates are 6, 12, 24, 48, 96, 384 and 1536-well plates. In connection with the multi-well plates separate plates may be used for each sample but it is preferred that the different samples are present in separate wells of the same multi-well plate. The vessels, wells and multi-well plates may have a lid to close the same.

[0015] The yeast cells (i.e. cells of a yeast strain) as used in the method of the invention carry a yeast-surface display expression cassette in their genome, wherein the yeast-surface display expression cassette encodes a fusion protein comprising an antigen of interest being fused to a cell wall or plasma membrane anchored protein of yeast. In this connection the term “in their genome” means that the expression cassette is chromosomally integrated into the yeast genome, The yeast cells display on their cell surface an antigen of interest. Yeast surface display systems are established in the art; see, for example, Teymennet-Ram rez et al. (2022), Front. Bioeng. Biotechnol., Volume 9, https: / / doi.org / 10.3389 / fbioe.2021.794742. The yeast is preferably Saccharomyces cerevisiae. An expression cassette basically consists of a promoter, the open reading frame (ORF), and a terminator. Promoters used in yeast expression systems are known in the art; they can either be inducible or constitutive. In accordance with the present invention the ORF encodes a fusion protein comprising an antigen of interest being fused to a cell wall or plasma membrane anchored protein of yeast. The expression of the fusion protein is ensured by placing the ORF under the control of a promoter, preferably a constitutive promoter. Suitable promoters for yeast are known in the art (see, for example, the text book Synthetic Biology: Parts, Devices and Applications (2018); Chapter 6 “Constitutive and Regulated Promoters in Yeast: How to Design and Make Use of Promoters in S. cerevisiae”, https: / / doi.org / 10.1002 / 9783527688104.ch6). In the appended examples, for example, the GAL1 promoter and the pGPD promoter are used.

[0016] The fusion of the antigen of interest to a cell wall or plasma membrane anchored protein of the yeast cell ensures that the antigen of interest is displayed on the surface of the yeast cells, where they can be bound by the antibodies in the samples. It is to be understood that the conditions in step (a) are to be selected such that more antigen is displayed on yeast cells surface than antibodies against the antigen might be present in the samples. This is to ensure that essentially all the antibodies against the antigen are bound to the displayed antigens after step (a). The number of yeast cells to be employed in the method of the invention can be adjusted accordingly.

[0017] The cell wall or plasma membrane anchored protein of the yeast can either be a cell wall or plasma membrane anchored protein that is encoded by a naturally occurring yeast cell but may also be a cell wall or plasma membrane anchored protein that has been genetically engineered. The genetically engineered cell wall or plasma membrane anchored protein may be an exogenous cell wall or plasma membrane anchored protein that has be introduced in expressible form as gene into the yeast or genome or may be a mutated form of a cell wall or plasma membrane anchored protein that is encoded by a naturally occurring yeast cell. Also, synthetic constructs that contain crucial elements for the secretion and presentation of the antigen at the yeast surface may be used. It can also be a protein that is capable to bind to a cell wall or plasma membrane protein (frequently a yeast protein, e.g., AGA2), or a protein fragment or domain of a cell wall or a plasma membrane anchored protein or a synthetic or chimeric protein containing a secretion peptide signal and in some cases a C-terminal signal for addition of a glycosylphosphatidylinositol anchor to cell wall.

[0018] It is important to note that the expression cassette is in accordance with the invention integrated into the yeast genome and is thus, for example, not episomally encoded on a plasmid or vector. The inventors tried both genomic and episomal (plasmid) expression. Figure 1 shows that in the method of the invention the expression of the antigen encoded into the genome resulted in the homogeneous display of the antigen with small cell to cell variation, whereas the antigen expression encoded by a plasmid resulted in a broader distribution. As the accuracy of the method is strongly affected by variations in the number of antigens on the yeast surface, the use of yeast strains encoding antigens integrated into the yeast genome constitute an essential feature of the method of the invention. DNA barcodes within the yeast cells ensure that each sample can be distinguished from all other samples in the plurality of samples and, as explained above, optionally further ensure that different antigens of interest can be distinguished from each other. A DNA barcode is a DNA sequence that has been introduced into the yeast cells that is unique for each sample. In other words, the DNA barcode sequence for one sample cannot be found in any of the other samples being subjected to the multiplex method of the invention.

[0019] The means for separating the yeast cells from the sample after step (a) are not particularly limited. Preferred examples are centrifugation, or filtration or the use of magnetic beads that bind to the surface of the cells. Centrifugation is most preferred and illustrated in the examples.

[0020] The fluorescently labelled antibody binds to antibodies bound to the antigen and therefore may also be referred to as a secondary antibody that binds to the primary antibodies bound to the antigen. The secondary antibody preferably binds to the Fc part of primary antibodies. For example, in case the sample was human the secondary antibody preferably binds to the Fc part of human antibodies, more preferably the Fc part of human IgG antibodies. Such fluorescently labelled antibodies are commercially available. The fluorescently label can be, for example, FITC, Cy3, Cy5, Texas Red, Tamra or an Alexa fluor. As the fluorescently labelled antibody also two antibodies can be used that bind to each other. In this case a primary antibody binds to the antibodies bound to the antigen. This primary antibody is then bound by a fluorescently labelled secondary antibody. The binding between the primary and the secondary antibody is preferably ensured via streptavidin / avidin and biotin binding; i.e one antibody carries streptavidin / avidin and the other antibody biotin.

[0021] While the yeast cells can be pooled after step (b) or (c) they are preferably pooled after step (c). The inventors tried both options and better results regarding the accuracy of the method of the invention were obtained with pooling after step (c).

[0022] In the method of the invention the fluorescent intensity of each yeast cell is proportional to the number of antibodies being bound via the antigen of interest to the surface of the cell. Based on the fluorescent intensity the yeast cells can therefore be sorted or separated into different sub-pools or also called “bins” herein.

[0023] Optionally the sub-pools of cells are grown to increase the number of cells and facilitate later steps (e.g., DNA extraction and PCR). Growth media and conditions for yeast cells are well established in the art. Typically, a yeast culture medium includes peptone, yeast extract, and dextrose or glucose. Yeast cells are usually grown in incubators under aerobic conditions, constant agitation, and a temperature of 30°C. Also the isolation of DNA from yeast cells is standard in the art; see, for example, Looke et al. (2011), Biotechniques; 50(5): 325-328. Yeast DNA extraction kits are commercially available.

[0024] In the method of the invention the isolated DNA of each sub-pool is then subjected to sequencing of each of the unique DNA barcodes. Thereby the nature of barcodes is identified. In addition, the relative frequency of individual barcodes in each sub-pool is thereby quantified. This in turn also identifies the samples being present in each of the subpools.

[0025] This identification of the samples via the barcodes makes possible to quantify the antibodies against the antigen of interest in each of the plurality of samples in a multiplexed manner. This is because the mean fluorescent intensity of each sample is proportional to the number of antibodies to the antigen in each sample. The more antibodies to the target antigen were bound, the higher is the fluorescence intensity. In addition, the barcode read-count information obtained from the sequencing analysis is used to determine the quantities of antibodies against the target antigen for each sample within the plurality of samples.

[0026] By the operation step (h) the results of the quantification of antibodies against an antigen of interest in a plurality of samples are normalized for t cell to cell variation in the amount of the number of fusion proteins being displayed on the surface of the cell, noting that the fusion proteins comprise the antigen of interest.

[0027] Accordingly, in step (h) the mean fluorescent intensity is measured for each yeast cell population expressing an antigen of interest. In this respect it is preferred with increasing preference that the yeast cell population comprises at least 100 cells, at least 200 cells, at least 500 cells, at least 1000 cells, at least 2000 cells and at least 5000 cells.

[0028] Hence, the method of the present invention with all its discussed technical features provides for the simultaneous quantification of antibodies against a target antigen in multiple samples with high accuracy. As explained above, the method of the present invention can also be further multiplexed by unique barcodes by detecting antibodies being directed to a plurality of different antigens. The method of the present invention is also called “Seroseq methodology” in the appended examples. The examples demonstrate that the Seroseq methodology can advantageously be used to quantify antibodies directed to different antigens in a large number of samples (n=96), in a pooled analysis. As proof of principle the method has been put into practice with a cohort of human serum samples (n=96) that were tested for the quantity of anti-S and anti-NCP antibodies (and AGA2-scaffold, as negative control) of SARS CoV2. In contrast to previous applications where the complexity was given mainly by the number of different yeast cells (i.e., cells of different yeast strains) present in the library, in the method of the invention the complexity can be the result of two variables: the number of antigens of interest to be analyzed and the number of samples to be analyzed. Based on this, Seroseq can reach, for example, at least a multiplex level of about 50-100 antibodies that can be measured simultaneously in 100 samples per run. A scheme illustrating the steps of this technology is shown in Fig. 6. As illustrated, the technology is based on yeast-surface display technology and DNA-unique barcodes. As explained above, the yeast-surface display technology is used to represent an antigen of interest on the surface of the yeast cells. It is important to note that the expression cassette encoding the antigen of interest is expressed from the genome which in the context of the method of the invention results in the display of the antigen at high density and with little cell to cell variation; see Fig. 1. Hence, it was surprisingly found that genomic expression significantly improved the accuracy ofthe method of the invention. As also explained above, the barcode allows for multiplexing in the context of the method of the invention. The barcode allows to label yeast cells and pool them. Another important parameter ofthe method ofthe invention is the sorting into sub-pools (bins) according to the fluorescence intensity of the yeast cells that is proportional to the number of antibodies attached on the yeast surface and the use of sequencing analysis, preferably high- throughput sequencing analysis wherein the barcode read-count information can be used to determine the quantities of antibodies against the target antigen separately for each sample within the plurality of samples. Overall, the method of the invention is the result of the thoughtful combination of a sequence of technical steps that were matched, so that they finally result in a method for the simultaneous quantification of antibodies against a target antigen in multiple samples with high accuracy.

[0029] In accordance with a preferred embodiment of the invention the fusion protein further comprises an epitope tag, preferably at its C-terminus, wherein the epitope tag is preferably a MYC tag, a FLAG tag or a V5 tag.

[0030] Epitope tags are widely used in applications such as ELISA, Western blot, and immunoprecipitation. Epitope tags are generally short amino acid sequences that are fused onto proteins or peptides and are frequently used in antibody-based assays. In such antibody-based assays antibodies are used that specifically bind to the tag of choice (e.g., anti-myc antibodies).

[0031] The preferred myc-tag is the c-myc tag having 10 amino acids (EQKLISEEDL; SEQ ID NO: 3). The FLAG-tag has 8 amino acids (DYKDDDDK; SEQ ID NO: 7). The V5 tag has 9-14 amino acids (IPNPLLGLD or GKPIPNPLLGLDST; SEQ ID NOs 8 and 9).

[0032] In accordance with a further preferred embodiment of the invention the cell wall or plasma membrane anchored protein of yeast is the Aga2p protein of Saccharomyces cerevisiae, wherein the Aga2p protein preferably comprises SEQ ID NO: 1 or a sequence being at least 80% identical thereto.

[0033] The Aga2p protein of Saccharomyces cerevisiae is part of the a-agglutinin heterodimer composed of the anchor protein AGA1 p and its smaller binding subunit AGA2p. Agglutinins are cell surface glycoproteins that exist either as a- or alpha-agglutinins, which are expressed by a and alpha mating type cells, respectively. The interaction between a-agglutinin and alpha-agglutinin promotes cell-cell recognition and adhesion during the yeast mating process.

[0034] While the Aga2p protein of Saccharomyces cerevisiae is used in the examples, several other cell wall or plasma membrane anchored proteins can be used in yeast surface display systems. In this respect reference is made to Table 1 of Mei et al. (2017), Microbiological Research, 196:118-128. Target

[0035] Anchors Carriers protein Hosts Applications fusion site a-agglutinin a-agglutinin Saccharomyces MHC, LplA, GOx, HRP, TEV-

[0036] (Agal p) (Aga2p) terminal cerevisiae, P, SrtA; hGal-1 , hEPO, Pichia pastoris mIFN-p, mlFN-y, etc. a- a- Pichia pastoris, p-glucosidase, agglutinin agglutinin C terminal Saccharomyces hemagglutinin;

[0037] (Aga1 p) (Aga1 p) cerevisiae a-galactosidase

[0038] N or C Pichia pastoris,

[0039] Flo1 p phospholipase D; cellulolytic terminal Saccharomyces cerevisiae enzymes, a-galactosidase

[0040] Cwp1 p Cwp1 p C > t .erm .ina ,l Sacch .a .romy3ces a-ga .lac ,tos .id .ase

[0041] Cwp2p Cwp2p C > , term .ina ,l Sacch .a .romy3ces a-ga ,lac ,tos .id .ase

[0042] Sedl p Sedl p C „ t .erm .ina , Sacch .a .romy3ces a-ga . act .os .id ,ase

[0043] C „ , term .i . Saccharomyces , , . . nal . .3a-galactosidase

[0044] Tirl p Tirl p C > t .erm .ina .l Sacch .a .romy3ces a-ga ,lact .os .id ,ase

[0045] YCR89w YCR89w C „ t .erm .ina ,l Sacch .a .romy3ces a-ga ,lact ,os .id .ase

[0046] The sequence being at least 80% identical to SEQ ID NO: 1 is with increasing preference at least (for each number) 85%, 90%, 95%, 97%, 98% and 99% identical to SEQ ID NO: 1.

[0047] Nucleotide and amino acid sequence analysis and alignment in connection with the present invention are preferably carried out using the NCBI BLAST algorithm (Stephen F. Altschul, Thomas L. Madden, Alejandro A. Schaffer, Jinghui Zhang, Zheng Zhang, Webb Miller, and David J. Lipman (1997), Nucleic Acids Res. 25:3389-3402). BLAST can be used for nucleotide sequences (nucleotide BLAST) and amino acid sequences (protein BLAST). The skilled person is aware of additional suitable programs to align nucleic acid sequences.

[0048] In the appended examples the Aga2p protein of Saccharomyces cerevisiae is part of a fusion protein AGA2-S-RBD-MYC, wherein AGA2 is the Aga2p protein of Saccharomyces cerevisiae, S-RBD is the antigen of interest (spike protein receptor binding domain of SARS-CoV2, amino acids 331-531) and the MYC in an antigen tag.

[0049] The S-RBD is represented by SEQ ID NO: 2 and the MYC-tag is represented by SEQ ID NO: 3 (EQKLISEEDL). The AGA2 and the S-RBD are connected by the linker of SEQ ID NO: 4 (GGGGSGGGGSGGGGSAS). The entire fusion protein AGA2-S-RBD-MYC is represented by SEQ ID NO: 5. In the examples also a AGA2-NCP-MYC fusion protein is illustrated, wherein instead of S-RBD the nucleoprotein (NCP, amino acids 2-178) of SARS-CoV2 is used as the antigen of interest. The entire fusion protein AGA2-NCP-MYC is SEQ ID NO: 6.

[0050] The expression of the fusion proteins are preferably under the control of a strong constitutive promoter pGPD (also from S. cerevisiae) but other promoters are also suitable (e.g., GAL1, TEF, etc.). The expression can be terminated by a terminator sequence, for example and preferably, by the mating pheromone alpha-factor 1 (MF(ALPHA)1) terminator from S. cerevisiae.

[0051] In accordance with a further preferred embodiment of the invention the sorting in step (d) comprises fluorescent activated cell sorting (FACS).

[0052] Fluorescence-activated cell sorting (FACS), sometimes called fluorescence-assisted cell sorting, is a specialized type of flow cytometry that uses fluorescent markers to target and isolate cell groups. This cell sorting technique is commonly used in different research fields, including hematopoiesis, oncology, stem cell biology, among others.

[0053] In accordance with another preferred embodiment of the invention the sequencing in step (f) is next generation sequencing, preferably Nanopore or Illumina sequencing.

[0054] Sanger sequencing (also known as dideoxy or capillary electrophoresis sequencing) was developed by Frederick Sanger and his colleagues over 40 years ago in 1977. Sanger sequencing is commonly referred to as first-generation sequencing.

[0055] Second and third generation technologies are commonly referred to as next generation sequencing (NGS) technologies (for review see, for example, van Djik et al. (2014), “Ten years of next generation sequencing technology”, Trends Genet, 30(9):418-426; Slatko et al. (2018), “Overview of Next Generation Sequencing Technologies”, Curr Protoc Mol Biol.; 122(1):e59; and Arigrain (2021), “Beginner’s guide to next-generation sequencing”, The Biochemist, 43(6):58-64). Currently, the most commonly used NGS technologies are commercially provided by Illumina, PacBio and Oxford Nanopore Technology (ONT).

[0056] The technology of Oxford Nanopore Technology (ONT) is based on a different principle, the poreforming protein a-haemolysin. The pore being formed by a-haemolysin has an inner diameter of 1 nm, which is just large enough to allow a single strand of DNA (ssDNA) to pass through.

[0057] ONT flow cells are controlled by an application-specific integrated circuit (ASIC) which sits under an artificial membrane containing a plethora of these protein nanopores. A voltage is set across the membrane attracting negatively charged DNA molecules through the nanopores which obstructs the current across the membrane. Because the four bases of DNA have different shapes and sizes, their translocation through a nanopore leads to different current variations which constitute ONT’s raw sequencing signal (squiggle). A base-calling algorithm then converts the squiggles into a sequence, each intensity step corresponding to the multiple bases obstructing the pore at a certain time (Arigrain (2021), “Beginner’s guide to next-generation sequencing”, The Biochemist, 43(6):58-64). Illumina’s sequencing technology combines fluorescent microscopes with microfluidics devices allowing the sample and reagents to flow through and hybridize at the surface of a glass flow cell. The flow cells contain one or more channels through which the libraries flow. Their glass surface is etched with millions of nanowells grafted with synthetic DNA linkers complementary to the libraries’ adaptors. Upon flow cell loading, one library fragment hybridizes at the surface of each nanowell. To increase the nanowell signal from a single molecule to several thousands, the newly hybridized libraries are amplified using a surface PCR chemistry, also called isothermal or bridge amplification. After the formation of these monotemplate clusters, sequencing by synthesis (SBS) can start. Fluorescently labelled nucleotides containing a blocker group are added onto the flow cell and incorporated to the clustered fragments one base at a time. A chemistry pauses the library synthesis whilst a fluorescence image of the flow cell is recorded. After imaging, the fluorophore attached to the freshly incorporated nucleotide is cleaved, allowing a new cycle of synthesis and imaging to take place. An algorithm then performs the base calling by translating each image taken at each SBS cycle into a sequence (Arigrain (2021), “Beginner’s guide to next-generation sequencing”, The Biochemist, 43(6):58-64).

[0058] In accordance with a different preferred embodiment of the invention the barcode is a nucleotide sequence of 20 to 50 nucleotides, preferably 20-28 nucleotides.

[0059] The barcodes are sufficiently long, so that different barcodes for a plurality of samples can be generated that make the samples distinguishable over each other by sequencing the barcodes. In accordance with a different preferred embodiment of the invention the barcodes are a nucleotide sequence of 6 up to 50 nucleotides (of known identity). The barcodes are on the other sufficiently short such that they can be easily sequenced.

[0060] In accordance with a further preferred embodiment of the invention the number of sub-pools is 4-24, preferably 6-14 and most preferably 8-12.

[0061] The above number of sub-pools is sufficient to reduce the complexity of the entire pooled sample into sub-pools, collecting yeast cells with different fluorescent intensities of the fluorescent label - from low to high - that can be further processed into steps (f) to (h) of the invention in order to quantify the antibodies against the antigen of interest in each of the plurality of samples in a multiplexed manner and with high accuracy. As explained, the mean fluorescent intensity of each sample is proportional to the number of antibodies to the antigen in each sample. Therefore, the pools with fluorescent intensities from low to high also comprise samples with antibodies level to the target antigen from low to high.

[0062] In accordance with a further preferred embodiment of the invention after step (f) and before step (g) the part of isolated DNA comprising the unique DNA barcode is amplified by PCR, wherein the primers used in this amplification preferably comprise barcodes that allow distinguishing all amplicons from one subpool from all amplicons from all other sub-pools, wherein said barcode is preferably the DNA- sequence(s) at the 3’ and / or 5’-end of the amplicons.

[0063] Hence in accordance with this preferred embodiment of the invention a second barcode is used in the method that does not render the samples distinguishable from each other but rather the DNA that has been isolated from the different sub-pools with characteristic fluorescent intensities.

[0064] This is technically advantageous because it allows to sequence the DNA from the different sub-pools in step (g) in one reaction, since the barcodes serve for differentiating which sequence is from which subpool.

[0065] In accordance with a still further preferred embodiment of the invention the unique DNA barcodes are unique plasmid DNA barcodes or unique genomic DNA barcodes and are preferably unique genomic DNA barcodes.

[0066] While plasmid DNA barcodes and genomic DNA barcodes can be used, genomic DNA barcodes are preferred because they are stably integrated into the yeast cells and plasmid counter-selection (e.g., using synthetic drop-out selection growth media) is not required.

[0067] In accordance with a more preferred embodiment of the invention, wherein the yeast cells further comprise an epitope tag as described herein above, the yeast cells are contacted with a fluorescently labelled antibody binding to the epitope tag, wherein the fluorescent intensity of each cell is proportional to the number of fusion proteins being displayed on the surface of the cell, and wherein the fluorescent label of the antibody binding to the epitope tag is distinct from the fluorescent label of the antibody binding to antibodies bound to the antigen.

[0068] As discussed above, Figure 1 shows that in the method of the invention the expression from the genome resulted in the display of the antigen with little cell to cell variation. Hence, it was surprisingly found that the expression of the antigen encoded into the yeast genome significantly improved the accuracy of the method of the invention.

[0069] If needed, the little cell to cell variation can be balanced by the above more preferred embodiment of the invention. Here, the yeast cells comprise an epitope tag that is bound by fluorescently labelled antibody. This, results in fluorescently labelled yeast cells, wherein the fluorescent intensity of each one of the yeast cells is proportional to the number of fusion proteins being displayed on the surface of the cells. Hence, the measured fluorescent intensity can be used to normalize differences in the number of fusion proteins being displayed on the surface of different cells.

[0070] In accordance with this more preferred embodiment of the invention the following fluorescently labelled antibodies are used: 1) a fluorescently labelled antibody binding to antibodies bound to the antigen, and 2) a fluorescently labelled antibody binding to the epitope tag. It is common practice in the art to select distinct fluorescent labels that are spectrally distinct, so that they can be detected in parallel, e.g., by a FACS machine being equipped with at least two lasers (but also required filters and detectors) that can be used to excite and detect different fluorophores simultaneously.

[0071] In accordance with a further preferred embodiment of the invention, the sample is (i) a biological sample or an industrial sample, or (ii) is tissue or body fluid, wherein the body fluid is preferably blood, plasma, serum, cerebrospinal fluid, or saliva.

[0072] While the nature of the samples to be used in the method of the invention is not particularly limited the above samples are preferred.

[0073] A biological sample is generally a sample that can be obtained from cell(s) or an organism. An industrial sample is generally a sample that is produced in an industrial process, such as a manufacturing process.

[0074] The tissue sample can be from any tissue or body site, such as bone, cartilage, skin, heart, liver, colon, gut, spleen, kidney, etc. The tissue sample can be homogenized body tissue, more preferably a cell- free suspension of a homogenized body tissue. The body fluid can be any fluid sample that can be obtained. The most common body fluids are blood, saliva, semen, vaginal fluids, mucus, and urine. Preferred are blood, plasma, serum, cerebrospinal fluid and saliva. The blood sample can be whole blood, serum, and plasma, wherein serum is most preferred.

[0075] In accordance with a further preferred embodiment of the invention, the sample is a preadsorbed sample that has been depleted for antibodies that bind to (i) yeast surface components; (ii) one or more elements of the yeast-surface display expression cassette, wherein said one or more elements are devoid of the antigen of interest; (iii) selected (sub-)domains of the antigen of interest, and / or (iv) a protein containing at least one amino acid substitution or post-translational modification relative to the antigen of interest.

[0076] In accordance with a further more preferred embodiment of the invention, the method of the above preferred embodiment, further comprises the step of preparing the preadsorbed sample by contacting the sample with yeast cells that (i) do not display the antigen of interest; (ii) display only selected (sub-) domains thereof; and / or (iii) display a protein containing at least one amino acid substitution or post- translational modification relative to the antigen of interest on their surface.

[0077] Humans and other animals are frequently exposed to yeast and therefore samples may contain antibodies that recognize yeast cells, in particular the surface components of the yeast cell wall. To get rid of such anti-yeast antibodies that might interfere with the accuracy of the method of the invention the samples can be preadsorbed samples that have been depleted for antibodies that bind to (i) yeast surface components and / or (ii) one or more elements of the yeast-surface display expression cassette. This can be done by contacting the samples with yeast cells that do not display the antigen of interest. For instance, in the appended examples samples were contacted in a preadsorption step with wild-type yeast cells or yeast cells expressing the yeast surface display (YSD) scaffold protein (AGA2) but not the antigen(s) of interest (see Fig. 5 and 6). Hence, the yeast cells that do not display the antigen of interest may be wild-type yeast cells, but preferred yeast cells expressing the yeast surface display (YSD) scaffold platform (e.g., AGA2-linker-tag) but not the antigen(s) of interest. Instead of such yeast cells also such yeast cell lysates or synthetic constructs that contain crucial elements for the secretion and presentation of the antigen at the yeast surface may be used in the preadsorbtion step.

[0078] In addition or as an alternative, samples may be a preadsorbed with selected (sub-)domains of the antigen of interest and / or (iv) a protein containing at least one amino acid substitution or post- translational modification relative to the antigen of interest by contacting the samples with selected (sub- ) domains of the antigen of interest; and / or (iii) protein containing at least one amino acid substitution or post-translational modification relative to the antigen of interest on their surface.

[0079] Such a preadsorbtion with selected (sub-)domains of the antigen of interest has the technical effect that the antibodies in the sample that bind to the selected (sub-)domains of the antigen of interest can no longer bind to the antigen of interest. On the other hand, antibodies within the samples that bind to other (sub-)domain(s) and antigens than the selected (sub-)domains of the antigen of interest can still bind to the antigen of interest. Thereby samples can, for example, be checked for the presence of antibodies that bind to epitopes affected by specific amino acid substitutions / post-translational modifications within the antigen of interest.

[0080] Similarly, preadsorption with protein containing at least one amino acid substitution or post-translational modification relative to the antigen of interest on their surface blocks the binding of antibodies in the sample that bind to the antigen of interest as the result of amino acid substitution^) or post-translational modification relative to the antigen of interest. The at least one amino acid substitution is with increasing preference 10 or less, 5 or less, 3 or less, 1 or 2 and 1 amino acid substitution^). The nature of post- translational modification is not particularly limited and can be, for example, ubiquitinylation, methylation, acetylation, lipidation, glycosylation (e.g., N-linked or O-linked), glypiation or phosphorylation.

[0081] As regards the embodiments characterized in this specification, in particular in the claims, it is intended that each embodiment mentioned in a dependent claim is combined with each embodiment of each claim (independent or dependent) said dependent claim depends on. For example, in case of an independent claim 1 reciting 3 alternatives A, B and C, a dependent claim 2 reciting 3 alternatives D, E and F and a claim 3 depending from claims 1 and 2 and reciting 3 alternatives G, H and I, it is to be understood that the specification unambiguously discloses embodiments corresponding to combinations A, D, G; A, D, H; A, D, I; A, E, G; A, E, H; A, E, I; A, F, G; A, F, H; A, F, I; B, D, G; B, D, H; B, D, I; B, E, G; B, E, H; B, E, I; B, F, G; B, F, H; B, F, I; C, D, G; C, D, H; C, D, I; C, E, G; C, E, H; C, E, I; C, F, G; C, F, H; C, F, I, unless specifically mentioned otherwise.

[0082] Similarly, and also in those cases where independent and / or dependent claims do not recite alternatives, it is understood that if dependent claims refer back to a plurality of preceding claims, any combination of subject-matter covered thereby is considered to be explicitly disclosed. For example, in case of an independent claim 1 , a dependent claim 2 referring back to claim 1 , and a dependent claim 3 referring back to both claims 2 and 1 , it follows that the combination of the subject-matter of claims 3 and 1 is clearly and unambiguously disclosed as is the combination of the subject-matter of claims 3, 2 and 1 . In case a further dependent claim 4 is present which refers to any one of claims 1 to 3, it follows that the combination of the subject-matter of claims 4 and 1 , of claims 4, 2 and 1 , of claims 4, 3 and 1 , as well as of claims 4, 3, 2 and 1 is clearly and unambiguously disclosed.

[0083] The figures show.

[0084] Figure 1. Generation of yeast surface display (YSD) strains encoding for SARS-CoV2 antigens. (A) Schematic representation of SARS-CoV2 Spike (S) and Nucleoprotein (NCP) proteins indicating protein domains used for yeast surface display expression. On the right, scheme illustrating YSD strain expressing the AGA2-S-RBD-tag fusion protein and binding antibody. (B) Yeast surface protein expression profile for AGA2-S-RBD-HA strains in which the fusion protein is either encoded in a plasmid or is chromosomally integrated. Logarithmically growing cells stained with anti-HA antibody were analyzed using flow cytometry analysis (FACS). 95% confidence intervals (Cl) are indicated with brackets; median fluorescence intensity is indicated with a dotted line.

[0085] Figure 2. Detection of human IgG antibodies against SARS-CoV2 S-RBD (A) or NCP (B) proteins in human sera using YSD and fluorescence microscopy. Yeast surface display strains expressing S-RBD (331 -531 aa) or NCP (2-178aa) protein domains (labeled with anti-MYC antibody) were mixed in a 1 :10 ratio with wild-type "empty" cells. Human sera from COVID-19 infected or vaccinated donors contain antibodies that recognize the S-RBD domain (A, white cells). Antibodies recognizing the NCP protein were exclusively found in COVID-19 infected donors (B, white cells).

[0086] Figure 3. Quantification of SARS-CoV2 S and NCP antibodies using yeast surface display and fluorescence-activated cell sorting (YSD-FACS) protocol. (A) Quantification of SARS-CoV-2 S and NCP antibody levels in COVID-19-vaccinated, -infected, -vaccinated and infected, or in negative samples (pre-pandemic). Circles and triangles indicate antibody titer as Iog10 YSD-FACS arbitrary units (a.u.) for S-RBD and NCP, respectively. The median value for each group is indicated with a straight horizontal line. (B) Correlation of anti-S antibody titer determined in 57 samples using YSD-FACS (y-axis) or a commercial ELISA assay (Euroimmun) (x-axis). R2= Pearson's correlation coefficient. (C) Comparison of the dynamic range of YSD-FACS (black filled circles, y-axis left) or ELISA (hollow triangles, y-axis right) by measuring a dilution series (x-axis) of a seropositive sample.

[0087] Figure 4. Longitudinal analysis of S-RBD antibody levels using YSD-FACS. (A) Follow-up anti-S-RBD antibody levels in donors (n=11) two weeks and three months after vaccination (2 x BioNTech / Comirnaty). (B) Longitudinal study of S-RBD antibody levels in one donor over a two-year period. Blood samples were collected at indicated time points and later processed for S-RBD antibody quantification. Figure 5. A fraction of the antibodies generated upon vaccination (BioNTech / Comirnaty) recognize S- RBD amino acids L452, and T478 that are mutated in S-RBD-delta variant. Sera from COVID-19- vaccinated, -vaccinated and infected (with SARS-CoV2-delta) or a sample pre-pandemic were preadsorbed either with baker's yeast, YSD-S-RBD or YSD-S-RBD-delta strains. Preadsorbed sera were then used for recognition of YSD S-RBD-WT by fluorescence microscopy (A) or FACS (B). Signal corresponding to S-RBD antibodies was highest when baker's yeast was used for preadsorption. In contrast, S-RBD antibodies were no longer detectable after preadsorption with a strain expressing the S-RBD domain. A small fraction of S-RBD antibodies were not depleted when a strain expressing the S-RBD-delta variant was used for preadsorption, indicating that the remaining signal comes from antibodies that recognize at least one epitope composed in part by amino acid L452 and / or T478.

[0088] Figure 6. Seroseq: Multiplexed serology protocol based on YSD-FACS and deep-sequencing. Human samples (blood, plasma, serum, saliva, or others) were preadsorbed with wild-type yeast cells or yeast expressing the YSD scaffold protein (AGA2) but not the antigen(s) of interest. Preadsorbed serum was incubated with YSD strain expressing the antigen of interest (e.g., SARS-CoV2-NCP) and containing a unique barcode that is linked to the position in the 96-well plate (e.g., YSD-BC1). Antibodies that remained attached to the YSD strains were recognized with an anti-human IgG fluorescently labeled antibody. Labeled yeast cells were pooled and sorted into bins according to their fluorescence intensity that is proportional to the number of antibodies attached on the yeast surface. DNA was isolated from the pool of cells present in each bin, and high-throughput sequencing analysis was used to determine the distribution of each barcode among the different bins. The barcode read-count information was used to determine the antibody titer per sample.

[0089] Figure 7. Multiplexed quantification of SARS-CoV2 S and NCP antibody levels using Seroseq. (A) Representative examples of barcode read-count distribution normalized to the total number of reads ("norm_counts") indicative to antibodies targeting the AGA2-scaffold YSD protein used as negative control ("neg. Ctrl."), SARS-CoV2 S-RBD or NCP protein domains. Samples 1 , 2 and 3 correspond to sera obtained from donors pre-pandemic, after COVID-19 infection or from an individual that was COVID-19 vaccinated but did not had the disease, respectively. (B) Correlation of antibody titer determined by YSD-FACS (Singleplex) and Seroseq (Multiplex) methodology for 96 sera screened for antibodies targeting AGA2-scaffold protein (neg. control), S and N protein domains.

[0090] Figure 8. Generation of chromosomally integrated YSD strains. YSD strains were generated using a generic acceptor strain that is transformed with DNA fragments encoding for the antigen of interest. The acceptor strain contains integrated at a chromosomal locus (H0M3) the following acceptor module conformed by: pGPD promoter driving expression of S. cerevisiae a-agglutinin AGA2 subunit, spacer ((GGGGS)s; SEQ ID NO: 10), S. paradoxus CYC1 terminator, l-Scel motif (indicated with a black diamond), S. cerevisiae URA3 (including endogenous promoter and terminator), partial hygromycin resistance-encoding gene (hph) missing the first 435 nucleotides and promoter (hphAN), terminator from S. paradoxus ALG9 gene. In addition, the acceptor strain overexpresses the AGA 1 gene, which encodes for a-agglutinin anchor subunit that is linked to AGA2 via disulfide bonds; and also expresses the l-Scel endonuclease that is under control of the galactose inducible promoter S. cerevisiae GAL1. Yeast cells (grown in galactose to induce l-Scel expression) were transformed with two synthetic DNA fragments encoding for the antigen of interest and a partial sequence of the hph gene missing the last 453 nucleotides (hphAC) [1], Both DNA fragments contain sequences that on one side share homology with the chromosomal integration site (linker and hphAN), and on the other side have homology between both DNA fragments (MYC-tag sequence). l-Scel-induced double strand break was repaired by in vivo homologous recombination, resulting in URA3 gene excision, reconstitution of the hph-gene, and the AGA2-linker-antigen-MYC fusion protein. Yeast cells that underwent proper recombination were selected by their resistance to 5-fluoroorotic acid (5-FOA) and hygromycin.

[0091] Figure 9. Reduction in the number of YSD cells used during S-RBD antibody detection leads to higher number of antibodies bound molecules per yeast cell surface. Cells expressing the S-RBD protein domain (tagged with an HA-tag) were diluted with wild-type yeast cells (empty) as indicated and were stained with a serum from a COVID-19 infected donor (supplemented with rat anti-HA antibody, clone 3F10). Anti-HA and human anti-S-RBD IgG antibodies were recognized with rat-Alexa594 (anti-HA, white cells) and anti-human lgG-Alexa488 (anti-S-RBD, white cells). Cells were either analyzed by fluorescence microscopy (A) or FACS (B).

[0092] Figure 10. Testing reproducibility of YSD-FACS assay. Correlation of two independent YSD-FACS determinations (repeat 1 and 2) of the antibody levels against SARS-CoV2 S-RBD (A) and NCP (B) proteins in human sera (n=12).

[0093] Figure 11. Generation of a barcoded (BC) YSD library using a donor plasmid. YSD acceptor strain was transformed with a barcoded plasmid library encoding the antigen (AG) of interest. Induction of l-Scel endonuclease cleaves at the l-Scel motifs at the chromosome in between the linker and URA3 gene, as well as at two positions in the donor plasmid (indicated with black diamonds). In vivo homologous recombination results in an YSD cassette encoding for the AG of interest, carrying a unique BC upstream the reconstituted hph gene. Cells that successfully recombined are resistant to 5-FOA and hygromycin.

[0094] Figure 12. Testing reproducibility of the Seroseq assay. Quantification of antibody levels (SARS-CoV2 S, NCP and MYC) in human samples (n=18) using Seroseq. Identical samples were tested in duplicate using different barcodes.

[0095] The Examples illustrate the claimed invention.

[0096] Example 1 - Results

[0097] During the COVID-19 pandemic it was decided to establish a yeast-surface-display (YSD)-based methodology that could be used for the detection and quantification of antibodies directed to SARS- CoV2 viral proteins. Ideally, it was aimed to distinguish between people that have been infected, vaccinated (or both) or have not been in contact with the virus (or viral antigens).

[0098] As most EU-approved vaccines use the Spike (S) protein as immunogen, fragments of the S protein were expressed, among them, the Spike receptor binding domain (S-RBD) corresponding to amino acids 331-531 (Fig. 1A). In addition, it was attempted to express different fragments of the SARS-CoV2 Nucleoprotein (NCP), which was expected to elicit an immune response in patients that have been already infected. Use of the YSD platform plasmid pYD1 described by Boder et al (Boder, Nat Biotech 1997) was made, in which the antigen of interest is expressed as a fusion protein of the cell wall protein AGA2. Fluorescence-activated cell sorting (FACS) experiments performed with strains carrying pYD1- derived plasmid encoding for the AGA2-S-RBD fusion protein (tagged at the C-terminus with an HA tag) revealed a specific signal, although with a high variability in terms of expression among the whole yeast population (Fig. 1 B). This expression variability has been documented in the past [2], and it is in part due to inefficient segregation of the plasmids during cell division, leading to cells without plasmid and others with more than 1 copy of the plasmid per cell. To achieve a more homogeneous protein expression, YSD strains were created carrying the YSD expression cassette chromosomally integrated into the yeast genome. To facilitate the generation of YSD strains (without the need to perform molecular cloning), YSD acceptor strains were engineered that permit in a one-step procedure the integration of the antigen directly into a genomic locus, followed by selection for successful recombination resulting in the AGA2-antigen fusion protein (Fig.8). As seen in Fig. 1 B, the chromosomally integrated YSD cassette encoding the AGA2-S-RBD fusion protein resulted in higher and more homogeneous protein expression (according to HA-tag signal, 95% confidence intervals (Cl) for plasmid-based expression=9- 2757; 95% Cl chromosome-integrated=390-5037).

[0099] Next, a protocol was developed to test for the presence of antibodies directed to S-RBD and NCP viral antigens in human sera. Since humans are frequently exposed to yeast and therefore contain antibodies that recognize yeast cells, in particular the surface components of the yeast cell wall, the protocol includes a preadsorption step of the sera that is for the removal from the sera all kind of antibodies that react with yeast cells / components that could interfere with later detection steps. After preadsorption, sera are then incubated with the YSD cells. Human antibodies attached to the yeast surface are recognized with an anti-human IgG fluorescently labeled antibody. To increase the detection sensitivity the number of YSD cells that are incubated with preadsorbed sera was reduced, and they were added instead wild-type ("empty") cells to facilitate the centrifugation of the YSD cells during wash steps. The expression of the antigen was visualized by co-staining with antibodies that recognize the tag (HA or MYC) placed at the very C-terminus of the YSD fusion protein.

[0100] First, fluorescence microscopy was used to visualize antibodies bound to YSD cells expressing either S-RBD, NCP or just the AGA2-scaffold protein. Sera samples represent four groups in regard COVID- 19 serology: negative sera collected before the COVID-19 pandemic (n=14); vaccinated (BioNTech / Comirnaty) but not infected (n=5); infected but not vaccinated (n=5), and vaccinated and infected (n=13). Representative examples of the images are shown in Fig.2. Cells expressing the S-RBD (Fig. 2A) or NCP (Fig.2B) fusion proteins were stained with an anti-MYC antibody (labeled in red). Human sera from COVID-19 infected or vaccinated donors resulted in a positive staining of S-RBD expressing cells (Fig. 2A, S-RBD human IgG, white cells). This signal was not observed in the mock control (no sera), in cells that do not express the antigen (MYC-negative cells) or when negative control sera (before the pandemic) were used. Interestingly, donors from the vaccinated group showed stronger anti-SRBD signal compared to those that were only infected, suggesting that vaccination results in a much higher immune response against S-RBD. On the other hand, only the sera from COVID-19 infected donors (but not those that were only vaccinated) recognized YSD strains expressing the NCP fusion protein. This last observation is in agreement with the fact that most frequently used COVID-19 vaccines in Europe (including Comirnaty, Moderna, AstraZeneca and Janssen) use exclusively the Spike (S) protein as immunogen. These results were further validated for all four groups of donors by using fluorescence- activated cell sorting (FACS) analysis (Fig.3A).

[0101] Besides having homogeneous YSD protein expression (Fig. 1 B), another factor that affects the sensitivity of our method is the number of YSD cells that are incubated with a defined amount of serum. As shown in Fig.9 at low YSD cell density (e.g., 1 :10 dilution or higher), a higher anti-S-RBD antibody signal was observed using fluorescence microscopy (Fig.9A) or FACS (Fig.9B). This indicates that the number of anti-S-RBD antibodies present in serum is rather limited relative to the number of yeast displayed antigens; thus, decreasing the number of YSD cells (for a constant amount of serum) will lead to higher number of attached antibody molecules per yeast cell.

[0102] To learn more about the sensitivity and specificity of this YSD-based methodology, it was compared with an EU-certified enzyme linked immunosorbent assay (ELISA) from the company Euroimmun that is frequently used for SARS-CoV2 serology. Analysis of 57 human sera by using both methodologies resulted in an excellent correlation (r2=0.94) and similar sensitivity (Fig. 3B). By using this data and the linear curve of the graph (y=0.77x-0.66) is possible to convert our YSD-FACS a.u. into "Binding Affinity Units" or BAU, which are frequently used in diagnostic. Further comparisons using serial serum dilutions suggested that the YSD-FACS assay has a broader linear range, especially at high antibody concentrations (Fig. 3C).

[0103] Moreover, the reproducibility of the YSD-FACS assay was tested by performing two independent measurements for SARS-CoV2 S and NCP antibody levels using a set of human sera (n=12) (Fig. 10A and 10B). For both datasets (SARS-CoV2 S and NCP antibody levels) a very good correlation among replicates was observed (r2=0.99 and r2=0.97 for SARS-CoV2 S and NCP, respectively), indicating the high reproducibility of the assay.

[0104] The YSD-FACS methodology was used to test in a small cohort how S-RBD antibody titer changes over time. Forthis, S-RBD antibody levels were measured in a group of donors (n=11) two weeks or 3 months after having received the second vaccination (BioNTech / Comirnaty) (Fig. 4). For this group, the levels of S-RBD antibodies two weeks after the second vaccination showed a mean value of 950 YSD-FACS a.u. and three months after the levels reduced by 63%, with a mean value of 350 YSD-FACS a.u. In addition, in one donor the S-RBD antibody levels were followed-up over a period of 2 years (Fig. 4B). At early time points (pre-pandemic) S-RBD antibodies higher than the threshold observed with the negative control YSD strain were not detect. However, already two weeks after the 2ndBioNTech immunization S-RBD antibodies were detectable. The antibody levels decreased after 3 months and remaining at constant levels for about 6 months. For this donor, an about 5-fold further increase in the titer was observed, 8 days after receiving the 3rdBioNTech immunization (booster), levels that remained constant for at least 40 days (Fig. 4B).

[0105] One additional application of this invention is that it can be used to identify antibodies that recognize specific epitopes within an antigen, for example those epitopes that changed over the course of the pandemic, becoming characteristic of certain mutant variants (e.g., Delta, Omicron, etc.). One pilot experiment that supports this application is presented in Fig. 5. For this type of procedure, instead of using a strain that only expresses the scaffold AGA2 protein, the human sera were either preadsorbed with baker's yeast (empty), a YSD strain expressing the wild-type S-RBD domain or a strain expressing the S-RBD delta variant carrying L452R and T478K mutations. Preadsorbed sera from different donors including: a negative control, COVID-19 vaccinated or vaccinated and infected with the delta variant was used to detect the S-RBD wild-type antigen. Sera preadsorbed with baker's yeast (empty) resulted in the strongest signal for anti-SRBD antibodies. In contrast, sera preadsorbed with cells expressing wild-type S-RBD resulted in complete depletion of S-RBD specific antibodies. Preadsorption performed with the S-RBD delta strain resulted in a strong but incomplete depletion of S-RBD antibodies (Fig. 5A and B). This residual signal is unlikely to be consequence of an incomplete depletion when using the S-RBD delta strain for preadsorption because S-RBD protein expression is not affected by these two mutations (data not shown). Instead, the residual signal corresponds to antibodies that recognize at least one epitope conformed by amino acids L452 or T478 in S-RBD, which is altered by L452R and / or T478K mutations. Interestingly, this residual S-RDB signal is higher in sera from vaccinated (BioNTech) donors compared to the person infected with the delta variant (and vaccinated), despite that the latter had about 3-times higher antibody levels compared to the serum from the vaccinated donors (levels measured after baker's yeast preadsorption). These observations suggest that the SARS-CoV2-delta infected (and vaccinated) donor has a considerable number of antibodies that recognize delta specific epitopes.

[0106] Example 2 - Seroseq: Multiplexed quantification of antibodies in human samples using YSD- FACS and high-throughput sequencing analysis

[0107] It was previously demonstrated by the inventors that by combining YSD and FACS analysis it is possible to quantify SARS-CoV2 S antibodies in human samples over a wider range of sera concentrations than ELISA. However, the simultaneous quantification of antibodies in multiple samples remained difficult, as every antigen has to be analyzed independently. To solve this problem, a multiplexed methodology (Seroseq) was developed that allows the quantification of antibodies in multiple patient samples. A scheme illustrating the most important steps of this technology is shown in Fig. 6. To quantify antibodies directed to one antigen in multiple samples, many different unique barcodes were introduced into an YSD strain expressing one specific antigen (e.g., NCP protein). Then about hundred clones were isolated and sequenced in each one of the clones the individual barcode. These barcoded strains were then used in a 96 well format, where in each well of the 96 well plate contained cells expressing the same antigen but distinguished by well-specific barcodes that are used to "label" preadsorbed sera of 96 patients. The procedure was conducted as before, and the wells were analyzed individually using flow cytometry to quantify the antibody titer as done previously (Fig. 3A). In addition, a fraction of the cells from all the wells were pooled and used fluorescence activated cell sorting (FACS) to subdivide them into sub-populations (termed "bins", e.g., bins 1-8) with similar fluorescence signals. In this way, cells that have low or no human antibodies attached are sorted in bin 1 ; whereas cells loaded with a high number of antibodies are sorted in subsequent bins (2, 3...8), according to the cell fluorescence intensity.

[0108] Next, DNA is extracted from the cells pooled in each bin and is used as template for a PCR reaction to amplify the region containing the barcodes that are linked to the position in the 96-well array of YSD strains and consequently, the patient sample. During this PCR amplification step, additional barcodes are introduced to retain information about the fluorescence intensity bin in which each barcoded yeast strain was sorted. These PCR products are then pooled and analyzed by high-throughput deepsequencing. After demultiplexing and read count of normalization of the sequencing data, the antibody abundance in each individual sample is estimated from the distribution of the corresponding read counts across fluorescence intensity bins. This enables the generation of quantitative profiles for antibodies directed to multiple antigens in many patient samples.

[0109] As proof of principle a cohort of human sera (n=96) was tested for S and NCP antibodies (and AGA2- scaffold, as negative control) with Seroseq as well as with the YSD-FACS protocol. To perform the Seroseq antibody quantification, three YSD-barcoded collections were generated expressing: the AGA2-scaffold protein, the SARS-CoV2 S and the SARS-CoV2 NCP antigen, all three as AGA2-fusion proteins. Each one of these collections consisted in 96 independent strains expressing the same antigen but marked with different barcodes, arrayed in a 96-well plate. These three barcoded (BC) YSD collections were merged into one 96-well plate, in such a way that each 96 well contained three BC- YSD strains (AGA2, AGA2-S and AGA2-NCP) that could be used to track the antigen and the position in the 96-well plate (sample).

[0110] Our cohort of sera included negative and positive controls for SARS-CoV2, some of them in duplicate to test for variations within the assay. The threshold for a positive antibody signal was determined by using YSD strains expressing the AGA2-scaffold protein (negative control).

[0111] Three representative examples after sequencing and data analysis (one negative sera collected prepandemic, one COVID-19 infected and one COVID-19 vaccinated but not infected) are shown in Fig. 7A. The counts for each specific barcode identified after sequencing (normalized to the total number of reads) are shown in the y-axis for each of the eight bins (x-axis).

[0112] For sample 1 , which corresponds to a negative serum for COVID-19, an accumulation of most reads was found in bin 1 , indicative of the absence of antibodies against AGA2, S and NCP antigens. Sample 2, which corresponds to a COVID-19 infected donor, showed an accumulation of reads in bins 4 and 3, for S and NCP antigens, respectively, indicative of the presence of antibodies against both antigens. Sample 3, obtained from a donor that was vaccinated but had no COVID-19 infection, showed an accumulation of reads for the S antigen in bin 7, indicative of higher levels of anti-S-RBD antibodies compared to sample 2. For sample 3, the reads corresponding to the NCP antibodies are mainly found in bin 1 , indicative of the lack of this type of antibodies. All these observations are in agreement with our previous findings (Fig. 2 and Fig. 3A) and support the specificity and sensitivity of the assay. All the data obtained for this cohort using the Seroseq and YSD-FACS method was compared side-by-side, showing an excellent correlation (Fig. 7B, r2=0.95).

[0113] Thus, herein it is shown that the Seroseq methodology can be used to quantify antibodies directed to different antigens in many samples (n=96), in a pooled analysis. For future applications it will be important to determine the maximum plexing level that can be achieved, without compromising the quality of the data.

[0114] Previous studies have demonstrated that it is feasible to use barcoded yeast libraries with a complexity of 5.000-10.000 strains, for sorting and later sequencing of the barcodes to obtain information about protein stability [3], In contrast to this previous application where the complexity was given mainly by the number of strains present in the library, in our Seroseq methodology the complexity is the result of two variables: the number of YSD antigens and the number of samples to be analyzed. Based on this, it is likely that Seroseq will reach at least a multiplex level of about 50-100 antibodies that can be measured simultaneously in 100 samples per run.

[0115] Example 3 - Methods

[0116] Yeast strains and plasmids

[0117] All strains used in this study are listed in Table 1 and derived from ESM356-1 [4] (S. cerevisiae S288C, MATa ura3-52 leu2A1 his3A200 trp1A63, which is a spore from strain FY1679).

[0118] Yeast surface display was performed using the AGA1-AGA2-system [5] with some modifications. With exception of the strain used in in Fig. 1 B that contains a plasmid (pYD1) encoding for the YSD protein, all other strains harbor a YSD cassette integrated into the genome. For plasmid based-expression of SARS-CoV2-S-RBD-331-531 aa-HA antigen (Fig.l B), the plasmid pHHB1380 was transformed into the HHY8151 strain, that overexpress the AGA1 under the control of the GAL1 promoter (generated with plasmid pl U211 as previously described by Boder et al. [5]). Additional strains used in this study, with exception of strains used in Fig. 1 B, have been modified to overexpress the AGA1 gene and the AGA2-antigen fusion protein, by placing both genes under control of the constitutive pGPD promoter. The strain HHY8556 (used in Fig. 1 B) that expresses the AGA2-S- RBD-331-531 aa-HA fusion protein was generated with the acceptor strain HHY8465 co-transformed with two PCR fragments: one using as template plasmid pHHB1435 and primers HHP4808 (5'- CCCCATAAACACACAGTATG-3') and HHP4539 (5'-GTACGAGCTAAAAGTACAGTG-3'), and a second PCR using as template plasmid pHHB1400 and primers HHP4801 (5'-

[0119] ACTACATGGCGTGATTTCATATG-3') and HHP4815 (5'-GCTAGCACACCATTCCGCC-3'). Transformants were selected for 5-FOA and hygromycin resistance.

[0120] YSD expression of the AGA2-S-RBD-331-531 aa-MYC fusion protein (Figs.2A, 3A-3C and S3A) was done with strain HHY8805, generated using the acceptor strain HHY8699 transformed with the donor plasmid pHHB1487 (encoding for S-RBD-331-531 aa-MYC), induced on galactose for l-Scel cleavage and selected on hygromycin and 5-FOA resistance.

[0121] YSD expression of the AGA2-N-2-178aa-MYC fusion protein (Figs. 2B, 3A-3C and S3B) was done with strain HHY8809, generated using the acceptor strain HHY8699 transformed with the donor plasmid pHHB1489 (encoding for N-2-178aa-MYC), induced on galactose for l-Scel cleavage and selected for hygromycin and 5-FOA resistance.

[0122] YSD expression of the AGA2-MYC fusion protein (used as negative control) (Figs. 3A-3C) was done with strain HHY8793, generated using the acceptor strain HHY8699 transformed with the donor plasmid pHHB1485 (encoding for MYC), induced on galactose for l-Scel cleavage and selected for hygromycin and 5-FOA resistance.

[0123] The strain used for sera preadsorption (Figs. 2A, 3A-3C) that expresses the AGA2-scaffold (but no tag / antigen) was generated as described above using as acceptor strain HHY8699 and as donor plasmid pHHB1486.

[0124] The YSD strain HHY8693 expressing AGA2-S-RBD-331-531 aa-HA fusion protein (Fig. 5A-5B) was generated with the acceptor strain HHY8631 and the donor plasmid pHHB1466, induced on galactose for l-Scel cleavage and selected for hygromycin and 5-FOA resistance. The YSD strain HHY8695 that express the AGA2-S-RBD-delta variant (Fig. 5B) was generated as previously , using HHY8631 as acceptor strain and pHHB1467 as donor plasmid. Expression of the AGA2-S-RBD-HA (Figs. 4A-4B, and S2A-S2B) was done with strain HHY8559 generated from HHY8397 co-transformed with two PCR fragments: one using as template pHHB1390 and primers HHP4808 (5'-CCCCATAAACACACAGTATG-3') and HHP4539 (5'-

[0125] GTACGAGCTAAAAGTACAGTG-3'), and a second PCR using pHHB1392 as template and primers HHP4801 (5'- ACTACATGGCGTGATTTCATATG-3') and HHP4815 (5'-GCTAGCACACCATTCCGCC- 3'). Plasmids used in this study are listed in Table 2. Background strain and acceptor module

[0126] The acceptor strain used for the generation of strains carrying chromosomally integrated YSD cassettes corresponds to HHY8699 (MATa ura3-52::HIS3 leu2A1::hisG.pGAL1-l-Scel trp1A63 his3A200 natNT2.pGPD-AGA 1 H0M3:: pGPD-AGA2-l-SCEI-tCYC-URA3-hphAN-Talg9).

[0127] The strain HHY8699 contains several genes (acceptor module) that facilitate the integration of a DNA fragment encoding forthe antigen, and the selection of correct recombinants. The acceptor module was amplified as one PCR fragment from the plasmid pHHB1468 with primers HHP4744 (5'- CTATCGCATTTTATCACAGAAGCTTTCATTTTTTTTAACTTTTACgagctcagtttatcattatcaa-3') and

[0128] HHP4874 (5'-

[0129] GGGAAAATTAATATTCTATCATTAAAGTGAAGAAGAAAGGTGGAggcgttgaggatccatgacacact-3'), where the sequence in lowercase binds to the plasmid template and the remaining sequence correspond to sequences used for homolog recombination at the H0M3 locus.

[0130] The acceptor module consists of the following elements:

[0131] • the AGA2 gene under the control of the constitutive GPD promoter

[0132] • (GGGGS)3linker

[0133] • terminator sequence of the CYC1 gene from S. paradoxus

[0134] • the recognition sequence for the l-Scel endonuclease

[0135] • the URA3 gene with its endogenous promoter and terminator

[0136] • a fragment of the hygromycin resistance gene (amino acids 146-342) missing part of the N- terminal region (hphAN)

[0137] • terminator sequence of the ALG9 gene from S. paradoxus

[0138] The acceptor strain HHY8699 was used for the generation of YSD strains expressing either the SARS- CoV2-S and N epitopes or just the expression of the AGA2 platform, used as negative control.

[0139] Addition of galactose induces the expression ofthe l-Scel endonuclease, creating a double strand break at the l-Scel recognition site in between the CYC1 terminator and URA3 gene (Figure 8).

[0140] The acceptor strain was transformed either with DNA fragments encoding for the antigen of interest (Fig.8) or with a barcoded-plasmid-library encoding SARS-CoV2-S, -N or the scaffold display platform AGA2-MYC (Fig.11). After induction of the l-Scel endonuclease, the double strand break at the chromosomal locus (H0M3) is repaired by homologous recombination using as template the transformed DNA fragments or barcoded-donor plasmids. The in vivo repair ofthe l-Scel-induced double strand break leads to excision ofthe URA3 gene and reconstitution ofthe hph gene. Cells that underwent successful recombination were selected for 5-FOA and hygromycin resistance. Cells carrying the YSD cassette drive expression of the AGA2-linker-antigen-MYC fusion protein.

[0141] Donor plasmids

[0142] The following donor plasmids were used for the construction of barcoded plasmid libraries.

[0143] • pHHB1485: donor plasmid encoding for the AGA2-MYC-based surface display platform

[0144] • pHHB1489: donor plasmid encoding for AGA2-SARS-CoV2-N-2-178aa-MYC (for strains displaying SARS-CoV-2 nucleocapsid protein) pHHB1525: donor plasmid with SARS-CoV2-S-RBD-331-531 aa (for strains displaying SARS-

[0145] CoV-2 spike protein)

[0146] All the donor plasmids contain the following elements:

[0147] • the kanMX selection marker

[0148] • the short recognition sequence for the l-Scel endonuclease

[0149] • truncated sequence of AGA2 ORF (23-261 nt)

[0150] • GS linker

[0151] • MYC tag

[0152] • terminator sequence of alpha-mating factor leader from S. cerevisiae

[0153] • EcoRI + Xmal restriction enzyme sites for cloning barcodes

[0154] • hphAC sequence coding for a N-terminal fragment (amino acids 1-145) of the hph marker

[0155] • the recognition sequence for the l-Scel endonuclease

[0156] • terminator sequence of the SMC4 gene from S. paradoxus

[0157] • the URA3 gene with its endogenous promoter and terminator from K. lactis

[0158] Generation of barcoded donor plasmid libraries and barcoded yeast libraries

[0159] A DNA oligo library (Sigma) of 69 bp containing a central 30-mer random sequence (5'- CACGACGCTCTTCCGATCTNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNAGATCGGAAGAGCA CACGTC-3') was amplified using Herculase II fusion DNA polymerase with primers: 5'- CTAGTCgaattcACACTCTTTCCCTACACGACGCTCTTCCGATCT-3' and 5'-

[0160] CAGATGcccgggGACTGGAGTTCAGACGTGTGCTCTTCCGATCT-3'. To reduce potential bias introduced by PCR amplification, ten independent PCR reactions (25 pl each) were performed. Cycling conditions were: 95 °C 3' denaturation, followed by 12 cycles of 95 °C 20" denaturation, 62 °C 20" annealing, 68 °C 30" extension, and a final 68 °C 4 min extension hold. The reactions were pooled and cleaned-up using NucleoSpin Gel and PCR clean-up columns. The amplified barcodes were cut with EcoRI and Xmal and subcloned into plasmids pHHB1485, pHHB1489 and pHHB1525, previously digested with EcoRI and Xmal. Ligation reactions were electroporated into E. coli Topl OF'. At least 1.8x104independent E. co / / transformants were obtained for each plasmid library. E. co / / transformants were pooled and used for plasmid isolation. DNA plasmid libraries were transformed into the yeast acceptor strain H8699 (previously grown in the presence of galactose to induce l-Scel expression) and were selected on YPD+200 mg / L G418. At least 120 independent transformants (per library) were grown in 2% glucose-complete synthetic media supplemented with 5-FOA (1 g / L) and hygromycin (300 mg / L) to select forthe correct recombinants. Single clones were isolated and the region containing the barcode was amplified by PCR with primers HHP5013 (5'-CGTCAAGACTGTCAAGGAG-3') and HHP5014 (5'- CTGCCTGCCATATTTCAATTTG-3') and Sanger sequenced. 96 validated strains were arranged in a 96-well format before storage at -80°C. Preparation of barcoded yeast surface display libraries (sufficient to prepare 24 x 96 well plates expressing AGA2-MYC, AGA2-S and AGA2-N epitopes)

[0161] Yeast libraries were pinned from the 96-well plates into YPD-agar plates using the Sanger RoTor, and were incubated overnight at 30°C. Next day, cells were pinned into 96-well flat bottom plates containing 150 pl YPD and were grown overnight at 30°C under shaking at 750 rpm. Next day, 50 pl of the overnight cultures were transferred to a 96-deep well plate (96 x 1 ml) using a Liquidator-200. The cell density across of the whole plate was measured using TECAN and Nanodrop instruments. If necessary, wells that have eithertoo high or low number of cells were adjusted. Cultures were grown at 30°C for 5h under shaking (750 rpm). In addition, a 200 ml overnight culture of the empty strain ESM356-1 (used as cellcarrier in our analysis) was diluted 10-times and the culture was grown also for 5h. After growing the ESM356-1 cells for 5h, 3.15x106cells were transferred with a Liquidator96 to a 96-well V-shape plate. At the same time, 4.5x104cells of each barcoded strain (AGA2-MYC, AGA2-S and AGA2-N) were transferred to the previous plate containing the ESM356-1 cells. After mixing these cells, each well contained 70% of the empty strain (ESM356-1) and 10% of AGA2-MYC, 10% of AGA2-S and 10% of AGA2-N strains carrying specific barcodes used to distinguish among three antigens (AGA2-MYC, AGA2-S or AGA2-N) and 96 different samples. Plates were centrifuged 5 min at 4.000 rpm, the supernatant was removed with a Liquidator96 and the plates were stored at -80°C until were used for quantification of antibodies.

[0162] Singleplex-quantification of antibodies in human samples using yeast surface display and flow cytometry (YSD-FACS)

[0163] Preadsorption of human samples (blood, sera or plasma) using yeast cells was performed as follows. 2.5 pl blood, serum or plasma was incubated with 5x108yeast cells (strain HHY8796 that express the AGA2-surface display platform, excluding the MYC-tag and antigen) in 250 pl 1X TBS (containing 0.5 pg / ml anti-MYC antibody, clone 4A6) for 2 hours at 4°C under rocking. Next, samples were centrifuged for 5 min at 4.000 rpm and 50 pl of the supernatant (preadsorbed sera) was transferred with a Liquidator96 to the wells containing the yeast surface display strains expressing the antigens of interest. Yeast surface display cells were resuspended with the preadsorbed sera by using the Liquidator96. Plate containing the yeast-sera suspension was rotated overnight at 4°C using a Reax2 overhead shaker. Yeast cells were centrifuged 5 min at 4.000 rpm, washed four times with 150 pl TBS, resuspended in 50 pl TBS containing anti-human lgG-Alexa488 (SA5-10126) and anti-mouse H+L- Alexa594 (A11032) (both at a final dilution of 1 :100), and incubated for 30 min at 4°C. Yeast-antibody suspension was washed four times with 150 pl TBS, resuspended in 150 pl TBS and were analyzed using a BD FACS Canto™ Flow cytometer with a high-throughput sampler (HTS) module, equipped for the detection of green fluorescent proteins (excitation: 488nm, long pass 505, bandpass:530 / 30) and red-fluorescent proteins (excitation: 561 nm, long pass 600 , bandpass: 609 / 62). Rainbow Calibration particles (Sphero TM, 556286) were included in every experiment for calibration / normalization purposes. Fluorescence microscopy

[0164] Cells were inoculated in 5 ml SC medium and grown overnight. Next morning cells were diluted to OD6OO= 0.2 in 20 ml SC medium and grown for 5 hours. Yeast cells were incubated for 30 min at room temperature with preadsorbed seras (and anti-MYC antibody) according to protocol described above. Next, cells were washed 3 times with TBS and were then incubated for 30 min at 4°C with anti-human IgG-Alexa 488 (SA5-10126) and anti-mouse H+L-Alexa 594 (A11032) secondary antibodies, both used at a 1 :100 dilution in TBS. Cells were washed 3 times with TBS, and then attached to glass-bottom 96- well microscopy plates (MGB096-1-2-LG-L, MatriPlate) using Bioconext / Concanavalin A coating, as described previously [6], Images were taken with a Nikon Ti-E epifluorescence microscope equipped with a 60x ApoTIRF oil-immersed objective (1.49 NA, Nikon), a 2048 x 2048 pixel (6.5 pm), an sCMOS camera (Flash4, Hamamatsu), and an autofocus system (Perfect Focus System, Nikon) with bright field, 469 / 35 excitation and 525 / 50 emission filters, or 542 / 27 excitation and 600 / 52 emission filters (all from Semrock, with exception of 525 / 50 which is from Chroma).

[0165] Multiplexed quantification of antibodies using FACS sorting and sequencing analysis

[0166] The multiplexed quantification of antibodies was performed as described previously for YSD-FACS- quantification with the following modifications. After the third wash step of the secondary antibodies, all the samples were pooled and washed one more time with 14,4 ml TBS. Cells were centrifuged and the yeast pellet was resuspended in 4 ml TBS for further FACS-sorting using a BD FACS Aria™ III Cell Sorter equipped for the detection of green fluorescent proteins (excitation: 488nm, long pass 502, bandpass: 530 / 30) and red-fluorescent proteins (excitation: 561 nm, long pass 600, bandpass: 610 / 20).

[0167] A total of 2x106MYC+ (Alexa594) cells were sorted in 8 fractions (bins) according to the log - transformed fluorescence emission intensity on the Alexa488 channel (anti-human IgG signal). The number of sorted cells per bin was adjusted according to the percentage of cells found in each bin. Sorted cells were grown overnight in YPD medium and then harvest for genomic DNA isolation.

[0168] In addition, an aliquot of the cells was used to check for cell viability by sorting 384 single cells into an YPD-agar plate. Cell viability was about 40% and was calculated based on the number of viable colonies after 3 days incubation at 30°C.

[0169] Genomic DNA isolation

[0170] Yeast cells were resuspended in 500 pl S-buffer (10mM K2HPO4 pH 7.2, 10 mM EDTA, 50 mM p- mercaptoethanol) and incubated with zymolyase 100T at 37°C for 30 min. Then, 100 pl lysis buffer (25 mM Tris-HCI pH 7.5, 25 mM EDTA, 2.5% (w / v) SDS) were added, vortexed and incubated at 65°C for 30 min. Proteins were precipitated by addition of 166 pl 3M potassium acetate, followed by 10 min incubation on ice and centrifugation for 10 min at 4°C 14.000 rpm. DNA was precipitated from the supernatant by addition of 800 pl cold 100% ethanol and centrifugation. The DNA pellet was washed with 500 pl 70% ethanol (v / v) and resuspended in 40 pl dH2O. Next, the DNA was treated with 1 pl RNase A 10 mg / ml at 37°C for 30 min, followed by addition of 800 pl 70% EtOH and centrifugation. The DNA pellet was resuspended in 25-50 pl dH2O. Amplification of barcodes and Nanopore sequencing

[0171] The DNA obtained from each bin (pool of cells) was used as template for PCR amplification. A 450 bp genomic region containing a 30-mer barcode (unique for each antigen and position per well) was amplified using NEB LongAmp Hot Start 2x master mix and primers containing additional barcodes used to distinguish bins 1-8. A list containing primers used for the barcoding (in addition to the ones in Nanopore PCR kit SQK-PSK004) are indicated in Table 3. PCR reactions were conducted using 2-4 pl genomic DNA, and the following PCR settings: 94°C T denaturation, followed by 25 cycles: 94°C 30" denaturation, 55°C 30" annealing, 65°C 30" extension, followed by 65°C 5'. Equivalent amount of PCR products for bins 1-8 were pooled, and 50 pl of this pooled PCR was purified using AMPure XP magnetic beads.

[0172] Purified DNA was quantified with Qubit fluorometer and 30 ng were used for a second PCR reaction using Nanopore PCR kit (SQK-PSK004). Following PCR settings were used: 94°C 1 ' denaturation, followed by 30 cycles: 94°C 30" denaturation, 62°C 30" annealing, 65°C 30" extension, followed by 65°C 5'. The PCR was purified with AMPure XP magnetic beads and 100 fmol of the PCR product were used for sequencing library preparation according to manufacture's protocol (Oxford Nanopore Technologies). DNA library was loaded on a MinlON Mk1 b device with a SpotON Flow Cell.

[0173] Analysis of Nanopore sequencing data and read counting

[0174] Each one of the DNA fragments analyzed by Nanopore sequencing contain three unique barcodes: one internal barcode (30N) characteristic of the antigen and the position in the 96-well plate, and two flanking barcodes (24N) that allow the assignment antigen / sample to one of the eight fluorescence intensity bins.

[0175] MinlON data were base-called using the Albacore Sequencing pipeline Software v2.0 (Oxford Nanopore Technologies). For data analysis, a custom script was used to extract and de-multiplex informative sequence segments from all the reads based on the identification of the internal barcode (30N) and at least one of the distal barcodes (24N) characteristic of each bin.

[0176] Table 1. Yeast strains used in the examples

[0177] Table 2. List of plasmids used in the examples.

[0178] Table 3. List of primers used for Nanopore sequencing of the barcoded yeast surface display strains.

[0179] Underlined is indicated sequence corresponding to Nanopore ONG adaptors. In lower capital is indicated sequence homology upstream (Mat-alpha terminator, forward) and downstream (pTEF promoter, reverse) of the barcode located downstream the yeast surface display cassette.

[0180] Literature

[0181] 1. Meurer, M., Duan, Y.Q., Sass, E., Kats, I., Herbst, K., Buchmuller, B.C., Dederer, V., Huber, F., Kirrmaier, D., Stefl, M., et al. (2018). Genome-wide C-SWAT library for high-throughput yeast genome tagging. Nat Methods 15, 598-600.

[0182] 2. Jensen, N.B., Strucko, T., Kildegaard, K.R., David, F., Maury, J., Mortensen, U.H., Forster, J., Nielsen, J., and Borodina, I. (2014). EasyClone: method for iterative chromosomal integration of multiple genes in Saccharomyces cerevisiae. Fems Yeast Res 14, 238-248. 3. Kats, I., Khmelinskii, A., Kschonsak, M., Huber, F., Kniess, R.A., Bartosik, A., and Knop, M. (2018). Mapping Degradation Signals and Pathways in a Eukaryotic N-terminome. Mol Cell 70, 488-+.

[0183] 4. Knop, M., and Schiebel, E. (1998). Receptors determine the cellular localization of a gammatubulin complex and thereby the site of microtubule formation. Embo J 17, 3952-3967. 5. Boder, E.T., and Wittrup, K.D. (1997). Yeast surface display for screening combinatorial polypeptide libraries. Nat Biotechnol 15, 553-557.

[0184] 6. Khmelinskii, A., Knop, M. (2014). Analysis of protein dynamics with tandem fluorescent protein timers. Methods Mol Biol. 1174:195-210.

Claims

CLAIMS1 . A multiplexed method for the quantification of antibodies in a plurality of samples, said antibodies being capable of binding to an antigen of interest in a plurality of samples, wherein each of the plurality of samples is in a separate vessel or well of a multi-well plate comprising(a) contacting each of the plurality of samples with yeast cells carrying a yeast-surface display expression cassette in its genome, wherein the yeast-surface display expression cassette encodes a fusion protein comprising the antigen of interest being fused to a cell wall or plasma membrane anchored protein of yeast, and wherein the yeast cells for each sample of the plurality of sample can be distinguished from all other yeast cells from all other of the plurality of samples by a unique DNA barcode being present in the yeast cells for each sample,(b) separating the yeast cells from the sample, preferably by centrifugation, or filtration or magnetic beads that bind to the surface of the yeast cells,(c) contacting the yeast cells with a fluorescently labelled antibody binding to antibodies bound to the antigen of interest,(d) pooling the yeast cells after step (b) or (c), preferably after step (c),(e) sorting the yeast cells into different sub-pools based on the fluorescent intensity of each fluorescently-labeled yeast cell, wherein the fluorescent intensity of each cell is proportional to the number of antibodies being bound to the antigen of interest on the surface of the cell and optionally growing the sub-pools of cells,(f) isolating the DNA from the yeast cells in each sub-pool,(g) subjecting the isolated DNA of each sub-pool to sequencing of each of the unique DNA barcodes, thereby quantifying the relative frequency of individual barcodes in each subpool and thereby identifying the samples being present in each of the subpools, and(h) quantifying the abundance of antibodies bound to the antigen of interest in each of the plurality of samples based on the mean fluorescent intensity measured for each yeast cell population expressing the antigen of interest and carrying a unique barcode, according to the distribution of the barcodes across the different fluorescent intensity bins for each sample, wherein the mean fluorescent intensity is proportional to the abundance of antigenspecific antibodies being present in each sample.

2. The method of claim 1 , wherein the fusion protein further comprises an epitope tag, preferably at its C-terminus, wherein the epitope tag is preferably a MYC tag, a FLAG tag or a V5 tag.

3. The method of claim 1 or 2, wherein the cell wall or plasma membrane anchored protein of yeast is the Aga2p protein of Saccharomyces cerevisiae, wherein the Aga2p protein preferably comprises SEQ ID NO: 1 or a sequence being at least 80% identical thereto.

4. The method of any one of claims 1 to 3, wherein the sorting in step (d) comprises fluorescentactivated cell sorting (FACS).

5. The method of any one of claims 1 to 4, wherein the sequencing in step (f) is next generation sequencing, preferably Nanopore or Illumina sequencing.

6. The method of any one of claims 1 or 5, wherein the barcode is a nucleotide sequence of 20 to 50 nucleotides.

7. The method of any one of claims 1 to 6, wherein the number of sub-pools is 4-24, preferably 6- 14 and most preferably 8-12.

8. The method of any one of claims 1 to 7, wherein after step (f) and before step (g) the part of isolated DNA comprising the unique DNA barcode is amplified by PCR, wherein the primers used in this amplification preferably comprise barcodes that allow distinguishing all amplicons from one sub-pool form all amplicons from all other sub-pools, wherein said barcode is preferably the DNA- sequence(s) at the 3’ and / or 5’-end of the amplicons.

9. The method of any one of claims 1 to 8, wherein the unique DNA barcodes are unique plasmid DNA barcodes or unique genomic DNA barcodes and are preferably unique genomic DNA barcodes.

10. The method of any one of claims 1 to 9, wherein the yeast cells further comprise an epitope tag according to claim 2, wherein the yeast cells are contacted with a fluorescently labelled antibody binding to the epitope tag, wherein the fluorescent intensity of each cell is proportional to the number of fusion proteins being displayed on the surface of the cell, and wherein the fluorescent label of the antibody binding to the epitope tag is distinct from the fluorescent label of the antibody binding to antibodies bound to the antigen.

11. The method of any one of claims 1 to 10, wherein the sample is (i) a biological sample or an industrial sample, or (ii) is tissue or body fluid, wherein the body fluid is preferably blood, plasma, serum, cerebrospinal fluid or saliva.

12. The method of any one of claims 1 to 11 , wherein the sample is a preadsorbed sample that has been depleted for antibodies that bind to(i) yeast surface components;(ii) one or more elements of the yeast-surface display expression cassette, wherein said one or more elements are devoid of the antigen of interest;(iii) selected (sub-)domains of the antigen of interest, and / or(iv) a protein containing at least one amino acid substitution or post-translational modification relative to the antigen of interest.

13. The method of claim 12, further comprising the step of preparing the preadsorbed sample by contacting the sample with yeast cells that(i) do not display the antigen of interest; (ii) display only selected (sub-)domains thereof; and / or(iii) display a protein containing at least one amino acid substitution or post-translational modification relative to the antigen of interest on their surface.

Images