EP4162069A1

EP4162069A1 - Method of single-cell analysis of multiple samples

Info

Publication number: EP4162069A1
Application number: EP21730550.7A
Authority: EP
Inventors: Carolina DALLETT; Razika Hussein; Garry P. Nolan; Maeve E. O'HUALLACHAIN; Sedide OZTURK; Sri Ramakrishna PALADUGU
Original assignee: F Hoffmann La Roche AG; Roche Diagnostics GmbH
Current assignee: F Hoffmann La Roche AG; Roche Diagnostics GmbH
Priority date: 2020-06-08
Filing date: 2021-06-01
Publication date: 2023-04-12
Also published as: WO2021249816A1

Abstract

The invention provides a method and compositions for single cell analysis of cells present in multiple samples assisted by sample pooling and sample barcoding.

Description

Method Of Single-Cell Analysis of Multiple Samples

FIELD OF THE INVENTION

The invention relates to the field of single cell analysis. More specifically, the invention relates to the use of barcodes in single cell analysis.

BACKGROUND OF THE INVENTION

Single-cell analysis by combinatorial indexing techniques like Quantum Barcoding (QBC) facilitates single-cell barcoding of large numbers of cells with higher cell throughput. The Quantum Barcoding method is disclosed e.g., in the U.S. Patent No. 10,144,950. QBC allows analysis of large numbers of individual cells at reduced cost compared with other NGS-based single-cell analysis technologies. However, some configurations of QBC are difficult to apply to samples having a small number of cells. It is desirable to be able to pool such limited sample volumes for performing the QBC analysis.

While methods of barcoding cellular samples are known in the fields of flow cytometry and mass cytometry, none have been applied in the context of combinatorial nucleic acid labelling such as QBC.

There is a need for a method of combinatorial single-cell analysis that enables pooling multiple samples of cells.

SUMMARY OF THE INVENTION

The invention is a method of single-cell analysis wherein multiple samples are labeled with a sample barcode and pooled into a single sample.

In some embodiments, the invention is a method of detecting a plurality of targets in a plurality of cells originating from multiple samples, the method comprising: binding to one or more (e.g., a plurality of) targets in a plurality of cells in multiple samples one or more of unique binding agents that are each specific for one of the targets; attaching a sample-identifying oligonucleotide to the unique binding agent; pooling the multiple samples to form a pooled sample; adding multiple subcode oligonucleotides to the sample-identifying oligonucleotide in the plurality of cells in an ordered manner during successive rounds of split-pool synthesis wherein the subcode oligonucleotide in each round anneals adjacently to the subcode oligonucleotide from a previous round via an annealing region, and covalently linking the adjacently annealed subcode oligonucleotides to each other to create in each cell, a unique cell-originating barcode; and detecting the unique cell-originating nucleotide code consisting of the unique binding agent, the sample-identifying barcode, and the unique cell-originating barcode thereby detecting the presence of the targets in the plurality of cells and identifying the sample of origin of the cells. Detecting may be by sequencing with an optional pre-sequencing amplification step.

In some embodiments, in each round of split-pool synthesis, the subcode oligonucleotide anneals to the annealing region located on a splint oligonucleotide adjacently to the annealed subcode oligonucleotide from a previous round and is ligated to the annealed subcode oligonucleotide from the previous round. In some embodiments, one round of split-pool comprises: splitting the pooled sample into multiple reaction volumes, each volume containing a subcode oligonucleotide; attaching the subcode oligonucleotide in the reaction volume; and pooling the reaction volumes.

In some embodiments, the unique binding agent comprises an antibody, e.g., an antibody conjugated to an antibody-identifying nucleic acid barcode, which is detected by sequencing.

In some embodiments, the sample-identifying oligonucleotide attaches to the antibody-identifying nucleic acid barcode, e.g., by annealing a splint oligonucleotide to the antibody-identifying nucleic acid barcode, wherein the splint oligonucleotide comprises an annealing region for the sample-identifying oligonucleotide in ligatable proximity to the antibody-identifying nucleic acid barcode; annealing the sample- identifying oligonucleotide to the splint oligonucleotide; and ligating the antibody- identifying nucleic acid barcode to the sample-identifying oligonucleotide.

In some embodiments, the unique binding agent is a nucleic acid probe. In some embodiments, the sample-identifying oligonucleotide attaches to the nucleic acid probe, e.g., by annealing a splint oligonucleotide to the nucleic acid probe, wherein the splint oligonucleotide comprises an annealing region for the sample-identifying oligonucleotide in ligatable proximity to the nucleic acid probe; annealing the sample-identifying oligonucleotide to the splint oligonucleotide; and ligating the nucleic acid probe to the sample-identifying oligonucleotide.

In some embodiments, the invention is a method of detecting a plurality of targets in a plurality of cells originating from multiple samples, the method comprising: binding to one or more targets in a plurality of cells in multiple samples a one or more of unique binding agents that are each specific for one of the targets; attaching to the plurality of cells in each of the multiple samples a sample-identifying oligonucleotide; pooling the multiple samples to form a pooled sample; adding multiple subcode oligonucleotides to the unique binding agent and to the sample- identifying barcode in the plurality of cells of the pooled sample in an ordered manner during successive rounds of split-pool synthesis wherein the subcode oligonucleotide in each round anneals adjacently to the subcode oligonucleotide from a previous round via an annealing region, and covalently linking the adjacently annealed subcode oligonucleotides to each other to create in each cell, a unique cell- originating barcode; detecting in each cell two unique cell-originating nucleotide codes: the first code consisting of the unique binding agent and the unique cell- originating barcode, and the second code consisting of the sample-identifying barcode and the same unique cell-originating barcode, thereby detecting the presence of the targets in the plurality of cells and identifying the sample of origin of the cells.

In some embodiments of this method, the unique binding agent is a nucleic acid probe. In some embodiments of this method, the unique binding agent comprises an antibody conjugated to an antibody-identifying nucleic acid barcode. In some embodiments, the unique binding agent is a primary antibody and the sample- identifying nucleic acid is attached to a secondary antibody capable of binding to the unique binding agent. In some embodiments, the secondary antibody is class-specific for the class of antibodies that includes the unique binding agent. In some embodiments, the secondary antibody is species-specific for the species from which the unique binding agent is derived. In some embodiments, the sample-identifying oligonucleotide is attached to the surface of the cells. In some embodiments, the sample-identifying oligonucleotide is conjugated to a moiety capable of interacting with the surface of the cells. The moiety can be biotin and the method further comprises coating the surface of the cells with streptavidin prior to contacting the sample with a plurality of sample- identifying oligonucleotide. The moiety can also be a fatty acid residue or a cholesterol moiety capable of forming a hydrophobic interaction with the membrane of the cells. The moiety can also be a maleimide moiety capable of reacting with amino groups present in cell membrane proteins of the cells. The moiety can also be a phosphine moiety capable of reacting with carbohydrate residues associated with cell membrane proteins of the cells.

In some embodiments, the sample identifying nucleic acid barcode is conjugated to an antibody specific to the type of cells including the plurality of cells in which the plurality of targets are to be detected. In some embodiments, the sample identifying nucleic acid barcode is conjugated to an anti-CD45 antibody and the plurality of cells in which the plurality of targets are to be detected are mammalian leukocytes. In some embodiments, the sample identifying nucleic acid barcode is conjugated to an anti-CD3 antibody and the plurality of cells in which the plurality of targets are to be detected are mammalian T-cells. In some embodiments, the sample identifying nucleic acid barcode is conjugated to an anti-CD298 antibody and to an anti-beta2 microglobulin antibody and the plurality of cells in which the plurality of targets are to be detected are human cells. In some embodiments, the sample identifying nucleic acid barcode is conjugated to an anti-VAMP7 antibody and the plurality of cells in which the plurality of targets are to be detected are human cells.

In some embodiments of this method, the subcode oligonucleotides anneal to annealing regions located on a splint oligonucleotide adjacently to the annealed subcode oligonucleotide from a previous round and are ligated to the annealed subcode from the previous round. In some embodiments of this method, one round of split-pool comprises: splitting the pooled sample into multiple reaction volumes, each volume containing a subcode oligonucleotide; attaching the subcode oligonucleotide in the reaction volume; and pooling the reaction volumes. In some embodiments of this method, the splint oligonucleotide anneals to the unique binding agent and separately, to the sample-identifying oligonucleotide and the first subcode oligonucleotide anneals adjacently to the unique binding agent and separately, adjacently to the sample-identifying oligonucleotide.

In some embodiments, the invention is a kit for detecting a plurality of targets in a plurality of cells originating from multiple samples, the kit comprising: a unique binding agent capable of binding to a cellular target; a set of sample-identifying oligonucleotides; a set of nucleic acid subcodes; a splint oligonucleotide comprising annealing regions capable of annealing to the annealing regions of the subcodes, and annealing regions of the sample-identifying oligonucleotides. In some embodiments, the unique binding agent in the kit is an antibody conjugated to an antibody- identifying nucleic acid barcode comprising an annealing region for the splint oligonucleotide. In some embodiments, the kit further comprises a secondary antibody conjugated to the sample-identifying oligonucleotide, wherein the secondary antibody is capable of binding to the unique binding agent. In some embodiments, the sample-identifying oligonucleotides in the kit are conjugated to a sample-tagging antibody specific for the type of cells including the plurality of cells in which the plurality of targets are to be detected. In some embodiments, the sample-tagging antibody in the kit is selected from an anti-CD45 antibody, an anti- CD3 antibody, an anti-CD298 antibody, an anti-beta 2 microglobulin antibody, and an anti-VAMP7 antibody.

In some embodiments, the unique binding agent in the kit is a nucleic acid probe. The nucleic acid probe may comprise a target-specific region and an annealing region capable of annealing to the splint oligonucleotide.

In some embodiments, the sample-identifying oligonucleotides in the kit comprise an annealing region capable of annealing to the unique binding agent.

In some embodiments, the sample-identifying oligonucleotides in the kit are capable of binding to the surface of the cells. In some embodiments, the sample-identifying oligonucleotides are conjugated to a moiety capable of interacting with the surface of the cells. In some embodiments, the moiety is biotin and the kit further comprises reagents for coating the surface of the cells with streptavidin prior to contacting the sample with a plurality of sample-identifying oligonucleotides. In some embodiments, the moiety is a fatty acid residue or a cholesterol moiety capable of forming a hydrophobic interaction with the membrane of the cells. In some embodiments, the moiety is a maleimide moiety capable of reacting with amino groups present in cell membrane proteins of the cells. In some embodiments, the moiety is a phosphine moiety capable of reacting with carbohydrate residues associated with cell membrane proteins of the cells. In some of these embodiments, the kit further comprising reagents for attaching the sample-identifying oligonucleotide to the surface of the cells.

In some embodiments, the kit further comprises a DNA ligase. In some embodiments, the kit further comprises amplification primers capable of annealing to the splint oligonucleotide.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure l is a diagram of a workflow of the first embodiment of the invention.

Figure 2 1 is a diagram of a workflow of the second embodiment of the invention.

Figure 3 is a UMAP clustering of data obtained from a pool of four samples, each sample labeled by staining with antibodies conjugated to sample-specific barcodes.

Figure 4 is a biaxial plot of data obtained from a pool of two samples, each labeled by non-specific crosslinking of streptavidin to cells and binding of biotin modified sample-specific oligos to streptavidin.

DETAILED DESCRIPTION OF THE INVENTION

The term "nucleic acid" refers to a nucleotide polymer, and unless otherwise limited, includes known analogs of natural nucleotides that can function in a similar manner (e.g., hybridize) to naturally occurring nucleotides.

The terms "polynucleotide", "nucleotide", "nucleotide sequence", "nucleic acid" and "oligonucleotide" are used interchangeably. They refer to a polymeric form of nucleotides of any length, either deoxyribonucleotides or ribonucleotides, or analogs thereof. Polynucleotides may have any three dimensional structure, and may perform any function, known or unknown. The following are non-limiting examples of polynucleotides: coding or non-coding regions of a gene or gene fragment, intergenic DNA, loci (locus) defined from linkage analysis, exons, introns, messenger RNA (mRNA), transfer RNA, ribosomal RNA, short interfering RNA (siRNA), short-hairpin RNA (shRNA), micro-RNA (miRNA), small nucleolar RNA, ribozymes, complementary DNA (cDNA), which is a DNA representation of mRNA, usually obtained by reverse transcription of messenger RNA (mRNA) or by amplification; DNA molecules produced synthetically or by amplification, genomic DNA, recombinant polynucleotides, branched polynucleotides, plasmids, vectors, isolated DNA of any sequence, isolated RNA of any sequence, nucleic acid probes, and primers. A polynucleotide may comprise modified nucleotides, such as methylated nucleotides and nucleotide analogs. If present, modifications to the nucleotide structure may be imparted before or after assembly of the polymer. The sequence of nucleotides may be interrupted by non nucleotide components. A polynucleotide may be further modified after polymerization, such as by conjugation with a labeling component. Polynucleotide sequences, when provided, are listed in the 5' to 3' direction, unless stated otherwise.

The term "probe" refers to an oligonucleotide capable of binding to a target nucleic acid of generally through complementary base pairing, although perfect complementarity is not required thus forming a duplex structure. The probe binds or hybridizes to a "probe binding site." The probe can be labeled with a detectable label to permit detection of the probe, particularly once the probe has hybridized to its complementary target. Alternatively, however, the probe may be unlabeled, but may be detectable by specific binding with a ligand that is labeled, either directly or indirectly.

The term "epitope" and "target molecule" are used interchangeably herein to refer to the molecule of interest (parts of it or the whole molecule) being detected and/or quantified by the methods described herein.

The term “Quantum Barcoding” or “QBC” refers to a method of detecting multiple targets in individual cells described in US10950144 and O’Huallachain, M. et al, Ultra-high throughput single-cell analysis of proteins and RNAs by split-pool synthesis. Nature Commun. Biol. 3, 213 (2020). Briefly, the method involves binding to individual cells in a mixture a target-specific agent (an antibody or a nucleic acid probe) and assembling a unique cell-specific barcode on each bound agent. The target-specific agent and the cell-specific barcode are detected in a mxiture (e.g., by DNA sequencing). QBC is able to detect a target in multipe cells in a mixture on a single-cell level. Furthermore, QBC is able to simultaneously detect multiple targets in multipe cells in a mixture without sroting, isolating or encapsulating individual cells.

Single-cell analysis by combinatorial indexing techniques like Quantum Barcoding (QBC) facilitate analysis of large numbers of individual cells with higher cell throughput. Many embodiments of the Quantum Barcoding method are disclosed e.g., in the U.S. Patent No. 10,144,950 which is incorporated herein by reference in its entirety. QBC allows analysis of large numbers of individual cells at reduced cost compared with other NGS-based single-cell analysis technologies. Some configurations of single-cell analysis by combinatorial indexing (QBC) require large inputs of cells, for example > 1 million cells, which is prohibitive for applications using precious samples that have limited sample amounts. To reduce the burden of large cell inputs, samples can be marked with a sample barcode as described here and pooled before entering the combinatorial indexing QBC workflow. This technique can improve cell yield for each pooled sample compared with processing the samples separately. An added benefit of sample barcoding and sample pooling is a reduction in batch effects.

Methods of sample barcoding of cell-containing samples have been described in connection with other methods of single-cell analysis including flow-cytometry and mass cytometry. See e.g., Krutzik, P. O. & Nolan, G. P. Fluorescent cell barcoding in flow cytometry allows high-throughput drug screening and signaling profiling. Nat. Methods (2006). doi:10.1038/nmeth872; Krutzik, P. O., et ah, Fluorescent cell bar coding for multiplex flow cytometry. Curr. Protoc. Cytom. (2011). doi:10.1002/ 0471142956. cy063 ls55; Bodenmiller, B. etal. Multiplexed mass cytometry profiling of cellular states perturbed by small-molecule regulators. Nat. Biotechnol. (2012). doi:10.1038/nbt.2317; or Han, G., et al. Metal-isotope-tagged monoclonal antibodies for high-dimensional mass cytometry. Nat. Protoc. (2018). doi : 10.1038/s41596-018-0016-7. Similar sample barcoding techniques have been applied to flow cytometry and mass

The present disclosure is a method of single-cell analysis by Quantum Barcoding with the extra step of pooling multiple samples of cells wherein each cell has been labeled with a sample-identifying barcode.

The present invention involves a method of handling cells from a sample. In some embodiments, the sample is derived from a subject or a patient. In some embodiments the sample may comprise a fragment of a solid tissue or a solid tumor derived from the subject or the patient, e.g ., by biopsy. The sample may also comprise body fluids (e.g, urine, sputum, serum, plasma or lymph, saliva, sputum, sweat, tear, cerebrospinal fluid, amniotic fluid, synovial fluid, pericardial fluid, peritoneal fluid, pleural fluid, cystic fluid, bile, gastric fluid, intestinal fluid, or fecal samples) that may contain cells. In some embodiments, the sample is a dissociated solid tissue including solid tumor. The sample may comprise whole blood or blood fractions where normal or tumor cells may be present. In other embodiments, the sample is a cultured sample, e.g, a tissue culture containing cells. In some embodiments, the cells of interest in the sample are infectious agents such as bacteria, protozoa or fungi.

Nucleic acids, proteins or other markers of interest may be present in the cells and are the target of the single cell analysis in multiple samples disclosed herein. Each nucleic acid target is characterized by its nucleic acid sequence. Each protein target is characterized by its amino acid sequence and by its epitopes recognized by specific antibodies. In some embodiments, the target nucleic acid target contains a locus of a genetic variant, e.g, a polymorphism, including a single nucleotide polymorphism or variant (SNP of SNV), or a genetic rearrangement resulting e.g, in a gene fusion. In some embodiments, a protein target contains an amino-acid change resulting in the creation of a unique epitope. In some embodiments, the target nucleic acid or target protein comprises a biomarker, i.e., a gene or protein antigen whose expression or variants are associated with a disease or condition. For example, the target nucleic acids and proteins can be selected from panels of disease-relevant markers described in U.S. Patent Application Ser. No. 14/774,518 filed on September 10, 2015. Such panels are available as AVENIO ctDNA Analysis kits (Roche Sequencing Solutions, Pleasanton, Cal.) In other embodiments, the target nucleic acids or proteins are characteristic of a particular organism and aids in identification of the organism or a characteristic of the pathogenic organism such as drug sensitivity or drug resistance. In yet other embodiments, the target nucleic acid or protein is a unique characteristic of a human subject, e.g., a combination of HLA or KIR sequences defining the subject’s unique HLA or KIR genotype. In yet other embodiments, the target nucleic acid is a somatic sequence such as a rearranged immune sequence representing an immunoglobulin (including IgG, IgM and IgA immunoglobulin) or a T-cell receptor sequence (TCR). In yet another application, the target is a fetal sequence present in maternal blood, including a fetal sequence characteristic of a fetal disease or condition or a maternal condition related to pregnancy. For example, the target could be one or more of the autosomal or X-linked disorders described in Zhang et al. (2019) Non-invasive prenatal sequencing for multiple Mendelian monogenic disorders using circulating cell-free fetal DNA, Nature Med. 25(3):439.

In some embodiments, the target is a nucleic acid (including mRNA, microRNA, viral RNA, cellular DNA or cell-free DNA (cfDNA) including circulating tumor DNA (ctDNA)).

In some embodiments, the target is a protein expressed in the cell or on the cell surface. For example, the protein target may be cell-surface protein identifying a cell type. In some embodiments, the cell surface protein is a lymphocyte surface protein selected from inhibitory receptors (such as Pdcdl, Havrcr2, Lag3, CD244, Entpdl, CD38, CD101, Tigit, CTLA4), cell surface receptors (such as TNFRSF9, TNFRSF4, Klrgl, CD28, Icos, IL2Rb, IL7R) or chemokine receptors (such as CX3CR1, CCL5, CCL4, CCL3, CSF1, CXCR5, CCR7, XCL1 and CXCL10). In some embodiments, the proteins are selected from CD4, CD8, CD11, CD 16, CD 19, CD20, CD45, CD56 and CD279.

In some embodiments, one target is detected in the plurality of cells present in multiple samples. In other embodiments, multiple targets are detected simultaneously in the plurality of cells present in multiple samples. For example, 2, 3, 4, 5, 6, 7, 8, 9, 10, 20, 30 or more targets are detected in the plurality of cells present in multiple samples.

In some embodiments, the invention is an improved method of detecting one or more targets in a plurality of individual cells by quantum barcoding (QBC) described in U.S. Patent 10,144,950. Specifically, the invention is a method of detecting one or more targets in a plurality of individual cells in a pool of multiple samples containing the cells.

In some embodiments, the invention is a method of detecting a plurality of targets in a plurality of cells originating from multiple samples. Referring to Figurel, the cells in multiple separate samples are contacted with a plurality of unique binding agents. The unique binding agents can include one or more antibodies or one or more nucleic acids. In some embodiments, the unique agents include a combination of nucleic acids and antibodies. Each of the unique agents is specific for one of the targets.

The unique agents are conjugated to a sample-identifying nucleic acid barcode that distinguishes each of the samples among the multiple samples. In some embodiments, the sample-identifying nucleic acid barcode is 5 or more nucleotides. In some embodiments, the barcode is 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 or more nucleotides in length. In some embodiments, barcodes are randomly generated sequences. In other embodiments, the barcodes are pre-designed sequences. In some embodiments, the barcodes are designed to be sufficient edit distance apart as described in Levenshtein V.I., (1966) Binary codes capable of correcting deletions, insertions and reversals, Soviet Physics Doklady 10:707.

In some embodiments, the unique-binding agent is an antibody that is conjugated to an antibody-identifying nucleic acid barcode. Methods to attach nucleic acids to antibodies are known, e.g., Gullberg et ah, PNAS 101 (22): pages 228420-8424 (2004); Boozer et al, Analytical Chemistry, 76(23): pages 6967-6972 (2004) or Kozlov et al., Biopolymers 5: 73 (5): pages 621 — 630 (2004).

In this embodiment, identifying the antibody is possible by sequencing the antibody- identifying nucleic acid barcode. The sample-identifying nucleic acid barcode may attach to the antibody-identifying nucleic acid barcode directly, e.g., by annealing of complementary regions, or via a splint oligonucleotide where the splint anneals to the antibody identifying barcode and the split further contains annealing regions for the sample-identifying barcodes and also contains annealing regions for the assembly of the unique cell-associated barcodes from subcodes as described below.

In some embodiments, the unique binding agent is a nucleic acid probe. In this embodiment, the sample-identifying nucleic acid barcode may attach to the nucleic acid probe directly, e.g., by annealing of complementary regions, or via a splint oligonucleotide where the splint anneals to the nucleic acid probe and the split further contains annealing regions for the sample-identifying barcodes and also contains annealing regions for the assembly of the unique cell-associated barcodes from subcodes as described below.

In the splint oligonucleotide, the annealing regions for the sample-identifying barcode and the subcodes of the cell-associated barcodes are arranged to allow annealing of barcoded in ligatable proximity to each other to allow ligation of the end of the nucleic acid probe (or antibody-associated barcode) to the end of the sample-identifying barcode and further, ligating the other end of the sample- identifying barcode to the first subcode of the unique cell-associated barcode described below.

Referring further to Figure 1, multiple samples are pooled to form a pooled sample. The targets in each cell of the pooled sample are now labeled with a unique cell- originating barcode to detect the presence of each target in each individual cell. In some embodiments, the sample-identifying nucleic acid barcodes serve as anchors to assemble a unique cell-originating barcode in each cell of the pooled sample. The assembly occurs by attaching the barcodes to one another directly or via a splint oligonucleotide.

The method of assembling a unique cell-originating barcode involves a step-wise assembly of the nucleic acid barcode from oligonucleotide subunits in an ordered manner during successive rounds of split-pool synthesis. Multiple variations of this method are disclosed in the U.S. Patent No. 10,144,950, which is incorporated herein in its entirety. Briefly, the split-pool method includes the steps of splitting the pooled sample into multiple reaction volumes, each volume containing a subcode oligonucleotide; attaching the subcode oligonucleotide to the latest subcode oligonucleotide present in the cell in the reaction volume; and pooling the reaction volumes for the next round of splitting. Attaching involves the subcode oligonucleotides in each round annealing adjacently to the subcode oligonucleotide from a previous round via an annealing region and covalently linking the adjacently annealed subcode oligonucleotides to each other. In some embodiments, the subcode oligonucleotides anneal to annealing regions located on a single splint oligonucleotide. The splint contains distinct annealing regions for each round of annealing to accommodate multiple successive rounds of annealing without cross reaction with subcode oligonucleotides from the previous round. The subcodes in each round anneal adjacently to the annealed subcode from a previous round and are ligated to the annealed subcode from the previous round. In some embodiments, the splint oligonucleotide first anneals to an annealing region present in the sample- identifying oligonucleotide, which serves as an anchor for assembling the unique cell-originating barcode. In some embodiments, the splint oligonucleotide first anneals to an annealing region present in the unique binding agent which serves as an anchor for assembling the unique cell-originating barcode. At the completion of the split-pool rounds, each cell acquires a unique cell-originating barcode composed of a unique combination of subcodes. Each labeled entity in one cell (e.g., the unique binding agent or the sample-identifying oligonucleotide) receives the same unique cell-originating barcode characteristic of that cell.

Referring to Figure 1, the method further comprises a step of detecting the unique nucleic acid codes comprising the unique binding agent (represented by the nucleic acid probe or the antibody-identifying barcode), the sample-identifying barcode, and the unique cell-originating barcode thereby detecting the presence of the targets in the plurality of cells while identifying the sample of origin for the cells. The detection may be by sequencing.

In some embodiments, the nucleic acids comprising unique nucleic acid code are amplified prior to sequencing.

Referring now to Figure 2, in some embodiments, the invention is a method of detecting a plurality of targets in a plurality of cells originating from multiple samples comprising an alternative sequence of steps as compared to Figure 1. Specifically, as in Figure 1, the method starts with binding to the targets in a plurality of cells in multiple samples a plurality of unique binding agents that are each specific for one of the targets. Next, unlike in the method shown in Figure 1, as shown in Figure 2, the sample-identifying nucleic acid barcodes are added to a location different from the unique binding agents. Instead, Figure 2 shows a method where the sample-identifying nucleic acid barcode is attached elsewhere to the cells.

In some embodiments, the sample-identifying nucleic acid is attached to an antibody distinct from the unique binding agent antibody. In some embodiments, the unique binding agent is a primary antibody and the sample-identifying nucleic acid is attached to a secondary antibody capable of binding to the unique binding agent (primary antibody). In some embodiments, the secondary antibody is class-specific for the class of antibodies that includes the primary antibody serving as a unique binding agent. In some embodiments, the secondary antibody is species-specific for the species from which the unique binding agent (primary antibody) is derived.

In yet another embodiment, the sample identifying nucleic acid is conjugated to an antibody specific to the type of cells analyzed by the method disclosed herein. Examples of such antibodies are shown in Table 1.

Table 1. Cell type-specific antibodies used for sample bar coding

In yet other embodiments, the sample-identifying oligonucleotide is attached directly to the cell surface. The sample-identifying nucleic acid is capable of attaching to the surface of the cells when the cells are contacted with the sample- identifying nucleic acid. In some embodiments, the sample-identifying nucleic acid is conjugated to a moiety capable of interacting with the surface of the cells.

In some embodiments, to the interacting moiety is streptavidin. The cell surface is modified with streptavidin and the sample-identifying oligonucleotide includes one or more biotinylated nucleotides. In some embodiments, streptavidin is added to the cell surface by reacting streptavidin with amino groups on the surface of the cell (e.g., epsilon amino groups of lysine in cell-membrane proteins). An effective way of conjugating streptavidin (or other avidin derivatives) to the cell surface involve maleimide activation of streptavidin and thiolation of amino groups of cell surface proteins to enable a reaction between a free sulfhydryl group and maleimide, see Espeel P. and Du Prez F.E. (2015) One-pot multi-step reactions based on thiolactone chemistry: a powerful synthetic tool in polymer chemistry , Eur. Polymer J. 62:247. In some embodiments, surface proteins are biotinylated and contacted with streptavidin, see Ho, V.H.B., et al, (2009) The precise control of cell labelling with streptavidin paramagnetic particles , Biomaterials 30:6548. Commercial reagents are conveniently available for streptavidin conjugation, e.g., Streptavidin Conjugation Kit -Lightning Link (Abeam, Cambridge, Mass.)

In some embodiments, the interacting moiety is a hydrophobic moiety capable of non-covalent hydrophobic interaction i.e., insertion into a cell membrane. In some embodiments, the hydrophobic moiety is a fatty acid residue or a cholesterol moiety. For example, oligonucleotides can be joined with a palmitoyl or stearoyl residue via an amino alkyl linker. Such oligonucleotide conjugates are stably integrated into a cell membrane and form hybrids with complementary nucleic acids, see Borisenko, G., et al. (2009) DNA modification of the cell surface , Nucl. Acids Res. 37:e28.

In some embodiments, the interacting moiety is a reactive moiety capable of forming a covalent bond with amino groups of cell membrane proteins. For example, 5'- thiol modified oligonucleotides can be conjugated to an NHS-PEG-maleimide cross-linker which reacts with available amino groups of cell membrane proteins, see Hsiao, S.C., et al., (2010) Direct Cell Surface Modification with DNA for the Capture of Primary Cells and the Investigation ofMyotube Formation on Defined Patterns , Langmuir: the ACS journal of surfaces and colloids, 25(12), 6985. In some embodiments, the interacting moiety is a reactive moiety capable of forming a covalent bond with carbohydrates associated with cell surface proteins. For example, an oligonucleotide conjugated to a biotinylated phosphine reacts with an azido-modified sialic acid on the cell surface to form a as described in Saxon E. and Bertozzi C.R., (2000) Cell Surface Engineering by a Modified Staudinger Reaction Science 287:2007.

Referring to the workflow illustrated in Figure 2, the step of attaching unique binding agents and the step of attaching sample-identifying nucleic acids can occur in any sequence. In some embodiments, the unique binding agent and the sample- identifying barcode are attached independently, e.g., the sample-identifying nucleic acids area attached to modified cell surface. In such embodiments, the step of attaching unique binding agents and the step of attaching sample-identifying nucleic acids can occur in any order, e.g., the step of attaching unique binding agents occurs can occur simultaneously, prior to or following the step of attaching sample- identifying nucleic acids. In other embodiments, the sample-identifying nucleic acids attach to a modality interacting with the unique binding agent (e.g., a secondary antibody or an annealing region interacting with a nucleic acid probe). In such embodiments, the step of attaching unique binding agents may occur prior to the step of attaching sample-identifying nucleic acids. In variations of this embodiment, the sample-identifying nucleic acids may interact the unique binding agent prior to contacting the unique binding agent to the cells of each sample.

Referring further to Figure 2, after the unique binding agent is added to the cells and the sample-identifying nucleic acid barcode is added to the cells, the method comprises the step of pooling multiple samples to form a pooled sample. The cells in the pool sample undergo the process of assembly of a unique cell-originating barcode as described in Figure 1. However, in Figure 2, in each cell, the unique cell-originating barcodes are assembled in parallel on the unique binding agent and on the sample-identifying barcode. Similar to the process in Figure 1, Figure 2 shows that the unique cell originating barcode is assembled by a method of split- pool synthesis where subcode oligonucleotides anneal adjacently to each other and are covalently linked to form the unique cell-originating barcode in each cell. In each cell, identical unique cell-originating barcodes are assembled on the unique binding agent and on the sample-identifying barcode. The method further comprises detecting the unique binding agent, the sample-identifying barcode and the unique cell-originating barcodes attached thereto. The detection may be by sequencing with or without a prior amplification step. Correlating the unique cell-originating barcodes on the sample-identifying barcode and the unique binding agent allows detecting the presence of the target in the cell and the sample origin of the cell.

In some embodiments, the invention is a kit for detecting a plurality of targets in a plurality of cells originating from multiple samples, the kit comprising: a set of sample-identifying oligonucleotides; a set of nucleic acid subcodes; a splint oligonucleotide comprising annealing regions capable of annealing to the annealing regions of the nucleic acid subcodes, and annealing regions of the sample- identifying oligonucleotides.

In some embodiments, the kit further comprises a unique binding agent. In other embodiments, the custom unique binding agent is provided by the user of the kit. In some embodiments, the unique binding agent included in the kit is a nucleic acid probe having a target-specific region and an annealing region capable of annealing to the splint oligonucleotide. In some embodiments, the unique binding agent included in the kit is an antibody conjugated to an oligonucleotide with an annealing region capable of annealing to the splint oligonucleotide. In some embodiments, the antibody is conjugated to the sample-identifying oligonucleotide.

In some embodiments, the kit further comprises a DNA ligase. In some embodiments, the kit further comprises reagents for amplification selected from amplification primers capable of amplifying the unique cell-originating barcodes, a thermostable nucleic acid polymerase and nucleotide precursors.

In some embodiments, the forward amplification primer included in the kit is capable of annealing to the last-to-be-added subcode within the cell-originating barcodes, and the reverse amplification primer included in the kit is capable of annealing to the UBA (e.g., to the nucleic acid probe or to the antibody-identifying nucleic acid). One of skill in the art would appreciate that primer-binding sites can be engineered into each of the last-to-be-added subcode, the nucleic acid probe and the antibody-identifying nucleic acid when these oligonucleotides are designed. In some embodiments, the forward and reverse amplification primers included in the kit each comprise a universal primer binding site. In such embodiments, the kit further comprises universal amplification primers. In some embodiments, the universal primers are specific for a particular sequencing platform and are readily available. Such universal primers may be omitted from the kit.

In some embodiments, the unique binding agent antibody is not conjugated to a sample identifying oligonucleotide. Instead, the kit further comprises a sample tagging antibody conjugated to the sample-identifying oligonucleotide. In some embodiments, the sample-tagging antibody is a secondary antibody capable of binding to the unique binding agent antibody. In some embodiments, the sample tagging antibody is a class-specific or species-specific antibody capable of binding to the class of antibodies or the constant region of antibodies of the species of origin represented by the unique binding agent antibody. In some embodiments, the sample-tagging antibody is specific for the type of cells including the cells to be analyzed with the kit. A non-limiting list of examples of cell-type-specific antibodies that can be included in th kit as cell-tagging antibodies is in Table 1. The cell-type specific sample tagging antibodies may be specific for one of the following: an anti-CD45 antibody, an anti-CD3 antibody, an anti-CD298 antibody, an anti-beta 2 microglobulin antibody, and an anti-VAMP7 antibody.

The method of detecting targets in a plurality of individual cells present in a pool of multiple samples containing the cells is not limited to the exemplary application listed above but can be used in any diagnostic, prognostic, therapeutic, patient stratification, drug development, treatment selection, and screening process that involves attaching barcodes to cells and where pooling samples containing the cells is desired. EXAMPLES

Example 1. Adding sample barcodes to cell-specific antibodies

In this example, human PBMCs were fixed with 16% formaldehyde solution (ThermoFisher Scientific, 28908) by adding it directly to cells to a final concentration of 1.6%. The cells were incubated for 10 min at room temperature under gentle agitation to fix the cells. Cells were stored in IX PBS with 10% DMSO at -80°C until proceeding with the next steps. Approximately 20 million fixed PBMCs were pelleted and the storage solution aspirated and discarded. Cells were washed two times with SME buffer (0.5% BSA, 0.02% NaN3, 5mM EDTA in lx PBS).

Cells were divided evenly into four separate tubes for binding to commercial antibodies against CD298 and beta-2 microglobulin conjugated to sample- identifying barcode oligonucleotides. Cells in each tube were re-suspended in cell surface blocking reactions containing 10 ul of Human TruStain FcX (BioLegend, 422301), 10 ul of Salmon Sperm DNA (Invitrogen, AM9680), 0.5 M NaCl, and SME buffer to a final volume of 100 ul. The blocking reactions were incubated at room temperature for 30 minutes with gentle rotation. After the 30-minute blocking incubation, 0.5 ug of two different oligo-conjugated antibodies were added. The oligonucleotide conjugated to the antibody contains a capture sequence, an antibody clone-specific barcode sequence, and a PCR handle compatible with Illumina® sequencing reagents and instruments (TotalSeq-B Hashtag antibodies (BioLegend, 394631, 394633, 394635, 394637, 394641, 394643, 394645, 394647)). After the antibodies were added to each tube, the final volume was brought to 200 ul with SME buffer and NaCl (0.62 M final concentration). The antibody staining reaction was incubated at room temperature for 1.5 hours with gentle rotation. After staining, the cells were washed two times with SME buffer supplemented with 0.5 M NaCl. The four cell samples were pooled.

An additional blocking reaction, staining with antibodies to other surface proteins, and single-cell barcoding by QBC was done as described in O’Huallachain, M. et al. Ultra-high throughput single-cell analysis of proteins and RNAs by split-pool synthesis. Nature Commun. Biol. 3, 213 (2020). The results are shown in Figure 3. The UMAP clustering of data is obtained from pooled samples formed from four samples as described above. The cells have undergone single-cell combinatorial indexing and sample barcoding by staining with antibodies conjugated to sample-specific barcodes. The four clusters indicate that the four cell samples could be separated by their sample barcode tags during the data analysis.

Example 2. Adding biotinylated sample barcodes to streptavidin-modified cell surface.

In this example, human PBMCs were fixed with 16% formaldehyde solution (Thermo Scientific, 28908) by adding it directly to cells to a final concentration of 1.6% and incubated for 10 min at room temperature under gentle agitation to fix the cells. Cells were stored in IX PBS with 10% DMSO at -80°C until proceeding with the next steps. Approximately 25 million fixed PBMCs were pelleted and the storage solution aspirated and discarded.

Cells were re-suspended in 62 ul of lx PBS for modifying protein amino groups on the cell surface. 10 ul of LL-Modifier reagent (Novus Biologicals, 7080030) was added to the cells in lx PBS and the solution was mixed gently. 10 ug of Lyophilized Lightning-Link mix containing streptavidin (Novus Biologicals, 7080030) was resuspended in 100 ul of lx PBS. 52 ul of the resuspended Lightning-Link mix was removed from the vial. Cells with LL-Modifier reagent were pipetted directly into the remaining 48 ul of resuspended Lightning-Link mix. The cells and Lightning- Link mixture were incubated for 2 hours protected from light with gentle rotation. At the end of the 2 hours 10 ul of LL-Quencher reagent (Novus Biologicals, 7080030) was added to the mixture and incubated at room temperature for 30 minutes with gentle rotation. Cells were washed with SME buffer (0.5% BSA, 0.02% NaN3, 5mM EDTA in lx PBS).

Cells were split into two tubes with approximately 8 million cells each in 70 ul of SME for binding sample-identifying oligonucleotides. A unique biotin-modified hashtag oligo with an anchor region for later annealing to QBC single-cell barcodes was added to each tube to a final concentration of 20 nM. The reaction volume was brought to 100 ul with SME buffer and NaCl to a final concentration of 0.6 M. The reaction was incubated at room temperature for 30 minutes followed by 3 washes with SME buffer. Each cell sample was re-suspended in a cell surface blocking reaction containing 12.5 ul of Human TruStain FcX (BioLegend, 422301), 12.5 ul of Salmon Sperm DNA (Invitrogen, AM9680), 0.5 M NaCl, 0.05 uM Biotin (Sigma, B4639) and SME buffer to a final volume of 50 ul. The blocking reactions were incubated at room temperature for 30 minutes with gentle rotation. After cell surface blocking, the two samples were pooled for antibody staining and single-cell barcoding by QBC as described previously in O’Huallachain, M. etal. (2020) supra. The results are shown in Figure 4. The biaxial plot of data is obtained from single cell combinatorial indexing with samples pooled after sample barcoding by non specific crosslinking of streptavidin to cells and binding of biotin-modified sample- specific oligos. The two cell sample populations are clearly distinguished from each other with 38.48% of detected cells having Biotin_Hashtag_2 sequence and 48.92% of cells detected with Biotin_Hashtag_3 sequence.

Claims

1. A method of detecting a plurality of targets in a plurality of cells originating from multiple samples, the method comprising: a) binding to one or more targets in a plurality of cells in multiple samples one or more of unique binding agents that are each specific for one of the targets; b) attaching a sample-identifying oligonucleotide to the unique binding agent; c) pooling the multiple samples to form a pooled sample; d) adding multiple subcode oligonucleotides to the sample-identifying oligonucleotide in the plurality of cells in an ordered manner during successive rounds of split-pool synthesis wherein the subcode oligonucleotide in each round anneals adjacently to the subcode oligonucleotide from a previous round via an annealing region, and covalently linking the adjacently annealed subcode oligonucleotides to each other to create in each cell, a unique cell-originating barcode; e) detecting the unique cell-originating nucleotide code consisting of the unique binding agent, the sample-identifying barcode, and the unique cell-originating barcode thereby detecting the presence of the targets in the plurality of cells and identifying the sample of origin of the cells.

2. The method of claim 1, wherein detecting is by sequencing and optionally, the unique cell originating nucleotide code is amplified prior to sequencing.

3. The method of claim 1-2, wherein a plurality of targets is detected in each cell.

4. The method of claim 1-3, wherein in each round of split-pool synthesis, the subcode oligonucleotide anneals to the annealing region located on a splint oligonucleotide adjacently to the annealed subcode oligonucleotide from a previous round and is ligated to the annealed subcode oligonucleotide from the previous round.

5. The method of claim 1-3, wherein one round of split-pool comprises: splitting the pooled sample into multiple reaction volumes, each volume containing a subcode oligonucleotide; attaching the subcode oligonucleotide in the reaction volume; and pooling the reaction volumes.

6. The method of claim 1-5, wherein the unique binding agent comprises an antibody.

7. The method of claim 1-5, wherein the unique binding agent is a nucleic acid probe.

8. A method of detecting a plurality of targets in a plurality of cells originating from multiple samples, the method comprising: a) binding to one or more targets in a plurality of cells in multiple samples a one or more of unique binding agents that are each specific for one of the targets; b) attaching to the plurality of cells in each of the multiple samples a sample-identifying oligonucleotide; c) pooling the multiple samples to form a pooled sample; d) adding multiple subcode oligonucleotides to the unique binding agent and to the sample-identifying barcode in the plurality of cells of the pooled sample in an ordered manner during successive rounds of split-pool synthesis wherein the subcode oligonucleotide in each round anneals adjacently to the subcode oligonucleotide from a previous round via an annealing region, and covalently linking the adjacently annealed subcode oligonucleotides to each other to create in each cell, a unique cell- originating barcode; e) detecting in each cell two unique cell-originating nucleotide codes: the first code consisting of the unique binding agent and the unique cell- originating barcode, and the second code consisting of the sample- identifying barcode and the same unique cell-originating barcode, thereby detecting the presence of the targets in the plurality of cells and identifying the sample of origin of the cells.

9. The method of claim 8, wherein the unique binding agent is a nucleic acid probe.

10. The method of claim 8, wherein the unique binding agent comprises an antibody conjugated to an antibody-identifying nucleic acid barcode.

11. The method of claim 8, wherein the sample-identifying oligonucleotide is attached to the surface of the cells.

12. The method of claim 8, wherein the sample-identifying oligonucleotide is conjugated to a moiety capable of interacting with the surface of the cells

13. The method of claim 8, wherein the sample identifying nucleic acid barcode is conjugated to an antibody specific to the type of cells including the plurality of cells in which the plurality of targets are to be detected.

14. The method of claim 8, wherein the subcode oligonucleotides anneal to annealing regions located on a splint oligonucleotide adjacently to the annealed subcode oligonucleotide from a previous round and are ligated to the annealed subcode from the previous round.

15. The method of claim 8, wherein one round of split-pool comprises: a) splitting the pooled sample into multiple reaction volumes, each volume containing a subcode oligonucleotide; b) attaching the subcode oligonucleotide in the reaction volume; and c) pooling the reaction volumes

16. A kit for detecting a plurality of targets in a plurality of cells originating from multiple samples, the kit comprising: f) a unique binding agent capable of binding to a cellular target; g) a set of sample-identifying oligonucleotides; h) a set of nucleic acid subcodes; i) a splint oligonucleotide comprising annealing regions capable of annealing to the annealing regions of the subcodes, and annealing regions of the sample-identifying oligonucleotides.

17. The kit of claim 16, wherein the unique binding agent is an antibody conjugated to an antibody-identifying nucleic acid barcode comprising an annealing region for the splint oligonucleotide.

18. The kit of claim 16, wherein the unique binding agent is a nucleic acid probe.

19. The kit of claim 16, wherein the sample-identifying oligonucleotides comprise an annealing region capable of annealing to the unique binding agent.

20. The kit of claim 16, wherein the sample-identifying oligonucleotides are capable of binding to the surface of the cells.

21. The kit of claim 16, wherein the sample-identifying oligonucleotides are conjugated to a moiety capable of interacting with the surface of the cells