Method and product

Abstract

42.170.173772 / 01 100 Abstract The present invention provides a method of determining whether one or more nucleic acid analyte, each of which comprises a unique barcode, which is administered to a cell is transported to a target subcellular location. Assessment is achieved by examination of the location of the analytes by identification of their barcodes or by identifying which analytes, by virtue of their barcodes, have been modified by interaction with an entity in the subcellular location, particularly kinases. Related methods as well as analytes and controls for this purpose are also provided.

Description

[0001] 42.170.173772 / 01

[0002] Method and Product

[0003] The present invention relates to methods of determining if one or more nucleic acid analyte administered to a cell is transported to a target subcellular location. This is achieved by using barcodes to uniquely identify each analyte and by mapping the location of those analytes. Mapping may be based on cellular fractionation or on functional interactions between the analytes and entities in the cell. This provides useful information on the functional properties of the analytes, in particular their interaction with key cellular mechanisms and subcellular compartments, and not simply their gross biodistribution. Compositions and products for this purpose are also provided.

[0004] Background

[0005] Oligonucleotide therapeutics (txONs) have demonstrated immense potential for the treatment of a wide variety of diseases by directly controlling protein production within specific cells in the body. Despite much progress, the need for safe, specific and efficacious methods to ensure therapeutic RNA and / or DNA reaches its target cells, and the subcellular compartment in which it is active, still represents a major hurdle for their advancement. A challenge in designing and testing the efficient delivery of txONs is the low correlation between in vitro studies and in vivo biodistribution and efficacy, and as a consequence the current low-throughput nature of traditional in vivo studies.

[0006] The introduction of barcodes overcomes the challenge of low-throughput in vivo studies, by allowing for the experimental co-administration (or ‘multiplexing’) of large libraries of delivery systems in the same animal. This also eliminates variability within groups and reduces the need for large in vivo experiments. It has, for instance, previously been shown that multiplexed studies using DNA barcodes to screen lipid nanoparticle (LNP) formulations (Dahlman et al., Proc. Natl. Acad. Sci., 2017, 114(8), p2060-2065) and / or oligonucleotide barcodes that can be used for unencapsulated (and / or conjugated) delivery (WO2024 / 069235) can provide a wealth of information regarding the individual biodistribution of dozens to thousands of delivery systems in vivo.

[0007] However, while biodistribution (and pharmacokinetics) is an important metric to understand the pharmaceutical potential of naked oligonucleotides, RNA and / or DNA delivery systems and other biomolecules, it is also important to understand how the activity (and wider pharmacodynamics) of these compounds is affected by differential sub-cellular distribution after initial cellular uptake. For example, it is known that the biodistribution of naked or formulated RNA therapeutics to a specific tissue in vivo does not necessarily have the predicted influence on gene expression (such as gene knockdown, also referred to as ‘KD’) within that tissue. In fact tissues with very low accumulation of certain therapeutics showed efficient functional delivery, in some cases even at a higher rate than tissues with 42.170.173772 / 01 substantially higher accumulation (Rhym et al., Nat. Biomed. Eng., 2023, 7 (7), p901-910). Hence, developing barcodes that enable measurement of both distribution AND functional activity (or a proxy for functional activity) upon cellular uptake within the same design are of utmost importance.

[0008] Dahlman's team has reported the implementation of a universal DNA barcoding system which enabled the delivery quantification of chemically distinct LNPs loaded with small interfering RNA and their application for mouse model in vivo screening (including genetic knockout models) (Sanchez et al., Nano Lett., 2022, 22(12), p4822-4830; Dahlman et al., 2017, supra).

[0009] The method is based on the use of a rapid microfluidic mixing system to encapsulate nucleic acid barcodes inside nanoparticles and the administration of the co-formulation as a single pool to mice. The barcodes are then recovered from the tissues and deep sequenced to obtain counts for the barcodes in each sample of interest. Each nanoparticle is associated with a unique DNA barcode sequence, allowing for the measurement of several nanoparticles’ (NPs) biodistribution in a single animal.

[0010] Major limitations of the described technology are the following: i) the DNA barcodes cannot be applied to non-encapsulated delivery systems such as conjugates owing to nuclease sensitivity and poor cellular uptake; ii) because the length of the barcoded strands is greater than 50 nucleotides, they cannot be directly conjugated to smaller RNA therapeutics such as small interfering RNA (siRNA) and antisense oligonucleotides (ASOs) without distorting distribution, thereby restricting barcode use to applications where they must be co-delivered; and iii) the primary readout of the technology is biodistribution. Functional uptake can be assessed through complementary assays (e.g. by fluorescence- activated cell sorting (FACS) of cells with reduced expression of an siRNA target or increased expression of an exogenous protein encoded by delivered messenger RNA (mRNA)) or through single-cell sequencing approaches that combine transcriptome profiling with barcode detection (Dobrowolski et al., Nat. Nanotechnol., 2022, 17, p871-879). However, these functional measurements do not directly identify the best-performing delivery candidates, as barcode identity is linked to functional activity under the assumption that the most abundant barcode corresponds to the most functional candidate, an assumption that may not always hold true. This limits both the accuracy and scalability of the approach.

[0011] Rhym et al. (2023, supra) described a high-throughput functional barcoding method which uses liquid chromatography with tandem mass spectrometry to detect peptide barcodes translated from LNP encapsulated mRNAs. The technology was applied for the evaluation of a library of about 400 nanoparticle formulations with 384 unique ionizable lipids while using only nine mice. Although well suited to screen encapsulated mRNA delivery and 42.170.173772 / 01 providing a direct functional readout through mRNA-driven exogenous protein expression, this strategy is not applicable to non-formulated delivery systems, nor can it capture functional uptake of therapeutic oligonucleotides working through different molecular mechanisms such as RNA interference (RNAi), ribonuclease H (RNaseH)-mediated antisense activity, RNA editing, splice-switching and any other mechanism of action not relying on active translation of the payload.

[0012] Crucially, while these barcoding methods have been successfully applied to the multiplex quantification of biodistribution in vivo and have even been extended to functional readouts, they remain limited in scope. Dahlman’s DNA barcoding approach depends on codelivery and on assumptions that may not consistently identify the most functional candidate, while peptide barcoding offers a direct readout but only for translation-dependent payloads. As a result, neither approach is generalisable to smaller txONs or to txON conjugates with other biomolecules such as antibodies, since the barcodes must remain largely unmodified (in the case of DNA barcodes to ensure amplifiability by polymerase chain reaction (PCR), in the case of peptide barcodes to permit translation) and require encapsulation in LNPs or adeno-associated viruses (AAVs) to prevent nuclease degradation.

[0013] Barcoded, or DEL(DNA Encoded Library)-like, approaches have also been utilized for the screening of RNA modifications in vitro (WO2023 / 3174845 and WO2023 / 194358). However, the applicability of these methods for in vivo screening is limited by: i) the degradation of compounds in vivo by nucleases, ii) the reliance on expressing (recombinant) polymerase (e.g. T3, Sp6, T7 RNA polymerase, Phi29 DNA polymerase, Syn5, or alphavirus replicase) in target cells, and iii) the oligonucleotide payload and the barcode are linked through a bead that will likely influence distribution, geometry and in vivo compatibility, meaning conjugation to moieties other than oligonucleotides for screening is not feasible based on the current described protocol.

[0014] Methods and compositions for the design and use of smaller and chemically modified oligonucleotide barcodes that can be used for multiplexed in vivo studies of conjugated and unencapsulated molecules have been described in WO2024 / 069235.

[0015] The present invention is concerned with the detection of barcodes (in nucleic acid analytes) which are (or have been) in subcellular compartments such as the cytoplasm and / or nucleus. This provides a method of determining whether a particular analyte has reached a particular subcellular compartment. This has various uses. In particular it acts as a proxy for determining whether an analyte, which may include, or mimic a functional molecule, is able to achieve its biological function. For example, txONs need to reach the cytoplasm and / or nucleus to exert their biological function. Given that endosomal escape and intracellular trafficking into the appropriate subcellular compartment is a prerequisite for txON activity, and is in fact the major rate-limiting step in the delivery of biological 42.170.173772 / 01 therapeutics (Dowdy, RNA, 2023, 29, p396-401), the level of cytosolic and / or nuclear penetration is a direct measure of delivery efficiency and may even, indirectly, serve as a proxy for therapeutic activity. For this reason, the barcodes, when present in analytes, may be considered effective ‘Activity Barcodes’. The analytes comprising the barcode may be multiplexed in vivo and their take-up within the cell assessed as described hereinafter. By appropriate selection of the analyte and barcode, information on functional activity (that enables the agent to have the expected biological activity) may be obtained and used for designing future products and / or predicting likely activity.

[0016] For example, to assess a variety of txONs a plurality of different analytes comprising the barcode may be designed which closely mimic different txONs in size, shape, sequence pattern, modification pattern and / or overall design, including conjugate payload(s). This ensures that the analytes comprising the barcodes have similar pharmacokinetic profiles to, and can be used as surrogates of, a given txON modality of interest for screening purposes. This allows for high-throughput in vivo screening. Also disclosed herein are methodologies for the normalization and validation of these analytes comprising barcodes in biological settings. The methods of the invention i) are compatible with RNA-sequencing pipelines, including single-cell sequencing, ii) collect specific functional distribution readouts, and iii) remove dependency on specific target genes since the readout is not based on a direct activity readout, but rather specific measurements of the recovered barcode, allowing the individual assessment of each one of multiple barcode-associated conjugates (referred to herein as analytes) in a single mixture.

[0017] Exemplary methods of performing the method are described hereinafter. By way of example Figure 1 illustrates the use of endogenous 5’ phosphorylation as a proxy for productive analyte (referred to in the figure as the barcoded oligonucleotide) uptake. Only analytes that are released into the cytoplasm (and / or nucleus) are phosphorylated by endogenous kinases in the cytoplasm (and / or nucleus). The analytes (carrying barcodes) are captured and a library generated which incorporates a phosphorylation-dependent step (e.g. ligation, not shown) to ensure selective detection of the monophosphorylated analytes. Figure 2 shows an alternative mechanism in which nuclear accumulation is used as a proxy for productive analyte uptake. Some analytes escape endolysosomal entrapment, reach the cytoplasm and then enter the nucleus for example via passive diffusion through the nuclear pore complex. The nuclei are isolated and the analytes extracted and identified based on the barcodes they carry. Nuclei isolation ensures selective detection of only those analytes that manage to traffic and accumulate in the nucleus. 42.170.173772 / 01

[0018] Statement of invention

[0019] In a first aspect, the present invention provides a method of determining whether one or more nucleic acid analyte administered to a cell is transported to a target subcellular location comprising the steps of: a) administering one or more nucleic acid analyte to said cell, each analyte comprising a nucleotide sequence barcode unique to said analyte; b) collecting said cell or a portion thereof and assessing whether one or more barcode of the administered one or more analyte is, or was, present in the target subcellular location, thereby determining which analyte was transported to the target subcellular location, wherein said assessment is made by: i) collection of said cell or a portion thereof comprising said target subcellular location and identifying if one or more of said barcode is present in said location, or ii) determining whether one or more of said analyte comprising a barcode has interacted with an entity in said subcellular location.

[0020] In a preferred aspect said analyte comprises a nucleic acid molecule with less than 200, e.g. less than 150 or less than 100, nucleotides. In another preferred aspect, said analyte comprises a nucleic acid molecule with less than 40 nucleotides.

[0021] As used herein, entities recited in the singular include those in the plural, and vice versa, e.g. reference to a cell includes the use of cells. Reference to “one or more” as used herein refers to 1, 2, 3, 4, 5, 6, for example, and other examples as described hereinafter. Analyte and barcode

[0022] A “nucleic acid” molecule or analyte refers to a molecule which consists of, or includes, an oligonucleotide or polynucleotide sequence, i.e. a polymer of linked nucleosides each of which can be independently modified or unmodified.

[0023] The “nucleotide sequence” of such a polynucleotide sequence denotes the order of nucleobases or nucleotides, described with a succession of letters using standard nomenclature, in 5’ to 3’ direction unless otherwise specified. A “nucleotide base” (or nucleobase) is a heterocyclic pyrimidine or purine compound that is a component of a nucleotide, and includes the primary purine bases adenine and guanine, and the primary pyrimidine bases cytosine, thymine, and uracil. A nucleobase may further be modified to include, without limitation, universal bases, hydrophobic bases, promiscuous bases, size- expanded bases, and fluorinated bases (see, e.g., Modified Nucleosides in Biochemistry, Biotechnology and Medicine, Herdewijn, P. ed. Wiley-VCH, 2008). The synthesis of such modified nucleobases (including phosphoramidite compounds that include modified nucleobases) is known in the art.

[0024] As used herein, "nucleic acid analyte", refers to a nucleic acid molecule that is the 42.170.173772 / 01 target of an analysis method aimed at detecting, identifying and / or quantifying the analyte in the subcellular location of a cell.

[0025] As referred to herein the nucleic acid analyte comprises a nucleotide sequence barcode. When more than one analyte is used a barcode unique to each analyte (relative to other analytes that may be present in a set or pool or used in a method or use of the invention) to allow their discrimination is employed.

[0026] In some embodiments, the analyte may contain only one barcode. In other embodiments, the analyte contains more than one barcode, for example 2, 3, 4, 5, 6, 7, 8, 9 or more barcodes (n). Each of the second, third, fourth, fifth, sixth, seventh, eighth, ninth or nth barcode may comprise a sequence identical or different to the first barcode. Alternatively, each of the barcode sequences may be identical or different to at least one or more other barcode sequences in the analyte.

[0027] As a minimum, the nucleic acid analyte consists of a barcode. However, it may additionally comprise or contain additional regions, components or moieties. For example, the barcode can appear as a 5' or 3' extension at the end of the analyte. Alternatively, the barcode may be included in an internal loop, hairpin or bulge region of the analyte. In another instance, the barcode may be included in an internal double-helical motif of the analyte.

[0028] In another instance, the barcode and / or analyte may be co-formulated with or otherwise non-covalently associated with a delivery system, such as an LNP or AAV particle.

[0029] In a preferred embodiment the nucleic acid analyte may contain a barcode flanked by sequences at the 5’ and / or 3’ ends. The flanking sequences may be from 1 to 50, e.g. from 5-20 sequences in length. These flanking sequences may be attached via normal nucleotide linkages or modified versions thereof, as described herein, or may be attached via alternative linkers, conveniently by conjugation.

[0030] As used herein, “linked” or “conjugated” means that two molecules are joined by a covalent bond or are associated via noncovalent bonds (e.g., hydrogen bonds or ionic bonds). The “linker” or “linking group” is one or more atoms that connect one molecule or portion of a molecule to another molecule or portion of a molecule. Linking groups may comprise any number of atoms or functional groups. In some embodiments, linking groups may not facilitate any biological or pharmaceutical response, and merely serve to link two biologically active molecules. The linker can be any suitable group for coupling the sequences (or moiety / cargo to the analyte as described hereinafter).

[0031] The linkers may be cleavable, e.g. to allow release of a cargo molecule. Suitable linker molecules include, but are not limited to thiol cleavable linkers such as dithiobismaleimidoethane, 1,4-bis[3-(2-pyridyldithio)propionamido]butane and 3-(2- pyridyldithio)propionyl hydrazide, base-cleavable linkers such as bis[2-(N-succinimidyl- 42.170.173772 / 01 oxycarbonyloxy)ethyl] sulfone or hydroxylamine-cleavable linkers such as (ethylene glycol bis(succinimidyl succinate)). Reversible click moieties, i.e. the Meldrum’s acid derivative, 5- (bis(methylthio)methylene)-2,2-dimethyl-1,3-dioxane-4, 6-dione may be used to crosslink cargo molecules (Diehl et al., Nat. Chem., 2016, p968-973). Dicer substrates may also be used, either on their own or in combination with the above.

[0032] Non-cleavable linkers may also be employed where the remainder of the molecule (e.g. cargo molecule) does not require cytosolic release; these include, but are not limited to, thiol-reactive maleimides (e.g. 1,8-bismaleimido-diethyleneglycol, 1 ,11-bismaleimido- triethyleneglycol, 1,4-bismaleimidobutane, bismaleimidohexane, bismaleimidoethane, tris(2- maleimidoethyl)amine), thiol / amine reactive linkers (e.g. N-a-maleimidoacet-oxysuccinimide ester, N-p-maleimidopropyl-oxysuccinimide ester, N-e-maleimidocaproic acid, N-y- maleimidobutyryl-oxysuccinimide ester, succinimidyl 4-(N-maleimidomethyl)cyclohexane-1- carboxy-(6-amidocaproate), succinimidyl 6-(3(2-pyridyldithio)propionamido)hexanoate, m- maleimidobenzoyl-N-hydroxysuccinimide ester, succinimidyl iodoacetate, succinimidyl (4- iodoacetyl)aminobenzoate, polyethylene glycol (PEG)ylated, long-chain SMCC (succinimidyl 4-(N-maleimidomethyl)cyclohexane-1- carboxylate) crosslinkers, succinimidyl 4-(p- maleimidophenyl)butyrate and sulfo-NHS equivalents) and hydroxyl / thiol reactive linkers (e.g. p-maleimidophenyl isocyanate).

[0033] In some embodiments, the linking may be performed post-synthetically via click chemistry. Such conjugation reactions are known in the art and are given, for example, in US 2022 / 0370635, the contents of which are incorporated herein by reference. If such conjugation reactions are used, the linkage between the different components may include products that result from, but are not limited to, copper (l)-catalyzed alkyne-azide cycloaddition (CuAAC), strain-promoted alkyne-azide cycloaddition (SPAAC), ruthenium- catalyzed azide-alkyne cycloaddition (RuAAC), inverse electron demand Diels-Alder reaction (IEDDA), Sulfur Fluoride Exchange (SuFEx), strain-promoted alkyne-nitrone cycloaddition (SPANG), ydrazine / oxime ether formation, thiol-ene radical reaction, thiol-yne radical reaction, thiol-Michael addition reaction, thiol-isocyanate reaction, thiol-epoxide click reaction, nucleophilic ring opening reactions (spring-loaded reactions) and traceless Staudinger ligation.

[0034] The flanking sequences may provide additional functionality, e.g. for performance of methods as described hereinafter. The analyte may also be attached to or associated with other components or moieties which may themselves not be nucleic acid in nature. Thus, for example the analyte may carry a cargo molecule, as described hereinafter. The analyte may also carry other moieties which may modify the functional properties of the analyte, as described hereinafter. 42.170.173772 / 01

[0035] The nucleotide sequence “barcode” refers to a single or double stranded oligonucleotide sequence, which may be contiguous or non-contiguous, that is designed to unambiguously identify a particular chemical entity, or its chemical composition, usually to discriminate between multiple different entities in a multiplex assay. When linked directly or indirectly to a chemical entity, the barcode can be used to quantify the amount of that particular entity present in a sample, even within complex biological materials such as tissues, blood and urine. To aid identification, where it is possible to do so (e.g. when larger barcodes are used, e.g. barcodes longer than about 15 or about 18 nucleotides) the barcode may provide a synthetic sequence that shares no substantial identity or complementarity with any endogenous DNA or RNA in the biological species under study.

[0036] In the present invention the barcode can be identified in a sample and is indicative of the presence of the analyte. The nucleic acid barcode may be covalently or noncovalently linked or otherwise associated, encapsulated, embedded or incorporated with or within the analyte. When more than one analyte is to be used, the barcode in each analyte is unique to that analyte, i.e. it can be discriminated from other barcodes in other analytes in the pool or set or as used in the method or use of the invention. This is conveniently achieved by each barcode having a different sequence, i.e. in at least one nucleotide in the barcode, e.g. in at least 2, 3, 4 or more nucleotides. To allow identification, each analyte provides a synthetic sequence that shares no substantial identity or complementarity with any endogenous DNA or RNA in the biological species under study.

[0037] The barcode principally acts as a marker of the analyte to allow its identification, but the nucleotides comprised in the barcode may have a dual function and may additionally offer some properties to the analyte. Thus for example, the barcode may affect the transport and / or activity of the analyte. For example one or more nucleotides in the barcode may interact with an entity in the cell, or aid that interaction. Equally one or more nucleotides in the barcode may assist with targeting or trafficking of the analyte.

[0038] The barcode may also include sequences of interest which reflect the sequence of molecules to be screened, i.e. to assess whether their distribution and / or activity is affected during the process of uptake and transport to a target subcellular location. In some embodiments the barcode may be an integral part of a molecule to be screened and ascribed an identification role merely based on its sequence or use.

[0039] The flanking sequences in said analyte may also provide these properties. By way of example an oligonucleotide which is the subject of the screening method may comprise or be attached to the barcode.

[0040] Furthermore, the flanking sequences may provide supplementary properties not provided by the barcode sequence. For example they may provide a sequence which may be recognised by a complementary nucleic acid sequence. This may allow the binding of 42.170.173772 / 01 that complementary nucleic acid e.g. for capture, detection or amplification. The recognition sequence may be common in more than one analyte to allow processing, i.e. only the barcode is unique.

[0041] Thus for example the analyte may comprise a poly(A) or poly(dT) sequence to enable simultaneous reverse transcription and amplification of barcode and total mRNA sequences. In yet another embodiment, the analyte may comprise an adapter sequence or primer binding site (PCR handle), or any other sequence, that allows direct sample processing with a minimal number of steps (for example, but not limited to, TruSeqTM adapter sequences, TruSeq™ RNA PCR primer sequences, NEBNext® adapter sequences, Nextera transposase adapter sequences, Nextera PCR primer sequences, AmpliSeq adapter sequences, or TruSight adapter sequences, 10X Genomics Capture sequence 1 , or 10X Genomics Capture sequence 2). In a preferred aspect such sequences are Illumina- compatible.

[0042] As referred to herein a “complementary” sequence is one that has the ability of an oligonucleotide or polynucleotide to hybridize (form base pair hydrogen bonds (under mammalian physiological conditions or similar conditions in vitro) and form specific duplex or double helical structures following established physiological pairing rules commonly known in the art). Complementary sequences include Watson-Crick base pairs or non-Watson-Crick base pairs and include natural or modified nucleotides or nucleotide mimics, at least to the extent that the above hybridization requirements are fulfilled. Sequence identity or complementarity is independent of modification.

[0043] Thus, in a preferred aspect the analyte additionally comprises one or more of: i) a universal recognition moiety which is invariant in the one or more analyte; ii) a moiety aiding interaction with said entity in said subcellular location in the cell; iii) a targeting moiety aiding targeting to a particular location in the cell; and / or iv) a cargo molecule.

[0044] A nucleic acid analyte comprising one or more of the above moieties is still considered a nucleic acid analyte. Thus, for example, if the cargo molecule interacts with the entity in the subcellular location, this is still considered interaction of the nucleic acid analyte with the entity.

[0045] The moiety / cargo molecule may be attached at the 3’ or 5’ end of the analyte or may be attached to a nucleotide within the analyte sequence. In some embodiments, a moiety / cargo molecule may be attached at more than 1 site, and the moieties / cargo molecules and / or linkers used at each site may be the same or different. Moieties, linkers and conjugation methods are known in the art and also described hereinbefore.

[0046] These moieties / cargo molecules may appear on one or more or all analytes when a plurality of analytes are used in said methods. For example, if their utility is to be tested in a 42.170.173772 / 01 screening method they may appear on only one or a portion of the tested analytes. In contrast, if they are not the component being tested, and are instead used as an integral part of the detection method, for example, they may appear on all of the analytes.

[0047] As referred to herein a “universal recognition moiety” is a portion of one or more nucleotide of said analyte to which a binding partner may bind. This binding partner may be a complementary hybridization or capture probe, for example, and has utility in downstream processes. Preferred examples of such moieties are the universal recognition sequences set forth in Table 2 (SEQ ID NOs: 225-236) which provide preferred aspects of the invention.

[0048] Entities in the subcellular location are described in more detail hereinafter. They may include entities such as kinases (or other enzymes that may achieve phosphorylation or whose activity results in the oligonucleotide substrate presenting a free phosphate group, e.g. by appropriate cleavage, referred to collectively herein as phosphorylating enzymes) or RNA-binding proteins, by way of example.

[0049] As used herein, the term "moiety" refers to a portion of a molecule with a specific function. This can include functional groups that are able to undergo specific reactions. A “moiety aiding interaction with said entity” is a component which may improve or facilitate the interaction. This may be by targeting the analyte to the entity or by providing a primary or secondary component necessary for the interaction. For example, when the interaction is the action of an enzyme on a substrate the moiety may provide all or part of the substrate or a necessary co-factor for the enzymatic reaction. In some cases the nucleic acid analyte may contain all components necessary for the interaction, but in other cases assistance may be offered by a further attached moiety.

[0050] A “targeting moiety aiding targeting to a particular location” improves targeting of the analyte to a selected location. This may include transfer through the cell membrane or at any point in its transfer to the final target subcellular location. The targeting moiety may appear on all analytes when that targeting is required for performance of the method of the invention. In a particular example, the targeting moiety may be a nuclear targeting moiety as described hereinafter. Other examples include 2’-O-hexadecyl (C16) which may be conjugated to siRNA to improve targeting and / or uptake.

[0051] Cargo molecule

[0052] A “cargo” molecule refers to a molecule which may be attached to the analyte. Analytes carrying a cargo molecule may a) consist of a nucleic acid barcode and an attached cargo molecule, or b) comprise a nucleic acid barcode and an attached cargo molecule. In the latter embodiment, the nucleic acid analyte additionally contains other nucleic acid sequences that may be present to provide a specific function (e.g. targeting or interacting with an entity) or may constitute a mimic of a test therapeutic oligonucleotide. 42.170.173772 / 01

[0053] The cargo molecule may be present as the test molecule, i.e. may consist of the test therapeutic oligonucleotide or reflect cargo to be attached to the therapeutic oligonucleotide, or may be presented to improve features of the analyte, e.g. its stability, solubility or transport. One or more, e.g. 2, 3 or 4 or more cargo molecules may be present, each of which may be the same or different.

[0054] Examples of moieties or cargo molecules attached to nucleic acid analytes are shown in Figures 7 and 8. In Figure 7 the moieties are attached to the nucleic acid analytes via linkers, which may be as described hereinbefore. Figure 8 provides an example in which the moiety or cargo molecule is a lipid. Figure 8B illustrates a barcode attached to a modified oligonucleotide. This conjugate may together be considered a nucleic acid analyte without the addition of a further moiety / cargo molecule (i.e. it is a nucleic acid molecule) or may be considered a nucleic acid analyte (barcode) to which a further moiety / cargo molecule is attached, in this case a modified oligonucleotide.

[0055] Examples of cargo molecules include oligonucleotides, small molecules (e.g. integrin ligand), lipids (e.g. C16), carbohydrates (e.g. GalNAc), peptides (e.g. cell-penetrating peptide), proteins (e.g. antibody). The oligonucleotide may be selected from the group consisting of mRNA, gRNA / CRISPR (guide RNA / Clustered Regularly Interspaced Short Palindromic Repeats), siRNA, ASO, miRNA (microRNA), InRNA (long noncoding RNA), shRNA (short hairpin RNA), saRNA (small activating RNA), AD-gRNA (adenosine deaminase acting on RNA (ADAR)-guiding RNA).

[0056] As referred to herein a “lipid” is a moiety or molecule that is soluble in nonpolar solvents. The term lipid includes amphiphilic molecules comprising a polar, water-soluble head group and a hydrophobic tail. Lipids can be of natural or synthetic origin. Non-limiting examples of lipids include fatty acids (e.g., saturated fatty acids, monounsaturated fatty acids, and polyunsaturated fatty acids), glycerolipids (e.g., monoacylglycerols, diacylglycerols, and triacylglycerols), phospholipids (e.g., phosphatidylethanolamine, phosphatidylcholine, and phosphatidylserine), sphingolipids (e.g., sphingomyelin), and cholesterol esters.

[0057] The terms "polypeptide", "peptide", and "protein", may be used interchangeably to refer to a string of at least three amino acids linked together by peptide bonds. The term "carbohydrate" refers to a polymer of sugars. As used herein, the term "small molecule", refers to an organic or inorganic compound, either synthesized in the laboratory or found in nature, which has a molecular weight of less than about 2000 g / mol, or less than about 1000 g / mol, or even less than about 500 g / mol. 42.170.173772 / 01

[0058] Features of barcode / analyte

[0059] Conveniently, the nucleic acid barcode is between 5-200, e.g. 5-100 nucleotides in length, for example 5-50 nucleotides in length. In certain analytes, the barcode may be between 6-12 nucleotides in length.

[0060] The full nucleic acid analyte is conveniently between 5 and 200 nucleotides in length, e.g. between 5 and 150 nucleotides in length, e.g. 5-100, 5-50 or 6-12 nucleotides in length.

[0061] The barcode and / or analyte may comprise a single-stranded portion, a doublestranded portion, a triple-stranded portion, a quadruple-stranded portion, a quintuple- stranded portion or combinations thereof.

[0062] The nucleic acid analyte and / or barcode may be single-stranded or double-stranded. The analyte and / or barcode may be a DNA or RNA molecule, e.g. selected from an mRNA, siRNA or antisense oligonucleotide (ASO). In another alternative, the nucleic acid analyte and / or barcode may be a mimic of such molecules.

[0063] Mimics and modified oligonucleotide analytes

[0064] Analytes for use according to the invention may act as mimics of particular therapeutic molecules. In a preferred aspect the methods of the invention allow for one or more functional properties of a test oligonucleotide, for which the analyte is a mimic, to be assessed. As such they may include molecules which mimic a particular molecule (such as an mRNA, siRNA or antisense oligonucleotide), but have modifications relative to the parent molecule. As referred to herein a “mimic” is a molecule which has one or more of the same sequence, structure, modifications (e.g. of individual nucleotides) or length as the parent molecule and is derivable therefrom. Such mimics may be prepared by post- synthesis / isolation modification of an oligonucleotide, e.g. the parent oligonucleotide without affecting functionality, except as described hereinafter.

[0065] The mimic may also be a molecule which retains one or more of the functional features of the oligonucleotide but is presented in the context of a different structure, e.g. using nucleic acid analogs or biorelevant monomers. Not only may oligomers of unnatural nucleotides or other organic building blocks be used, but also carbohydrates, heterocyclic or macrocyclic compounds or any organic molecule that comprises structural elements and conformation that provides a molecular electrostatic surface that mimics the same properties of the 3-dimensional conformation of the oligonucleotide may be used and prepared by methods known in the art. Thus the mimics may bear little or no resemblance to a nucleotide backbone.

[0066] Mimics may comprise an entirely synthetic non-oligonucleotide form or may retain one or more elements of the oligonucleotide on which it is based, e.g. by derivatizing one or more nucleotide or replacing one or more nucleotide with alternative non-nucleotide components. Structural elements considered redundant for the function of the 42.170.173772 / 01 oligonucleotide may be minimized to retain a scaffold function only or removed where appropriate.

[0067] When mimics retain one or more oligonucleotide elements, i.e. more than one nucleotide, such nucleotides may be replaced with a non-standard or structural analogue thereof. Nucleotides retained in the sequences may also be derivatised or modified (e.g. labelled, or otherwise modified as described hereinafter) as long as the functional properties of the oligonucleotide (which is being mimicked) for use according to the invention are retained. The mimics are referred to as being "derivable from" a certain oligonucleotide sequence. By this it is meant that the mimic is designed with reference to a defined oligonucleotide sequence, such that it retains one or more of the features of the oligonucleotide which are essential for its function, e.g. its sequence, length, nucleotide modifications. Such features may be provided by non-oligonucleotide components or one or more of the nucleotides or the bonds linking said nucleotides of the oligonucleotide may be modified so as to improve certain functions of the oligonucleotide such as stability or nuclease resistance, while retaining the structural features of the oligonucleotide which are essential for its function.

[0068] Examples of non-standard or structural analogue nucleotides which may be used are described below.

[0069] It will be appreciated that in screening methods the full functionality of a test therapeutic oligonucleotide may be undesirable. Thus whilst a mimic may have one or more functions in common with the test therapeutic oligonucleotide, e.g. size, overall structure, sequence, such that its transport and interaction with entities in the cell may be assessed by methods of the invention, it may be deficient in one or more functions of the parent molecule, e.g. unable to complete its full biological functionality. Providing the analyte retains at least one or more functional properties of the parent molecule from which it is derived it may still be considered a mimic.

[0070] In some embodiments, the nucleic acid analyte and / or barcode comprises one or more modified nucleotides, relative to unmodified RNA or DNA. In certain embodiments, each nucleotide of said nucleic acid barcode is a modified nucleotide. A modified nucleotide is a nucleotide other than a ribonucleotide (2'-hydroxyl nucleotide) or deoxyribonucleotide (2’-deoxy nucleotide).

[0071] In some embodiments, the modification encompasses a modified sugar moiety, a modified nucleobase, a modified internucleoside linkage or a combination thereof. For example, in some embodiments, the modified oligonucleotide comprises at least one modified sugar moiety, at least one modified nucleobase and at least one modified internucleoside linkage. In other embodiments, the modified oligonucleotide comprises at least one modified sugar moiety and at least one modified internucleoside linkage, but no 42.170.173772 / 01 modified nucleobases. In certain embodiments, the modified oligonucleotide comprises at least one modified internucleoside linkage, but no modified sugar moieties and no modified nucleobases.

[0072] The one or more modifications introduced can confer desirable properties to the modified oligonucleotide, for example and without limitation enhanced nuclease resistance, reduced immunogenicity, increased thermal stability, and improved affinity and / or specificity in base-pairing. They may further affect and / or modulate protein binding, bioavailability, circulation time, toxicity, biodistribution, tissue retention, cellular uptake, endosomal escape and / or other pharmacokinetic properties.

[0073] Methods for the synthesis of oligonucleotides and modified oligonucleotides are well established in the art and described for example in “Current protocols in nucleic acid chemistry,” Beaucage, S. L. et al. (Edrs.), John Wiley & Sons, Inc., New York, NY, USA, which is hereby incorporated herein by reference.

[0074] The modified analytes comprising oligonucleotides provided herein may contain one or more asymmetric centers and thus give rise to enantiomers, diastereomers, and other stereoisomeric configurations that may be defined, in terms of absolute stereochemistry, as (R) or (S), a or p such as for sugar anomers, or as (D) or (L) such as for amino acids or sugars. Included herein are all such possible isomers, as well as their racemic and optically pure forms.

[0075] In certain embodiments of the invention, substantially all of the nucleotides of an analyte of the invention are modified. In other embodiments of the invention, all of the nucleotides of an analyte or substantially all of the nucleotides of an analyte are modified, i.e. , not more than 5, 4, 3, 2, or 1 unmodified nucleotides are present in a strand of the analyte.

[0076] In certain embodiments of the invention, a modification of the sugar moiety comprises a modification at the 2' position (i.e., a nucleoside with a group other than a hydroxyl group or a deoxy group at the 2' position of the sugar ring). Exemplary 2' modifications include, but are not limited to: 2'-fluoro (2'-F), 2'-O-methyl (2'-0Me), 2'-O- methoxyethyl (2'-MOE), 2'-O-propargyl, 2'-azido, 2'-amino, 2'-ethyl, 2'- aminoethyl (EA), 2'-O- propylamine, 2'-O-[2-(methylamino)-2-oxoethyl] (2-0-NMA), 2'-O-[2-[(N,N- dimethylamino)oxy]ethyl] (2 -DMAOE), 2'-O-[2-[2-(N,N-dimethylamino)ethoxy]ethyl] (2 - DMAEOE), 2'-aminopropoxy, 2'-pivaloyloxymethyl (PivOM), 2'-phenylisobutyryloxymethyl (PiBuOM), 2'-O-[2-(guanidinium)ethyl] (2'-O-GE), 2'-O-(N-(aminoethyl)carbamoyl)methyl (2'- AECM), and 2'-O-hexadecyl. In some embodiments, 2' modifications are selected from 2'-O-, 2'-S-, or 2'-N-alkyl; 2'-O-, 2'-S-, or 2'-N-alkenyl; 2'-O-, 2'-S- or 2'-N-alkynyl; 2'-O-alkyl-O-alkyl; or heterocycloalkyl, wherein the alkyl, alkenyl and alkynyl can be substituted or unsubstituted Ci to C20 alkyl or C2to C20 alkenyl and alkynyl; and wherein the substituent groups of 42.170.173772 / 01 substituted alkyl, alkenyl and alkynyl groups may independently be selected from hydroxyl, amino, alkoxy, carboxy, benzyl, phenyl, nitro, thiol, thioalkoxy (S-alkyl), halogen, alkyl, aryl, alkenyl and alkynyl. Examples include, but are not limited to, 2'-O-[(CH2)n-O]m-CH3, 2'-O- (CH2)n-NH2, and 2'-O-(CH2)n-O-NH2, where n and m are positive integers independently selected from 1 to about 10.

[0077] In certain embodiments of the invention, a modification of the sugar moiety comprises a modification at the 4' position. A modification at the 4’ position may be the only modification of the sugar moiety or may be used in combination with a modification at the 2’ position. Exemplary 4' modifications and 2', 4' modifications include, but are not limited to, 4'- thio (4'-S); 4'-fluoro (4'-F); 4'-O-methyl (4'-OMe); 2'-F,4'-OMe; 2',4'-di-OMe; 2'-OMe,4'-F; 2'- OMe,4'-S; 4'-aminomethyl (4'-AM); 4'-aminoethyl; 2'-F,4'-AM; 2'-OMe,4'-AM, 4'-C- guanidinomethyl, 4'-C-guanidinocarbohydrazidomethyl (see e.g. https: / / doi.org / 10.1089 / nat.2019.0792 (Malek-Adamian et al., Nucleic Acid Ther., 2019, 29(4), p187-194) and https: / / doi.org / 10.1080%2F15476286.2022.2052641 (Gangopadhyay & Gore, RNA Biol., 2022, 1 , p452-467), which are hereby incorporated by reference).

[0078] In some embodiments, the analytes of the invention include one or more bicyclic sugar moieties, in which the 2'-carbon and the 4'-carbon (or the 2'-carbon and the T-carbon, or the 3'-carbon and the 5'-carbon, or the 3'-carbon and the 4'-carbon) of the sugar ring are linked via a bridging moiety. Examples of 4' to 2' bridging moieties include, but are not limited to: 4'-(CH2)-2' (methylene); 4'-(CH2)2-2' (ethylene); 4'-(CH2)3-2' (propylene); 4'-(CH2) — 0-2' (methylene-oxy) (also referred to as locked nucleic acid or O-LNA); 4'-(CH2) — S-2' (methylene-thio) (also referred to as S-LNA); 4'-(CH2) — N(R)-2' (methylene-amino) (also referred to as amino-LNA); 4'-(CH2) — N(OCH3)-2' (and analogs thereof; see e.g., U.S. Pat. No. 8,278,425 or W02008 / 150729); 4'-(CH2)— N(R)— 0-2' (methylene-amino-oxy) (see, U.S. Pat. No. 7,427,672); 4'-(CH2)— O— N(R)-2' (methylene-oxy-amino); 4'-(CH2)— O— N(CH3)2-2' (or analogs thereof; see e.g., U.S. Patent Publication No. 2004 / 0171570); 4'-(CH2)2— 0-2' (ethylene-oxy) (also referred to as ethylene-bridged nucleic acid or ENA); 4'-CH(CH3) — 0-2' (methyl-methylene-oxy) (also referred to as constrained ethyl or cEt); 4'-(CH2) — CH(CH3)-2' (methyl carbocyclic) (see, e.g., Chattopadhyaya et al., J. Org. Chem., 2009, 74, p118-134); 4'-CH(CH2OCH3) — 0-2' (and analogs thereof; see, e.g., U.S. Pat. No. 7,399,845); 4 - C(CH3)(CH3) — 0-2' (and analogs thereof; see e.g., U.S. Pat. No. 8,278,283 or W02009 / 006478); 4'-CH2— O— N(CH3)2-2' (see, e.g., U.S. Patent Publication No. 2004 / 0171570); 4'-(CH2)— C(=CH2)-2' (and analogs thereof; see, e.g., U.S. Pat. No. 8,278,426); wherein each R is independently H, Ci to C2o alkyl, or a protecting group. The entire contents of each of the foregoing references are hereby incorporated herein by reference. Many other bicyclic and tricyclic sugar and sugar surrogate ring systems are known in the art that can be used in the modified oligonucleotides described herein. 42.170.173772 / 01

[0079] Additional representative patents and publications that teach the preparation of nucleotides with bridged bicyclic or tricyclic sugar moieties include, but are not limited to, the following: US Patent Nos. 6,268,490; 6,525,191; 6,670,461 ; 6,770,748; 6,794,499; 6,998,484; 7,053,207; 7,034, 133;7, 084, 125; 7,399,845; 7,427,672; 7,569,686; 7,741 ,457; 8,022,193; 8,030,467; 8,278,425; 8,278,426; 8,278,283; US 2008 / 0039618; US 2009 / 0012281; https: / / doi.org / 10.1039 / C5CC02024G (Yamaguchi et al., Chem. Comm., 2015, 51, p9737- 9740); and https: / / doi.org / 10.3390 / molecules25071732 (Komine et al., Molecules, 2020, 25(7), p1732), the entire contents of each of which are hereby incorporated herein by reference.

[0080] The ribose moiety may be substituted for another sugar moiety - for example, but without limitation: 1 ,5,-anhydrohexitol to produce hexitol nucleic acid (HNA), threose to produce threose nucleic acid (TNA), arabinose to produce arabino nucleic acid (ANA), anitol to produce anitol nucleic acid, or manitol to produce manitol nucleic acid (see Leumann, Bioorg. & Med. Chem., 2002, 10, p841-854). The ribose molecule can also be replaced with non-sugar moieties such as cyclohexene to produce cyclohexene nucleic acid (CeNA), glycol to produce glycol nucleic acids (GNA) or a morpholine ring system to produce morpholino nucleic acids. In some embodiments, the sugar surrogates may be additionally modified, for example by adding or altering various substituent groups as described above for modifications at the 2' position such as 2'-deoxy-2'-fluoro-P-d-arabinonucleic acid (2'- FANA), 3'-fluoro HNA (F-HNA) or N-fluoroalkylated morpholinos (see https: / / doi.org / 10.1002 / chem.202203248 (Debreczeni et al., Chem. Eur. J., 2022, 29(11), p1- 15)).

[0081] In some embodiments, the modification of the sugar moiety comprises an unlocked acyclic nucleic acid (UNA), wherein at least one of the covalent carbon-carbon or carbonoxygen bonds of the sugar moiety is removed. For example, the covalent carbon-carbon bond between the C2' and C3' carbons of the sugar may be removed (see https: / / doi.org / 10.1093 / nass / nrn068 (Jensen et al., Nucl. Acids Symp. 2008, 52(1), p133- 134) and Fluter et a / ., Mol. Biosyst., 2009, 5, p838-843 hereby incorporated by reference). Representative patents that teach the preparation of UNA include, but are not limited to, U.S. Pat. No. 8,314,227; and U.S. Patent Publication Nos. 2013 / 0096289; 2013 / 0011922; and 2011 / 0313020, the entire contents of each of which are hereby incorporated herein by reference.

[0082] Preferred modifications at the ribose group include, and may be independently, for each occurrence, selected from 2'-O-methyl, 2'-fluoro, 2'-F-arabino, 2'-methoxyethyl (MOE), 2'-amino, 2' deoxy, 2'-O-al lyl , locked nucleic acid, unlocked nucleic acid, 2',4'-constrained 2'- O-ethyl-bridged nucleic acid, arabinose, hexose, cyclohexenyl nucleic acid, hexitol nucleic acid, glycol nucleic acid, 4'-thioribonucleoside, and 4'-C-aminomethyl-2'-O-methyl. In certain 42.170.173772 / 01 aspects, the modification at the phosphate group is independently for each occurrence selected from phosphorothioate (PS), phosphorodithioate, alkylated phosphorothioate, boranophosphate, methylphosphonate, phosphoramidate, or amide backbone. In certain embodiments, the modification at the nucleobase group is independently for each occurence selected from 5- methylcytosine, 5-hydroxymethylcytosine, 5-methyluracil, 5-ribosyluracil (pseudouracil), 7- methylguanine, inosine, xanthine, hypoxanthine, 3-methylcytidine, dihydrouridine, N6- methyldeoxyadenosine, N4-methylcytosine, 5-hydroxymethylcytosine, 5- formylcytosine, and 5- carboxylcytosine.

[0083] As mentioned above, modifications may be selected to mimic therapeutic and diagnostic ssDNA or ssRNA, e.g. ASO, ssiRNA (single-stranded siRNA), sgRNA and aptamers.

[0084] In some embodiments, said nucleic acid analyte and / or barcode comprises at least 2, 3, 4, 5, 6, 7, 8, 9, or 10 contiguous 2'-O-methyl nucleotides.

[0085] In some embodiments, said nucleic acid analyte and / or barcode comprises at least 2, 3, 4, 5, 6, 7, 8, 9, or 10 consecutive 2'-methoxyethyl nucleotides. In some embodiments, said nucleic acid analyte and / or barcode comprises at least one locked nucleic acid (LNA) nucleotide. In some embodiments, each internucleotide linkage of said nucleic acid analyte and / or barcode is a phosphodiester internucleotide linkage or a phosphorothioate internucleotide linkage. In some embodiments, at least one internucleotide linkage of said nucleic acid analyte and / or barcode is a phosphorothioate internucleotide linkage.

[0086] In some embodiments, the nucleotides at positions 1-2, 1-3, 1-4, 1-5, 1-6, 1-7 or 1-8 from the 3' end of said nucleic acid analyte and / or barcode are connected to adjacent nucleotides via phosphorothioate linkages. In some embodiments, the nucleotides at positions 1-2, 1-3, 1-4, 1-5, 1-6, 1-7 or 1-8 from the 5' end of said nucleic acid analyte and / or barcode are connected to adjacent nucleotides via phosphorothioate linkages. In some embodiments, the terminal 2, 3, 4, 5, 6, 7, 8, 9 or 10 nucleotides at both the 5' and 3' terminus of said nucleic acid analyte and / or barcode, independently, are linked with phosphorothioate internucleotide linkages.

[0087] Examples of analytes (i.e. nucleic acid molecules including barcodes) that may be used in methods of the invention and methods of their production are described in WO2024 / 069235 which is incorporated herein by reference (see for example Figures 4, 5, 11-15, and 23, and Example 3 and the descriptions of barcodes that may be used).

[0088] In a preferred aspect of the invention, neither the analyte nor the barcode is encapsulated in a nanoparticle. As referred to herein, encapsulation refers to the analyte / barcode being wholly contained within a structure that protects it from access to the exterior which may aid its stability and / or transport. The analyte may thus be delivered in the naked, unencapsulated form. In this scenario, in one alternative a delivery agent may be 42.170.173772 / 01 used, such as an aptamer, a small molecule, a peptide, a polypeptide or a lipid. In some cases, the analyte and / or barcode may be encapsulated, e.g. in a lipid nanoparticle. Thus in a further alternative the analyte and / or the barcode is encapsulated in a nanoparticle. A "nanoparticle" refers to particles having dimensions that are measured on the nanometer scale. Viral transduction (e.g. adenovirus or AAV) may also be used

[0089] In methods according to the invention, one or more analyte may be used. As noted above, when more than one analyte is used at least one barcode unique to each analyte is used. When more than one analyte is used this may be considered a plurality or a library of analytes. Such a library includes at least 2, preferably at least 5, 10, 20, 30, 40, 50 or 100 analytes, e.g. up to 10, 50, 100 or 200 analytes. This allows multiplexing and the selective identification of which barcode, and hence analyte, from the library reached the subcellular location of interest. Thus in a preferred aspect, more than one analyte is administered to said cell (e.g. having the number of analytes indicated above for the library), and said method determines which analyte is transported to a target subcellular location. This may be just one or more than one analyte. When the presence of more than one barcode is assessed, this may mean that the barcode of more than one analyte is assessed, or more than one barcode of each analyte is assessed, or a combination of both.

[0090] Exemplary analytes which act as mimics for various therapeutic molecules of interest (ASO, siRNA) are shown in Figure 4. As is evident the analyte may include various modified nucleotides, the barcode sequence may be placed internally, or a terminal end of the analyte and the analyte itself may take various forms including single stranded and double stranded and may include a hairpin structure.

[0091] Assessment of the barcode

[0092] The barcode sequences are determined for example, using techniques including, but not limited to, next-generation sequencing, microarray, quantitative PCR, hybridization (such as plate-oligo fluorescence (POF) assay), in-situ hybridization (such as fluorescence in situ hybridization (FISH)), and branched DNA assays. In a preferred aspect they are determined by sequencing, including bulk, single-cell or spatial sequencing. For single cell sequencing cells may be separated for analysis prior to lysis, e.g. in microfluidic drops, nanowells or hydrogels by way of example.

[0093] Suitable sequencing techniques include, but are not limited to, bridge amplification sequencing / Solexa (Illumina), Ion semiconductor sequencing, GenapSys sequencing, combinatorial probe-anchor synthesis (cPAS), sequencing by ligation (SOLiD sequencing), single molecule real-time sequencing (SMRT), Heliscope single molecule sequencing, nanopore sequencing, pyrosequencing, and Sanger sequencing.

[0094] In some embodiments, the detection and quantification of the nucleic acid barcodes described herein requires amplification of the one or more barcodes. For example, 42.170.173772 / 01 sequencing library preparation for deep sequencing applications on Illumina instruments generally involves PCR amplification. Those skilled in the art will appreciate that any amplification method known in the art, including isothermal amplification methods, may be applied without departing from the scope of the present disclosure.

[0095] In some embodiments, the innate activity of certain DNA-dependent DNA polymerases can be exploited for simultaneous, 1-step reverse transcription and amplification of analytes comprising barcodes. Suitable DNA polymerases are described in WO2024 / 069235.

[0096] In some embodiments, amplification and quantitative detection of nucleic acid barcodes can be performed simultaneously. Exemplary amplification methods that allow for direct quantification during amplification include, but are not limited to, quantitative real-time polymerase chain reaction (qPCR and qRT-PCR), digital PCR, digital droplet PCR (ddPCR), ligation-dependent probe amplification (MLPA), digital MLPA (dMLPA), real-time multiplex loop-mediated isothermal amplification (RT-LAMP), RCA-assisted CRISPR / Cas9 cleavage (RACE), RCA-FRET DNA assay, RCA-assisted single-molecule flow cytometry and the like. Other amplification techniques include, but are not limited to, rolling circle amplification (RCA), recombinase polymerase amplification (RPA), loop-mediated isothermal amplification (LAMP), exponential amplification reaction (EXPAR), duplex-specific nuclease signal amplification (DSNSA), nicking enzyme amplification reaction (NEAR), nucleic acid sequence- based amplification (NASBA), strand displacement amplification (SDA), multiple displacement amplification (MDA), helicase-dependent amplification (HOA), hybridization chain reaction (HCR), catalyzed hairpin amplification (CHA), and single chimeric primer isothermal amplification (SPIA).

[0097] By way of example, in methods described herein isolated analytes may be amplified and sequenced by subjecting the product (e.g. a ligation product) to reverse transcription to generate a cDNA, which method may be primed by an oligonucleotide complementary to a sequence presented in the analyte or attached to the analyte, e.g. via ligation; followed by PCR amplification of the cDNA followed by sequencing or hybridization as appropriate. Cells

[0098] In methods of the invention the analytes are administered to a cell. That cell may be present in a subject or a cell-containing sample. As such the method may be performed in vitro, ex vivo or in vivo. When performed in vivo preferably the method is a non-therapeutic and / or a non-surgical method and / or is performed on non-human animals. A therapeutic method is considered one which provides a partial or full curative effect on an animal which has a disease, condition or disorder which is considered abnormal. Surgical methods are considered those which require the intervention of a trained physician. 42.170.173772 / 01

[0099] The term “in vivo" refers to events that occur within a living multi-cellular organism, such as a human and a non-human animal.

[0100] Subjects according to the invention include mammals, particularly humans and non- human animals such as primates, mice, rats, dogs, pigs, sheep, guinea pigs, goats, rabbits, horses, cows and cats.

[0101] The cell may be any eukaryotic cell (e.g. from a subject described hereinbefore) and is provided in a liquid sample to which the analyte(s) may be applied. The cells may be present in suspension or they may be fixed to a solid support. These cells may be derived from a number of sources including from tissue, fluid, solid or waste samples from subjects or cultured cells and may be used as obtained or processed to enrich or isolate the cells. Suitable sources include blood plasma, serum or whole blood, urine, faeces, lacrimal fluid, sputum, semen, cerebrospinal fluid, saliva, culture swabs, skin samples and tissue samples. Tissue samples may be obtained from organs and tissues such as liver, lung, heart, kidney, pancreas, spleen, skin, adipose tissue, eye, brain, spinal cord, muscle, testes, ovaries, uterus, mouth, oesophagus, stomach, small intestine, large intestine, and gallbladder, by way of example. Optionally the cells can be tumours, including malignant or benign. Administration

[0102] The analyte(s) is administered to the cell by any convenient means. When administration is to a cell in vitro or ex vivo, administration refers to contacting the cell with the analyte(s). This may be achieved by introducing the analyte(s) into the liquid containing the cell or otherwise directly applying the analyte(s) to the cells or other material in contact with those cells. When performing administration in vivo this may be by any convenient method of delivery suitable for delivery of a nucleic acid analyte. The analyte may be administered systemically or locally relative to the target cells of interest. For local delivery topical administration may be performed. In certain embodiments of the invention, the analyte(s) may be administered intravenously, intramuscularly, subcutaneously, intrathecally, intracerebrally, intraventricular, intraspinal, intradermally, intranasally, intratracheally, orally, transcutaneously, or mucosally.

[0103] Conveniently, for in vivo administration, the analyte is provided in the form of a composition comprising also pharmaceutically acceptable excipients, carriers or diluents.

[0104] "Pharmaceutically acceptable" as used herein refers to ingredients that are compatible with other ingredients of the compositions as well as physiologically acceptable to the recipient. The nature of the composition and carriers or excipient materials, dosages etc. may be selected in routine manner according to choice and the desired route of administration, etc. Dosages may likewise be determined in routine manner and may depend upon the nature of the molecule, age of patient, mode of administration etc. 42.170.173772 / 01

[0105] Pharmaceutical compositions as disclosed herein include liquid solutions or syrups, solid compositions such as powders, granules, tablets or capsules, creams, ointments and any other style of composition commonly used in the art. Suitable pharmaceutically acceptable diluents, carriers and excipients for use in such compositions are well known in the art.

[0106] The dose of the analyte to be used may be provided in the amount of from 1 pg / kg to 100 mg / kg, e.g. 10 pg / kg to 10 mg / kg body weight, for example from 0.5 to 5 mg / kg (referring to total analyte used rather than per analyte if more than one analyte is used). The clinician / researcher will be able to calculate an appropriate dose for a subject based on all relevant factors, e.g. age, height, weight, and condition of the subject and the analyte to be administered. Typically sub-therapeutic doses may be used.

[0107] Conveniently a library of analytes are administered to the cells or subject at one time, e.g. 5 to 100, such as 15 to 50 analytes at one time.

[0108] Following administration, the subject or cell(s) which has been brought into contact with the analyte(s) is maintained for a sufficient period of time to allow the analyte(s) to be taken up by the cell and to be transported to the subcellular compartment. The time for this to be achieved will vary depending on a number of factors including the environment of the cell (particularly if the cell is in a subject) and the target subcellular compartment. By way of example, in in vivo methods, this period may be from 60 minutes to 24 hours, or from 7 to 28 days. For in vitro methods, by way of example the contacting period to allow transport may be from 10 to 120 minutes, or from 6 to 72 hours.

[0109] Target subcellular location

[0110] The target subcellular location as referred to herein is a spatially defined location within the cells, preferably bounded by a membrane to provide an organelle within the cell. A “target” location depends on the intent of the method and may be selected appropriately. Conveniently, in a preferred aspect of the invention, the subcellular location is the cytoplasm or the nucleus. In the alternative, the subcellular location may be other organelles of the cells, including the endosome or lysosome. The presence of the analyte in these locations may be indicative of likely functional activity, as described hereinafter. In some cases the presence of the analyte in a particular location may be compared to its presence in another location, e.g. comparing its presence in endosomes (and or lysosome) to cytoplasm (and / or nucleus) to track transport through the cell. In yet other cases, the presence of the analyte in a particular location may be compared to total analyte presence within the cells, i.e. its presence across all subcellular locations and / or organelles within the cells. 42.170.173772 / 01

[0111] Transportation

[0112] The analyte is “transported” to the target subcellular location by passive or active transport mechanisms within the cell. This may be affected by components provided on the analyte, e.g. if nuclear localization peptides are conjugated to the analyte. Collecting said cell or portion thereof

[0113] In accordance with the method of the invention after administration of the analyte(s) and maintenance to allow the analyte(s) to reach the target subcellular location, the cell(s) is collected. In the case of an in vitro or ex vivo method, the sample in which the method has been performed may require no additional collection as it already contains the cells of interest. However, further enrichment or isolation of the cells may be performed, e.g. by separation from liquid or other material in the sample. In the case of in vivo methods, a sample is obtained from the subject to which the analyte(s) has been administered, e.g. a fluid, solid or waste sample, in particular a bodily sample, such as a tissue sample, as discussed hereinbefore.

[0114] Prior to, or during, the assessment for the presence of one or more barcode the cells may be further processed, e.g. by permeabilization, lysis and / or collection of one or more organelles or components, e.g. collection of the cytoplasm or nucleus. This largely depends on the intended method of assessment. The collection of less than the complete cell constitutes collection of a “portion” of the cell which generally includes the organelles or components present at the site of the target subcellular location to allow its analysis. In cases where direct analysis of the target subcellular location is to be performed to determine if the barcode and hence analyte is present, necessarily this is separated from other subcellular locations. Thus, for example, the nucleus may be isolated if that is the target subcellular location. However, if the determination of the location is based on interaction with an entity at that location which has occurred by the time the cells are collected, the whole cells or portions that contain the molecules required for confirming that interaction may be collected.

[0115] Assessing whether one or more barcode is present in the target subcellular location

[0116] The assessment may be quantitative or qualitative, i.e. a binary result which denotes whether a barcode / analyte is, or was, present or not in the target subcellular location, or a quantitative result that indicates how much of the barcode / analyte is, or was, present in the target subcellular location. This may be useful for example if the transport of analytes with different properties are being screened.

[0117] The assessment is made by determining if a) one or more barcode is present, i.e. detectable in the target subcellular location, or b) if one or more analyte interacted with an entity in the target subcellular location. 42.170.173772 / 01

[0118] The former may be determined by assessing directly if the sequence of the barcode is present in the sample. Conveniently this is achieved by appropriate sequencing techniques as described hereinbefore.

[0119] In the alternative the assessment is made based on the interaction of the analyte with an entity in the target subcellular domain.

[0120] The entity is present, or was present, in one or more (preferably one), but not all locations in the cell. Interaction with that entity may thus be used as evidence that the analyte has been in that location. The entity produces, on interaction with an analyte (including any attached moiety or cargo molecule), a product that may be detected which is evidence of the interaction. That product may be detected directly or used to generate a detectable signal. Thus, in one aspect, the present invention provides an assessment which comprises determining whether one or more of said analyte has interacted with an entity in said subcellular location.

[0121] The entity may be, by way of example, an RNA-binding protein, a protein, an enzyme, a ligand or a receptor. The entity may be selected based on the analyte or vice versa. In a preferred aspect, the entity is not a polymerase. Furthermore, whilst the entity may be provided to the cell, e.g. exogenously or via recombinant expression, in a preferred embodiment the entity is present in the cell endogenously. In a further preferred aspect, the analyte is not amplified in said cell, though it may be amplified after lysis of said cell for the purpose of analysis.

[0122] By way of a first example, the entity may be an enzyme that i) processes an oligonucleotide in such a way that a free phosphate group is presented on the oligonucleotide substrate, e.g. left behind on at least one of the one or more generated oligonucleotide products, e.g. an endo- or exoribonuclease, or that ii) can transfer a phosphate group to an oligonucleotide, e.g. preferably a kinase, collectively referred to herein as phosphorylating enzymes. In this embodiment, the phosphorylating enzyme, e.g. kinase, is present in a particular subcellular location. The analyte, when present in that location, is phosphorylated. The phosphorylated analyte may be detected as evidence of its presence in that location. This method forms a preferred aspect of the invention and is illustrated in Figure 1.

[0123] Phosphorylation as a proxy for endosomal escape

[0124] Enzymes that produce an oligonucleotide substrate with a free phosphate group (i.e. are capable of leaving behind a phosphate group on an oligonucleotide or transferring a phosphate to an oligonucleotide) include some endo- and exoribonucleases, and kinases.

[0125] Polynucleotide kinases (PNKs) are a class of enzymes that catalyzes the transfer of a monophosphate from a nucleoside triphosphate (NTP; usually ATP) to the 5' end of nucleic acid polymers (Saito et al., Genome Biol. Evol., 2019, 11(10), p2713-2726). PNKs 42.170.173772 / 01 phosphorylate the 5'-hydroxyl groups of double-stranded RNA (dsRNA), single-stranded RNA (ssRNA), double-stranded DNA (dsDNA) and double-stranded DNA:RNA hybrids, but not ssDNA. dsRNA is phosphorylated more efficiently than dsDNA, and the RNA component of a DNA: RNA hybrid is phosphorylated more efficiently than the DNA component.

[0126] PNKs are essential for RISC-mediated target cleavage by non-phosphorylated siRNAs. Clp1 is a polyribonucleotide 5-hydroxyl kinase endogenously expressed in eukaryotes including Homo sapiens (Hs), Mus musculus (Mm), Caenorhabditis elegans (Ce), Drosophila melanogaster (Dm), Arabidopsis thaliana (At), Schizosaccharomyces pombe (Sp), and Saccharomyces cerevisiae (Sc). It is involved in pre-tRNA (precursor transfer RNA) splicing and mRNA 3 -end formation. Other known enzymes similar to Clp1 are Nol9, and Grc3 which are present in some eukaryotes and are involved in pre-rRNA processing.

[0127] It has been previously demonstrated that immunodepletion of hCLP1 (human CLP1) in HeLa cytoplasmic extracts renders siRNAs unable to efficiently cleave their target RNA, in contrast to control cytoplasmic extracts. This is likely due to impaired incorporation of the siRNA into the RNA-induced silencing complex (RISC) (Weitzer et al., Nature, 2007, 447(7141), p222-226). Later it was shown that siRNA efficiency was decreased in Clp1 kinase-dead (Clp1K / K) murine cells, and this effect was not rescued by the overexpression of other kinases. These findings indicate that in mouse cells it is Clp1 that mainly acts to phosphorylate the 5' end of RNAs in the siRNA pathway, with no apparent involvement of Nol9.

[0128] 5’-phosphorylation of siRNA may therefore be used as a surrogate marker for endosomal release of siRNA, as an indirect indication of exposure to the cytoplasmic compartment where 5'-hydroxyl kinase (or another enzyme which achieves phosphorylation) resides. Previous literature reports that the 5’-phosphorylated guide strands of siRNA in liver is around 0.1-0.3% of the total level of guide strands in hepatocytes (Trubetskoy et al., Nucleic Acids Res. 2017, 45(3), p1469-1478). An approximate 16% of the total 5’- phosphorylated guide strands are then incorporated into RISC.

[0129] PNK-mediated phosphorylation as readout for endosomal escape can be applied to all classes of short RNAs. In fact hCLPpI phosphorylates double and single stranded RNAs between 59 and 39 nucleotides in length, both with 2-nucleotide overhangs ds RNAs and blunt-ended stranded, and substrates as short as 4 nucleotides (Weitzer et al., supra). Hence the method may be used for quantifying analytics beyond siRNAs, such as ASO gapmers, splice-switching oligonucleotides (SSOs) and saRNA.

[0130] Depending on the enzyme, e.g. kinase, selected and its location, phosphorylation by an enzyme, e.g. kinase may occur in another compartment, e.g. the nucleus. References to a cytoplasm as used herein may also refer to other compartments. 42.170.173772 / 01

[0131] Thus in a preferred aspect, the assessment is made by determining whether one or more of said analyte has interacted with a phosphorylating enzyme, e.g. a kinase, in said target subcellular location comprising the additional steps of: c) collecting at least the one or more analyte which has been phosphorylated; d) determining the barcode of each analyte that has been phosphorylated to determine which analyte was transported to the target subcellular location.

[0132] In a particularly preferred aspect, the subcellular location is the cytoplasm (and / or nucleus) and said assessment is made by determining whether one or more of said analyte has interacted with a phosphorylating enzyme, e.g. a kinase, in the cytoplasm (and / or nucleus) comprising the additional steps of: c) collecting at least the one or more analyte which has been phosphorylated; d) determining the barcode of each analyte that has been phosphorylated to determine which analyte was transported to the cytoplasm (and / or nucleus).

[0133] In step c) the cell is optionally lysed or permeabilized.

[0134] The method may include the collection of just the one or more analyte which has been phosphorylated or also other analytes which may not have been phosphorylated. In the latter case, discrimination between the phosphorylated and non-phosphorylated analytes is performed in step d).

[0135] In a preferred aspect the kinase is selected from Clp1 , T4 polynucleotide kinase (PNK), mammalian polynucleotide kinase (mPNK), Nol9, and Grc, preferably Clp1.

[0136] To achieve this method, the analyte for use in the method is appropriately selected to allow phosphorylation. Conveniently the analyte or the portion to be phosphorylated has one or more of the following features: a) it is ssRNA, dsRNA or a double-stranded RNA:DNA hybrid; b) where it is double stranded it may have a 5’ overhang or short (1-4 nt) 3’ overhang or may have blunt ends; c) has a length of 8-80, preferably 14-40 nucleotides; d) the first two 5’ nucleotides adopt an A-form helical conformation with the ribose in C3’ endo orientation (this excludes DNA and certain modifications such as UNA and 2’-F- ANA); e) an available 5’OH or 3’OH group for phosphorylation; f) nucleotide modifications to mimic therapeutic or diagnostic ssRNA, e.g. 2’F, 2’OMe, 2’MOE, LNA, PS (providing they do affect downstream processing).

[0137] As an alternative to an available 5’OH or 3’OH group, a terminal modification may be provided which is removed in target cells, e.g. an m7G cap which is removed in cells.

[0138] Optionally the analyte may comprise a 5’ terminal non-variable shield of 1-3 nucleotides to avoid sequence-dependent bias when detecting the barcode. 42.170.173772 / 01

[0139] In line with the method of the invention each analyte includes a barcode. In the present embodiment it is preferably 3-40, e.g. 5-12 nucleotides in length. A universal recognition sequence may also be provided. In some embodiments, the universal recognition sequence is complementary to a capture oligonucleotide which may be provided on magnetic beads to allow capture. In an alternative, the universal recognition sequence may be complementary to a reverse transcription (RT) primer, by way of example. In some cases, the capture oligonucleotide may double as an RT primer.

[0140] Preferred modifications of the nucleic acid analyte include alternating 2’F and 2’OME modifications with 3’ terminal phosphorothioate backbone modifications on each side. Modifications may be used to stabilize the required conformation.

[0141] The analyte is designed to meet substrate selectivity criteria of the enzyme, e.g. kinase, of interest, e.g. the Cl p1 kinase. For example, RNA is phosphorylated by this kinase more efficiently than DNA, and dsRNA with 5’ overhangs or blunt ends is phosphorylated more efficiently than dsRNA with 3’ overhangs. The substrate must have a minimum length of 2 nucleotides and structural data on Caenorhabditis elegans Clp1 revealed that the first two 5’ nucleotides are involved in binding to the enzyme.

[0142] Preferred analytes that may be used are single-stranded RNA oligonucleotides with a 5’OH group and which are 8 and 80 nucleotides in length, preferably between 14-40 nucleotides in length. In other embodiments, the analyte is a single-stranded DNA oligonucleotide with a 5’OH group and which is between 8 and 120 nucleotides in length, preferably between 14-40 nucleotides in length, wherein the first two or more nucleotides are unmodified RNA nucleotides or modified nucleotides that adopt the standard C3’-endo pucker.

[0143] In other embodiments, the analyte may be a double stranded molecule in which the second strand is partially (>50%), substantially (>80%) or perfectly complementary to the first strand. The complementary strand can be RNA or DNA and can include one or more modified nucleotides and / or one or more non-phosphodiester linkages. The first and second strands can be of the same length or of different lengths. In the duplex, the first strand may have a 5’ overhang, a blunt end or a 3’ overhang. The duplex may be an RNA: RNA homoduplex or an RNA:DNA heteroduplex.

[0144] Example sequences of analytes which include barcodes are shown in Table 1. 42.170.173772 / 01

[0145] Table 1 :

[0146] Example analyte sequences. Barcodes are highlighted in bold. rA / rC / rG / rll indicates unmodified 2’OH RNA, lower-case a / c / g / u indicates 2’F (2’-fluoro), mA / mC / mG / mll indicates 2’0Me (2’-methoxy), / 2MOErA / / 2MOErG / indicates 2’MOE (2’-methoxyethyl) and * indicates PS (phosphorothioate). 2 / 01 2 / 01 2 / 01 2 / 01 2 / 01 2 / 01 42.170.173772 / 01

[0147] Table 2 provides examples of universal recognition sequences that may be used in this or other embodiments.

[0148] Table 2: Example universal recognition sequences 42.170.173772 / 01

[0149] Table 3 provides examples of second strands that may be used in double stranded designs.

[0150] Table 3: Example complementary strand sequences for double-stranded designs. mA / mC / mG / mll indicates 2’0Me (2’-methoxy), / 5AmMC6 / indicates 5' amino modifier 06 and NHSTetrazine / / 5AmMC6 / indicates a / 5AmMC6 / group reacted with an NHS-activated tetrazine.

[0151] Determining the barcode of each analyte that has been phosphorylated to determine which analyte was transported to the target subcellular compartment (e.g. the cytoplasm and / or nucleus) may be performed by any convenient method. Ultimately, phosphorylated analytes need to be discriminated from those that have not been phosphorylated. This may be achieved by selective capture and / or amplification, by way of example. In one example of a suitable method after collecting of the cells, material including the analytes is collected, e.g. the RNA of the sample is collected. 42.170.173772 / 01

[0152] The sample may contain a mixture of phosphorylated and unphosphorylated analytes. The analytes may be collected if they contain a universal recognition moiety for doing so, e.g. they may be captured by binding to a capture oligonucleotide. A phosphorylation-dependent reaction is then performed (in the alternative, or in addition the capture step may be performed after subsequent steps). The phosphorylation-dependent reaction may be an enzymatic ligation reaction by virtue of a ligase, a chemical ligation reaction or may be activation of the 5’ phosphate group for chemical ligation (e.g. using a condensing reagent selected from the group of 1-(3-dimethylaminopropyl)-3- ethylcarbodiimide hydrochloride (EDC), 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDCI) (Hirano & Komatsu, RSC ADV., 2022, 12(38), p24471-24477; Boos et al., Nucleic Acids Res., 2013, 41(15), e145), cyanogen bromide (Zhang et al., Mol. Cell. Biol., 2014, 34(12), p2318-2329), imidazole derivatives (Moreno et al., Nucleic Acids Res., 2009, 37(6), p1925-1935; Hill et al., ACS Omega, 2023, 8(43), p40463-40481) or 1 -hydroxybenzotriazole (HOAt)). Chemical ligation may be phosphoramidate ligation, wherein the 5’ phosphate group on the barcode is activated using EDC and chemically ligated to an adapter oligonucleotide containing a 3’ amine group.

[0153] By way of example, adapters may be bound via ligation to only those analytes which have been phosphorylated. The use of adapters is illustrated in Figures 5 and 6.

[0154] Suitable adapters include those shown below in Table 4.

[0155] Table 4: Exemplary Adapters. / 56-FAM / indicates 5’ 6-carboxyfluorescein, N indicates mixed DNA bases (A, C, G, T) and rN indicates mixed RNA bases (rA, rC, rG, rll). 42.170.173772 / 01

[0156] Ligation may be achieved, e.g. by use of T4 RNA ligase 1. This enzyme catalyzes the ligation of a 5’-phosphoryl-terminated nucleic acid donor (carried on the analyte) to a 3’ hydroxyl-terminated nucleic acid acceptor (termed here the 5’ Adapter) through the formation of a 3’ — > 5’ phosphodiester bond with hydrolysis of ATP to AMP and PPi. In the alternative a ligase such as T3 DNA Ligase, SplintR Ligase, 5' AppDNA / RNA Ligase, T4 RNA Ligase 1 , T4 RNA Ligase 2 or RtcB ligase may be used for 3’ adapter ligation. The 5’ adapter ligation reaction can be performed in a single step or as a 2-step reaction (e.g. 5’ adapter preadenylation followed by adapter ligation).

[0157] The adapter may contain other features, such as a primer binding site for PCR amplification; a degenerate unique molecular identifier sequence of at least 5 nucleotides in length; and / or an internal sample identifier (“barcode” sequence) to allow for sample pooling, for example.

[0158] Adapter-ligated analytes may then be amplified, e.g. by reverse transcription and cDNA amplification. Sequencing of the analytes may then be performed.

[0159] In the alternative, other modes of detection of phosphorylation may be used, e.g. via peptide nucleic acid (PNA) hybridization (e.g. a fluorescent sense strand PNA probe binding to RNAi duplex guide strands, Trubetskoy et al., supra) and use of anion exchange high performance liquid chromatography for analysis to separately quantify 5’-phosphorylated and non-phosphorylated barcoded analytes.

[0160] Thus in a preferred aspect in this method in the step of determining the barcode of each analyte that has been phosphorylated to determine which analyte was transported to the target subcellular compartment (e.g. the cytoplasm and / or nucleus), a phosphorylationdependent reaction is performed.

[0161] In a particularly preferred aspect, this embodiment of the invention provides a method of determining whether one or more nucleic acid analyte administered to a cell is transported to the target subcellular compartment (e.g. the cytoplasm and / or nucleus) comprising the steps of: a) administering one or more nucleic acid analyte to said cell, each analyte comprising a nucleotide sequence barcode unique to said analyte; 42.170.173772 / 01 b) collecting said cell or a portion thereof; c) collecting at least the one or more analyte, which has been phosphorylated by interaction with a phosphorylating enzyme, e.g. kinase, in said target subcellular compartment (e.g. the cytoplasm and / or nucleus); d) determining the barcode of each analyte that has been phosphorylated to determine which analyte was transported to the target subcellular compartment (e.g. the cytoplasm and / or nucleus).

[0162] Preferably said analyte comprises a nucleic acid molecule with less than 200, 150 or 100 nucleotides.

[0163] In said method, collecting the analytes which have been phosphorylated comprises collecting at least the analytes that have been phosphorylated as well as, optionally, the nonphosphorylated analytes. In the latter case discrimination between the two groups is performed in step d). If only the phosphorylated analytes are collected in step c), the step may be considered as isolation of the phosphorylated analytes and step d) simply requires determination of the barcodes. Collection may be achieved by binding to a solid support, e.g. on the basis of a universal recognition sequence present in the analyte.

[0164] Discrimination between the phosphorylated and non-phosphorylated analytes is performed by a phosphate-dependent reaction, for example by binding, e.g. covalently, a molecule which recognises a phosphorylated analyte but not a non-phosphorylated analyte. As noted above, this may be performed in step c) or step d). As described herein this may conveniently be achieved by use of an adapter (e.g. which is as described hereinbefore, which is capable of reacting with phosphorylated but not non-phosphorylated analytes, and which preferably has a primer site for PCR amplification) which may be ligated to the phosphorylated analyte, e.g. via T4 RNA ligase 1. Preferred adapters are as set out in Table 4.

[0165] The analytes that have been collected or isolated may be amplified, e.g. using a reverse transcriptase followed by PCR.

[0166] As described above, in some cases, both the phosphorylated analytes and the non- phosphorylated analytes are assessed. This allows a determination of the total cellular uptake (gross uptake) relative to those analytes that are transported to the cytoplasm, for example (cytosolic uptake). In this case, the analytes obtained from the cells may be subject to exogenous phosphorylation so that they are all phosphorylated before assessment. To do so, as described in Example 3, a portion of the sample may be subject to manual or exogenous phosphorylation and then followed by subsequent analysis steps, e.g. adapter ligation steps. This provides a measure of gross uptake and may be compared to the cytosolic uptake results in which no exogenous phosphorylation is performed.

[0167] A preferred example of the steps of the method may therefore be described as: 42.170.173772 / 01 a) administering one or more nucleic acid analyte to said cell, each analyte comprising a nucleotide sequence barcode unique to said analyte and a universal recognition sequence; b) collecting said cell or a portion thereof; c) collecting at least the one or more analyte, which has been phosphorylated by interaction with a phosphorylating enzyme, e.g. a kinase, in said target subcellular compartment (e.g. the cytoplasm and / or nucleus), by capture to a solid support carrying a sequence complementary to the universal recognition sequence; d) contacting said one or more analyte with a molecule which recognises a phosphorylated analyte but not a non-phosphorylated analyte; e) optionally amplifying analytes which covalently attach (i.e. bind) to said molecule which recognises a phosphorylated analyte but not a non-phosphorylated analyte; f) determining the barcode of each analyte that has been phosphorylated to determine which analyte was transported to the target subcellular compartment (e.g. the cytoplasm and / or nucleus).

[0168] As noted above, step c) may include capture of non-phosphorylated analytes which are discriminated from phosphorylated analytes in step d). As such step c) requires the capture of at least the phosphorylated analytes.

[0169] A “molecule which recognises a phosphorylated analyte but not a non- phosphorylated analyte” is a molecule which may bind to or interact with the phosphorylated analyte or is capable of doing so with the assistance of another molecule, e.g. an adapter as described hereinbefore which forms a covalent bond with said phosphorylated analyte by virtue of a ligase.

[0170] Thus in a preferred aspect: in step c) the cell is lysed or permeabilized; in step d) the molecule which recognises a phosphorylated analyte is an adapter which is covalently linked (or bound) to said phosphorylated analyte by virtue of a ligase; in step e) the amplification is performed by reverse transcription followed by amplification (e.g. PCR amplification); and / or in step f) the determination is performed by sequencing.

[0171] A “solid support” as described herein is any suitable support to which capture moieties may be attached, preferably a bead, e.g. as described in the Examples.

[0172] As described hereinbefore endogenously phosphorylated analytes in the sample may be compared to those taken up by the cell by use of an exogenous phosphorylation step. Thus in a further preferred aspect, the method comprises step b) as follows: i) collecting said cell or a portion thereof; 42.170.173772 / 01 ii) collecting the one or more analyte that has been taken up by said cell; iii) dividing said collected one or more analyte into at least two aliquots; iv) contacting at least one aliquot with an exogenous enzyme (e.g. kinase) to allow phosphorylation of said one or more analyte; and step c) collecting from a first aliquot (to which no exogenous enzyme has been added), at least the one or more analyte, which has been phosphorylated by interaction with an endogenous phosphorylating enzyme, e.g. kinase, in said target subcellular compartment (e.g. the cytoplasm and / or nucleus), and separately collecting from a second aliquot (to which exogenous enzyme has been added) at least the one or more analyte, which has been phosphorylated by interaction with an exogenous phosphorylating enzyme, e.g. kinase, or an endogenous phosphorylating enzyme, e.g. kinase (i.e. the second aliquot includes both endogenously and exogenously phosphorylated analytes), by capture to a solid support carrying a sequence complementary to the universal recognition sequence. Each aliquot may then be subjected to steps d) and f) (and optionally e)), above. The results from the different aliquots may be used to determine the proportion (e.g. percent) of analytes that are taken up by the cell that reach the target subcellular compartment (e.g. the cytoplasm and / or nucleus).

[0173] As referred to herein “taken up” refers to uptake of the analytes from the exterior to the interior of the cell. An exogenous phosphorylating enzyme, e.g. kinase, may be any phosphorylating enzyme, e.g. kinase, as described hereinbefore which is added to the aliquot in vitro under conditions suitable to allow phosphorylation of the collected analytes. Interaction with an RNA-binding protein

[0174] In a second example of interaction with an entity in a particular target subcellular location, one or more analyte may interact with an RNA-binding protein. In this case, at least a portion of the analyte is an RNA molecule or mimic to allow binding to the protein.

[0175] A myriad of endogenous (and recombinant) RNA-binding proteins are known in the field. Many of them possess peculiar sequence- and / or structure-dependent substrate recognition mechanisms that guide substrate specificity and selectivity. Two prominent examples include Ago2, a predominantly cytoplasmic RNA endonuclease that is part of the RISC complex and specifically binds to short double-stranded RNA, as well as Cas9, a bacterial and archeal DNA endonuclease that recognises the stem-loop structure of the crRNA-tracrRNA (CRISPR RNA-trans-activating CRISPR RNA) complex (https: / / doi.org / 10.1146 / annurev-biophys-062215-010822 (Jiang & Doudna, Ann. Rev. Biophys., 2017, 46, p505-529), https: / / doi.org / 10.1177 / 2472630320922813 (Jiang et al., SLAS Tech., 2021, 26(1), p92-102)). The high affinity between target RNAs and their cognate RNA-binding proteins can be exploited to detect analytes that reach a subcellular 42.170.173772 / 01 compartment in which therapeutic RNAs are functional. By using analytes that include at least a sequence that mimic the natural substrates of cellular RNA-binding proteins, analytes that are bound to their cognate protein can be enriched using antibodies or column-based purification approaches, serving as a proxy for evidence of functional analyte uptake by the cell and transport to a target subcellular compartment.

[0176] Conjugated siRNAs and ASOs, following in vivo administration and tissue distribution, are naturally trapped in the endosomal compartment limiting their cytosolic exposure and subsequent loading into RISC or general target engagement. Specifically for siRNA, a successful siRNA activity requires the administered siRNA to be internalized by the desired cells, to escape from endosomes if endocytosed, and to be loaded into RSIC and specifically binding directly to Ago2 (Pei et al., RNA, 2010, 16(12), p2553-2563).

[0177] The antisense strand (AS) loaded into RISC is the functional player in target mRNA silencing, through recruiting target mRNA to RISC by sequence-specific complementary pairing and then degradation. The amount of siRNA AS loaded into RISC closely correlates with target mRNA knockdown. Therefore determining whether an RNA molecule has reached the relevant subcellular location and optionally quantifying the amount of AS loaded into RISC, provides a valuable prediction for the therapeutic siRNA activity.

[0178] Thus in methods of the invention, analytes that include a sequence which mimics natural binding partners of RNA-binding proteins can be used and extracted for detection using standard immunoprecipitation methods. In some embodiments, the analytes are designed to mimic binding partners of Ago2. In that case, quantification may be achieved by using immunoprecipitation with an antibody against Ago2 (Gao et al., STAR Protoc., 2022, 3(3), 101596; Xu et al., Nucleic Acid Ther., 2024, 34(4), p199-210, https: / / doi.org / 10.1089 / nat.2023.0067). If activity quantification is additionally required this may be achieved by, for example, stem-loop PCR or northern blotting.

[0179] The analyte may be a double-stranded RNA oligonucleotide that is between 14 to 29 nucleotides in length, preferably between 17-23 nucleotides in length. Both strands can include one or more modified nucleotides and / or one or more non-phosphodiester linkages. Preferably the analyte should comprise a 5’ phosphate or metabolically stable phosphate analogue and the first nucleobase should preferably be uracil or adenine (https: / / www.ncbi.nlm.nih.gov / pmc / articles / PMC7812868 / , Shinohara et al., RNA, 2021, 27(2), p163-173). The 5’ end of the analyte should further have a higher thermodynamic stability than the 5’ end of the complementary strand, for example by having a higher A / ll content. In some embodiments, the duplex may have a 3’Ull or 3’IIG overhang (https: / / www.sciencedirect.com / science / article / abs / pii / S0968089624002396, Li et al., Biorg. Med. Chem., 2024, 110, p117825). 42.170.173772 / 01

[0180] Nuclear localization

[0181] In an alternative embodiment of the invention, the method may be used to determine if the analytes have entered the nucleus. Since transmission to the nucleus requires passage through the cytoplasm, this is also indicative of transport to the cytoplasm. Figure 2 Illustrates the use of nuclear accumulation as a proxy for productive analyte uptake. This method has the advantage that sequencing techniques based on sequencing nucleic acid material in the nucleus can be used.

[0182] The isolation of certain cell types characterized by intrinsic fragility, large size and tight interconnections, brings difficulties in traditional isolation methodologies, hence making their transcriptome profiling difficult and less reliable. Single-nucleus RNA sequencing (snRNA-seq) is used as an alternative or complementary approach for cells that are difficult to isolate (Kim et al., J. Pathol. Transl. Med., 2023, 57(1), 52). snRNA-seq has been applied to several tissues and cell types (e.g. kidney, heart, liver, adipocytes, and myofibers) and for the majority of the tested tissues, snRNA-seq is more powerful at recovering sensitive cell types, reducing the bias towards immune cell types typical in scSEQ.

[0183] Chemical ligation-quantitative polymerase chain reaction (CL-qPCR) is a technique suitable for the detection of chemically modified oligonucleotides and oligonucleotide conjugates to quantify the oligonucleotides in cells and subcellular compartments following free uptake (Boos et al., Nucleic Acids Res., 2013, 41(15), e145). CL-qPCR has been designed to be compatible with highly modified oligonucleotides of short length. The technique requires the synthesis of probes for each single sequence reducing the high throughput capabilities. In addition the detection limits are higher compared to snRNA-seq readout.

[0184] Short single-stranded ASOs are known to passively diffuse through the nuclear core complexes and bind to nuclear components, leading to nuclear accumulation within minutes after microinjection (https: / / doi.org / 10.1093%2Fnar%2Fgky1158, Buntz et al., Nucl. Acids Res., 2019, 47(2), p953-969). Passive diffusion of ASOs can occur either freely or bound to proteins, and is an energy-independent process. In some cases, single-stranded phosphorothioate ASOs are actively trafficked into the nucleus or exported from the nucleus through an energy-dependent, nuclear pore complex-mediated transport mechanism (https: / / doi.Org / 10.1093%2Fnar%2F28.2.582, Lorenz et al., Nucl. Acids Res., 2000, 28(2), p582-592). It has further been shown that short double-stranded oligonucleotides, such as siRNAs, are rapidly imported into the nucleus upon microinjection, transfection or lipofection (https: / / doi.org / 10.1093 / nar / gkm694, Jarve et al., Nucl. Acids Res., 2007, 35(18), e124). SiRNA-like variants containing longer 5' overhangs (5nt-6nt long) tend to accumulate in the nucleus, like ssRNA. Oligonucleotides with 9nt 5' overhangs (which can be extended by 42.170.173772 / 01 design) behave similarly to single-stranded RNAs and hence accumulate in the nuclei (Hirano et al., RSC Adv., 2022, 12(38), p24471-24477). A single-stranded oligonucleotide adopts a flexible and compactly folded random-coil structure. As a result, a single-stranded oligonucleotide is expected to pass through nuclear pores from the cytoplasm.

[0185] The method of the invention may be used to determine if specific analytes (acting as mimics of test oligonucleotide therapeutics) enter the cytoplasm and then the nucleus. This acts as a proxy for determining the take-up and processing of such analytes. This may include the use of nuclear localization techniques if the test oligonucleotide therapeutics might take advantage of those techniques and the methods are used to test their efficacy. However, in some circumstances it may be desirable to transport analytes to the nucleus to allow for sequencing. This may be achieved by using nuclear localization techniques. In this case, the output from the method confirms that the analytes were released into the cytoplasm, as this is a necessary first step before nuclear uptake.

[0186] Thus, analytes of the invention may carry a nuclear targeting moiety. As noted above this may form part of the analyte as indicative of the test oligonucleotide therapeutic or may be attached to the analyte to allow its nuclear uptake. As referred to herein a “nuclear targeting moiety” improves the nuclear localization of the analyte of which it is part or to which it is attached. This may include moieties that prevent transport out of the nucleus such as that facilitated by exportin-5.

[0187] The nuclear targeting moiety may be a nucleotide sequence. Nuclear localization sequences that may be used include an RNA motif consisting of the pentamer sequence AGCCC with sequence restrictions at positions -8 (T or A) and -3 (G or C) relative to the first nucleotide of the pentamer (Zhang et al., Mol. Cell. Biol., 2014, 34(12), p2318-2329). The presence of this sequence in the analyte effectively programs the analyte to preferentially accumulate in the nuclei. Alternative sequences which favour uptake to the nucleus and avoid transport out of the nucleus are known in the art.

[0188] In an alternative embodiment, the nuclear targeting moiety is a 2,2,7- trimethylguanosine cap (m3G-CAP) which may be provided at the 5’ end of the nucleic acid analyte and provides a nuclear import mechanism with the m3G signal as an adapter which confers nuclear targeting capabilities (Moreno et al., Nucleic Acids Res., 2009, 37(6), p1925- 1935).

[0189] In a further alternative, a nuclear localization peptide may be used, preferably conjugated to the analyte. Such peptides are well known in the art and have been conjugated to oligonucleotides (especially splice-switching oligonucleotides) to improve their nuclear localization and as a consequence, their biological activity (Hill et al., ACS Omega, 2023, 8(43), p40463-40481). Such peptides may be readily attached to the analyte without affecting processing, including sequencing. Preferably the peptide is attached covalently and 42.170.173772 / 01 the linkage is not cleavable. In a preferred aspect the peptide is attached at the 3’ end of the analyte.

[0190] For performance of the method in this embodiment, as noted above the analyte may include modifications to aid nuclear transfer and / or may carry a nuclear targeting moiety. The analyte may be as described above, and conveniently has at least two or more of the following features: a) single or double stranded, preferably ssRNA or dsRNA; b) has a length of 8-80, preferably 14-40 nucleotides; c) has a barcode of 3-40 nucleotides: d) may include a hairpin portion; e) in double stranded molecules has a 5’ overhang of at least 3 nucleotides and / or a stem length of less than 14 nucleotides to avoid recognition by exportin-5; f) an RNA motif for nuclear localization, preferably AGCCC, RCCTCC (where R is A or G) or CTG repeats; g) a sequence that recruits a protein that undergoes nuclear import or retention, e.g. Cas9; h) a 5’-m3G cap; i) carries a nuclear targeting moiety, preferably a nuclear localization peptide or sequence; j) nucleotide modifications to mimic therapeutic or diagnostic ssRNA, e.g. 2’F, 2’OMe, 2’MOE, LNA, PS (providing they do not affect downstream processing), preferably all internucleotide linkages (or at least 14 consecutive linkages) may be phosphorothioate linkages. Mimics are as described hereinbefore and in the present context the mimic is a molecule carrying said nucleotide modification which mimics one or more properties of the therapeutic or diagnostic ssRNA, particularly the functional properties. The mimic does not affect downstream processing, i.e. it is a functionally similar molecule insofar as it is processed in the same or a similar manner, e.g. without variation beyond that which may be expected between different experiments. Appropriate testing in the test system of interest (which may be in vivo or in vitro) may be performed to compare processing with the molecule that is the subject of the mimic.

[0191] The two or more features (e.g. 3, 4, 5, 6, 7, 8, 9 or 10) are preferably selected from a) to c) , e) and j), and optionally may also include one or more of d), or f) to i).

[0192] As noted above, Figure 2 provides an outline method for performance of this embodiment of the invention. After administration of the analytes to the cell and maintaining the cells to allow them to reach the target subcellular location (in this case the nucleus), the nuclei are isolated, i.e. removed from the other material in the cell, particularly endosomal portions of the cell. 42.170.173772 / 01

[0193] Appropriate techniques for isolation of nuclei are well known in the art.

[0194] In particular, competent buffers with nonionic detergents that disrupt cell membranes, but preserve nuclear membranes, may be used for nuclear isolation (Kim et al., J. Pathol. Transl. Med., 2023, 57(1), p52). Mechanical forces to lyse the cell may be used, e.g. a Dounce homogenizer or other types of tissue lysers. Appropriate RNase inhibitors may be used, during and after the isolation process, to avoid analyte degradation. After isolation, microscopy may be used to confirm the presence of intact nuclei.

[0195] The nuclei may be used directly in sequencing methods or may be further processed to remove or enrich the analytes. The barcodes of the analytes may then be determined, e.g. sequenced, as appropriate. This may include amplification technologies to improve sensitivity. Depending on the upstream processing steps, appropriate sequencing methods that may be used include known bulk RNA sequencing and single-nucleus RNA sequencing methods.

[0196] Further aspects of the invention

[0197] In an alternative embodiment the invention provides one or more analyte as defined herein for use as a diagnostic to determine whether one or more analyte administered to a subject is transported to a target subcellular location within said cell. The terms have the meanings and preferred features as described herein. As referred to herein a diagnostic refers to deriving information on the analyte’s behaviour in the subject to which it is administered to allow the derivation of diagnostic information.

[0198] In a yet further aspect, the invention provides a method of determining whether one or more nucleic acid analyte comprising a nucleotide sequence barcode unique to said nucleic acid analyte is, or was, present in a target subcellular location in a cell, comprising a) obtaining a sample containing said cell to which said one or more analyte has been administered; b) performing an assessment as set out in step b) as described hereinbefore.

[0199] The sample may be obtained from a subject or from the cell-containing sample to which the analytes have been administered. Suitable samples are as described hereinbefore. The terms have the meanings and preferred features as described herein.

[0200] In a still further aspect, the invention provides a method of determining whether one or more nucleic acid analyte has entered a cell and been released from the endosome (and / or lysosome) by transporting said one or more analyte to the nucleus for detection, wherein said analyte is as defined herein, comprising the steps of: a) administering said one or more analyte to said cell (which may be in vitro, ex vivo or in vivo) or obtaining a sample containing said cell to which said one or more analyte has been administered (preferably in vivo, e.g. administered to a subject and a sample is collected); 42.170.173772 / 01 b) collecting said cell or a portion thereof, isolating said nucleus and assessing whether one or more barcode is present in said nucleus and thereby determining which of the one or more analyte entered said cell and was released from the endosome (and / or lysosome). The terms have the meanings and preferred features as described herein.

[0201] In a yet further aspect, the invention provides a method for detecting the presence of one or more nucleic acid analyte in a target subcellular location after administration to a cell, comprising the steps of: a) administering one or more nucleic acid analyte to said cell, each analyte comprising a nucleotide sequence barcode unique to said analyte as defined hereinbefore; b) collecting said cell or a portion thereof and assessing whether one or more barcode of the administered one or more analyte is, or was, present in the target subcellular location, thereby determining which analyte is, or was, present in the target subcellular location, wherein said assessment is made by: i) collection of said cell or a portion thereof comprising said target subcellular location and identifying if one or more of said barcode is present in said location, or ii) by determining whether one or more of said analyte has interacted with an entity in said subcellular location. The terms have the meanings and preferred features as described herein.

[0202] Multiplexing

[0203] The methods and uses of the invention are primarily useful for the performance of multiplex screening assays. To achieve this, more than one nucleic acid analyte is used. Each analyte comprises at least one barcode unique to the analyte (relative to other analytes to be used in the method or which are in the pool) to allow its detection. This effectively provides a library or plurality of barcoded nucleic analytes which may be used for detection.

[0204] Each of the barcodes in the different analytes differ by at least one nucleotide as described hereinbefore. The analytes may have features in common, e.g. cargo molecules or universal recognition moieties, but differ from one another insofar as they have different barcodes. They may include one or more additional modifications for which the barcode is a marker (e.g. a specific test sequence which is not the barcode). However, in some cases, if the barcode itself effectively provides the mimic for testing (i.e. is a dual purpose barcode), this may be the sole difference between the analytes.

[0205] The barcode serves as a label for the identity of every analyte such that it can be used to distinguish between hundreds of different analytes in a pool of analytes, enabling detection and analysis of a plurality of analytes in a single assay (multiplexing). The sequence of each barcode is designed to be sufficiently unique to allow its unambiguous differentiation from all other barcodes in the pool. 42.170.173772 / 01

[0206] As used herein, the term "multiplex" or "multiplexing" refers to the parallel analysis of pooled analytes in methods and uses of the invention.

[0207] In particularly preferred embodiments, the nucleic acid barcode is designed such that no two barcoded analytes, in a pool / set / library or used in a method, share the same barcode sequence. Therefore, any two analytes out of a given set will be distinguishable by their unique barcodes and their barcodes will differ in at least one nucleotide position. In a preferred embodiment of the disclosure, each barcode will differ in at least 3 nucleotide positions from all other barcodes in the set.

[0208] In some embodiments, the incorporation of a nucleic acid barcode unique to each analyte, or a composition containing it, allows for the identification of the identity or source of pooled analytes or samples containing them in multiplexed assays. The number of analytes processed in parallel determines the minimum number of barcodes to be included in a given pool of analytes / samples. For example, if in a given case the pool of samples or analytes to be processed simultaneously comprises 2, 5, 12, 48, 96, or 384 samples, a set of at least 2 or more, 5 or more, 12 or more, 48 or more, 96 or more, or 384 or more, respectively, distinct barcodes will be required for unambiguous sample / analyte identification.

[0209] It will be readily apparent to those skilled in the art that the length of the barcode may limit the number of samples that can be processed in parallel. For a given barcode length L, the maximum possible number of unique barcode sequences is 4AL. If for instance, a barcode length of 3 nucleotides is selected, the number of barcodes that can be multiplexed is no more than 64. In practice, the maximum number of unique barcode sequences appropriate for a desired application may be lower due to experimental constraints. Considerations for the rational design of a set of barcode sequences suitable for deep sequencing applications are set forth in WO2024 / 069235.

[0210] In some embodiments, the nucleic acid barcode is used as an identifier of the chemical composition of an analyte. It may correlate with one or multiple chemical characteristics of the analyte including, but not limited to, a particular type of modification, a particular pattern (or arrangement) of multiple modifications, a particular number of modifications present, a particular physicochemical property or a particular set of physicochemical properties, a particular conjugation chemistry used for cargo attachment, a particular type of chemical linker introduced during intermolecular conjugation, a particular coating molecule or mix of coating molecules, a particular method used for vehicle assembly, or combinations thereof. It may further correlate with non-chemical properties such as shape (for example, but not limited to, helix, circle, triangle, square, pentagon, hexagon, heptagon, octagon, nonagon, decagon, cube, cylinder, sphere, pyramid, prism, tetrahedron, or octahedron), dimension (for example, but not limited to, diameter, length, width, height, depth, radius, circumference, area, volume, or combinations thereof), nucleic 42.170.173772 / 01 acid structure (for example, but not limited to, duplex, triplex, quadruplex, hairpin, loop, bulge, pseudoknot, 2-way junction, 3-way junction, 4-way junction, 5-way junction, 6-way junction, or combinations thereof), particular cargo load (for example, but not limited to, oligonucleotides, including NATs, peptides, proteins, including antibodies, lipids, carbohydrates, polymers, fluorophores, and combinations thereof) or number of covalently or non-covalently attached cargoes, and the number of molecular constituents of the analyte, or combinations thereof. Alternatively, or in combination with the above, a barcode may uniquely identify a particular production batch of delivery vehicles or a particular pretreatment method (for example, but not limited to, incubation with a particular protein or mix of proteins, incubation in a particular buffer, medium or biological matrix, a particular storage condition, a particular time point of a stability experiment, or combinations thereof). Normalization / validation methods

[0211] Barcode detection, whether by next-generation sequencing (NGS) or another method, can be affected by a range of technical and systemic biases. For example, RNA yields can differ across experimental conditions. Enzymatic processing steps such as reverse transcription, adapter ligation and PGR amplification can be more efficient for some sequences than others, and the enzymes used may be sensitive to slight differences in RNA secondary structure or the presence of modified nucleotides. Moreover, standard normalization methods in NGS rely on the assumption that most barcodes are not differentially abundant across samples, which does not hold true in the use cases described herein.

[0212] To correct for potential biases and enable quantitative measurement of barcode concentrations, the present invention provides a method for determining the absolute number of barcoded molecules wherein a set of synthetic “spike-ins” is used as an internal standard. The method comprises the following steps: a) manufacturing a set of at least 3, e.g. at least 9 synthetic internal standard analytes, wherein the sequence of each internal standard analytes contains a unique barcode and wherein the overall design of each internal standard analyte matches the design of the barcodes and analytes used in a given experiment in their functionalized form (e.g. if the barcode design is a duplex with a 9-nt 5’ overhang and full PS backbone in the nucleic acid pipeline, the internal standard analyte should also be a duplex with a 9-nt 5’ overhang and full PS backbone; OR if the barcode is a ssRNA strand with a 5’OH group used in a 5’ phosphorylation experiment, the spike-in / internal standard analyte should be a ssRNA strand with a 5’ phosphate group to reflect the functionalized form of the barcode); b) mixing the set of synthetic internal standard analytes in known ratios so that the concentration of each internal standard analyte in the mix is known and there are at least 3 different concentrations present; 42.170.173772 / 01 c) adding a fixed volume of the internal standard analyte mixture to each sample (after cell or nuclei lysis); d) performing barcode detection, and e) determining the absolute amount of barcode molecules using the internal standard analytes as reference values. In some embodiments step a) may be excluded if using pre-made standards.

[0213] In some embodiments, a control sample comprising the internal standard analyte library in water (or cell lysate or tissue lysate) is included to calculate barcode-specific detection efficacy scores.

[0214] The internal standards may be used for validation or normalization of the methods of the invention. Thus, in methods described hereinbefore, the methods may additionally comprise the inclusion of one or more (preferably at least 3, 4, 5, 6, 10, 15, e.g. up to 10 or up to 20 or up to 30) synthetic internal standard analyte as a control in the step(s) in which the one or more barcode of the analyte is determined. The control is included for normalization and / or for absolute quantification of test analytes Products for use in methods of the invention

[0215] The invention also extends to products for use in the invention. Such products have the features as described hereinbefore and all such described products are embodiments of the invention. Accordingly the present invention provides a plurality of nucleic acid analytes, wherein each analyte: a) is a nucleic acid molecule with less than 500 (preferably less than 300, 200, 150 or 100) nucleotides; b) comprises a different barcode unique to said analyte, wherein said barcode sequence is from 4 to 20 nucleotides in length; c) comprises an invariant universal recognition sequence; and optionally one or more of: i) a 5’OH or 3’OH group amenable to phosphorylation; ii) the first two 5’ nucleotides of the analyte adopt an A-form helical conformation with the ribose in the C3’ endo orientation; iii) a synthetic 5’-m3G cap; and iv) a nuclear targeting moiety, preferably a nuclear localization peptide or a sequence selected from AGCCC, RCCTCCC or CTG repeats. Preferably the analytes have two or more of features i), ii) and iii). Such analytes may be unphosphorylated (e.g. having a group amendable to phosphorylation as described above), or may be phosphorylated, e.g. at the 5’ end. Exemplary phosphorylated analytes are set forth in Table 9 (SEQ ID NOs: 263- 265) which may be provided in single stranded or double stranded form (Table 10). 42.170.173772 / 01

[0216] In a preferred aspect the plurality of analytes contains one or more synthetic internal standard analytes as controls, wherein optionally the standard analytes are provided separately.

[0217] In a further preferred aspect the plurality of analytes includes at least one nucleic acid analyte as set forth in Table 1 or any other analyte as described herein, particularly in the Examples.

[0218] In a preferred aspect, each analyte additionally comprises one or more of: i) a moiety aiding interaction with an entity present in a subcellular location; and / or ii) a targeting moiety aiding targeting to a particular location in a cell.

[0219] The analytes and their preferred features are as described hereinbefore.

[0220] As used herein, unless specifically identified in a structure as having a particular conformation, for each structure in which asymmetric centers are present and thus give rise to enantiomers, diastereomers, or other stereoisomeric configurations, each structure disclosed herein is intended to represent all such possible isomers, including their optically pure and racemic forms. For example, the structures disclosed herein are intended to cover mixtures of diastereomers as well as single stereoisomers.

[0221] The person of ordinary skill in the art would readily understand and appreciate that the compounds and compositions disclosed herein may have certain atoms (e.g., N, O, or S atoms) in a protonated or deprotonated state, depending upon the environment in which the compound or composition is placed. Accordingly, as used herein, the structures disclosed herein envisage that certain functional groups, such as, for example, OH, SH, or NH, may be protonated or deprotonated. The disclosure herein is intended to cover the disclosed compounds and compositions regardless of their state of protonation based on the environment (such as pH), as would be readily understood by the person of ordinary skill in the art.

[0222] The invention also extends to any analyte, or analyte comprising or associated with a cargo molecule as described herein, particularly as set forth in the Examples. In particular the analyte may have a sequence as set forth in any of SEQ ID NOs: 1 to 224 (Table 1) or 266 or 267 (Table 13). For the generation of double stranded analytes, sequences such as provided in Table 3 (SEQ ID NOs: 237 to 242), Table 13A (SEQ ID NOs: 266 and 267), Table 15A (SEQ ID NO: 272) or Table 16A (SEQ ID NO: 273-275) may be used to generate duplexes. Preferred double stranded analytes are set forth in Tables 5 , 8A, 13B, and 16B, which may be conjugated to cargo molecules as described herein, particularly as described in Table 8B and Table 17. Preferred molecules are as described in the Examples.

[0223] The methods described in the Examples form further preferred aspects of the invention. All combinations of the preferred features described above are contemplated, 42.170.173772 / 01 particularly as described in the Examples. The invention will now be described in more detail in the following non-limiting Examples with reference to the following drawings in which:

[0224] Figure 1 is a schematic illustration outlining the use of endogenous 5’ phosphorylation as a proxy for productive analyte (referred to in the figure as the barcoded oligonucleotide) uptake. (1) Cells internalize double-stranded analytes via endocytic pathways, leading to analyte accumulation in endosomal or lysosomal compartments. (2) A fraction of the analytes is able to escape the endolysosomal entrapment and reaches the cytoplasm. Endogenous polynucleotide kinases localized in the cytoplasm then phosphorylate a subset of the analytes that escaped. (3) After cell lysis, the monophosphorylated analytes (as well as the remaining non-phosphorylated analytes, not shown) are enriched by hybridization using magnetic beads functionalized with capture oligos and subjected to library preparation and next-generation sequencing. A phosphorylation-dependent step during library preparation (e.g. ligation, not shown) ensures selective detection of the monophosphorylated analytes.

[0225] Figure 2 is a schematic illustration outlining the use of nuclear accumulation as a proxy for productive analyte (referred to in the figure as the barcoded oligonucleotide) uptake. (1) Cells internalize single-stranded analytes via endocytic pathways, leading to analyte accumulation in endosomal or lysosomal compartments. (2) A fraction of the analytes is able to escape the endolysosomal entrapment, reaches the cytoplasm and enters the nucleus for example via passive diffusion through the nuclear pore complex. (3) The nucleus is isolated using a standard single-nuclei isolation protocol, and the remaining analytes are then extracted and subjected to library preparation and next-generation sequencing. Nuclei isolation ensures selective detection of analytes that manage to traffic and accumulate in the nucleus, whilst analytes in the cytoplasmic and endolysosomal compartments are efficiently removed.

[0226] Figure 3 is a schematic illustration outlining the use of Ago2 immunoprecipitation for the enrichment of analytes (referred to in the figure as the barcoded oligonucleotide) that have escaped the endosome. (1) Cells internalize double-stranded analytes via endocytic pathways, leading to analyte accumulation in endosomal or lysosomal compartments. (2) A fraction of the analytes is able to escape the endolysosomal entrapment and reaches the cytoplasm, where the escaped analytes interact with proteins such as Ago2 and associated proteins (RISC complex). (3) The Ago2-bound analytes are selectively enriched by standard immunoprecipitation or column purification protocols, and the isolated analytes are then extracted and subjected to library preparation and next-generation sequencing. Only 42.170.173772 / 01 analytes that reach the cellular compartment(s) in which the target protein is expressed are enriched, resulting in the selective detection of analytes with productive uptake.

[0227] Figure 4 shows exemplary analytes that may be used in methods of the invention. In particular, analytes that mimic test substances are shown. The placement of the barcode sequences within the analyte sequence is shown (squares) as well as the presence of modified nucleotides. Atypical modifications that deviate from the standard modification of the test substance can be introduced in some cases to suppress undesired siRNA-like activity of the analytes.

[0228] Figure 5 outlines an exemplary library preparation workflow for bulk RNA sequencing of short analytes, containing barcodes positioned at or near the 5’ terminus, for the selective amplification of endogenously 5’ phosphorylated molecules. Adapter ligation directly on the RNA strand allows sample multiplexing already at the cDNA level, reducing sample numbers for further processing. CS = complementary strand; PBS = primer binding sequence; P5 / P7 - primer sequences.

[0229] Figure 6 outlines an alternative (to Figure 5) in which the analyte is modified by phosphorylation at the 5’ end when the barcode appears at or near the 3’ end. In this workflow both 3’ and 5’ adapters are used and reverse transcription is performed in solution (as opposed to solid phase).

[0230] Figure 7 is a schematic illustration depicting optional conjugation sites for the covalent or non-covalent attachment of additional moieties (also referred to as cargo) to single-stranded (top) or double-stranded (bottom) barcoded analytes. In some embodiments, only a single conjugation site is used.

[0231] Figure 8 is a schematic illustration depicting: A) A barcode / analyte-lipid conjugate, consisting of a barcode / analyte region, a linker region (attached via click chemistry or other chemical linkers known in the art) and a lipid. Examples of suitable lipid are shown. B) A barcode / analyte conjugated to a highly modified oligonucleotide, wherein the oligonucleotide is modified either at the 2’ position, the phosphate backbone or the nucleotide base.

[0232] Figure 9 shows qPCR amplification curves for libraries prepared from a double-stranded analyte (DX0001), generated in the presence (PNK+) or absence (PNK-) of PNK treatment. Samples lacking the analyte (no template control, NTC) or the 5’ adapter (Adapter-) were included as controls. 42.170.173772 / 01

[0233] Figure 10 shows tapestation traces of libraries generated in the presence or absence of PNK treatment, as indicated. Samples were subjected to 10 cycles cDNA pre-amplification and 12-15 cycles of indexing PCR. Lanes to the left (B1 to E1) show cDNA derived from rat brain tissue samples upon intrathecal barcode / analyte administration, middle lanes show cDNA derived from double-stranded (F1 to G1) and single-stranded (H1 to A2) barcodes / analytes spiked into tissue lysates from untreated rats, and lanes B2 and C2 show control libraries produced from single-stranded barcodes / analytes spiked into untreated tissue lysates in the absence of adapters. Products include the expected main product of -183 bp and a side product of -175 bp. The predominant product is shown with an arrow, i.e. the correct product in lanes B1 , D1 , F1 and H1 and the side product in C1, E1 , G1, A2 and B2.

[0234] Figure 11 shows the tapestation traces (A) and electropherogram (B) of a seguencing-ready, pooled library prior to seguencing. The pool comprised 28 individual libraries prepared using the workflow shown in Figure 5. The pooled library was of high-guality and matched the expected peak size (amplicon size: 193 bp).

[0235] Figure 12 shows a bar plot of RNAseg read counts of barcodes that were transfected in A549 cells either with 5’OH groups (in grey) or pre-phosphorylated 5’P groups (in black). As a control, a single-stranded analyte was spiked into cell lysate in a proteinase K-containing buffer (to block residual kinase activity). The numbers shown at the top of the plot refer to the SEQ. IDs of the parent 5’OH analyte (regardless of the 5’ modification, see Tables 9 and 10 for more details on the analytes used). Counts were normalised by 20 repeats of rarefaction to account for differences in seguencing depth across libraries and enable crosssample comparisons. Abbreviations ds and ss indicate double-stranded and single-stranded analytes, respectively.

[0236] Figure 13 shows in vitro validation results of the bulk analyte seguencing approach. Internal standards were used to assess assay performance. (A) Accuracy and lower limit of detection determination by linear regression analysis of 15 serially diluted internal standards. Nine technical replicates were measured. Dots represent individual data points. (B) Observed vs expected coefficient of variation based on Poisson sampling (dotted line).

[0237] Figure 14 shows regression and cross-validation curves for the absolute guantification of analytes. A set of 28 serially diluted internal standards was used to guantify the abundance of 15 cargo-conjugated analytes spiked into rat tissue samples at known concentrations. (A) Linear regression plot of the serially diluted internal standards. The regression eguation and 42.170.173772 / 01

[0238] R2 values are shown for each of 3 independent replicates (n = 3). (B) Predicted vs actual plot following absolute quantification of cargo-conjugated analytes by normalization against internal standards. Error bars represent the standard deviations of indexing PCR duplicates. MFE indicates the mean fold error across 3 independent replicates (n = 3).

[0239] Figure 15 shows in vitro validation of analyte detection in isolated nuclei. (A) Bar plot of fluorescence intensities measured across isolated nuclei and non-nuclear fractions (designated as “rest”). Fluorescence values were obtained by fluorescence hybridization assay, with the signal correlating to siRNA abundance in the respective samples. Samples lacking the siRNA (no template control, NTC) or comprising plain buffer (blank) were included as controls. AU, arbitrary units. (B) Tapestation traces of sequencing libraries generated from non-nuclear supernatant (middle lane) and isolated nuclei (right lane), obtained from A549 cells transfected with 10 nM of single-stranded analyte (SEQ. ID 14).

[0240] Figure 16 compares the dose-response relationship of siRNA-mediated gene knockdown and intracellular 5’ phosphorylation. (A) Sigmoidal dose-response curve illustrating siRNA- mediated knockdown of PPIB mRNA as a function of siRNA concentration. Each data point represents the mean of three independent replicate measurements (n = 3), with error bars indicating the standard deviation. Data were fitted using a non-linear regression model, with the half-maximal inhibition value (IC50) and the coefficient of determination (R2) of the fit indicated in the figure. (B) Non-saturating dose-response curve illustrating cytosolic uptake of a barcoded siRNA mimic as a function of transfection concentration. Individual data points correspond to duplicate measurements from three independent biological replicates (n = 3).

[0241] Figure 17 compares the dose-response relationship of total cellular analyte internalization (gross uptake) against the dose-response relationship of intracellular 5’ phosphorylation (functional uptake). For each concentration, three data points are shown, each representing the mean of two technical replicates derived from the same biological sample. Error bars indicate the standard deviation across the three independent biological replicates (n = 3). Data were fitted using a linear regression model, with shaded ribbons representing the 95% confidence intervals of the fit. The coefficient of determination (R2) for each regression is indicated in the figure.

[0242] Figure 18 shows the effect of hydroxychloroquine treatment on siRNA-mediated gene knockdown, intracellular 5’ phosphorylation levels (functional uptake) and total cellular internalization (gross uptake) in A549 cells. Measurements were performed 24 hours after gymnotic administration of cholesterol-conjugated siRNA or analytes (1 pM) in the presence 42.170.173772 / 01

[0243] (CQ+) or absence (CQ-) of hydroxychloroquine co-treatment (100 pM). (A) Bar plot of PPIB gene knockdown mediated by cholesterol-conjugated siRNA (DX0079). llntr indicates cells that did not receive siRNA. Relative gene expression levels were determined by quantitative PCR and normalized to RPLPO as the reference gene. Individual data points from three biological replicates (n = 3) are shown, with error bars representing the standard deviation. Statistical significance is indicated in the figure, with ** denoting p < 0.01 and *** denoting p < 0.001. (B) Bar plot of functional and total cellular uptake of single-stranded, barcoded analyte (SEQ. ID 14). Absolute uptake levels were calculated from targeted RNA sequencing data via normalization to internal standards. Individual data points represent technical duplicates derived from a single sample (n = 1) obtained by pooling material from three biological replicates.

[0244] Figure 19 shows the kinetics of analyte internalization and intracellular phosphorylation upon endosomal escape in A549 cells. (Top) Absolute uptake over time for up to 24 hours after analyte transfection (10 nM), as obtained by bulk RNA sequencing via normalization to internal standards. Lines represent posterior mean values and shaded ribbons represent 95% credibility intervals obtained from Bayesian inference. (Bottom) Percent endosomal escape calculated as the ratio between functional and gross uptake. The line indicates mean percentages. Error bars represent 95% confidence intervals estimated by BCa bootstrapping. Individual data points correspond to duplicate measurements from three independent biological replicates (n = 3, each indicated by a different shape).

[0245] Figure 20 shows a dual-axis bar plot comparing target gene knockdown by Sodl-targeting siRNA (left y-axis) with cytosolic concentration of a barcoded siRNA-mimicking analyte (right y-axis) across rat lumbar spinal cord (LSC) and brainstem (BS) tissues. Individual data points correspond to six independent biological replicates derived from different animals (n = 6). Error bars represent the mean ± standard deviation for gene knockdown and the geometric mean x geometric standard deviation for cytosolic uptake.

[0246] Figure 21 shows the results of a multiplexed in vivo screen for functional delivery of barcoded, cargo-conjugated analytes in rat brainstem tissues. Concentrations of escaped barcodes were obtained by sequencing of endogenously 5’ phosphorylated analytes, and read counts were converted to absolute quantities by normalization to internal standards. Depicted are geometric means, with error bars representing geometric standard deviation. Individual data points correspond to five independent biological replicates derived from different animals (n = 5). 42.170.173772 / 01

[0247] Figure 22 shows the tapestation electropherogram (A) of a sequencing-ready, pooled library prior to sequencing, and a bar plot (B) depicting cytosolic concentrations of escaped barcodes following intratracheal installation of dual-cargo conjugates in mice. Facet labels 1x and 2x refer to the design of the cargo conjugates, with 1x indicating the presence of a single, barcoded analyte strand whilst 2x indicates two copies thereof. Bar heights represent geometric means, with error bars show the geometric standard deviation. Individual data points correspond to five independent biological replicates derived from different animals (n = 5).

[0248] EXAMPLES

[0249] Example 1 : Oligonucleotide synthesis and duplex assembly (Methodology for later Examples)

[0250] A) Oligonucleotide synthesis

[0251] DNA oligonucleotide strands were obtained from Integrated DNA Technologies UK Ltd (IDT). RNA oligonucleotide strands were either purchased from WuXi AppTec (HongKong) or synthesized at 1-10 pmol scale using a K&A synthesizer (H-16). All protocols were modified depending on the sequence requirements. Phosphoramidites and CPGs with standard protecting groups were purchased from ChemGenes and Glen Research. The detritylation step was carried out with 3% TCA in DCM, followed by coupling with 0.1 M phosphoramidite solutions and 0.25 M BMT in MeCN. Capping was performed using THF / lutidine / acetic anhydride (80 / 10 / 10) as capping A and 16% N-methylimidazole in THF as capping B, respectively. The oxidation step was accomplished with 0.02 M iodine solution in THF / Pyr / water (90.6 / 0.4 / 9).

[0252] All synthesized oligonucleotides were cleaved and deprotected using aq. methylamine / ammonium hydroxide solution (1 :1) for 3 h at RT for a solid support with a first base attached or for 1 h at 65 °C for a universal CPG. The removal of tert-butyl silyl protecting groups was performed by incubating an intermediate product in DMSO Et3N.3HF for 3 h at 65 °C. Crude oligonucleotides were subsequently precipitated from ethanolic solution containing sodium acetate. After 2 h at -70°C the precipitate was harvested by 25 min centrifugation at 4°C (14,000 rpm). The supernatant was separated, and the remaining pellet was washed repeatedly with 70% EtOH. After a final wash, the crude sample was dried under vacuum in a speedvac and redissolved in water for purification.

[0253] Crude RNA strands were purified either by IEX-HPLC or by IP-RP HPLC. 42.170.173772 / 01

[0254] I EX was carried out with a preparative DNAPac PA200 (ThermoFisher), 22 x 250 mm column, or PL-SAX (Agilent) 22x150 mm 1000 A column at 75 °C with a flow rate of 15 mL / min and UV detection at 260 nm. Elution was performed with a linear gradient selected based on crude impurity profile, determined by analytical testing using either a DNAPac PA200RS LIPLC column or PL-SAX analytical column. Buffer A: 25 mM Tris HCI, pH 8.0, 20% acetonitrile, 10 mM sodium perchlorate; buffer B: 25 mM Tris HCI, pH 8.0, 20% acetonitrile, 600 mM sodium perchlorate, OR, Buffer A: 25 mM Tris HCI, pH 8.0, 20% acetonitrile, 25 mM sodium chloride; buffer B: 25 mM Tris HCI, pH 8.0, 20% acetonitrile, 1.5 M sodium chloride.

[0255] RP-HPLC was carried out with a BEH C18 300A (Waters) 19 x 150 mm at 60 °C, with a flow rate of 25 mL / min and UV detection at 260 nm. Buffer A: TEAA (0.1 M, pH = 7); buffer B: MeCN, OR, Buffer A: HAA (0.1 M, pH 7); buffer B: MeCN.

[0256] Fractions containing RNA were assessed for purity by analytical PAGE, IEX and RP-HPLC, then pooled and subjected to final QC on PAGE, IEX and RP-HPLC, acetonitrile removed in vacuo. The purified oligos were then desalted with Gel-Pak desalting columns (Glen). The solution was lyophilized, and the RNA dissolved in nuclease-free water for concentration determination by UV absorbance and quality assessment via denaturing PAGE.

[0257] B) Duplex assembly

[0258] Barcoded RNA duplexes (i.e. double-stranded analytes with barcodes) were generated by combining equimolar RNA solutions of a barcoded strand (see Table 1 for exemplary sequences) and a fully or partially complementary strand (see Table 3 for exemplary sequences) in 1 x PBS. The two strands were annealed using a temperature ramp from 95 °C to 15 °C in approx. 20 minutes (at a rate of 0.1 °C every ~3 seconds) to form a 10 pM stock solution of double-stranded analytes. A non-exhaustive list of duplexes generated is provided in Table 5.

[0259] Table 5: Double-stranded analytes along with their corresponding strand IDs. Strand 1 is the barcode-containing strand. .170.173772 / 01 42.170.173772 / 01

[0260] C) Conjugation of cargo molecules

[0261] In some embodiments, RNA cargo molecules were covalently conjugated to single- or double-stranded analytes using Inverse Electron-Demand Diels-Alder reaction (I EDDA) as previously described (WO2024 / 069235). Briefly, a 5’ amino modified RNA strand (e.g. SEQ. ID 239, diluted to a final concentration of 100-200 pM in 50% v / v DMSO and 20% v / v 0.5 M bicarbonate buffer, pH = 8.4) was treated with a heterobifunctional tetrazine-NHS ester (5-20 eq, dissolved to a stock solution of 100 mM in anhydrous DMSO), the mixture was agitated for 1-3 hours at 30°C and the resulting tetrazine-NHS oligo (e.g. SEQ. ID 240) was purified by RP-HPLC or EtOH precipitation. A 5’ norbornene modified cargo RNA strand (e.g. SEQ. ID 261) (5 nmol, 1.0 eq, 1400 pM final concentration) was then reacted with the tetrazine- NHS oligo (15 nmol, 1.6 eq) in 1 * PBS buffer for 12 hours at ambient temperature. The 42.170.173772 / 01 product was purified by I EX chromatography using a DNAPac PA10022 x 250 mm column at 75 °C, at a flow rate of 25 mL / min.40% to 60% B in 30 min (A: 0.1 M NaCI pH 7, B: 1.0M NaCI). Fractions containing the product were concentrated and desalted.

[0262] To assemble double-stranded analytes conjugated to an RNA duplex as cargo, stoichiometric amounts of the constituting strands (comprising a barcoded strand as shown in Table 1, an RNA-RNA IEDDA conjugate produced as described above and a strand fully or partially complementary to the conjugated cargo strand e.g. SEQ. ID 262) were mixed in PBS or artificial cerebrospinal fluid at a final construct concentration of 10 pM. The mixture was then heated at 95 °C for 5 min followed by a gradual cooldown to 4 °C (2.5 °C / min) using a PCR thermocycler.

[0263] Example 2: Bulk RNA seq library preparation (Methodology for later Examples)

[0264] A method for the detection of in cellulo 5’ phosphorylated RNA analytes by bulk RNA sequencing is provided. A schematic diagram of the workflow is presented in Figure 5. Primers and sequencing adapters used are shown in Table 6. Buffers used for hybridization capture and bead washing are listed in Table 7.

[0265] Table 6: List of oligonucleotides used during various steps of the bulk RNAseq library preparation protocol. Oligos were obtained from IDT. / 5AmMC6 / indicates 5' amino modifier C6, and * indicates phosphorothioate. 42.170.173772 / 01

[0266] Table 7: Composition of buffers used for hybridization capture and bead washing steps.

[0267] A) Functionalization of RT beads for hybridization capture

[0268] A total of 13.5 mg Dynabeads M-270 Carboxyl (Thermo Fisher, #14306D) were washed thrice with 1x phosphate buffered saline (PBS), resuspended in propylphosphonic anhydride (T3P) solution (16 mg / mL of T3P in DMF) and incubated under shaking (630 rpm) for 10 min at 25 °C. Then, 10 nmol of capture oligo (see Table 6) was added, the mixture was buffered with 4-(2-hydroxyethyl)piperazine-1-ethanesulfonic acid (HEPES, pH 8) at a final concentration of 25 mM and incubated under shaking (630 rpm) for 2 hours at 25 °C. The conjugated beads were washed thrice with 1 x PBS to remove unreacted oligo and conjugation yield was determined by measuring the DNA concentration in the supernatant of the wash fractions using an Epoch microplate spectrophotometer (BioTek / Agilent, USA). The beads were deactivated by treatment with 50 mM tris(hydroxymethyl)aminomethane (Tris, pH 7.4) for 15 min to block the unreacted surface (as per manufacturer’s protocol), followed by two washes with 1x PBS to remove Tris. The functionalized beads were adjusted to a concentration of 5mg / mL in 1 x PBS and stored at 4°C.

[0269] Prior to use, 0.20 mg of functionalized RT beads were allotted per sample, washed twice in 150 pL Bead Binding Buffer and resuspended in 150 pL Bead Binding Buffer.

[0270] B) Hybridization-based barcode / analyte enrichment

[0271] Per sample, equal volumes of cellular lysate and Bead Binding Buffer were mixed and added to 0.20 mg of RT beads from step A. The mixture was heated to 95 °C for 10 min to disrupt secondary structures and incubated for 30 min at room temperature to capture the barcoded RNA strands. Following three washes with Bead Wash Buffer (see Table 7), the beads were treated with 5 U of RNaseH (New England Biolabs, #M0297) for 20 minutes at 37 °C to remove non-specifically bound, unmodified cellular RNA. (This step is optional and can be skipped, especially in cases where barcoded strands contain one or more unmodified RNA 42.170.173772 / 01 nucleotides). Beads were washed thrice with Bead Wash Buffer in preparation for the ligation step.

[0272] C) 5’ Adapter ligation and cDNA synthesis

[0273] To ensure the selective detection of in cellulo 5’ monophosphorylated barcoded strands, a phosphorylation-dependent 5’ ligation step was performed. Adapter sequences (see Table 4 / 6) were designed to be chimeric DNA / RNA oligos comprising a PCR handle for amplification and a 6-nt sample identifier sequence to allow for sample pooling post-ligation. The sample identifier sequence was flanked by 5 degenerate DNA bases and 8 degenerate RNA bases, respectively, which function as unique molecular identifier (UMI) for PCR duplicate detection and removal. Barcoded strands lacking a 5’ monophosphate (for example due to endosomal entrapment) cannot serve as substrates for T4 RNA ligase and hence remain unligated, preventing their amplification by PCR and thus their detection by sequencing.

[0274] Ligation of 5’ adapters was performed in 10.5 pL of ligation mix [1 T4 RNA Ligase Reaction Buffer, 4.76 mM ATP, 28.6% PEG8000, 10 pmol of 5’ adapter and 60 U T4 RNA Ligase 1 (ssRNA Ligase), High Concentration (NEB, #M0437M)]. Following incubation for 1 h at 37 °C, the samples were placed on a magnetic separation stand (NEB, #S1515S) and the supernatant was discarded. Beads were washed three times with 200 pL Bead Wash Buffer to remove any unligated adapters, pooled and resuspended in 25 pL of reverse transcription mix (1x First-Strand Buffer, 0.5 mM dNTPs, 1 mM DTT, 7.5 U SUPERase ln RNase Inhibitor (Invitrogen, #AM2696), 125 U SuperScript™ IV Reverse Transcriptase (Invitrogen, #18090050)). Reverse transcription was performed for 15 min at 50 °C to generate solidphase first-strand cDNA. Beads were washed three times with 200 pL Bead Wash Buffer to remove the RNA template strand.

[0275] D) PCR amplification, QC and sequencing

[0276] The solid-phase cDNA generated in step C was amplified by polymerase chain reaction (PCR). Beads were resuspended in 25 pl cDNA amplification mix (1 * Q5 Hot Start High- Fidelity Master Mix (NEB, #M0494S), 0.5 pM F PCR primer, 0.5 pM R PCR primer) and amplification was performed using the following programme: 30 s at 98 °C, 10 cycles of 10 s at 98 °C, 5 s at 68 °C and 5 s at 72 °C followed by a final extension at 72 °C for 2 min and cooling to 4 °C.

[0277] Beads were placed on a magnetic separation stand and the supernatant containing the amplified, now double-stranded DNA was purified using ProNex Size-Selective Purification 42.170.173772 / 01

[0278] System (Promega, #NG2001) with a ratio of 3.1X to remove excess primers. Cycling conditions for indexing PCR were determined by amplifying 1 pl of purified DNA via quantitative PCR in a 15 pL reaction volume (1 * Powerllp™ SYBR® Green Master Mix (Applied Biosystems, #A25742), 0.3 pM P5 and P7 primers), whereby the quantification cycle (Cq) value obtained was used as the cycle number applied to final library amplification during indexing PCR. Indexing PCR was performed in duplicate in a total volume of 50 pL of amplification mix (1* Powerllp™ SYBR® Green Master Mix (Applied Biosystems, #A25742), 6 pL of purified pre-amplified DNA, 5 pL of universal P5 and P7 Illumina indexing primers (xGen™ UDI 10nt Primer Plates 1-4, IDT, #10008052)), and the following PCR programme was used: 30 s at 98 °C and predetermined cycles of 10 s at 98 °C, 10 s at 68 °C and 10 s at 72 °C followed by final extension at 72 °C for 2 min and cooling to 4 °C.

[0279] The indexed libraries were then purified using ProNex Size-Selective Purification System (Promega, #NG2001) by performing two rounds of size selection with ratios of 1.65X and 1.55X, respectively, to enrich correctly sized amplicons whilst removing by-products as well as excess primers. Quality control was performed either on a 2100 Bioanalyzer Instrument (Agilent) using High Sensitivity D1000 Reagents (Agilent, #5067-5585) or on a TapeStation 4200 instrument (Agilent) using High Sensitivity D1000 Reagents (Agilent, #5067-5593) as per manufacturer’s protocol. The libraries were quantified using a Qubit 4 Fluorometer (Thermo Fisher Scientific), and pooled for sequencing. Sequencing was performed on an Illumina Novaseq instrument.

[0280] E) Sequencing readout

[0281] Sequencing results were processed using a custom analysis pipeline to identify barcodes and quantify either their gross uptake, their phosphorylated fraction or both. For this, first read quality filtering and trimming was performed. The remaining high-quality reads were demultiplexed based on the bulkSEQ sample barcodes with a mismatch tolerance of 1. Unique molecular identifier (UMI) sequences were extracted and corrected for sequencing errors based on abundance and quality score criteria. The reads were then mapped against the known list of reference barcodes present in the injected analyte library and counted to produce a count matrix. Reads that share the same UMI, sample barcode and analyte barcode were considered PCR duplicates and counted once. Where applicable, to ensure that count matrices could be compared across samples, the read counts were normalized against internal standard molecules that had been added to the libraries in known amounts. 42.170.173772 / 01

[0282] Example 3: Selective detection of 5’ monophosphorylated analytes in water

[0283] The use of bacteriophage T4 PNK to label 5' DNA or RNA ends with32P has been instrumental in the development of assays for the analysis of nucleic acid structure, molecular cloning and nucleic acid sequencing (Wang et al., EMBO J., 2002, 21(14), p3873- 3880). As a proof of concept interrogation of product formation within the Bulk RNA seq library preparation pipeline, initial tests were carried out in the presence or absence of T4 PNK to identify the size and identity of the expected product from the endogenous phosphorylation expected upon administration.

[0284] The methodology for Bulk RNA seq library preparation is outlined in Example 2. Where modifications of the methodology took place, those are referenced below. Primers and sequencing adapters used are shown in Tables 4 / 6. Buffers used for hybridization capture and bead washing are listed in Table 7.

[0285] A double-stranded analyte (DX0001, see Table 5) containing a barcoded strand (SEQ. ID 36, see Table 1) was spiked in 300 pL nuclease-free water at a final concentration of 0.1 pM. Four separate samples were generated. Half of those samples were then taken through our Bulk RNA seq library preparation as described in Example 2, while for the other half an additional step of 5’ phosphorylation was added before the 5’ Adapter ligation and cDNA synthesis step. This was performed by treating the beads with 10 II of T4 Polynucleotide Kinase (New England Biolabs, #M0201L) for 30 minutes at 37 °C according to manufacturer’s protocol.

[0286] Subsequently, all samples were taken through the 5’ adapter ligation and cDNA synthesis steps, whereby the same adapter sequence was used for all samples during ligation (Table 4, SEQ. ID 248) except for two control samples to which no adapter sequence was added (“Adapter-”). Processing for all samples then continued up to indexing PCR cycling condition determination via quantitative PCR (1 * Powerllp™ SYBR® Green Master Mix (Applied Biosystems, #A25742), 0.3 pM P5 and P7 primers) on a QuantStudio™ 5 Real-Time PCR system (Applied Biosystems) for 45 cycles.

[0287] Results

[0288] Quantitative PCR amplification curves of generated libraries are shown in Figure 9. Amplification was only observed in the water samples spiked with analytes when a 5’ phosphorylation step by incubation with PNK was added in the Bulk RNA seq library 42.170.173772 / 01 preparation pipeline (Figure 9, “PNK+”). In the absence of this additional in vitro phosphorylation step (Figure 9, “PNK-”) only weak, late-appearing non-specific amplification was detected in the graphs generated by the Design & Analysis software of QuantStudioTM 5 Real-Time PCR system. Further controls included the absence of adapter to be ligated on the 5’ flank of the barcoded strand which also resulted in no product formation regardless of whether the sample was pre-incubated with PNK (Figure 9, “PNK+ Adapter-”) or not (Figure 9, “PNK- Adapter-”). Results from this experiment confirm that only barcoded strands that have a phosphorylated 5’ terminus are detected, which intracellularly would only occur upon exiting the endosomes and encountering internal kinases in the cell’s cytoplasm.

[0289] Example 4: Detection of endogenously phosphorylated analytes in rat tissue samples

[0290] Following up from our in vitro tests presented in Example 3, detection of 5’ monophosphorylated analytes in tissue lysates, coming either from rats that had been administered with analytes conjugated to differently modified RNA duplexes as cargo (see Tables 8A and 8B), or that have been spiked-in with known concentrations of unconjugated analytes (DX0001 or SEQ. ID 36), was performed. The aim was to test if endogenously phosphorylated analytes were present in high enough abundance in treated tissues to enable efficient library preparation.

[0291] The protocol for preparing cargo-conjugated analytes is described in Example 1C. The methodology for Bulk RNA seq library preparation is outlined in Example 2. Primers and sequencing adapters used are shown in Tables 4 / 6. Buffers used for hybridization capture and bead washing are listed in Table 7.

[0292] Table 8A: Library of cargo-conjugated analytes for Example 4. 42.170.173772 / 01

[0293] Table 8B: Sequences of RNA strands used as cargo covalently attached to the analytes shown in Table 8A. mA / mC / mG / mll indicates 2’0Me (2’-methoxy), / i2P indicates internal 2’- O-propargyl modified nucleotides, / 5Norbornene / indicates 5’ norbornene modifier, / 32PG / indicates 3’ terminal 2’-O-propargyl modified G.

[0294] A) Intrathecal administration of cargo-conjugated analytes in rats

[0295] All in vivo experimental procedures were undertaken at Pharmidex Ltd and were subjected to the provisions of the United Kingdom Animals (Scientific Procedures) Act 1986 administered by the Home Office.

[0296] A library comprising 15 cargo-conjugated analytes (see Table 8) was formulated at up to 30 mg / ml in artificial cerebrospinal fluid (aCSF). At a minimum of 30 minutes before surgery, eight to ten week old female Sprague Dawley (SD) rats were subcutaneously administered 1 mg / kg of meloxicam and 0.1 mg / kg of buprenorphine. After anesthesia with isoflurane, rats were placed on a warm heating pad and treated with eye lubricant; the intrathecal (IT) injection site was shaved and disinfected; and an incision was made to expose the spinal column. Each animal was injected with 30 pl of the analyte library by lumbar puncture in the dorsal region of the spine between the L3-L5 vertebral space. Vehicle animals were injected 42.170.173772 / 01 with 30 pL of aCSF. The administration was performed with an insulin syringe. Proper placement of the needle was confirmed via CSF backflow in the hub of the needle.

[0297] Once administration was completed, gentle constant pressure on the plunger was maintained for 30 seconds; the incision was sutured and secured with tissue glue; and the rats were placed in sternal recumbency on a heating pad until recovery (Brown et al., Nat. Biotechnol., 2022, 40, p1500-1508). Seven days post IT administration, animals were culled and various CNS regions were collected, stored separately in RNAIater (Sigma, #R0901) at 4°C overnight and then stored at -70°C until tissue homogenization.

[0298] B) In vivo sample preparation compatible with phosphorylation readout:

[0299] Whole CNS regions were homogenized in 900 pL QuantiGene™ Homogenizing Solution (Invitrogen #QG0517) with added Proteinase K at a final concentration of 0.5 mg / mL (Invitrogen #25530049] for 15 min at 30 HZ per block side on a TissueLyser III (Qiagen) repeated until no tissue clumps were visible. Samples were then heated to 65°C for 30 minutes and vortexed for 1 min at maximum speed every 10 minutes. A subset of those tissue lysates coming from IT injected and untreated vehicle control animals was selected for downstream analysis and the rest were stored at -70°C.

[0300] For some of the test conditions used, barcoded strands, in single-stranded (SEQ. ID 36, see Table 1) or duplex configuration (DX0001 , see Table 5), were spiked in CNS tissue lysates originating from untreated vehicle control rats at a final concentration of 0.1 pM. These spiked-in tissue samples were processed alongside the injected rat tissue lysates referenced above for Bulk RNA Seq library preparation. The pipeline for half of those samples followed the method as outlined in Example 2, whereas for the rest of the samples an additional step of incubation with PNK for in vitro 5’ phosphorylation of the barcoded strands was implemented in the pipeline in a similar manner as presented in Example 3.

[0301] Tapestation readout

[0302] Sample libraries were run on a High Sensitivity D1000 Tapestation screentape and results are shown in Figure 10. Main (-183 bp) and non-specific side (-175 bp) products were observed. The predominant product is shown with an arrow, i.e. the correct product in lanes B1, D1 , F1 and H1 and the side product in C1, E1 , G1, A2 and B2. Despite the absence of an in vitro phosphorylation step, the main product was detected in CNS tissue samples from rats that have been administered a library of 15 barcoded analytes via IT injection (Figure 10, Lanes C1 and E1 , albeit contaminated with side product). In those samples, the observed main product was reduced compared to that from samples where an in vitro 5’ 42.170.173772 / 01 phosphorylation step was added in the pipeline (Figure 10, Lanes B1 and D1). This confirms that only a fraction of the delivered analytes efficiently exits the endosomes intracellularly, is 5’ phosphorylated by internal kinases and can, therefore, be detected using this method.

[0303] For the in vitro spiked-in lysate samples, in the absence of an artificial in vitro 5’ phosphorylation step (Figure 10, Lanes G1 [double-stranded, DX0001] and A2 [singlestranded, SEQ. ID 36]), only the non-specific side product was observed in agreement with our in vitro observations in Example 3 and the absence of phosphorylation taking place in the cytoplasm. Controls lacking the adapter sequence show no detectable main (correct) product regardless of whether the single-stranded analyte (SEQ. ID 36) was in vitro phosphorylated (Figure 10, Lane B2) or not (Figure 10, Lane C2) during the Bulk RNA seq library preparation.

[0304] Example 5: Use of endogenous 5’ phosphorylation as a proxy for productive analyte uptake in cells

[0305] As an additional proof-of-concept, we designed an experiment to assess if exogenously phosphorylated analytes transfected in cells (gross uptake) can be distinguished from endogenously phosphorylated analytes (cytosolic fraction) based on differential barcode abundance.

[0306] Reverse transfection of A 549 cells and cell lysate preparation

[0307] Human A549 cells (ATCC, CCL-185™) cultured in DMEM supplemented with 10% FBS were reverse transfected by adding a single analyte containing a barcoded strand at 10 nM final concentration complexed with Lipofectamine RNAiMAX® Transfection reagent (Invitrogen #13778150) to the cell suspension as per manufacturer’s protocol. Each barcoded strand used was unique in its sequence. Analytes were transfected either in single-stranded or double-stranded form (see Table 10). As technical controls, for each of the single-stranded analytes that were transfected their pre-phosphorylated counterparts were also included (see Table 9). Similarly, for one of the double-stranded analytes, there was also a pre-phosphorylated version of this duplex included in the transfection conditions (see Table 10). Untransfected control cells were also included. 42.170.173772 / 01

[0308] Table 9: Pre-phosphorylated analyte strands used in Example 5. / 5Phos / indicates 5’ phosphorylation, lower-case a / c / g / u indicates 2’F (2’-fluoro), mA / mC / mG / mll indicates 2’0Me

[0309] (2’-methoxy) and * indicates PS (phosphorothioate).

[0310] Table 10: Analytes for Example 5.

[0311] Cells were plated in 6-well plates at a density of 3.5 x 105cells / well and each condition was run in triplicate. After 24 hours of incubation with the respective barcoded strands, media from all wells was aspirated, cells were washed once with 1mL PBS and harvested after incubating with 500 pL TrypLE (Gibco #12605010) for 5 mins at 37°C in a cell culture incubator. 42.170.173772 / 01

[0312] Following incubation, trypsinization was halted by adding 2mL of DMEM / 10% FBS per well and cells were subsequently counted with a Trypan blue stain 0.4% (Invitrogen #T10282) on a Countess® II FL Automated Cell Counter (Life Sciences) as per manufacturer’s instructions. Equal cell numbers per condition were aliquoted, pelleted by centrifugation at 300 ref for 4 minutes, supernatant aspirated and subsequently snap frozen in liquid nitrogen and stored at -70°C freezer until further processing. Snap frozen cell pellets were lysed using the QuantiGene™ Sample Processing Kit for cultured cells (ThermoFisher #QS010) supplemented with Proteinase K (included in the same kit, QuantiGene™ Sample Processing Kit for cultured cells, ThermoFisher #QS010).

[0313] Lysis mixture from the kit was pre-warmed at 37°C for 30 minutes followed by gentle swirling. Working Lysis Mixture was prepared by adding 10 pL of Proteinase K per volume (mL) of Lysis Mixture required. Working Lysis Mixture was then diluted by adding 2 volumes of nuclease-free water to 1 volume of Working Lysis Mixture. The volume of Diluted Working Lysis mixture required and added to the cell pellets was calculated per sample to achieve a density of 400 cells / pL. Following addition of Diluted Lysis Mixture, cell pellets were pipetted up and down 15 times, followed by vortexing at maximum speed for 1 min in order to achieve dissolution. Samples were incubated at 55°C for 30 minutes and once viscosity of the solution was similar to the Lysis Mixture the cell lysate were then processed for Bulk RNA seq library preparation as outlined in Example 2.

[0314] Sequencing readout

[0315] Tapestation traces confirming successful library preparation are shown in Figure 11. The sequencing data were analyzed as described in Example 2E. To account for differences in sequencing depth across libraries and allow for comparisons across samples (in the absence of internal standards), reads were randomly subsampled 20 times without replacement to approximately 1 Mio reads per sample. Counts were then plotted based on experimental conditions (Figure 12). As expected, externally phosphorylated (5’P) analytes were detected with higher efficiency than analytes with a 5’OH group. Moreover, 5'OH analytes transfected in live cells were detected more readily than the same 5’OH analytes spiked into cell lysate (background control), suggesting that both dsRNA and ssRNA barcodes are indeed phosphorylated inside cells and not by residual enzymatic activity in lysates. 42.170.173772 / 01

[0316] Example 6: Limit of detection determination

[0317] Internal standards (i.e., synthetic spike-ins) can be leveraged to assess the dynamic range and lower limit of detection of RNA sequencing assays. We used a pool of 15 RNA duplexes as internal standards, each containing a unique 8-nt barcode sequence with varying GC contents, covering a 55concentration range to measure sensitivity, accuracy, and biases in the bulk RNA sequencing workflow for barcode detection.

[0318] Preparation of spike-in RNA beads

[0319] Following annealing (as described in Example 1 B), 5 sub-pools containing a mix of 3 equimolar barcoded duplexes each were prepared, subjected to a 1:5 serial dilution in nuclease-free water and combined at staggered concentrations into a single pool as shown in Table 11. The single pool was further diluted by a factor of 180 with nuclease-free water, then 50 pL of said diluted single pool were mixed with 100 pL of nuclease-free water and 150 pL of Bead Binding Buffer and added to 0.2 mg of pre-washed RT beads in 150 pL Bead Binding Buffer. After heating to 95 °C for 2 minutes to disrupt secondary structures, the mixture was incubated for 30 minutes at room temperature. The beads were washed thrice in Bead Wash Buffer (to remove unbound RNA) and resuspended in 1850 pL Bead Wash Buffer.

[0320] Table 11 : Preparation of spike-in mixture from 5 sub-pools, each containing a mix of 3 equimolar barcoded duplexes. Initial concentration and final number of molecules (nMolecules) reflect values for individual duplexes before pooling and in the final spike-in amount used, respectively. 42.170.173772 / 01

[0321] Bulk RNAseq library preparation

[0322] The bulkSEQ library preparation was carried out as described in Example 2 above, in nine independent technical replicates. For each replicate, 10 pL of spike-in beads were added to a sample of nuclease-free water followed by the addition of 0.20 mg of functionalized RT beads and a volume of Bead Binding Buffer equal to the initial sample volume. This was performed to replicate the experimental conditions for enriching barcoded strands (analytes) from cellular lysates.

[0323] Following hybridization capture and RNaseH treatment as described in Example 2B, the captured barcoded RNA strands were 5’ phosphorylated for 30 min at 37°C using 10 II of T4 Polynucleotide Kinase (New England Biolabs, #M0201) in the presence of 1 mM adenosine 5'-triphosphate (ATP, New England Biolabs, #P0756) and 50 II of SUPERase ln RNase Inhibitor (Invitrogen, #AM2696). Then, the beads were washed thrice with Bead Wash Buffer and the subsequent steps of 5’ adapter ligation, reverse transcription, cDNA amplification, QC and sequencing were performed as described in Example 2C-D.

[0324] Results

[0325] Raw sequencing data were analyzed as described in Example 2E. The absolute number of reads observed per barcode was plotted against the known number of internal standard molecules and the results are shown in Figure 13A. On logarithmic scale, sequencing read counts displayed a strong linear correlation (Pearson's r = 0.90) with internal standard input amounts, demonstrating that internal standards can be used as normalizers for absolute barcode quantification. The threshold for estimating the lower limit of detection (LLOD) was empirically set to ~5 reads per barcode, yielding a LLOD of <1 amol.

[0326] As shown in Figure 13B, the coefficient of variation (CV) across technical PCR replicates increases with decreasing number of spike-in input molecules. The observed pattern roughly follows a Poisson distribution, indicating that assay variance at low input concentrations is likely due to stochastic sampling noise and not primarily driven by technical noise of the assay. Overall, the data illustrate that RNA barcodes may be identified according to the invention with good sensitivity and accuracy. 42.170.173772 / 01

[0327] Example 7: Barcode quantification by bulk RNAseq using standard analytes

[0328] To assess the use of synthetic standards for absolute quantification of barcode abundance in tissues, a first pool comprising 15 cargo-conjugated analytes (with RC001 as cargo, see Table 8B) and a separate second pool comprising 28 internal standards were spiked at known concentrations into rat brain tissue samples obtained from wild-type vehicle animals. By comparing the analyte concentrations calculated from normalization to internal standards against the known input analyte concentrations, quantification accuracy can be determined.

[0329] Tissue preparation

[0330] Eight to ten week old female Sprague Dawley (SD) rats were given an intrathecal administration of 30 pl aCSF (artificial cerebrospinal fluid, Bio-Techne Ltd, #3525) (vehicle control) as previously described (Example 4A). Animals were sacrificed 7 days postadministration, brain tissues were dissected and stored in RNAIater (Merck, #R0901) overnight before the whole region was homogenized in 900 pL lysis buffer (QuantiGene Homogenizing Solution (Invitrogen, #QG0517), 0.5 mg / mL Proteinase K (Invitrogen, #QG0517)) using a TissueLyser III (Qiagen) according to the manufacturer's instructions.

[0331] Preparation of analyte and standard pools

[0332] Cargo-conjugated analytes were prepared as described in Example 1C and comprised the duplexes listed in Table 12A covalently attached to the RNA cargo duplex RC001 (see Table 8B). Internal standards (see Table 12B) were used in duplex form and assembled as described in Example 2A. Each oligo contained a unique 8-nt identifier sequence.

[0333] To generate the analyte pool, 5 sub-pools containing a mix of 3 equimolar cargo-conjugated barcoded duplexes each were prepared, subjected to a 1 :2 serial dilution in nuclease-free water and combined at staggered concentrations as shown in Table 12A. The analyte pool spanning 5 orders of magnitude on a Iog2 scale was further diluted by a factor of 30 with nuclease-free water and stored at -70 °C.

[0334] To assemble the pool of internal standards, 9 sub-pools containing between 1 to 4 equimolar barcoded duplexes each were prepared, subjected to a 1 :2 serial dilution in nuclease-free water and combined at staggered concentrations as outlined in Table 12B. The standard pool, spanning a concentration range of 9 orders of magnitude on a Iog2 scale, was further diluted by a factor of 80 with nuclease-free water and stored at -70 °C. 42.170.173772 / 01

[0335] Table 12A: Preparation of the analyte pool. Columns “Initial concentration” and “final nMolecules” reflect values for individual duplexes and do not refer to the overall sub-pool. “Initial” and “final” indicate the values before pooling and after spike-in, respectively. The indicated duplexes were used as conjugates linked to cargo RC001 by IEDDA.

[0336] Table 12B: Preparation of the standard pool. Columns “Initial concentration”, “final amol” and “final nMolecules” reflect values for individual duplexes and do not refer to the overall subpool. “Initial” and “final” indicate the values before pooling and after spike-in, respectively. 42.170.173772 / 01

[0337] Bulk RNAseq library preparation

[0338] The bulkSEQ library preparation was carried out as described in Example 2 above. Three independent biological replicates were prepared. For each replicate, 10 pL of the analyte pool and 10 pL of the internal standard pool were added to 450 pL of tissue lysate. Then, 0.20 mg of functionalized RT beads were added and a volume of Bead Binding Buffer equal to the initial sample volume.

[0339] Following hybridization capture and RNaseH treatment as described in Example 2B, the captured barcoded RNA strands were 5’ phosphorylated for 30 min at 37°C using 10 II of T4 Polynucleotide Kinase (New England Biolabs, #M0201) in the presence of 1 mM adenosine 5'-triphosphate (ATP, New England Biolabs, #P0756) and 50 II of SUPERase ln RNase Inhibitor (Invitrogen, #AM2696). The beads were washed thrice with Bead Wash Buffer and the subsequent steps of 5’ adapter ligation, reverse transcription, cDNA amplification, QC and sequencing were performed as described in Example 2C-D.

[0340] Results

[0341] Raw sequencing data were analyzed as described in Example 2E. The log-transformed number of reads observed per internal standard was plotted against its known log- transformed input concentration and a linear regression model was applied (Figure 14A). Although the linear model fit the data reasonably well overall (r2> 0.79), some internal standards were detected more efficiently than others. To correct for these differences, barcode-specific amplification efficiency factors were calculated and used as scaling factors. Following scaling, the adjusted read counts of analyte barcodes were normalized against the internal standards for absolute quantification using a multinomial model and the so-obtained 42.170.173772 / 01 inferred concentration was plotted against the known input concentration of each analyte (Figure 14B) to determine the accuracy of the model. The multinomial model reliably predicted the concentration of the cargo-conjugated analytes with a mean fold error (MFE) of 1.03.

[0342] Example 8: Analyte detection in isolated nuclei

[0343] In this experiment nucleic acid analytes that had been released from endolysosomal vesicles and transported to the nucleus were detected by analyte detection in the nucleus.

[0344] Reverse transfection of A 549 cells

[0345] Reverse transfection of A549 cells was carried out using Lipofectamine RNAiMAX (Invitrogen #13778150) as described under Example 5. Transfections were performed in triplicate in 6-well plates at a final concentration of 10 nM of single-stranded analyte (SEQ. ID 14) or double-stranded PP / B- targeting siRNA (Hassler et al., Nucleic Acids Res, 2018, 46(5), p2185-2196). Untransfected cells were included as a negative control.

[0346] One day after transfection, media from all wells was aspirated, cells were washed once with 1 mL PBS and detached by incubation with 500 pLTrypLE (Gibco #12605010) for 5 min at 37°C. Trypsinization was halted by adding 2 mL of DMEM / 10% FBS per well, triplicates were pooled and cells were counted with a Trypan blue stain 0.4% (Invitrogen #T10282) on a Countess® II FL Automated Cell Counter (Life Sciences) as per manufacturer’s instructions.

[0347] Single nuclei isolation

[0348] Single nuclei were isolated according to the established protocol from 10x Genomics for nuclei isolation from single-cell suspension (CG000124, Rev F, https: / / cdn.10xgenomics.com / image / upload / v1660261285 / support- documents / CG000124_Demonstrated_Protocol_Nuclei_isolation_RevF.pdf). Briefly, cells were centrifuged at 300 ref for 5 min at 4°C, after which the supernatant was removed. The cell pellet was resuspended in 200 pl of Lysis Buffer [10mM Tris-HCI (pH 7.4), 10mM NaCI, 3mM MgCh, 0.025% pre-diluted Lysis Reagent (Pre-diluted Lysis Reagent: 25% IGEPAL CO-630 (Sigma, #542334)] and incubated on ice for 1-5 minutes. Following incubation, 800 pL of Nuclei Wash & Resuspension Buffer [1% Bovine Serum Albumin (BSA) Solution, 0.2 U / pL RNase Inhibitor (40U / pL stock) in 1* PBS]) were added and mixed by pipetting up and down. The cell solution was centrifuged at 500 ref, for 10 min, at 4°C and the supernatant was removed and kept aside. The pellet was resuspended in 1 mL of Nuclei Wash & Resuspension Buffer, centrifuged at 500 ref, for 10 min, at 4°C and the supernatant was 42.170.173772 / 01 once more removed and kept aside. This last step was repeated once more. The pellet was then resuspended in 1 ml of Nuclei Wash & Resuspension Buffer aiming for a final concentration of 700-1200 nuclei / pl. Nuclei were counted with a Trypan blue stain 0.4% (Invitrogen #T10282) on a Countess® II FL Automated Cell Counter (Life Sciences) as per manufacturer’s instructions.

[0349] Analyte detection in isolated nuclei by single-nuclei sequencing

[0350] Although not performed in this method, analytes may be detected by single-nuclei sequencing. In such a method nuclei are sequenced according to 10x Genomics’ user guide for Chromium Next GEM Single Cell 5' Reagent Kits v2 (Dual Index) with Feature Barcode technology for CRISPR Screening (CG000510 Rev B). Briefly, Chromium Next Gem Chip K (10x Genomics, #1000287) is loaded with a customized master mix [comprising a suitable primer and Poly-dT RT Primer (10x Genomics, #2000007) in place of CRISPR poly-dT RT Primer Mix (10x Genomics, #2000593)] and single nuclei suspension for a target recovery of 5000 nuclei as per the Cell Suspension Volume Calculator Table. Next, Gel Beads and Partitioning Oil are loaded as instructed and the Chip is run in Chromium. After the run, GEMs are incubated in a thermal cycler for reverse transcription (53°C for 45 min, 85°C for 5 min, hold at 4°C) and subjected to Post-GEM-RT clean-up with Dynabeads followed by cDNA amplification (98°C for 45 sec; 12 cycles of 98°C for 20 sec, 63°C for 30 sec, 72°C for 1 min; 1 cycle of 72°C for 1 min, hold at 4°C). Amplified cDNA is cleaned up with SPRIselect reagent kit (Beckman Coulter, #B23318) as instructed, with the pellet used for 5’ Gene Expression library construction (as per protocol without modifications) and the supernatant kept for CRISPR library construction with minor deviations from the protocol as outlined below.

[0351] First, supernatant clean-up is performed with SPRIselect and the obtained material is used as input for Feature PCR, whereby Feature SI Primers 4 (10x Genomics, #2000592) in the Feature PCR Mix are replaced by custom primer sequences to achieve analyte amplification. Post Feature PCR, the samples are cleaned up using SPRIselect and the obtained material is used as input for Sample Index PCR. Amplified indexed products are cleaned up using double-sided size selection with SPRIselect, subjected to quality control by TapeStation and sequenced together with the 5’ Gene Expression library according to 10x Genomics’ recommendations. 42.170.173772 / 01

[0352] Analyte detection in isolated nuclei by bulk RNA sequencing

[0353] Following single nuclei isolation, both isolated nuclei and kept aside supernatant (nonnuclear fraction) were homogenized using the QuantiGene™ Sample Processing Kit for cultured cells (ThermoFisher #QS010). Bulk RNA seq library preparation was performed as described in Example 2 with one exception: after hybridization capture and RNaseH treatment, the beads were incubated with 10 II of T4 Polynucleotide Kinase (New England Biolabs, #M0201 L) for 30 minutes at 37 °C to enable efficient adapter ligation in the subsequent step. Final libraries were quality controlled on a TapeStation 4200 instrument (Agilent) using High Sensitivity D1000 Reagents (Agilent, #5067-5593) as per manufacturer’s protocol.

[0354] Small interfering RNA detection by fluorescence hybridization

[0355] Following single nuclei isolation, both isolated nuclei and kept-aside supernatant (nonnuclear fraction) were mixed with a 5’ biotinylated capture probe and a 3’ Digoxigenin- modified detection probe at a final concentration of 1.5 nM and hybridized on a thermal cycler under the following conditions: 90 °C for 5 minutes, 40 °C for 30 min, and a final hold at 12 °C. The hybridized material was transferred onto a streptavidin coated black plate (Thermo-Fisher, #15503) and incubated for 30 min at room temperature. Unbound material was removed through 3 washes with KPL wash solution (Sera Care, #50-63-04) followed by the addition of AttoPhos substrate (Promega, #S1000) according to the manufacturer's instructions. The plate was incubated for a further 30 min at room temperature in the dark and then fluorescence was measured on a plate reader at an excitation wavelength of 435 nm and an emission wavelength of 555 nm.

[0356] Results

[0357] Nuclei suspensions were assessed microscopically to determine nuclei concentration, size and suspension quality. Suspensions were largely free of debris and nuclei aggregates, indicating good quality.

[0358] Nuclei isolated from cells transfected with double-stranded siRNA contained detectable levels of siRNA (Figure 15A). As expected, nuclear siRNA levels were lower than those observed in the non-nuclear fraction, suggesting selective detection of functionally delivered molecules. Non-transfected controls showed no siRNA uptake.

[0359] Aside from siRNA, single-stranded barcoded analytes were also detectable in isolated nuclei. Tapestation traces confirming successful library preparation from nuclear and non- 42.170.173772 / 01 nuclear fractions are shown in Figure 15B. Differences in band intensity are not reflective of analyte concentrations and cannot be interpreted quantitatively.

[0360] Example 9: Method for Ago2 pulldown

[0361] The general methodology for performance of the invention in which Ago2 pulldown is used as the readout is shown in Figure 3. Ago2 immunoprecipitation (IP) has been previously described in literature (Lin et al., EMBO reports, 2024, 25(5), p2441-2478).

[0362] The following assay has been adapted from that method. Tissue comprising cells to which the analytes have been administered is homogenized in Lysis buffer which consists of 50 mM Tris-HCI (pH 7.5), 150 mM NaCI, 2mM EDTA, 0.5% Triton-X 100 and protease inhibitor (Roche, REF11873580001).

[0363] To prepare homogenate liver lysate, for example, 1.5 mL lysis buffer and 3.75 pL RNase inhibitor, Murine (New England BioLabs®, M0314S) is mixed with mouse liver tissue, and one 5 mm magnetic bead (QIAGEN, 69997) is added to the mixture. The mixture is then shaken for 4 mins, 30 Hz to obtain the lysate.

[0364] Dynabeads™ Protein G beads (Invitrogen by Thermo Fisher Scientific, 10004D) are first washed using lysate buffer twice and finished by adding the same volume of lysis buffer.

[0365] 500 pL tissue lysate and 15 pL Protein G beads are mixed with 10 pL Ago2 antibody or mouse IgG (control IP) and the mixture is rotated for 2 hours at 4°C . Then the mixture is washed with 250 pL lysis buffer containing 0.63 pL RNase inhibitor 4 times and with 500 pL lysis buffer without RNase inhibitor once. Finally, 200 pL 1x PBS containing 0.25% Triton X is added to the washed mixture and the mixture is heated at 95°C, 600 RPM for 5 mins to elute the captured AGO2-bound material from the beads. The samples are stored under - 20°C for further characterization. Plate-oligo fluorescence (POF) assay is used to quantify the results of Ago2 immunoprecipitation. Additionally, TaqMan® microRNA (miRNA) assay is applied to confirm the effectiveness of the designed Ago2 immunoprecipitation assay. The procedure of the assay follows the manufacturer’s instructions.

[0366] Example 10: Preparation of a pre-phosphorylated standard pool

[0367] As shown in Example 6 and Example 7, internal standards can be leveraged both for quality control purposes and for absolute barcode quantification. In our methods for detecting in 42.170.173772 / 01 cellulo 5’ phosphorylated RNA analytes pre-phosphorylated RNA internal standards may be used. Pre-phosphorylated internal standards may be prepared enzymatically as described in this Example. Alternative methods for enzymatic or non-enzymatic 5’ monophosphorylation, such as solid-phase synthesis, are known in the art and can be used instead without deviating from the scope of this invention.

[0368] Non-radioactive phosphorylation of ssRN A with T4 PNK

[0369] Non-phosphorylated barcoded strands were 5’ monophosphorylated by incubating 5 nmol of each strand with 167 units of T4 PNK (New England Biolabs, #M0201 L) in the presence of 1 mM ATP (New England Biolabs, #P0756) in a total volume of 833 pL (1X T4 PNK reaction buffer) for 120 minutes at 37°C. The enzyme was heat-inactivated by incubating at 65°C for 20 minutes. After letting samples cool down to room temperature, reactions were transferred to an AcroPrep Advance 96-well filter plate with 30kDA MWCO (Cytiva #8165), washed with 100 pl of nuclease-free water and the filtrate transferred to an AcroPrep 24-well 3kDa MWCO Omega™ ultrafiltration plate (Cytiva #97051). Successful phosphorylation was confirmed by LC-MS. This method was found to successfully phosphorylate a variety of different analytes with at least 78% conversion. The phosphorylated analytes may be hybridized to complementary strands to form duplexes.

[0370] Example 11 : Dose-response relationship

[0371] Small interfering RNAs typically show a sigmoidal dose-response relationship with minimal knockdown at low log(siRNA concentrations), a steep increase in knockdown as log(siRNA concentration) increases, followed by a plateau at high log(siRNA concentrations). Given that endosomal escape is a rate-limiting step and prerequisite to siRNA-mediated target gene knockdown, one would also expect the levels of 5’ phosphorylated, cytosolic siRNA to roughly scale with dose - not linearly and not in perfect correlation with activity levels, but generally following similar directional trends. To test this hypothesis, cells were transfected either with PP / B-targeting siRNA or double-stranded, siRNA-resembling barcodes over a range of concentrations and target gene knockdown or cytosolic uptake levels were measured, respectively.

[0372] Reverse transfection of A549 cells was carried out as described under Example 5. The volume of Lipofectamine RNAiMAX® transfection reagent (Invitrogen #13778150) was kept constant across all siRNA or barcode doses. The transfected strands and duplexes are listed 42.170.173772 / 01 in Table 13A and 13B. Untransfected cells were included as a negative control. Each condition was performed in independent triplicates.

[0373] Table 13A: Sense and antisense strands of PP / B-targeting siRNA used in Example 11.

[0374] / 5Cy3 / indicates 5’ Cyanine3 (https: / / eu.idtdna.com / site / catalog / modifications / product / 1106), lower-case a / c / g / u indicates 2’F (2’-fluoro), mA / mC / mG / mU indicates 2’0Me (2’-methoxy), and * indicates PS (phosphorothioate).

[0375] Table 13B: Duplexes used in Example 11 , along with final transfection concentrations.

[0376] Cell lysis and one-step qPCR

[0377] After 48 hours, media from all wells was aspirated, cells were washed once with 100 pL cold PBS and processed using the Luna® Cell Ready One-Step RT-qPCR Kit (NEB #E3030S) as per manufacturer’s instructions. Briefly, cells were lysed for 10 min at 37°C in Cell Lysis Mix supplemented with DNAse I (included in kit NEB #E3030S) and lysis was stopped by adding 10X Luna Cell Ready Stop Solution. The lysates were incubated for 5 min at 25°C and quantitative PCR was performed in a 15 pL reaction volume comprising Luna Universal One- Step Reaction Mix, Luna WarmStart RT Enzyme Mix, forward and reverse primers (0.5 pM final concentration, see Table 14) and 1:10 diluted cell lysate as template. The following cycling conditions were used on a QuantStudio™ 5 Real-Time PCR system: 10 min at 55°C (1 cycle), 1 min at 95°C (1 cycle), 45 cycles of 10 sec at 95°C followed by 30 sec at 60°C.

[0378] Table 14: qPCR primer sequences 42.170.173772 / 01

[0379] Cell lysis and Bulk RNA seq library preparation

[0380] After 24 hours of transfection, media from all wells was aspirated, cells were washed once with 1 mL PBS and detached by incubation with 500 pLTrypLE (Gibco #12605010) for 5 min at 37°C. Trypsinization was halted by adding 2 mL of DMEM / 10% FBS per well, triplicates were pooled and cells were counted with a Trypan blue stain 0.4% (Invitrogen #T10282) on a Countess® II FL Automated Cell Counter (Life Sciences) as per manufacturer’s instructions. Equal cell numbers per condition (-500K cells) were aliquoted, pelleted by centrifugation at 300 ref for 4 minutes and snap frozen in liquid nitrogen. Snap frozen cell pellets were lysed using the QuantiGene™ Sample Processing Kit for cultured cells (ThermoFisher #QS010) supplemented with Proteinase K (included in same kit, QuantiGene™ Sample Processing Kit for cultured cells, ThermoFisher #QS010), as described in Example 5.

[0381] Once viscosity of the solution was similar to the Lysis Mixture, each sample was split in half. One half was processed according to the Bulk RNA seq library preparation workflow as described in Example 2 to enable detection of functional uptake (endogenously phosphorylated barcodes only). Prior to hybridization capture, lysates were spiked with 10 pL of a 1 :100 dilution of a pre-phosphorylated internal standard pool in nuclease-free water. The other half was processed slightly differently to allow detection of total cellular uptake (endogenously phosphorylated and unphosphorylated barcodes). Lysates were spiked with 10 pL of an undiluted pre-phosphorylated internal standard pool in nuclease-free water. Hybridization capture and RNaseH treatment were carried out as described in Example 2B, then the beads were incubated with 10 U of T4 Polynucleotide Kinase (New England Biolabs, #M0201 L) for 30 minutes at 37 °C according to the manufacturer's protocol. As a technical positive control for this 5’ phosphorylation step, 10 pL of an 8.8 nM solution of DX0079 in nuclease-free water (or approximately 5.3 x 1010molecules) were added to one volume of Quantigene cell homogenization solution (ThermoFisher #QS010) along with 10 pL of the undiluted pre-phosphorylated internal standard pool. 42.170.173772 / 01

[0382] For all samples, ligation of 5’ adapters was performed in twice the usual volume (21 pL) of ligation mix to make sure all beads were fully covered. The subsequent steps of reverse transcription, cDNA amplification, QC and sequencing were performed as described in Example 2C-D. Raw sequencing data were analyzed as described in Example 2E and Example 7. To achieve absolute quantification, read counts were scaled by barcode-specific amplification efficiency factors and normalized against internal standards.

[0383] Results

[0384] The dose-response relationship for PP / B-targeting siRNA (DX0078) is shown in Figure 16A. The curve follows a sigmoidal shape and half-maximal gene knockdown (IC50) was achieved at an siRNA concentration of ~0.3 nM. Doses of 5 nM led to near-complete (>95%) knockdown, indicating saturation. In contrast, no saturation was observed for functional uptake levels even at doses as high as 50 nM (Figure 16B). This suggests that in A549 cells, the intracellular knockdown machinery saturates faster than the endocytotic system and total cytosolic load continues to increase with dose even when functional activity plateaus.

[0385] Total cellular uptake was found to behave similarly to cytosolic, 5’ phosphorylated analytes over the concentration range tested (Figure 17). Although significant variability was observed across biological replicates, these data confirm that functional uptake measurements are dose-dependent and the assay can hence be used to rank analytes by endosomal escape efficiency.

[0386] Example 12: Effect of hydroxychloroquine on intracellular barcode uptake

[0387] Chloroquine and its derivative, hydroxychloroquine, are lysosomotropic agents known for their ability to enhance endosomal escape of nucleic acid therapeutics (Desai et al., Molecules, 2024, 29(13), 3131 ; Gilleron et a!., Nucleic Acids Res., 2015, 43(16), p7984- 8001). For example, chloroquine treatment has been shown to lower the IC50 of cholesterol- conjugated siRNAs in HeLa cells, and the observed knockdown improvement has been associated with increased membrane damage to endolysosomal vesicles (Du Rietz et al., Nat. Commun., 2020, 11, 1809), indicative of vesicle rupture. Here, we employed hydroxychloroquine co-treatment as a positive control to validate if phosphorylation status is a suitable proxy for endosomal escape of gymnotically administered oligonucleotides.

[0388] Gymnotlc uptake and hydroxychloroquine treatment

[0389] Human A549 cells (ATCC, CCL-185™) cultured in DMEM supplemented with 10% FBS were seeded in 6-well plates at a density of 350k cells / well. After 24 hours (1 day post- 42.170.173772 / 01 seeding), cholesterol-conjugated siRNA or single-stranded analytes (see Table 15A, 15B and 15C) were added at a final concentration of 1 pM in 1.5 mL / well. Cells left untreated were included as a control. Half of the samples further received hydroxychloroquine (CQ) at a final concentration of 100 pM. All conditions were done in triplicate and incubated for 24h at 37°C.

[0390] Table 15A: Analyte strands used in Example 12. Lower-case a / c / g / u indicates 2’F (2’-fluoro), mA / mC / mG / mll indicates 2’OMe (2’-methoxy), / 3CholTEG / indicates a 3’ cholesteroltriethyleneglycol (TEG) modifier, and * indicates PS (phosphorothioate).

[0391] Table 15B: Cholesterol-conjugated siRNA used in Example 12, targeting PPIB.

[0392] Table 15C: Analytes for Example 5.

[0393] After 24 hours of incubation with the respective analytes in the presence or absence of CQ (48 hours post-seeding), media from all wells was aspirated, cells were washed once with 1 mL PBS and detached by incubation with 500 pLTrypLE (Gibco #12605010) for 5 min at 37°C. Trypsinization was halted by adding 2 mL of DMEM / 10% FBS per well, triplicates were pooled and cells were counted with a Trypan blue stain 0.4% (Invitrogen #T10282) on a Countess® II FL Automated Cell Counter (Life Sciences) as per manufacturer’s instructions. 42.170.173772 / 01

[0394] Cell lysis and one-step qPCR

[0395] Of the samples treated with DX0079 (and untreated controls), equal cell numbers per condition (32K cells) were aliquoted, pelleted by centrifugation at 300 ref for 4 minutes and snap frozen in liquid nitrogen. Snap frozen cell pellets were processed using the Luna® Cell Ready One-Step RT-qPCR Kit (NEB #E3030S) as per manufacturer’s instructions. Briefly, cell pellets were resuspended in Cell Resuspension Solution (1 :25 dilution of Luna Cell Ready RNA Protection Reagent in cold 1X PBS) and lysed for 10 min at 37°C in Cell Lysis Mix supplemented with DNAse I (included in kit NEB #E3030S). Lysis was stopped by adding 10X Luna Cell Ready Stop Solution, the lysates were incubated for 5 min at 25°C and quantitative PCR was performed in a 15 pL reaction volume comprising Luna Universal One- Step Reaction Mix, Luna WarmStart RT Enzyme Mix, forward and reverse primers (0.4 pM final concentration, see Table 14) and 1:10 diluted cell lysate as template. The following cycling conditions were used on a QuantStudio™ 5 Real-Time PCR system: 10 min at 55°C (1 cycle), 1 min at 95°C (1 cycle), 45 cycles of 10 sec at 95°C followed by 30 sec at 60°C.

[0396] Cell lysis and Bulk RNA seq library preparation

[0397] Of samples treated with analyte 14 (and untreated controls), equal cell numbers per condition (684K cells) were aliquoted, pelleted by centrifugation at 300 ref for 4 minutes and snap frozen in liquid nitrogen. Snap frozen cell pellets were lysed using the QuantiGene™ Sample Processing Kit for cultured cells (ThermoFisher #QS010) supplemented with Proteinase K (included in same kit, QuantiGene™ Sample Processing Kit for cultured cells, ThermoFisher #QS010), as described in Example 5 above.

[0398] Once viscosity of the solution was similar to the Lysis Mixture, the cell lysates were processed for Bulk RNA seq library preparation as outlined in Example 2 with slight modifications as detailed below. To enable the detection of functional uptake (endogenously phosphorylated barcodes only) and total cellular uptake (endogenously phosphorylated and unphosphorylated barcodes) simultaneously, the samples were split into 2 groups as described in Example 11.

[0399] In group 1 (for functional uptake detection), lysates were spiked with 10 pL of a 1 :100 dilution of a pre-phosphorylated internal standard pool in nuclease-free water prior to hybridization capture.

[0400] In group 2 (for total cellular uptake detection), lysates were spiked with 10 pL of the undiluted pre-phosphorylated internal standard pool instead so that final spike-in amounts 42.170.173772 / 01 are 100x higher than in group 1 , and the samples were incubated with 10 II of T4 Polynucleotide Kinase (New England Biolabs, #M0201 L) for 30 minutes at 37 °C prior to adapter ligation. As a technical positive control, 10 pL of an 8.8 nM solution of SEQ. ID 14 in nuclease-free water were added to one volume of Quantigene cell homogenization solution (ThermoFisher #QS010) along with 10 pL of the undiluted pre-phosphorylated internal standard pool.

[0401] In all samples from both groups, ligation of 5’ adapters was performed in twice the usual volume (21 pL) of ligation mix to make sure all beads were fully covered and could be resuspended more easily. To prevent adapter bias, the same adapter (Table 4, Seq ID 249) was used across all conditions and samples were not pooled post-ligation. The subsequent steps of reverse transcription, cDNA amplification, QC and sequencing were performed as described in Example 2C-D. Raw sequencing data were analyzed as described in Example 2E and Example 7. To achieve absolute quantification, read counts were scaled by barcodespecific amplification efficiency factors and normalized against internal standards.

[0402] Results

[0403] Relative gene expression results as measured by quantitative PCR are shown in Figure 18A. In line with existing literature, hydroxychloroquine addition (CQ+) to cells incubated with cholesterol-conjugated siRNA improved target gene knockdown by approximately 30% as compared to controls without hydroxychloroquine (CQ-). Chloroquine addition to cells in the absence of siRNA did not affect target gene expression, indicating that the observed KD improvement in the presence of siRNA is siRNA-specific and not caused by chloroquine itself, likely due to enhanced endosomal escape of siRNA.

[0404] Results from bulk RNA sequencing for the detection of in cellulo phosphorylated barcodes as a proxy for functional uptake in the presence or absence of hydroxychloroquine are shown in Figure 18B. As hypothesized, hydroxychloroquine addition increased functional barcode uptake, with approximately 320% more barcodes detected in CQ+ as opposed to CQ- samples. A similar trend was also observed for total cellular barcode uptake, which reflects both phosphorylated and unphosphorylated barcodes (including barcodes trapped in endolysosomal vesicles). This may reflect the effect of CQ on analyte uptake or retention. Therefore, the observed increase in functional barcode uptake cannot confidently be attributed to an increase in endosomal escape levels, but rather seems to be correlated with an increase in total cellular uptake. 42.170.173772 / 01

[0405] Example 13: Time course of endosomal escape

[0406] Having found that 5’ phosphorylated analyte detection scaled with dose in a manner similar to total cellular uptake levels (Example 11), and hydroxychloroquine treatment increased cytosolic analyte signal concomitant with total cellular uptake (Example 12), it was important to test if the two readouts were, in fact, interdependent. While analyte internalization usually occurs more or less instantly upon lipoplex-mediated transfection (Hirsch etal., Nucleic Acids Res., 2015, 43(9), p4650-4660), endosomal release is thought to proceed gradually through discrete release events (Hedlund etal., Nat. Commun., 2023, 14, 1075). To compare the kinetics of barcode uptake and endosomal escape, and to validate if 5’ phosphorylation levels can serve as a proxy for endosomal escape, analyte uptake and subcellular distribution were traced over multiple time points post-transfection.

[0407] Forward transfection of A 549 cells and sequencing

[0408] Three separately cultured batches of A549 cells (biological replicates) were seeded in 6-well plates at a density of 250k cells / well. On the next day, cells were transfected with doublestranded analyte DX0001 at a final concentration of 10 nM using Lipofectamine RNAiMAX® transfection reagent (Invitrogen #13778150) as per manufacturer’s instructions. Untransfected cells were included as a control. Cells were harvested at the indicated time points (1 min, 20 min, 1 h, 3 h, 6 h, 24 h) by washing cells twice with PBS and adding Lysis Solution (QuantiGene™ Sample Processing Kit for cultured cells [ThermoFisher #QS010]). Plates were sealed and frozen at -20°C until downstream processing, which involved incubation at at 54°C for 45 minutes in the presence of Proteinase K (ThermoFisher #QS010) for complete cell lysis.

[0409] Once viscosity of the solution was similar to the Lysis Mixture, the cell lysates were processed for Bulk RNA seq library preparation as described in Example 11, with half of the lysate used for the detection of 5’ phosphorylated analytes (functional uptake only), and the other half used for the detection of total cellular uptake.

[0410] Raw sequencing data were analyzed as described in Example 2E and Example 7. To achieve absolute quantification, read counts were scaled by barcode-specific amplification efficiency factors and normalized against internal standards. 42.170.173772 / 01

[0411] Results

[0412] Cellular internalization of double-stranded analyte DX0001 occurred almost instantaneously, with high levels of barcode uptake detected already at the 1 min time point and concentrations plateauing within the first hour of transfection (Figure 19, top panel). In contrast, 5’ phosphorylation was barely detectable for the first 60 minutes and peaked only 6 hours after transfection. At this time point, between 1-4% of transfected material was present in 5’ phosphorylated form (Figure 19, bottom panel), representing the fraction of analyte that successfully escaped the endosome. These numbers are in line with existing literature estimates (Gilleron et al., Nat. Biotechnol., 2013, 31, p638-646; Wittrup et al., Nat.

[0413] Biotechnol., 2015, 33, p870-876) and confirm that 5’ phosphorylation levels can be used as reliable indicators of subcellular delivery into the cytosol.

[0414] Saturation curve analysis confirmed that the samples were sequenced to a saturating depth.

[0415] Example 14: Correlation between 5’ phosphorylation and knockdown activity

[0416] Conjugation of 2'-O-hexadecyl (C16) to siRNAs has previously been shown to enable safe, potent and durable silencing in the central nervous system of rodents and non-human primates (Brown et al., Nat. Biotechnol., 2022, 40, p1500-1508). Notably, knockdown activity was highest in the spinal cord (near the site of injection) and gradually decreased across outer and inner brain regions. To test if 5’ phosphorylation levels follow a similar gradient and correlate with activity, rats were administered with C16-conjugated siRNAs or barcoded C16-conjugated siRNA mimics and silencing levels were compared against 5’ phosphorylated analyte levels between lumbar spinal cord and brainstem tissues.

[0417] C16-conjugated siRNA targeting Sod1 and barcoded siRNA mimic were prepared by solidphase synthesis as described previously (Brown et al., Nat. Biotechnol., 2022, 40, p1500- 1508). Sequences and duplex formulations of the tested analytes are provided in Table 16. Intrathecal administrations were performed as detailed in Example 4A. DX0080 was administered at a single dose of 20 nmol (-300 mg) for activity readout and animals (n=6) were sacrificed 28 days post-dosing. DX0081 was administered at a single dose of 1 nmol (-6.7 mg) and animals (n=5) were sacrificed 7 days post-dosing. 42.170.173772 / 01

[0418] Table 16A: Analyte strands used in Example 14. (C16u) indicates C16-modified uridine. Lower-case a / c / g / u indicates 2’F (2’-fluoro), mA / mC / mG / mll indicates 2’OMe (2’-methoxy), / 5VP indicates 5'-(E)-vinylphosphonate and * indicates PS (phosphorothioate).

[0419] Table 16B: siRNA and double-stranded analyte used in Example 14.

[0420] Lumbar spinal cord and brainstem tissues were processed as described in Example 4B. Knockdown measurements were performed by branched DNA assay following the method described in Coles et al., Nucleic Acid Ther., 2016, 26(2), p86-92. Bulk RNA seq library preparation and sequencing was performed as described in Example 2. Prior to hybridization capture, lysates were spiked with 10 pL of a 1 :100 dilution of a pre-phosphorylated internal standard pool in nuclease-free water to enable absolute quantification.

[0421] Results

[0422] As expected based on previous literature, C16-conjugated Sod1 siRNA (DX0080) achieved higher gene knockdown in lumbar spinal cord than in brainstem (Figure 20, light grey bars). Similarly, more 5’ phosphorylated analytes derived from C16-conjugated siRNA mimic (DX0081) were detected in lumbar spinal cord as compared to brainstem (Figure 20, dark grey bars), indicating higher levels of endosomal escape in the former. This confirms that 5’ phosphorylation levels do correlate with activity across tissues in vivo and can be used to identify regions with higher functional delivery. 42.170.173772 / 01

[0423] Example 15: In vivo screening

[0424] Following validation for the 5’ phosphorylation detection method (Examples 3-14), a plurality of barcoded analytes (selected from those in Table 8A) conjugated to differently modified RNA duplexes as cargo (Table 8B) was administered intrathecally in rats. The aim was to identify if the analytes were able to efficiently escape from endosomes.

[0425] The protocol for preparing cargo-conjugated analytes is described in Example 1C.

[0426] Intrathecal administration was performed as detailed in Example 4A. Each animal received a total dose of 13.4 nmol of co-formulated conjugate mixture, corresponding to a dose of 0.96 nmol per individual conjugate. Vehicles injected with aCSF served as control. Rats (n=5) were sacrificed 7 days post-dosing and brainstem tissues were homogenized as described in Example 4B. Bulk RNA seq library preparation and sequencing were performed according to Example 2, with half of the lysates processed separately for total cellular uptake detection as explained in Example 11.

[0427] Raw sequencing data were analyzed as described in Example 2E and Example 7. To achieve absolute quantification, read counts were scaled by barcode-specific amplification efficiency factors and normalized against internal standards.

[0428] Results

[0429] Figure 21 shows the screening results ranked by 5’ phosphorylated analyte abundance as detected by bulk RNA sequencing. CC00006 was the most efficient at escaping endosomes. Together, these results demonstrate the feasibility of identifying molecules with potent endosomal escape capability in a multiplexed, high-throughput fashion directly in vivo.

[0430] Example 16: Detection of analytes with 2 cargo molecules

[0431] To test if the detection of 5’ monophosphorylated analytes is affected by the presence of multiple cargo molecules, conjugates comprising either two siRNA-mimicking barcoded analytes attached to a cargo scaffold (CC00016) or one siRNA-mimicking barcoded analyte attached to a dual-cargo, siRNA-conjugated scaffold (CC00017) were synthesized (Table 17A). The protocol for preparing cargo-conjugated analytes is described in Example 1C. 42.170.173772 / 01

[0432] A) Intratracheal instillation of cargo-conjugated analytes in mice

[0433] Female BALB / c mice (6-8 weeks old) were positioned supine, and a flexible, blunt-ended cannula attached to a syringe was gently inserted through the mouth into the trachea, confirmed by smooth passage and lack of resistance. The test substance (40 pL, 10 nmol of CC00016 or CC00017) was then instilled into the trachea. Vehicles instilled with PBS served as control. After instillation, the cannula was withdrawn, and the mice were observed. Animal weight was monitored until sacrifice. Sacrifice was performed 14 days later by administration of a lethal dose of pentobarbitone and exsanguinated by the abdominal aorta, to ensure that no blood clots were formed in the lungs. Lungs were collected, stored separately in RNAIater (Sigma, #R0901) at 4°C overnight and then stored at -70°C until tissue homogenization.

[0434] Table 17A: Library of cargo-conjugated analytes for Example 15.

[0435] 42.170.173772 / 01

[0436] Table 17B: Sequences of RNA strands used as cargo covalently attached to the analytes shown in Table 17A. 5’VP indicates 5'-(E)-vinylphosphonate.

[0437] B) Tissue homogenization, library preparation and sequencing

[0438] Whole right lung lobes were homogenized as described in Example 4B. Bulk RNA seq library preparation and sequencing was performed according to Example 2. Lysates were spiked with 10 pL of a 1 :100 dilution of a pre-phosphorylated internal standard pool in nuclease-free water prior to hybridization capture.

[0439] Raw sequencing data were analyzed as described in Example 2E and Example 7. To achieve absolute quantification, read counts were scaled by barcode-specific amplification efficiency factors and normalized against internal standards.

[0440] Results

[0441] Successful library preparation was confirmed by Tapestation (Figure 22A). 5’ monophosphorylated analytes of both conjugates (CC00016 and CC00017), representing the fraction of endosomally escaped molecules that had interacted with an endogenous kinase in the target subcellular location, were detected at similar levels (Figure 22B). In conclusion, the presence of multiple cargo molecules had no effect on analyte detection.

Claims

42. 170.173772 / 01Claims1. A method of determining whether one or more nucleic acid analyte administered to a cell is transported to a target subcellular location comprising the steps of: a) administering one or more nucleic acid analyte to said cell, each analyte comprising a nucleotide sequence barcode unique to said analyte; b) collecting said cell or a portion thereof and assessing whether one or more barcode of the administered one or more analyte is, or was, present, in the target subcellular location(s), thereby determining which analyte was transported to the target subcellular location(s), wherein said assessment is made by: i) collection of said cell or a portion thereof comprising said target subcellular location and identifying if one or more of said barcode is present in said location or, ii) determining whether one or more of said analyte comprising a barcode has interacted with an entity in said subcellular location.

2. The method as claimed in claim 1, wherein said analyte comprises a nucleic acid molecule with less than 200 nucleotides, preferably less than 150 nucleotides^3. The method as claimed in claim 1 or 2, wherein said cell is in a subject or a sample, preferably a mammalian subject.

4. The method as claimed in any one of claims 1 to 3, wherein neither the analyte nor the barcode is encapsulated in a nanoparticle.

5. The method as claimed in any one of claims 1 to 3, wherein the analyte and / or the barcode is encapsulated in a nanoparticle.

6. The method as claimed in any one of claims 1 to 5, wherein the assessment of whether the barcode is, or was, present in the target subcellular location is achieved by sequencing the barcode.

7. The method as claimed in claim 6, wherein the barcode is a single or double stranded DNA or RNA molecule.

8. The method as claimed in any one of claims 1 to 7, wherein the analyte and / or barcode comprises nucleotides which have been modified, preferably one or more modified42.170.173772 / 01 nucleotide selected from 2’F, 2’0Me, 2’MOE, a locked nucleotide or a phosphorothioate backbone modification.

9. The method as claimed in any one of claims 1 to 8, wherein the analyte and / or barcode is a mRNA, siRNA or antisense oligonucleotide (ASO), saRNA, miRNA, aptamer or a mimic thereof.

10. The method as claimed in any one of claims 1 to 9, wherein the analyte additionally comprises one or more of: i) a universal recognition moiety which is invariant in the one or more analyte; ii) a moiety aiding interaction with said entity in said subcellular location in the cell; iii) a targeting moiety aiding targeting to a particular location in the cell; and / or vi) a second cargo molecule.

11. The method as claimed in any one of claims 1 to 10, wherein the analyte is a single or double stranded nucleic acid molecule.

12. The method as claimed in any one of claims 1 to 11, wherein said analyte consists of said barcode or comprises said barcode with a flanking 3’ and / or 5’ sequence.

13. The method as claimed in any one of claims 1 to 12, wherein more than one analyte is administered to said cell, and said method determines which analyte is transported to a target subcellular location.

14. The method as claimed in any one of claims 1 to 13, wherein said subcellular location is the endosome, lysosome, cytoplasm or the nucleus.

15. The method as claimed in claim 14, wherein said assessment is made by determining whether one or more of said analyte has interacted with a phosphorylating enzyme, e.g. a kinase, in the target subcellular compartment (preferably the cytoplasm and / or nucleus) comprising the additional steps of: c) collecting at least the one or more analyte which has been phosphorylated; d) determining the barcode of each analyte that has been phosphorylated to determine which analyte was transported to the target subcellular compartment (preferably the cytoplasm and / or nucleus).

16. The method as claimed in claim 15, wherein the kinase is Clp1.42.170.173772 / 0117. The method as claimed in any one of claims 1 to 14, wherein said assessment comprises isolation of said nucleus.

18. The method as claimed in claim 17, wherein the barcodes are RNA molecules that are identified by single-nucleus RNA sequencing.

19. The method as claimed in claim 17 or 18, wherein the analyte additionally comprises a nuclear targeting moiety, preferably a nuclear localization sequence or peptide.

20. The method as claimed in any one of claims 1 to 14, wherein said assessment comprises determining whether one or more of said analyte comprising a barcode has interacted with an entity in said subcellular location.

21. The method as claimed in claim 20, wherein the entity is a kinase.

22. The method as claimed in claim 20, wherein the entity is selected from an RNA- binding protein, a protein, an enzyme, a ligand or a receptor, wherein said entity is preferably not a polymerase.

23. The method as claimed in any one of claims 1 to 21 , wherein said method comprises the steps: a) administering one or more nucleic acid analyte to said cell, each analyte comprising a nucleotide sequence barcode unique to said analyte and a universal recognition sequence; b) collecting said cell or a portion thereof; c) collecting at least the one or more analyte, which has been phosphorylated by interaction with a phosphorylating enzyme, e.g. a kinase, in said target subcellular compartment (preferably the cytoplasm and / or nucleus), by capture to a solid support carrying a sequence complementary to the universal recognition sequence; d) contacting said one or more analyte with a molecule which recognises a phosphorylated analyte but not a non-phosphorylated analyte; e) optionally amplifying analytes which covalently attach to said molecule which recognises a phosphorylated analyte but not a non-phosphorylated analyte; f) determining the barcode of each analyte that has been phosphorylated to determine which analyte was transported to the target subcellular compartment.42.170.173772 / 0124. The method as claimed in claim 23, wherein: in step c) the cell is lysed or permeabilized; in step d) the molecule which recognises a phosphorylated analyte is an adapter which is covalently linked to said phosphorylated analyte by virtue of a ligase; in step e) the amplification is performed by reverse transcription followed by amplification; and / or in step f) the determination is performed by sequencing.

25. The method as claimed in any one of claims 1 to 21 , wherein said method comprises: a) administering one or more nucleic acid analyte to said cell, each analyte comprising a nucleotide sequence barcode unique to said analyte and a universal recognition sequence; b) i) collecting said cell or a portion thereof; ii) collecting the one or more analyte that has been taken up by said cell; iii) dividing said collected one or more analyte into at least two aliquots; iv) contacting at least one aliquot with an exogenous phosphorylating enzyme, e.g. kinase, to allow phosphorylation of said one or more analyte; and c) collecting from a first aliquot to which no exogenous phosphorylating enzyme has been added at least the one or more analyte, which has been phosphorylated by interaction with an endogenous phosphorylating enzyme, e.g. a kinase, in said target subcellular compartment (e.g. the cytoplasm and / or nucleus), and separately collecting from a second aliquot to which said exogenous phosphorylating enzyme has been added at least the one or more analyte, which has been phosphorylated by interaction with said exogenous phosphorylating enzyme, e.g. kinase, or an endogenous phosphorylating enzyme, e.g. kinase, by capture to a solid support carrying a sequence complementary to the universal recognition sequence, and subjecting each aliquot to steps d) and f) and optionally e) as set forth in claim 23 or 24.

26. The method as claimed in any one of claims 1 to 25, wherein said method is performed in vitro or is a non-therapeutic and / or non-surgical method performed in vivo, preferably on non-human animals.

27. One or more analyte as defined in any one of claims 1 , 2, 4-12 or 19 for use as a diagnostic to determine whether one or more analyte administered to a subject is transported to a target subcellular location within said cell.42.170.173772 / 0128. A method of determining whether one or more nucleic acid analyte comprising a nucleotide sequence barcode unique to said nucleic acid analyte is, or was, present in a target subcellular location in a cell, comprising a) obtaining a sample containing said cell to which said one or more analyte has been administered; b) performing an assessment as set out in step b) in any one of claims 1 to 26.

29. The method as claimed in claim 28, wherein said one or more analyte is as defined in any one of claims 1 , 2, 4-12 or 19, and / or said cell is as defined in claim 3, and / or said subcellular location is as defined in claim 14, and / or said assessment is as defined in any one of claims 15-18 or 20-26.

30. A method of determining whether one or more nucleic acid analyte has entered a cell and been released from the endosome and / or lysosome by transporting said one or more analyte to the nucleus for detection, wherein said analyte is as defined in any one of claims 1 , 2, 4-12 or 19, comprising the steps of: a) administering said one or more analyte to said cell or obtaining a sample containing said cell to which said one or more analyte has been administered; b) collecting said cell or a portion thereof, isolating said nucleus and assessing whether one or more barcode is present in said nucleus and thereby determining which of the one or more analyte entered said cell and was released from the endosome and / or lysosome.

31. The method as claimed in claim 30, wherein said cell is as defined in claim 3, and / or said assessment is as defined in claim 18.

32. A method for detecting the presence of one or more nucleic acid analyte in a target subcellular location after administration to a cell, comprising the steps of: a) administering one or more nucleic acid analyte to said cell, each analyte comprising a nucleotide sequence barcode unique to said analyte as defined in any one of claims 1 , 2, 4-12 or 19; b) collecting said cell or a portion thereof and assessing whether one or more barcode of the administered one or more analyte is, or was, present in the target subcellular location, thereby determining which analyte is, or was, present in the target subcellular location, wherein said assessment is made by:42.170.173772 / 01 i) collection of said cell or a portion thereof comprising said target subcellular location and identifying if one or more of said barcode is present in said location, or ii) determining whether one or more of said analyte has interacted with an entity in said subcellular location.

33. The method as claimed in claim 32, wherein said one or more analyte is as defined in any one of claims 1 , 2, 4-12 or 19, and / or said cell is as defined in claim 3, and / or said subcellular location is as defined in claim 14, and / or said assessment is as defined in any one of claims 15-18, or 20 to 24.

34. The method or use as claimed in any one of claims 1 to 33, wherein in the step(s) in which the one or more barcode of the analyte is determined, one or more synthetic internal standard analyte is included as a control.

35. A plurality of nucleic acid analytes, wherein each analyte: a) is a nucleic acid molecule with less than 500 nucleotides; b) comprises a different barcode unique to said analyte, wherein said barcode sequence is from 4 to 20 nucleotides in length; c) comprises an invariant universal recognition sequence; and optionally one or more of: i) a 5’OH or 3’OH group amenable to phosphorylation; ii) the first two 5’ nucleotides of the analyte adopt an A-form helical conformation with the ribose in the C3’ endo orientation; iii) a synthetic 5’-m3G cap; and iv) a nuclear targeting moiety, preferably a nuclear localization peptide or a sequence selected from AGCCC, RCCTCCC or CTG repeats.

36. The plurality of nucleic acid analytes as claimed in claim 35, wherein said plurality of analytes contains one or more synthetic internal standard analytes as controls, wherein optionally said standard analytes are provided separately.

37. The plurality of analytes as claimed in claim 35 or 36, wherein said plurality of analytes includes at least one nucleic acid analyte as set forth in Table 1.

38. The plurality of analytes as claimed in any one of claims 35 to 37, wherein each analyte additionally comprises one or more of: i) a moiety aiding interaction with an entity present in a subcellular location; and / or42.170.173772 / 01 ii) a targeting moiety aiding targeting to a particular location in a cell.

39. The plurality of analytes as claimed in any one of claims 35 to 38 wherein said analyte is as defined in any one of claims 1 , 2, 4-12 or 19.

Images