This application claims the benefit of European Patent Applications EP20171439.1, filed 24 April 2020, and EP20185061.7, filed 09 July 2020, all of which are incorporated herein by reference.

The present invention relates to a diagnostic method for cancer based on an ADP-ribosylation signal in an isolated sample of tissue of the patient. The present invention further relates to a method for detection of ADP-ribosylation in a cell and to a polypeptide, which binds to ADP- ribosylation.

Background of the Invention

The inventors discovered that ADP-ribosylation is prognostic marker in cancer diagnosis. The present application shows that this marker is useful in several different cancer models. The data are gained from several hundreds of patients, which makes them statistically more relevant than many other studies.

The detection of ADP-ribosylation is performed with immunohistochemistry on histological tissue slides. The pipeline for cancer diagnosis, nearly always goes via a histological analysis performed by a pathologist who scores the tissue biomarker stained by immunohistochemistry. Although protein ADP-ribosylation was first described in the early 1960s, ADP-ribosylation was traditionally studied and identified in vitro via the incorporation of radioactive ADPr or ADPr-analogs. For a long time, only antibodies recognizing poly-ADP-ribosylation (PAR) were available, which restricted the ability to detect only PAR events by immunoblotting or immunofluorescence. Only recently the use of ADP-ribose binding domains, like the macro domain of the Archaeoglobus fulgidus Af1521 fused to an Fc-fragment or anti ADP-ribose antibody that can detect mono-ADP-ribosylation, became available (Gibson, B.A. et ai.

(2017), Biochemistry, 56, 6305-6316.)

The inventors’ methodology offers the possibility to the pathologist to directly observe and then score the ADPR signal in the tissue biopsies. Moreover, the methodology can colocalize the ADPR signal with other biomarkers by immunofluorescence-multiplexing, which is the new frontier of cancer diagnosis (with detection of currently up to 7 biomarkers at the same time on a tissue biopsy, saving time and patient tissues).

Summary of the invention

A first aspect of the invention relates to a method for diagnosis of cancer in a patient, or a method of determining the prognosis of a cancer patient, or a method of assigning a patient to an outcome group, or a method of assigning a patient to a treatment regimen, said method comprising the steps of a. providing an isolated tissue sample of said patient, wherein the isolated tissue sample comprises a plurality of cells; b. fixating said plurality of cells, thereby permeabilizing the cells’ membrane; c. contacting, i.e. staining, said cells with a polypeptide ADPR binder, wherein the ADPR binder is capable to specifically and selectively non-covalently bind ADP-ribosylated biomolecules, particularly wherein the biomolecules are peptides or polypeptides or nucleic acid molecules, more particularly wherein the biomolecules are polypeptides, and wherein the ADPR binder comprises a detectable label, or a binding moiety allowing specific labelling of the ADPR binder with a second binder, which comprises a detectable label; and optionally, washing off excess first ADPR binder and contacting the sample with the second binder; d. detecting the amount and location of ADP-ribosylated biomolecules inside the isolated tissue sample; e. Assigning, as a function of where and/or how much ADPR binder is detected in the isolated tissue sample o a likelihood of having or developing cancer to said patient, or o assigning a likelihood of prognosis to said patient; o assigning the patient to an outcome group or o assigning the patient to treatment with an anticancer treatment.

A second aspect of the invention relates to a method for detection of ADP-ribosylated peptides or polypeptides in a cell.

In one alternative of this second aspect, the method comprises the steps of a. providing an isolated tissue sample of said patient, wherein the isolated tissue sample comprises a plurality of cells; b. permeabilizing the cells’ membrane; c. in a labelling step, contacting said cells with an ADPR binder, wherein the ADPR binder is capable to specifically bind ADP-ribosylated biomolecules; d. in a detection step, detecting the amount and location of ADP-ribosylated biomolecules inside the isolated tissue sample.

In another alternative of this second aspect, the method comprises the steps of a. providing a cell from an isolated tissue sample or from cell culture; b. contacting said cell with an ADPR binder, wherein said ADPR binder comprises SEQ ID NO 001 or said ADPR binder comprises an ADP-ribosyl-binding sequence having ³90%, particularly >92%, or ³94%, more particularly ³96%, even more particularly ³98%, most particularly ³99% identity to SEQ ID NO 001, wherein said ADP-ribosyl-binding sequence contains the residues Glu35 and Arg145, and wherein said ADP-ribosyl- binding sequence has at least 85% (particularly 90%, 92%, 94%, 96% most particularly ³98%) of the ADP-ribosyl-binding activity of SEQ ID NO 001 ; c. detecting ADP-ribosylated peptides or polypeptides.

A third aspect of the invention relates to a polypeptide comprising a. a polypeptide sequence of SEQ ID NO 001 or a polypeptide sequence having >90%, particularly ³92%, more particularly >94%, even more particularly >96%, more particularly >98%, most particularly ³99% identity to SEQ ID NO 001 and containing the residues Glu35 and Arg145, and wherein said polypeptide sequence has at least 85% (particularly 90%, 92%, 94%, 96% most particularly ³98%) of the ADP-ribosyl-binding activity of SEQ ID NO 001, and b. a detectable label, particularly a detectable label selected from a tag sequence, a fluorescent or luminescent protein moiety and a fluorescent dye.

A fourth aspect of the invention relates to a system to carry out the method of any one of aspects 1 or 2, particularly a system for determining the content of ADP-ribosylated biomolecules in a sample obtained from a patient.

Detailed Description of the Invention

Terms and definitions

For purposes of interpreting this specification, the following definitions will apply and whenever appropriate, terms used in the singular will also include the plural and vice versa. In the event that any definition set forth below conflicts with any document incorporated herein by reference, the definition set forth shall control.

The terms “comprising,” “having,” “containing,” and “including,” and other similar forms, and grammatical equivalents thereof, as used herein, are intended to be equivalent in meaning and to be open ended in that an item or items following any one of these words is not meant to be an exhaustive listing of such item or items, or meant to be limited to only the listed item or items. For example, an article “comprising” components A, B, and C can consist of (i.e., contain only) components A, B, and C, or can contain not only components A, B, and C but also one or more other components. As such, it is intended and understood that “comprises” and similar forms thereof, and grammatical equivalents thereof, include disclosure of embodiments of “consisting essentially of’ or “consisting of.”

Where a range of values is provided, it is understood that each intervening value, to the tenth of the unit of the lower limit, unless the context clearly dictate otherwise, between the upper and lower limit of that range and any other stated or intervening value in that stated range, is encompassed within the disclosure, subject to any specifically excluded limit in the stated range. Where the stated range includes one or both of the limits, ranges excluding either or both of those included limits are also included in the disclosure.

Reference to “about” a value or parameter herein includes (and describes) variations that are directed to that value or parameter per se. For example, description referring to “about X” includes description of “X.”

As used herein, including in the appended claims, the singular forms “a,” “or,” and “the” include plural referents unless the context clearly dictates otherwise.

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art (e.g., in cell culture, molecular genetics, nucleic acid chemistry, hybridization techniques and biochemistry). Standard techniques are used for molecular, genetic and biochemical methods (see generally, Sambrook et al., Molecular Cloning: A Laboratory Manual, 4th ed. (2012) Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. and Ausubel et al., Short Protocols in Molecular Biology (2002) 5th Ed, John Wiley & Sons, Inc.) and chemical methods.

The term polypeptide in the context of the present specification relates to a molecule consisting of 50 or more amino acids that form a linear chain wherein the amino acids are connected by peptide bonds. The amino acid sequence of a polypeptide may represent the amino acid sequence of a whole (as found physiologically) protein or fragments thereof. The term "polypeptides" and "protein" are used interchangeably herein and include proteins and fragments thereof. Polypeptides are disclosed herein as amino acid residue sequences.

The term peptide in the context of the present specification relates to a molecule consisting of up to 50 amino acids, in particular 8 to 30 amino acids, more particularly 8 to 15amino acids, that form a linear chain wherein the amino acids are connected by peptide bonds.

Amino acid residue sequences are given from amino to carboxyl terminus. Capital letters for sequence positions refer to L-amino acids in the one-letter code (Stryer, Biochemistry, 3 ^rd ed. p. 21). Lower case letters for amino acid sequence positions refer to the corresponding D- or (2R)-amino acids. Sequences are written left to right in the direction from the amino to the carboxy terminus. In accordance with standard nomenclature, amino acid residue sequences are denominated by either a three letter or a single letter code as indicated as follows: Alanine (Ala, A), Arginine (Arg, R), Asparagine (Asn, N), Aspartic Acid (Asp, D), Cysteine (Cys, C), Glutamine (Gin, Q), Glutamic Acid (Glu, E), Glycine (Gly, G), Histidine (His, H), Isoleucine (lie, I), Leucine (Leu, L), Lysine (Lys, K), Methionine (Met, M), Phenylalanine (Phe, F), Proline (Pro, P), Serine (Ser, S), Threonine (Thr, T), Tryptophan (Trp, W), Tyrosine (Tyr, Y), and Valine (Val, V).

The term variant refers to a polypeptide that differs from a reference polypeptide, but retains essential properties. A typical variant of a polypeptide differs in its primary amino acid sequence from another, reference polypeptide. Generally, differences are limited so that the sequences of the reference polypeptide and the variant are closely similar overall and, in many regions, identical. A variant and reference polypeptide may differ in amino acid sequence by one or more modifications (e.g., substitutions, additions, and/or deletions). A substituted or inserted amino acid residue may or may not be one encoded by the genetic code. A variant of a polypeptide may be naturally occurring such as an allelic variant, or it may be a variant that is not known to occur naturally.

In the context of the present specification, the terms sequence identity and percentage of sequence identity refer to a single quantitative parameter representing the result of a sequence comparison determined by comparing two aligned sequences position by position. Methods for alignment of sequences for comparison are well-known in the art. Alignment of sequences for comparison may be conducted by the local homology algorithm of Smith and Waterman, Adv. Appl. Math. 2:482 (1981), by the global alignment algorithm of Needleman and Wunsch, J. Mol. Biol. 48:443 (1970), by the search for similarity method of Pearson and Lipman, Proc. Nat. Acad. Sci. 85:2444 (1988) or by computerized implementations of these algorithms, including, but not limited to: CLUSTAL, GAP, BESTFIT, BLAST, FASTA and TFASTA. Software for performing BLAST analyses is publicly available, e.g., through the National Center for Biotechnology-Information (http://blast.ncbi.nlm.nih.gov/).

One example for comparison of amino acid sequences is the BLASTP algorithm that uses the default settings: Expect threshold: 10; Word size: 3; Max matches in a query range: 0; Matrix: BLOSUM62; Gap Costs: Existence 11 , Extension 1 ; Compositional adjustments: Conditional compositional score matrix adjustment. One such example for comparison of nucleic acid sequences is the BLASTN algorithm that uses the default settings: Expect threshold: 10; Word size: 28; Max matches in a query range: 0; Match/Mismatch Scores: 1.-2; Gap costs: Linear. Unless stated otherwise, sequence identity values provided herein refer to the value obtained using the BLAST suite of programs (Altschul et al., J. Mol. Biol. 215:403-410 (1990)) using the above identified default parameters for protein and nucleic acid comparison, respectively.

Reference to identical sequences without specification of a percentage value implies 100% identical sequences (i.e. the same sequence). In the context of the present specification, the term antibody refers to whole antibodies including but not limited to immunoglobulin type G (IgG), type A (IgA), type D (IgD), type E (IgE) or type M (IgM), any antigen binding fragment or single chains thereof and related or derived constructs. A whole antibody is a glycoprotein comprising at least two heavy (H) chains and two light (L) chains inter-connected by disulfide bonds. Each heavy chain is comprised of a heavy chain variable region (VH) and a heavy chain constant region (CH). The heavy chain constant region of IgG is comprised of three domains, CH1 , CH2 and CH3. Each light chain is comprised of a light chain variable region (abbreviated herein as VL) and a light chain constant region (CL). The light chain constant region is comprised of one domain, CL. The variable regions of the heavy and light chains contain a binding domain that interacts with an antigen. The constant regions of the antibodies may mediate the binding of the immunoglobulin to host tissues or factors, including various cells of the immune system (e.g., effector cells) and the first component of the classical complement system. Similarly, the term encompasses a so- called nanobody or single domain antibody, an antibody fragment consisting of a single monomeric variable antibody domain.

In the context of the present specification, the term fragment crystallizable (Fc) region is used in its meaning known in the art of cell biology and immunology; it refers to a fraction of an antibody comprising, if applied to IgG, two identical heavy chain fragments comprised of a CH2 and a CH3 domain, covalently linked by disulfide bonds.

The term specific binding in the context of the present invention refers to a property of ligands that bind to their target with a certain affinity and target specificity. The affinity of such a ligand is indicated by the dissociation constant of the ligand. A specifically reactive ligand has a dissociation constant of £ 10 ⁷ mol/L when binding to its target, but a dissociation constant at least three orders of magnitude higher in its interaction with a molecule having a globally similar chemical composition as the target, but a different three-dimensional structure.

A polymer of a given group of monomers is a homopolymer (made up of a multiple of the same monomer); a copolymer of a given selection of monomers is a heteropolymer constituted by monomers of at least two of the group.

A first aspect of the invention relates to a method for diagnosis of cancer in a patient, or a method of determining the prognosis of a cancer patient, or a method of assigning a patient to an outcome group, or a method of assigning a patient to a treatment regimen. The method comprises the steps of a. providing an isolated tissue sample of said patient, wherein the isolated tissue sample comprises a plurality of cells; b. in a fixating step, fixating said plurality of cells, thereby permeabilizing the cells’ membrane; c. in a contacting step, contacting, i.e. staining, said cells with a polypeptide ADPR binder, wherein the ADPR binder is capable to specifically and selectively non-covalently bind ADP-ribosylated biomolecules, and wherein the ADPR binder comprises a detectable label, or a binding moiety allowing specific labelling of the ADPR binder with a second binder, which comprises a detectable label; and optionally, washing off excess first ADPR binder and contacting the sample with the second binder; d. in a detection step, detecting the amount and location of ADP-ribosylated biomolecules inside the isolated tissue sample; e. in an assigning step, assigning, as a function of where and/or how much ADPR binder is detected in the isolated tissue sample o a likelihood of having or developing cancer to said patient, or o assigning a likelihood of prognosis to said patient, or o assigning the patient to an outcome group or o assigning the patient to treatment with an anticancer treatment.

In certain embodiments, the ADPR binder is capable of binding mono-ADP-ribosylated biomolecules.

In certain embodiments, the ADPR binder does not specifically react to poly-ADP-ribosylated biomolecules.

While the ADPR binders described here have the capacity to bind both mono- and poly-ADP- ribosylation structures (MAR and PAR, respectively), it is their capacity to bind mono-ADP- ribosylated (MARylated) biomolecules which renders them superior to PAR-specific regents. Thus, it is the MAR-binding capacity which allows the prognostic analysis of histological patient samples. MAR and PAR structures are generated by different enzymes within the cells and these enzymes have different subcellular localization. The inventors found that it is the cytoplasmic MAR signal which correlates with different patient outcome data.

In certain embodiments, the ADP-ribosylated biomolecules are ADP-ribosylated peptides or ADP-ribosylated polypeptides or ADP-ribosylated nucleic acid molecules. In certain embodiments, the ADP-ribosylated biomolecules are ADP-ribosylated polypeptides.

In certain embodiments, o a high likelihood of having or developing cancer is assigned to said patient, or o a more severe prognosis is assigned to said patient; or o treatment with an anticancer treatment is assigned to said patient; if said cells in said isolated tissue sample of said patient are stained weakly with said ADPR binder in comparison to control tissue (healthy tissue of the same tissue origin) as judged by a blinded experienced pathologist.

In praxis, the isolated tissue sample is compared with healthy tissue on the same array. Quantification is performed by a trained pathologist or by an image-evaluation software.

In immunohistochemistry, a tiered approach (e.g., grades of 0, 1+, 2+, 3+; none, little, some, strong) is often used (Aeffner et al., (2017) Archives of Pathology & Laboratory Medicine, Vol. 141 , No. 9, pp. 1267-1275.). Immunohistochemical (IHC) assays performed on formalin-fixed paraffin-embedded (FFPE) tissue sections traditionally have been semi-quantified by pathologist visual scoring of staining. Rizzardi et al. (Rizzardi et al., Diagn Pathol 7, 42 (2012)) demonstrated that computer-aided methods to classify image areas of interest (e.g., carcinomatous areas of tissue specimens) and quantify IHC staining intensity within those areas can produce highly similar data to visual evaluation by a pathologist. Thus, a “weak” or “strong” staining is a clear distinction for an expert in pathology or for computer-aided evaluation methods.

A particular advantage of the method of the first aspect is that tumor cells can be identified independently of their cell cycle status and their proliferation activity. In addition, the method of the first aspect is a general classification method for the existence and severity of a tumor, independently of its origin. This is in contrast to the main markers used in histochemistry. Thus, the method of the first aspect is a useful addition to current standards or even a stand-alone tool.

Also, the method of the first aspect can be used if only a very small amount of cell material is available. Many other methods highly depend on the cells’ growth pattern, which is not needed for the method of the invention.

In certain embodiments, said ADPR binder comprises SEQ ID NO 001. In certain embodiments, said ADPR binder comprises an ADP-ribosyl-binding sequence having ³90% identity to SEQ ID NO 001 , wherein said ADP-ribosyl-binding sequence contains the residues Glu35 and Arg145. In certain embodiments, said ADPR binder comprises an ADP-ribosyl-binding sequence having ³92% identity to SEQ ID NO 001, wherein said ADP- ribosyl-binding sequence contains the residues Glu35 and Arg145. In certain embodiments, said ADPR binder comprises an ADP-ribosyl-binding sequence having ³94% identity to SEQ ID NO 001 , wherein said ADP-ribosyl-binding sequence contains the residues Glu35 and Arg145. In certain embodiments, said ADPR binder comprises an ADP-ribosyl-binding sequence having ³96% identity to SEQ ID NO 001, wherein said ADP-ribosyl-binding sequence contains the residues Glu35 and Arg145. In certain embodiments, said ADPR binder comprises an ADP-ribosyl-binding sequence having ³98% identity to SEQ ID NO 001, wherein said ADP-ribosyl-binding sequence contains the residues Glu35 and Arg145. In certain embodiments, said ADPR binder comprises an ADP-ribosyl-binding sequence having ³99% identity to SEQ ID NO 001 , wherein said ADP-ribosyl-binding sequence contains the residues Glu35 and Arg145.

In certain embodiments, said ADP-ribosyl-binding sequence additionally contains one residue selected from Arg15, Cys74, Leu97, Val103, Gly105, Gly110, and Asp162. In certain embodiments, said ADP-ribosyl-binding sequence additionally contains two residues selected from Arg15, Cys74, Leu97, Val103, Gly105, Gly110, and Asp162. In certain embodiments, said ADP-ribosyl-binding sequence additionally contains three residues selected from Arg15, Cys74, Leu97, Val103, Gly105, Gly110, and Asp162. In certain embodiments, said ADP- ribosyl-binding sequence additionally contains four residues selected from Arg15, Cys74, Leu97, Val103, Gly105, Gly110, and Asp162. In certain embodiments, said ADP-ribosyl- binding sequence additionally contains five residues selected from Arg15, Cys74, Leu97, Val103, Gly105, Gly110, and Asp162. In certain embodiments, said ADP-ribosyl-binding sequence additionally contains six residues selected from Arg15, Cys74, Leu97, Val103, Gly105, Gly110, and Asp162. In certain embodiments, said ADP-ribosyl-binding sequence additionally contains all residues Arg15, Cys74, Leu97, Val103, Gly105, Gly110, and Asp162.

The above-mentioned ADP-ribosyl-binding sequence has a certain ADP-ribosyl-binding activity. In certain embodiments, the above-mentioned ADP-ribosyl-binding sequence has at least 85% of the ADP-ribosyl-binding activity of SEQ ID NO 001. In certain embodiments, the above-mentioned ADP-ribosyl-binding sequence has at least 90% of the ADP-ribosyl-binding activity of SEQ ID NO 001. In certain embodiments, the above-mentioned ADP-ribosyl-binding sequence has at least 92% of the ADP-ribosyl-binding activity of SEQ ID NO 001. In certain embodiments, the above-mentioned ADP-ribosyl-binding sequence has at least 94% of the ADP-ribosyl-binding activity of SEQ ID NO 001. In certain embodiments, the above-mentioned ADP-ribosyl-binding sequence has at least 96% of the ADP-ribosyl-binding activity of SEQ ID NO 001. In certain embodiments, the above-mentioned ADP-ribosyl-binding sequence has at least 98% of the ADP-ribosyl-binding activity of SEQ ID NO 001.

The method of the first aspect is even more reliable in detecting and classifying tumors when used with the ADPR binder comprising SEQ ID NO 001, as shown in Example 8.

In certain embodiments, the cancer is selected from renal cell carcinoma, breast cancer, ovarian cancer and colon cancer. In certain embodiments, the cancer is renal cell carcinoma.

In certain embodiments, the cells are contacted with an additional binder. In certain embodiments, the additional binder is selected from GLUT1, CA IX, Ki-67, Mib-1, and p53. GLUT 1 is aberrantly expressed in several tumor types, as e.g. in some types of colon, ovarian or breast cancer.

CA IX is over-expressed in VHL mutated clear cell renal cell carcinoma (ccRCC) and hypoxic solid tumors, but is low-expressed in normal kidney and most other normal tissues. It may be involved in cell proliferation and transformation.

Ki-67 is a cellular marker for proliferation. During interphase, the Ki-67 antigen can be exclusively detected within the cell nucleus, whereas in mitosis most of the protein is relocated to the surface of the chromosomes. Ki-67 protein is present during all active phases of the cell cycle (G1, S, G2, and mitosis), but is absent in resting (quiescent) cells (GO).

Mib-1 is a cellular marker for proliferation. It is directed at the same antigen as Ki-67. p53 is upregulated in many types of cancer and thus, it serves as a marker for malignancy.

In certain embodiments, the isolated tissue sample is a biopsy of a neoplasm of said patient. In certain embodiments, the isolated tissue sample is a biopsy of a solid neoplasm of said patient.

A second aspect of the invention relates to a method for detection of ADP-ribosylated peptides or polypeptides in a cell, comprising the steps of a. providing a cell from an isolated tissue sample or from cell culture; b. contacting said cell with an ADPR binder; c. detecting ADP-ribosylated peptides or polypeptides.

In certain embodiments, said ADPR binder comprises SEQ ID NO 001. In certain embodiments, said ADPR binder comprises an ADP-ribosyl-binding sequence having ³90% identity to SEQ ID NO 001, wherein said ADP-ribosyl-binding sequence contains the residues Glu35 and Arg145. In certain embodiments, said ADPR binder comprises an ADP-ribosyl-binding sequence having ³92% identity to SEQ ID NO 001 , wherein said ADP- ribosyl-binding sequence contains the residues Glu35 and Arg145. In certain embodiments, said ADPR binder comprises an ADP-ribosyl-binding sequence having ³94% identity to SEQ ID NO 001 , wherein said ADP-ribosyl-binding sequence contains the residues Glu35 and Arg145. In certain embodiments, said ADPR binder comprises an ADP-ribosyl-binding sequence having ³96% identity to SEQ ID NO 001, wherein said ADP-ribosyl-binding sequence contains the residues Glu35 and Arg145. In certain embodiments, said ADPR binder comprises an ADP-ribosyl-binding sequence having ³98% identity to SEQ ID NO 001, wherein said ADP-ribosyl-binding sequence contains the residues Glu35 and Arg145. In certain embodiments, said ADPR binder comprises an ADP-ribosyl-binding sequence having ³99% identity to SEQ ID NO 001 , wherein said ADP-ribosyl-binding sequence contains the residues Glu35 and Arg145.

In certain embodiments, said ADP-ribosyl-binding sequence additionally contains one residue selected from Arg15, Cys74, Leu97, Val103, Gly105, Gly110, and Asp162. In certain embodiments, said ADP-ribosyl-binding sequence additionally contains two residues selected from Arg15, Cys74, Leu97, Val103, Gly105, Gly110, and Asp162. In certain embodiments, said ADP-ribosyl-binding sequence additionally contains three residues selected from Arg15, Cys74, Leu97, Val103, Gly105, Gly110, and Asp162. In certain embodiments, said ADP- ribosyl-binding sequence additionally contains four residues selected from Arg15, Cys74, Leu97, Val103, Gly105, Gly110, and Asp162. In certain embodiments, said ADP-ribosyl- binding sequence additionally contains five residues selected from Arg15, Cys74, Leu97, Val103, Gly105, Gly110, and Asp162. In certain embodiments, said ADP-ribosyl-binding sequence additionally contains six residues selected from Arg15, Cys74, Leu97, Val103, Gly105, Gly110, and Asp162. In certain embodiments, said ADP-ribosyl-binding sequence additionally contains all residues Arg15, Cys74, Leu97, Val103, Gly105, Gly110, and Asp162.

The above-mentioned ADP-ribosyl-binding sequence has a certain ADP-ribosyl-binding activity. In certain embodiments, the above-mentioned ADP-ribosyl-binding sequence has at least 85% of the ADP-ribosyl-binding activity of SEQ ID NO 001. In certain embodiments, the above-mentioned ADP-ribosyl-binding sequence has at least 90% of the ADP-ribosyl-binding activity of SEQ ID NO 001. In certain embodiments, the above-mentioned ADP-ribosyl-binding sequence has at least 92% of the ADP-ribosyl-binding activity of SEQ ID NO 001. In certain embodiments, the above-mentioned ADP-ribosyl-binding sequence has at least 94% of the ADP-ribosyl-binding activity of SEQ ID NO 001. In certain embodiments, the above-mentioned ADP-ribosyl-binding sequence has at least 96% of the ADP-ribosyl-binding activity of SEQ ID NO 001. In certain embodiments, the above-mentioned ADP-ribosyl-binding sequence has at least 98% of the ADP-ribosyl-binding activity of SEQ ID NO 001.

In certain embodiments of the first or second aspect, the detection of ADP-ribosylated biomolecules or peptides or polypeptides is non-nuclear, which means that only the signal outside of the nucleus is considered. In certain embodiments of the first or second aspect, the detection of ADP-ribosylated biomolecules or peptides or polypeptides is in the cytoplasm. Cytoplasm includes all organelles outside of the nucleus. In certain embodiments of the first or second aspect, the detection of ADP-ribosylated biomolecules or peptides or polypeptides is in mitochondria. In certain embodiments of the first or second aspect, the detection of ADP- ribosylated biomolecules or peptides or polypeptides is in the endoplasmatic reticulum (ER). In certain embodiments of the first or second aspect, the detection of ADP-ribosylated biomolecules or peptides or polypeptides is in stress granules. Stress granules are dense aggregations in the cytosol composed of proteins and RNAs that appear when the cell is under stress.

The ADPR signal in the experiments of the Examples showed a signal in the cytoplasm, not in the nucleus, which presumably stems from mitochondria based on immunofluorescence co- stainings with mitochondrial markers using TMA. It is assumed that the majority of the signal results from mono-ADPR, because enzymes catalysing mono-APDR were reported to be in the cytoplasm and in mitochondria (Hopp et al., Cells. 2019 Aug 13;8(8).), while poly-ADPR was reported to be mainly in the nucleus (Ray Chaudhuri et al., Nat Rev Mol Cell Biol. 2017 Oct; 18(10):610-621.).

A third aspect of the invention relates to a polypeptide comprising a. a polypeptide sequence comprising or essentially consisting of the sequence denoted as SEQ ID NO 001, or a variant thereof retaining its essential binding qualities and showing the key residues Glu35 and Arg145, and b. a detectable label.

The inventors have surprisingly found that the variant of the ADP binding moiety given as SEQ ID NO 001 , having the residues Glu35 and Arg145, has far superior binding characteristics than previously characterized ADPR binding domains, and enables diagnostic procedures detecting ADPR modification of biomolecules, particularly of proteins, with far higher sensitivity and specificity than previously known ligands.

In order to enable detection of the binder, the polypeptide needs to be modified. Any label allowing detection may be considered. The most salient examples of a detectable label include, but are not necessarily limited to: a tag sequence, allowing the specific and selective attachment of a secondary binder that in turn is labelled in a way to enable detection, for example by optical means; an enzymatic activity such as a fluorescent or luminescent protein moiety allowing direct detection, and a fluorescent dye attached to the polypeptide.

Tag sequences are of particular interest, as specific, labelled binders are readily available for a number of such tags.

In certain embodiments, the polypeptide comprises a tag sequence selected from an Fc-tag (a fragment crystallizable derived from an immunoglobulin), a Strep-tag, a glutathione-S- transferase-tag, a green fluorescent protein-tag, a BCCP-tag, a FI_AG-tag, an HA-tag, a Myc- tag, a maltose binding protein-tag, a Nus-tag, a thioredoxin-tag, a CRDSAT-tag, a poly- glutamate-tag, a calmodulin-tag, an ALFA-tag, an Avi-tag, a C-tag, an E-tag, an NE-tag, an S- tag, an SBP-tag, aSpot-tag, a T7-tag, a Ty-tag, a V5-tag, a VSV-tag, an Xpress-tag, and a TC- tag, particularly said tag sequence is an Fc-tag.

In certain embodiments, the polypeptide comprises a fluorescent protein tag. One example is green fluorescent protein (GFP) from Aequorea victoria and derivatives thereof, such as enhanced blue fluorescent protein (EBFP), enhanced blue fluorescent protein 2 (EBFP2), azurite, mKalamal, sirius enhanced green fluorescent protein (EGFP), emerald, superfolder avGFP, T-sapphire yellow fluorescent protein (YFP), enhanced yellow fluorescent protein (EYFP), citrine, venus, YPet, topaz, SYFP, mAmetrine enhanced cyan fluorescent protein (ECFP), mTurquoise, mTurquoise2, cerulean, CyPet, SCFP.

A fluorescent protein for practicing the invention may also be selected from the group comprising fluorescent protein from Discosoma striata and derivatives thereof: mTagBFP,

TagCFP, AmCyan, Midoriishi Cyan, mTFP1

- Azami Green, mWasabi, ZsGreen, TagGFP, TagGFP2, TurboGFP, CopCFP, AceGFP

- TagYFP, TurboYFP, ZsYellow, PhiYfP

Kusabira Orange, Kusabira Orange2, mOrange, mOrange2, dTomato, dTomato-Tandem, DsRed, DsRed2, DsRed-Express (T1), DsRed-Express2, DsRed-Max, DsRed-Monomer, TurboRFP, TagRFP, TagRFP-T mRuby, mApple, mStrawberry, AsRed2, mRFP1 , JRed, mCherry, eqFP611 , tdRFP611, HcRedl, mRaspberry tdRFP639, mKate, mKate2, katushka, tdKatushka, HcRed-Tandem, mPlum, AQ143.

Fluorescent proteins also comprise proteins derived from alpha-allophycocyanin from the cyanobacterium Trichodesmium erythraeum such as small ultra-red fluorescent protein.

In other embodiments, the polypeptide comprising SEQ ID NO 001 or a variant thereof comprises a dye molecule, exemplified but not limited to, octadecyl rhodamine B, 7-nitro-2- 1,3-benzoxadiazol-4-yl, 4-acetamido-4'-isothiocyanatostilbene-2,2' disulfonic acid, acridine and derivatives, 5-(2'-aminoethyl)aminonaphthalene-1-sulfonic acid (EDANS), 4-amino-N-(3- [vinylsulfonyl]phenyl)naphthalimide-3, 6-disulfonate dilithium salt, N-(4-anilino-1- naphthyl)maleimide, anthranilamide, BODIPY, Brilliant Yellow, coumarin and derivatives, cyanine dyes, cyanosine, 4',6-diaminidino-2-phenylindole (DAPI), bromopyrogallol red, 7- diethylamino-3-(4'-isothiocyanatophenyl)-4-methylcoumarin, diethylenetriamine pentaacetate, 4,4'-diisothiocyanatodihydro-stilbene-2,2'-disulfonic acid, 4,4'-diisothiocyanatostilbene-2,2'- disulfonic acid, dansylchloride, 4-dimethylaminophenylazophenyl-4'-isothiocyanate (DABITC), eosin and derivatives, erythrosin and derivatives, ethidium, fluorescein, 5-carboxyfluorescein (FAM), 5-(4,6-dichlorotriazin-2-yl)aminofluorescein (DTAF), 2',7'-dimethoxy-4'5'-dichloro-6- carboxyfluorescein, fluorescein isothiocyanate, X-rhodamine-5-(and 6)-isothiocyanate (QFITC or XRITC), fluorescamine, IR-144 (2-[2-[3-[[1,3-dihydro-1,1-dimethyl-3-(3-sulfopropyl)-2H- benz[e]indol2-ylidene]ethylidene]-2-[4-(ethoxycarbonyl)- 1-piperazinyl]-1-cyclopenten-1- yl]ethenyl]-1 ,1-dimethyl-3-(3-sulforpropyl)-1 H-benz[e]indolium hydroxide, inner salt, compound with n,n-diethylethanamine(1:1), CAS No.: 54849-69-3), 5-chloro-2-[2-[3-[(5-chloro- 3-ethyl-2(3H)-benzothiazol-ylidene)ethylidene]-2-(diphenylam ino)-1-cyclopenten-1- yl]ethenyl]-3-ethyl benzothiazoliu perchlorate (IR140), malachite green isothiocyanate, 4- methylumbelliferone, ortho cresolphthalein, nitrotyrosine, pararosaniline, phenol red, B- phycoerythrin, o-phthaldialdehyde, pyrene, pyrene butyrate, succinimidyl 1 -pyrene, butyrate quantum dots, Reactive Red 4 (Cibacron Brilliant Red 3B-A), rhodamine and derivatives, 6- carboxy-X-rhodamine (ROX), 6-carboxyrhodamine (R6G), lissamine rhodamine B sulfonyl chloride rhodamine (Rhod), rhodamine B, rhodamine 123, rhodamine X isothiocyanate, sulforhodamine B, sulforhodamine 101, sulfonyl chloride derivative of sulforhodamine 101 (Texas Red), N,N,N',N'tetramethyl-6-carboxyrhodamine (TAMRA) tetramethyl rhodamine, tetramethyl rhodamine isothiocyanate (TRITC), riboflavin, rosolic acid, terbium chelate derivatives, Cyanine-3 (Cy3), Cyanine-5 (Cy5), Cyanine-5.5 (Cy5.5), Cyanine-7 (Cy7), IRD 700, IRD 800, Alexa 647, La Jolta Blue, phthalo cyanine, and naphthalo cyanine.

In certain embodiments, said polypeptide sequence comprises SEQ ID NO 001. In certain embodiments, said polypeptide sequence comprises an ADP-ribosyl-binding sequence having ³90% identity to SEQ ID NO 001, wherein said ADP-ribosyl-binding sequence contains the residues Glu35 and Arg145. In certain embodiments, said polypeptide sequence comprises an ADP-ribosyl-binding sequence having >92% identity to SEQ ID NO 001 , wherein said ADP-ribosyl-binding sequence contains the residues Glu35 and Arg145. In certain embodiments, said polypeptide sequence comprises an ADP-ribosyl-binding sequence having ³94% identity to SEQ ID NO 001 , wherein said ADP-ribosyl-binding sequence contains the residues Glu35 and Arg145. In certain embodiments, said polypeptide sequence comprises an ADP-ribosyl-binding sequence having ³96% identity to SEQ ID NO 001, wherein said ADP-ribosyl-binding sequence contains the residues Glu35 and Arg145. In certain embodiments, said polypeptide sequence comprises an ADP-ribosyl-binding sequence having ³98% identity to SEQ ID NO 001, wherein said ADP-ribosyl-binding sequence contains the residues Glu35 and Arg145. In certain embodiments, said polypeptide sequence comprises an ADP-ribosyl-binding sequence having ³99% identity to SEQ ID NO 001, wherein said ADP- ribosyl-binding sequence contains the residues Glu35 and Arg145. In certain embodiments, said ADP-ribosyl-binding sequence additionally contains one residue selected from Arg15, Cys74, Leu97, Val103, Gly105, Gly110, and Asp162. In certain embodiments, said ADP-ribosyl-binding sequence additionally contains two residues selected from Arg15, Cys74, Leu97, Val103, Gly105, Gly110, and Asp162. In certain embodiments, said ADP-ribosyl-binding sequence additionally contains three residues selected from Arg15, Cys74, Leu97, Val103, Gly105, Gly110, and Asp162. In certain embodiments, said ADP- ribosyl-binding sequence additionally contains four residues selected from Arg15, Cys74, Leu97, Val103, Gly105, Gly110, and Asp162. In certain embodiments, said ADP-ribosyl- binding sequence additionally contains five residues selected from Arg15, Cys74, Leu97, Val103, Gly105, Gly110, and Asp162. In certain embodiments, said ADP-ribosyl-binding sequence additionally contains six residues selected from Arg15, Cys74, Leu97, Val103, Gly105, Gly110, and Asp162. In certain embodiments, said ADP-ribosyl-binding sequence additionally contains all residues Arg15, Cys74, Leu97, Val103, Gly105, Gly110, and Asp162.The above-mentioned ADP-ribosyl-binding sequence has a certain ADP-ribosyl- binding activity. In certain embodiments, the above-mentioned ADP-ribosyl-binding sequence has at least 85% of the ADP-ribosyl-binding activity of SEQ ID NO 001. In certain embodiments, the above-mentioned ADP-ribosyl-binding sequence has at least 90% of the ADP-ribosyl-binding activity of SEQ ID NO 001. In certain embodiments, the above-mentioned ADP-ribosyl-binding sequence has at least 92% of the ADP-ribosyl-binding activity of SEQ ID NO 001. In certain embodiments, the above-mentioned ADP-ribosyl-binding sequence has at least 94% of the ADP-ribosyl-binding activity of SEQ ID NO 001. In certain embodiments, the above-mentioned ADP-ribosyl-binding sequence has at least 96% of the ADP-ribosyl-binding activity of SEQ ID NO 001. In certain embodiments, the above-mentioned ADP-ribosyl-binding sequence has at least 98% of the ADP-ribosyl-binding activity of SEQ ID NO 001.

The polypeptide of the third aspect is particularly useful for carrying out the methods of the first and second aspect.

A fourth aspect of the invention relates to a system to carry out the method of any one of aspects 1 or 2. In particular, the system according to the invention is designed and configured to the method for determining the content of ADP-ribosylated biomolecules in a sample obtained from a patient, wherein an isolated tissue sample of the patient, the isolated tissue sample comprising a plurality of cells having a permeabilized cell membrane is contacted with an ADPR binder capable to specifically bind ADP-ribosylated biomolecules, and the system is fitted to detect the amount and location of ADP-ribosylated biomolecules inside the isolated tissue sample.

Wherever alternatives for single separable features such as, for example, an isotype protein or a diagnostic method are laid out herein as “embodiments”, it is to be understood that such alternatives may be combined freely to form discrete embodiments of the invention disclosed herein. Thus, any of the alternative embodiments for a diagnostic method may be combined with any isotype protein mentioned herein.

The invention is further illustrated by the following examples and figures, from which further embodiments and advantages can be drawn. These examples are meant to illustrate the invention but not to limit its scope.

Description of the Figures

Fig. 1 Characterization of evolved Af1521 macro domain (A) Pull-down experiment of unmodified and ADP-ribosylated H2B peptides (biotinylated and bound to streptavidin Sepharose beads) of WT Af1521 with the sequence: MEVLFEAKVGDITLKLAQGDITQYPAKAIVNAANKRLEHGGGVAYAIAKACAGDA GLYTEISKKAMREQFGRDYIDHGEVVVTPAMNLEERGIKYVFHTVGPICSGMWS EELKEKLYKAFLGPLEKAEEMGVESIAFPAVSAGIYGCDLEKVVETFLEAVKNFK GSAVKEVALVIYDRKSAEVALKVFERSL (SEQ ID NO 022) and eAf1521 with the sequence:

MEVLFEAKVGDITLRLAQGDITQYPAKAIVNAANERLEHGGGVAYAIAKACAGDA GLYTEISKKAMREQFGRDCIDHGEVVVTPAMNLEERGIKYVLHTVGPVCGGMW SGELKEKLYKAFLGPLEKAEEMGVESIAFPAVSAGIRGCDLEKVVETFLEAVKDF KGSAVKEVALVIYDRKSAEVALKVFERSL (SEQ ID NO 001) under increasing salt concentration. Input lanes are the same for all measured conditions. The proteins were detected with SDS-PAGE followed by Coomassie staining. (B) Sequence alignment of WT Af1521 and eAf1521. Point mutations in eAf1521 are highlighted. The substrate binding region is underlined. K35E and Y145R substitutions are marked in grey. (C) Structural comparison of the evolved macro domains eAf1521 (mustard) and WT Af1521 (turquoise; PDB ID: 2BFQ). (D, upper panel) Detailed representation of the terminal-ribose of the ADPr with the point mutations Y145R and K35E in its vicinity. The C1 ^' -linked oxygen is shown in purple. (D, lower panel left) Surface rendering of the WT Af1521 ADPr binding site. The 1141 sidechain is highlighted. (D, lower panel right) Corresponding view for the eAf1521 macro domain, with the 1141 sidechain rotated to entrap the diphosphate moiety. (E) Pull down experiment using unmodified and ADP-ribosylated H2B peptides (biotinylated and bound to streptavidin Sepharose beads) of WT Af1521-K35E- Y145R, e Af 1521-E35K-R145Y and eAf1521-l144G under increasing salt concentration. Input lanes are the same for all measured conditions. The proteins were analyzed by SDS-PAGE followed by Coomassie staining.

Fig. 2 Stronger enrichment of global ADP-ribosylome by eAf1521 compared to WT Af1521 (A, B) Volcano plots comparing ADP-ribosylated peptides enriched from untreated and H2C>2-treated HeLa cells with either WT Af1521 oreAf1521. (C) Venn diagrams displaying the overlap of ADP-ribosylated proteins enriched from untreated (left panel) or hhC reated HeLa cells (right panel) with either WT Af1521 or eAf1521. (D) Distribution of ADPr on different amino acid acceptor sites (D, E, K, R, S, Y). (E) Venn diagram representing the overlap of ADP-ribosylated proteins enriched with eAf1521 using different starting material amounts (5 mg, 10 mg and 20 mg). (F) Scatter plot comparing the number of ADP-ribosylated proteins identified via eAf1521 enrichment using different starting material (5 mg, 10 mg and 20 mg).

Fig. 3 eAf1521 more strongly enriches for Serine-ADP-ribosylation (A) Box plot representing MS1 intensities of all ADPr peptides of untreated and H2C>2-treated HeLa cell lysate enriched with either WT Af1521 or eAf2151. (B) Scatter plots depicting MS1 ADPr peptide intensities of selected S-ADPr modified peptides (PARP1: SKGQVKEEGINKSEK (SEQ ID NO 002), HNRNPA1:

SSGPYGGGGQYFAKPR (SEQ ID NO 023), LMNA: SGAQASSTPLSPTR (SEQ ID NO 024), RBMX: SDLYSSGR (SEQ ID NO. 025)). The modification confidence of the amino acid acceptor sites are >95%. (C) Scatter plots depicting MS1 ADPr peptide intensities of selected R-ADPr peptides identified in all untreated and H2O2- treated HeLa lysates (P4HB: VDATEESDLAQQYGVRGYPTI K (SEQ ID NO 026), PDIA3: RLAPEYEAAATR (SEQ ID NO 027)). The modification confidence of the amino acid acceptor sites are > 95%.

Fig. 4 Fc-eAf1521 is more sensitive than Fc-WT Af1521 in detecting mono- and poly-ADP-ribosylated target proteins (A) Immunoblot analyses of different substrates with either Fc-eAf1521 or Fc-WT Af1521 (untreated and H ₂ 0 ₂ -treated HeLa cells including PJ34 treatment as control for ADP-ribosylation inhibition; unmodified, poly-ADP-ribosylated and mono-ADP-ribosylated ARTD1 and unmodified and mono-ADP-ribosylated ARTD8cat). Detection was performed with IRDye 680 goat anti-mouse antibody. Immunoblot analyses to be compared were performed at the same time and with the same exposure. (B) Immunofluorescence analysis of HeLa cells cotreated with H2O2 and 20 mM Olaparib stained with either Fc-WT Af1521 or Fc-eAf1521, followed by Alexa 488-labeled goat anti-mouse antibody.

Fig. 5 ADPR signal in kidney tissues. (A) Kidney tissue TMA stained with anti-ADPR antibody. Scoring: negative (1), weak (2), moderate (3), elevate (4) and strong (5). (B) ADPR staining of healthy kidney epithelial cells on the left and of RCC tissue on the right. On the top, the overview (scale bar 100 pm, 10x), on the bottom a ROI (scale bar 10 pm, 40x). (C) Semi-quantitative staining analysis, Chi-square test, p- value < 0.0001, N=367. (D) RCC tissues stained with anti-ADPR antibody. Scores: 1 (negative), 2 (weak), 3 (moderate), 4 (elevate) and 5 (strong). On the top the overview of the tissue spot in the TMA (scale bar 100 pm, 10x), on the bottom the ROI (scale bar 10 pm, 40x). (E) Kaplan-Meier survival plot of RCC patients stratified based on ADPR signal intensity as follows: strong (5 - red), moderate (4, 3, 2 - yellow) and weak (1 - blue). Mantel-Cox test, p-value=0.0163, N=294.

Fig. 6 Mitochondrial ADPR (mtADPR) signal in papillary RCC subtype. (A) ADPR staining in papillary RCC subtype and related scores. Upper pictures overview (scale bar 100 pm, 10x), lower pictures ROI (scale bar 10 pm, 40x). (B) Kaplan- Meier survival plot patients with papillary RCC stratified based on mtADPR signal intensity: high mtADPR (scores 5-4 - red), low ADPR (scores 3-2 - blue), Mantel- Cox test, p-value = 0.0107, N=41. (C) Contingency, Fisher’s exact test to assess the association between mtADPR scores high/low and ISUP grade in papillary RCC (p-value=0.0147, N=41).

Fig. 7 mtADPR signal in breast cancer (BrCa) and its subtypes. (A) BrCa TMA stained with anti-ADPR antibody overview. (B) BrCa tissues stained with anti- ADPR antibody. Scores: negative/weak (1- blue), moderate (2 - yellow), strong (3

- red). On the top the overview of the tissue spot in the TMA (scale bar 100 pm, 10x), on the bottom the ROI (scale bar 10 urn, 40x). (C) Kaplan-Meier overall survival plot of BrCa patients stratified based on mtADPR signal intensity as follows: strong (3 - red), moderate (2 - yellow) and weak (1 - blue). Mantel-Cox test, p-value < 0.0001, N=877. (D) invasive ductal BrCa tissues stained with anti- ADPR antibody. Scores: negative/weak (1- blue), moderate (2 - yellow), strong (3

- red). On the top the overview of the tissue spot in the TMA (scale bar 100 pm, 10x), on the bottom the ROI (scale bar 10 urn, 40x). (E) Kaplan-Meier overall survival plot of invasive ductal BrCa patients stratified based on mtADPR signal intensity as follows: strong (3 - red), moderate (2 - yellow) and weak (1 - blue). Mantel-Cox test, p-value = 0.0001, N=663. (F) correlation between mtADPR scores (negative/weak - blue, moderate - yellow, strong - red) and tumor grade in invasive ductal BrCa. Stages: early (1, 2); late (3, 4). Chi-square test, p-value = 0.0066, N=700. (G) invasive lobular BrCa tissues stained with anti-ADPR antibody. Scores: negative/weak (1- blue), moderate (2 - yellow), strong (3 - red). On the top the overview of the tissue spot in the TMA (scale bar 100 pm, 10x), on the bottom the ROI (scale bar 10 urn, 40x). (H) Kaplan-Meier overall survival plot of invasive lobular BrCa patients stratified based on mtADPR signal intensity as follows: strong (3 - red), moderate (2 - yellow) and weak (1 - blue). Mantel-Cox test, p-value = 0.0021, N=121.

Fig. 8 mtADPR signal in high grade serous ovarian cancer. (A) OvCa TMA stained with anti-ADPR antibody overview. (B) High grade serous OvCa tissues stained with anti-ADPR antibody. Scores: negative/weak (1- blue), moderate (2 - yellow), strong (3 - red). On the top the overview of the tissue spot in the TMA (scale bar 100 urn, 10x), on the bottom the ROI (scale bar 10 urn, 40x). (C) Kaplan-Meier overall survival plot of high grade serous OvCa patients stratified based on mtADPR signal intensity as follows: strong (3 - red), moderate (2 - yellow) and weak (1 - blue). Mantel-Cox test, p-value = 0.0088, N=68.

Fig. 9 mtADPR signal in colon tissues. (A) Colon tissues stained with anti-ADPR antibody. Scores: negative/weak (1- blue), moderate (2 - yellow), strong (3 - red). On the top, colon cancer (CoCa) tissues with tissue overview (scale bar 100 pm, 10x) and below the ROI (scale bar 10 pm, 40x). On the bottom, colon epithelium with tissue overview (scale bar 100 pm, 10x) and below the ROI (scale bar 10 pm, 40x). (B) Semi-quantitative staining analysis, Chi-square test, p-value = 0.0034, N=257. (C) Correlation between mtADPR scores: negative/weak (1- blue), moderate (2 - yellow), strong (3 - red) and tumor stage in CoCa. Stages: early (1 , 2); late (3, 4). Chi-square test, p-value = 0.0087, N=234.

Fig. 10 eAf1521 (mtADPR) signal in RCC. (A) Kidney tissues stained with eAf1521

(ADPR). Scores: 1 (negative/weak), 2 (moderate) and 3 (strong). On the top, RCC tissues with above overview (scale bar 100 pm, 10x) and below the ROI (scale bar 10 pm, 40x). On the bottom, kidney epithelium with above tissue overview (scale bar 100 pm, 10x) and below the ROI (scale bar 10 pm, 40x). (B) Semi-quantitative staining analysis, Chi-square test, p-value = 0.0249, N=364. Low scores (1-2, fair), high scores (3, dark). (C) Kaplan-Meier overall survival plot of RCC patients stratified based on eAf1521 (mtADPR) signal intensity as follows: strong (3 - red), moderate (2 - yellow) and weak/negative (1 - blue). Mantel-Cox test, p-value < 0.0001, N=293. (D) Correlation between mtADPR scores obtained with eAf1521 (mtADPR) in RCC tissues (red - strong; yellow - moderate and blue - weak/negative) and tumor stage in RCC. Stages: early (1 , 2); late (3, 4). Chi-square test, p-value = 0.0004, N=309. (E) Correlation between mtADPR scores obtained with eAf1521 (ADPR) in RCC tissues (red - strong; yellow - moderate and blue - weak/negative) and ISUP grade in RCC. Stages: early (1 , 2); late (3, 4). Chi-square test, p-value < 0.0001, N=297. Fig. 11 eAf1521 (mtADPR) signal in ccRCC. (A) ccRCC stained with eAf1521 (mtADPR). Scores: 1 (negative/weak), 2 (moderate) and 3 (strong). On the top, ccRCC tissues with above overview (scale bar 100 p , 10x) and below the ROI (scale bar 10 urn, 40x). (B) Kaplan-Meier overall survival plot of ccRCC patients stratified based on eAf1521 (mtADPR) signal intensity as follows: strong (3 - red) and low (2, 1 - green). Mantel-Cox test, p-value = 0.0496, N=42. (C) Correlation between mtADPR scores obtained with eAf1521 (mtADPR) in ccRCC tissues (red - strong; yellow - moderate and blue - weak/negative) and tumor stage in ccRCC. Stages: early (1 , 2); late (3, 4). Chi-square test, p-value = 0.0029, N=242. (D) Correlation between mtADPR scores obtained with eAf1521 (ADPR) in ccRCC tissues (red - strong; yellow - moderate and blue - weak/negative) and ISUP grade in ccRCC. Stages: early (1, 2); late (3, 4). Chi-square test, p-value = 0.0002, N=241.

Fig. 12 eAf1521 (mtADPR) signal in papRCC. (A) papRCC stained with eAf1521

(mtADPR). Scores: 1 (negative/weak), 2 (moderate) and 3 (strong). On the top, papRCC tissues with above overview (scale bar 100 pm, 10x) and below the ROI (scale bar 10 pm, 40x). (B) Kaplan-Meier overall survival plot of papRCC patients stratified based on mtADPR signal intensity as follows: strong (3 - red) and low (2, 1 - green). Mantel-Cox test, p-value = 0.0496, N=42.

Fig. 13 Correlation ADPR and eAf1521 in RCC. (A) Correlation of signal intensity ADPR vs eAf1521 (ADPR) in RCC. eAf1521 scores: weak/negative (1, yellow), moderate (2, orange) and strong (3, brown). Chi-square test, N=330. (B) Correlation of signal intensity ADPR vs eAf1521 (ADPR) in RCC. eAf1521 scores: weak/negative (1 - yellow) and high (2, 3 - gold). Chi-square test for trend, p-value = 0.0001 , N=330. (C) Correlation of signal intensity ADPR vs eAf1521 (ADPR) in RCC. eAf1521 scores: low (1, 2 - beige) and strong (3 - gold). Chi-square test for trend, p-value < 0.0001 , N=330.

Fig. 14 Ribosome display evolves macro domain eAf1521. (A) ELISA analysis of eAf1521 variants against the unmodified and ADPr H2B peptide. The ELISA was performed after the 4 ^th round of error-prone PCR followed by selection using ribosome display. AU, absorbance unit. (B) Exemplary SDS-PAGE gel of pull down experiment using unmodified and ADP-ribosylated H2B peptides (biotinylated and bound to streptavidin Sepharose beads) of Af1521 WT. The proteins were analyzed by SDS-PAGE followed by Coomassie staining. The whole SDS-PAGE gel confirms equal concentration of peptides are used and that the elution was sufficient for the binding assay. (C) Pull-down experiment using ADP- ribosylated H2B peptides (biotinylated and bound to streptavidin Sepharose beads) of WT Af1521 and eAf1521 under increasing concentration of free ADPr. The proteins were analyzed by SDS-PAGE followed by Coomassie staining. (D) View of the eAf1521 bound ADP-ribose showing the electron density around the ADP-ribose as well as side chains N34, E35 and R145, contoured at 1.5 o.

Fig. 15 Comparison of MS-based identification of ADP-ribosylated peptides and proteins using WT Af1521 or eAf1521. Venn diagram depicting the overlap of ADP-ribosylated proteins in H2C>2-treated HeLa cells lysate in different ADPr proteomic studies. For comparison with Hendriks et al. 2019 dataset only the closest experimental conditions Trypsin and EThcD were included.

Fig. 16 Fc-WT Af1521 and eAf1521 as a tool for Western Blot analysis. Immunoblot analyses to be compared were performed at the same time and with the same exposure. (A) Dot blot analyses of free PAR chains with either Fc-eAf1521 or Fc- WT Af1521. Immunoblot analyses to be compared were performed at the same time and with the same exposure. Detection was performed with IRDye 680 goat anti-mouse antibody. (B) Immunoblot analyses of oligo-ADP-ribosylated and poly- ADP-ribosylated ARTD1 with either Fc-eAf1521 or Fc-WT Af1521. The automodification reaction of ARTD1 was performed with either 3 mM or 300 pM NAD ⁺ for 30 min at 37°C. Immunoblot analyses to be compared were performed at the same time and with the same exposure. Detection was performed with IRDye 680 goat anti-mouse antibody. (C) Pull-down experiment using GST-Af1521 WT, GST-eAf1521 and GST of poly-ADP-ribosylated ARTD1. Only 10% of input and unbound fractions were loaded. The proteins were detected by immunoblot analysis stained with 10H antibody (PAR antibody) and followed by IRDye 680 goat anti-mouse antibody.

Fig. 17 shows panADPR staining (Millipore) staining in RCC. (A) IHC staining of kidney tissues with anti-panADPR binding reagent (Millipore) and related intensity scores: strong (red, 3), moderate (yellow, 2) and weak/negative (blue, 1). On the top, selected TMA cores (scale bar 100 pm, 10x), on the bottom, ROIs (scale bar 20 pm, 40x). (B) Kaplan-Meier survival plot of (from left to right) clear cell (Mantel-Cox test, N=221), chromophobe (Mantel-Cox test, N=15) and papillary RCC (Mantel- Cox test, N=39) with patients stratified based on the panADPR signal intensity.

Fig. 18 shows PAR staining (Enzo) in RCC. (A) IHC staining of kidney tissues with anti- PAR antibody (Enzo) and related intensity scores: strong (red, 3), moderate (yellow, 2) and weak/negative (blue, 1). On the top, selected TMA cores (scale bar 100 pm, 10x), on the bottom, ROIs (scale bar 20 pm, 40x). (B) Kaplan-Meier survival plot of (from left to right) clear cell (Mantel-Cox test, N=216), chromophobe (Mantel-Cox test, N=16) and papillary RCC (Mantel-Cox test, N=43) with patients stratified based on the PAR signal intensity.

Examples

Example 1: In vitro selection of an AH521 macro domain with 1000-fold increased affinity for

ADP-ribose

To decipher cellular ADP-ribosylomes and identify the corresponding ADPr acceptor sites, the inventors recently co-developed a mass spectrometry (MS)-based approach that employs the Af1521 macro domain to enrich ADP-ribosylated peptides for the sample of interest. To improve the binding affinity of wildtype (WT) Af1521 for ADPr and, thus, the detection of ADP-ribosylated peptides or proteins, error-prone PCR driven ribosome display selection of Af1521 was performed. As target, the inventors used an H2B peptide that was synthetically ADP-ribosylated with a N-glycosidic linkage on Q at position 2 of the peptide (Methods Mol Biol, 687, 283-306; Angew Chem Int Ed Engl, 55:10634-10638). After performing the error-prone PCR, the Af1521 mutants were in vitro transcribed, translated (such that they do not leave the ribosome) and subsequently selected using the ADP- ribosylated peptide. After every selection round, the enriched mRNA was isolated and amplified by RT-PCR. The enriched pools were recloned into the ribosome display vector pRDV which served as template for the next round of selection. Four rounds of selection using the modified peptide coupled either to streptavidin-coated plates (for rounds 1-3) or magnetic beads (for round 4) were performed with increasing stringency. This was achieved by extending the washing times in round 2, 3 and 4, reducing the target concentration from 200 nM ADPr H2B peptide to 100 nM in round 2 and to 20 nM in round 3 and 4. Off-rate selection was also implemented using unmodified H2B peptide for round 3 and auto-ADP- ribosylated ARTD10 for round 4 (Ahmad, S. et al. (2016), Sci Rep, 6, 28922; Zahnd, C. et al. (2010) Protein Eng Des Sel, 23, 175-184; Dreier, B. et al. (2011) J Mol Biol, 405, 410-426.)

After 4 ribosome display selection iterations (i.e. using peptide coupled streptavidin-containing plates for iteration 1-3 or magnetic beads for iteration 4), the pools were subcloned into an E. coli expression vector for subsequent analysis of the expressed mutated Af1521 domains by ELISA, using the same unmodified and modified H2B peptides and its unmodified counterpart.

88 out of 92 clones (> 95%) displayed specific and significant ADP-ribosylated peptide binding properties (binding signals > 0.5 absorbance units (All)) compared to the unmodified peptide (Fig. 14A). Due to the large number of positive binders, the inventors randomly chose 10 clones and subcloned them as GST-fusion proteins. After expression and purification, increasingly stringent pull-down assays using unmodified and ADP-ribosylated H2B peptides were performed to further characterize the candidate binders (Fig. 1A, Fig. 14B).

While binding of WT AH521 to ADP-ribosylated peptide was compromised with salt concentrations of 200 mM and lost with 400 mM, the inventors observed that one of the eAf1521 macro domains candidates was still able to bind the modified peptide very well in the presence of 400 mM NaCI. This demonstrates that the affinity of eAf1521 for the ADP- ribosylated H2B peptide had strongly improved. Sequencing revealed that this eAf1521 isoform contains nine point mutations (Fig. 1B). Based on the available structure of WT Af1521 , the inventors found that one of these mutations, Y145R, occurs within the ADPr binding region of the macro domain.

In addition, pull-down experiments with either GST-WT Af1521 or GST-eAf1521 and the modified H2B peptides in the presence of increasing concentrations of free ADPr as competitor confirmed the strong binding of eAf1521 compared to WT Af1521 (Fig. 14C). Low concentrations of free ADPr completely inhibited WT Af1521 binding to the ADP-ribosylated peptide, while eAf1521 ADP-ribosylated peptide binding was still observed even in the presence of 10x more free ADPr relative to the modified peptide concentration.

To quantify the binding affinities of WT and eAf1521 to ADPr the inventors performed surface plasmon resonance (SPR) measurements. For these experiments, recombinantly expressed and purified His-tagged WT or eAf1521 were immobilized on a (multi)NTA modified NiHCIOOOM chip and their affinities for soluble ADPr were analyzed. These kinetic analyses revealed that the D of ADPr for WT Af1521 was ~3 mM, while eAf1521 displayed a D of ~3 nM (Tab. 1) and thus a ~1000-fold affinity increase.

As the Af1521 macro domain also exhibits hydrolysis activity that can compromise modified protein detection and enrichment potentials, the inventors compared the catalytic activities of WT Af1521 and eAf1521 using an in vitro ADP-ribosylation assay. To this end, the catalytic domain of ARTD8 (ARTD8cat) was auto-modified using radiolabeled ³² P-NAD ⁺ , residual NAD ⁺ removed and the resulting mono-ADP-ribosylated ARTD8cat used to define the hydrolytic activities of WT Af1521 or eAf1521. While both tested enzymes remained inactive at 4°C even after 2 h incubation, the condition used for ADP-ribosylome enrichment for MS analysis, both appeared to demodify ARTD8cat at 37 °C. Together, these data suggest that the 9 point mutations introduced into eAf1521 did not abolish the ADPr hydrolase activity. Example 2: K35E and Y145R mutations in eAf1521 form a salt bridge that facilitate additional interactions with ADPrthat enhance its affinity for ADPr

To gain further mechanistic insights into the superior affinity of eAf1521 for ADPr, the inventors studied the structural changes induced by the 9 point mutations. To this end, the inventors solved the crystal structure of eAf1521 at 1.82 A resolution. The overall structure of the eAf1521 macro domain was virtually identical to WT Af1521 (Fig. 1C). The two models align with a root mean square difference of 0.3 A (calculated over all 192 shared Ca positions). Nevertheless, two notable features could explain the improved affinity of eAf1521 for ADP- ribosylated targets. First, the orientation of the terminal i.e. nicotinamide-linked ribose moiety of the eAf1521-bound ADPr was rotated such that the ribose C-T oxygen, which serves as the anchor point for the target side chain, resided closer to the macro domain surface (Fig. 1D). This change was facilitated by the K35E and Y145R amino acid substitutions. The R145 sidechain formed a salt bridge with the E35 carboxyl and thus provides a rigid boundary for terminal ribose binding while allowing hydrogen bond formation between the arginine Ns atom and the C-4’ (ring-forming) oxygen. In the WT Af1521 , the terminal ribose site is delineated by the Y145 side chain, which is situated between the ribose carbon ring on one face, and 1102 side chain on the other. Indeed, this novel eAf1521 amino acid arrangement appears to contribute directly to strengthening the interaction with the proximal ribose. The hydrogen bonding interaction between R145 of eAf1521 and the bound proximal ribose also changes the rotamer of the 1144 side chain. In the eAf1521 structure, the 1144 sidechain adopts a rotamer that bridges the central phosphates of ADPr and apparently appears to trap the ADPr (Fig. 1D and Fig. 14D) but may not energetically contribute to the binding. This 1141 rotamer formed as a consequence of the altered orientation of the eAf1521 bound terminal ribose described above.

To confirm the importance of the newly generated salt bridge between R145 and E35, the inventors replaced the arginine with a shorter lysine residue (R145K), which lacks the Ns atom H-bonding to the C-4’ (ring-forming) oxygen. Binding experiments with the modified H2B peptide revealed that the GST-eAf1521-R145K mutant was no longer able to bind to the ADP- ribosylated peptide, comparable to a known non-binding Af1521 macro domain mutant (Af1521-G42E). These data confirmed the importance of the R145 residue and showed that the positive charge by itself (R>K) was not sufficient to maintain the high affinity for ADPr peptides, but that the guanidino group is essential, contacting both E35 and C-4’ (ring-forming) oxygen.

To further test whether E35 and R145, and not the other 7 mutations, are the main contributors to the observed increase in affinity of eAf1521 for ADPr, the inventors introduced K35E and Y145R into WT Af1521 and vice versa (Fig. 1E). Binding assays confirmed that the GST-eAf1521-E35K-R145Y mutant lost its increased affinity and now bound ADP-ribosylated peptides comparable to WT Af1521. In further support of this hypothesis, GST-WT Af1521- K35E-Y145R displayed significantly augmented affinity for the ADP-ribosylated H2B peptide (Fig. 1 E), which became comparable to eAf1521 (Fig 1A).

Finally, to define the influence 1144 tunnel formation with the ADPr pyrophosphate had on eAf1521 ADPr binding, the inventors generated I144G eAf1521 mutants and performed additional binding assays as described above (Fig. 1 E). The binding of eAf1521-H44G remained similar compared to eAf1521 indicating that 1144 and the formation of the tunnel did not contribute to the increased binding we observed with eAf1521. Together, these results provide strong evidence that the two mutations K35E and Y145R, and as a consequence the newly generated salt bridge, solely contribute to the increased binding affinity of eAf1521.

Example 3: eAf1521 enriches ADP-ribosylation present in oenotoxic stressed HeLa cells to a larger extent than its WT counterpart

The experiments described above were all carried out on free ADPr or a synthetically ADP- ribosylated H2B peptide. To investigate whether the in vitro selection procedure could have been specifically biased towards this H2B peptide, the inventors compared the HeLa cell H2O2- induced ADP-ribosylated peptide identification capacities of WT Af1521 and eAf1521 using the inventors’ established mass spectrometry work flow. This well-established workflow was applied to both untreated and H ₂ 0 ₂ -treated HeLa cell lysates with either WT Af1521 or eAf1521. Comparison of the untreated and H ₂ 0 ₂ -treated HeLa cell lysates confirmed a strong induction of ADP-ribosylation at the peptide and protein levels, presumably via H ₂ 0 ₂ -induced activation of nuclear ARTD1. Interestingly, eAf1521 enriched significantly more ADP- ribosylated peptides from both untreated and H ₂ 0 ₂ -treated HeLa cell lysates (Fig. 2A and B). In the untreated HeLa samples, ADP-ribosylation levels are typically not detectable by conventional anti-PAR antibody immunofluorescence. Indeed, >6 times more ADP-ribosylated proteins were detected in untreated HeLa lysates with eAf1521 compared to WT Af1521 (56 compared to 9 ADPr proteins; Fig. 2C, left panel). In agreement with this, the number of distinct ADP-ribosylated proteins identified in H ₂ 0 ₂ -treated HeLa cell lysates, which are used by researchers in the community as the “gold-standard” ADP-ribosylome due to its known high ADP-ribosylation levels, was 2-fold greater when the enrichments were performed with eAf1521 rather than WT Af1521 (419 vs. 185 ADPr proteins; Fig. 2C, right panel).

Together, these data indicate that the higher enrichment efficiency of eAf1521 resulted in an increase in the number of modified proteins identified within a complex biological sample. When comparing the overlap of the identified ADP-ribosylomes from either WT or eAf1521 , the inventors observed over 95% of all ADPr proteins after H ₂ 0 ₂ -treatment identified with WT Af1521 were also identified via enrichment with eAf1521 (Fig. 2C, right panel). The remaining ADP-ribosylated proteins identified only with WT Af1521 lie within the dynamic range of the MS method.

To further exclude that the selection strategy of eAf1521 resulted in any bias in the enrichment of ADPr modified proteins, comparison of the inventors’ eAf1521 ADP-ribosylome with another ADPr proteomic study using WT Af1521 for enrichment of H2C>2-treated HeLa lysate was performed. This comparison revealed a high degree overlap of ADPr proteins, strongly implying that the selection strategy did not result in any apparent bias (Fig. 15)

Together, these results suggest that the observed eAf1521 ADPr affinity enhancements are not specific to the modified H2B peptide it was selected against and demonstrate that eAf1521 ADP-ribosylated peptide enrichment capacities remain ubiquitous for this PTM. To investigate the ADPr acceptor sites, the inventors restricted their searches of EThcD spectra to the potential acceptor sites S, R, K, D, E, and Y, as the evidence provided by Hendriks et al, 2019 indicated that other ADPr acceptor amino acids (e.g. C, H and T) are not very abundant in either untreated and H2C>2-treated Hela cells (Hendriks, I. A. et al. (2019), Mol Cell Proteomics, 18, 1010-1026). Furthermore, in agreement with previous findings, ADPr acceptor site analysis revealed that S was the major modified amino acid identified in the HeLa H2C>2-treated samples (Fig. 2D).

The current MS-based workflows employed by the community for the identification of cellular ADP-ribosylome rely on large input material. This has remained a major limitation with respect to the type of samples that are feasible to analyze and has especially limited the analysis of samples derived from clinical settings (e.g. patient biopsies). To investigate whether performing eAf1521 -based MS analysis could reduce starting material requirements, the inventors used the same workflow with 5 mg, 10 mg or 20 mg of H2C>2-treated HeLa lysate as input material and carried out label-free quantification analyses. Three independent MS sample measurements identified similar amounts of ADP-ribosylated proteins. Comparison of the identified ADP-ribosylated proteins enriched by eAf1521 with the corresponding lysate amounts revealed that the procedure was reproducible and with all three tested concentrations the number of identified ADPr proteins increased when enriching with eAf1521 (Fig. 2E). 20 or 10 mg identified comparable numbers of modified proteins, nevertheless reducing the input material by a factor of four (to 5mg) still allowed identification of 80% of the ADP-ribosylated proteins identified with the maximal input material, implying that the highest abundant ADPr proteins are still detectable (Fig. 2F). Together, these data suggest that eAf1521-based enrichments of input materials whose quantities are less than ideal could still lead to the acquisition of valuable ADP-ribosylome data that may prove fruitful, especially within disease and/or clinical contexts. Example 4: Modified peptide and ADP-ribose acceptor site identifications were strongly improved due to higher MS spectral intensities

When analyzing the MS signal intensities of all ADP-ribosylated peptides, the inventors observed strong increases in the intensities in H2C>2-treated HeLa cell lysates compared to the untreated condition (Fig. 3A). Furthermore, global ADPr peptide intensities were enhanced independent of the condition (untreated vs. H2C>2-treated) when the enrichment was carried out with eAf1521. Ultimately, this led to significant increases in the number and purity of ADPr peptide spectra matches (PSMs) identified with eAf1521 compared to WT Af1521.

To further investigate peptides with different ADPr acceptor sites and the contribution of either WT Af1521 or eAf1521 , the inventors selected ADPr peptides that were confidently modified on either S or Arginine (R) residues with a site localization probability of > 95% (Fig. 3B and C). The inventors observed that MS signal intensities generally increased for S-ADPr peptides after treatment with H2O2. These S-ADPr peptides belong to proteins that localize to the nucleus and are most likely targets of H2C>2-activated ARTD1. Interestingly, some S-ADPr peptides shown here (ARTD1/PARP1 , SKGQVKEEGINKSEK (SEQ ID NO 002), S507; HNPNPA1, SSGPYGGGGQYFAPR (SEQ ID NO 003), S337) were also identified in the untreated HeLa cell lysates, indicating that ARTD1 regulates cellular processes via ADP- ribosylation also under basal conditions using the same modification sites. In contrast, the intensities of the detected R-ADPr peptides did not change after H ₂ 0 ₂ -treatment (Fig. 3C). Moreover, the R-ADPr modified proteins are located in the cytoplasm, suggesting that these targets are not modified by ARTD1 and are not affected by H ₂ 0 ₂ -treatment.

In addition, the inventors compared the ability of WT Af1521 and eAf1521 to enrich S- ADPr versus R-ADPr peptides. Most of the S-ADPr modified peptide intensities were significantly enhanced using eAf1521 compared to WT Af1521 , which was not the case for the R-ADPr peptides (Fig. 3B, C). Nevertheless, eAf1521 was still capable of enriching R-ADPr modified peptides to the same relative extent than Af1521 WT.

Example 5: eAf1521 improves mono- and oolv-ADP-ribosylated protein detection via immunoblotting

The increased affinity of the eAf1521 for ADPr and the subsequent increase in the identification of ADP-ribosylated proteins the inventors observed under all conditions in the proteomic studies, prompted them to test whether eAf1521 could be used to detect cellular ADP- ribosylation by immunoblotting. To this end, WT Af1521 and eAf1521 were fused to an Fc- domain of lgG2a to generate a bivalent fusion protein. The new constructs were expressed in HEK 293T cells and purified from the supernatants using a Ni-NTA bead-based purification, exploiting the C-terminal His tag behind an HA tag on the Fc domain. The fusion constructs were first characterized by immunoblots, testing against three different extracts: i) a HeLa cell lysate that was treated with H2O2 alone or together with the broad PARP-inhibitor PJ34, ii) in vitro auto-modified poly-ADP-ribosylated ARTD1 that was subsequently left untreated or treated with the enzyme poly(ADP-ribose)-glycohydrolase PARG to reduce PARylation to mono-ADP-ribosylated ARTD1 , or iii) in vitro auto-modified mono-ADP-ribosylated ARTD8cat (Fig. 4A). In all tested conditions, eAf1521 recognized the modified proteins to a stronger extent irrespective of whether the proteins were mono- or poly- ADP-ribosylated. Treatment of samples with PJ34 reduced the ADP-ribosylation signal after H2O2 treatment, indicating that the signals detected were indeed dependent on ADP- ribosylation. The same experiment was performed with different concentrations of the purified Af1521-Fc constructs. The detection differences between WT Af1521 or eAf1521 could be confirmed at all concentrations; at the highest concentration used (500 ng/mL) eAf1521 signals were stronger than WT. Moreover, at lower concentrations of WT Af1521 signals were not detectable (125 ng/mL or 31.25 ng/mL), whereas at eAf1521 concentrations as low as 31.25 ng/mL both poly- or mono-ADP-ribosylated proteins were still detected. Thus, demonstrating that eAf1521 also displays a ~16-fold increase in immunoblot detection sensitivity when compared to WT Af1521. Finally, the inventors also tested the specificity of Fc-WT Af1521 and Fc-eAf1521 constructs by performing competition experiments with ADPr against in vitro auto- modified poly-ADP-ribosylated and mono-ADP-ribosylated ARTD1. Co-incubation of the inventors’ Fc-proteins with free ADPr reduced the signal on immunoblots confirming the specificity towards ADP-ribosylated proteins. To specifically define the oligo-ADP-ribosylation detection sensitivity, the inventors auto-modified ARTD1 in vitro using different NAD ⁺ concentrations to generate either oligo- or poly-ADP-ribosylated ARTD1. Subsequent immunoblot analysis confirmed that eAf1521 detects both oligo- and poly-ADP-ribosylated ARTD1 similarly to WT Af1521 (Fig. 16A). Next, the inventors aimed to determine the enrichment capability of eAf1521 towards poly-ADP-ribosylated proteins. Therefore, the inventors performed for poly-ADP-ribosylated ARTD1 pulldown assays using our GST-fusion constructs (Fig. 16B). Poly-ADP-ribosylated ARTD1 was enriched comparably using both WT Af1521 and eAf1521 GST-fusion proteins compared to the control of GST alone. Finally, the inventors tested whether eAf1521 also detects free PAR chains that were isolated from poly- ADP-ribosylated ARTD1 digested with Proteinase K. Dot blot analysis revealed that Fc- eAf1521 indeed also recognized isolated PAR chains as efficiently as Fc-WT Af1521 (Fig. 16C). Together these data indicate that the mutations introduced into eAf1521 increased its affinity towards ADPr without altering the preference for mono-, oligo- or poly-ADP-ribosylated targets. Example 6: Immunofluorescence analysis of cellular ADP-ribosylation using the eAf1521 revealed extranuclear ADP-ribosylation under untreated conditions

The increased sensitivity observed in immunoblotting raised the question whether the same constructs would also show higher sensitivity when used for immunofluorescence detection to improve cellular ADP-ribosylation studies in cells. HeLa cells were left untreated or treated with H2O2 for 10 minutes, fixed with 4% PFA and, subsequently, incubated with Fc-WT Af1521 or Fc-eAf1521 (Fig. 4B). Treatment with H2O2 led to stronger signals of nuclear ADP-ribosylation when detected with eAf1521 compared to WT Af1521 (Fig. 4B). When subtracting the ADPr WT Af1521 signal from the same condition detected by eAf1521 , the overall enhancement was still evident, confirming the enhanced ability of eAf1521 to detect cellular ADP-ribosylation.

To further investigate the specificity of the newly generated tool towards ADP- ribosylation, the inventors pretreated HeLa cells with the ADP-ribosylation inhibitor Olaparib (Evers et ai. (2008) Clin Cancer Res, 14, 3916-3925.) to inhibit ARTD1. Inhibition of ARTD1 eliminated the nuclear signal detected in H2C>2-treated HeLa cells with both WT Af1521 and eAf1521, suggesting that WT Af1521 and eAf1521 were specifically recognizing ADP- ribosylation (Fig. 4B). Furthermore, the inventors performed a co-incubation experiment of WT Af1521 and eAf1521 in the presence of either 1 mM free ADPr or another nucleotide (e.g. GTP) on H2C>2-treated HeLa cells. The detected nuclear signal was reduced by co-incubation of both our Fc fusion proteins (i.e. Fc- WT Af1521 and -eAf1521) with ADPr but not with GTP, further confirming the detection specificity towards ADP-ribosylation. Interestingly, the inventors observed also a weak nuclear and extranuclear signal with eAf1521 in untreated HeLa cells that was barely detectable by WT AH521 (Fig. 4B). In addition, treatment with Olaparib strongly reduced the nuclear signal while the extranuclear ADP-ribosylation signal was not affected, suggesting that the observed ADP-ribosylation is likely catalyzed by different ADP- ribosyltransferases.

Example 1: Anti-ADPR antibody in cancer prognosis

In Figure 5, the inventors demonstrate that healthy kidney epithelial cells (that normally give rise to the tumor) are generally characterized by a stronger mtADPR (mitochondrial ADP- ribosylation) signal than kidney cancer tissues. Based on this, it may be speculated that the transformation of a healthy epithelial cell to a cancer cell in kidney leads to a loss in mtADPR. The loss of mtADPR becomes more impressive when the tumor becomes more aggressive, and the patients are characterized by a poor prognosis.

In Figure 6, the inventors report the case of papillary RCC, the second most frequent RCC subtype, which is very aggressive. By re-stratifying the patients, the inventors could observe that a high mtADPR signal associates with a better patient prognosis, therefore with a longer overall survival. In Figure 7, the inventors focus on breast cancer and its subtypes invasive lobular and invasive ductal. Strong mtADPR signal associates with a better overall survival of the patients in this cancer type and its subtypes.

In Figure 8, the inventors focus on ovarian cancer, particularly in its subtype high grade serous. Also in this tumor type, the inventors observe that a strong tADPR signal is a favorable factor that is associated with a better patient prognosis.

In Figure 9, the inventors focus on colon tissues. As in kidney, the inventors observed that healthy colon epithelial cells display generally a stronger mtADPR signal intensity than colon cancer tissues. In colon cancer tissues, since the inventors did not have patient survival data, they could associate the mtADPR signal intensity to the tumor stage, a parameter which indicates the development of the tumor, and that it is considered favorable when low and a bad prognostic factor when it is high. In colon cancer, the inventors observed that that a strong mtADPR signal is a characteristic of most of the early stages colon cancer, and therefore mtADPR has to be still considered a favorable prognostic factor also for this tumor type.

Example 8: eAf1521 in cancer prognosis

To better characterize the RCC cancer, the inventors used another tool to stain ADPR: instead of the anti-ADPR antibody, the inventors stained the same RCC TMA with eAf1521. The inventors repeated the same staining and scoring as performed previously with the anti-ADPR antibody.

In Figure 10, the inventors stained RCC TMA with eAf1521 and they quantified the mitochondrial ADPR (mtADPR) signal intensity. As observed with anti-ADPR antibody, the inventors observe that mtADPR signal intensity is generally stronger in healthy kidney epithelium than in RCC. In kidney cancer, moreover mtADPR signal intensity correlates with a good prognosis. However, differently than before, here the patient stratification seems to be more homogeneous than what observed previously with the anti-ADPR antibody. Furthermore, with eAf1521 the inventors are also able to stratify the patients according to the tumor stage and ISUP grade, which are two of the most important traditional prognostic markers: low grade and early stage RCCs are frequently characterized by a weak/negative mtADPR signal intensity. The ability of stratifying patients based on overall survival, tumor stage and ISUP grade makes eAf1521 a very promising tool for prognostic.

In Figure 11, the inventors show that also in the most common RCC subtype, clear cell RCC (ccRCC), a strong mtADPR signal intensity observed with eAf1521 correlates with a better patient overall survival. The same results about tumor stage and ISUP grade, was observed also in ccRCC. The fact that the same phenotype observed for RCC can be identified also in ccRCC, makes eAf1521 an even better tool for RCC prognosis than the anti-ADPR antibody. In Figure 12, the inventors show that also in the second most common RCC subtype, papillary RCC (papRCC), a strong mtADPR signal intensity observed with eAf1521 correlates with a better patient overall survival, as observed after staining with anti-ADPR antibody. Due to the small patient cohort, the inventors stratified the patients in two groups (strong, 3, red - low, 1- 2, green).

In Figure 13, the inventors show that in RCC the signal intensity obtained with anti-ADPR antibody correlates with the one given by the eAf1521. Tumors scored as strong ADPR are mostly scored as strong eAf1521, and tumors scored as weak ADPR are very frequently scored as weak with eAf1521.

Example 9: pan-ADPR binding reagent

The inventors stained the same RCC TMA with another binding reagent that recognizes pan- ADPR (Conrad et al. Mol Cancer Ther. 2020 Jan;19(1):282-291.) using the wildtype form of eAf1521. RCC patients’ stratification based on high/low pan-ADPR signal intensity showed no correlation between the pan-ADPR signal intensities and patient overall survival in the three RCC subtypes papillary (N=39), clear cell (N=221) and chromophobe RCC (ISM 5, Figure 17). The data shows that the eAf1521 is clearly superior to the wildtype Af1521 in regard to binding (i.e. intensities) and also in the prognostic histological evaluation of cancer tissue.

Example 10: Anti-polv-ADPR antibody

Furthermore, the inventors stained the same RCC TMA with an antibody that recognizes only poly-ADPR (Enzo (Kawamitsu et al. (1984), Biochemistry, 23, 3771-3777), anti-PAR) and blindly scored the extranuclear PAR signal intensity. The inventors did not observe any correlation between the PAR intensity scores and patient overall survival in the three most important RCC subtypes (papillary N=43, clear cell N=216 and chromophobe N=16 (Figure 18). This finding may indicate that among the two ADPR types, cytosolic MAR rather than PAR might be crucially involved in the RCC patient overall survival.

Discussion

The IHC stainings performed with the anti-ADPR antibody show a clear prognostic potential in several cancer types. The signal intensity observed in the stainings represent an indication of the cytoplasmic/mitochondrial ADPR (mtADPR) levels in the tissues, therefore the stronger the signal, the higher the ADPR levels, and vice versa. Associating the mtADPR levels (1- weak, 2- moderate, 3- strong) to the patients’ survival rate, the inventors observed that patients with a longer life expectation are also characterized by a stronger mtADPR signal in the cancer biopsies, and vice versa. The inventors observed a clear correlation between mtADPR levels and patient overall survival in: renal cell carcinoma (RCC), breast cancer, and ovarian cancer (high grade serous). In colon cancer (patient survival data were missing) the inventors could associate strong mtADPR signal to early stage tumors, indicating that mtADPR levels may decrease with tumor development (and therefore decreasing in patients characterized by a lower overall survival). In kidney and colon, the inventors could observe that stronger mtADPR signals were also characteristic of healthy epithelium (epithelial cells are the source of cancer in these organs) whereas cancer tissues display also weak signals. eAf1521 (ADPR) and anti-ADPR antibody can both identify healthy kidney epithelial tissues as characterized by a more frequent strong signal. eAf1521 (ADPR) and anti-ADPR antibody both indicate a strong mtADPR signal as a favourable prognostic factor, thus high ADPR levels are associated with a higher patient overall survival. eAf1521 (ADPR) and anti-ADPR antibody signals correlate. In renal cell carcinoma, eAf1521 (ADPR) staining allows for a better patient distribution/stratification, with more linear curves in the survival plot. In clear cell RCC (ccRCC), the most frequent RCC subtype, only eAf1521 (ADPR) can display a statistically significant difference in patient survival associated to staining intensity. In papillary RCC (papRCC), the second most frequent RCC subtype and very aggressive, both eAf1521 (ADPR) and anti- ADPR antibody show that the higher is the mtADPR signal intensity, the better is the prognosis for the patient. Only eAf1521 (ADPR) shows a correlation between mtADPR signal intensity and tumor stage and ISUP grade (tumor grade and stage are both traditional prognostic markers). However anti-ADPR antibody can identify an association between mtADPR signal intensity and ISUP grade in papRCC.

Materials and Methods

Cell culture

HeLa cells (Kyoto) and HEK293T cells were cultured in Dulbecco's modified Eagle's medium (DMEM), supplemented with 10% fetal calf serum (FCS) and 1% pen ici 11 i n/strepta vi d i n at 37 °C with 5% CO2. To induce ADP-ribosylation, HeLa cells were either untreated or treated with 1 mM H2O2 in PBS containing 1 mM MgCh for 10 min. Pretreatment with ADP-ribosylation inhibitors (e.g. Olaparib) was performed for 30 min before treatment with H2O2 and continued during H2C>2-treatment at the indicated concentrations: PJ-34 (10 mM), Olaparib (20 pM).

Engineering of the AH521 macro domain and ribosome selection and screening

To generate the library WT Af1521 was cloned into the pRDV vector (Dreier and Pluckthun (2012) Methods Mol Biol, 805, 261-286.), using the restriction enzymes BamHI and EcoRI (NEB). In order to introduce random mutations error prone PCR was applied using the pRDV-specific primers T7B and tolAk using either 0, 3, 6 and 10 pM of the nucleotide analogs dPTP and 8-oxo-dGTP each and Platinum Taq Polymerase (Invitrogen) as previously described (Dreier and Pluckthun (2012) Methods Mol Biol, 805, 261-286). For the PCR using the outer primers the following conditions applied: initial denaturation 3 min at 95°C, for amplification 40 cycles with 30 sec at 95°C, 30 sec at 50°C and 1 min at 72°C followed by a final extension at 72°C for 5 min.

The four PCR reactions resulting in different mutational loads were pooled and used as template for the in vitro transcription reaction using T7 RNA polymerase (Fermentas) and a home-made transcription buffer exactly as described (Dreier and Pluckthun (2012) Methods Mol Biol, 805, 261-286). The resulting RNA was purified using lllustra Microspin G-50 columns (GE). Ten pg of purified RNA were used for the in vitro translation reaction in a volume of 110 m I containing home-made S30 extract and premixZ/methionine. In vitro translation was performed at 37°C for 15 min before the reaction was stopped as described (Dreier and Pluckthun (2012) Methods Mol Biol, 805, 261-286). From the stopped in vitro translation reaction 4 times 100 mI were used for the first selection round, only 2 times 100 mI for successive rounds which were pooled after the elution step. To remove unspecific binders a pre-selection step was included in round 2 and 3, but omitted from the initial selection round using a BSA-blocked streptavidin-coated (Immunopure, Pierce) well of a 96 well Maxisorp plate (Nunc) for 30 min at 4°C. The reaction was then directly transferred to a fresh well additionally containing the biotinylated, streptavidin-immobilized ADP-ribosylated H2B peptide. The binding reaction was performed at 4°C for 1 h. In the initial round 200 nM of peptide was used, while in round 2 the target concentration was reduced to 100 nM and in round 3 and 4 to 20 nM, respectively. The selection wells were washed 6 times (2 brief washes followed by 4 washing steps with a 2 min incubation) with ice-cold 300 mI WTB buffer (50 mM Tris-acetate, 150 mM NaCI, 50 mM magnesium acetate, 0.05 % Tween-20, pH 7.5). For all other rounds the washing was prolonged to two fast washes and four washing steps of 10 min incubation each.

Elution was performed by addition of twice 100 mI EB buffer (50 mM Tris-acetate, 150 mM NaCI, 25 mM EDTA) followed by RNA purification and DNAsel treatment using the HighPure RNA isolation kit (Roche) as described previously (Dreier and Pluckthun (2012) Methods Mol Biol, 805, 261-286). Afterwards a reverse transcription was performed essentially as described (Dreier and Pluckthun (2012) Methods Mol Biol, 805, 261-286) using the recovered mRNA and subsequently amplified by PCR using the scaffold-specific primers Af1521_pR D V_fwd_2 5’-gacaaaggatccATGGAACGGCGTAC-3’ (SEQ ID NO 004) and Af1521_pRDV_ Eco_rev 5’-CTTTGAGAGGAGTCTTgaattcgga-3’ (SEQ ID NO 005) with an annealing temperature of 50°C and 35 cycles using Vent polymerase (NEB). The PCR products were cloned into the pRDV vector using the restriction enzymes BamHI and EcoRI (NEB). Respective bands were gelpurified using the Gel Purification Kit (QIAGEN) and cDNA fragments were digested with SamHI and EcoRI followed by a clean-up using the PCR purification kit (QIAGEN) and ligated into pRDV. The resulting enriched pool served as template for the next round of selection using error-prone PCR and the primers T7B and TolAk. In total four rounds of selection were performed. After the second round of selection the washes were prolonged (2 direct washes, 4x 10 min). In round 3 and 4, an off-rate selection was implemented (Dreier and Pluckthun (2012) Methods Mol Biol, 805, 261-286).

In round 3, error-prone PCR was applied to increase diversity, but no error-prone PCR was used in round 4, which instead served to efficiently enrich for generated high-affinity H2B binders. While round 1-3 were carried out with target immobilized on plates, in round 4 the selection was performed in solution MyOne T 1 streptavidin coated magnetic beads (Thermo Scientific) as previously described (Dreier and Pluckthun (2012) Methods Mol Biol, 805, 261- 286). For round 3, a completion with a 1000-fold access of competitor was performed. Afterwards, the wells were directly washed twice and additionally 4 times for 10 min. For round 4, 20 nM of biotinylated ADP-ribosylated H2B peptide in the presence of 20 mM of ADP-ribosylated ARTDIOcat (GST-fusion protein expressing the catalytic domain of ARTD10/PARP10 that is able to catalyzed auto-mono-ADP-ribosylation) was used and incubated for 1 h to increase the specificity towards the ADPr. A 30 min capture of Af1521 mutants that still were bound to the biotinylated peptide was performed using streptavidin- coated magnetic beads as previously described (Dreier and Pluckthun (2012) Methods Mol Biol, 805, 261 -286). Af 1521 mutants still binding to the biotinylated H2B peptide were eluted with EB buffer followed by RNA purification and RT-PCR. After performing the last iteration, the recovered mRNA was reversed transcribed and subsequently amplified by PCR. The PCR products were cloned into the expression vector pDST67 a pQE30 (QIAGEN) derivative (Steiner et al. (2008) J Mol Biol, 382, 1211-1227.) containing a MRGS-FMag using the restriction enzymes BamHI and Pstl. After transformation of E. coli XL-1 Blue 92 single colonies were picked from selection plates and expressed in a 96 well format in 2xYT media containing 100 pg/ml ampicillin and 1% glucose for 4 h at 37°C after induction using 0.5 mM IPTG. Cells were harvested by centrifugation for 10 min at 4000 rpm and cells were lysed by addition of 50 pi B-PERII cell lysis buffer (Pierce) as previously described (Dreier and Pluckthun (2012) Methods Mol Biol, 805, 261-286). The lysate was diluted with 950 mI PBS- TB (PBS: 137 mM NaCI, 30 mM KCI, 80 mM Na ₂ HP0 ₄ , 15 mM KH ₂ P0 _4, pH 7.4 containing 0.1 (v/v) Tween20 and 0.2% (w/v) BSA) and cleared by centrifugation for 10 min at 4000 rpm. In order to identify initial hits an ELISA was performed against the biotinylated and ADP- ribosylated H2B peptide and unmodified peptide as control. The binding was detected using the MRGS-FMag (Dreier and Pluckthun (2011) Methods Mol Biol, 687, 283-306.). Therefore 96-well MaxiSorp plates (Nunc) were coated with 100 mI 66 nM Immunopure streptavidin (Pierce) in PBS overnight at 4°C. After washing twice with PBS, the wells were blocked with PBS/0.5% BSA for 1 h at room temperature (RT). The biotinylated peptides were immobilized at a concentration of 100 nM in PBS-TB at RT for 1 h. After three washes with PBS-T, 100 mI of diluted lysate (90 mI PBS-TB plus 10 mI lysate) was added to the target- coated wells and incubated at RT for 1 h. Following three washes with PBS-T peptide-bound Af1521 mutants were detected using a mouse- anti-RGS(H)4 primary antibody (QIAGEN) followed by a goat-anti-mouse secondary antibody coupled to alkaline phosphatase (Sigma). After addition of the substrate (pNPP), absorbance at 405 nm was determined (Dreier and Pluckthun (2012) Methods Mol Biol, 805, 261-286). Ten clones were randomly selected for sequencing.

Cloning and protein purification

Randomly mutagenized Af1521 candidates were cloned into pGEX6P-1 (Addgene) using the restriction enzymes BamHI and EcoRI (NEB). Mutants of eAf1521 and WT Af1521 were obtained by site-directed mutagenesis (G42E, forward primer GAGCACGGCGAAGGGGTGGC (SEQ ID NO 006), reverse primer GCCACCCCTTCGCCGTGCTC (SED ID NO 007); R145K: forward primer

GCTGGGATAAAAGGCTGTGATCTG (SEQ ID NO 008), reverse primer

CAGATCACAGCCTTTTATCCCAGC (SEQ ID NO 009)). AH521-K35E, forward primer 5’- GCCAACGAGAGGCTGG-3’ (SEQ ID 011), reverse primer 5’-CCAGCCTCTCGTTGGC-3’ (SEQ ID 012); eAf1521-E35E, forward primer 5’-GCCAACAAGAGGCTGG-3’ (SEQ ID 013), reverse primer 5’-CCAGCCTCTTGTTGGC-3’ (SEQ ID 14); eAf1521-H44G, forward primer 5’- CTGGGGGACGCGGC-3’ (SEQ ID 015), reverse primer 5’-GCCGCGTCCCCCAG-3’ (SEQ ID 016), WT AH521-Y145R forward primer 5’-CTGGGATACGCGGCTGTG-3’ (SEQ ID 017), reverse primer 5’-CACAGCCGCGTATCCCAG-3’ (SEQ ID 018); eAf1521-R145Y, forward primer 5’-CTGGGATATACGGCTGTG-3’ (SEQ ID 019), reverse primer 5’- CACAGCCGTATATCCCAG-3’ (SEQ ID 020). His-tagged WT eAf1521 (with a N-terminal His tag) were constructed in the bacterial expression vector pET19b by GenScript (Piscataway, NJ, USA). For generation of Fc fusion constructs, fragments encoding the IL-2 secretion signal and an engineered mouse lgG2a Fc domain (described in (Gortz et ai. (2015) Sci Rep, 5, 14685), Fc domain originally from pFUSE-mlgG2ae1-Fc, InvivoGen) fused to a C-terminal HA and a HiS ₆ -tag, Af1521 WT and eAf1521 were sequence-optimized and customly synthesized (GeneArt, ThermoFisher Scientific). The fragment containing the IL-2 secretion signal and the Fc domain with HA and His tags was cloned into pcDNA5/FRT/TO (Invitrogen) using the restriction enzymes Hindi 11 and Xhol (NEB). The Af1521-encoding fragments were cloned in between the secretion signal and the Fc tag via Kpnl and BamHI sites.

Bacterial expression vectors were transformed into E. coli BL21 , and protein expression was induced by adding 1 mM IPTG at Oϋboo 0.4-0.6 for 3 h at 30 °C. Batch purification of GST- tagged or His-tagged proteins was carried out using glutathione Sepharose 4B beads (GE Healthcare) or ProBond™ Nickel-Chelating Resin (Thermo Fisher Scientific) according to the manufacturer's manual. Fc fusion domains were expressed by transfecting HEK293T cells with the mammalian expression vectors using calcium phosphate. Roughly 6 hrs after transfection, the medium was removed and replaced with fresh DM EM containing 1% of FCS. 2 days after transfection, the medium was collected and again replaced with fresh DM EM containing 1% of FCS. The collected supernatants were filtered through a 0.45 pm mesh, to eliminate residual cells. Batch purification of His-tagged Fc fusion domains from the supernatant was carried out using ProBond™ Nickel-Chelating Resin (Thermo Fisher Scientific) according to the manufacturer's manual. Expression and purification of all recombinant proteins were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) followed by Coomassie staining or immunoblotting.

Pulldown using ADP-ribosylated H2B peptide

To test the binding of the evolved Af1521 candidates towards ADP-ribosylation, biotinylated ADP-ribosylated or non-modified H2B peptides (Kistemaker et al. (2016) Angew Chem Int Ed Engl, 55, 10634-10638.) were bound to streptavidin Sepharose high-performance beads (GE Healthcare). For each pull-down, 5 pL of beads were washed three times in binding buffer (50 mM NaCI, 50 mM Tris-HCI pH 8, 0.05% NP-40) and incubated overnight at 4 °C in 1 ml_ binding buffer with 2 pg of the modified or unmodified H2B peptide. Afterwards, beads were washed three times with 1 ml_ of in incubation buffer (1x protease inhibitor cocktail (Roche), 50 mM Tris-HCI pH 8, 0.05% NP-40) containing different salt concentration (50 mM, 200 mM and 400 mM NaCI). 2 pg of the recombinant protein was incubated with the beads in 1 ml_ of incubation buffer for 3 hrs at 4 °C. After centrifugation at 1,500 g for 5 min, the supernatant containing unbound protein was added on 5 pL prewashed glutathione Sepharose 4B beads (GE Healthcare) and additionally incubated for 2 hrs at 4 °C. Subsequently, all beads were washed three times with incubation buffer before analysis by SDS-PAGE followed by Coomassie blue staining.

Crystallization, data collection and structure refinement Protein purification and crystallization

11 mg of IMAC-purified N-terminal His-tagged eAf1521 was further purified by SEC (Sephacryl-100; GE Healthcare) in 20 mM Tris pH 7.5, 300 mM NaCI, 10% glycerol, 2 mM TCEP. The main peak fractions were concentrated by ultrafiltration in Vivaspin cartridges (Sartorius). Crystallization conditions were identified using the JCSG+ crystal screen (Qiagen) and sitting drop vapor diffusion. Crystals grew at 4 °C in droplets consisting of 0.1 pL protein solution (23.1 mg/ml_ including 2 mM ADP-ribose) and 0.2 pL of well solution (25% w/v PEG3350, 0.1 M Bis-Tris, pH 5.5, and either 0.2 M (NH ₄ ) ₂ S0 ₄ or 0.2 M NaCI). Crystals were briefly transferred to cryo solution (well solution supplemented with 15% glycerol, 0.2 M NaCI, and 5 mM ADPr) and then stored under liquid nitrogen. Data collection, structure solution, and refinement

Diffraction data were collected at the Diamond synchrotron radiation source, Didcot, UK, at beamline i24, at 0.96862A. The inventors used AutoPROC (Vonrhein et al. (2011) Acta Crystallogr D Biol Crystallogr, 67, 293-302) for data processing and Phaser (McCoy et al. (2007) J Appl Crystallogr, 40, 658-674.) for phasing. The structures were solved by molecular replacement using the WT Af1521 structure (PDB ID: 2BFQ, (19)) as model template. Crystals grown in presence of NH ₄ SO ₄ (spacegroup P6 ₁ ) yielded data down to 1.23 A; however, the ADPr binding site was involved in crystal contacts. To exclude artifacts, the inventors abandoned refinement of this model. Crystals grown in presence of NaCI (spacegroup C2) diffracted to 1.82 A and the model was refined using Buster (Bricogne et al. (2018). version 2.10.3. ed. Global Phasing Ltd., Cambridge, United Kingdom.). The progress of refinement was monitored using decreasing R and R free values. The model was analyzed with Molprobity (Chen et al. (2010) Acta Crystallogr D Biol Crystallogr, 66, 12-21). Coordinates and structure factors have been deposited to the Protein Data Bank (PDB ID: 6FX7).

Surface Plasma Resonance (SPR) Measurements

SPR measurements were performed at 20 °C in HBS buffer (10 mM HEPES, 150 mM NaCI, 0.005 % Tween-20, 25 mM EDTA, pH 7.6) using a Biacore T200 instrument (GE Healthcare). Proteins were immobilized on a (multi)NTA derivatized polycarboxylate hydrogel NiHCIOOOM chip according to a protocol recommended by the chip manufacturer (Xantec, Dusseldorf, Germany). First, a 0.5 M EDTA solution (pH 8.5) was injected for 300 s, followed by a 120 s buffer injection, a surface activation step with 5 mM NiCL for 60 s, and a 120 s injection of eAf1521 (cone) and 150 s WT Af1521 (cone) respectively, both containing a hexa-His tag. This was followed by a 120 s injection of immobilization buffer. All immobilization steps were performed in immobilization buffer (10 mM HEPES, 150 mM NaCI, 0.005 % Tween-20, 50 pM EDTA) at a flow rate of 5 pl_/ min. This results in a surface density of 390-520 and 4650 RU for eAf1521 and WT Af 1521 , respectively.

In the double referenced binding experiments, two-fold dilution series of 5 and 7 concentrations of ADPr were injected for 30 s and 120 s for WT and eAf1521 , respectively, at a flow rate of 30 pL/ min in HBS buffer. ADPr concentrations were in the range of 2500 nM-39 nM for binding to WT Af1521 and 25 nM-1.25 nM for eAf1521, respectively. Sensorgrams measured for WT Af1521 at the ADPr concentrations measured for eAf1521 were not detectable, confirming that the engineering process indeed evolved a macro domain with higher affinity to ADPr. At higher ADPr concentrations, sensorgrams were deviating from 1:1 kinetics. For this reason, eAf1521 measurements were conducted with dilution series of only 3 concentrations (25 nM-6.25 nM). In order to obtain a reliable data set, 3 flow cells with immobilized eAf1521 at densities of 520 RU, 500 RU, and 390 RU, resp., were used in parallel. Additionally, WT Af1521 was measured under the conditions used for eAf1521. The chip surface was regenerated by injection of 0.5 M EDTA for 5 min. The chip was regenerated by a 5 min injection of 0.5 M EDTA in water.

Uncoated flow cell 1 of the sensor chip was used as a reference. Data were evaluated using Biacore software version 2.0.3. Sensorgrams were fitted using a 1 :1 kinetic model. Sensorgrams of Af1521 were additionally evaluated using a steady-state model. The equilibrium dissociation constant of WT Af1521 calculated for steady state (ss) conditions was 4 times larger, and an overestimation of the affinity in kinetic experiments is possible, presumably caused by rebinding as a common surface effect often observed in SPR measurements. The binding behavior of eAf1521 was governed by a 600 times slower dissociation and a 2 times faster association step compared to WT AH521. Calculation of the dissociation rate constant using KD (SS) resulted in a more reliable value (Tab. 1).

In vitro ADP-ribosylation and de-modification assays

In vitro ADP-ribosylation assays were performed based on previously described methods (Rosenthal and Hottiger (2014) Front Biosci (Landmark Ed), 19, 1041-1056., Abplanalp et al. (2018) Methods Mol Biol, 1813, 205-213). For auto-ADP-ribosylation assays recombinant ARTD1 (10 pmol) was incubated in reaction buffer (RB; 50 mM Tris-HCI pH 7.4, 4 mM MgCh and 250 mM dithiothreitol (DTT)) with 100 mM NAD ⁺ and 200 nM of double-stranded annealed 40 bp long oligomer (5'-TGCGACAACGATGAGATTGCCACT ACTTGAACCAGTGCGG-3' (SEQ ID NO 010), 5’-CCGCACTGGTTCAAGTAGTGGCAATCTCATCGTTGTCGCA-3’ (SEQ ID 021)) for 15 min at 37 °C. Recombinant ARTD8cat was incubated in RB buffer with either 100 pM NAD ⁺ or 200 nM [ ³² P] NAD ⁺ (Perkin Elmer) for 30 min at 37 °C. These reactions were stopped via the addition of SDS-buffer or by filtering through an lllustra MicroSpin G-50 column (GE Healthcare) according to the manufacturer’s protocol. De-modification assays were performed in RB buffer. For de-modification of ARTD1, the auto-modified recombinant proteins were incubated with 10 pmol PARG for 30 min at 37 °C. The hydrolysis activities of WT Af1521 or eAf1521 were tested by incubating auto-modified ARTD8cat with either 10 pmol of recombinant WT Af1521 oreAf1521 for 2 hrs at 4 °C or 37 °C. De-modification of auto-modified ARTD8cat was then visualized by SDS-PAGE and autoradiography. For generation of PAR chains poly-ADP-ribosylated ARTD1 was digested using Proteinase K at 42°C for 1 h.

ADPr-Peptide Enrichment

ADPr-Peptide enrichments were carried out as previously described (Leutert etai. (2018) Cell Rep, 24, 1916-1929 e1915.) with the following protocol modifications. After reducing potentially PARylated peptides to MARylated peptides using the enzyme PARG, the affinity enrichment of ADP-ribosylated peptides was either performed using either WT Af1521 or eAf1521 for 2 hrs at 4 °C. Both enriched samples were then prepared for MS analysis as described previously (Martello et al. (2016) Nat Commun, 7, 12917.).

Liquid Chromatography and Mass Spectrometry Analysis

Identification of ADP-ribosylated peptides from untreated and H2C>2-treated HeLa cells was performed on an Orbitrap Fusion Tribrid mass spectrometer (Thermo Fisher Scientific), coupled to an ACQUITY M class UPLC liquid chromatograph (Waters). The inventors applied an ADPr product-dependent analysis called HCD-PP-EThcD (Bilan et al. (2017) Anal Chem, 89, 1523-1530.). Briefly, the method includes high-energy data-dependent HCD, followed by high-quality HCD and EThcD MS/MS when two or more ADPr fragment peaks (136.0623, 250.0940, 348.07091, and 428.0372) were observed in the HCD scan. A detailed description of the MS parameters can be found in (Bilan et al. (2017) Anal Chem, 89, 1523-1530.). Solvent compositions in channels A and B were 0.1% formic acid in water and 0.1% formic acid in acetonitrile, respectively. Peptides were loaded onto a nanoEase M/Z Symmetry (Waters) trap column, 180 pm x 20 mm, packed with C18 material, 5 pm, 100 A, and separated on an analytical nanoEase M/Z HSS T3 Column (Waters, 75 pm x 200 mm) packed with reverse- phase C18 material (Waters, 1.8 pm, 100 A). Peptides were eluted over 110 min at a flow rate of 300 nL/min. A linear elution gradient protocol from 3% to 25% B for 95 min, followed by 35% B for 5 min and a wash step at 95% B for 5 min, respectively, was used.

Quantification and statistical analysis Mass Spectrometry Data Analysis and Label Free Quantification

MS1 -based label-free quantification (LFQ) was performed by applying Progenesis Ql for Proteomics software (v. 3.0.6039.34628, Nonlinear Dynamics, Purham, NC) with default settings and the following exceptions. Peptide ions were filtered for charges ranging from +2 - +5. A maximum of the top 5 ranked MS/MS spectra per peptide ion were exported with the most intense 200 peaks per spectrum with activated charge-deconvolution and deisotoping option as a Mascot generic formatted file (MGF). MS/MS spectra were searched with Mascot for each type of fragmentation (HCD and EThcD). Mascot searches were carried out as previously described (Leutert et al. (2018) Cell Rep, 24, 1916-1929 e1915) with the following protocol modifications. The MGFs were searched against the target-decoy UniProtKB human database (taxonomy 9606, canonical sequences and reviewed entries only, downloaded on 2019/07/06). N-Terminal protein acetylation was set as a variable modification. S, R, K, D, E and Y residues were set as variable ADPr acceptor sites. The neutral losses from the ADPr 249.0862 Da, 347.0631 Da, and 583.0829 Da were scored in HCD fragment ion spectra (Gehrig et al., manuscript in preparation). The Mascot search results were imported into Scaffold and filtered for protein and peptide FDR values of 2% and 1% respectively. When multiple precursors were observed for the same peptide, the values were summed up to obtain the total intensity level of the peptide.

For amino acid acceptor site analysis on protein and peptide level, MS and MS/MS spectra were converted to Mascot generic format (MGF) using Proteome Discoverer, v2.1 (Thermo Fisher Scientific, Bremen, Germany). Separate MGF files were created from the raw file for each type of fragmentation (HCD and EThcD) using a dedicated rule in the converter control (Barkow-Oesterreicher et al. (2013) Source Code Biol Med, 8:3). . Mascot was used as described above and the Mascot search results were imported into Scaffold 4 software (version 4.8.4). Peptides were considered correctly identified when a Mascot score >15 and a Mascot delta score >5 were obtained. These settings ensured a FDR lower than 1% at the PSM level. For post-translational modification (PTM) localization site probability estimation ScaffoldPTM software (Version 3.2.0) was used invoking the site localization algorithm Ascore (Beausoleil et al. (2006) Nat Biotechnol, 24, 1285-1292.), including neutral losses for HCD fragment ion spectra. Razor peptides, spectra that belong to more than one peptide, were not included for further analysis on the peptide and protein level. Proteins that contained the same peptides and could not be differentiated based on MS/MS analysis were only reported once for further analysis on the protein level. For the ADP-ribosylation site analyses, peptides identified with EThcD fragmentation, having a localization score >95%, were used if not stated otherwise. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE (Perez-Riverol et al. (2019) Nucleic Acids Res, 47, D442-D450) partner repository with the dataset identifier PXD016686.

Bioinformatic Analyses

Statistical analysis, scatter plots, volcano plot analysis were performed using Prism 8. Normalized LFQ intensities were imported. For statistical analysis, the Iog10 transformed and normalized MS1 signal intensity of 3 biochemical replicates were compared using a student t- test with *, p<0.05; **, p <0.005; ***, p < 0.0005, two-sided. Venn diagrams were generated using biovenn (http://www.biovenn.nl).

Pull-down using PARylated proteins

To test the binding of our eAf1521 towards PARylation, recombinant His-tagged ARTD1 was in vitro poly-ADP-ribosylated as described above. 25 pmol of automodified ARTD1 was incubated with 125 pmol of either WT Af 1521 , eAf1521 or GST in 1 ml binding buffer (1% BSA,

50 mM NaCI, 50 mM Tris-HCI pH 8, 0.05% NP-40) at 4°C for 1 h. 10 pl_ prewashed glutathione

Sepharose 4B beads (GE Healthcare) were added and additionally incubated for 1 h at 4°C.

After centrifugation at 1,500 g for 5 min, the supernatant containing unbound PARylated

ARTD1 was added on 10 pL prewashed ProBond™ Nickel-Chelating Resin (Thermo Fisher

Scientific) and additionally incubated for 2 h at 4°C. Subsequently, all beads were washed three times with binding buffer and one additional time with binding buffer lacking BSA before analysis by immunoblotting.

Immunoblotting

Untreated or treated HeLa cells were lysed with RIPA buffer (50 mM Tris-HCI pH 7.4, 400 mM NaCI, 1% NP-40, 0.1% Na-deoxycholate, 1x protease inhibitor cocktail (Roche), 10 mM PJ- 34), sonicated and centrifuged at 16,000 g for 10 min.

HeLa lysates or recombinant proteins were mixed with SDS-buffer, boiled at 95 °C for 5 min. After separation by SDS-PAGE, a wet-transfer onto PVDF membrane was performed. The membranes were blocked with 5% milk in TBS-T for 1 hr at room temperature. Primary antibodies were diluted in 5% milk in TBS-T and incubated at 4 °C overnight. After three washes with TBS-T for 5 min, the secondary antibody (in TBS-T) was incubated for 1 hr at RT and the membranes were additionally washed 3x with TBS-T. For dot blot analysis, proteins were vacuum blotted onto a nitrocellulose membrane that was further blocked in milk and stained with antibodies as described above.

The bands or dots were visualized using the Odyssey infrared imaging system (LICOR). The following primary and secondary antibodies were used for immunoblot and dotblot analyses: anti-tetra-His (1:1000, Qiagen), Fc-WT Af1521 (1:400, 500 ng/mL,), Fc- eAf1521 (1:400, 500 ng/mL), IRDye 800CW goat anti-rabbit IgG (1 :15,000, LI-COR, P/N 925- 32211), and IRDye 680RD goat anti-mouse IgG (1:15,000, LI-COR, P/N 925-68070). Molecular weights are indicated by the PageRuler Plus Prestained Protein Ladder (Thermo Scientific).

Immunofluorescence

For immunofluorescence (IF) experiments, HeLa cells were grown on glass coverslips. After treatment, HeLa cells were fixed with 4% PFA for 15 min at room temperature and permeabilized for 10 min at room temperature in PBS supplemented with 0.2% Triton-X100 (Sigma Aldrich). After blocking the cells with PBS supplemented with 10% of goat serum for 1 hr, the cells were incubated with the primary antibody (diluted in blocking solution) overnight at 4 °C. The cells were washed 2x with PBS and subsequently incubated with the secondary antibody (diluted in blocking solution) for 2 h at room temperature. After two 5 min washes with PBS, the cells were incubated with 0.1 pg/mL DAPI in PBS for 20 min at room temperature. The cells were additionally washed twice with PBS for 5 min. Then the coverslips were briefly washed in distilled water and mounted on glass slides using 5.5 pL Mowiol solution per coverslip. The following primary and secondary reagents were used for IF analyses: Fc-WT Af1521 (1:400, 500 ng/mL), Fc-eAf1521 (1 :400, 500 ng/mL). Tumor Micro Arrays (TMAs)

Human cancer tissue microarrays (TMA) containing breast carcinomas, ovarian carcinomas, renal cell carcinomas, colorectal carcinomas (one punch of 0.6 mm diameter per sample) were constructed as described (Kononen et al. , Nat Med. 1998 Jul;4(7):844-7.). TMA sections (2.5 pm) were on glass slides. The samples were retrieved from the archives of the Department of Pathology and Molecular Pathology, University Hospital Zurich (Zurich, Switzerland) between the years 1993 to 2013. All patients were patients of the USZ. For each tumor, one representative tumor tissue block, with a minimum of 1 cm tumor diameter, was re-evaluated using hematoxylin and eosin-stained sections. Tumor samples with necrosis and high content of inflammatory cells were excluded. Only those cases with representative tumor regions that contained at least 70% tumor cells were selected for the TMA construction. This study was approved by the local commission of ethics (BASEC-Nr_2016-00811). All tumors were reviewed by pathologists of the Department of Pathology and Molecular Pathology specialized in their field. Classification, grading and staging was performed according to current TNM and WHO classification. Tissue sections were pre-treated with Tris-EDTA-Borate Buffer at 100°C for 60 min (CC1 standard protocol, Ventana), incubated with ADPR antibody for 44 min at 37°C. Tumors were considered strong (+3), moderate (+2) or weak/negative (+1) for cytoplasmic ADP ribosylation in the tumor cells.

The TMAs were stained in University Hospital Zurich (USZ) with Ultra Discovery Ventana machine (using the standard protocol) by using the anti-ADPR antibody. eAf1521 was run on the Leica Bond machine.

The scoring of the immunohistochemistry (IHC) signal was based on the following parameters:

• The inventors focused only on ADPR signal intensity of extranuclear (cytoplasmic/mitochondrial) signal (mtADPR).

• The inventors scored only the mtADPR intensity of tumor cells (infiltrating cells such as fibroblasts, immune cells, etc. are not considered).

• The scores 1-2-3-(4-5) were assigned according to the signal intensity of the staining [1 (weak/negative), 2 (moderate), 3 (strong)].

• The inventors assigned the scores blindly (without knowing the patient data) to each biopsy.

• The inventors received the approval for the scoring methodology ex post by the head of the department of pathology, University Hospital Zurich. Staining protocol

Tumor tissue micro-arrays containing formalin-fixed paraffin-embedded tissue blocks were cut and then stained in an Ultra Discovery device from Ventana. The tissue was pre-treated using the CC1 standard protocol (60 min at 100°C in Tris-EDTA/borate buffer). The polyclonal rabbit anti-ADPR antibody was used at 1 pg/ml (1:500 dilution). Incubation time was 30 min. The staining was revealed using the UltraMap-Rabbit DAB kit. For eAf1521 staining and the tissue was pre-treated. The eAf1521-Fc was used at 0.8 pg/ml. Stained slides were scanned using a NanoZoomer (Hamamatsu Photonics, Shizuoka, Japan).

Anti-ADPR antibody

For generation of the polyclonal anti-ADPR antibody, rabbits were immunized with a terminal peptide from histone H2B tail chemically modified at one amino acid side chain to carry an analogue of ADP-ribose. Similar syntheses were used in e.g. (Liu et al. Org Biomol Chem. 2019 Jun 5 ; 17(22) : 5460-5474 ; Liu et al. Angew Chem Int Ed Engl. 2018 Feb 5;57(6):1659- 1662; and van der Heden van Noort et al., J Am Chem Soc. 2010 Apr 14;132(14):5236-40.). The immune serum was then negatively purified on an affinity column carrying the unmodified peptide and subsequently positively purified on an affinity column carrying the modified peptide.

Sequence of eAf 1521

MEVLFEAKVGDITLRLAQGDITQYPAKAIVNAANERLEHGGGVAYAIAKACAGDAGL YTEISK KAMREQFGRDCIDHGEVVVTPAMNLEERGIKYVLHTVGPVCGGMWSGELKEKLYKAFLGP LEKAEEMGVESIAFPAVSAGIRGCDLEKVVETFLEAVKDFKGSAVKEVALVIYDRKSAEV ALK VFERSL (SEQ ID NO 001)

Table 1:

Kinetic data for binding of ADPr to eAf1521 and Af1521 WT, respectively A dissociation rate constant, /c _0ff , was calculated by using KD from steady state measurement and the association rate constant, k _on . Statistical error from fitting ss (steady state).

Probe

Probe Model k _off / s ¹ KΏI nM Y ² density

WT M1521 4650 1:1 2.36±0.02 0.678±0.003 2879±9 0.028 33.3 ss 9710±1950 0.068 93.2 eAf 1521 520 1:1 5.94±0.02 0.001115±0.000002 1.88±0.01 0.042 9.6 eAfl521 500 1:1 5.63±0.06 0.001203±0.000007 2.14±0.01 0.041 10.2 eAf 1521 390 1:1 5.15±0.04 0.001000±0.000006 1.94±0.01 0.032 9.5

43

SUBSTITUTE SHEET RULE 26