Assay technologies facilitating drug discovery for ADP-ribosyl writers, readers and erasers.

ADP-ribosylation is a post-translational modification catalyzed by writer enzymes - ADP-ribosyltransferases. The modification is part of many signaling events, can modulate the function and stability of target proteins, and often results in the recruitment of reader proteins that bind to the ADP-ribosyl groups. Erasers are integral actors in these signaling events and reverse the modification. ADP-ribosylation can be targeted with therapeutics and many inhibitors against writers exist, with some being in clinical use. Inhibitors against readers and erasers are sparser and development of these has gained momentum only in recent years. Drug discovery has been hampered by the lack of specific tools, however many significant advances in the methods have recently been reported. We discuss assays used in the field with a focus on methods allowing efficient identification of small molecule inhibitors and profiling against enzyme families. While human proteins are focused, the methods can be also applied to bacterial toxins and virus encoded erasers that can be targeted to treat infectious diseases in the future.

Reconstitution of the DTX3L-PARP9 complex reveals determinants for high-affinity heterodimerization and multimeric assembly.

Ubiquitination and ADP-ribosylation are post-translational modifications that play major roles in pathways including the DNA damage response and viral infection. The enzymes responsible for these modifications are therefore potential targets for therapeutic intervention. DTX3L is an E3 Ubiquitin ligase that forms a heterodimer with PARP9. In addition to its ubiquitin ligase activity, DTX3L-PARP9 also acts as an ADP-ribosyl transferase for Gly76 on the C-terminus of ubiquitin. NAD+-dependent ADP-ribosylation of ubiquitin by DTX3L-PARP9 prevents ubiquitin from conjugating to protein substrates. To gain insight into how DTX3L-PARP9 generates these post-translational modifications, we produced recombinant forms of DTX3L and PARP9 and studied their physical interactions. We show the DTX3L D3 domain (230-510) mediates the interaction with PARP9 with nanomolar affinity and an apparent 1 : 1 stoichiometry. We also show that DTX3L and PARP9 assemble into a higher molecular weight oligomer, and that this is mediated by the DTX3L N-terminal region (1-200). Lastly, we show that ADP-ribosylation of ubiquitin at Gly76 is reversible in vitro by several Macrodomain-type hydrolases. Our study provides a framework to understand how DTX3L-PARP9 mediates ADP-ribosylation and ubiquitination through both intra- and inter-subunit interactions.

The SARS-CoV-2 Nsp3 macrodomain reverses PARP9/DTX3L-dependent ADP-ribosylation induced by interferon signaling.

SARS-CoV-2 nonstructural protein 3 (Nsp3) contains a macrodomain that is essential for coronavirus pathogenesis and is thus an attractive target for drug development. This macrodomain is thought to counteract the host interferon (IFN) response, an important antiviral signalling cascade, via the reversal of protein ADP-ribosylation, a posttranslational modification catalyzed by host poly(ADP-ribose) polymerases (PARPs). However, the main cellular targets of the coronavirus macrodomain that mediate this effect are currently unknown. Here, we use a robust immunofluorescence-based assay to show that activation of the IFN response induces ADP-ribosylation of host proteins and that ectopic expression of the SARS-CoV-2 Nsp3 macrodomain reverses this modification in human cells. We further demonstrate that this assay can be used to screen for on-target and cell-active macrodomain inhibitors. This IFN-induced ADP-ribosylation is dependent on PARP9 and its binding partner DTX3L, but surprisingly the expression of the Nsp3 macrodomain or the deletion of either PARP9 or DTX3L does not impair IFN signaling or the induction of IFN-responsive genes. Our results suggest that PARP9/DTX3L-dependent ADP-ribosylation is a downstream effector of the host IFN response and that the cellular function of the SARS-CoV-2 Nsp3 macrodomain is to hydrolyze this end product of IFN signaling, rather than to suppress the IFN response itself.

Activities and binding partners of E3 ubiquitin ligase DTX3L and its roles in cancer.

Ubiquitination is a protein post-translational modification that affects protein localisation, stability and interactions. E3 ubiquitin ligases regulate the final step of the ubiquitination reaction by recognising target proteins and mediating the ubiquitin transfer from an E2 enzyme. DTX3L is a multi-domain E3 ubiquitin ligase in which the N-terminus mediates protein oligomerisation, a middle D3 domain mediates the interaction with PARP9, a RING domain responsible for recognising E2 ∼ Ub and a DTC domain has the dual activity of ADP-ribosylating ubiquitin and mediating ubiquitination. The activity of DTX3L is known to be modulated by at least two different factors: the concentration of NAD+, which dictates if the enzyme acts as a ligase or as an ADP-ribosyltransferase, and its binding partners, which affect DTX3L activity through yet unknown mechanisms. In light of recent findings it is possible that DTX3L could ubiquitinate ADP-ribose attached to proteins. Different DTX3L-protein complexes have been found to be part of multiple signalling pathways through which they promote the adhesion, proliferation, migration and chemoresistance of e.g. lymphoma, glioma, melanoma, and prostate cancer. In this review, we have covered the literature available for the molecular functions of DTX3L especially in the context of cancer biology, different pathways it regulates and how these relate to its function as an oncoprotein.

Design, synthesis and evaluation of inhibitors of the SARS-CoV-2 nsp3 macrodomain.

A series of amino acid based 7H-pyrrolo[2,3-d]pyrimidines were designed and synthesized to discern the structure activity relationships against the SARS-CoV-2 nsp3 macrodomain (Mac1), an ADP-ribosylhydrolase that is critical for coronavirus replication and pathogenesis. Structure activity studies identified compound 15c as a low-micromolar inhibitor of Mac1 in two ADP-ribose binding assays. This compound also demonstrated inhibition in an enzymatic assay of Mac1 and displayed a thermal shift comparable to ADPr in the melting temperature of Mac1 supporting binding to the target protein. A structural model reproducibly predicted a binding mode where the pyrrolo pyrimidine forms a hydrogen bonding network with Asp22 and the amide backbone NH of Ile23 in the adenosine binding pocket and the carboxylate forms hydrogen bonds to the amide backbone of Phe157 and Asp156, part of the oxyanion subsite of Mac1. Compound 15c also demonstrated notable selectivity for coronavirus macrodomains when tested against a panel of ADP-ribose binding proteins. Together, this study identified several low MW, low µM Mac1 inhibitors to use as small molecule chemical probes for this potential anti-viral target and offers starting points for further optimization.

Macrodomain Binding Compound MRS 2578 Inhibits Alphavirus Replication.

Alphaviruses are positive-strand RNA viruses causing febrile disease. Macrodomain-containing proteins, involved in ADP-ribose-mediated signaling, are encoded by both host cells and several virus groups, including alphaviruses. In this study, compound MRS 2578 that targets the human ADP-ribose glycohydrolase MacroD1 inhibited Semliki Forest virus production as well as viral RNA replication and replicase protein expression. The inhibitor was similarly active in alphavirus trans-replication systems, indicating that it targets the viral RNA replication stage.

Analogs of TIQ-A as inhibitors of human mono-ADP-ribosylating PARPs.

The scaffold of TIQ-A, a previously known inhibitor of human poly-ADP-ribosyltransferase PARP1, was utilized to develop inhibitors against human mono-ADP-ribosyltransferases through structure-guided design and activity profiling. By supplementing the TIQ-A scaffold with small structural changes, based on a PARP10 inhibitor OUL35, selectivity changed from poly-ADP-ribosyltransferases towards mono-ADP-ribosyltransferases. Binding modes of analogs were experimentally verified by determining complex crystal structures with mono-ADP-ribosyltransferase PARP15 and with poly-ADP-ribosyltransferase TNKS2. The best analogs of the study achieved 10-20-fold selectivity towards mono-ADP-ribosyltransferases PARP10 and PARP15 while maintaining micromolar potencies. The work demonstrates a route to differentiate compound selectivity between mono- and poly-ribosyltransferases of the human ARTD family.

Structural basis for DNA break recognition by ARTD2/PARP2.

Human ARTD2 (or PARP2) is an ADP-ribosyltransferase, which is catalytically activated by binding to damaged DNA. ARTD2 subsequently ADP-ribosylates itself and other proteins, initiating a cascade of events leading to DNA repair. In contrast to ARTD1, the founding member of the enzyme family, ARTD2 does not have specialized zinc-fingers for detecting DNA damage. The domain organization of ARTD2 includes disordered N-terminus, WGR and catalytic domains. However, the N-terminus of ARTD2 is not strictly required for the DNA dependent activity. While it is known that ARTD2 requires the WGR domain for efficient DNA binding and subsequent catalytic activation, the mechanism of DNA damage detection and subsequent catalytic activation are not completely understood. Here, we report crystal structures of ARTD2 WGR domain bound to double-strand break mimicking DNA oligonucleotides. Notably, the crystal structures revealed DNA binding mode of ARTD2 involving DNA end to end interaction. Structures demonstrate how ARTD2 recognizes nicked DNA, how it interacts with the 5'-phosphate group, and how it can mediate joining of DNA ends in vitro. Extensive mutagenesis of the ARTD2-DNA interface combined with activity, binding, and stoichiometry measurements demonstrate that the WGR domain is the key for DNA break detection.

Inhibitor screening assay for neurexin-LRRTM adhesion protein interaction involved in synaptic maintenance and neurological disorders.

Synaptic adhesion molecules, including presynaptic neurexins (NRXNs) and post-synaptic leucine-rich repeat transmembrane (LRRTM) proteins are important for development and maintenance of brain neuronal networks. NRXNs are probably the best characterized synaptic adhesion molecules, and one of the major presynaptic organizer proteins. The LRRTMs were found as ligands for NRXNs. Many of the synaptic adhesion proteins have been linked to neurological cognitive disorders, such as schizophrenia and autism spectrum disorders, making them targets of interest for both biological studies, and towards drug development. Therefore, we decided to develop a screening method to target the adhesion proteins, here the LRRTM-NRXN interaction, to find small molecule probes for further studies in cellular settings. To our knowledge, no potent small molecule compounds against the neuronal synaptic adhesion proteins are available. We utilized the AlphaScreen technology, and developed an assay targeting the NRXN-LRRTM2 interaction. We carried out screening of 2000 compounds and identified hits with moderate IC50-values. We also established an orthogonal in-cell Western blot assay to validate hits. This paves way for future development of specific high affinity compounds by further high throughput screening of larger compound libraries using the methods established here. The method could also be applied to screening other NRXN-ligand interactions.

Adenosine analogs bearing phosphate isosteres as human MDO1 ligands.

The human O-acetyl-ADP-ribose deacetylase MDO1 is a mono-ADP-ribosylhydrolase involved in the reversal of post-translational modifications. Until now MDO1 has been poorly characterized, partly since no ligand is known besides adenosine nucleotides. Here, we synthesized thirteen compounds retaining the adenosine moiety and bearing bioisosteric replacements of the phosphate at the ribose 5'-oxygen. These compounds are composed of either a squaryldiamide or an amide group as the bioisosteric replacement and/or as a linker. To these groups a variety of substituents were attached such as phenyl, benzyl, pyridyl, carboxyl, hydroxy and tetrazolyl. Biochemical evaluation showed that two compounds, one from both series, inhibited ADP-ribosyl hydrolysis mediated by MDO1 in high concentrations.

Development and structural analysis of adenosine site binding tankyrase inhibitors.

Tankyrases 1 and 2, the specialized members of the ARTD protein family, are druggable biotargets whose inhibition may have therapeutic potential against cancer, metabolic disease, fibrotic disease, fibrotic wound healing and HSV viral infections. We have previously identified a novel tankyrase inhibitor scaffold, JW55, and showed that it reduces mouse colon adenoma formation in vivo. Here we expanded the scaffold and profiled the selectivity of the compounds against a panel of human ARTDs. The scaffold also enables a fine modulation of selectivity towards either tankyrase 1 or tankyrase 2. In order to get insight about the binding mode of the inhibitors, we solved crystal structures of the compounds in complex with tankyrase 2. The compounds bind to the adenosine pocket of the catalytic domain and cause changes in the protein structure that are modulated by the chemical modifications of the compounds. The structural analysis allows further rational development of this compound class as a potent and selective tankyrase inhibitor.

Structural basis for regulation of the human acetyl-CoA thioesterase 12 and interactions with the steroidogenic acute regulatory protein-related lipid transfer (START) domain.

Acetyl-CoA plays a fundamental role in cell signaling and metabolic pathways, with its cellular levels tightly controlled through reciprocal regulation of enzymes that mediate its synthesis and catabolism. ACOT12, the primary acetyl-CoA thioesterase in the liver of human, mouse, and rat, is responsible for cleavage of the thioester bond within acetyl-CoA, producing acetate and coenzyme A for a range of cellular processes. The enzyme is regulated by ADP and ATP, which is believed to be mediated through the ligand-induced oligomerization of the thioesterase domains, whereby ATP induces active dimers and tetramers, whereas apo- and ADP-bound ACOT12 are monomeric and inactive. Here, using a range of structural and biophysical techniques, it is demonstrated that ACOT12 is a trimer rather than a tetramer and that neither ADP nor ATP exert their regulatory effects by altering the oligomeric status of the enzyme. Rather, the binding site and mechanism of ADP regulation have been determined to occur through two novel regulatory regions, one involving a large loop that links the thioesterase domains (Phe(154)-Thr(178)), defined here as RegLoop1, and a second region involving the C terminus of thioesterase domain 2 (Gln(304)-Gly(326)), designated RegLoop2. Mutagenesis confirmed that Arg(312) and Arg(313) are crucial for this mode of regulation, and novel interactions with the START domain are presented together with insights into domain swapping within eukaryotic thioesterases for substrate recognition. In summary, these experiments provide the first structural insights into the regulation of this enzyme family, revealing an alternate hypothesis likely to be conserved throughout evolution.

Structure-based design, synthesis and evaluation in vitro of arylnaphthyridinones, arylpyridopyrimidinones and their tetrahydro derivatives as inhibitors of the tankyrases.

The tankyrases are members of the PARP superfamily; they poly(ADP-ribosyl)ate their target proteins using NAD(+) as a source of electrophilic ADP-ribosyl units. The three principal protein substrates of the tankyrases (TRF1, NuMA and axin) are involved in replication of cancer cells; thus inhibitors of the tankyrases may have anticancer activity. Using structure-based drug design and by analogy with known 3-arylisoquinolin-1-one and 2-arylquinazolin-4-one inhibitors, series of arylnaphthyridinones, arylpyridinopyrimidinones and their tetrahydro-derivatives were synthesised and evaluated in vitro. 7-Aryl-1,6-naphthyridin-5-ones, 3-aryl-2,6-naphthyridin-1-ones and 3-aryl-2,7-naphthyridin-1-ones were prepared by acid-catalysed cyclisation of the corresponding arylethynylpyridinenitriles or reaction of bromopyridinecarboxylic acids with ß-diketones, followed by treatment with NH3. The 7-aryl-1,6-naphthyridin-5-ones were methylated at 1-N and reduced to 7-aryl-1-methyl-1,2,3,4-tetrahydro-1,6-naphthyridin-5-ones. Cu-catalysed reaction of benzamidines with bromopyridinecarboxylic acids furnished 2-arylpyrido[2,3-d]pyrimidin-4-ones. Condensation of benzamidines with methyl 1-benzyl-4-oxopiperidine-3-carboxylate and deprotection gave 2-aryl-5,6,7,8-tetrahydropyrido[4,3-d]pyrimidin-4-ones, aza analogues of the known inhibitor XAV939. Introduction of the ring-N in the arylnaphthyridinones and the arylpyridopyrimidinones caused >1000-fold loss in activity, compared with their carbocyclic isoquinolinone and quinazolinone analogues. However, the 7-aryl-1-methyl-1,2,3,4-tetrahydro-1,6-naphthyridin-5-ones showed excellent inhibition of the tankyrases, with some examples having IC50=2nM. One compound (7-(4-bromophenyl)-1-methyl-1,2,3,4-tetrahydro-1,6-naphthyridin-5-one) showed 70-fold selectivity for inhibition of tankyrase-2 versus tankyrase-1. The mode of binding was explored through crystal structures of inhibitors in complex with tankyrase-2.

Exploration of the nicotinamide-binding site of the tankyrases, identifying 3-arylisoquinolin-1-ones as potent and selective inhibitors in vitro.

Tankyrases-1 and -2 (TNKS-1 and TNKS-2) have three cellular roles which make them important targets in cancer. Using NAD(+) as a substrate, they poly(ADP-ribosyl)ate TRF1 (regulating lengths of telomeres), NuMA (facilitating mitosis) and axin (in wnt/ß-catenin signalling). Using molecular modelling and the structure of the weak inhibitor 5-aminoiso quinolin-1-one, 3-aryl-5-substituted-isoquinolin-1-ones were designed as inhibitors to explore the structure-activity relationships (SARs) for binding and to define the shape of a hydrophobic cavity in the active site. 5-Amino-3-arylisoquinolinones were synthesised by Suzuki-Miyaura coupling of arylboronic acids to 3-bromo-1-methoxy-5-nitro-isoquinoline, reduction and O-demethylation. 3-Aryl-5-methylisoquinolin-1-ones, 3-aryl-5-fluoroisoquinolin-1-ones and 3-aryl-5-methoxyisoquinolin-1-ones were accessed by deprotonation of 3-substituted-N,N,2-trimethylbenzamides and quench with an appropriate benzonitrile. SAR around the isoquinolinone core showed that aryl was required at the 3-position, optimally with a para-substituent. Small meta-substituents were tolerated but groups in the ortho-positions reduced or abolished activity. This was not due to lack of coplanarity of the rings, as shown by the potency of 4,5-dimethyl-3-phenylisoquinolin-1-one. Methyl and methoxy were optimal at the 5-position. SAR was rationalised by modelling and by crystal structures of examples with TNKS-2. The 3-aryl unit was located in a large hydrophobic cavity and the para-substituents projected into a tunnel leading to the exterior. Potency against TNKS-1 paralleled potency against TNKS-2. Most inhibitors were highly selective for TNKSs over PARP-1 and PARP-2. A range of highly potent and selective inhibitors is now available for cellular studies.

Discovery of potent and selective nonplanar tankyrase inhibiting nicotinamide mimics.

Diphtheria toxin-like ADP-ribosyltransferases catalyse a posttranslational modification, ADP-ribosylation and form a protein family of 17 members in humans. Two of the family members, tankyrases 1 and 2, are involved in several cellular processes including mitosis and Wnt/ß-catenin signalling pathway. They are often over-expressed in cancer cells and have been linked with the survival of cancer cells making them potential therapeutic targets. In this study, we identified nine tankyrase inhibitors through virtual and in vitro screening. Crystal structures of tankyrase 2 with the compounds showed that they bind to the nicotinamide binding site of the catalytic domain. Based on the co-crystal structures we designed and synthesized a series of tetrahydroquinazolin-4-one and pyridopyrimidin-4-one analogs and were subsequently able to improve the potency of a hit compound almost 100-fold (from 11 µM to 150 nM). The most potent compounds were selective towards tankyrases over a panel of other human ARTD enzymes. They also inhibited Wnt/ß-catenin pathway in a cell-based reporter assay demonstrating the potential usefulness of the identified new scaffolds for further development.

Managing macromolecular crystallographic data with a laboratory information management system.

Protein crystallography is an established method to study the atomic structures of macromolecules and their complexes. A prerequisite for successful structure determination is diffraction-quality crystals, which may require extensive optimization of both the protein and the conditions, and hence projects can stretch over an extended period, with multiple users being involved. The workflow from crystallization and crystal treatment to deposition and publication is well defined, and therefore an electronic laboratory information management system (LIMS) is well suited to management of the data. Completion of the project requires key information on all the steps being available and this information should also be made available according to the FAIR principles. As crystallized samples are typically shipped between facilities, a key feature to be captured in the LIMS is the exchange of metadata between the crystallization facility of the home laboratory and, for example, synchrotron facilities. On completion, structures are deposited in the Protein Data Bank (PDB) and the LIMS can include the PDB code in its database, completing the chain of custody from crystallization to structure deposition and publication. A LIMS designed for macromolecular crystallography, IceBear, is available as a standalone installation and as a hosted service, and the implementation of key features for the capture of metadata in IceBear is discussed as an example.

High-throughput screening assay for PARP-HPF1 interaction inhibitors to affect DNA damage repair.

ADP-ribosyltransferases PARP1 and PARP2 play a major role in DNA repair mechanism by detecting the DNA damage and inducing poly-ADP-ribosylation dependent chromatin relaxation and recruitment of repair proteins. Catalytic PARP inhibitors are used as anticancer drugs especially in the case of tumors arising from sensitizing mutations. Recently, a study showed that Histone PARylation Factor (HPF1) forms a joint active site with PARP1/2. The interaction of HPF1 with PARP1/2 alters the modification site from Aspartate/Glutamate to Serine, which has been shown to be a key ADP-ribosylation event in the context of DNA damage. Therefore, disruption of PARP1/2-HPF1 interaction could be an alternative strategy for drug development to block the PARP1/2 activity. In this study, we describe a FRET based high-throughput screening assay to screen inhibitor libraries against PARP-HPF1 interaction. We optimized the conditions for FRET signal and verified the interaction by competing the FRET pair in multiple ways. The assay is robust and easy to automate. Validatory screening showed the robust performance of the assay, and we discovered two compounds Dimethylacrylshikonin and Alkannin, with µM inhibition potency against PARP1/2-HPF1 interaction. The assay will facilitate the discovery of inhibitors against HPF1-PARP1/2 complex and to develop potentially new effective anticancer agents.

Discovery of 2-Amide-3-methylester Thiophenes that Target SARS-CoV-2 Mac1 and Repress Coronavirus Replication, Validating Mac1 as an Antiviral Target.

The COVID-19 pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus has made it clear that further development of antiviral therapies will be needed. Here, we describe small-molecule inhibitors for SARS-CoV-2 Mac1, which counters ADP-ribosylation-mediated innate immune responses. Three high-throughput screening hits had the same 2-amide-3-methylester thiophene scaffold. We studied the compound binding mode using X-ray crystallography, allowing us to design analogues. Compound 27 (MDOLL-0229) had an IC50 of 2.1 µM and was selective for CoV Mac1 proteins after profiling for activity against a panel of viral and human proteins. The improved potency allowed testing of its effect on virus replication, and indeed, 27 inhibited replication of both murine hepatitis virus (MHV) prototypes CoV and SARS-CoV-2. Sequencing of a drug-resistant MHV identified mutations in Mac1, further demonstrating the specificity of 27. Compound 27 is the first Mac1-targeted small molecule demonstrated to inhibit coronavirus replication in a cell model.

Synthetically accessible de novo design using reaction vectors: Application to PARP1 inhibitors.

De novo design has been a hotly pursued topic for many years. Most recent developments have involved the use of deep learning methods for generative molecular design. Despite increasing levels of algorithmic sophistication, the design of molecules that are synthetically accessible remains a major challenge. Reaction-based de novo design takes a conceptually simpler approach and aims to address synthesisability directly by mimicking synthetic chemistry and driving structural transformations by known reactions that are applied in a stepwise manner. However, the use of a small number of hand-coded transformations restricts the chemical space that can be accessed and there are few examples in the literature where molecules and their synthetic routes have been designed and executed successfully. Here we describe the application of reaction-based de novo design to the design of synthetically accessible and biologically active compounds as proof-of-concept of our reaction vector-based software. Reaction vectors are derived automatically from known reactions and allow access to a wide region of synthetically accessible chemical space. The design was aimed at producing molecules that are active against PARP1 and which have improved brain penetration properties compared to existing PARP1 inhibitors. We synthesised a selection of the designed molecules according to the provided synthetic routes and tested them experimentally. The results demonstrate that reaction vectors can be applied to the design of novel molecules of biological relevance that are also synthetically accessible.

Oligomerization mediated by the D2 domain of DTX3L is critical for DTX3L-PARP9 reading function of mono-ADP-ribosylated androgen receptor.

Deltex proteins are a family of E3 ubiquitin ligases that encode C-terminal RING and DTC domains that mediate interactions with E2 ubiquitin-conjugating enzymes and recognize ubiquitination substrates. DTX3L is unique among the Deltex proteins based on its N-terminal domain architecture. The N-terminal D1 and D2 domains of DTX3L mediate homo-oligomerization, and the D3 domain interacts with PARP9, a protein that contains tandem macrodomains with ADP-ribose reader function. While DTX3L and PARP9 are known to heterodimerize, and assemble into a high molecular weight oligomeric complex, the nature of the oligomeric structure, including whether this contributes to the ADP-ribose reader function is unknown. Here, we report a crystal structure of the DTX3L N-terminal D2 domain and show that it forms a tetramer with, conveniently, D2 symmetry. We identified two interfaces in the structure: a major, conserved interface with a surface of 973 Å2 and a smaller one of 415 Å2. Using native mass spectrometry, we observed molecular species that correspond to monomers, dimers and tetramers of the D2 domain. Reconstitution of DTX3L knockout cells with a D1-D2 deletion mutant showed the domain is dispensable for DTX3L-PARP9 heterodimer formation, but necessary to assemble an oligomeric complex with efficient reader function for ADP-ribosylated androgen receptor. Our results suggest that homo-oligomerization of DTX3L is important for the DTX3L-PARP9 complex to read mono-ADP-ribosylation on a ligand-regulated transcription factor.