Multivalent binding of the tardigrade Dsup protein to chromatin promotes yeast survival and longevity upon exposure to oxidative damage

In reconstituted assays recombinant Dsup protects chromatinized plasmid DNA from DSBs caused by hydroxyl radicals²², so we asked if Dsup expression protected the in vivo yeast genome from oxidative DNA damage. 8-oxoguanine (8-OHdG) is generated when ROS react with DNA²⁷, so we quantified this base modification after transient exposure to H₂O₂ and observed a significant reduction in the presence of Dsup (Fig. 1c).

ROS and oxidative damage increase with age, and reducing oxidative damage extends the lifespan of multiple species (yeast, worms, fruit flies, mice²⁸), while elevated ROS production shortens lifespan²⁹. We thus asked if Dsup expression could extend yeast lifespan. In otherwise WT yeast Dsup had a negligible impact on chronological lifespan (the length of time a cell survives in a non-dividing state; Supplementary Fig. 1a), while replicative lifespan (the maximum number of times a cell can divide) was slightly reduced (Supplementary Fig. 1b). Cells lacking the superoxide dismutase (SOD) genes are deficient in their ability to process both endogenous and exogenous ROS. As a result, they accumulate oxidative stress and damage, such that yeast lacking Sod1 have a shortened replicative lifespan³⁰. When expressed in sod1Δ yeast, Dsup significantly increased replicative lifespan (Fig. 1d), suggesting enhanced survival and longevity in the face of chronic oxidative damage.

Free-radical scavengers are effective at protecting yeast from oxidative stress and extending lifespan³⁵, so we investigated if Dsup mediates such a role. Redox-sensitive GFPs (roGFPs) are excited at 405 nm in oxidizing but 488 nm in reducing conditions, so relative emissions after excitation at [405/488 nm] report on changes in redox state. In this manner, H₂O₂ treatment of yeast cells caused the redox state to increase in the cytoplasm and nucleus (as determined with compartment specific reporters: respectively roGFP2-Grx1³⁶ or roGFP2-Grx1-NLS) (Supplementary Fig. 2 and Fig. 3c). Of note, Dsup (WT), Dsup HMGN 3 R/3E and Dsup ΔHMGN ΔC + NLS each reduced the initial and post-H₂O₂ treatment redox state in the cytoplasm and nucleus (all relative to EV; Empty vector) (Fig. 3c). Importantly, these data show that while sequences within the Dsup N-terminal region (aa 1-359) reduce free-radical levels, this is insufficient to confer enhanced survival in response to H₂O₂ since Dsup ΔHMGN ΔC + NLS cells were not resistant to this genotoxin (Fig. 3a, b).

The presence of tardigrade Dsup in human and plant cells alters transcription factor binding and gene expression in response to DNA damage^5,10. This is suggestive that Dsup binds certain areas of the genome to influence transcription. Alternatively, to have the greatest physically protective effect from oxidative DNA damage, Dsup might be expected to uniformly coat chromatin. To investigate these possibilities, we used Cleavage Under Targets & Release Using Nuclease (CUT&RUN)³⁷ to map 6His-Dsup-FLAG localization (by anti-FLAG), and observed that Dsup (WT) enriched across the yeast genome, without bias or selectivity (i.e., not showing gaps, peaks or domains; Fig. 4b). Of note, the ability of CUT&RUN to map transcriptionally active promoters with anti-H3K4me3 was unaffected by Dsup (compare Empty vector (EV) and Dsup (WT)), indicating minimal impact on local chromatin structure (Fig. 4b). This was confirmed by micrococcal nuclease (MNase) digestion of chromatin from yeast −/+ Dsup, where we observed only minor differences in efficiency (Supplementary Fig. 3).

We next compared CUT&RUN across Dsup alleles, first noting that relative DNA yields post MNase digestion (prior to adapter ligation) were consistently Dsup (WT) » Dsup HMGN-3R/3E > Dsup ΔHMGN ΔC + NLS > EV) (Supplementary Fig. 4). This is mirrored in the CUT&RUN data, where Dsup HMGN-3R/3E showed less enrichment than Dsup (WT) across all genomic regions, while Dsup ΔHMGN ΔC + NLS resembled Empty vector (Fig. 4b; all data group scaled after normalizing to E. coli spike-in to allow comparisons of global changes in factor binding).

Taken together, these data indicate that Dsup binds without obvious bias across the genome in a manner dependent on its C-terminus (which includes the HMGN-like motif). Further, while mutation of conserved arginines in the Dsup HMGN-like motif (-3R/3E) reduced chromatin binding and/or increased its turnover (revealed by the long incubation steps of CUT&RUN but not direct chromatin fractionation (Fig. 4a)), the allele still conferred protection from oxidative DNA damage (Fig. 3a, b). These data indicate that both the HMGN-like motif and distal C-terminal sequences contribute to global chromatin binding.

Of interest, Dsup expression per se led to transcriptional changes for 220 genes relative to Empty vector control (Group 2: Fig. 5 and Supplementary Data File 3). This included a subset of genes involved in amino acid biosynthesis, though it did not extend to increased abundance of Gcn4 transcriptional activator (and so did not represent a 'General Amino Acid Response'; Supplementary Fig. 5). However, Group 2 also included genes proximal to telomeres, or encoded from Ty transposable elements. Both are usually actively repressed locations in budding yeast⁴¹, suggesting Dsup expression may lead to some opening of repressive chromatin; in line with the minor increase in MNase digestion efficiency also observed in these cells (Supplementary Fig. 3).

Together, this suggests that cells expressing Dsup are not transcriptionally poised to respond to oxidative stress, nor do they have an enhanced transcriptional response, including of DNA damage response genes, when it is encountered (Fig. 5). As such transcriptional profiling provides no explanation for the ability of Dsup to promote yeast cell survival in the face of oxidative damage.

Having ruled out a transcriptional role for Dsup in protecting yeast from oxidative damage (Fig. 5), we set out to further characterize its interaction with chromatin (Fig. 4). To interrogate potential mode(s) of engagement, we used the Captify in vitro chemiluminescent assay⁴². Here the biotinylated target (e.g., free DNA or fully defined mononucleosome: Supplementary Data File 1D) couples to streptavidin-donor beads while epitope-tagged Query (e.g., various forms of 6His-Dsup-FLAG; Fig. 2a, Supplementary Fig. 6 and Supplementary Data File 1E²²) couples to anti-tag acceptor beads. After mixing potential reactants the donor beads are excited at 680 nm, releasing a singlet oxygen that causes emission (520–620 nm) in proximal acceptor beads: this luminescent signal is directly proportional to the amount of [Donor - Acceptor] bridged by the [Target: Query] interaction (Supplementary Fig. 7). Binding is quantified by plotting Alpha Counts (fluorescence) as a function of protein concentration and expressed as relative EC50 (for all EC₅₀^rel from this study see Supplementary Data File 4).

We next used Captify to examine the contribution of various Dsup regions for nucleosome / DNA engagement. Here Dsup ∆HMGN ∆C showed no binding (Supplementary Fig. 8); not unexpected given our findings from cell fractionation and CUT&RUN (Fig. 4 and Supplementary Fig. 4). We thus created mutants to isolate the contributions of the HMGN-like motif (Dsup HMGN -3R/3E and Dsup HMGN-8A) or adjacent C-terminal region (Dsup ∆C) (Fig. 2a and Supplementary Data File 1E). Relative to Dsup (WT), each mutant had reduced binding to free DNA and nucleosomes that was undetectable in the presence of salDNA (compare Figs. 6b and 6c). Interestingly Dsup HMGN-3R/3E and -8A each preferred free DNA while Dsup ∆C preferred nucleosomes, indicating the mutated regions differentially contribute to chromatin engagement. Of note, while both HMGN mutants showed the same binding profile, charge reversing -3R/E bound nucleosomes ~two-fold weaker than charge neutralizing -8A (EC₅₀^rel 61.53 vs. 32.47 nM: Supplementary Data File 4), as might be expected given residue charge is central to how the HMGN-like motif engages the nucleosome acidic patch³².

Plasmid pRS306-PTDH3-Dsup was created by Gibson cloning of amplified DNA fragments following kit directions (NEB Gibson Assembly® Cloning Kit). In brief, pRS306⁶² was digested with SacI and BglII. The TDH3 promoter was PCR amplified (primers in Supplementary Data File 1A) from yeast genomic DNA with pTDH3_SacI_F (giving homology to Sac1 digested end of pRS306) and pTDH3_R (giving homology to 5' end of Dsup gene). The R. varieornatus Dsup gene (aa1-445; encoding protein accession P0DOW4, Supplementary Data File 1E) including N-terminal 6xHis and C-terminal FLAG epitope tags was amplified from plasmid pET21b-nHis6-Rvar-DSUP-cFLAG (kind gift from James Kadonaga²²), with primers Rvar_Dsup_F and Rvar_Dsup_R, respectively giving homology to the TDH3 promoter 3' end and ADH1 terminator 5' end. The ADH1 terminator was amplified from yeast genomic DNA using primers tADH1_F and tADH1_BglII_R, respectively giving homology to the Dsup gene 3' end and BglII digested pRS306 5' end.

pRS306-PTDH3-Dsup was digested with MfeI and integrated to BY4741⁶³ at the endogenous TDH3 promoter to make yeast strain RGY002 (pTDH3-6His_Dsup_FLAG: all yeast strains and their phenotypes in Supplementary Data File 1B). Mutations to integrated Dsup, including insertion of a stop codon to derive the 'Empty vector' strain (RAY149), were by CRISPR-Cas9 mediated genome editing⁶⁴. Primer sequences to generate guide RNAs⁶¹ and HDR template DNA are in Supplementary Data File 1A. Yeast were handled using standard methods. Dsup expressing strains were grown in SC-ura media (unless otherwise indicated).

P415TEF cyto roGFP2-Grx1-NLS was made from p415TEF cyto roGFP2-Grx1 (kind gift from Tobias Dick; Addgene plasmid # 65004)³⁶ by traditional cloning. First, the roGFP2-Grx1 sequence was PCR-amplified from plasmid p415TEF cyto roGFP2-Grx1 using primers that added a 2xNLS sequence (PKKKRKVPKKKRKV) to the Grx1 C-terminus. The resulting PCR product was digested with BamHI and HindIII and ligated to similarly digested plasmid p415TEF cyto roGFP2-Grx1 (Supplementary Data File 1B). All yeast strains and plasmids are available on request.

Exponentially growing yeast cells ( ~ 10⁷ at OD_λ600 0.8–1.0) were collected by centrifugation, washed once with water, and flash frozen in liquid nitrogen. Pellets were resuspended in 100 μL modified Laemmli buffer⁶⁵, boiled for five minutes (mins), and clarified by centrifugation. Proteins in the supernatant were resolved by 10% SDS-PAGE, membrane transferred, and immunoblotted with antibodies to FLAG (Sigma F1804; 1:1,000), GAPDH (Sigma A9521; 1:10,000), H2B (Abcam ab1790; 1:5000), H2A (Active Motif 39235; 1:2000), GCN4 (Absolute Antibody Ab00436-1.1; 1:1000), or H3 (Abcam ab1791; 1:1000). Uncropped unprocessed blots are provided in the Source Data.

30 mL yeast cultures were grown at 30 °C in shaking flasks to OD_λ600 0.6. Cells were harvested by centrifugation, and half of each culture resuspended in 15 mL of fresh SC-ura media −/+10 mM H₂O₂. After two hours at 30 °C cells were harvested by centrifugation, and genomic DNA isolated (Thermo Scientific Yeast DNA extraction kit) and resuspended in 50 μL nuclease-free water. DNA was quantified (NanoDrop spectrometer), samples diluted to 2 mg/mL, and sequentially incubated at 95 °C for five mins and on ice for 10 min. 50 μg DNA was then sequentially incubated with nuclease P1 (NEB; 1 U at 37 °C for two hours in provided buffer), alkaline phosphatase (NEB Quick CIP; 10 U at 37 °C for one hour in provided buffer supplemented with 100 mM Tris pH 8), at 95 °C for 10 min, and centrifuged at 6000 x g for five mins. Samples were then DNA quantified (NanoDrop spectrometer) and normalized.

ELISA to measure 8-hydroxy 2-deoxyguanosine (8-OhdG) was performed as per kit instructions (Abcam ab201734). For each sample 15 μg DNA was assayed in triplicate (three independent cultures for each condition) and absorbance at 450 nm measured by plate reader.

Yeast cultures were grown to saturation overnight in YPD at 30 °C and diluted to OD_λ600 0.1–0.2. The OD_λ600 of cultures from three independent colonies per genotype were measured every 30 min, and growth curves fitted with an exponential regression in Microsoft Excel. Doubling times were calculated as the slope of the curve during exponential phase and compared across independent cultures using a Student's t-test.

Indirect immunofluorescence of yeast cells was carried out as previously⁶⁶. 2.5 OD of early-mid log phase cells (OD_λ600 0.5–0.6) were crosslinked in 4% formaldehyde for 20 min at room temperature, then spheroplasted with 500 μg/mL Zymolyase 100 T for 30 min at 30 °C with rotation. Spheroplasted cells were applied to a 10-chamber poly-lysine coated microscope slide and permeabilized by incubation in methanol at −20 °C for six min, immediately followed by incubation in acetone at −20 °C for 30 seconds. After blocking in 5% BSA, slides were incubated with anti-FLAG (Sigma F1804; 1:1,000), anti-H2A (Abcam ab18255; 1:1.000) or anti-GAPDH (Sigma A9521; 1:5,000). After washing, slides were incubated with Alexa Fluor® 594 or 488 secondary antibodies (BioLegend), and coverslips mounted using ProLong™ Gold Antifade Mountant with DAPI (Invitrogen). Images were taken using an Olympus BX63 Fluorescence Microscope with a DP80 Camera and 60X objective.

To measure the response to acute hydrogen peroxide (H₂O₂) exposure, yeast cells were grown in YPD media to mid-log (OD_λ600 0.5−1.0), harvested by centrifugation, and resuspended to OD_λ600 0.6 in fresh media containing H₂O₂ (0, 4, 6, or 8 mM). After 90 min incubation (30 °C with shaking), cultures were washed and serial dilutions spread on SC-ura agar plates. After two days at 30 °C, colonies were counted and averaged across three technical replicates. Three independent experiments were performed from separate starting colonies, and statistical analyses performed using a Student's t-test.

To measure the response to chronic H₂O₂ exposure, yeast cells were grown in liquid culture until mid-log, harvested by centrifugation, and resuspended in sterile water to OD_λ600 1.0. Five-fold serial dilutions were made in a 96-well plate and spotted with a sterile 6x8-prong manifold onto YPD agar plates containing indicated concentrations of H₂O₂. Similar methods were used to evaluate sensitivity to methyl methanesulfonate (MMS; at indicated concentrations in YPD) and Zeocin (at indicated concentrations in YPD). To measure sensitivity to ultraviolet light, yeast serial dilutions on YPD plates were exposed to UV (doses (J/cm²) indicated in figures) using a crosslinker (Stratalinker). Plates were incubated for three days at 30 °C.

Yeast cells were grown overnight to early-mid log (OD_λ600 0.2−0.6) and diluted to OD 0.1 in freshly-filtered YPD. This inoculum was added to an automated dissection chip (iBiochips) as per manufacturer's instructions to achieve single cell loading. Light microscopy images of cells were acquired every 20 min over four days using an Evos FL Auto two-cell imaging microscope and associated software (ImageJ). At least 50 cells were counted per condition, with survival curves calculated in Graphpad Prism 9, and statistical analyses performed with a log-rank test.

Yeast chronological lifespan was measured as previously⁶⁷. Data is presented as average and standard deviation across three independent cultures (each an average of two technical replicates).

Yeast cells expressing cytoplasmic or nuclear localized (Supplementary Fig. 2) roGFP (redox sensing GFP) from respective plasmids p415TEF cyto roGFP2-Grx1 (GlutaRedoXin 1)³⁶ or roGFP2-Grx1-NLS were grown in SC-LEU media to mid-log (OD_λ600 0.6−0.8) and diluted to OD_λ600 0.6 in 5 mL flow cytometry tubes. Fluorescence at 405 nm and 488 nm was measured on a flow cytometer (BD Biosciences BD® LSR II) immediately before addition of H₂O₂ (to 4 mM), and at indicated time points. Data is presented as the mean and standard deviation of 405/488 nm values for each timepoint (independent triplicates) using FlowJo.

Exponentially growing yeast cells ( ~ 4 × 10⁸ at OD_λ600 0.8−1.0) were collected by centrifugation, washed once with ice cold 10% glycerol, and flash frozen. After thawing on ice, the cell pellet was washed (100 mM Tris pH 9.4, 10 mM DTT), resuspended in the same buffer, and rested on ice for 10 min. Cells were pelleted by centrifugation, washed in spheroplasting buffer (10 mM HEPES, 1.2 M Sorbitol, 0.5 mM PMSF), resuspended in spheroplasting buffer containing 56 μg/mL Zymolyase 100 T (US Biological), and incubated at 30 °C for one hour with rotation (until a 50 μL aliquot mixed with 1 mL 10% SDS had an OD_λ600 ~ 10% of the starting value). Spheroplasts were collected by centrifugation (1500 g for 2 min) and sequentially washed with spheroplasting buffer and wash buffer (1 M sorbitol, 20 mM Tris pH 7.5, 20 mM KCL, 2 mM EDTA, 0.5 mM PMSF, 0.1 μM spermine, 0.25 μM spermidine, 1:100 Calbiochem Protease Inhibitor Cocktail Set IV). Cells were gently resuspended and lysed in 250 μL Lysis Buffer (wash buffer with 400 mM sorbitol) for 10 min on ice.

Half of the volume after lysis (Input) was mixed with 5x Laemmli buffer, while the other half was pelleted at 14,000 x g for 15 min. The supernatant was collected (non-chromatin fraction) and mixed with 5x Laemmli buffer, and the pellet (chromatin fraction) resuspended in 1x Laemmli buffer. All fractions were boiled for 5 min, clarified by centrifugation, 7.5% of the total volume resolved by 12.5% SDS-PAGE, membrane transferred, and immunoblotted with anti-FLAG (Sigma F1804; 1:1,000) to detect Dsup. Successful fractionation was confirmed with anti-H2A (Abcam ab18255; 1:5,000) as a chromatin bound protein, and anti-GAPDH (Sigma A9521; 1:20,000) as a non-chromatin bound protein.

Nuclei from yeast cells expressing Dsup alleles (Supplementary Data File 1B) were purified as previously⁶⁸ with minor modifications. In brief, yeast cells were grown to mid-log (OD_λ600 0.6−0.8) in 500 mL SC-ura and collected by centrifugation. Cells were resuspended to 500 μL, spheroplasted with Zymolyase 100 T (2 mg/mL; 37 °C for 30 min), nuclei isolated as previously⁶⁸, and 1 mL aliquots (5 × 10⁷ nuclei) slow-frozen overnight in an isopropanol chamber at −80 °C.

For CUT&RUN⁶⁹, nuclei were rapidly thawed (2–3 min at 37 °C), and 100 μL of suspension used per reaction with the CUTANA™ ChIC/CUT&RUN Kit (EpiCypher). In brief, after immunotethering (of pAG-MNase to Rabbit IgG, anti-H3K4me3, or anti-FLAG: Supplementary Data File 1C), MNase digestion was performed (4 °C for 2 h) and DNA eluted in 12 μL final volume. 5 ng of DNA was used to prepare sequencing libraries with the Ultra II DNA Library Prep Kit (NEB #E7645L). Libraries were sequenced on an Illumina NextSeq 2000 platform, obtaining an average of ~1.1 million paired-end (PE) reads per reaction (Supplementary Data File 2).

PE fastq files were aligned to the sacCer3 reference genome (Bowtie2⁷⁰), filtered from duplicate (SAMtools⁷¹), multi-aligned (broadinstitute.github.io/picard), and exclusion list reads⁷², and the resulting unique reads for comparable reactions normalized by a spike-in scaling factor (1/ % E.coli reads) (BEDTools v2.30.0⁷³) to further RPKM (Reads Per Kilobase per Million mapped reads) -normalized bigwig files (DeepTools). Peak visualization from bigwig files was by Integrative Genomics Viewer (IGV). CUT&RUN sequence data is available from NCBI Gene Expression Omnibus (accession number GSE237436). The CUT&RUN analyses were performed independently three times with consistent results.

Overnight cultures of yeast strains −/+Dsup (RAY149 and RGY002 respectively: Supplementary Data File 1B) were grown at 30 °C in SC-URA medium. Cultures were diluted to OD_λ600 0.2 in 30 ml SC-URA media, grown to mid-log (OD_λ600 0.6–0.8), standardized to OD_λ600 0.5 and split to 12 × 2 mL aliquots. An untreated triplicate (T0) was collected by centrifugation, washed in water and flash frozen in liquid nitrogen. To the remaining aliquots in triplicate H₂O₂ (Fisher Scientific BP2633-500) was added (0, 4, or 8 mM) and cultures incubated at 30 °C for 30 min. Cells were collected by centrifugation, washed in water, and flash frozen.

RNA extraction was performed with the MasterPure Yeast RNA Purification Kit (Epicentre MPY03010) using the standard protocol including DNA removal step. RNA quality, concentration and purity was determined by Tapestation (Agilent 5067-5576) and nanodrop (Fisher Scientific), and material stored at −80 °C until library preparation. For each biological replicate, three technical replicates of non-strand specific libraries were prepared from 200 ng mRNA (Fast RNA-seq Lib Prep Kit v2; Abclonal RK20306) and sequenced to an average depth of 25 million paired-end 150 x 150 bp. Library preparation and sequencing services were provided by Novogene.

Sequencing read and base-calling quality were assessed by FastQC v0.11.9 (www.bioinformatics.babraham.ac.uk/projects/fastqc/↗), and paired-end fastq files aligned to the sacCer3 reference genome (Bowtie2⁷⁰). Files were triple filtered from duplicate (SAMtools⁷¹), multi-aligned (broadinstitute.github.io/picard), and exclusion list reads⁷² before further analysis. Gene FPKM (Fragment Per gene-length Kilobase per Million filtered reads) were computed with Homer v4.11⁷⁴. Technical and biological replicate FPKM reproducibility were assessed by linear regression R-squared values. To improve the statistical power of differential gene expression analyses, technical and biological replicates were combined by EdgeR v3.34.0⁷⁵. Genes were significantly differentially expressed if their false discovery rate was <0.05 and the fold-change was greater or less than two (log₂ <−1 or >1).

To generate the heat map in Fig. 5 the union of 868 significant genes from the ten pairwise log₂ ratios was clustered along each row (gene) by K-means and hierarchically by each column (pairwise ratios) using Cluster 3.0⁷⁶. The optimal number of clusters were heuristically determined by the lowest K-means value that resulted in non-redundant cluster patterns. To functionally annotate the clusters, gene set enrichment analysis (GSEA) was performed on each of the five clusters with DAVID Knowledgebase v2024q2^77,78 (all data in Supplementary Data File 3). RNA-sequence data is available from NCBI Gene Expression Omnibus (accession number GSE294109).

Yeast cultures were grown to mid-log phase and crosslinked using 1% (w/v) formaldehyde in either minimal media or water. Cell pellets equivalent to 50 ml of 0.85 OD culture were collected, resuspended in 1 ml lysis buffer (50 mM HEPES-KOH pH 8.0, 150 mM NaCl, 2 mM EDTA, 1% (v/v) Triton X-100, 0.1% w/v sodium deoxycholate), mixed with 1 ml zirconia beads and lysed using a Mini-Bead beater. The resulting extracts were spun in a bench-top centrifuge (16,000 g for 10 min at 4 °C) and supernatants discarded. Chromatin-containing pellets were resuspended, washed once with NP-S buffer (0.5 mM Spermidine, 0.075% (v/v) IGEPAL, 50 mM NaCl, 10 mM Tris-Cl pH 7.5, 5 mM MgCl₂, 1 mM CaCl₂) and resuspended in NP-S buffer + 1 mM β-mercaptoethanol (β-ME). 5 ml cell equivalent of each sample was removed as Input. To parallel 5 ml cell equivalents 20U of MNase (Worthington Biochemical corporation) was added, and reactions incubated at 37 °C for 15 or 20 min. Digestions were halted by EDTA (to 10 mM) and incubating on ice for 10 min, following by centrifugation (16,000 g for 10 min at 4 °C) to collect the supernatant containing digested DNA fragments. Crosslinks were removed from Input and MNase digested samples by resuspending in NP-S buffer + 1 mM β-ME, mixing with an equivalent volume of 2x proteinase K buffer (40 mM Tris-Cl pH 7.5, 40 mM EDTA, 2% (w/v) SDS), adding 3 µl of 20 mg/ml proteinase K, and incubating at 65 °C overnight. DNA was purified from each sample using phenol-chloroform-isoamyl alcohol extraction and isopropanol precipitation. Material (and DNA size standards) were resolved on a 1.3% agarose gel and stained with ethidium bromide before image capture (Supplementary Fig. 3).

All mononucleosomes (EpiCypher; Supplementary Data File 1D) were created from fully-defined (PTM or mutant) octamers wrapped by 5' biotinylated 147 × 601 DNA (Supplementary Data File 1E) unless otherwise stated, with PTMs confirmed by mass-spectrometry and immunoblotting (if an antibody was available)^42,79. Histone tail truncations were by direct expression of the indicated histone prior to octamer assembly (H3.1 NΔ2, H3.1 NΔ32, H3.3 NΔ32 or H4 NΔ15), or trypsin digestion of an assembled unmodified nucleosome (tail-less).

Dsup alleles for recombinant protein expression were cloned (GenScript) into pET-21a expression vectors with N-terminal 6x Histidine and C-terminal FLAG epitope tags (Supplementary Data File 1E). For expression, individual colonies from freshly transformed E.coli BL21(DE3) (T7 Express lysY/I^q for HMGN-8A) were grown to saturation overnight and used to inoculate 1 L LB supplemented with 100 µg/mL carbenicillin. Cultures were grown to OD_λ600 ~ 0.7 (200 rpm shaking at 37 °C), induced with 0.5 mM IPTG (200 rpm shaking at 16 °C overnight), and harvested by centrifugation (4000 g for 20 min at 4 °C). Pellets were resuspended in 12.5 mL wash buffer (50 mM Tris pH 7.5, 500 mM NaCl) per liter of culture and transferred to 50 mL conical tubes for centrifugation (4000 g for 15 min at 4 °C). Supernatants were decanted and cell pellets frozen at −80 °C.

Dsup (WT) and HMGN 3 R/3E were purified essentially as previously²²: In brief, cell pellets from a 2 L culture were thawed, resuspended in 25 mL lysis buffer (50 mM Tris pH 7.5, 500 mM NaCl, 1 mM PMSF, 0.2 mg/mL lysozyme, 0.2% Triton X-100, and Roche protease inhibitor tablet (one per 50 mL lysis buffer)), and incubated at 4 °C for 1 h with nutating. Lysates were sonicated (20 sec pulse on, 40 sec pulse off, 30% amplitude, 3 min total: Qsonica Q500) and clarified by centrifugation (31,000 g for 20 min at 4 °C). Lysate soluble fraction was mixed with 15 mL nickel resin (HisPur™ Ni-NTA: Thermo Scientific) equilibrated in cold Buffer A (50 mM Tris pH 7.5, 500 mM NaCl, 5 mM imidazole) and batch bound at 4 °C for 30−60 min with nutating. Subsequent steps of column purification were performed using a gravity column at room temperature with ice cold buffers. Unbound material was collected as flow-through, the column washed with 5 column volumes (CV) Buffer A, and eluted with 4 CV Buffer B (50 mM Tris pH 7.5, 500 mM NaCl, 500 mM imidazole), collecting one fraction per CV. Fractions E1 and E2 were combined, treated with 1 µl Benzonase (Millipore Sigma) on ice for ~ 15 min, mixed with 1 mL Pierce™ Anti-DYKDDDDK Affinity Resin (Thermo Scientific) equilibrated in Buffer C (50 mM Tris pH 7.5, 150 mM NaCl, 1 mM EDTA, 10% glycerol), and incubated at 4 °C for 1 hour with nutating. Subsequent steps of column purification were performed using a gravity column at room temperature with ice cold buffers. Unbound material was collected as flow-through, and the resin washed with 5 CV Buffer C and eluted with 10 CV Buffer D (50 mM Tris pH 7.5, 150 mM NaCl, 1 mM EDTA, 10% glycerol, 0.3 mg/mL 3X FLAG peptide). Aliquots were analyzed by SDS-PAGE / Coomassie staining, and fractions containing Dsup proteins combined in SnakeSkin dialysis tubing and dialyzed at 4 °C (three 2 L buffer exchanges (overnight, 2 hrs, 2 hrs) of 50 mM Tris pH 7.5, 150 mM NaCl, 20% glycerol, 1 mM DTT). Proteins were collected and additional buffer exchanges (to concentrate and remove the FLAG peptide) were by centrifugation (Vivaspin 20 (Cytiva) 10 K MWCO centrifugal filter). Aliquots were dispensed (0.1 mL), flash frozen in a dry ice / 100% ethanol bath, and stored at −80 °C.

Dsup ΔC and HMGN-8A were purified as above with minor changes: Cell pellets from a 2 L culture were lysed and clarified, and the soluble fraction decanted to a small beaker. 10% polyethyleneimine (branched, ~M.N. 60,000, 50 wt % aqueous solution pH 7.6; Thermo Scientific) was added dropwise to a final concentration of 0.1% while stirring at 4 °C. After continued stirring for ~ 45 min, lysate was clarified by centrifugation (31,000 g for 20 min at 4 °C) and soluble material decanted to a new tube for subsequent purification. Proteins were acid-eluted from the anti-DYKDDDDK Affinity Resin with 5 CV buffer E (0.1 M glycine pH 2.8) and immediately neutralized with 0.5 M Tris pH 8.5. Positive fractions were pooled for further purification.

The assay previously known as dCypher™ is now named Captify™, with no distinction in how the assay is performed or its capabilities. The interaction of WT or mutant 6His-Dsup-FLAG (kind gift from James Kadonaga²²: aka. the Queries [Supplementary Data File 1E]) with free DNA (147 × 601 Widom sequence) or fully defined nucleosomes (the Targets: Supplementary Data File 1D) was assayed by Captify on the Alpha (Amplified luminescence proximity homogeneous assay) platform as previously^42,79 with minor modifications.

Dsup queries (5 μL) were serially titrated in duplicate against a fixed concentration of target (5 μL, 10 nM biotinylated nucleosome or 2.5 nM free DNA (147 × 601)) in 384-well plates and incubated for 30 min. A 10 μL mix of AlphaScreen streptavidin Donor (Revvity, 6760002) and nickel-chelate Acceptor beads (Revvity, AL108M) was added and incubated for a further 60 min. All incubations were at room temperature in subdued lighting. Alpha counts were measured using a PerkinElmer 2104 EnVision plate reader (680 nm laser excitation, 570 nm emission filter ± 50 nm bandwidth).

[Query: Target] binding was examined over a range of assay conditions (20 mM Tris pH 7.5, 0.01% BSA, 0.01% NP-40, 1 mM DTT with additives as noted), including the impact of ionic strength (50–250 mM NaCl) and competitor salmon sperm DNA (salDNA; 0–20 μg/mL). Binding curves [Query: Target] were generated using a non-linear 4PL curve fit in Prism 9.0 (GraphPad) to yield EC₅₀^rel values^42,79 (Supplementary Data File 4). Where necessary, values beyond the Alpha hook point (indicating bead saturation / competition with unbound Query) were excluded and top signal constrained to average max signal for Target. In cases where signal never reached plateau, those were constrained to the average max signal within the assay (relative to unmodified nucleosome). In cases where a targets max signal never achieved half of max signal relative to unmodified nucleosome, an EC₅₀^rel was deemed not determinable (ND). In cases where a targets max signal never surpassed a two-fold increase, an EC₅₀^rel was deemed ND at less than lowest concentration tested (e.g., ND; <1 nM).

Each experiment in this study was repeated independently three times with similar results. No data were excluded from the analyses. No statistical method was used to predetermine sample size. The investigators were not blinded to allocation during experiments and outcome assessment.

Further information on research design is available in the linked to this article. Nature Portfolio Reporting Summary