Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
A Capped Tudor Domain within a Core Subunit of the Sin3L/Rpd3L Histone
Deacetylase Complex Binds Nucleic Acids
Ryan Dale Marcum, Joseph Hsieh, Maksim Giljen, Yongbo Zhang, and Ishwar Radhakrishnan*
Department of Molecular Biosciences, Northwestern University, Evanston, IL 60208-3500
*Address correspondence to: i-radhakrishnan@northwestern.edu;
Contact information for coauthors: ryanmarcum@u.northwestern.edu;
Joseph.Hsieh@cuanschutz.edu; maksimgiljen2020@u.northwestern.edu;
ybzhang@northwestern.edu
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 9, 2021. ; https://doi.org/10.1101/2021.08.09.455673doi: bioRxiv preprint
Abstract
Chromatin-modifying complexes containing histone deacetylase (HDAC) activities play critical
roles in the regulation of gene transcription in eukaryotes. These complexes are thought to lack
intrinsic DNA-binding activity, but according to a well-established paradigm, they are recruited
via protein-protein interactions by gene-specific transcription factors and post-translational
histone modifications to their sites of action on the genome. The mammalian Sin3L/Rpd3L
complex, comprising more than a dozen different polypeptides, is an ancient HDAC complex
found in diverse eukaryotes. The subunits of this complex harbor conserved domains and motifs
of unknown structure and function. Here we show that Sds3, a constitutively associated subunit
critical for the proper functioning of the complex, harbors a type of Tudor domain that we
designate the capped Tudor domain (CTD). Unlike canonical Tudor domains that bind modified
histones, the Sds3 CTD binds to nucleic acids that can form higher-order structures such as G-
quadruplexes, and shares similarities with the knotted Tudor domain of the Esa1 histone
acetyltransferase (HAT) that was previously shown to bind single-stranded RNA. Our findings
expand the range of macromolecules capable of recruiting the Sin3L/Rpd3L complex and draws
attention to potentially new roles for this HDAC complex in transcription biology.
Keywords: transcriptional corepressor; chromatin-modifying complex; protein-nucleic acid
interactions; solution NMR
Abbreviations: BRMS1: breast cancer metastasis suppressor 1; BRMS1L: BRMS1-like; CTD:
C-terminal domain/capped Tudor domain; HDAC: histone deacetylase; HAT: histone
acetyltransferase; lncRNA: long non-coding RNA; MBP: maltose-binding protein; NGS: next-
generation sequencing; RMS: root-mean-square; RMSD: RMS deviation; SELEX: systematic
evolution of ligands by exponential enrichment; EMSA: electrophoretic mobility shift assay
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 9, 2021. ; https://doi.org/10.1101/2021.08.09.455673doi: bioRxiv preprint
Introduction
Post-translational modifications of core histones constitute a common molecular mechanism for
regulating transcription by modulating DNA template accessibility to RNA polymerases,
regulatory factors, and other effectors (1,2). Among various post-translational modifications,
acetylation of lysine residues is not only abundant but also one that is characterized by high
turnover, consistent with its central role in the dynamic induction and repression of genes (3).
Deacetylation of histones in mammals is mediated in large part by histone deacetylases
(HDACs) 1, 2, and 3 (4-6). These enzymes, found in at least six giant multiprotein complexes
including the Sin3L/Rpd3L, Sin3S/Rpd3S, NurD, LSD1-CoREST, MiDAC, and SMART/NCoR
complexes (7-11), exert their effects following recruitment to specific sites on the genome by
DNA-bound transcription factors and/or specific histone modifications .
The Sin3L/Rpd3L complex is the prototypical HDAC complex found in organisms as diverse
as yeast and human (8,12). The complex plays fundamental roles in mammalian biology,
regulating a wide array of genes involved in the cell cycle, differentiation, metabolism, and stem
cell maintenance (13-15). The 1.2-2 MDa mammalian complex harbors at least 10 constitutively
associated subunits including Sin3A/B, HDAC1/2, RBBP4/7, Sds3/BRMS1/BRMS1L,
SAP30/SAP30L, ING1b/ING2, SAP130a/b, ARID4A/B, FAM60A, and SAP25 (paralogous
proteins in this list are separated by a '/'). The first five subunits on the list comprise the core
complex because of their essential roles in complex assembly and stability (16-19); these
subunits along with the ING subunits have orthologs in yeast. Whereas the RBBP, ING, and
ARID4 subunits harbor WD-40, PHD, and Royal Family domains that bind unmodified and
modified histones, the other subunits of the complex harbor conserved domains of unknown
structure and function.
In the course of our studies to define the molecular roles of the key subunits of the
Rpd3L/Sin3L complex, we previously described a novel zinc finger motif shared by the SAP30
and SAP30L subunits of this complex that we later showed turbocharges HDAC activity in
response to small-molecule effectors such as inositol phosphates derived from membrane lipids
(20,21). We also showed how the Sds3 subunit provides a dimerization function for the complex
that involves a region that assembles into a two-stranded antiparallel coiled-coil helix (22). We
further showed that the subunit plays a critical role in core complex assembly by engaging
directly and independently with Sin3 and HDACs; the subunit and its paralogs have been
implicated in interactions with other subunits of the complex as well as with sequence-specific
DNA transcription factors (22-24). At the cellular level, the subunit plays a critical role in the
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 9, 2021. ; https://doi.org/10.1101/2021.08.09.455673doi: bioRxiv preprint
proper segregation of chromosomes during cell division by targeting the Sin3L/Rpd3L complex
to pericentric heterochromatin (16). Recently, the subunit has also been implicated in resolving
co-transcriptionally generated R-loops (featuring RNA-DNA hybrids and single-stranded DNA)
that are a major source of genomic instability (25).
Two paralogs of Sds3 have been described including the breast cancer metastasis
suppressor 1 (BRMS1) and a BRMS1-like protein called BRMS1L that share many domains
found in Sds3 (24,26). All three proteins are found in Sin3L/Rpd3L complexes and share certain
key structural and functional features but are not functionally redundant. For example, disruption
and downregulation of BRMS1 and BRMS1L is associated with metastasis of multiple types of
cancers, whereas overexpression suppresses this effect through a mechanism involving
repression of several metastasis-associated protein-coding and microRNA genes (27-31).
However, Sds3 overexpression fails to compensate for BRMS1 deletion or epigenetic silencing
in breast cancer and does not suppress metastasis (32); the molecular basis for this observation
remains obscure (33,34), warranting deeper structural and functional studies to understand the
molecular roles of Sds3 and its paralogs.
Here, we describe the structure of another conserved domain of an unknown function in the
Sds3 subunit that is shared with one of its paralogs, BRMS1L, but not with BRMS1. Our
structural and biochemical analyses suggest that the domain broadly shares an SH3-like β-
barrel fold found within many chromatin-binding transcription factors but instead of binding
chromatin, the domain binds nucleic acids including both RNA and DNA.
Results
A Conserved C-terminal Domain (CTD) of Unknown Structure & Function in Sds3 and BRMS1L
Sequence analysis of Sds3 and BRMS1L orthologs from human to zebrafish revealed an ~80
residue region at the C-termini of the respective proteins with a pattern of conservation that
suggested a well-conserved, independently folded domain (Figure 1a); the next well-conserved
segment corresponding to the previously characterized Sin3-interaction domain (SID) resides
~30 residues N-terminal of this domain (22). Searches conducted using the mouse Sds3 protein
of the RCSB PDB database using BLAST as well as the repository of templates in the SWISS-
MODEL homology modeling server failed to identify any bona fide structural homologs for this
domain (i.e., no suitable templates in the so-called safe zone for homology modeling).
Sds3 CTD Adopts a Unique Variation of a Common Fold
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 9, 2021. ; https://doi.org/10.1101/2021.08.09.455673doi: bioRxiv preprint
To gain insights into the structure and function of the CTD, we expressed and purified a
recombinant protein corresponding to residues 250-326 of mouse Sds3. A 1H-15N correlated
NMR spectrum of this protein was characterized by narrow and well-dispersed resonances
indicative of a folded domain (Supplementary Figure S1). The solution NMR structure of the
domain was determined using a combination of 1H-1H NOE-based distance and backbone
chemical shift-based torsion angle restraints (Table 1). Structure determination and refinement
resulted in an ensemble of 20 converged conformers with average RMSDs in ordered regions of
0.52 Å and good agreement with experimental restraints and excellent backbone and covalent
geometry (Figure 1b; Table 1). The domain comprises eight strands and a short helix (Figure
1c). Except at the N- and C-termini and the loop connecting β4 and β5, the conformers adopt
highly similar backbone conformations. Strands β4 to β8 form an antiparallel five-stranded
closed β-barrel fold, reminiscent of SH3-domains, with one mouth of the barrel capped by the
three-stranded β-sheet formed by β1, β2, and β3.
To test the reliability of predictions by de novo methods, the mouse Sds3 sequence was
submitted to the Robetta server for tertiary structure prediction using TrRefineRosetta (35). The
top five solutions returned by this method all had the same overall fold (Figure 1d), and
remarkably, the backbone RMSD for the 65 Cα atom pairs involved in the best-fit superposition
between the top solution and the representative structure from the NMR ensemble was 0.84 Å
(Figure 1e). Thus, although homology modeling (aka comparative modeling) methods failed to
detect a suitable template for modeling, Rosetta could readily and reliably predict the structure
of this domain.
To gain insights into the domain's function, we sought to establish the closest homolog at
the structural level by searching the RCSB PDB database using DALI (36). Although DALI
returned many hits, the one in the PDB25 database with the highest Z-score (6.9) corresponded
to the bromo adjacent homology (BAH) domain of the Zea mays protein ZMET2, a DNA
cytosine-5-methyltransferase ((37); Figure 2a). A best-fit superposition of all the regular
secondary structural elements shared by the two proteins, except for the slightly elongated
helical segment in Sds3, yielded a backbone RMSD of 1.87 Å. The ZMET2 BAH domain binds
to histone H3K9me2, in a manner reminiscent of methyllysine/methylarginine-binding by the so-
called Royal domains including the Tudor, MBT, chromobarrel, and PWWP domains ((38);
Supplementary Figure S2). Comparisons with a representative structure for each of these four
members of the Royal family suggested that the Sds3 CTD might be evolutionarily closest to the
Tudor domain with a best-fit superposition of backbone atoms of 1.39 Å (Figure 2b). The Sds3
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 9, 2021. ; https://doi.org/10.1101/2021.08.09.455673doi: bioRxiv preprint
CTD harbors two features that distinguish it from regular Tudor domains: a longer helical
segment linking the two C-terminal strands of the β-barrel and a three-stranded β-sheet at the
N-terminus that closes one edge of the barrel, thereby capping it. Therefore, we refer to the
Sds3 domain as a 'capped' Tudor domain (CTD).
Sds3 CTD is not a Histone-binding Module
Unlike the aforementioned BAH and Royal family domains, the Sds3 capped Tudor domain
lacks an aromatic cage near the remaining edge of the β-barrel for binding methyllysine
residues (Figures 2a-2c and Supplementary Figure S2). Indeed, the domain in this region is
devoid of aromatic residues barring one (Phe320; Figure 2c). To test whether the domain could
bind to post-translationally modified histones, we screened the MODified™ histone peptide
array using purified protein. Although an anti-Myc antibody bound to the Myc peptide that was
included in the array as a positive control, no binding was detected for Sds3 to any of the
histone peptides, both modified and unmodified, in the array (Figure 2d). To complement these
findings, NMR titrations of 15N-labeled Sds3 CTD were conducted with dimethyllysine,
trimethyllysine, acetyllysine, and dimethylarginine as well as with an unmodified histone H3
peptide (residues 1-42). However, none of these compounds produced any discernible
perturbations in the NMR spectra. Collectively, these results suggest that the Sds3 CTD has no
apparent histone-binding activity.
Sds3 CTD Surface Properties Suggest a Role in Nucleic Acid Binding
To gain clues into its molecular function, we then analyzed the surface properties of the Sds3
capped Tudor domain. Given the considerably high levels of sequence conservation (Figure
1a), mapping the information onto the molecular surface was not especially insightful. However,
since Tudor domains have been implicated in functions other than histone binding, such as
nucleic acid binding, we calculated the electrostatic potential using APBS (39) and mapped it
onto the molecular surface of the domain. Doing so revealed an overwhelmingly electropositive
or neutral surface with multiple, discrete patches that were strongly electropositive (Figure 3a).
Since the Sin3L/Rpd3L complex functions, in part, by directly engaging with nucleosomes (40),
we first asked whether the domain could bind to the well-characterized acidic patch on the
surface of nucleosomes. Instead of using mononucleosomes, we used the histone H2A-H2B
heterodimer (41), which is a well-established surrogate for the acidic patch in NMR titration
experiments with Sds3 CTD. Once again, no discernible perturbations could be detected in the
NMR spectra, ruling out a potential role for the capped Tudor domain in nucleosome binding.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 9, 2021. ; https://doi.org/10.1101/2021.08.09.455673doi: bioRxiv preprint
Since the knotted Tudor domain of Esa1 was previously shown to bind RNA (42), we asked
whether the Sds3 CTD could have a similar function. To deduce potential RNA-binding motifs,
SELEX experiments were performed starting with a 20mer randomized library (43). Samples of
the RNA library following three rounds of selection and amplification were incubated with
increasing amounts of MBP-tagged Sds3 CTD in electrophoretic mobility shift assays (EMSAs;
Figure 3b). Although no clear mobility shifts were observed in these experiments, samples of
the library following six rounds of selection, amplification, and incubation with MBP-Sds3 CTD
yielded a clear band whose mobility was significantly reduced compared to that of the free RNA.
These bands were observed at micromolar concentrations of MBP-Sds3 CTD, implying a
modest affinity interaction. Following reverse transcription and amplification, the RNA library
from this round was sent for next-generation sequencing (NGS). A total of 42 x 106 reads were
obtained, out of which ~10 x 106 were deemed to be of high quality for motif detection using the
MEME suite. Because MEME can only handle a maximum of 500,000 sequences, the reads
were randomly assigned to 10 datasets, each comprising 500,000 sequences. The five most
statistically significant motifs in each dataset reported by MEME were compiled
(Supplementary Table S1) and those that were found in more than three datasets are listed in
Table 2. Somewhat unexpectedly and despite six rounds of enrichment, a single dominant motif
did not emerge from these analyses. The most prevalent was a 7-residue motif (HGTGGTK;
where H is A/C/T and K is G/T) found on average in 4.2% of the sequences. Remarkably, the
other motifs deduced from these analyses were significantly enriched in Ts and Gs.
Sds3 CTD Binds G-quadruplexes
Since Ts and Gs are commonly found in G-quadruplexes, we asked whether a T2G4 DNA
quadruplex might interact with Sds3 CTD (note that we chose to perform these experiments with
DNA rather than RNA because both molecules can form similar quadruplex structures). We first
recorded 1H NMR spectra to confirm G-quadruplex formation for this sequence, which as
expected, is characterized by the presence of four narrow, imino proton resonances in the 10-
11.5 ppm region emanating from each of the four tetrads ((44); Supplementary Figure S3).
Four additional resonances of reduced intensity are observed in the imino proton region
suggesting the formation of two types of quadruplexes that most likely differ in strand direction.
The addition of one equivalent of T2G4 to 15N-Sds3 CTD induced significant perturbations in the
NMR spectrum (Figure 4a) with a few resonances shifting to new positions and several others
undergoing various degrees of broadening. Interestingly, the minor quadruplex species is
relatively unperturbed in the presence of Sds3 CTD, implying that the interaction with the major
species is specific (Supplementary Figure S3). Similarly, titrations with a random 10mer self-
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 9, 2021. ; https://doi.org/10.1101/2021.08.09.455673doi: bioRxiv preprint
complementary DNA duplex (5'-GCGAATTCGC-3') elicited only modest perturbations in the
spectrum characterized by chemical shift deviations of 0.014 ± 0.012 ppm (Supplementary
Figure S4). Additionally, virtually no perturbations were noted in the titration with a random
RNA 8mer sequence (5'-AACUGUCG-3'). Collectively, these results indicate that Sds3 CTD
preferentially associates with certain G-quadruplexes over double-stranded DNA or single-
stranded RNA sequences.
To identify the region of Sds3 CTD involved in binding to the G-quadruplex, we quantified
the peak intensity ratios in the holo and apo HSQC spectra and mapped them on to the
molecular surface of the CTD. The strongest perturbations were observed for a contiguous
surface formed largely by residues in strands β6, β7, the loop preceding β6, and the sole helix
connecting β7 and β8 (Figure 4b). This surface is distinct from the one located at the edge of
the barrel that is commonly used by the BAH domain and the Royal family domains to engage
with chromatin targets (Figure 2a, 2b, and Supplementary Figure S2). Interestingly, the three-
stranded β-sheet formed by the capping motif of the CTD does not show significant spectral
perturbations, implying that this novel feature is not essential for binding nucleic acids, at least
those that were tested in this study.
Discussion
HDAC containing chromatin-modifying complexes frequently contain many protein subunits with
the non-enzymatic subunits widely thought to impart genome targeting specificity, especially
since HDACs exhibit little sequence specificity themselves for acetylated targets. Two common
mechanisms of genome targeting involve protein-protein interactions with sequence-specific
DNA binding factors and/or engagement with specific post-translational modifications on
histones. An especially intriguing feature of the Sin3L/Rpd3L complex is that the core subunits,
including Sin3, Sds3, and SAP30, harbor domains of unknown structure and function that are
narrowly distributed and found only in the respective orthologs and paralogs. The Sin3 subunit
performs a scaffolding function for the assembly of the complex by engaging directly with most
of the subunits while also providing multiple surfaces for direct engagement with DNA-bound
factors. The SAP30 subunit is involved in turbocharging the catalytic activity of HDAC1 whereas
Sds3 is thought to impart stability to the complex while also providing a dimerization function
and interaction sites for DNA-bound factors and other subunits of the complex.
The discovery of a type of Tudor domain in Sds3 was unexpected and initially suggested an
unrecognized function for the subunit in chromatin binding. However, as our subsequent studies
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 9, 2021. ; https://doi.org/10.1101/2021.08.09.455673doi: bioRxiv preprint
have shown, Sds3 CTD shares more in common with another type of Tudor domain found in the
Esa1 histone acetyltransferase previously shown to bind single-stranded RNA than with
canonical Tudor domains (42). Although these non-canonical Tudor domains share a backbone
RMSD of only 1.98 Å, with the highest level of structural similarity in the region spanning β4-β7
of Sds3 CTD, both domains bind to nucleic acids on the body of the barrel (Figure 5). The
involvement of overlapping surfaces of the β-barrel in these distantly related domains is
particularly striking. Even more interesting is the presence of conserved tryptophan and tyrosine
residues at the protein-nucleic acid interface of both proteins (Figure 5), although the exact
locations of these residues are not conserved between these domains. The latter likely reflects
the different specificities of the domains for their target(s). Both the involvement of aromatic
residues and their location on the surface of β-sheets are defining features shared with RNA-
recognition motifs (45). Thus, both knotted and capped Tudor domains appear to have
independently acquired nucleic acid-binding functions through a process of convergent
evolution. Finally, since Esa1 is a histone acetyltransferase and a member of the NuA4 HAT
complex (46), it is intriguing that these non-canonical Tudor domains are found in complexes
with opposing enzymatic activities that produce contrasting transcriptional outcomes. Even more
striking, the essential requirement for Esa1 in yeast can be bypassed through deletion of the
gene encoding Sds3 (47).
There is precedent for RNA mediating the recruitment of at least one other chromatin-
modifying complex, the LSD1-CoREST complex (48). The complex harbors both a histone
demethylase (LSD1) as well as HDACs 1 and 2 and is recruited to the telomeric regions by a
long non-coding RNA (lncRNA) that associates with chromatin in these regions to facilitate
telomere silencing and heterochromatin formation (49). The lncRNA harbors many repeats of
the 5'-UUAGGG-3' sequence that forms intramolecular G-quadruplexes that in turn is critical for
efficient interactions with LSD1 (50,51). Since G-quadruplexes have been found to localize to
heterochromatin and gene knockout studies implicate Sds3 in the proper establishment of
pericentric heterochromatin (16,52), it is tempting to speculate that the Sin3L/Rpd3L complex
may be recruited to these regions through an analogous mechanism involving potentially G-
quadruplex or other higher-order nucleic acid structures to promote gene silencing and
heterochromatin formation at centromeres. A similar molecular mechanism may be operative in
the context of R-loops where multiple subunits of the Sin3L/Rpd3L complex have been
implicated in resolving these structures generated during transcription (25).
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 9, 2021. ; https://doi.org/10.1101/2021.08.09.455673doi: bioRxiv preprint
Although both Sds3 and BRMS1L share a CTD, the orthologous proteins define separate
clades consistent with their distinct patterns of sequence conservation (Figure 1a), implying that
the domains, while sharing a similar function, likely encode different specificities for their targets.
Interestingly, the N-terminal three-stranded β-sheet that forms the capping motif that is unique
to these Tudor domains does not seem to be involved in nucleic acid binding for the sequences
that were tested. However, this structural motif also harbors conserved, solvent-exposed
tyrosine and tryptophan residues (Y263 and W268) that could potentially be involved in binding
other nucleic acid targets.
In conclusion, we have described a new type of Tudor domain that appears to have evolved
to perform non-canonical functions such as binding nucleic acids. Although our results suggest
that the Sds3 CTD has a preference for certain G-quadruplexes, more detailed studies focused
on this domain are needed to definitively assess this preference over other higher-order nucleic
acid structures as well as to identify the actual biologically relevant nucleic acid partner(s)
targeted by the domain. If confirmed, it would only be the second instance of an HDAC-
containing chromatin-modifying complex implicated in direct recruitment by higher-order nucleic
acid structures, expanding the repertoire of macromolecules that could function in this manner.
Our findings thus draw attention to potentially new and underappreciated roles for both Sds3
and the Sin3L/Rpd3L HDAC complex in transcription biology.
Acknowledgements: This work was supported by grants from the American Heart Association
to I.R. (17GRNT33680167) and R.D.M. (16PRE27260041) and from the National Institutes of
Health for upgrading the 600 MHz NMR console (S10 OD012016). We thank members of the
Radhakrishnan lab for critical comments. We are grateful to the Robert H. Lurie Comprehensive
Cancer Center at Northwestern for supporting structural biology research. We thank Yawen Bai
at the NCI for providing the plasmid encoding histone H2A-H2B heterodimer.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 9, 2021. ; https://doi.org/10.1101/2021.08.09.455673doi: bioRxiv preprint
Experimental Methods
Construct Generation, Protein Expression and Purification
The coding sequence of mouse Sds3 CTD (residues 250-326) was sub-cloned into the
pMCSG7, pMCSG9, and pMCSG10 bacterial expression vectors (53). His6-tagged-CTD
encoded by the pMCSG7 vector was expressed at 16 °C in BL21(DE3) cells and subsequently
purified via Ni2+-affinity chromatography. Cell pellets were resuspended in lysis buffer (50 mM
Tris, pH 8.0, 200 mM NaCl, 1 mM TCEP, 1 mM phenylmethylsulfonyl fluoride (PMSF), 1 µM
leupeptin, 1 mM pepstatin and 0.1% Triton X-100) and lysed by sonication. After centrifugation,
the lysate was loaded onto a Ni2+-affinity resin (Sigma), washed with high salt (800 mM NaCl)
and eluted with 300 mM imidazole. The His6-tag was removed by incubating the protein with
TEV protease overnight at 4 °C, the samples concentrated and further purified via size
exclusion chromatography using a Superdex 75 GL column (GE Healthcare) and a running
buffer comprising 20 mM Tris (pH 8.0), 200 mM NaCl, and 1 mM DTT. Uniformly 15N- and/or
13C-labeled proteins were produced following the same procedure except they were grown in
M9 minimal media supplemented with 15N-ammonium sulfate and/or 13C-glucose.
MBP- and GST-tagged proteins encoded by the pMCSG9 and pMCSG10 vectors were
expressed and purified in a similar manner as the His6-tagged Sds3 CTD with the following
changes. GST-Sds3 CTD was purified with glutathione sepharose (GE Healthcare) and eluted
using 25 mM glutathione, while MBP-Sds3 was purified with amylose resin (New England
Biolabs) and eluted with 25 mM maltose. All proteins were stored at 4 °C until they were used.
NMR Spectroscopy and Structure Determination
All NMR spectra were acquired at 25 °C on a 600 MHz Agilent DD2 spectrometer. Sds3 CTD
samples in the range of 350 μM in 20 mM sodium phosphate buffer (pH 6.0) containing 50 mM
NaCl, 1 mM DTT and 10% D2O were used to acquire NMR data. 3D HNCACB, CBCA(CO)NH,
C(CO)NH-TOCSY, HNCO, and 15N-NOESY-HSQC NMR spectra were acquired for sequence-
specific backbone resonance assignments (54). Data processing was performed using Felix
(Felix NMR) and the peaks in these spectra were picked in NMRFAM-Sparky (55) and
submitted to I-PINE for peak assignment (56). All the assignments were checked manually for
accuracy. Side chain assignments were performed manually using 3D HCCH-COSY and
HCCH-TOCSY spectra acquired in D2O; the sample for these experiments was generated by
exchanging the buffer from H2O to D2O. Aromatic resonances were assigned based on a careful
analysis of 2D 1H-13C aromatic HSQC, (HB)CB(CGCD)HD, (HB)CB(CGCDCE)HE (57) and 1H-
1H NOESY spectra recorded in D2O.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 9, 2021. ; https://doi.org/10.1101/2021.08.09.455673doi: bioRxiv preprint
Backbone φ and ψ dihedral angle restraints for structure calculations were derived from a
combined analysis of the 1H, 13C, 13C, 13C', and backbone 15N chemical shifts using TALOS+
(58); only residues with reliability scores of 10 in secondary structural elements were restrained.
1H-1H NOE-based distance restraints were derived from three spectra, including 3D 15N-edited
NOESY (τm = 80 ms) recorded in H2O, 3D 13C-edited aliphatic NOESY (τm = 60 ms), and 2D 1H-
1H NOESY (τm = 75 ms) recorded in D2O.
Structures were determined using ARIA 1.2 in conjunction with CNS 1.1 starting from an
initial structure with extended backbone conformation (59-61). All NOEs were calibrated
automatically and were assigned iteratively by ARIA; the assignments were checked manually
for errors after each run. Eighty conformers were calculated; 40 conformers with the lowest
restraint energies were refined in a shell of water and the 20 conformers with the lowest
restraint energies and violations and ideal covalent geometry were selected. The final
conformers were analyzed using CNS (59), PROCHECK (62), and scripts written in-house.
Histone Peptide Array
His6-tagged Sds3 CTD was incubated with a MODifiedTM histone peptide array (Active Motif) at
a concentration of 15 M. After a 2 h incubation period, the array was washed and probed with
an anti-His primary antibody (Thermo Fisher, MA121315, 1:1000 dilution), after which anti-
mouse HRP-conjugated secondary antibody (Thermo Fisher Scientific, #OB617005, 1:1000
dilution) was used. The array was imaged using West Pico chemiluminescent substrate
(Thermo Scientific, #34080) and a Syngene Pxi chemiluminescent imager. The screen was
performed in duplicate as a test of reproducibility.
NMR Titrations
Sds3 CTD samples in the 150 to 230 μM range in 20 mM sodium phosphate buffer (pH 6.0)
containing 50 mM NaCl, 1 mM DTT, and 10% D2O were used for the NMR titration experiments.
1H-15N HSQC NMR spectra were acquired following the addition of excess dimethyllysine,
trimethyllysine, acetyllysine, and dimethylarginine (Sigma; all compounds were used without
further purification). NMR titrations were also conducted with an unmodified histone H3 peptide
(residues 1-42), purified histone H2A-H2B heterodimer, DNA and RNA oligonucleotides. The
oligonucleotides were purchased from Integrated DNA Technologies and Dharmacon and used
without further purification. Data processing and analysis were performed using Felix (Felix
NMR) and NMRFAM-Sparky (55).
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 9, 2021. ; https://doi.org/10.1101/2021.08.09.455673doi: bioRxiv preprint
SELEX
SELEX (systematic evolution of ligands by exponential enrichment; (43) experiments were
conducted with GST-tagged Sds3 CTD. An RNA library for selection experiments was obtained
from TriLink Biotechnologies that consisted of random 20mer sequences flanked by adapters of
known sequence for reverse transcription, PCR amplification, and sequencing. To preclear the
RNA library of non-specific interactions with GST, the RNA library was initially incubated with
purified GST immobilized on glutathione sepharose beads. The flow-through containing
unbound RNA was then incubated with purified GST-Sds3 CTD immobilized on glutathione
sepharose beads. The beads were washed extensively, and the protein was digested with
proteinase K. Bound RNA was then purified using phenol/chloroform extraction. RNA was
reverse transcribed and then PCR amplified. The PCR template was used to transcribe RNA
and the process was repeated six times to enrich Sds3 CTD binding sequences.
After the final round of selection, PCR products were submitted to the Northwestern NUSeq
core facility. Sequencing reads were generated using an Illumina SR75 sequencer. Sequences
were trimmed to remove adapters using Cutadapt (63) and filtered by quality (Fast QC) in the
Galaxy bioinformatics suite (64). Ten datasets of 500,000 randomly selected sequences from
~106 high quality reads were extracted using the Galaxy bioinformatics suite for motif analysis.
Motifs were identified and analyzed using the MEME suite (65). The top five motifs from each
dataset were compiled and those that appeared in more than three datasets were deemed
significant for inclusion in Table 2.
EMSAs
Samples were prepared by mixing increasing concentration of MBP-tagged Sds3 CTD with 1
μM total RNA from rounds 3 and 6 (the final round) of SELEX. EMSAs were performed using
5% native PAGE gels with 0.5x TB buffer (50 mM Tris, 50 mM boric acid). Gels were
equilibrated for 30 minutes before samples were loaded onto the gel. Samples were run at 4 °C
for 90 min and then stained 30 min with SYBR™ Gold (ThermoFisher). Gels were imaged on a
fluorescence Typhoon imager with the excitation and emission set to 480 nm and 520 nm,
respectively.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 9, 2021. ; https://doi.org/10.1101/2021.08.09.455673doi: bioRxiv preprint
Table 1. NMR Structure Determination Statistics
Restraint Statistics
NOE-based distance restraints
1456
Unambiguous NOE-based restraints
1295
Intra-residue
778
Sequential (| i – j | = 1)
232
Medium-range (1 < | i – j | 4)
59
Long-range (| i – j | > 4)
226
Ambiguous NOE-based restraints
161
Hydrogen bonding distance restraints
32
Torsion angle restraints
(63 φ, 63 ψ)
Structure Quality of NMR Ensemble
Restraint satisfaction
RMS differences for distances (Å)
0.0079 ± 0.0009
RMS differences for torsion angles (°)
0.3133 ± 0.0790
Deviations from ideal covalent geometry
Bond lengths (Å)
0.0035 ± 0.0001
Bond angles (°)
0.4361 ± 0.0146
Impropers (°)
1.3050 ± 0.1113
Ramachandran plot statistics (%)
Residues in most favored regions 84.2
Residues in additional allowed regions 15.1
Residues in generously allowed regions 0.5
Residues in disallowed regions 0.3
Average Atomic RMSDs from Average Structure (Å)
All atoms
1.76
All atoms except in disordered regions
a
1.40
Backbone atoms (N, C
α
, C')
All residues
1.04
All residues except disordered regions
a
0.52
All residues in secondary structural elements 0.44
a: disordered regions include two non-native residues at the N-terminus in addition to residues
250-251, 279-283 and residue 326 at the C-terminus of the domain
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 9, 2021. ; https://doi.org/10.1101/2021.08.09.455673doi: bioRxiv preprint
Table 2. Statistically Significant Sequence Motifs Identified by MEME from NGS Data
Motif Dataset Number
of Sites Probability
E-value
HGTGGTT 2 33555 0.067110 6.10 x 10-16
CGTGGTT 4 13453 0.026906 8.10 x 10
-
11
MGTGGTT 5 26004 0.052008 1.60 x 10
-
14
CGTGGTT 8 14162 0.028324 5.50 x 10
-
06
CGTGGTK 9 19955 0.039910 1.30 x 10
-
08
MGTGGTT 10 21703 0.043406 1.60 x 10
-
12
CCGTTTGTGGTGCGTTTTT 2 2766 0.005532 1.50 x 10-27
4 2851 0.005702 2.50 x 10
-
35
6 2828 0.005656 8.20 x 10
-
36
10 2916 0.005832 5.70 x 10
-
31
CCGTTTGTGGTGCGTTTTTG 1 2703 0.005406 1.10 x 10-29
3 2761 0.005522 3.40 x 10
-
23
5 3339 0.006678 7.20 x 10
-
31
7 2347 0.004694 9.20 x 10
-
24
8 3217 0.006434 1.80 x 10
-
25
9 2797 0.005594 1.60 x 10
-
36
GGCGTTGTCCGTGGTTTGTG 1 1753 0.003506 2.30 x 10-10
3 2616 0.005232 1.60 x 10
-
24
5 1851 0.003702 1.80 x 10
-
16
6 1896 0.003792 1.70 x 10
-
25
8 3858 0.007716 1.40 x 10
-
19
9 2551 0.005102 1.20 x 10
-
17
10 1566 0.003132 3.80 x 10
-
23
GTCTGTGGTTGGTCTTGGCT 3 295 0.000590 5.30 x 10-10
6 319 0.000638 4.10 x 10
-
05
8 478 0.000956 5.20 x 10
-
07
IUPAC codes: H: A/C/T; M: A/C; K: G/T
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 9, 2021. ; https://doi.org/10.1101/2021.08.09.455673doi: bioRxiv preprint
Figure Legends
Figure 1. Solution structure of Sds3 CTD and comparison with prediction. (a) A CLUSTAL Ω-
guided multiple sequence alignment of a ~80-residue region at the C-terminus of Sds3 and
BRMS1L orthologs from various eukaryotes. Two views differing by a 180° rotation around the
vertical axis of (b) a best-fit backbone superposition of the ensemble of 20 conformers and (c)
the corresponding representative structure. (d) A best-fit backbone superposition of the top five
solutions returned by Rosetta. (e) A best-fit backbone superposition of the representative NMR
structure with the highest-ranked Rosetta prediction. The cartoon on top of panel a identifies the
locations of various secondary structural elements in the solution structure.
Figure 2. Structural relatives and insights into a potential function for the Sds3 CTD. (a) A best-
fit backbone superposition of Sds3 CTD (blue) with its relative ZMET2 BAH domain (yellow) as
determined by DALI. The BAH domain binds to a dimethyllysine-containing histone peptide with
the modified residue (rendered transparently) binding to a 'cage' formed by the side chains of
three aromatic residues (shown in spacefilling representation). The BAH domain features two
long insertions in the loop regions of Sds3 CTD connecting the β1 and β2 strands and the β6
and β7 strands; neither of them is shown for clarity. (b) A best-fit backbone superposition of
Sds3 CTD (blue) with the PHF1 Tudor domain (green) that binds to a trimethyllysine-containing
histone peptide. The modified lysine is rendered transparently while the aromatic side chains of
the four residues forming the 'cage' are rendered in spacefilling mode. (c) Two views, identical
to those shown in panels a and b, of the Sds3 CTD with the side chains shown in stick
representation to illustrate the general lack of aromatic side chains (except for F320) on one
edge of the barrel that constitutes the canonical binding pocket for modified lysines and
arginines. (d) A binding screen conducted with purified His6-tagged Sds3 CTD and the
MODified™ histone peptide array. The results illustrate a complete lack of histone-binding
activity for the CTD. The sole dark spot in the array corresponds to a Myc peptide, included in
the array as a positive control, detected by an anti-Myc antibody.
Figure 3. Analysis of the surface properties of Sds3 CTD suggests a nucleic acid-binding
function. (a) An electrostatic potential map calculated using APBS and projected on to the
molecular surface of Sds3 CTD. The poses for the two views are identical to those shown in
figure 2. (b) Electrophoretic mobility shift assays conducted using MBP-tagged Sds3 CTD and
the RNA library following round 3 (left) and round 6 (right) of the SELEX experiments. The
library from round 6 was reverse transcribed and sent for NGS.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 9, 2021. ; https://doi.org/10.1101/2021.08.09.455673doi: bioRxiv preprint
Figure 4. Sds3 CTD binds to a G-quadruplex. (a) 1H-15N HSQC spectra of Sds3 CTD recorded
in the absence (blue) and presence (magenta) of 1 equivalent of T2G4 G-quadruplex DNA.
Strongly perturbed resonances are annotated. To facilitate an objective comparison between
the holo and apo spectra, the contour thresholds were adjusted using the peak intensities of the
'unperturbed' resonances. (b) Front and back views of the molecular surface of Sds3 CTD
colored according to the ratio of the raw peak intensities in the holo and apo spectra (i.e.,
Iholo/Iapo). The surface is rendered semi-transparently to help identify the underlying residue.
Note that the peak intensities of all resonances were diminished because of the larger size of
the resulting complex and due to sample dilution caused by the addition of the quadruplex.
Figure 5. A side-by-side comparison following a best-fit backbone superposition of the Esa1
knotted Tudor domain (KTD; left; PDB ID: 2RO0)) with the Sds3 capped Tudor domain (right).
The views highlight the residues that form the RNA and G-quadruplex binding surfaces inferred
from NMR titration experiments.
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 9, 2021. ; https://doi.org/10.1101/2021.08.09.455673doi: bioRxiv preprint
References
1. Gardner, K. E., Allis, C. D., and Strahl, B. D. (2011) OPERating ON Chromatin, a Colorful
Language where Context Matters. J. Mol. Biol. 409, 36-46
2. Bowman, G. D., and Poirier, M. G. (2015) Post-translational modifications of histones that
influence nucleosome dynamics. Chem. Rev. 115, 2274-2295
3. Evertts, A. G., Zee, B. M., Dimaggio, P. A., Gonzales-Cope, M., Coller, H. A., and Garcia, B.
A. (2013) Quantitative dynamics of the link between cellular metabolism and histone
acetylation. J. Biol. Chem. 288, 12142-12151
4. Moser, M. A., Hagelkruys, A., and Seiser, C. (2014) Transcription and beyond: the role of
mammalian class I lysine deacetylases. Chromosoma 123, 67-78
5. Yang, X. J., and Seto, E. (2008) The Rpd3/Hda1 family of lysine deacetylases: from bacteria
and yeast to mice and men. Nat. Rev. Mol. Cell Biol. 9, 206-218
6. Hayakawa, T., and Nakayama, J. (2011) Physiological roles of class I HDAC complex and
histone demethylase. Journal of Biomedicine and Biotechnology 2011, 129383
7. Xue, Y., Wong, J., Moreno, G. T., Young, M. K., Cote, J., and Wang, W. (1998) NURD, a
novel complex with both ATP-dependent chromatin-remodeling and histone deacetylase
activities. Mol. Cell 2, 851-861
8. Smith, K. T., Sardiu, M. E., Martin-Brown, S. A., Seidel, C., Mushegian, A., Egidy, R.,
Florens, L., Washburn, M. P., and Workman, J. L. (2012) Human family with sequence
similarity 60 member A (FAM60A) protein: a new subunit of the Sin3 deacetylase complex.
Mol. Cell. Proteomics 11, 1815-1828
9. You, A., Tong, J. K., Grozinger, C. M., and Schreiber, S. L. (2001) CoREST is an integral
component of the CoREST- human histone deacetylase complex. Proc. Natl. Acad. Sci. U.
S. A. 98, 1454-1458
10. Bantscheff, M., Hopf, C., Savitski, M. M., Dittmann, A., Grandi, P., Michon, A. M., Schlegl, J.,
Abraham, Y., Becher, I., Bergamini, G., Boesche, M., Delling, M., Dumpelfeld, B., Eberhard,
D., Huthmacher, C., Mathieson, T., Poeckel, D., Reader, V., Strunk, K., Sweetman, G.,
Kruse, U., Neubauer, G., Ramsden, N. G., and Drewes, G. (2011) Chemoproteomics
profiling of HDAC inhibitors reveals selective targeting of HDAC complexes. Nat. Biotechnol.
29, 255-265
11. Guenther, M. G., Barak, O., and Lazar, M. A. (2001) The SMRT and N-CoR corepressors
are activating cofactors for histone deacetylase 3. Mol. Cell. Biol. 21, 6091-6101
12. Carrozza, M. J., Florens, L., Swanson, S. K., Shia, W. J., Anderson, S., Yates, J.,
Washburn, M. P., and Workman, J. L. (2005) Stable incorporation of sequence specific
repressors Ash1 and Ume6 into the Rpd3L complex. Biochim. Biophys. Acta 1731, 77-87;
discussion 75-76
13. Silverstein, R. A., and Ekwall, K. (2005) Sin3: a flexible regulator of global gene expression
and genome stability. Curr. Genet. 47, 1-17
14. Grzenda, A., Lomberk, G., Zhang, J. S., and Urrutia, R. (2009) Sin3: master scaffold and
transcriptional corepressor. Biochim. Biophys. Acta 1789, 443-450
15. Adams, G. E., Chandru, A., and Cowley, S. M. (2018) Co-repressor, co-activator and
general transcription factor: the many faces of the Sin3 histone deacetylase (HDAC)
complex. Biochem. J. 475, 3921-3932
16. David, G., Turner, G. M., Yao, Y., Protopopov, A., and DePinho, R. A. (2003) mSin3-
associated protein, mSds3, is essential for pericentric heterochromatin formation and
chromosome segregation in mammalian cells. Genes Dev. 17, 2396-2405
17. Fleischer, T. C., Yun, U. J., and Ayer, D. E. (2003) Identification and characterization of
three new components of the mSin3A corepressor complex. Mol. Cell. Biol. 23, 3456-3467
18. Banks, C. A. S., Thornton, J. L., Eubanks, C. G., Adams, M. K., Miah, S., Boanca, G., Liu,
X., Katt, M. L., Parmely, T. J., Florens, L., and Washburn, M. P. (2018) A Structured
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 9, 2021. ; https://doi.org/10.1101/2021.08.09.455673doi: bioRxiv preprint
Workflow for Mapping Human Sin3 Histone Deacetylase Complex Interactions Using Halo-
MudPIT Affinity-Purification Mass Spectrometry. Mol. Cell. Proteomics 17, 1432-1447
19. Banks, C. A. S., Zhang, Y., Miah, S., Hao, Y., Adams, M. K., Wen, Z., Thornton, J. L.,
Florens, L., and Washburn, M. P. (2020) Integrative Modeling of a Sin3/HDAC Complex
Sub-structure. Cell Rep 31, 107516
20. He, Y., Imhoff, R., Sahu, A., and Radhakrishnan, I. (2009) Solution structure of a novel zinc
finger motif in the SAP30 polypeptide of the Sin3 corepressor complex and its potential role
in nucleic acid recognition. Nucleic Acids Res. 37, 2142-2152
21. Marcum, R. D., and Radhakrishnan, I. (2019) Inositol phosphates and core subunits of the
Sin3L/Rpd3L histone deacetylase (HDAC) complex up-regulate deacetylase activity. J. Biol.
Chem. 294, 13928-13938
22. Clark, M. D., Marcum, R., Graveline, R., Chan, C. W., Xie, T., Chen, Z., Ding, Y., Zhang, Y.,
Mondragon, A., David, G., and Radhakrishnan, I. (2015) Structural insights into the
assembly of the histone deacetylase-associated Sin3L/Rpd3L corepressor complex. Proc.
Natl. Acad. Sci. U. S. A. 112, E3669-3678
23. Shi, X., Seldin, D. C., and Garry, D. J. (2012) Foxk1 recruits the Sds3 complex and
represses gene expression in myogenic progenitors. Biochem. J. 446, 349-357
24. Meehan, W. J., Samant, R. S., Hopper, J. E., Carrozza, M. J., Shevde, L. A., Workman, J.
L., Eckert, K. A., Verderame, M. F., and Welch, D. R. (2004) Breast cancer metastasis
suppressor 1 (BRMS1) forms complexes with retinoblastoma-binding protein 1 (RBP1) and
the mSin3 histone deacetylase complex and represses transcription. J. Biol. Chem. 279,
1562-1569
25. Salas-Armenteros, I., Perez-Calero, C., Bayona-Feliu, A., Tumini, E., Luna, R., and Aguilera,
A. (2017) Human THO-Sin3A interaction reveals new mechanisms to prevent R-loops that
cause genome instability. EMBO J. 36, 3532-3547
26. Nikolaev, A. Y., Papanikolaou, N. A., Li, M., Qin, J., and Gu, W. (2004) Identification of a
novel BRMS1-homologue protein p40 as a component of the mSin3A/p33(ING1b)/HDAC1
deacetylase complex. Biochem. Biophys. Res. Commun. 323, 1216-1222
27. Gong, C., Qu, S., Lv, X. B., Liu, B., Tan, W., Nie, Y., Su, F., Liu, Q., Yao, H., and Song, E.
(2014) BRMS1L suppresses breast cancer metastasis by inducing epigenetic silence of
FZD10. Nature communications 5, 5406
28. Zhang, S., Lin, Q. D., and Di, W. (2006) Suppression of human ovarian carcinoma
metastasis by the metastasis-suppressor gene, BRMS1. Int. J. Gynecol. Cancer 16, 522-
531
29. Seraj, M. J., Samant, R. S., Verderame, M. F., and Welch, D. R. (2000) Functional evidence
for a novel human breast carcinoma metastasis suppressor, BRMS1, encoded at
chromosome 11q13. Cancer Res. 60, 2764-2769
30. Hurst, D. R. (2012) Metastasis suppression by BRMS1 associated with SIN3 chromatin
remodeling complexes. Cancer Metastasis Rev. 31, 641-651
31. Smith, P. W., Liu, Y., Siefert, S. A., Moskaluk, C. A., Petroni, G. R., and Jones, D. R. (2009)
Breast cancer metastasis suppressor 1 (BRMS1) suppresses metastasis and correlates with
improved patient survival in non-small cell lung cancer. Cancer Lett. 276, 196-203
32. Silveira, A. C., Hurst, D. R., Vaidya, K. S., Ayer, D. E., and Welch, D. R. (2009) Over-
expression of the BRMS1 family member SUDS3 does not suppress metastasis of human
cancer cells. Cancer Lett. 276, 32-37
33. Hurst, D. R., and Welch, D. R. (2011) Unraveling the enigmatic complexities of BRMS1-
mediated metastasis suppression. FEBS Lett. 585, 3185-3190
34. Zimmermann, R. C., and Welch, D. R. (2020) BRMS1: a multifunctional signaling molecule
in metastasis. Cancer Metastasis Rev. 39, 755-768
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 9, 2021. ; https://doi.org/10.1101/2021.08.09.455673doi: bioRxiv preprint
35. Yang, J., Anishchenko, I., Park, H., Peng, Z., Ovchinnikov, S., and Baker, D. (2020)
Improved protein structure prediction using predicted interresidue orientations. Proc. Natl.
Acad. Sci. U. S. A. 117, 1496-1503
36. Holm, L. (2020) Using Dali for Protein Structure Comparison. Methods Mol. Biol. 2112, 29-
42
37. Du, J., Zhong, X., Bernatavichute, Y. V., Stroud, H., Feng, S., Caro, E., Vashisht, A. A.,
Terragni, J., Chin, H. G., Tu, A., Hetzel, J., Wohlschlegel, J. A., Pradhan, S., Patel, D. J.,
and Jacobsen, S. E. (2012) Dual binding of chromomethylase domains to H3K9me2-
containing nucleosomes directs DNA methylation in plants. Cell 151, 167-180
38. Xu, C., Cui, G., Botuyan, M. V., and Mer, G. (2015) Methyllysine Recognition by the Royal
Family Modules: Chromo, Tudor, MBT, Chromo Barrel, and PWWP Domains. . in Histone
Recognition (Zhou, M. M. ed.), Springer. pp 49-82
39. Jurrus, E., Engel, D., Star, K., Monson, K., Brandi, J., Felberg, L. E., Brookes, D. H., Wilson,
L., Chen, J., Liles, K., Chun, M., Li, P., Gohara, D. W., Dolinsky, T., Konecny, R., Koes, D.
R., Nielsen, J. E., Head-Gordon, T., Geng, W., Krasny, R., Wei, G. W., Holst, M. J.,
McCammon, J. A., and Baker, N. A. (2018) Improvements to the APBS biomolecular
solvation software suite. Protein Sci. 27, 112-128
40. Vermeulen, M., Carrozza, M. J., Lasonder, E., Workman, J. L., Logie, C., and Stunnenberg,
H. G. (2004) In vitro targeting reveals intrinsic histone tail specificity of the Sin3/histone
deacetylase and N-CoR/SMRT corepressor complexes. Mol. Cell. Biol. 24, 2364-2372
41. Zhou, Z., Feng, H., Hansen, D. F., Kato, H., Luk, E., Freedberg, D. I., Kay, L. E., Wu, C.,
and Bai, Y. (2008) NMR structure of chaperone Chz1 complexed with histones H2A.Z-H2B.
Nat. Struct. Mol. Biol. 15, 868-869
42. Shimojo, H., Sano, N., Moriwaki, Y., Okuda, M., Horikoshi, M., and Nishimura, Y. (2008)
Novel structural and functional mode of a knot essential for RNA binding activity of the Esa1
presumed chromodomain. J. Mol. Biol. 378, 987-1001
43. Manley, J. L. (2013) SELEX to identify protein-binding sites on RNA. Cold Spring Harb
Protoc 2013, 156-163
44. Adrian, M., Heddi, B., and Phan, A. T. (2012) NMR spectroscopy of G-quadruplexes.
Methods 57, 11-24
45. Maris, C., Dominguez, C., and Allain, F. H. (2005) The RNA recognition motif, a plastic
RNA-binding platform to regulate post-transcriptional gene expression. FEBS J. 272, 2118-
2131
46. Avvakumov, N., and Cote, J. (2007) The MYST family of histone acetyltransferases and
their intimate links to cancer. Oncogene 26, 5395-5407
47. Torres-Machorro, A. L., and Pillus, L. (2014) Bypassing the requirement for an essential
MYST acetyltransferase. Genetics 197, 851-863
48. Khalil, A. M., Guttman, M., Huarte, M., Garber, M., Raj, A., Rivea Morales, D., Thomas, K.,
Presser, A., Bernstein, B. E., van Oudenaarden, A., Regev, A., Lander, E. S., and Rinn, J. L.
(2009) Many human large intergenic noncoding RNAs associate with chromatin-modifying
complexes and affect gene expression. Proc. Natl. Acad. Sci. U. S. A. 106, 11667-11672
49. Porro, A., Feuerhahn, S., and Lingner, J. (2014) TERRA-reinforced association of LSD1 with
MRE11 promotes processing of uncapped telomeres. Cell Rep 6, 765-776
50. Patel, D. J., Phan, A. T., and Kuryavyi, V. (2007) Human telomere, oncogenic promoter and
5'-UTR G-quadruplexes: diverse higher order DNA and RNA targets for cancer therapeutics.
Nucleic Acids Res. 35, 7429-7455
51. Hirschi, A., Martin, W. J., Luka, Z., Loukachevitch, L. V., and Reiter, N. J. (2016) G-
quadruplex RNA binding and recognition by the lysine-specific histone demethylase-1
enzyme. RNA 22, 1250-1260
52. Hoffmann, R. F., Moshkin, Y. M., Mouton, S., Grzeschik, N. A., Kalicharan, R. D., Kuipers,
J., Wolters, A. H., Nishida, K., Romashchenko, A. V., Postberg, J., Lipps, H., Berezikov, E.,
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 9, 2021. ; https://doi.org/10.1101/2021.08.09.455673doi: bioRxiv preprint
Sibon, O. C., Giepmans, B. N., and Lansdorp, P. M. (2016) Guanine quadruplex structures
localize to heterochromatin. Nucleic Acids Res. 44, 152-163
53. Eschenfeldt, W. H., Lucy, S., Millard, C. S., Joachimiak, A., and Mark, I. D. (2009) A family
of LIC vectors for high-throughput cloning and purification of proteins. Methods in molecular
biology 498, 105-115
54. Ferentz, A. E., and Wagner, G. (2000) NMR spectroscopy: a multifaceted approach to
macromolecular structure. Q Rev Biophys 33, 29-65
55. Lee, W., Tonelli, M., and Markley, J. L. (2015) NMRFAM-SPARKY: enhanced software for
biomolecular NMR spectroscopy. Bioinformatics 31, 1325-1327
56. Lee, W., Bahrami, A., Dashti, H. T., Eghbalnia, H. R., Tonelli, M., Westler, W. M., and
Markley, J. L. (2019) I-PINE web server: an integrative probabilistic NMR assignment
system for proteins. Journal of biomolecular NMR 73, 213-222
57. Lohr, F., and Ruterjans, H. (1996) Novel Pulse Sequences for the Resonance Assignment
of Aromatic Side Chains in 13C-Labeled Proteins. J. Magn. Reson. B 112, 259-268
58. Shen, Y., Delaglio, F., Cornilescu, G., and Bax, A. (2009) TALOS+: a hybrid method for
predicting protein backbone torsion angles from NMR chemical shifts. J. Biomol. NMR 44,
213-223
59. Brünger, A. T., Adams, P. D., Clore, G. M., DeLano, W. L., Gros, P., Grosse-Kunstleve, R.
W., Jiang, J. S., Kuszewski, J., Nilges, M., Pannu, N. S., Read, R. J., Rice, L. M., Simonson,
T., and Warren, G. L. (1998) Crystallography & NMR system: A new software suite for
macromolecular structure determination. Acta. Crystallogr. D Biol. Crystallogr. 54, 905-921.
60. Linge, J. P., Habeck, M., Rieping, W., and Nilges, M. (2003) ARIA: automated NOE
assignment and NMR structure calculation. Bioinformatics 19, 315-316.
61. Linge, J. P., Habeck, M., Rieping, W., and Nilges, M. (2004) Correction of spin diffusion
during iterative automated NOE assignment. J. Magn. Reson. 167, 334-342
62. Laskowski, R. A., Rullmannn, J. A., MacArthur, M. W., Kaptein, R., and Thornton, J. M.
(1996) AQUA and PROCHECK-NMR: programs for checking the quality of protein structures
solved by NMR. J. Biomol. NMR 8, 477-486.
63. Martin, M. (2011) Cutadapt Removes Adapter Sequences From High-Throughput
Sequencing Reads. EMBnet 17
64. Afgan, E., Baker, D., Batut, B., van den Beek, M., Bouvier, D., Cech, M., Chilton, J.,
Clements, D., Coraor, N., Gruning, B. A., Guerler, A., Hillman-Jackson, J., Hiltemann, S.,
Jalili, V., Rasche, H., Soranzo, N., Goecks, J., Taylor, J., Nekrutenko, A., and Blankenberg,
D. (2018) The Galaxy platform for accessible, reproducible and collaborative biomedical
analyses: 2018 update. Nucleic acids research 46, W537-W544
65. Bailey, T. L., Boden, M., Buske, F. A., Frith, M., Grant, C. E., Clementi, L., Ren, J., Li, W.
W., and Noble, W. S. (2009) MEME SUITE: tools for motif discovery and searching. Nucleic
acids research 37, W202-208
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 9, 2021. ; https://doi.org/10.1101/2021.08.09.455673doi: bioRxiv preprint
260 270 280 290 300 310 320
M.musculus (250-326)
H.sapiens (250-326)
B.taurus (250-326)
G.gallus (235-311)
X.laevis (243-319)
D.rerio (240-316)
M.musculus (249-323)
H.sapiens (249-323)
B.taurus (249-323)
G.gallus (247-322)
X.laevis (248-322)
D.rerio (249-323)
AQ RFEAR IEDGKLYYDKRWYHKSQAIYLESKDNQKL SCVISSVGANEIWVRKTSDSTKMRIYVGQLQRGLFVIRRRS
AQ RFEAR IEDGKLYYDKRWYHKSQAIYLESKDNQKL SCVISSVGANEIWVRKTSDSTKMRIYLGQLQRGLFVIRRRS
AQ RFEAR IEDGKLYYDKRWYHKSQAIYLESKDNQKL SCVISSVGANEIWVRKTSDSTKMRIYLGQLQRGLFVIRRRS
AQ RFEAR IEDGKLYYDKRWYHKSQAIYLESKENTK ISCVISSVGANEIWVRKTSDSTKMRIYLGQLQRGVFVIRRRS
VQRF EAR IEDGK LYYDKRWYHKSQAIYLESKDNNKMSCVISSVGNNEIWVRKSSDSTKVRIYLGQLQKGLFVIRRRS
SQRYEAR IEEGKLYYDKRWYHKSQAIYLESKENTK ISCVISSVGTNEIWVRKTSDSTKMRIYLGQLQRGAFIIRRRS
KHLHSARSEEGRL YYD GEWY IRGQTICIDRKDECPTSAVITT INHDEVWFKRP-DGS KSKLYISQLQKGKYSIKHS -
KHLHSARSEEGRL YYD GEWY IRGQTICIDKKDECPTSAVITT INHDEVWFKRP-DGS KSKLYISQLQKGKYSIKHS -
KHLHSARSEEGRL YYD GEWY IRGQTICIDKKDECPTSAVITT INHDEVWFKRP-DGS KSKLYISQLQKGKYSIKHS -
KHLHSARSEEGRL YYD GEWY GRGQTIYIDKKDECPTSAIITT INHDEVWFKRP-DGS KSKLYISQLQKGKYSIKHNH
KHQH SARSEEGRLHYDGEWY GRGQTICIDKKDEFPTSAVITT INSDEVWFKRQ-DGS KSKLYISQLQKGKYSIKHI-
KLQHNARSEDGR LFYD GE WY SRGQAITIDKKDEYPTSAVITT INHDEVWFKRV-DGS KSKLYVSQLQKGKYTVKHA-
Sds3BRMS1L
a
b
c
d
e
N
C
N
C
N
N
N
N
N
β2
β1
β3
β4
β5
β6
β7
α1
β8
C
β1
β2
β3
β4
β5
β6
β7
α1 β8
β2
β3
β4
β5
β6
β7
α1
β8
C
β1
β1 β2 β3 β4 β5 β6 β7 β8α1
Figure 1
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 9, 2021. ; https://doi.org/10.1101/2021.08.09.455673doi: bioRxiv preprint
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 9, 2021. ; https://doi.org/10.1101/2021.08.09.455673doi: bioRxiv preprint
0.1 0.5 1 5 10
MBP-Sds3 CTD (µM)
0 0.01 0.05
MBP-Sds3 CTD (µM)
0 0.01 0.05 0.1 0.5 1 5 10
MBP-Sds3 CTD (µM)
0 0.01 0.05
MBP-Sds3 CTD (µM)
0 0.01 0.05
Round 3 Round 6
a
b
Figure 3
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 9, 2021. ; https://doi.org/10.1101/2021.08.09.455673doi: bioRxiv preprint
W298
R309
E296
M308
G313
N295
A294
D259
10 9 8 76
130
125
120
115
110
15N
(ppm)
1H (ppm)
-T2G4/+T2G4
G313
Q316sc
N295sc
I297
N295
Q314sc
L315
M308
K271
V312
R309
Q273
D259
L262
Y269
Y311
A250
S286
H270
Q316
A294
a
b
Y311
V312
Q316
Figure 4
18.75% 12.5% 6.25%
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 9, 2021. ; https://doi.org/10.1101/2021.08.09.455673doi: bioRxiv preprint
Figure 5
Esa1 KTD Sds3 CTD
N
C
N
W298
M308
R309
Y311
E296
I297
N295 V312
A294
L315
Q316
W66
Y53
L63
R62
K61
.CC-BY-NC-ND 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 9, 2021. ; https://doi.org/10.1101/2021.08.09.455673doi: bioRxiv preprint
