Non-canonical DNA structures, notably G-quadruplexes and i-motifs, draw significant attention because biological evidence suggests that they play crucial roles in a variety of disease-related biological processes. G-quadruplex DNA is composed of planar guanine tetrads which are engaged in efficient p-p stacking and are further stabilized by monovalent central cation (e.g. K⁺ or Na⁺). Sequences with G-quadruplex forming potential are present in telomeres and in oncogene promoters, according to bioinformatics studies. I-motifs are intercalated hemi-protonated cytosine-rich structures formed in the C-rich sequences. Naturally, such sequences are present in the regions complimentary to the G-rich parts of genome.

In this work we have investigated nine variants of telomeric DNA with the repeat (TTGGGG)n from the organism Tetrahymena thermophila using biophysical and x-ray crystallographic studies. Biophysical characterization showed that all sequences folded into stable GQs which adopted a variety of conformations, most commonly parallel and hybrid. Native PAGE suggested that most of the sequences form multiple species in the presence of potassium. All species, but one, are monomolecular. We successfully crystallized two variants, TET25 (resolution 1.56 Å) and TET26 (three crystal forms with resolution 1.99 and 1.97 Å) and solved the structures via molecular replacement. TET25 adopted a hybrid (3+1) conformation with a four G-tetrad core, three lateral loops, one propeller loop, and 5’ snapback. TET26 fold into a parallel GQ conformation with a four G-tetrad core and three TT propeller loops. We have also investigated binding of N-methylmesoporphyrin IX (NMM) to all sequences and crystallized three variants with NMM. NMM induces parallel fold in all sequences. Both crystal structures display 5’-5’ dimers of parallel GQs with NMM bonded to the 3’ G-tetrad. NMM binds GQ with one of its faces and another NMM molecule with another. Our structural data demonstrate great plasticity of the telomeric sequence from the telomeric region of T. thermophila where small variation in the overhang length and composition leads to drastically distinct GQ structures.

I will also share our progress toward the structure an i-motif DNA from the HRAS oncogene promoter as well as the structure of repetitive DNA (CAGAGG)n from difficult-to-replicate regions of the mouse genome implicated in replication stress. Our findings have potential to contribute to the development of new and efficient anticancer therapies.

The simple elegance of the Watson-Crick DNA model reported in 1953[1] belies an underlying complexity that is central to all life. However, about thirty years then elapsed before the true complexity of DNA was revealed in high resolution crystals structures of oligonucleotides. In these structures, DNA was captured in three distinct helical forms, Z [2], B [2], and A [3], providing the first evidence for the remarkable ability of DNA to adopt different stable conformations influenced by nucleobase sequence. Since then, our understanding of the fundamental properties of DNA has been challenged further with efforts to expand the genetic code through the creation of unnatural nucleobases. These new entities include nucleobases that pair strictly through hydrophobic interactions [4, 5] and those that pair through hydrogen bonding interactions [6]. The latter nucleobases were created by Benner and coworkers and are referred to as the Artificially Expanded Genetic Information System (AEGIS) [7]. AEGIS takes advantage of alternative hydrogen bonding arrangements between Watson-Crick like pairs, a large purine-like nucleobase and a small pyridimine-like nucleobase that exclusively pair to one another rather than natural nucleobases. This concept has produced an expanded genetic code, Hachimoji DNA [8] comprising 8 letters, 4 natural and 4 unnatural, and most recently Alien DNA, comprising 4 unnatural nucleobases. These systems including unnatural base pairs (UBPs) expand the structural landscape of DNA through the creation of duplexes that do not conform to known helical forms.

In previous work, we have reported structures including up to 6 UBPs within 16 bp duplex DNA structures [9]. Our most recent work on Alien DNA includes structures with 12 UBPs of 16 base pairs (almost Alien DNA) contained within the structure captured in B-like and A-like helical forms. The B-like structures were obtained through the use of our host-guest system, which is selective for DNA sequences that can adopt helical forms that are more similar to B than A-form DNA. In this system, the N-terminal fragment of Moloney murine leukemia virus reverse transcriptase serves as the host and a 16-mer DNA duplex as the guest [10]. Using this system, we have determined structures of numerous DNA sequences at relatively high resolution (1.6-1.8 Å) including now one with 12 UBPs. One of the almost Alien DNA sequences including 12 UBPs has crystallized in three different crystal forms, two that diffract to 1.2 Å, providing the first very detailed structural information for these UBPs including sugar conformations. These latest structures of almost Alien DNA will be presented here along with comparative analyses with natural and other less Alien DNA structures including UBPs.

[1] Watson, J.D. and F.H. Crick, (1953). Nature 171, 737-8.

[2] Wang, A.H., G.J. Quigley, F.J. Kolpak, J.L. Crawford, J.H. van Boom, G. van der Marel and A. Rich, (1979). Nature 282, 680-6.

[3] Heinemann, U., H. Lauble, R. Frank and H. Blocker, (1987). Nucleic Acids Res 15, 9531-50.

[4] Leconte, A.M., G.T. Hwang, S. Matsuda, P. Capek, Y. Hari and F.E. Romesberg, (2008). J Am Chem Soc 130, 2336-43.

[5] Hirao, I., M. Kimoto, T. Mitsui, T. Fujiwara, R. Kawai, A. Sato, Y. Harada and S. Yokoyama, (2006). Nat Methods 3, 729-35.

[6] Yang, Z., D. Hutter, P. Sheng, A.M. Sismour and S.A. Benner, (2006). Nucleic Acids Res 34, 6095-101.

[7] Sefah, K., Z. Yang, K.M. Bradley, S. Hoshika, E. Jimenez, L. Zhang, G. Zhu, S. Shanker, F. Yu, D. Turek, W. Tan and S.A. Benner, (2014). Proc Natl Acad Sci U S A 111, 1449-54.

[8] Hoshika, S., N.A. Leal, M.J. Kim, M.S. Kim, N.B. Karalkar, H.J. Kim, A.M. Bates, N.E. Watkins, Jr., H.A. SantaLucia, A.J. Meyer, S. DasGupta, J.A. Piccirilli, A.D. Ellington, J. SantaLucia, Jr., M.M. Georgiadis and S.A. Benner, (2019). Science 363, 884-887.

[9] Georgiadis, M.M., I. Singh, W.F. Kellett, S. Hoshika, S.A. Benner and N.G. Richards, (2015). J Am Chem Soc 137, 6947-55.

[10] Cote, M.L., S.J. Yohannan and M.M. Georgiadis, (2000). Acta Crystallogr D Biol Crystallogr 56, 1120-31.

Apurinic/apyrimidinic endonuclease 1 (APE1) is a well-known endonuclease specifically targeting an AP site to initiate base excision repair. Interestingly, APE1 also bears 3′-to-5′ exonuclease activity that shows very different catalytic properties and cellular functions. The 3'-to-5' exonuclease activity of APE1 is responsible for processing matched/mismatched terminus of duplex DNA in various DNA repair pathways, as well as for nucleoside analogs removal associated with drug resistance. Due to the limited information of APE1’s exonucleolytic catalysis, its fundamental roles in various DNA repair pathways and in drug resistance are poorly understood. In addition, how APE1 exonucleolytically recognizes and processes the terminus of duplex DNA without base preference remain unclear. We determined the first two APE1-dsDNA complex structures, which displayed a dsDNA end-binding mode. Integration of our structures, biochemical assays, and molecular dynamics simulation reveals the general rules of APE1 in handling various dsDNA substrates. The DNA binding-induced RM (Arg176 and Met269) bridge formation in active site and DNA-binding modes transition between matched and mismatched termini of dsDNA compose the exquisite machinery for substrate selection, binding, and digestion. Our studies pave the way for understanding the dsDNA terminal-processing-related cellular functions and drug resistance mechanisms of APE1.

(Ref: Nat Commun. 12, 601 (2021). https://doi.org/10.1038/s41467-020-20853-2. )

Structural characterization of clinically reported missense mutations identified in BRCA1

Neha Mishra^{1, 2}, Suchita Dubey^{1, 2} Ashok K Varma^{1, 21}Advanced centre for Treatment, Research and Education in cancer, Kharghar, Navi Mumbai, Maharashtra – 410210, INDIA, ²Homi Bhabha National Institute, Training School Complex, Anushaktinagar, Mumbai, Maharashtra – 400094, INDIA nmishra@actrec.gov.in

Germline loss of function mutations in Breast Cancer Susceptibility gene 1 (BRCA1) and Breast Cancer Susceptibility gene 2 (BRCA2) are known to be responsible for Hereditary Breast and Ovarian Cancer Syndrome (HBOCS). BRCA1 encodes 1863 amino acids consisting of N-terminus RING domain, intrinsically disordered central DNA binding region and highly conserved two tandem repeats of BRCTs constituting the C-terminus domain (CTD)[1]. The RING domain forms a heterodimer with BRCA1-associated RING domain protein 1 (BARD1) and acts as E3- Ubiquitin ligase. However, C-terminal of the BRCA1 is known to interact with proteins containing the consensus sequences of pS-X-X-F motif to mediate different complex formation at the time of Double Strand Break Repair (DSBR). Majority of the missense mutations are found in the BRCT, RING domain and few in the central domain of BRCA1. Several studies have been performed to classify such variants of uncertain significance (VUS) in BRCA1 as pathogenic or neutral but the exact molecular mechanism of pathogenicity still remains to be deciphered[2, 3]. The aim of the present study is to evaluate the structural significance of these missense mutations located in the central and C-terminus functional domains of BRCA1 using biophysical, biochemical and in-silico tools. It was found that BRCA1 Arg866Cys in the central region, Thr1691Arg and Gly1801Asp in the BRCT domain show conformational alterations. The central non-specific DNA binding domain has also been evaluated for its conformational changes in the presence and absence of super-coiled DNA. However, a reduced DNA binding ability was observed for the mutant as compared to the wild- type protein. Further, the central region of BRCA1 has been assessed for its intrinsically disordered behaviour. Addition of 2,2,2-trifluoroethanol (TFE) led to gain of structure of the central region and therefore, less susceptibility towards proteolysis. The mutations have been characterized with the help of Size exclusion chromatography (SEC), Circular Dichroism spectroscopy, nano DSF, EMSA. With this information we would further extend our studies for protein-protein interactions of the wild type and mutant proteins using ITC, SPR and co-crystallization. The reported results will enhance our understanding towards the fundamental structural differences arising due to cancer predisposing missense mutations.

Keywords: Hereditary Breast Cancer; BRCA1; secondary structural changes; BRCT; Intrinsically disordered protein regions (IDPRs); DNA binding regions

Acknowledgement-Funding for this study was supported by Annual Scientific Fund from ACTREC-TMC. The authors thank the XRD facility at ACTREC for providing necessary support to this study.

[1] R. Roy, J. Chun, and S. N. Powell, “BRCA1 and BRCA2 : different roles in a common pathway of genome protection,” Nat. Rev. Cancer.

[2] R. W. Anantha et al., “Functional and mutational landscapes of BRCA1 for homology-directed repair and therapy resistance,” pp. 1–21, 2017.

[3] P. Bouwman, H. Van Der Gulden, I. Van Der Heijden, R. Drost, and C. N. Klijn, “RESEARCH ARTICLE A High-Throughput Functional Complementation Assay for Classifi cation of BRCA1 Missense Variants,” 2013.

A solid-solid phase transition (SSPT) occurs between distinguishable crystalline forms. SSPTs have been studied extensively in metallic alloys, inorganic salt or small organic molecular crystals, but much less so in biomacromolecular crystals. In particular, SSPTs involving large-scale molecular changes that are important to biological function are largely unexplored, yet may enhance our understanding of conformational space. Here, we report a systematic study of the ligand-induced SSPT in crystals of the adenine riboswitch aptamer RNA (riboA) using a combination of polarized video microscopy (PVM), solution atomic force microscopy (AFM), and time-resolved serial crystallography (TRX). The SSPT, driven by large conformational changes induced by ligand, transforms the crystal lattice from monoclinic (apo), to triclinic (intermediate lattice in a ligand-bound conformation), to orthorhombic (final bound conformational and lattice state). Using crystal structures of each state, we mapped out the changes to the crystal packing interfaces, which define the interplay between molecular conformation and crystal phase, which were corroborated by solution AFM. Using PVM to monitor changes in crystal birefringence, we characterized the kinetics of the SSPT in crystals of different sizes and ligand concentration. Together, these studies illustrate a practical approach for characterizing SSPT in biomacromolecular crystals involving large conformational changes, and provide useful spatiotemporal data for informing time-resolved crystallography experiments.

CD72 is an inhibitory co-receptor that negatively regulates B cell antigen receptor (BCR) signalling. The ligand-binding domain of CD72 at the extracellular region belongs to the C-type lectin-like domain (CTLD) superfamily. We have demonstrated that it recognizes the nuclear autoantigen Sm/RNP composed of proteins and RNA, and suppresses autoimmune diseases such as systemic lupus erythematosus [1]. The crystal structure of the ligand-binding domain of mouse CD72^a, a lupus-resistant allele, has been determined at 1.2 Å resolution. Electrostatic potential analysis of the molecular surface of CD72^a-CTLD suggest that charge distribution at the putative ligand-binding site may affect the binding affinity between CD72 and Sm/RNP.

We have determined the crystal structure of the ligand-binding domain of mouse CD72^c, a lupus-susceptible allele with reduced affinity to Sm/RNP. The obtained crystals were large enough for X-ray diffraction experiments of about 200 µm cubic, but clusters of hundreds or thousands of microcrystals (Fig. 1). Development of the micro focus X-ray beam and rapid automated data collection [2] and processing [3] systems at SPring-8 enabled us to obtain a full data set that allowed the successful structure determination and refinement at 2.5 Å resolution (Fig. 2). We took 1,400 of small-wedge (10 degree) data from 14 crystals. The data were classified based on the unit-cell dimensions or correlation coefficient between data and merged to a full data set for structure determination. Analysis of the hierarchical clustering of the small-wedge data shows that the crystal packing varies along with the c-axis direction, but no significant conformational variations were observed among the crystal structures.

The obtained structure reveals that substitutions of amino acids at the ligand-binding site do cause the inversion of the charge distribution of the molecular surface as we hypothesized. Charge repulsion between CD72^c-CTLD and strong negative charges of RNA of Sm/RNP would be the molecular mechanism of reduced affinity.

[1] Akatsu, C., Shinagawa, K., Numoto, N., Liu, Z., Ucar, A. K., Aslam, M., Phoon, S., Adachi, T., Furukawa, K., It,o N. & Tsubata, T. (2016) J. Exp. Med. 213,2691.

[2] Hirata, K., Yamashita, K., Ueno, G., Kawano, Y., Hasegawa, K., Kumasaka, T. & Yamamoto M. (2019) Acta Cryst. D75, 138.

[3] Yamashita, K., Hirata, K. & Yamamoto, M. (2018) Acta Cryst. D74, 441.