research papers\(\def\hfill{\hskip 5em}\def\hfil{\hskip 3em}\def\eqno#1{\hfil {#1}}\)

Journal logoSTRUCTURAL
BIOLOGY
ISSN: 2059-7983

Racemic crystal structures of A-DNA duplexes

crossmark logo

aUniversité de Bordeaux, CNRS, Bordeaux Institut National Polytechnique, CBMN (UMR 5248), 33600 Pessac, France, bDepartment of Pharmacy and Center for Integrated Protein Science, Ludwig-Maximilians-University, Munich, Germany, and cUniversité de Bordeaux, CNRS, INSERM, Institut Européen de Chimie et Biologie (UAR3033 and US001), 33600 Pessac, France
*Correspondence e-mail: ivan.huc@cup.lmu.de

Edited by K. Diederichs, University of Konstanz, Germany (Received 27 December 2021; accepted 10 April 2022; online 9 May 2022)

The ease with which racemic mixtures crystallize compared with the equivalent chiral systems is routinely taken advantage of to produce crystals of small molecules. However, biological macromolecules such as DNA and proteins are naturally chiral, and thus the limited range of chiral space groups available hampers the crystallization of such molecules. Inspiring work over the past 15 years has shown that racemic mixtures of proteins, which were made possible by impressive advances in protein chemical synthesis, can indeed improve the success rate of protein crystallization experiments. More recently, the racemic crystallization approach was extended to include nucleic acids as a possible aid in the determination of enantiopure DNA crystal structures. Here, findings are reported that suggest that the benefits may extend beyond this. Two racemic crystal structures of the DNA sequence d(CCCGGG) are described which were found to fold into A-form DNA. This form differs from the Z-form DNA conformation adopted by the chiral equivalent in the solid state, suggesting that the use of racemates may also favour the emergence of new conformations. Importantly, the racemic mixture forms interactions in the solid state that differ from the chiral equivalent (including the formation of racemic pseudo-helices), suggesting that the use of racemic DNA mixtures could provide new possibilities for the design of precise self-assembled nanomaterials and nanostructures.

1. Introduction

Racemic mixtures crystallize much more frequently as racemates in non-Sohncke (i.e. enantiogenic) space groups than as conglomerates. They also tend to be more amenable to crystallization than the equivalent enantiopure samples (Jacques et al., 1994[Jacques, J., Collet, A. & Wilen, S. (1994). Enantiomers, Racemates and Resolutions. Malabar: Krieger Publishing Co.]; Matthews, 2009[Matthews, B. W. (2009). Protein Sci. 18, 1135-1138.]). Crystallography in the chemistry world (i.e. small-molecule crystallography) commonly involves racemates or achiral molecules. In contrast, biological macromolecular crystallography involves the study of naturally chiral molecules, such as DNA and proteins, which do not have access to non-Sohncke space groups and are thus limited with respect to packing options (Wukovitz & Yeates, 1995[Wukovitz, S. W. & Yeates, T. O. (1995). Nat. Struct. Mol. Biol. 2, 1062-1067.]). However, impressive improvements in chemical peptide (and protein) synthesis over the past twenty years have enabled the generation of racemic protein mixtures, thereby enabling racemic crystallographic methods. Indeed, racemic protein mixtures have been shown to crystallize more readily, invariably forming racemic crystals and enabling successful structure determination when this had failed with the natural L-amino-acid backbone (Yeates & Kent, 2011[Yeates, T. O. & Kent, S. B. H. (2011). Annu. Rev. Biophys. 41, 41-61.]). This racemic approach to the crystallization of naturally chiral protein molecules was subsequently shown to be equally effective for the crystallization of DNA thanks to the commercial availability of affordable L-DNA sequences, with a variety of structural motifs now shown to form racemic crystals, enabling successful high-resolution structures to be determined (Doi et al., 1993[Doi, M., Inoue, M., Tomoo, K., Ishida, T., Ueda, Y., Akagi, M. & Urata, H. (1993). J. Am. Chem. Soc. 115, 10432-10433.]; Rypniewski et al., 2006[Rypniewski, W., Vallazza, M., Perbandt, M., Klussmann, S., DeLucas, L. J., Betzel, C. & Erdmann, V. A. (2006). Acta Cryst. D62, 659-664.]; Mandal et al., 2014[Mandal, P. K., Collie, G. W., Kauffmann, B. & Huc, I. (2014). Angew. Chem. Int. Ed. 53, 14424-14427.], 2016[Mandal, P. K., Collie, G. W., Srivastava, S. C., Kauffmann, B. & Huc, I. (2016). Nucleic Acids Res. 44, 5936-5943.]; Drozdzal et al., 2016[Drozdzal, P., Gilski, M. & Jaskolski, M. (2016). Acta Cryst. D72, 1203-1211.]). Although a limited number of racemic DNA structures are available, some trends can be noted, including that racemates generally (although not always) crystallize in conditions under which the single enantiomer has been shown to crystallize, that equal or better quality crystallographic data are obtained from the racemates, and that the DNA conformations are the same in single-enantiomer and racemic DNA crystals grown under similar conditions.

Racemic DNA crystallography is thus of potential use to the structural biology field by enabling successful structure determination (particularly for challenging systems; Mandal et al., 2016[Mandal, P. K., Collie, G. W., Srivastava, S. C., Kauffmann, B. & Huc, I. (2016). Nucleic Acids Res. 44, 5936-5943.]). In parallel, the combination of the physico-chemical properties of DNA (folding, stability, programmability and self-recognition) with access to all 230 space groups through the use of racemic DNA mixtures may give access to novel materials and/or self-assembled nano­structures. Indeed, beyond the variety of naturally occurring DNA structures (Neidle, 2021[Neidle, S. (2021). J. Biol. Chem. 296, 100553.]), DNA has also been studied extensively as a tool for the construction of precise self-assembled nano­materials and nano-objects with intriguing properties (Seeman, 2003[Seeman, N. C. (2003). Nature, 421, 427-431.], 2010[Seeman, N. C. (2010). Annu. Rev. Biochem. 79, 65-87.]; Rothemund, 2006[Rothemund, P. W. K. (2006). Nature, 440, 297-302.]; Aldaye et al., 2008[Aldaye, F. A., Palmer, A. L. & Sleiman, H. F. (2008). Science, 321, 1795-1799.]; Andersen et al., 2009[Andersen, E. S., Dong, M., Nielsen, M. M., Jahn, K., Subramani, R., Mamdouh, W., Golas, M. M., Sander, B., Stark, H., Oliveira, C. L. P., Pedersen, J. S., Birkedal, V., Besenbacher, F., Gothelf, K. V. & Kjems, J. (2009). Nature, 459, 73-76.]; McLaughlin et al., 2011[McLaughlin, C. K., Hamblin, G. D. & Sleiman, H. F. (2011). Chem. Soc. Rev. 40, 5647-5656.]). In this vein, here we report two racemic crystal structures of the DNA sequence d(CCCGGG). This sequence forms highly ordered racemic crystals of an A-form duplex DNA, in contrast to the Z-form DNA observed in the crystal structure of the enantiopure equivalent (Gautham et al., 1999[Gautham, N., Karthe, P. & Krishnaswamy, S. (1999). Perspectives in Structural Biology, edited by M. Vijayan, N. Yathindra & A. Kolaskar, pp. 593-601. Hyderabad: Indian Academy of Sciences/Universities Press.]). In addition, we observe the formation of racemic pseudo-helices within the crystal lattice, as well as the formation of large, water-filled channels. These racemic DNA structures and their unique, racemate-specific features thus suggest that racemic mixtures of DNA may enable the formation of solid-state organizations that are inaccessible to chiral samples.

2. Materials and methods

2.1. Crystallization and diffraction data collection

The L- and D-forms of the DNA sequence d(CCCGGG) were synthesized and purified by ChemGenes Corporation (USA). 2 mM DNA solutions of each enantiomer were prepared using ultrapure water. The enantiopure solutions were annealed at 353 K for 20 min and gradually cooled overnight to 293 K. After annealing, a racemic mixture was prepared by mixing the enantiopure solutions in an equimolar ratio. Crystallization experiments were carried out by the hanging-drop vapour-diffusion method at 293 K in hanging drops composed of 0.5 µl racemic DNA sample plus an equal volume of crystallization reagent: 2 M ammonium sulfate plus 100 mM Tris buffer pH 8.5 for crystal form 1 and 1.8 M ammonium sulfate plus 100 mM MES buffer pH 6.5 and 10 mM cobalt(II) chloride for crystal form 2 (Fig. 1[link]).

[Figure 1]
Figure 1
Crystals of D/L-d(CCCGGG) viewed under polarized light. X-ray diffraction measurements revealed that the crystals in (a) and (b) belonged to the achiral space groups [R\bar 3] and P21/n, respectively.

For low-temperature diffraction measurements, crystals were mounted in nylon loops and vitrified in a stream of cold nitrogen gas at 150 K. Initial diffraction patterns confirmed the crystals to contain ordered DNA and revealed the presence of randomly oriented salt microcrystals. The DNA crystals were soaked briefly in 4 M trimethylamine N-oxide (TMO), which prevented the rapid formation of ammonium sulfate crystals and provided good cryoprotection (Mueller-Dieckmann et al., 2011[Mueller-Dieckmann, C., Kauffmann, B. & Weiss, M. S. (2011). J. Appl. Cryst. 44, 433-436.]). X-ray diffraction data were measured using a microfocus rotating-anode Rigaku FR-X diffracto­meter with Cu Kα radiation and a hybrid pixel detector (Dectris PILATUS 200K). Crystals 1 and 2 diffracted to resolutions of 2.48 and 2.80 Å, respectively. The diffraction data were indexed, integrated and scaled using the XDS package (Kabsch, 2010[Kabsch, W. (2010). Acta Cryst. D66, 125-132.]) and data statistics are summarized in Table 1[link].

Table 1
Summary of X-ray diffraction data and refinement

Values in parentheses are for the outer shell.

Crystal form 1 2
Data collection
 Space group [R\bar 3] P21/n
a, b, c (Å) 105.41, 105.41, 56.13 48.88, 41.43, 71.96
α, β, γ (°) 90.00, 90.00, 120.00 90.00, 97.47, 90.00
 Resolution range (Å) 23.91–2.48 (2.57–2.48) 31.49–2.80 (2.90–2.80)
 Total reflections 15782 (1279) 13035 (1332)
 Unique reflections 8101 (729) 6839 (685)
 Multiplicity 1.9 (1.8) 1.9 (1.9)
 Completeness 0.97 (0.89) 0.99 (0.99)
 Mean I/σ(I) 10.71 (2.02) 11.19 (2.47)
Rmerge (%) 5.74 (42.49) 7.30 (33.93)
Rmeas (%) 8.11 (60.09) 10.32 (47.99)
 CC1/2 0.997 (0.926) 0.990 (0.924)
 Wilson B factor (Å2) 49.67 49.82
Refinement
 No. of reflections 8101 (725) 6839 (685)
Rwork (%) 24.96 (36.60) 27.95 (41.87)
Rfree (%) 26.83 (42.57) 31.82 (52.96)
 No. of non-H atoms
  Total 539 751
  DNA 480 720
  Ions 2 Co, 1 Cl
  TMO 10 20
  Water 49 8
 R.m.s.d., bond lengths (Å) 0.003 0.007
 R.m.s.d., angles (°) 0.39 0.56
 Mean B factor (Å2) 34.27 49.17
 PDB code 6gn2 6gn3

2.2. Structure determination and refinement

Crystal 1 belonged to the achiral trigonal space group [R\bar 3{\hbox {:}}H], with unit-cell parameters a = b = 105.41, c = 56.13 Å. The E-statistics distribution for this crystal indicates that the structure is centrosymmetric (Marsh, 1995[Marsh, R. E. (1995). Acta Cryst. B51, 897-907.]). The structure was determined by molecular replacement (MR) using Phaser (McCoy et al., 2007[McCoy, A. J., Grosse-Kunstleve, R. W., Adams, P. D., Winn, M. D., Storoni, L. C. & Read, R. J. (2007). J. Appl. Cryst. 40, 658-674.]) from the CCP4 suite (Winn et al., 2011[Winn, M. D., Ballard, C. C., Cowtan, K. D., Dodson, E. J., Emsley, P., Evans, P. R., Keegan, R. M., Krissinel, E. B., Leslie, A. G. W., McCoy, A., McNicholas, S. J., Murshudov, G. N., Pannu, N. S., Potterton, E. A., Powell, H. R., Read, R. J., Vagin, A. & Wilson, K. S. (2011). Acta Cryst. D67, 235-242.]). MR was successful using the crystal structure of d(CCCCGGGG)2 (PDB entry 1vt5; Haran et al., 1987[Haran, T. E., Shakked, Z., Wang, A. H.-J. & Rich, A. (1987). J. Biomol. Struct. Dyn. 5, 199-217.]) as a search model, after failing when using the Z-DNA conformation of the single-enantiomer crystal of D-d(CCCGGG)2 (Gautham et al., 1999[Gautham, N., Karthe, P. & Krishnaswamy, S. (1999). Perspectives in Structural Biology, edited by M. Vijayan, N. Yathindra & A. Kolaskar, pp. 593-601. Hyderabad: Indian Academy of Sciences/Universities Press.]). To build the MR model, the octa­nucleotide DNA strands were shortened to six residues by deletion of their first and eighth nucleotides. For MR, a multiple-copy search was carried out for two, three and four duplexes in the asymmetric unit. The MR solution, with a translation-function Z (TFZ) score of 9.1 and a log-likelihood gain (LLG) score of 118, revealed two independent D-DNA duplexes (hereafter named helices 1.1 and 1.2) in the asymmetric unit. The initial maps showed clearly resolved electron density for the duplex and allowed the unambiguous identification of the chirality of symmetry-related D- and L-enantiomers. The MR solution was refined using phenix.refine (Afonine et al., 2012[Afonine, P. V., Grosse-Kunstleve, R. W., Echols, N., Headd, J. J., Moriarty, N. W., Mustyakimov, M., Terwilliger, T. C., Urzhumtsev, A., Zwart, P. H. & Adams, P. D. (2012). Acta Cryst. D68, 352-367.]) in the Phenix suite (Liebschner et al., 2019[Liebschner, D., Afonine, P. V., Baker, M. L., Bunkóczi, G., Chen, V. B., Croll, T. I., Hintze, B., Hung, L.-W., Jain, S., McCoy, A. J., Moriarty, N. W., Oeffner, R. D., Poon, B. K., Prisant, M. G., Read, R. J., Richardson, J. S., Richardson, D. C., Sammito, M. D., Sobolev, O. V., Stockwell, D. H., Terwilliger, T. C., Urzhumtsev, A. G., Videau, L. L., Williams, C. J. & Adams, P. D. (2019). Acta Cryst. D75, 861-877.]) with maximum-likelihood targets and NCS restraints (torsion angle). The restraint dictionary for nucleic acids was compiled by Parkinson et al. (1996[Parkinson, G., Vojtechovsky, J., Clowney, L., Brünger, A. T. & Berman, H. M. (1996). Acta Cryst. D52, 57-64.]). 10% of the unique reflections were used to calculate Rfree (Brünger, 1992[Brünger, A. T. (1992). Nature, 355, 472-475.]). Refinement of the coordinates was initially carried out with rigid-body refinement and individual coordinate refinement using simulated annealing. Subsequently, individual coordinate refinement was carried out in real space. Individual isotropic refinement of atomic displacement parameters (ADPs) was performed in combination with translation–libration–screw (TLS) refinement with the parameters generated in phenix.refine (Afonine et al., 2012[Afonine, P. V., Grosse-Kunstleve, R. W., Echols, N., Headd, J. J., Moriarty, N. W., Mustyakimov, M., Terwilliger, T. C., Urzhumtsev, A., Zwart, P. H. & Adams, P. D. (2012). Acta Cryst. D68, 352-367.]) using each DNA strand as a separate TLS group. After each refinement step, visual inspection of the model and the electron-density maps was carried out in Coot (Emsley et al., 2010[Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. (2010). Acta Cryst. D66, 486-501.]) using 2FoFc and difference Fourier (Fo − Fc) maps. The difference Fourier maps were contoured at the 5σ level in order to place ions or small molecules from crystallization reagents or cryoprotectants and at the 3σ level to identify water molecules. After refinement of the DNA strands, positive peaks were observed above 7σ in the difference Fourier map at two locations (Supplementary Fig. S1). In both locations, modelling and refinement was attempted with fully occupied water, sulfate ions and Tris buffer (from the crystallization reagent). In all cases the difference Fourier map showed residual electron density. Fitting of TMO (the cryoprotectant) into the electron density was carried out using LigandFit (Terwilliger et al., 2007[Terwilliger, T. C., Adams, P. D., Moriarty, N. W. & Cohn, J. D. (2007). Acta Cryst. D63, 101-107.]). Restraint generation and optimization for TMO was performed using eLBOW (Moriarty et al., 2009[Moriarty, N. W., Grosse-Kunstleve, R. W. & Adams, P. D. (2009). Acta Cryst. D65, 1074-1080.]) in the Phenix suite (Liebschner et al., 2019[Liebschner, D., Afonine, P. V., Baker, M. L., Bunkóczi, G., Chen, V. B., Croll, T. I., Hintze, B., Hung, L.-W., Jain, S., McCoy, A. J., Moriarty, N. W., Oeffner, R. D., Poon, B. K., Prisant, M. G., Read, R. J., Richardson, J. S., Richardson, D. C., Sammito, M. D., Sobolev, O. V., Stockwell, D. H., Terwilliger, T. C., Urzhumtsev, A. G., Videau, L. L., Williams, C. J. & Adams, P. D. (2019). Acta Cryst. D75, 861-877.]. On the basis of ligand correlation score, peak height and refinement to reasonable B factors, we justify the assignment of these positive peaks as TMO. The fact that TMO was not co-crystallized and the inherent quality of the data presumably explain why the modelling of some TMO molecules remains poor. 49 water molecules were added throughout the various stages of refinement (Supplementary Fig. S1). The final R factor and Rfree were 25.86% and 27.23%, respectively.

Crystal 2 belonged to the achiral monoclinic space group P21/n, with unit-cell parameters a = 48.88, b = 41.43, c = 71.96 Å, β = 97.47°. The structure was solved by MR using a double helix (helix 1.1) from the trigonal form described above. For MR, a multiple-copy search was carried out for two, three and four duplexes in the asymmetric unit. The MR solution, with a translation-function Z (TFZ) score of 8.0 and a log-likelihood gain (LLG) score of 202, revealed three independent D-DNA duplexes (named helices 2.1, 2.2 and 2.3) in the asymmetric unit. The refinement protocol was similar to that implemented for the trigonal form described above. After refinement of the DNA strands, positive peaks were observed above 7σ in the difference Fourier FoFc map at multiple locations (Supplementary Fig. S2). The two highest positive peaks were coordinated to the N7 atom of guanine G11 in helix 2.1 and in helix 2.2. Crystallographic studies reported that the cobalt(II) ion binds exclusively to N7 of guanine by coordination (Gao et al., 1993[Gao, Y., Sriram, M. & Wang, A. H.-J. (1993). Nucleic Acids Res. 21, 4093-4101.]). These two strong positive peaks were assumed to be, and were identified as, cobalt(II) ions from the crystallization reagent. The equivalent G11 in helix 2.3 was devoid of a strong positive peak. The average Co–N7 distance was 2.35 Å. The cobalt ions were refined with full occupancy, but due to the poor quality of the data the coordination sphere around them was not modelled. The remaining positive peaks were modelled as a chlorine ion and TMO. The density for the chlorine ion was broader than a water peak and was further away from the DNA atoms (3.5 Å from the N4 atom of cytosine). Modelling this peak with water gave residual density in the FoFc map. Fitting and refinement of TMO molecules was as described above. Eight water molecules were added in the final stages of refinement (Supplementary Fig. S2). The final R factor and Rfree were 27.95% and 31.82%, respectively. Further refinement did not lead to better convergence or to an improvement in the refinement statistics. One may note that the refinement R values were high compared with typical macromolecular structure depositions. This difference is at least in part inherent to centric data (Wilson, 1950[Wilson, A. J. C. (1950). Acta Cryst. 3, 397-398.]).

The coordinates and structure factors were deposited in the Protein Data Bank (Berman et al., 2000[Berman, H. M., Westbrook, J., Feng, Z., Gilliland, G., Bhat, T. N., Weissig, H., Shindyalov, I. N. & Bourne, P. E. (2000). Nucleic Acids Res. 28, 235-242.]) with accession codes 6gn2 and 6gn3. Raw X-ray data for the two crystal structures have been deposited with Zenodo (at https://doi.org/10.5281/zenodo.6397509 for PDB entry 6gn2 and https://doi.org/10.5281/zenodo.6397557 for PDB entry 6gn3). The program 3DNA (Lu & Olson, 2003[Lu, X.-J. & Olson, W. K. (2003). Nucleic Acids Res. 31, 5108-5121.]) was used to calculate helical parameters. Figures were prepared using PyMOL (DeLano, 2002[DeLano, W. L. (2002). PyMOL. https://www.pymol.org.]). Root-mean-square deviation (r.m.s.d.) values were determined using SUPERPOSE (Krissinel & Henrick, 2004[Krissinel, E. & Henrick, K. (2004). Acta Cryst. D60, 2256-2268.]).

2.3. Circular-dichroism (CD) spectroscopy

CD spectra were recorded on a Jasco J-810 spectrometer using quartz cells with a 2 mm optical path length. Scans were acquired at 293 K as an average of four measurements over the range 200–320 nm, with a scanning speed of 50 nm min−1, a response time of 0.5 s, a data pitch of 0.2 nm and a bandwidth of 2 nm. 100 µM DNA solutions were prepared in 2.0 M ammonium sulfate, 0.1 M Tris buffer pH 8.5 (crystallization condition 1), 1.8 M ammonium sulfate, 0.1 M MES buffer pH 6.5, 0.01 M cobalt(II) chloride (crystallization condition 2) and 1.0 M barium chloride, 50 mM sodium cacodylate pH 6.9, 0.001 M spermine·4HCl (the crystallization condition for the Z-form; Gautham et al., 1999[Gautham, N., Karthe, P. & Krishnaswamy, S. (1999). Perspectives in Structural Biology, edited by M. Vijayan, N. Yathindra & A. Kolaskar, pp. 593-601. Hyderabad: Indian Academy of Sciences/Universities Press.]). Data for solutions of each crystallization condition were baseline-corrected for signal contributions due to buffer, salts and spermine·4HCl.

3. Results and discussion

Previously, crystallization of the enantiopure D-d(CCCGGG) sequence was reported by Gautham and coworkers using the hanging-drop method at a high salt concentration (1 M barium chloride plus 50 mM sodium cacodylate buffer pH 6.9, 1 mM spermine; Gautham et al., 1999[Gautham, N., Karthe, P. & Krishnaswamy, S. (1999). Perspectives in Structural Biology, edited by M. Vijayan, N. Yathindra & A. Kolaskar, pp. 593-601. Hyderabad: Indian Academy of Sciences/Universities Press.]). Despite the lack of multiple alternating pyrimidine–purine base steps, this enantiopure sequence was shown by Gautham and coworkers to form a Z-DNA duplex. We used similar conditions to attempt to crystallize the racemic mixture [i.e. D/L-d(CCCGGG)]; however, we were unsuccessful in obtaining crystals. We next turned to a sparse-matrix crystallization screening approach, which successfully identified two related conditions suitable for crystal growth. X-ray diffraction analyses of the two crystal forms indicated that the crystals belonged to the achiral space groups [R\bar 3] and P21/n, in line with previous findings describing the tendency of racemic DNA mixtures to form racemic crystals rather than conglomerates (Mandal et al., 2014[Mandal, P. K., Collie, G. W., Kauffmann, B. & Huc, I. (2014). Angew. Chem. Int. Ed. 53, 14424-14427.]). Both crystal structures were solved by molecular replacement, with the resulting refined electron-density maps being of sufficiently high quality to allow the chirality of the DNA strands to be unambiguously determined. Surprisingly, both crystal forms of D/L-d(CCCGGG)2 reported here revealed DNA structures adopting A-type duplex conformations (Fig. 2[link]). Analysis of the duplex geometry reveals parameters that are all in line with expectations for A-DNA, namely a rise per residue of ∼2.8–3.0 Å, a base-pair inclination of ∼9–13° and a preference for C3′-endo sugar puckers (Supplementary Table S2). To our knowledge, this is the first report of a racemic crystal structure of A-form DNA.

[Figure 2]
Figure 2
Asymmetric units of D/L-d(CCCGGG)2 in two racemic crystal forms: (a) [R\bar 3] and (b) P21/n. A-DNA helices are shown as cartoons. Trimethylamine N-oxide, cobalt ions (pink) and chlorine ions (green) are shown as spheres. Bound waters are omitted for clarity.

The adoption of A-form duplexes by the racemic form of this sequence was unexpected, as previous spectroscopic studies of the equivalent enantiopure sequence [i.e. D-d(CCCGGG)] revealed B-type duplex DNA, albeit with non­standard deoxyribose conformations and internucleotide stacking geometry (Wolk et al., 1989[Wolk, S., Thurmes, W. N., Ross, W. S., Hardin, C. C. & Tinoco, I. Jr (1989). Biochemistry, 28, 2452-2459.]). In addition, as stated above, previous crystallographic studies of the enantiopure d(CCCGGG) sequence revealed an unusual Z-form conformation with nonstandard Z-form geometry (Gautham et al., 1999[Gautham, N., Karthe, P. & Krishnaswamy, S. (1999). Perspectives in Structural Biology, edited by M. Vijayan, N. Yathindra & A. Kolaskar, pp. 593-601. Hyderabad: Indian Academy of Sciences/Universities Press.]). The two strands in this duplex had different backbone conformations: the phosphate groups of one strand had a zigzag arrangement, while those in the other strand formed a smooth continuous helix. A study prior to this had also revealed the enantiopure d(CCCGGG) sequence to adopt a conformation intermediate between A-DNA and B-DNA when bound to the antibiotic nogalamycin (Cruse et al., 1996[Cruse, W. B. T., Saludjian, P., Leroux, Y., Léger, G., Manouni, D. E. & Prangé, T. (1996). J. Biol. Chem. 271, 15558-15567.]). We measured CD spectra of D-d(CCCGGG) under the conditions that gave rise to the two A-form and the Z-form crystals (Supplementary Fig. S3). Under all conditions, the spectra had the characteristics of B-form DNA, with a negative band near 240 nm and a positive band above 260 nm. The absence of a strong negative band near 290 nm suggested no strong contribution from a Z-form, and the absence of a negative band below 220 nm indicated that the A-form was not prevalent (Kypr et al., 2009[Kypr, J., Kejnovská, I., Renčiuk, D. & Vorlíčková, M. (2009). Nucleic Acids Res. 37, 1713-1725.]). Thus, both the racemic A-form and enantiopure Z-form appear to reflect effects that occur only in the solid state, a phenomenon that has been proposed for other nucleic acid structures (Sheehan et al., 2019[Sheehan, J. H., Smith, J. A., Pallan, P. S., Lybrand, T. P. & Egli, M. (2019). Crystals, 9, 532.]). The racemic A-form duplex crystal structures that we report here therefore add to the array of conformations adopted by this DNA sequence, providing further evidence of the considerable structural polymorphism of this motif. Perhaps paradoxically, despite this apparent structural polymorphism, the five independent A-type duplexes present within these two crystal structures reveal a remarkable degree of structural similarity, as indicated by structural alignments (Fig. 3[link] and Supplementary Table S1), despite the differences in space group and crystal packing.

[Figure 3]
Figure 3
Superposition of all D-DNA molecules found in the two structures of D/L-d(CCCGGG)2. Helices 1.1 and 1.2 (space group [R\bar 3]) are coloured red and green, respectively. Helices 2.1, 2.2 and 2.3 (space group P21/n) are coloured blue, yellow and magenta, respectively. R.m.s.d.s are reported in Supplementary Table S1.

Analysis of the crystal-packing interactions reveals further surprising features, with the lattices of both crystal forms reported here showing both homochiral and heterochiral pseudo-continuous helices (Fig. 4[link], Supplementary Figs. S6 and S7), as well as linear water-filled channels (Supplementary Fig. S8). Heterochiral pseudo-continuous helices are helices in which D-DNA duplexes and L-DNA duplexes stack end-to-end in an alternating manner, with an inversion of the helix handedness at every helix–helix contact. Previous reports of racemic DNA sequences have revealed a predominance of homochiral pseudo-helices within the crystal lattices (Mandal et al., 2014[Mandal, P. K., Collie, G. W., Kauffmann, B. & Huc, I. (2014). Angew. Chem. Int. Ed. 53, 14424-14427.]; Supplementary Fig. S4), with few reports of heterochiral pseudo-helical packing (Doi et al., 1993[Doi, M., Inoue, M., Tomoo, K., Ishida, T., Ueda, Y., Akagi, M. & Urata, H. (1993). J. Am. Chem. Soc. 115, 10432-10433.]; Rypniewski et al., 2006[Rypniewski, W., Vallazza, M., Perbandt, M., Klussmann, S., DeLucas, L. J., Betzel, C. & Erdmann, V. A. (2006). Acta Cryst. D62, 659-664.]; Supplementary Fig. S5) and none for A-form DNA. Thus, the racemic D/L-d(CCCGGG) sequence studied here reveals packing interactions that have not previously been observed, suggesting that racemic mixtures of DNA sequences may generate assemblies and higher-order structures that are not accessible to equivalent enantiopure systems.

[Figure 4]
Figure 4
Packing of L- and D-form A-DNA duplexes into heterochiral pseudo­helices within the lattice of crystal form 1. An almost identical pattern is observed in crystal form 2.

As mentioned in Section 1[link], artificial chiral DNA systems have been used to engineer nano-architectures such as nanocontainers (Sun et al., 2014[Sun, W., Boulais, E., Hakobyan, Y., Wang, W. L., Guan, A., Bathe, M. & Yin, P. (2014). Science, 346, 1258361.]) and DNA–protein hybrid shapes (Praetorius & Dietz, 2017[Praetorius, F. & Dietz, H. (2017). Science, 355, eaam5488.]). Rationally engineered DNA crystals have also been produced using L-DNA in order to confer them with resistance towards nuclease degradation (Simmons et al., 2017[Simmons, C. R., Zhang, F., MacCulloch, T., Fahmi, N., Stephanopoulos, N., Liu, Y., Seeman, N. C. & Yan, H. (2017). J. Am. Chem. Soc. 139, 11254-11260.]). Speculation that racemic DNA may be of benefit to such endeavours may not be entirely fanciful.

4. Conclusions

Here, we have reported two racemic crystal structures of the DNA sequence d(CCCGGG) revealing A-type duplexes, in contrast to previous structural descriptions of the enantiopure form of this sequence. Why does the racemic mixture of this hexameric DNA sequence adopt a different conformation in the solid state compared with the enantiopure equivalent? We cannot currently answer this question, but it appears that neither reflects the prevalent species in solution. Crystal-packing forces are thus likely to be at the origin of the lattices that are observed. The conditions required to capture the Z-form of the racemic d(CCCGGG) mixture in the solid state have yet to be discovered. However, what can be concluded with some confidence is that a racemic crystallography approach has allowed us to identify a structural form of a DNA sequence that has not previously been observed for the enantiopure equivalent. Our results add to other established uses of L-DNA including, for example, as interacting molecules with proteins (An et al., 2020[An, J., Choi, J., Hwang, D., Park, J., Pemble, C. W., Duong, T. H. M., Kim, K.-R., Ahn, H., Chung, H. S. & Ahn, D.-R. (2020). Chem. Commun. 56, 2186-2189.]; Vater & Klussmann, 2015[Vater, A. & Klussmann, S. (2015). Drug Discov. Today, 20, 147-155.]). Considering the potential for such systems to generate novel solid-state self-assemblies and nanomaterials, we believe that further investigation may be warranted.

5. Related literature

The following reference is cited in the supporting information for this article: Altona & Sundaralingam (1972[Altona, C. & Sundaralingam, M. (1972). J. Am. Chem. Soc. 94, 8205-8212.]).

Supporting information


Footnotes

Present address: AstraZeneca, 310 Cambridge Science Park, Cambridge CB4 0WG, United Kingdom.

Acknowledgements

This work has benefited from the facilities and expertise of the Biophysical and Structural Chemistry platform at IECB, CNRS UMS 3033, INSERM US001, Université de Bordeaux. Open access funding enabled and organized by Projekt DEAL.

Funding information

The following funding is acknowledged: European Research Council (grant No. ERC-2012-AdG-320892).

References

First citationAfonine, P. V., Grosse-Kunstleve, R. W., Echols, N., Headd, J. J., Moriarty, N. W., Mustyakimov, M., Terwilliger, T. C., Urzhumtsev, A., Zwart, P. H. & Adams, P. D. (2012). Acta Cryst. D68, 352–367.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationAldaye, F. A., Palmer, A. L. & Sleiman, H. F. (2008). Science, 321, 1795–1799.  Web of Science CrossRef PubMed CAS Google Scholar
First citationAltona, C. & Sundaralingam, M. (1972). J. Am. Chem. Soc. 94, 8205–8212.  CrossRef CAS PubMed Web of Science Google Scholar
First citationAn, J., Choi, J., Hwang, D., Park, J., Pemble, C. W., Duong, T. H. M., Kim, K.-R., Ahn, H., Chung, H. S. & Ahn, D.-R. (2020). Chem. Commun. 56, 2186–2189.  Web of Science CrossRef CAS Google Scholar
First citationAndersen, E. S., Dong, M., Nielsen, M. M., Jahn, K., Subramani, R., Mamdouh, W., Golas, M. M., Sander, B., Stark, H., Oliveira, C. L. P., Pedersen, J. S., Birkedal, V., Besenbacher, F., Gothelf, K. V. & Kjems, J. (2009). Nature, 459, 73–76.  Web of Science CrossRef PubMed CAS Google Scholar
First citationBerman, H. M., Westbrook, J., Feng, Z., Gilliland, G., Bhat, T. N., Weissig, H., Shindyalov, I. N. & Bourne, P. E. (2000). Nucleic Acids Res. 28, 235–242.  Web of Science CrossRef PubMed CAS Google Scholar
First citationBrünger, A. T. (1992). Nature, 355, 472–475.  PubMed Web of Science Google Scholar
First citationCruse, W. B. T., Saludjian, P., Leroux, Y., Léger, G., Manouni, D. E. & Prangé, T. (1996). J. Biol. Chem. 271, 15558–15567.  CrossRef CAS PubMed Web of Science Google Scholar
First citationDeLano, W. L. (2002). PyMOL. https://www.pymol.orgGoogle Scholar
First citationDoi, M., Inoue, M., Tomoo, K., Ishida, T., Ueda, Y., Akagi, M. & Urata, H. (1993). J. Am. Chem. Soc. 115, 10432–10433.  CrossRef CAS Web of Science Google Scholar
First citationDrozdzal, P., Gilski, M. & Jaskolski, M. (2016). Acta Cryst. D72, 1203–1211.  Web of Science CrossRef IUCr Journals Google Scholar
First citationEmsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. (2010). Acta Cryst. D66, 486–501.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationGao, Y., Sriram, M. & Wang, A. H.-J. (1993). Nucleic Acids Res. 21, 4093–4101.  CrossRef CAS PubMed Web of Science Google Scholar
First citationGautham, N., Karthe, P. & Krishnaswamy, S. (1999). Perspectives in Structural Biology, edited by M. Vijayan, N. Yathindra & A. Kolaskar, pp. 593–601. Hyderabad: Indian Academy of Sciences/Universities Press.  Google Scholar
First citationHaran, T. E., Shakked, Z., Wang, A. H.-J. & Rich, A. (1987). J. Biomol. Struct. Dyn. 5, 199–217.  CrossRef CAS PubMed Web of Science Google Scholar
First citationJacques, J., Collet, A. & Wilen, S. (1994). Enantiomers, Racemates and Resolutions. Malabar: Krieger Publishing Co.  Google Scholar
First citationKabsch, W. (2010). Acta Cryst. D66, 125–132.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationKrissinel, E. & Henrick, K. (2004). Acta Cryst. D60, 2256–2268.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationKypr, J., Kejnovská, I., Renčiuk, D. & Vorlíčková, M. (2009). Nucleic Acids Res. 37, 1713–1725.  Web of Science CrossRef PubMed CAS Google Scholar
First citationLiebschner, D., Afonine, P. V., Baker, M. L., Bunkóczi, G., Chen, V. B., Croll, T. I., Hintze, B., Hung, L.-W., Jain, S., McCoy, A. J., Moriarty, N. W., Oeffner, R. D., Poon, B. K., Prisant, M. G., Read, R. J., Richardson, J. S., Richardson, D. C., Sammito, M. D., Sobolev, O. V., Stockwell, D. H., Terwilliger, T. C., Urzhumtsev, A. G., Videau, L. L., Williams, C. J. & Adams, P. D. (2019). Acta Cryst. D75, 861–877.  Web of Science CrossRef IUCr Journals Google Scholar
First citationLu, X.-J. & Olson, W. K. (2003). Nucleic Acids Res. 31, 5108–5121.  Web of Science CrossRef PubMed CAS Google Scholar
First citationMandal, P. K., Collie, G. W., Kauffmann, B. & Huc, I. (2014). Angew. Chem. Int. Ed. 53, 14424–14427.  Web of Science CrossRef CAS Google Scholar
First citationMandal, P. K., Collie, G. W., Srivastava, S. C., Kauffmann, B. & Huc, I. (2016). Nucleic Acids Res. 44, 5936–5943.  Web of Science CrossRef CAS PubMed Google Scholar
First citationMarsh, R. E. (1995). Acta Cryst. B51, 897–907.  CSD CrossRef CAS Web of Science IUCr Journals Google Scholar
First citationMatthews, B. W. (2009). Protein Sci. 18, 1135–1138.  Web of Science CrossRef PubMed CAS Google Scholar
First citationMcCoy, A. J., Grosse-Kunstleve, R. W., Adams, P. D., Winn, M. D., Storoni, L. C. & Read, R. J. (2007). J. Appl. Cryst. 40, 658–674.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationMcLaughlin, C. K., Hamblin, G. D. & Sleiman, H. F. (2011). Chem. Soc. Rev. 40, 5647–5656.  Web of Science CrossRef CAS PubMed Google Scholar
First citationMoriarty, N. W., Grosse-Kunstleve, R. W. & Adams, P. D. (2009). Acta Cryst. D65, 1074–1080.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationMueller-Dieckmann, C., Kauffmann, B. & Weiss, M. S. (2011). J. Appl. Cryst. 44, 433–436.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationNeidle, S. (2021). J. Biol. Chem. 296, 100553.  Web of Science CrossRef PubMed Google Scholar
First citationParkinson, G., Vojtechovsky, J., Clowney, L., Brünger, A. T. & Berman, H. M. (1996). Acta Cryst. D52, 57–64.  CrossRef CAS Web of Science IUCr Journals Google Scholar
First citationPraetorius, F. & Dietz, H. (2017). Science, 355, eaam5488.  Web of Science CrossRef PubMed Google Scholar
First citationRothemund, P. W. K. (2006). Nature, 440, 297–302.  Web of Science CrossRef PubMed CAS Google Scholar
First citationRypniewski, W., Vallazza, M., Perbandt, M., Klussmann, S., DeLucas, L. J., Betzel, C. & Erdmann, V. A. (2006). Acta Cryst. D62, 659–664.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationSeeman, N. C. (2003). Nature, 421, 427–431.  Web of Science CrossRef PubMed Google Scholar
First citationSeeman, N. C. (2010). Annu. Rev. Biochem. 79, 65–87.  Web of Science CrossRef CAS PubMed Google Scholar
First citationSheehan, J. H., Smith, J. A., Pallan, P. S., Lybrand, T. P. & Egli, M. (2019). Crystals, 9, 532.  Web of Science CrossRef Google Scholar
First citationSimmons, C. R., Zhang, F., MacCulloch, T., Fahmi, N., Stephanopoulos, N., Liu, Y., Seeman, N. C. & Yan, H. (2017). J. Am. Chem. Soc. 139, 11254–11260.  Web of Science CrossRef CAS PubMed Google Scholar
First citationSun, W., Boulais, E., Hakobyan, Y., Wang, W. L., Guan, A., Bathe, M. & Yin, P. (2014). Science, 346, 1258361.  Web of Science CrossRef PubMed Google Scholar
First citationTerwilliger, T. C., Adams, P. D., Moriarty, N. W. & Cohn, J. D. (2007). Acta Cryst. D63, 101–107.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationVater, A. & Klussmann, S. (2015). Drug Discov. Today, 20, 147–155.  Web of Science CrossRef CAS PubMed Google Scholar
First citationWilson, A. J. C. (1950). Acta Cryst. 3, 397–398.  CrossRef IUCr Journals Web of Science Google Scholar
First citationWinn, M. D., Ballard, C. C., Cowtan, K. D., Dodson, E. J., Emsley, P., Evans, P. R., Keegan, R. M., Krissinel, E. B., Leslie, A. G. W., McCoy, A., McNicholas, S. J., Murshudov, G. N., Pannu, N. S., Potterton, E. A., Powell, H. R., Read, R. J., Vagin, A. & Wilson, K. S. (2011). Acta Cryst. D67, 235–242.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationWolk, S., Thurmes, W. N., Ross, W. S., Hardin, C. C. & Tinoco, I. Jr (1989). Biochemistry, 28, 2452–2459.  CrossRef CAS PubMed Web of Science Google Scholar
First citationWukovitz, S. W. & Yeates, T. O. (1995). Nat. Struct. Mol. Biol. 2, 1062–1067.  CrossRef CAS Web of Science Google Scholar
First citationYeates, T. O. & Kent, S. B. H. (2011). Annu. Rev. Biophys. 41, 41–61.  Web of Science CrossRef Google Scholar

This is an open-access article distributed under the terms of the Creative Commons Attribution (CC-BY) Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original authors and source are cited.

Journal logoSTRUCTURAL
BIOLOGY
ISSN: 2059-7983
Follow Acta Cryst. D
Sign up for e-alerts
Follow Acta Cryst. on Twitter
Follow us on facebook
Sign up for RSS feeds