Abstract
Basic side chains play crucial roles in protein-DNA interactions. In this study, using NMR spectroscopy, we investigated the dynamics of Arg and Lys side chains of the fruit fly Antennapedia homeodomain in the free state and in the complex with target DNA. We measured 15N relaxation for Arg and Lys side chains at two magnetic fields, from which generalized order parameters for the cationic groups were determined. Mobility of the R5 side chain, which makes hydrogen bonds with a thymine base in the DNA minor groove, was greatly dampened. Several Lys and Arg side chains that form intermolecular ion pairs with DNA phosphates were found to retain high mobility with the order parameter being < 0.6 in the DNA-bound state. Interestingly, some of the interfacial cationic groups in the complex were more mobile than in the free protein. The retained or enhanced mobility of the Arg and Lys side chains in the complex should mitigate the overall loss of conformational entropy in the protein-DNA association and allow dynamic molecular recognition.
Graphical Abstract

In protein-DNA complexes, arginine (Arg) and lysine (Lys) side chains interact with DNA phosphates to form intermolecular ion pairs, serving as crucial constituents of the molecular interfaces.1,2 The electrostatic interactions of basic side chains with DNA are important not only as a major driving force for protein-DNA association but also as a shape-readout mechanism to recognize particular DNA.2 Direct readout through hydrogen bonds between basic side chains and DNA bases are also common.1,2 Thus, basic side chains of proteins play crucial roles in association with and recognition of DNA.
Recent studies have shown that conformational mobility makes significant entropic contribution to the thermodynamics of macromolecular association.3,4 To understand DNA recognition by proteins from a thermodynamic viewpoint, the dynamic properties of basic side chains should be delineated. Some NMR methods have been developed for dynamics investigations of basic side chains.5–13 Using these methods, we recently studied changes in the mobility of the basic side chains of the Egr-1 zinc-finger protein upon association with its target DNA.14 Interestingly, the basic side chains that form ion pairs with the DNA phosphates were found to retain high mobility despite the simultaneous presence of hydrogen bonds and strong short-range electrostatic interactions. This trend was particularly remarkable for the Lys side chains. In contrast, the Arg side chains that directly interact with the DNA bases were found to lose substantial mobility upon binding. To examine whether or not these are general characteristics of basic side chains, other systems should also be studied.
For this purpose, we study the dynamics of the basic side chains of the fruit fly Antennapedia (Antp) homeodomain in the free and DNA-bound states. The Antp homeodomain is comprised of 60 residues, 18 of which are basic side chains (12 Arg and 6 Lys residues) (Figure 1). This protein recognizes DNA sequence, TAATGG, and binds to it with a dissociation constant on the order of 10−8 M at physiological ionic strength.15 Five crystal structures are available for the Antp homeodomain-DNA complexes.15–17 NMR structures for the DNA complex of a 68-residue construct of the Antp homeodomain are also available.18,19 These structures show that R3, R5, R28, R31, R43, K46, R53, K55, and K57 interact with DNA. We previously studied dynamics of Lys side chains of the same Antp homeodomain-DNA complex.20 In the current work, using NMR spectroscopy, we investigate dynamics of the Arg and Lys side chains in the free and DNA-bound states of the Antp homeodomain and analyze changes in side-chain mobility upon the molecular association.
Figure 1.

The basic side chains in the Antp homeodomain-DNA complex. (a) Sequences of the Antp homeodomain C39S mutant and the 15-bp DNA duplex used in this study. (b) A crystal structure of the Antp homeodomain-DNA complex (PDB 4XID)15.
Internal motions of Arg side-chain Nε-Hε groups
We compared the internal motions of the Arg guanidinium groups of the Antp homeodomain in the free and DNA-bound states under the same buffer conditions and temperature. Figure 2a shows the 1H-15N heteronuclear in-phase single quantum coherence (HISQC) spectra6 recorded for the Arg side-chain Nε-Hε moieties of the Antp homeodomain in the free and DNA-bound states at 25°C. Both samples were dissolved in a buffer of 20 mM potassium succinate (pH 5.8), 100 mM KCl, and 0.4 mM NaF (as a preservative). The Arg side chains exhibited well-isolated signals in the 1H-15N heteronuclear correlation spectra for both states. For the Arg side-chain Nε nuclei, we measured 15N longitudinal (R1) and transverse (R2) relaxation rates and heteronuclear NOE at the 1H frequency of 750 MHz. Heteronuclear NOE and 15N R1 data were collected at 600 MHz as well. These relaxation data are shown in Tables S1 and S2 in the Supporting Information (SI). Using these data, we determined the generalized order parameters (S2)21 for the Arg Nε-Hε bonds of the Antp homeodomain in the free and DNA-bound states, as described.14 The S2 parameter satisfies the inequalities of 0 ≤ S2 ≤ 1, and represents a measure of the degree of spatial restriction of internal motion.21 Changes in order parameters upon molecular association are related to changes in conformational entropy.4 The molecular rotational diffusion parameters (Table S3 in the SI) were determined from backbone 15N relaxation rates R1 and R2.14 The data of the generalized order parameters (S2) determined for Arg Nε-Hε bonds in the free protein and in the complex are shown in Figure 2b as well as Table S4 in the SI.
Figure 2.
NMR investigations on internal motions of Arg side-chain Nε-Hε moieties of the Antp homeodomain in the free and DNA-bound states. (a) Overlaid 1H-15N HISQC spectra6 recorded for the Arg side chains in the free protein (blue) and in the complex with 15-bp DNA (red). The region indicated by a dotted box is expanded in the inset. (b) NMR-derived order parameters S2 for Arg Nε-Hε bonds in the free protein (blue) and in the complex (red).
Change in mobility of Arg side chains upon protein-DNA association
The S2 data allow us to assess the change in dynamics of the Arg side chains upon DNA-binding. This is straightforward because these Arg order parameters were determined under identical conditions for the free and DNA-bound states. Among the Arg side chains, R5 showed the largest increase in order parameter S2, indicating that the mobility of this side chain becomes severely hampered upon the protein-DNA association. This side chain is deeply buried in DNA minor groove (see Figure 1b), and makes hydrogen bonds with the O2 atom of the first thymine base of the recognition sequence TAATGG. The large increase in the S2 parameter for the R5 Nε-Hε bond can be attributed to immobilization arising from hydrogen-bonding to the thymine base in the narrow space of DNA minor groove. R5 is extremely well conserved among the homeodomain proteins, and plays an important role in DNA shape recognition by these proteins.22,23 The side chain of R3, another residue important for DNA shape recognition,22 also exhibited a larger increase in the S2 parameter. In some crystal structures, the side chains of R3, R28, R31, R43, and R53 form contact ion pairs (CIPs) with DNA phosphates. Despite the short-range electrostatic interactions with DNA, R3, R28, and R43 side chains retain substantial mobility, exhibiting S2 < 0.5. Upon molecular association, the order parameters S2 of R3, R43, and R53 Nε-Hε bonds were found to increase by ~0.2, suggesting that the ion pair formation with DNA phosphate restricts the internal motions of these Arg side chains. Interestingly, R28 and R31 did not exhibit such immobilization. The S2 value for the R31 side chain was found to decrease by 0.12, indicating that this side chain becomes more mobile upon protein-DNA association. The R31 side chain in the free state exhibited a relatively large S2 value (= 0.77), probably due to hydrogen-bonding to E42 Oε as seen in some crystal structures of free homeodomains (e.g., PDB 1ENH and 1P7I). This interaction is broken due to attraction of the R31 guanidino cation to a DNA phosphate in the crystal structures of the Antp homeodomain-DNA complexes. Competition between E42 carboxylate and DNA phosphate for the R31 guanidino cation might mobilize the R31 side chain in the complex. The retained or enhanced mobility of Arg side chains upon protein-DNA association would be entropically favorable for binding.
Internal motions of Lys side-chain NH3+ groups
We previously determined the order parameters of the Lys side-chain NH3+ moieties in the Antp homeodomain-DNA complexes at pH 5.8 and 25°C.15,17,20 Under these conditions, the side-chain NH3+ moieties of the interfacial Lys residues, K46, K55, K57, and K58, in the complexes exhibited signals in 1H-15N HISQC spectra. In our previous studies, intermolecular CIPs were confirmed for K46, K55, and K57 by observation of the hydrogen-bond scalar couplings (h3JNP) between Lys side-chain 15N and DNA phosphate 31P nuclei.15,20,24 Although it would be ideal to compare the internal motions of the free and DNA-bound states under the identical conditions, unfortunately, the Antp homeodomain in the free state showed no observable signals from Lys NH3+ groups at pH 5.8 and 25°C due to rapid hydrogen exchange. However, the free Antp homeodomain at pH 4.5 and 2°C exhibited 1H-15N HISQC signals from all Lys NH3+ moieties (Figure 3a). Backbone 1H-15N TROSY spectra show that the protein remains folded under these conditions (Figure S1). As previously described,8,13 we measured 15N relaxation at the 1H-frequencies of 800 and 600 MHz, and determined order parameters of Lys side-chain NH3+ moieties of the free protein under these conditions. Because the complex was found to partially aggregate at the low pH and low temperature (data not shown), we conducted the Lys side-chain 15N relaxation analysis for the complex at pH 5.8 and 15°C, in addition to our previous analysis at 25°C.20 The 15N relaxation parameters measured for the free protein and the complex in the current study are shown in Tables S5 and S6 in the SI. Figure 3b shows the order parameters (S2axis) determined for Lys side-chain NH3+ moieties in the free protein and in the complex (values are shown in Table S7 in the SI).
Figure 3.

NMR investigations on internal motions of Lys side-chain NH3+ moieties of the Antp homeodomain in the free and DNA-bound states. (a) Lys NH3+-selective 1H-15N HISQC spectra6 recorded for the free protein (blue) and the complex with 15-bp DNA (red). (b) NMR-derived order parameters S2axis for Lys side chains in the free protein (blue) and in the complex (red). (c) Dynamic equilibrium between the CIP and SIP states.
Comparison of internal motions of Lys side chains upon protein-DNA association
The K46, K57, and K58 side chains in the DNA-bound state exhibited significantly larger order parameters than in the free state. To some degree, these differences may reflect differences in temperature (2°C vs. 15°C) and in pH (4.5 vs. 5.8). However, our previous study for a different system showed that temperature dependence of order parameters for Lys NH3+ moieties are very weak.25 In fact, in the current study, the order parameters measured for the Lys NH3+ moieties in the complex at 15°C were in good agreement with the previous values measured at 25°C. The difference in pH (i.e., 5.8 vs. 4.5) might affect the order parameters of Lys NH3+ moieties if they are interacting with Glu or Asp carboxylates (typical pKa ~4.0). Because the distance between the K55 Nζ and E59 Oε atoms is as short as 4.1 Å in the crystal structures, the difference in pH may affect the order parameter for the K55 NH3+ moiety.
Different motional characteristics of Arg and Lys side chains
The order parameters S2axis of Lys NH3+ groups are defined for the C3 symmetry axis, which corresponds to the Cε-Nζ bond. Although the same number of rotatable bonds (i.e., four covalent bonds) are involved between Lys Cα and Cε atoms and between Arg Cα and Nε atoms, the ranges of the measured order parameters were remarkably different for Lys Cε-Nζ and Arg Nε-Hε bonds. Compared to Lys Cε-Nζ bonds, Arg Nε-Hε bonds exhibited a wider range of order parameters (see Figures 2b vs. 3b) as well as larger changes in order parameters upon DNA binding (see Figure S2a in the SI). The same characteristic differences between Arg and Lys side chains were also seen in our previous study on the Egr-1 zinc-finger protein in the free and DNA-bound states.14 Judging from our current and previous studies, the mobility of Arg side chains seems to be more sensitive to the surrounding environment, compared to Lys side chains.
Retained mobility of basic side chains forming ion pairs with DNA phosphates
Our previous study on the Egr-1 zinc-finger protein in the free and DNA-bound states showed that the basic side chains forming ion pairs with DNA phosphates tend to retain substantial mobility.14 This trend was confirmed in the current study on the Antp homeodomain. The side chains R3, R28, R43, K46, K55, and K57 in the DNA-bound state exhibited order parameters S2 between 0.19 and 0.55, indicating substantial mobility, although they form intermolecular ion pairs with DNA. Although our current study suggests that the overall entropic change is negative for the Arg/Lys cationic moieties (see Figure S2b in the SI), their high mobility in the complex should mitigate the entropic loss. The high mobility of the interfacial basic side chains may also help proteins adaptively recognize DNA, even while the DNA undergoes significant conformational fluctuations such as BI–BII transitions.
Despite the simultaneous presence of the hydrogen bonds and strong short-range electrostatic interactions, why are these basic side chains so mobile? As we previously discussed based on theoretical and computational investigations,14,20,26 this high mobility can be attributed to the dynamic equilibria between the CIP and solvent-separated ion-pair (SIP) states (Figure 3c). The CIP and SIP states can be distinguished in terms of N…O distance between cationic and anionic groups.20 For Lys side-chain NH3+–DNA phosphate ion pairs, the free energy differences between CIP and SIP states were estimated to be 0.8–1.6 kcal/mol and the energy barriers for CIP→SIP transitions were estimated to be 2.2–3.2 kcal/mol.20 With the relatively small energy difference and barrier, transitions between the CIP and SIP states rapidly occur, making the ion pairs highly dynamic in a ps–ns timescale. Interestingly, a computational study showed that the free energy difference and barriers between CIP and SIP states of an Arg+–Glu− pair are significantly larger than those of a Lys+–Glu; pair,27 which seems to be qualitatively consistent with our experimental observation that Lys side chains tends to be more mobile than Arg side chains. It should also be noted that Lys and Arg side chains differ not only in number of hydrogen bonds, but also in desolvation energy, as discussed by Rohs et al.28 These differences may account for the higher mobility of Lys side chains.
Supplementary Material
Acknowledgments
Funding Sources
This work was supported by Grant R01-GM105931 from the National Institutes of Health (NIH). D.N. is a recipient of the Houston Area Molecular Biophysics Program fellowship (supported by NIH Grant T32-GM008280). No competing financial interests have been declared.
We thank Drs. Levani Zandarashvili, Chuanying Chen, and Montgomery Pettitt for useful discussion.
ABBREVIATIONS
- Antp
Antennapedia
- CIP
contact ion pair
- HISQC
heteronuclear in-phase single-quantum coherence
- SIP
solvent-separated ion pair
Footnotes
Author Contributions
The manuscript was written through contributions of all authors, who have given approval to the final version.
ASSOCIATED CONTENT
The Supporting Information is available free of charge on the ACS Publications website: the 15N relaxation and S2 data for the Arg and Lys side chains; molecular rotational diffusion parameters; and Materials and Methods.
References
- 1.Luscombe NM, Laskowski RA, Thornton JM. Nucleic Acids Res. 2001;29:2860–2874. doi: 10.1093/nar/29.13.2860. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Rohs R, Jin X, West SM, Joshi R, Honig B, Mann RS. Annu Rev Biochem. 2010;79:233–269. doi: 10.1146/annurev-biochem-060408-091030. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Caro JA, Harpole KW, Kasinath V, Lim J, Granja J, Valentine KG, Sharp KA, Wand AJ. Proc Natl Acad Sci U S A. 2017;114:6563–6568. doi: 10.1073/pnas.1621154114. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Wand AJ. Curr Opin Struct Biol. 2013;23:75–81. doi: 10.1016/j.sbi.2012.11.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Berglund H, Baumann H, Knapp S, Ladenstein R, Härd T. J Am Chem Soc. 1995;117:12883–12884. [Google Scholar]
- 6.Iwahara J, Jung YS, Clore GM. J Am Chem Soc. 2007;129:2971–2980. doi: 10.1021/ja0683436. [DOI] [PubMed] [Google Scholar]
- 7.Trbovic N, Cho JH, Abel R, Friesner RA, Rance M, Palmer AG., 3rd J Am Chem Soc. 2009;131:615–622. doi: 10.1021/ja806475k. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Esadze A, Li DW, Wang T, Brüschweiler R, Iwahara J. J Am Chem Soc. 2011;133:909–919. doi: 10.1021/ja107847d. [DOI] [PubMed] [Google Scholar]
- 9.Zandarashvili L, Li DW, Wang T, Bruschweiler R, Iwahara J. J Am Chem Soc. 2011;133:9192–9195. doi: 10.1021/ja202219n. [DOI] [PubMed] [Google Scholar]
- 10.Werbeck ND, Kirkpatrick J, Hansen DF. Angew Chem Int Ed Engl. 2013;52:3145–3147. doi: 10.1002/anie.201209385. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Anderson KM, Esadze A, Manoharan M, Bruschweiler R, Gorenstein DG, Iwahara J. J Am Chem Soc. 2013;135:3613–3619. doi: 10.1021/ja312314b. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Zandarashvili L, Esadze A, Iwahara J. Adv Protein Chem Struct Biol. 2013;93:37–80. doi: 10.1016/B978-0-12-416596-0.00002-6. [DOI] [PubMed] [Google Scholar]
- 13.Nguyen D, Lokesh GLR, Volk DE, Iwahara J. Molecules. 2017;22:1355. doi: 10.3390/molecules22081355. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Esadze A, Chen C, Zandarashvili L, Roy S, Pettitt BM, Iwahara J. Nucleic Acids Res. 2016;44:6961–6970. doi: 10.1093/nar/gkw531. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Zandarashvili L, Nguyen D, Anderson KM, White MA, Gorenstein DG, Iwahara J. Biophys J. 2015;109:1026–1037. doi: 10.1016/j.bpj.2015.07.032. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Fraenkel E, Rould MA, Chambers KA, Pabo CO. J Mol Biol. 1998;284:351–361. doi: 10.1006/jmbi.1998.2147. [DOI] [PubMed] [Google Scholar]
- 17.Nguyen D, Zandarashvili L, White MA, Iwahara J. Chembiochem. 2016;17:1636–1642. doi: 10.1002/cbic.201600265. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Billeter M, Qian YQ, Otting G, Muller M, Gehring W, Wüthrich K. J Mol Biol. 1993;234:1084–1093. doi: 10.1006/jmbi.1993.1661. [DOI] [PubMed] [Google Scholar]
- 19.Fernandez C, Szyperski T, Billeter M, Ono A, Iwai H, Kainosho M, Wüthrich K. J Mol Biol. 1999;292:609–617. doi: 10.1006/jmbi.1999.2987. [DOI] [PubMed] [Google Scholar]
- 20.Chen C, Esadze A, Zandarashvili L, Nguyen D, Montgomery Pettitt B, Iwahara J. J Phys Chem Lett. 2015;6:2733–2737. doi: 10.1021/acs.jpclett.5b01134. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Lipari G, Szabo A. J Am Chem Soc. 1982;104:4546–4559. [Google Scholar]
- 22.Abe N, Dror I, Yang L, Slattery M, Zhou T, Bussemaker HJ, Rohs R, Mann RS. Cell. 2015;161:307–318. doi: 10.1016/j.cell.2015.02.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Dror I, Zhou T, Mandel-Gutfreund Y, Rohs R. Nucleic Acids Res. 2014;42:430–441. doi: 10.1093/nar/gkt862. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Zandarashvili L, Esadze A, Kemme CA, Chattopadhyay A, Nguyen D, Iwahara J. J Phys Chem Lett. 2016;7:820–824. doi: 10.1021/acs.jpclett.6b00019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Zandarashvili L, Iwahara J. Biochemistry. 2015;54:538–545. doi: 10.1021/bi5012749. [DOI] [PubMed] [Google Scholar]
- 26.Iwahara J, Esadze A, Zandarashvili L. Biomolecules. 2015;5:2435–2463. doi: 10.3390/biom5042435. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Masunov A, Lazaridis T. J Am Chem Soc. 2003;125:1722–1730. doi: 10.1021/ja025521w. [DOI] [PubMed] [Google Scholar]
- 28.Rohs R, West SM, Sosinsky A, Liu P, Mann RS, Honig B. Nature. 2009;461:1248–1253. doi: 10.1038/nature08473. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.

