Role of aromatic amino acids in protein–nucleic acid recognition

Abstract

Statistical analysis of structures from the PBD has been used to examine the role that the aromatic amino acids play in protein-nucleic acid recognition. In protein-DNA complexes, the residues Phe and His are found to bind selectively to the DNA chain--Phe to A and T, and His to T and G. The preferred binding modes are identified, and the interactions involving Phe are shown to be important in the transcription process. In protein-RNA complexes, Phe is found to occur far less often and is instead replaced by Trp, which binds selectively to C and G, offering a possible mechanism for differentiation between the two nucleic acids. SASA analysis of the two sets of complexes suggests that all of the aromatic amino acids are more heavily involved in binding than would be expected on the balance of probability. Phe and Tyr occur approximately equal in both sets of data, whereas the proportions of His and Trp vary considerably, supporting the idea that these residues may be involved in differentiating between the two nucleic acids.

Topology of RNA–protein nucleobase–amino acid π–π interactions and comparison to analogous DNA–protein π–π contacts

“…Due to the importance of nucleic acid-protein interactions, a number of studies have analyzed experimental Xray crystal structures to understand the types of atomic level interactions between DNA Luscombe and Thornton 2002;Gromiha et al 2004aGromiha et al ,b, 2005Mao et al 2004;Lejeune et al 2005;Prabakaran et al 2006;Baker and Grant 2007;Sathyapriya et al 2008;Wilson et al 2014Wilson et al , 2015 or RNA (Allers and Shamoo 2001;Jones et al 2001;Treger and Westhof 2001;Cheng et al 2003;Jeong et al 2003;Lejeune et al 2005;Morozova et al 2006;Baker and Grant 2007;Ellis et al 2007;Bahadur et al 2008;Barik et al 2015) nucleotides and amino acids that govern nucleic acid recognition and binding. These studies have determined that nucleic acid-protein contacts span hydrogen bonding (direct or water mediated), ionic (salt-bridge or phosphate backbone), van der Waals, and hydrophobic interactions.…”

“…These studies have determined that nucleic acid-protein contacts span hydrogen bonding (direct or water mediated), ionic (salt-bridge or phosphate backbone), van der Waals, and hydrophobic interactions. Importantly, key differences between RNA and DNA recognition have become apparent (Allers and Shamoo 2001;Jones et al 2001;Lejeune et al 2005;Morozova et al 2006;Baker and Grant 2007;Bahadur et al 2008;Barik et al 2015). Indeed, the 2 ′ -hydroxyl group plays a role in RNA recognition (Bahadur et al 2008;Barik et al 2015), with estimates that this moiety is involved in approximately onequarter of all RNA-protein hydrogen bonds (Treger and Westhof 2001).…”

“…Previous analyses of nucleic acid-protein crystal structures reveal small distances between the nucleobases and π-containing amino acids (Jones et al 2001;Luscombe et al 2001;Treger and Westhof 2001;Lejeune et al 2005;Morozova et al 2006;Baker and Grant 2007;Ellis et al 2007;Wilson et al 2014;Barik et al 2015;Wilson et al 2015). Indeed, despite the involvement of the DNA nucleobases in stable stacking interactions within the double helix, an abundance of nucleobase-amino acid π-interactions (π-contacts) have been identified in DNA-protein complexes Lejeune et al 2005;Baker and Grant 2007;Wilson et al 2014;Wilson et al 2015).…”

“…Indeed, despite the involvement of the DNA nucleobases in stable stacking interactions within the double helix, an abundance of nucleobase-amino acid π-interactions (π-contacts) have been identified in DNA-protein complexes Lejeune et al 2005;Baker and Grant 2007;Wilson et al 2014;Wilson et al 2015). Since most RNA nucleobases are not stacked within duplexes, nucleobase-amino acid π-interactions can occur without requiring unfavorable conformational changes and with the added benefit of sequestering the bases from solvent.…”

“…Alternatively, analysis of van der Waals interactions in 89 RNA-protein complexes revealed that the most favored nucleotide-amino acid pairings include U:Y, A:F, and G:tryptophan (W) (Ellis et al 2007). Although three studies specifically considered nucleobase-amino acid π-π stacking interactions and consistently concluded that all aromatic (cyclic) amino acids are involved in this type of nucleobase contact, variations occur in the amino acid and nucleobase reported to most commonly participate in stacking contacts (Morozova et al 2006;Baker and Grant 2007;Barik et al 2015). Besides the aromatic amino acids, Morozova et al (2006) and Barik et al (2015) both report significant involvement of R in nucleobase π-π stacking interactions.…”