Commentary and Perspective (Invited)
What is Aromaphilicity?
2023 年 20 巻 1 号 論文ID: e200002
詳細
2023 年 20 巻 1 号 論文ID: e200002
Proteins and peptides have the ability to interact with various substances such as biological molecules and artificial objects. In principle, these interactions are attributed to the interplays of amino acid residues and peptide bonds with target substances and are often described in physical terms, including electrostatic interaction, hydrogen bond, and van der Waals interaction. However, in some practical cases, conceptual scales and indices for describing the nature of amino acids, such as the hydrophilicity scale and the hydropathy index, are more useful for understanding the interactions. In recent years, I have investigated the affinity of proteins and peptides for aromatic carbon materials, such as carbon nanotubes (CNTs) and graphene, and realized that this affinity is barely described by conventional scales and indices. After speculating whether a more suitable index for describing the affinity of amino acids for aromatic carbon materials is available in such a situation, I recognized the need for a new concept that describes such an affinity. Upon establishing this concept, I named the affinity for aromatic carbon material surfaces “aromaphilicity,” meaning an aromatic-loving nature. In this Commentary and Perspective, I summarized my recent works with my collaborators regarding physical interactions between amino acids (or amino acid residues) and aromatic carbon material surfaces and introduced a new index—aromaphilicity index—of amino acids. The aromaphilicity index is unique and distinct from conventional indices for amino acids, offering prospective applications as a universal index for describing the properties of amino acids.
The interactions of amino acids with aromatic carbon materials (such as CNTs and graphene) have been investigated by experiments and computer simulations. My collaborators and I have previously demonstrated the high affinity of aromatic amino acids for CNTs in an experimental study using a CNT-immobilized liquid chromatographic column. This affinity was determined by differences in retention time of amino acids between the CNT-immobilized column and CNT-free column (Figure 1) [1]. The affinity was in the order Trp>Tyr>Phe. By contrast, other hydrophobic amino acids such as Val, Leu, and Ile exhibited insignificant affinity for CNTs. These results suggest that aromatic amino acids have high aromaphilicity in the order Trp>Tyr>Phe, whereas the other hydrophobic amino acids have low or insignificant aromaphilicity.
(a) Schematic illustration of liquid chromatography of amino acids on CNT-immobilized column. (b) Relationship of retention factors (log k) of amino acids between CNT-immobilized (
The interactions of charged amino acids with CNT surfaces were investigated by an experiment on the dispersibility of CNTs [2]. Here, oligopeptides of Arg (R) and Lys (K), including R20, R10, R5, K20, K10, and K5, were used as dispersants for the CNTs. The dispersibility of CNTs in the presence of each oligopeptide was in the order R20>R10>K20>R5>K10≈K5≈0. The results of this experiment suggest that Arg has higher aromaphilicity than Lys.
Despite such experimental results, obtaining a set of relative aromaphilicity values for all proteinogenic amino acids using the same method is technically difficult. For example, the chromatographic method described above is inapplicable to amino acids with a charged side chain because the electrostatic interactions of the charged side chains with charged ligands (amino groups) on the matrixes affect the chromatographic retention of the amino acids. Additionally, the experiment regarding CNT dispersibility described above is inapplicable to peptides with uncharged side chains because CNTs coated with those peptides eventually form precipitants owing to the lack of electrostatic repulsion.
A possible approach for determining the relative aromaphilicity of every amino acid is by quantifying its affinity for aromatic carbonaceous surfaces with the help of computer simulations. Various methods such as quantum chemical and molecular dynamics (MD) simulations have been conducted to examine the interactions between amino acids and aromatic carbon materials, including CNTs and graphene. My collaborator and I have recently performed MD simulations to quantify the binding free energy of every proteinogenic amino acid to the aromatic carbonaceous surfaces in aqueous systems [3,4]. A set of the relative values of binding free energies to graphene was then defined as the aromaphilicity index of the amino acids. Consistent with the experimental results (Figure 1), Trp has the highest aromaphilicity index (Figure 2 and Table 1). Notably, Arg is in third place and exhibited a higher aromaphilicity index than Phe. Importantly, this index was correlated with experimental data on the affinity of 16 amino acids for CNTs in literature [5] with a correlation coefficient R2=0.789. Therefore, the aromaphilicity index is unique as a descriptor of the amino acids. The aromaphilicity index is correlated with the surface area of the planar moiety in amino acid side chains with an R2 value of 0.95, which suggests that aromaphilicity is primarily governed by van der Waals and π–π interactions. The aromaphilicity index is useful for visualizing the aromaphilicity profiles of arbitrary protein surfaces obtained from the Protein Data Bank, where the aromaphilicity hot spots should correspond to the feasible binding sites of the proteins to aromatic carbonaceous surfaces (Figure 2). In addition, aromaphilicity decreases with increasing curvature of the aromatic carbonaceous surfaces.
Aromaphilicity index of proteinogenic amino acids and aromaphilicity profile of hen egg white lysozyme [3].
| Amino acid | Aromaphilicity index (cm3 mol–1) | Molar residue refractivity (a.u.) | Polarizability | Polarizability parameter |
|---|---|---|---|---|
| Trp | 1.000 | 55.24 | 157.8 | 0.409 |
| Tyr | 0.850 | 44.34 | ― | 0.298 |
| Arg | 0.750 | 39.47 | 115.6 | 0.291 |
| Phe | 0.575 | 42.21 | 122.9 | 0.290 |
| His | 0.450 | 34.62 | 102.6 | 0.230 |
| Met | 0.325 | 34.45 | 102.1 | 0.221 |
| Gln | 0.300 | 30.37 | 91.2 | 0.180 |
| Asn | 0.200 | 26.09 | 79.8 | 0.134 |
| Ile | 0.200 | 31.87 | 95.2 | 0.186 |
| Cys | 0.200 | 48.58 | ― | 0.128 |
| Val | 0.150 | 26.73 | 81.5 | 0.140 |
| Ser | 0.125 | 19.16 | 61.2 | 0.062 |
| Leu | 0.125 | 31.59 | 94.5 | 0.186 |
| Pro | 0.125 | 23.74 | 73.5 | ― |
| Lys | 0.100 | 34.10 | 101.2 | 0.219 |
| Thr | 0.075 | 23.82 | 73.7 | 0.108 |
| Glu | 0.050 | 30.07 | 90.4 | 0.151 |
| Ala | 0.025 | 17.15 | 55.9 | 0.046 |
| Gly | 0.000 | 12.81 | 44.3 | 0.000 |
| Asp | –0.025 | 26.06 | ― | 0.105 |
In principle, Trp, Tyr, and Phe have a high aromaphilicity index owing to the aromatic groups in their side chains [6]. However, Arg has the third highest aromaphilicity index, which seems counterintuitive since the amino acid is generally categorized to be hydrophilic and non-aromatic.
The aromaphilicity index is correlated with the molar residue refractivity values (Table 1) [7]. In addition, the index is correlated with the polarizability values (Table 1), which is reasonable because polarizability is a function of refractivity. Importantly, refractivity and polarizability are generally related to the strength of van der Waals forces [8,9]. Accordingly, aromaphilicity is primarily governed by van der Waals interactions between amino acids and aromatic carbonaceous surfaces, that is, the aromaphilicity of amino acids does not originate from their “aromaticity.” In addition to van der Waals interactions, π–π and cation–π interactions with aromatic carbonaceous surfaces can contribute to aromaphilicity. Notably, certain amino acids do not follow the correlation; for example, the aromaphilicity index of cysteine is lower than that expected from the refractivity, which may be attributed to the steric hindrance of the sulfur atom due to its large radius.
Based on the mechanism described above, the high aromaphilicity of Arg is attributed to the characteristic group in the side chain—the guanidinium group. The guanidinium group possesses a planar structure with high refractivity and polarizability, and thus potentially has stable face-to-face interactions with aromatic carbonaceous surfaces through van der Waals and π–π interactions (and possibly hydrophobic interactions [13]). In applications, the affinity of Arg to the aromatic carbonaceous surfaces has been utilized in solubilizing various aromatic compounds in aqueous solutions (Table 2), in eluting proteins from liquid chromatography columns with aromatic ligands [14], and further in suppressing protein adsorption onto CNT surfaces [15].
| Compound | Transfer free energy (kJ/mol) |
|---|---|
| Caffeic acid | –5 |
| Alkyl gallatesb | –3 |
| Coumarin | –2 |
| Nucleobasesc | –1––2 |
| Tyrosine | –1 |
| Phenylalanine | –0.5 |
a The data were obtained from ref [14]. The compounds with a larger negative value are more effectively solubilized by arginine in aqueous solution. b Alkyl gallates include methyl, ethyl, propyl, and butyl gallates. c Nucleobases include adenine, guanine, cytosine, thymine, and uracil at pH 7.4.
The concept of aromaphilicity will also be applied to other compounds. My collaborators and I have recently demonstrated that nicotinamide adenine dinucleotide (NAD) has a high affinity for CNTs (Figure 3) [16]. Importantly, NAD contains aromatic groups in its chemical structure, which are potentially aromaphilic. Similar interaction patterns have been observed for DNA and small organic molecules [17,18]. A feasible approach to predict the aromaphilicity of arbitrary compounds will include a bottom–up approach for determining the aromaphilicity index at the functional group level. Incidentally, NADH (an oxidized form of NAD) can chemically react with CNTs in aqueous solutions, leading to the transformation of NADH to NAD+ (a reduced form of NAD) (Figure 3) [16].
Conformations of coenzyme (NADH) molecules on a CNT obtained by MD simulation and chemical reaction scheme demonstrated by an experiment. NAD+ was produced from NADH through electron transfer to CNTs. Reproduced from ref [16].
The aromaphilicity index is expected to be useful for understanding various phenomena occurring in proteins, peptides, and amino acids. Aromaphilicity was recently suggested to be related to the mechanism of liquid–liquid phase separation of proteins and peptides [19–21], which suggests the potential application of the index for describing and understanding residue–residue interactions in proteins and peptides. A perspective for future directions includes the design of amino acid sequences in proteins and peptides as biopharmaceuticals on the basis of the aromaphilicity index.
The evidence data generated and/or analyzed during the current study are available from the corresponding author on reasonable request.
I thank K. Shiraki and his laboratory members for their collaboration on experiments determining the aromaphilicity of amino acids using CNT-immobilized column chromatography. I thank T. Kameda for collaborating on MD simulations determining the aromaphilicity index and the affinity of NAD for CNTs.
