Influence of the conformations of αA-crystallin peptides on the isomerization rates of aspartic acid residues

https://doi.org/10.1016/j.bbapap.2020.140480Get rights and content

Highlights

  • Three-dimensional structural features of αA-crystallin peptides were predicted.

  • Peptide conformation was related to the rate of Asp-residue stereoinversion.

  • Solvent exposure was an important factor in the Asp-residue stereoinversion.

  • Distant residues may also affect the rate of Asp-residue stereoinversion.

Abstract

The isomerization rate of aspartic acid (Asp) residue is known to be affected by the three-dimensional structures of peptides and proteins. Although the isomerized Asp residues were experimentally observed, structural features which affect the isomerization cannot be elucidated sufficiently because of protein denaturation and aggregation. In this study, molecular dynamics (MD) simulations were conducted on three αA-crystallin peptides (T6, T10, and T18), each containing a single Asp residue with different isomerization rate (T18 > T6 > T10) to clarify the structural factors of Asp isomerization tendency. For MD trajectories, distances between side-chain carboxyl carbon of Asp and main-chain amide nitrogen of (n + 1) residue (Cγ–N distances), root mean square fluctuations (RMSFs), and polar surface areas for main-chain amide nitrogen of (n + 1) residues (PSAN) were calculated, because these structural features are considered to relate to the formations of cyclic imide intermediates. RMSFs and PSAN are indexes of peptide backbone flexibilities and solvent exposure of the amide nitrogen, respectively. The average Cγ–N distances of T10 was longer than those of the other two peptides. In addition, the peptide containing Asp residue with a higher isomerization rate showed higher flexibility of the peptide backbone around the Asp residue. PSAN for amide nitrogen in T18 were much larger than those of other two peptides. The computational results suggest that Asp-residue isomerization rates are affected by these factors.

Introduction

It has been believed that amino acids in living organisms were only l-enantiomers, not only l- but also d-amino acid residues have been detected from the biological peptides and proteins in aging tissues, such as crystallin in the lens [[1], [2], [3], [4], [5]], amyloid β in the brain [[6], [7], [8], [9]], and elastin in the aorta, ligament, skin, and intervertebral disc [[10], [11], [12], [13], [14]]. Since most peptides and proteins in living organisms are biosynthesized from l-amino acids, d-amino acid residues are considered to be formed by stereoinversion of l-amino acid residues [[15], [16], [17], [18]]. It is known that aspartic acid (Asp) residues are prone to stereoinversion as compared to other amino acid residues. Although the Asp decoded from the gene is only of l-α-form, structural isomerization from α- to β-form has been reported in aging tissues as well as stereoinversion from l- to d-form. That is, three types of isomerized Asp residues can be formed from l-α-Asp residue, i.e., l-β-, d-α-, and d-β-Asp residues. The formation of these isomerized Asp residues is suspected to dramatically change the three-dimensional (3D) structure of the peptides and proteins [[19], [20], [21]], and to cause several age-related diseases, e.g., cataract [[1], [2], [3], [4], [5]], Alzheimer's diseases [[6], [7], [8], [9]], and arteriosclerosis [10,13].

Asp-residue isomerization is believed to non-enzymatically proceed via the five-membered ring succinimide (Suc) intermediate (Scheme 1) [[15], [16], [17], [18]]. Suc intermediates are formed by the nucleophilic attack of the main-chain amide nitrogen of the C-terminal side adjacent residue, i.e., (n + 1) residue, to the Asp-residue side-chain carboxyl carbon with the release of one water molecule. Suc intermediates are prone to stereoinversion. Radkiewick et al. presumed that the Suc intermediate is likely to undergo stereoinversion compared to ordinary amino acid residues since the carbanion formed by the abstraction of the α-proton of Suc is stabilized by the resonance effect [22]. Some Suc intermediates in peptides and proteins convert from l- to d-form by keto-enol tautomerization. Subsequently, four types of Asp isomers (l-α-, l-β-, d-α-, and d-β-Asp) are formed by ring-opening reactions of l-/d-Suc intermediate. To date, the reaction mechanisms for Asp-residue isomerization has been proposed using quantum chemical calculation of simplified model molecules [[23], [24], [25], [26], [27], [28], [29], [30], [31], [32]]. Although Asp-residue isomerization proceeds non-enzymatically, the computational study by Catak et al. suggested that the activation barrier of Asp-residue isomerization in the absence of any catalysts was very high, and that some catalysts were required to proceed the reaction [23]. In addition, recently, it has been computationally shown that the miscellaneous molecules in the living organisms can catalyze Asp-residue stereoinversion, and that the calculated activation energies were approximately 20–30 kcal mol−1, which reproduced the experimental results [[33], [34], [35], [36]].

Until now, isomerized Asp residues have been found only in specific sites of limited protein. This fact indicates that not all Asp residues are equally isomerized, and that the particular environment around the Asp residue affects the rate of Asp-residue isomerization. Fujii et al. presumed that the smaller size of the (n + 1) residue, e.g., glycine (Gly), alanine (Ala), and serine (Ser), is preferred for the Asp isomerization; because the main-chain amide nitrogen has to nucleophilically attack the side-chain carboxyl carbon of Asp residue in order to Suc intermediate formation [15,16]. However, even though the (n + 1) residue of Asp140 in αB-crystallin is Gly which is the smallest amino acid, isomerization of Asp140 is hardly observed [2]. Additionally, it is reported that the Asp-residue isomerization in peptides proceed rapidly compared to that in proteins [36]. Thus, the reactivity for isomerization of Asp-residue is considered to be affected by the 3D structure of peptides and proteins.

Previously, Fujii et al. performed a kinetic study using three types of αA-crystallin peptides (T6, T10, and T18), i.e., model peptides with the same amino acid sequence as αA-crystallin, and showed that the rate of Asp-residue isomerization depended on the amino acid sequence [33]. Table 1 shows the amino acid sequences of αA-crystallin peptides and the rate of Asp-residue isomerization. However, for these peptides, no details are elucidated about the factors that affect the rate of Asp-residue isomerization. To investigate the relationship between the rates of Asp-residue isomerization and the conformations of αA-crystallin peptides, structural biological analyses of peptides are indispensable; however, structural biological experiments of them are difficult because of the protein denaturation and aggregation. In the present study, we investigated the influence of the conformation of αA-crystallin peptides on the rate of Asp-residue isomerization using MD simulations. Although the quantum chemical calculation is the only method which can accurately reproduce chemical reactions, it can only be conducted on peptides constituted of at most a few residues due to computational cost. In addition, since the αA-crystallin peptides will not be able to maintain a rigid conformation due to the lack of sufficient intramolecular interactions, the αA-crystallin peptides are considered to be able to form a wide variety of conformations. Therefore, the quantum chemical calculation is not suitable for analyze in this study. On the other hand, MD simulations can obtain not only static 3D structures but also conformational ensembles. To date, we have shown that MD simulations are useful for predicting the structural tendencies of short peptides [[37], [38], [39]]. Moreover, very recently, we actually simulated three types of elastin peptides, and succeeded in predicting the relationship between the environment around Asp residues in these elastin peptides and the rate of Asp-residue isomerization [40]. In the present study, we analyzed the conformational ensembles of three-types αA-crystallin peptides obtained by MD simulations, and predicted the factors which affect the rate of Asp-residue isomerization.

Section snippets

MD simulations

To eliminate initial bias, linear chains generated by tleap module of AmberTools were used for initial structures of αA-crystallin peptides for calculations. First, energy minimizations were performed for the initial structures, and the maximum number of minimization cycles was set to 2000. After energy minimizations, 200-ns MD simulations at temperature of 300 K were conducted at time step 1 fs, and the snapshots were extracted from every 500 fs. To control the temperature, Langevin dynamics

Cγ–N distance

The calculated average values and the standard deviations of Cγ–N distances for T6, T10, and T18 are shown in Table 2, and Fig. 2 shows box plots of the Cγ–N distances for each peptide. In the box plots, the lower limit of the inlier was defined as the first quartile minus 1.5 times the interquartile range, and the upper limit inlier was defined as the third quartile plus 1.5 times the interquartile range. Additionally, Fig. 3 illustrates the snapshots of T6, T10, and T18 with the shortest Cγ–N

Conclusion

In this study, we performed the MD simulations for three types of αA-crystallin peptides (T6, T10, and T18), and obtained the conformational ensembles. The conformational features of αA-crystallin peptides were evaluated by analyzing these ensembles. Although each of three αA-crystallin peptides contains a single Asp residue, the experimental rates of Asp-residue isomerization are significantly different (as shown in Table 1, the isomerization rates are T18 > T6 > T10).

In the T10 with the

Credit author statement

Tomoki Nakayoshi: Conceptualization, Methodology, Formal Analysis, Writing–Original Draft, Funding Acquisition. Koichi Kato: Methodology, Formal Analysis. Eiji Kurimoto: Validation, Investigation. Akifumi Oda: Conceptualization, Writing–Review & Editing, Project administration, Funding acquisition.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgement

This work was supported by grants-in-aid for scientific research [15H01064], [17K08257], and [19J23595] from the Japan Society for the Promotion of Science.

References (46)

  • J.-P. Ryckaert et al.

    Numerical integration of the cartesian equations of motion of a system with constraints: molecular dynamics of n-alkanes

    J. Comput. Phys.

    (1977)
  • N. Fujii et al.

    Simultaneous stereoinversion and isomerization at specific aspartic acid residues in αA-crystallin from human lens

    J. Biochem.

    (1994)
  • Y. Kaji et al.

    Localization of d-β-aspartic acid-containing proteins in human eyes

    Invest. Opthalmol. Vis. Sci.

    (2007)
  • M.Y.S. Hooi et al.

    Racemisation and human cataract. d-Ser, d-Asp/Asn and d-Thr are higher in the lifelong proteins of cataract lenses than in age-matched normal lenses

    AGE

    (2011)
  • M.Y.S. Hooi et al.

    Age-dependent racemization of serine residues in a human chaperone protein

    Protein Sci.

    (2013)
  • R. Shapira et al.

    Neuritic plaque amyloid in Alzheimer’s disease is highly racemized

    J. Neurochem.

    (1988)
  • T. Kubo et al.

    In vivo conversion of racemized β-amyloid ([d-Ser26]Aβ1–40) to truncated and toxic fragments ([d-Ser26]Aβ25–35/40) and fragment presence in the brains of Alzheimer's patients

    J. Neurosci. Res.

    (2002)
  • T. Kubo et al.

    β-Amyloid racemized at the Ser26 residue in the brains of patients with Alzheimer disease: implications in the pathogenesis of Alzheimer disease

    J. Neuropathol. Exp. Neurol.

    (2003)
  • S. Ritz-Timme et al.

    Age estimation based on aspartic acid racemization in elastin from the yellow ligaments

    Int. J. Legal Med.

    (2003)
  • S. Ritz-Timme et al.

    Aspartic acid racemization: evidence for marked longevity of elastin in human skin

    Brit. J. Derm.

    (2003)
  • R.C. Dobberstein et al.

    Aspartic acid racemisation in purified elastin from arteries as basis for age estimation

    Int. J. Legal Med.

    (2010)
  • N. Fujii et al.

    Homochirality and life

    Chem. Rec.

    (2004)
  • N. Fujii

    D-amino acids in living higher organisms

    Orig. Life Evol. Biosph.

    (2002)
  • View full text