Da Silva Haroldo C, De Almeida Wagner B
Laboratório de Química Computacional e Modelagem Molecular (LQC-MM), Departamento de Química Inorgânica, Instituto de Química, Universidade Federal Fluminense (UFF) Outeiro de São João Batista s/n, Campus do Valonguinho, Centro 24020-141 Niterói RJ Brazil
Departamento de Físico-Química, Instituto de Química, Pavilhão Haroldo Lisboa da Cunha, Universidade do Estado do Rio de Janeiro (UERJ) Rua São Francisco Xavier, 524, Maracanã 20550-013 Rio de Janeiro RJ Brazil
RSC Adv. 2024 Jun 18;14(27):19619-19635. doi: 10.1039/d4ra03430a. eCollection 2024 Jun 12.
Conformational analyses of organic compounds in solution still represent a challenge to be overcome. The traditional methodology uses the relative energies of the conformations to decide which one is most likely to exist in the experimental sample. The goal of this work was to deepen the approach of conformational analysis of flavonoid rutin (a well-known antioxidant agent) in DMSO solution. The methodology we used in this paper involves expanding the sample configuration space to a total of 44 possible geometries, using Molecular Dynamics (MD) simulations, which accesses structures that would hardly be considered with our chemical perception, followed by DFT geometry optimizations using the B97X-D/6-31G(d,p) - PCM level of theory. Spectroscopic and thermodynamic analyses were done, by calculating the relative energies and nuclear magnetic resonance (H-NMR) chemical shifts, comparing the theoretical and experimental H-NMR spectra (DMSO- ) and evaluating Mean Absolute Error (MAE). The essence of this procedure lies in searching for patterns, like those found in traditional DNA tests common in healthcare. Here, the theoretical spectrum plays the role of the analyzed human sample, while the experimental spectrum acts as the reference standard. In solution, it is natural for the solute to dynamically alter its geometry, going through various conformations (simulated here by MD). However, our DFT/PCM results show that a structure named 32 with torsion angles and manually rotated by approx. 20° showed the best theoretical-experimental agreement of H-NMR spectra (in DMSO- ). Relative energies benchmarking involving 16 DFT functionals revealed that the B97X-D is very adequate for estimating energies of organic compounds with dispersion of charge (MAE < 1.0 kcal mol, using post-Hartree-Fock MP2 method as reference). To describe the stability of the conformations, calculations of Natural Bonding Orbitals (NBO) were made, aiming to reveal possible intramolecular hydrogen bonds that stabilize the structures. Since van der Waals (vdW) interactions are difficult to be identified by NBO donations, the Reduced Density Gradient (RDG) were calculated, which provides 2D plots and 3D surfaces that describe Non-Covalent Interactions (NCI). These data allowed us to analyze the effect of dispersion interactions on the relative stability of the rutin conformations. Our results strongly indicate that a combination of DFT (B97X-D)-PCM relative energies and NMR spectroscopic criterion is a more efficient strategy in conformational analysis of organic compounds in solution.
溶液中有机化合物的构象分析仍然是一个有待克服的挑战。传统方法利用构象的相对能量来确定实验样品中最可能存在的构象。这项工作的目标是深化对黄酮类芦丁（一种著名的抗氧化剂）在二甲基亚砜（DMSO）溶液中的构象分析方法。我们在本文中使用的方法包括，通过分子动力学（MD）模拟将样品构型空间扩展到总共44种可能的几何结构，该模拟能够获取我们基于化学认知很难考虑到的结构，随后使用B97X-D/6-31G(d,p) - PCM理论水平进行密度泛函理论（DFT）几何优化。通过计算相对能量和核磁共振（H-NMR）化学位移、比较理论和实验H-NMR谱（DMSO- ）以及评估平均绝对误差（MAE），进行了光谱和热力学分析。这个过程的本质在于寻找模式，就像在医疗保健中常见的传统DNA检测中发现的那些模式一样。在这里，理论谱扮演被分析人类样品的角色，而实验谱则作为参考标准。在溶液中，溶质自然会动态改变其几何结构，经历各种构象（在这里通过MD模拟）。然而，我们的DFT/PCM结果表明，一种名为32的结构，其扭转角 和 手动旋转约20°后，显示出H-NMR谱（在DMSO- 中）最佳的理论与实验一致性。涉及16种DFT泛函的相对能量基准测试表明，B97X-D对于估计具有电荷分散的有机化合物的能量非常合适（以Hartree-Fock后MP2方法为参考，MAE < 1.0 kcal mol）。为了描述构象的稳定性，进行了自然键轨道（NBO）计算，旨在揭示稳定结构的可能分子内氢键。由于范德华（vdW）相互作用很难通过NBO供体来识别，因此计算了约化密度梯度（RDG），它提供了描述非共价相互作用（NCI）的二维图和三维表面。这些数据使我们能够分析色散相互作用对芦丁构象相对稳定性的影响。我们的结果有力地表明，DFT（B97X-D）-PCM相对能量和NMR光谱标准的结合是溶液中有机化合物构象分析中一种更有效的策略。
