Chemical decarboxylation kinetics and identification of amino acid standards by benchtop NMR spectroscopy

In this paper, we examined the competence of amino acids as standards for instrumental biochemical analysis. The chosen amino acids were first dissolved in various aquatic solutions and then measured in a benchtop NMR spectrometer, which is not a common choice in such analytical investigations. Analysis by mass spectrometry was used in addition. As part of these investigations, we examined and determined the stability of the amino acids ornithine, glutamic acid, alanine, glycine, proline, pyroglutamic acid, phenylalanine and trans-4-hydroxy-D-proline under critical basic and acidic pH conditions and under various other conditions. We observed that not all solutions of the amino acid standards remain stable under the given conditions and a chemical transformation takes place. Given our findings by mass spectroscopy, additional kinetic measurements were carried out with the benchtop NMR spectrometer. We discovered that pyroglutamic acid becomes unstable under basic conditions and decarboxylates to pyrrolidone.


Introduction
Amino acids describe the vast, complex and diverse group of molecules that include both an amino and a carboxyl group (Lu and Freeland 2016). They are the building blocks of proteins and can be considered the most important structures in the creation and proper function of life itself. The mystery of how amino acids were created in the primitive and extreme conditions of early earth moves scientist to this day (Miller 1953). It was discovered that not all prebiotically synthesized amino acids ended up in the standard alphabet and that not all structures of the standard alphabet could have been prebiotically synthesized. The standard alphabet consists of 20 amino acids abbreviated into a three letter or one letter code, which is intended to reduce the size of the data files needed to describe the sequence of the amino acids within a protein. Some amino acids, like proline, do not even meet the minimal requirements of an amino acid, because of its cyclic structure. The carbon here binds back to the 'backbone' nitrogen, generating a secondary amino group instead of a primary amine (Lu and Freeland 2016). Therefore, the analysis of those structures, free amino acids and proteins plays an important role in today's chemistry and biochemistry as well as in various other fields. Most suitable 1 3 for such structure and functional group analysis is NMR spectroscopy.
Chemical shifts in nuclear magnetic resonance (NMR) measure the locally induced magnetic field on a nucleus that is sensitive to the chemical environment (Friebolin 2011). They can be used as fingerprints for detailing the chemical composition of condensed matter. The analysis of chemical shifts plays an increasingly important role in structural peptide and protein analyses (Wishart et al. 1995). A systematic study of the NMR chemical shifts of free amino acids is, therefore, highly desirable and could lay the ground for studies of protein shifts in complex biological environments. The position of the resonance is the first step in the NMR-based structure determination of proteins. Because of limiting protein concentrations or their low stabilities, a new strategy centered around the evaluating of the NMR spectra of individual amino acids has been established (Bellstedt et al. 2013). It was found that, when calculating the NMR shifts, the effect of the environment on the chemical shifts of these amino acids corresponds to the influence of the intermolecular hydrogen bonds or electrical dipole field effects (Yoon et al. 2004). It was also found that in aqueous solutions the pH influences the phenomenon of chemical exchange. The chemical shifts are sensible toward specific intermolecular effects between the labile protons of amino acids and the ones of water (Khallouk et al. 2005).
The NMR spectroscopy with the benchtop device can be used for such analytic work and offers many advantages. Among other things, it has low acquisition and operating costs, is easy to use, has a high spectral resolution, no cryogens needed, is easy to use via Windows PC and is mobile and compact. Following these advantages, the device also allows smaller organizations the entry into this analysis method and could be used as a fully automated analyzer. However, negative aspects, such as the spectral width, must also be taken into account. The lower field strength of the systems also often requires the use of model-based approaches for spectra evaluation (Tables 1, 2).
This motivated the present work on the behavior and stability of amino acids and amino acid standards in benchtop NMR spectroscopy. We discovered that such solutions are potentially unstable under various conditions (Airaudo et al. 1987), and therefore, we have conducted certain investigations. By using an internal standard, the concentrations of amino acid standard solutions which in solid state are difficult to weigh were determined by quantitative NMR spectroscopy (Yamazaki et al. 2017). For these experiments, we choose six naturally occurring L-amino acids and the non-proteinogenic ornithine and 1 3 pyroglutamic acid. To remove any strong solvent signal in the 1 H-NMR spectra, the samples were measured with water suppression. Complementary mass spectroscopy is carried out for the solutions, since this analytical method is very useful for amino acids (Johnson 2011). In mass spectrometric experiments, we observed unexpected degradation processes and decided to investigate them further. In this paper, we present the investigation of amino acids and amino acid standards regarding their stability in extreme conditions.

Material and methods
Chemicals and reagents were purchased from Carl Roth and were used without further purification. The amino acids examined and used in this study were alanine, glutamic acid, glycine, ornithine, phenylalanine, proline, pyroglutamic acid, trans-4-hydroxy-D-proline. The structural formulas of the amino acids are shown below ( Fig. 1; Tables 3, 4).
No experiments on human participants or animals were carried out during the implementation of these experiments, nor were any human or animal rights violated (Tables 5, 6).

Sample preparation with water and buffer
The first set of amino acids was dissolved in 10 ml H 2 O by shaking and the second set in 5 ml 50 mM ammonium bicarbonate buffer (NH 4 HCO 3 ). An exception was the solution with trans-4-hydroxy-D-proline which was dissolved in 8 ml H 2 O and 3 ml ammonium bicarbonate buffer, due to material shortage of this amino acid. However, the concentration of all amino acids in the solutions was about 0.075 M. The buffer solution was prepared in deuterium oxide (D 2 O) with 0.4005 g NH 4 HCO 3 and 89.4 mg 3-(trimethylsilyl)-2,2,3,3tetradeuteropropionic acid sodium salt (TMSP-d 4 ). Before starting the NMR measurement, the pH of every water and buffer sample was determined with the potentiometric titrator 916 Ti-Touch which is used for routine analysis.

Sample preparation with NaOD
The amino acids were dissolved in 2 ml of a 4 M NaOD solution by shaking. From each sample, 0.6 ml was transferred to standard NMR tube and measured in the benchtop NMR spectrometer (Tables 7, 8).

Sample preparation with DCl
The amino acids were dissolved in 2 ml of 4 M DCl solution containing 21.8 mg TMSP-d 4 . From each sample, 0.6 ml was transferred to a standard NMR tube and measured in the benchtop NMR spectrometer. Additionally, 0.6 ml of the glutamic acid sample was filled into a sealed 5-mm standard NMR tube and heated for 30 min at 80 °C in an oil bath before the measurement. During the heating time, the temperature was controlled four times, to make sure that it stays constant.

Sample preparation for kinetics
Five similar TMSP-d 4 containing D 2 O and NaOD solutions were prepared for the kinetics measurement.
For a 4 M NaOD solution in D 2 O, 45.2 mg of TMSP-d 4 was dissolved in 5.25 ml. For the next three solutions, 1 M, 0.08 M and 0.16 M, 43.1 mg of TMSP-d 4 were dissolved in 5 ml. The 0.5 M NaOD and TMSP-d 4 solution was made by mixing the 1 M solution with pure D2O in 1:1 ratio.
For each kinetics measurement, pyroglutamic acid was dissolved in 1 ml of each of the basic solutions. 0.6 ml of these samples was transferred to a standard NMR tube and used for the reaction monitoring measurement in the NMR spectrometer.

General procedure for NMR measurements
After the preparations, about 0.6 ml of these samples was transferred into sealed 5-mm standard NMR tubes. The 1 H NMR spectra were then recorded on a Magritek Spinsolve 60 Carbon Benchtop NMR spectrometer. The Chemical shifts are reported in ppm relative to solvent signal (HDO: δH = 4.79 ppm). The spectra were recorded at a temperature of 25 °C with the 1D Proton standard protocol using 16 scans with a 90-degree excitation pulse (7 microseconds) covering a spectral range from 46 to −37 ppm (Magritek Spinsolve standard NMR conditions). 32 k data points are acquired with an acquisition time of 6.5 s and a repetition time (recycle delay) of 15 s. In advance to every long-time experiment (like kinetic measurements) and at the start of each working day, a so-called powershim has been performed within Spinsolve, where all available shims get optimized. The final scan records a spectrum and analyzes the linewidths at 50% and 0.55% of peak maximum, the Signal-to-Noise ratio (SNR), peak amplitude (signal) and the mean noise level. The settings of the 1D Presat protocol, which suppresses the large solvent peaks of water, exactly match the settings of the standard protocol. The frequency search was set to automatic, the SAT power was −60 dB, the SAT period was 3 s, and 2 dummy scans were carried out. Data processing has been performed with MestReNova version 14.1.0 including zeroth and first-order phase correction and baseline correction with a Bernstein polynomial fit of third order. It can be observed that the peaks by 5.60 ppm decrease over time till they completely disappear. Also new peaks by 4.50 ppm appear and increase in size over time

Mass spectroscopy
Electrospray ionization mass spectra (ESI) were performed on Sciex API QTRAP Mass Spectrometer (AB Sciex LLC, Framingham, MA, USA). The mass spectrometer was operated in the positive ion mode with an electrospray voltage of 5000 V at 200 °C, curtain gas at 25 psi, collision gas at 6 psi, nebulizing gas at 25 psi and auxiliary gas at 25 psi. All quadrupoles were working at unit resolution.
To carry out the Q1 mass spectra, all 4 M DCl and 4 M NaOD samples were diluted with MS-grade methanol.

NMR spectroscopy and the stability of the amino acids
If we look at the stability of the amino acids, we observed that seven of the eight amino acids tested were stable at given experimental conditions. The first H-NMR spectra of the simple water solution were poor to evaluate because the water signal was wide and covered the signals of the amino acids. The amino acid buffer solutions then were measured with water suppression. No signal and peaks of the amino acid were covered or distorted by the buffer. However, many amino acids show peaks near the 4.79 ppm mark of the wide water signal, which hindered the quantification of the integrals. All the amino acids also remained stable in the buffer solution for a week.
The spectra of the NaOD solution were easy to evaluate. The water peak was almost absent and shifted to around 5.30 ppm in most samples. The reason for the shift of the water peak in different samples is their pH-dependency. Some peaks of the samples also shifted or changed their distance to one another, for example, the peaks of trans-4-hydroxy-D-proline have split up considerably. After a week, the samples were measured a second time and it was discovered that the spectra of amino acid (7) had changed its appearance, indicating a chemical transformation. All the other amino acids remained stable in the basic solution.  Increase of pyrrolidone integral in 4 M NaOD soluƟon over Ɵme The spectra of the DCl solution were very similar to the measurements carried out in the NaOD solution. The water peak shifted again, but this time to around 6.00 ppm. Some of the amino acid peaks also shifted compared to the buffer spectra. As a second measurement after a week showed, all of the eight amino acids remained stable in acidic solution.
In order to calculate the concentration of the amino acids from the measured and processed spectra, we used the formula shown below. From the buffer solutions and DCl solutions runs, four spectra of sample were used to determinate the arithmetic mean of all the specific integrals of each amino acid. The mean value of the integrals is then divided by the number of protons and multiplied by the concentration of the standard.
As it can be seen from the following tables, the concentrations determined by the NMR measurement tend to have higher values than the calculated concentrations. When determining the concentrations using the NMR measurements, we have potential sources of error which could cause the results to deviate from the concentrations calculated from the gravimetric sample preparation. These potential sources of error include phase correction and baseline correction, which were carried out when evaluating the NMR spectra. bounds of integration. Those potential sources of error can influence the integrals, which in turn influence the determined concentrations.

Stability of glutamic acid in 4 M DCl solution
For future investigation of the stability, the glutamic acid with 4 M DCl solution was heated for 30 Minutes at 80 °C. The NMR spectrum result has shown no changes. So, it can be concluded that no change in the structure of the glutamic acid was caused by brief heating. In addition, the glutamic acid sample was also stable for a week in the refrigerator and for several hours at room temperature.

The kinetics of pyroglutamic acid
With the information from mass spectroscopy and additional research (Dose 1957;Derek et al. 1984;Claes et al. 2019;Steffen et al. 1991;Snider and Wolfenden 2000;Thanassi and Aminomalonic acid. 1970), we came to the conclusion that the pyroglutamic acid, after a few days in NaOD solution, is decarboxylating. Here the reaction equation that fits our hypothesis is shown in deuterated water (Fig. 2). The 4 M kinetic measurement was carried out at 30-min intervals and ran for four days, but the reaction ended on the second day.
The 1 M kinetic measurement was carried out at 30-min intervals and ran over a full eight days (Fig. 3).
The 0.5 M kinetics measurement was carried out at twohour intervals and ran for eleven days, after which the reaction was not fully completed. Even after multiple days of run time (>8 days), the kinetics measurements with the 0.08 M and 0.16 M NaOD solutions did not show any signs of a chemical transformation.
The overall process of the decarboxylation is a one molecular disintegration reaction. In this case, we assume the observed reaction is of (pseudo) first order, because the rate of reaction linearly depends only on the concentration of the pyroglutamate itself. This would align with the overall formation of pyrrolidone over time illustrated in the following graphs (Figs. 4, 5, 6, 7):

Mass spectroscopy
Since we observed changes in the NMR spectra of the pyroglutamic acid, we carried out mass spectroscopy for all amino acids in NaOD and DCl after the kinetic measurements. The results of mass spectroscopy support the above observation that decarboxylation or chemical transformation occurs in the case of pyroglutamic acid. Another supporting fact is that the masses of the other amino acid samples in NaOD and DCl showed no change in their mass in the mass spectra.

Conclusion
Almost every amino acid that was used in the experiment was stable at the critical pH conditions, with the exception of pyroglutamic acid. This shows that the amino acids have great potential as standards in the field of biochemical analysis particularly in NMR spectroscopy. They stay stable over time, in very acidic and very basic conditions, at high temperatures and have good recognizable peaks.
Keeping in mind that benchtop NMR spectroscopy cannot be the method of choice for everyday quantification and identification of amino acids due to more suitable methods such as HPLC-MS or GC-MS, the method can still be of great use for solving more complex analytical tasks. It is also a good option for quick measurements because it can be varied out easily.
Aknowledgements We like to thank the ReAching program of the faculty MLS for support.
Funding Open Access funding enabled and organized by Projekt DEAL.

Declarations
Conflicts of interest All authors certify that they have no affiliations with or involvement in any organization or entity with any financial interest or non-financial interest in the subject matter or materials discussed in this manuscript. All authors declare that they have no conflict of interest.

Data availability
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Code availability Code availability not applicable.
Ethical approval This article does not contain any studies with human participants or animals performed by any of the authors.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.