Capillary Zone Electrophoresis with Diode-Array Detection as a Tool to Monitor Variations in the Concentration of Organic Acids from the Tricarboxylic Acid Cycle in Human Urine

Abstract

The tricarboxylic acid (TCA) cycle is responsible for the generation of energy in the human body and may inform about the energy status in different physiological and pathological conditions. It was developed a method to quantify the TCA cycle intermediates (α-ketoglutaric, citric, fumaric, lactic, malic, and succinic acids) by capillary zone electrophoresis with diode array detector (CZE-DAD) in indirect mode. The background electrolyte consisted of 24 mmol L^–1 2,6-pyridine carboxylic acid, 76 mmol L^–1 β-alanine, and 4 mmol L^–1 hexadecyltrimethylammonium hydroxide (pH 4). The method was validated and presented a good linearity (R² < 0.9922) and adequate limits of detection (0.24-1.84 mg L^–1) and quantifcation (0.74-5.57 mg L^–1). The method was successfully applied to samples of individuals with obesity before being submitted to bariatric surgery and one year after the surgery. The citric, lactic, and malic acids were detected in the analyzed samples. To evaluate the changes in concentrations of the TCA intermediates, principal component analysis was carried out, and there was no signifcant difference between the two sets of samples. Altogether, the data obtained after applying this new method suggest that CZE-DAD could be used to detect and quantify the TCA cycle metabolites in human urine.

Keywords:
tricarboxylic acid cycle; organic acids; capillary zone electrophoresis; indirect method

Introduction

The tricarboxylic acid (TCA) cycle intermediates present the potential of being biomarkers of different health conditions. The TCA cycle is a complex series of chemical reactions that involve the conversion of citrate to isocitrate, α-ketoglutarate, succinyl-CoA, succinate, fumarate, malate, and oxaloacetate (Figure 1). The TCA cycle takes place within the mitochondria, which are organelles responsible for energy production in eukaryotic cells.² Mitochondria are involved in a range of cellular processes, including cellular respiration, calcium signaling, and apoptosis. Dysfunction in mitochondria has been linked to a range of diseases, including metabolic disorders, neurodegenerative diseases, and cancer.³

Figure 1
Overview of the tricarboxylic acid cycle. The compound structures of the organic acids involved in the process are presented (adapted from reference 1).

The TCA cycle is the main pathway for the aerobic oxidation of fatty acids, carbohydrates, and amino acids intermediates that are metabolized to acetyl-CoA before entering the cycle. The cycle is central to energy metabolism, being connected simultaneously with several other metabolisms, therefore maintaining its activity is essential for homeostasis. The intermediates of the cycle are also substrates necessary for reactions outside of the mitochondria, and when they are removed from the cycle, they must be replaced so the cycle can keep its continued function in a process called anaplerosis. The cycle can be stimulated or slowed down depending on the body’s energy demands.^{4 5 6} When the energy demands are increased or when aerobic energy production is reduced, lactic acid is produced by glucose oxidation under anaerobic conditions.⁷,⁸ Because of the importance of the cycle in the central metabolism of the mitochondria, the intermediates show potential as biomarkers for various health conditions, providing insights into cellular metabolism and helping to identify novel targets for therapeutic intervention in metabolic diseases.

The concentration of metabolites can be measured in body fluids, such as blood and urine, or even in cells and specific tissues. Variations in the concentration of the TCA cycle intermediates have been associated with the risk of type

II diabetes,⁹ a higher risk of arrhythmia¹⁰ and tumorigenesis.¹¹

Analysis in human urine indicated a relation between the

TCA cycle metabolites and prostate cancer,¹² urolithiasis¹³,¹⁴ and the progression of kidney diseases.^{15 16 17}

Continuous efforts have been made to determine these intermediates using different analytical techniques. The first choice to analyze organic acids is usually gas-chromatography (GC).¹⁸,¹⁹ However, analyses by GC require a derivatization step, which is time-consuming and increases the probability of analyte loss during the process.²⁰ To overcome this issue, liquid chromatography is preferred,^{21 22 23 24} but analyses by reversed-phase liquid chromatography (RP-LC) are difficult because the aliphatic carboxylic acids are polar, so they are weakly retained on the column. Different sample pretreatment can be used to enhance the separation in such cases, for example, using aldol condensation.²⁵ Another option, which is less laborious than changing the sample preparation method, is using ionexchange chromatography.²⁴,²⁶ Furthermore, an alternative method is hydrophilic interaction liquid chromatography (HILIC), which can provide a relatively strong retention of these metabolites. When combined with tandem mass spectrometry (MS/MS), it provides selective methods for the analysis of highly hydrophilic molecules.^{27 28 29 30 31} Although the excellent performance of HILIC-MS/MS, these instruments are expensive to acquire and maintain, besides, a large amount of toxic organic solvents are usually employed. A simpler and cheaper alternative is capillary zone electrophoresis (CZE). This technique requires small amounts of samples and reagents, and simple sample preparation. Another advantage of this technique is its capability to detect metabolites that can be ionized, like the organic acids of the TCA cycle.³² Since these organic acids are commonly found in some types of food samples, CZE methods have been applied to analyze them in coffee,³³ wine³⁴ and other grape-derived products.³⁵,³⁶

CZE has also been successfully applied to the analysis of organic acids in cell extracts.¹,³⁷

Therefore, our work aimed to develop and validate a method for the analysis of the intermediates of the TCA cycle, namely α-ketoglutaric, citric, fumaric, malic and succinic acids, and lactic acid in human urine. The method was applied to the identification and quantification of these metabolites in urine samples from individuals affected by obesity before and after being submitted to bariatric surgery.

Experimental

Reagents and solutions

All reagents were of analytical grade and solvents were of chromatographic purity. Aqueous solutions were prepared with ultrapure water with a resistivity of 18.2 MΩ cm (Milli-Q system, Millipore, Bedford, USA). The standards α-ketoglutaric (99% purity), citric (99% purity), fumaric (99% purity), lactic (85-90% purity), malic (99% purity), succinic (99% purity), and chloroacetic (99% purity) acids and the reagents 2,6-pyridine carboxylic acid (PDC) (99% purity), β-alanine (99% purity) and hexadecyltrimethylammonium hydroxide (CTAOH) were purchased from Sigma-Aldrich (St. Louis, USA).

Individual stock solutions of the selected carboxylic acids at 1000 mg L^–1 concentration were prepared by dissolving appropriate amounts of each acid with 50% methanol in deionized water and stored at 4 °C until analysis.

Instrumental

All experiments were conducted in a capillary electrophoresis system (model 7100, Agilent Technologies, Palo Alto, USA) equipped with a diode array detector (DAD) and data treatment software (HP ChemStation).

The separations were carried out in a fused-silica capillary (Polymicro Technologies, Phoenix, USA) of 100 cm total length (91.5 cm effective length × 75 μm internal diameter (I.D.) and 375 μm outer diameter (O.D.)). The background electrolyte (BGE) was composed of 24 mmol L^–1 PDC, 76 mmol L^–1 β-alanine, and 4 mmol L^–1 CTAOH (pH 4). It was prepared directly by weighing the reagents. Before the analysis, the capillary was conditioned by flushing 1 mol L^–1 NaOH (5 min), deionized water (5 min), and BGE (5 min). Between runs, the capillary was rinsed with the BGE for 1 min.

Samples were injected hydrodynamically (100 mBar 5 s^–1) at 20 °C and analyzed using indirect detection at 290 nm. The voltage applied to separation was –30 kV, with negative polarity on the injection side.

Human urine samples

All the measurements performed in human urine samples to validate the method here presented were performed under the ethical guidelines of the Ethics Committee for Human Research (CAAE 19792013.5.0000.0121) of the Universidade Federal de Santa Catarina (UFSC), Florianópolis, SC, Brazil, and subjects’ consent was obtained according to the Declaration of Helsinki.

Under the supervision of the professionals of the Endocrinology Service of the Polydoro Ernani de São Thiago University Hospital of UFSC, urine samples were collected from participants affected by obesity grade II with associated comorbidities or obesity grade III who were eligible for bariatric surgery. Samples were collected hours previous to the surgery and one year after. After collection, urine samples were immediately frozen at –80 °C and maintained at this temperature until sample preparation for analyses.

Sample preparation

A total of 32 urine samples from 16 individuals were analyzed, one sample collected before the surgery and the other, one year after the surgery. For deproteinization, aliquots of 400 µL urine sample were transferred to glass tubes to which 800 µL of an ice-cold 2:1 acetonitrile:methanol mixture was added. Samples were then vortexed for 10 s and centrifuged at 2,000 × g for 15 min at 4 °C. The resulting supernatant was collected, transferred to vials, and then frozen at –20 °C until analysis. Prior to analysis, the samples were thawed at room temperature and diluted with chloroacetic acid (internal standard-IS) whose final concentration was set to 30 mg L^-1. Samples were transferred to vials for analysis by CZE-DAD.

Method validation

For the method validation, the parameters assessed were system suitability, linearity, intra and inter-day precision, matrix effect, and accuracy. The method validation was performed based on the guidelines from AOAC International.³⁸

System suitability

The system suitability was evaluated by injecting a standard solution 10 consecutive times and calculating the coeffcient of variation (CV) of the mean of both the peak area and migration time corrected by the IS.

Linearity and matrix effect

The linearity was studied by the analysis of calibration curves prepared with standard solutions. For citric acid, the curves were prepared in seven concentration levels, with independent triplicates in each level, and the linear range was from 10 to 300 mg L^–1. For the other analytes, the curves were prepared in six concentration levels, also with three independent replicates in each level, and the linear range was from 5 to 55 mg L^–1 for each analyte. The ordinary least squares method was applied, and the statistical signifcance of the regression model was checked using the F test. The outliers were assessed by applying the Grubbs test and the Shapiro-Wilk test was used to evaluate the normality of the residuals.

The matrix effect was assessed by comparing the slopes obtained for each standard solution calibration curve and matrix addition calibration curves. The F test was used to verify if the variances were different or equal and then the adequate t-test was applied to confrm the presence or not of matrix effect.

Limits of detection and quantification

The limit of detection (LOD) and the limit of quantification (LOQ) were determined based on the signal-to-noise ratio obtained by analyte standard solutions. The LOD was determined as three times the signal-to-noise ratio and the LOQ was ten times the signal-to-noise ratio.

Precision

The repeatability (intra-day precision) was evaluated by the injection of a standard solution in three different concentration levels in three independent replicates. For the intermediate precision (inter-day precision) the same process was repeated over the course of three days. The results were expressed as CV (%).

Accuracy

The accuracy was assessed by measuring the recovery of three samples spiked at three concentration levels in the linear range of the calibration curve, for each analyte. All the measurements were performed in triplicate. The recovery was calculated as follows: Recovery (%) = (measured concentration – endogenous concentration)/spiked concentration × 100%. The results were expressed as the means of the recovery of the independent replicates in each level and the standard deviation (SD) of the replicates was calculated.

Application

The method was applied to the analysis of urine samples from patients with obesity before bariatric surgery and one year after bariatric surgery. The results were expressed as mg of each acid per mg of creatinine. To verify if there was a difference in concentrations of organic acids between the samples collected before the bariatric surgery and one year after bariatric surgery, a principal component analysis (PCA) was applied. The PCA was carried out in R Statistical Software³⁹ using the factoextra R package.⁴⁰

Results and Discussion

Method development and optimization

The method development was first approached by selecting an appropriate BGE pH. For that purpose, a plot of mobility versus pH was drawn for all analytes under investigation (Figure 2) using pK_a and electrophoretic mobilities obtained from a database compiled by Hirokawa et al.⁴¹ The IS selected was chloroacetic acid. The choice was based on the absence of this analyte in the human samples, and therefore, its effective mobility was also computed together with the mobilities of all analytes for optimization purposes. In the pH range between 3 and 4, the differences in mobility were maximized, proving this region to be more promising for optimization purposes.

Figure 2
Effective mobility versus pH curves for the system: α-ketoglutaric, citric acid, fumaric acid, lactic acid, malic acid, succinic acid, chloroacetic acid (IS), β-alanine, PDC (2,6-pyridine carboxylic acid).

The next step was to select appropriate BGE components, including its co-ion and counter-ion. In the determination of anions by CZE, the co-ion should have mobility as close as possible to the mobility of most of the analytes under consideration to minimize peak distortion.⁴²,⁴³ Since the analytes have low or no molar absorptivity, it was also imperative that the selected anion should be a chromophore so that indirect detection was contemplated. Besides that, adequate buffering capacity is necessary to reduce pH variations. Considering these characteristics, PDC was selected as co-ion, a chromophore that enabled the indirect detection mode in 290 nm. PDC’s effective mobility is also displayed in Figure 2. At the selected pH of 4, its effective mobility approached the average mobility of the analytes under investigation. To give adequate buffering capacity for the BGE in the selected pH, β-alanine was chosen as the counter-ion.

Urine has a high electrical conductivity (> 20 mS cm^–1) due to its high NaCl content, which may cause the spreading of the sample zone, resulting in peak broadening.⁴⁴,⁴⁵ To overcome this issue, we used a high concentration of PDC (24 mmol L^–1). This concentration of PDC in the BGE ensures a higher buffer capacity and enhances the separation, allowing the samples to be directly diluted and injected into the CZE system.

In the analysis of small anions, speed is usually sought and that is customarily achieved by reversing the direction of the electroosmotic fow. For this method, CTAOH, a cationic surfactant, was chosen. It is a BGE additive that promotes fow reversal at certain conditions through dynamic modifcation of the capillary wall. The separation of the standard solution sample is presented in Figure 3.

Figure 3
Electropherogram of a standard solution containing the analytes in the concentration of 55 mg L^–1, and the IS in the concentration of 30 mg L^{– 1} obtained using the validated method by CZE. Experimental conditions: BGE, 24 mmol L^–1 PDC, 76 mmol L^–1 β-alanine and 4 mmol L^–1 CTAOH (pH 4); fused-silica capillary, 100 cm total length (91.5 cm effective length), 75 μm I.D.; hydrodynamic injection 100 mBar 5 s^–1; voltage –30 kV; temperature 20 °C; indirect detection at 290 nm. Signals: chloroacetic acid (IS) (1), fumaric acid (2); α-ketoglutaric acid (3); malic acid (4); citric acid (5); lactic acid (6); succinic acid (7).

Validation

The proposed methodology was validated by determining its performance characteristics regarding selectivity, linearity, LOD, LOQ, injection precision, intra-and inter-day precision, and accuracy.

System suitability

To evaluate the instrumental precision, ten consecutive injections of a mixture containing all standards at 25 mg L^–1 were performed. The CV for the corrected peak areas were between 1.69 and 5.09% (Table 1). For the corrected migration time the CV for all analytes were lower than 0.13%. These results demonstrated that the instrumental system was adequate for validation and analysis.

Table 1
Analytical performance of the proposed CZE method for the analysis of urine sample

Parameter	α-Ketoglutaric acid	Citric acid	Fumaric acid	Lactic acid	Malic acid	Succinic acid
Instrumental precision, corrected peak area (CV; n = 10) / %	5.10	2.93	1.96	1.93	2.67	1.69
Instrumental precision, corrected migration time (CV; n = 10) / %	0.03	0.04	0.01	0.06	0.04	0.12
Intra-day precision, corrected peak area (CV; n = 9) / %	6.88-2.05	7.49-2.01	5.31-1.77	6.54-1.78	1.25-0.79	7.09-2.11
Intra-day precision, corrected migration time (CV; n = 9) / %	0.14-0.09	0.32-0.16	0.06-0.05	0.45-0.25	0.24-0.15	0.82-0.47
Inter-day precision, corrected peak area (CV; n = 27) / %	9.06-7.13	7.52-9.78	3.90-3.61	10.51-7.49	9.38-9.34	9.81-7.98
Inter-day precision, corrected migration time (CV; n = 27) / %	0.24-0.92	0.36-0.83	0.09-0.12	0.44-1.09	0.36-0.97	1.04-1.12
Linearity, linear range / (mg L–1)	5-55	10-300	5-55	5-55	5-45	5-55
Linearity, slope	0.0223	0.0224	0.0343	0.0418	0.0236	0.0537
Slope standard deviation	0.0006	0.0004	0.0003	0.0006	0.0012	0.0009
Linearity, intercept	0.0161	–0.0661	0.0092	0.0452	0.0610	0.0462
Intercept standard deviation	0.0200	0.0649	0.0091	0.0223	0.0346	0.0304
Linearity, determination coefficient (R2)	0.9973	0.9987	0.9998	0.9990	0.9922	0.9989
LOQ / (mg L–1)	3.34	5.57	0.74	2.38	1.09	1.94
LOD / (mg L–1)	1.10	1.84	0.24	0.78	0.36	0.64

• CV: coefficient of variance; LOQ: limit of quantification; LOD: limit of detection.

Linearity and matrix effect

The calibration curves were obtained by plotting the ratio of the peak area of the analyte to the IS and the analyte concentration. The linear least-square regression model was applied, and the results are presented in Table 1. The determination coefficients of all analytes were greater than 0.9973, indicating good linearity. The statistical significance of the regression model was evaluated, and the coefficients of the proposed model were statistically significant (F values greater than 381.7 and p-values lower than 2.93 × 10^–4). The Grubbs test was applied, and it confirmed that there were no outliers. The assumption that the residuals follow a normal distribution was confirmed by the Shapiro-Wilk test (Shapiro-Wilk test values close to 1 and p-values greater than 0.05), certifying that de proposed models were suitable to calculate the concentration of the analytes in the samples.⁴⁶,⁴⁷

For the matrix effect, the F test was applied to the slopes obtained for each standard solution calibration curve and matrix addition calibration curves. Then, the appropriate t-test was applied. Different means indicated that there was a matrix effect. Malic acid was the only analyte that presented matrix effect, making matrix-match calibration necessary.

LOD and LOQ

LOD and LOQ were obtained by recording signal-to-noise ratios. The LOD and LOQ data for all analytes are presented in Table 1. LOD were in the range of 0.24-1.84 mg L^–1 whereas LOQ were in the range of 0.74-5.57 mg L^–1.

Precision

For the intra-day precision, the CV was less than 7% for the peak areas corrected by the IS and less than 1% for the migration time for all analytes. For the inter-day precision, all analytes presented a CV lower than 11% for the peak areas and less than 1% for the migration time. Both intra and inter-day precisions were considered satisfactory. The precision data are presented in Table 1.

Accuracy

The accuracy was assessed according to the apparent recovery of a known amount of all standards at three concentration levels added to the urine samples. The recovery for all analytes ranged from 87.33 to 114.52% (Table S1, Supplementary Information (SI) section), confirming the method’s accuracy.

Sample analysis

To demonstrate the applicability of the proposed method, urine samples from individuals affected by obesity submitted to bariatric surgery were analyzed for the content of TCA intermediates (Figure 4). To confirm the presence of the analytes in the samples, the individual organic acids were identified in electropherograms by spiking the biological samples with standard solutions of each organic acid (Figure S1, SI section). The concentrations of the analytes in each sample were calculated using the models obtained with the calibration curves. The concentrations of the acids were then corrected by the concentration of creatinine by mL of urine and the results were expressed as mg of analyte per mg of creatinine in each urine sample.

Figure 4
Electropherogram obtained with the analysis of a urine sample using the validated method. The sample was from an individual affected by obesity after one year of being submitted to bariatric surgery. Experimental conditions: BGE, 24 mmol L^–1 PDC, 76 mmol L^–1 β-alanine and 4 mmol L^–1 CTAOH (pH 4); fused-silica capillary, 100 cm total length (91.5 cm effective length), 75 μm I.D.; hydrodynamic injection 100 mBar 5 s^–1; voltage –30 kV; temperature 20 °C; indirect detection at 290 nm. Signals and its concentrations: chloroacetic acid (IS) (1), 30 mg L^–1; malic acid (2), 10.50 mg L^–1; citric acid (3), 128.03 mg L^–1; lactic acid (4), 14.09 mg L^–1.

α-Ketoglutaric, fumaric, and succinic acids were not detected in any of the analyzed samples. These acids are rarely detected despite the biological matrix being analyzed. Methods using CZE were not able to detect them in cells¹,³⁷ and LC methods could not detect them in urine and plasma.²¹,²² Even when present in the sample, their concentrations are low. A LC method detected them in such low concentrations that they could not be quantified.²⁴ α-Ketoglutaric acid is difficult to detect due to its instability and pH sensitivity. It may be unstable under sample preparation conditions and consequently, the acid can undergo decarboxylation.⁴⁸

Regarding the other acids, a method using LC-MS/MS was able to quantify fumaric and succinic acids in serum, cells, and tissue and they presented a low concentration in comparison to the other acids from the cycle being analyzed.²³ Fumaric, succinic and α-ketoglutaric were also analyzed by HILIC-MS/MS in tissue and the concentrations found were also the lowest when compared to the other acids from the TCA cycle analyzed.³⁰ The low concentrations of these metabolites may be due to high flux or turnover rates.⁴⁹

The method we proposed in this work was able to detect and quantify low concentrations of the analytes in the standard samples. Therefore, we considered that these metabolites could not be quantified because their concentrations were below LOQ or due to their instability.

Lactic acid concentrations varied from 0.01 to 0.08 mg mg^–1, and malic acid concentrations were in the range of 0.01 to 0.03 mg mg^–1. Citric acid was the analyte with the highest concentration in the samples, and the one that presented the most variation in concentration, varying from 0.02 to 0.43 mg mg^–1. To evaluate if the different concentrations of citric, lactic, and malic acids found in the analyzed samples were due to metabolic changes related to bariatric surgery, a PCA analysis was carried out. The scores plot shows the variation explained by the first principal component (PC1) and second principal component (PC2) (Figure 5). The two principal components explained 89% of the total variance. However, the PCA did not classify the samples according to the period they were collected. Therefore, the differences between the two sets of samples were not strong enough to be detected. Hence, none of the acids can be used to distinguish the two sets of samples. The variation in the concentrations of the acids in the samples is more likely due to the natural biological variability between individuals because of general differences in metabolism, that are associated with the genetic background or nutritional intake.¹⁹

Figure 5
PCA scores plot of the 32 urine samples analyzed by the validated CZE method. The concentrations were corrected by the creatinine concentration in each sample. Post: samples collected one year after the bariatric surgery; Pre: samples collected before the bariatric surgery; PC: principal component.

Urinary metabolites are better indicators of kidney function while the state of the whole system is reflected in the serum metabolites.⁵⁰ Combined analyses of urine and serum metabolites might provide a better understanding of the metabolic changes involved in obesity. There were few subjects in this study for any conclusion to be drawn. Furthermore, there was no control group to verify if there was a difference in the concentration of the metabolites between healthy individuals and individuals affected by obesity. Nonetheless, the method developed allowed the simultaneous measurement of most of the TCA cycle intermediates and lactic acid and it has been proven to be efficient to quantify these metabolites in human urine.

The concentration of citric acid in blood and urine usually indicates the TCA cycle activity status since there is no known extra-renal elimination and oral intake has little influence.¹⁷ Citric acid and its metabolites have been associated with the development of prostate cancer.¹² Citric acid also plays an important role in urolithiasis and its urinary concentration could be useful in the early diagnosis of the formation of kidney stones. It forms stable soluble complexes with calcium, leading to a decrease in calcium oxalate and calcium phosphate formation, inhibiting the formation of kidney stones. Hence, lower concentrations of citric acid in urine can indicate the formation of kidney stones.¹³,¹⁴,⁵¹ Lower excretion of the TCA cycle metabolites has also been associated with non-diabetic chronic kidney disease.¹⁷ These metabolites have also shown the potential to be biomarkers of kidney impairment in the early stage of diabetic kidney disease.¹⁵,¹⁶ Kidneys present a high number of mitochondria thus, disruption of mitochondrial homeostasis may lead to reduced energy production, resulting in renal function impairment. Therefore, abnormalities of the TCA cycle detected in urine may be helpful to study the development and progression of different health conditions, especially diseases related to kidney function.¹⁵,¹⁶ Here we presented a method capable of analyzing these variations and that can be applied in various health research.

When compared to other methods in the literature¹,³⁷ that used CZE to detect the TCA cycle intermediates, our method has the advantage of using hydrodynamic injection. The electrokinetic injection is discriminative since the amount of analyte injected is dependent on the analyte’s mobility. In comparison, the hydrodynamic injection precision and accuracy are not strongly affected by experimental conditions. Therefore, it presents better repeatability and is less prone to measurement errors, making it more suitable for quantitative analysis. Moreover, our method offers a simpler sample pretreatment, making it less laborious and susceptible to analyte loss.

Conclusions

The proposed method was validated and provided suitable results for parameters such as linearity, precision, accuracy, LOD, and LOQ. Compared to other methods commonly used to quantify the TCA cycle intermediates in biological samples, this method has a relative simplicity in terms of instrumentation, and small sample consumption, besides the low generation of residues, contributing to green chemistry and also offering high separation efficiency. It was able to simultaneously determine five organic acids of the TCA cycle, and lactic acid, all essential metabolites to human metabolism. When compared to other methods in the literature that also used CZE, our method uses a more reliable sample injection mode and requires a simpler sample pretreatment.

The results obtained with this method along with its advantages evidence its potential to be employed in the analyses of urine, especially in research concerning human kidney health. Furthermore, the developed CZE-DAD method shows broad prospects to be applied in the analysis of the intermediates of the TCA cycle in other biological samples.

Supplementary Information

Supplementary data are available free of charge at http://jbcs.sbq.org.br as a PDF file.

Acknowledgments

The authors wish to thank Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq),

Fundação de Amparo à Pesquisa do Estado de Santa Catarina (FAPESC) (grant No. 2019TR0847) and INCT Catalysis for funding this research.

References

• ¹
  Markuszewski, M. J.; Otsuka, K.; Terabe, S.; Matsuda, K.; Nishioka, T.; J. Chromatogr. A 2003, 1010, 113. [Crossref]
  » Crossref
• ²
  Ryan, D. G.; Frezza, C.; O’Neill, L. A.; Curr. Opin. Biotechnol 2021, 68, 72. [Crossref]
  » Crossref
• ³
  Choi, I.; Son, H.; Baek, J.-H.; Life 2021, 11, 69. [Crossref]
  » Crossref
• ⁴
  Martínez-Reyes, I.; Chandel, N. S.; Nat. Commun 2020, 11, 102. [Crossref]
  » Crossref
• ⁵
  Owen, O. E.; Kalhan, S. C.; Hanson, R. W.; J. Biol. Chem 2002, 277, 30409. [Crossref]
  » Crossref
• ⁶
  Inigo, M.; Deja, S.; Burgess, S. C.; Annu. Rev. Nutr 2021, 41, 19. [Crossref]
  » Crossref
• ⁷
  Jones, T. E.; Pories, W. J.; Houmard, J. A.; Tanner, C. J.; Zheng, D.; Zou, K.; Coen, P. M.; Goodpaster, B. H.; Kraus, W. E.; Dohm, G. L.; Surgery 2019, 166, 861. [Crossref]
  » Crossref
• ⁸
  De-Cleva, R.; Cardia, L.; Vieira-Gadducci, A.; Greve, J. M.; Santo, M. A.; Arq. Bras. Cir. Dig 2021, 34, e1579. [Crossref]
  » Crossref
• ⁹
  Guasch-Ferré, M.; Santos, J. L.; Martínez-González, M. A.; Clish, C. B.; Razquin, C.; Wang, D.; Liang, L.; Li, J.; Dennis, C.; Corella, D.; Muñoz-Bravo, C.; Romaguera, D.; Estruch, R.; Santos-Lozano, J. M.; Castañer, O.; Alonso-Gómez, A.; Serra-Majem, L.; Ros, E.; Canudas, S.; Asensio, E. M.; Fitó, M.; Pierce, K.; Martínez, J. A.; Salas-Salvadó, J.; Toledo, E.; Hu, F. B.; Ruiz-Canela, M.; Am. J. Clin. Nutr 2020, 111, 835. [Crossref]
  » Crossref
• ¹⁰
  Bulló, M.; Papandreou, C.; García-Gavilán, J.; Ruiz-Canela, M.; Li, J.; Guasch-Ferré, M.; Toledo, E.; Clish, C.; Corella, D.; Estruch, R.; Ros, E.; Fitó, M.; Lee, C. H.; Pierce, K.; Razquin, C.; Arós, F.; Serra-Majem, L.; Liang, L.; Martínez-González, M. A.; Hu, F. B.; Salas-Salvadó, J.; Metabolism 2021, 125, 154915. [Crossref]
  » Crossref
• ¹¹
  Sciacovelli, M.; Frezza, C.; Free Radicals Biol. Med 2016, 100, 175. [Crossref]
  » Crossref
• ¹²
  Buszewska-Forajta, M.; Monedeiro, F.; Gołębiowski, A.; Adamczyk, P.; Buszewski, B.; Metabolites 2022, 12, 268. [Crossref]
  » Crossref
• ¹³
  Zhou, S.; Kong, L.; Wang, X.; Liang, T.; Wan, H.; Wang, P.; Anal. Chim. Acta 2022, 1191, 339178. [Crossref]
  » Crossref
• ¹⁴
  Sun, Q.; Tu, J.; Yaroshenko, I.; Kirsanov, D.; Legin, A.; Wang, P.; Procedia Chem 2016, 20, 52. [Crossref]
  » Crossref
• ¹⁵
  Lunyera, J.; Diamantidis, C. J.; Bosworth, H. B.; Patel, U. D.; Bain, J.; Muehlbauer, M. J.; Ilkayeva, O.; Nguyen, M.; Sharma, B.; Ma, J. Z.; Shah, S. H.; Scialla, J. J.; Metabolomics 2022, 18, 5. [Crossref]
  » Crossref
• ¹⁶
  Liu, J. J.; Liu, S.; Gurung, R. L.; Ching, J.; Kovalik, J. P.; Tan, T. Y. ; Lim, S. C.; J. Clin. Endocrinol. Metab 2018, 103, 4357. [Crossref]
  » Crossref
• ¹⁷
  Hallan, S.; Afkarian, M.; Zelnick, L. R.; Kestenbaum, B.; Sharma, S.; Saito, R.; Darshi, M.; Barding, G.; Raftery, D.; Ju, W.; Kretzler, M.; Sharma, K.; de Boer, I. H.; EBioMedicine 2017, 26, 68. [Crossref]
  » Crossref
• ¹⁸
  Vincent, A.; Savolainen, O. I.; Sen, P.; Carlsson, N. G.; Almgren, A.; Lindqvist, H.; Lind, M. V.; Undeland, I.; Sandberg, A. S.; Ross, A. B.; Mol. Nutr. Food Res 2017, 61, 1600400. [Crossref]
  » Crossref
• ¹⁹
  Denkert, C.; Budczies, J.; Weichert, W.; Wohlgemuth, G.; Scholz, M.; Kind, T.; Niesporek, S.; Noske, A.; Buckendahl, A.; Dietel, M.; Fiehn, O.; Mol. Cancer 2008, 7, 72. [Crossref]
  » Crossref
• ²⁰
  Tang, D.-Q.; Zou, L.; Yin, X.-X.; Ong, C. N.; Mass Spectrom. Rev 2016, 35, 574. [Crossref]
  » Crossref
• ²¹
  Kubota, K.; Fukushima, T.; Yuji, R.; Miyano, H.; Hirayama, K.; Santa, T.; Imai, K.; Biomed. Chromatogr 2005, 19, 788. [Crossref]
  » Crossref
• ²²
  Shurubor, Y. I.; Cooper, A. J. L.; Isakova, E. P.; Deryabina, Y. I.; Beal, M. F.; Krasnikov, B. F.; Anal. Biochem 2016, 503, 8. [Crossref]
  » Crossref
• ²³
  Rathod, R.; Gajera, B.; Nazir, K.; Wallenius, J.; Velagapudi, V. ; Metabolites 2020, 10, 103. [Crossref]
  » Crossref
• ²⁴
  Lu, S.; Sun, X.; Shi, C.; Zhang, Y. ; J. Chromatogr. A 2003, 1012, 161. [Crossref]
  » Crossref
• ²⁵
  Fan, J.; Liu, Y. ; Su, S.; Cao, Y. ; Yu, Y.; Acta Chromatogr 2021, 33, 322. [Crossref]
  » Crossref
• ²⁶
  Nozal, M. J.; Bernal, J. L.; Diego, J. C.; Gómez, L. A.; Higes, M.; J. Liq. Chromatogr. Relat. Technol 2003, 26, 1231. [Crossref]
  » Crossref
• ²⁷
  Virgiliou, C.; Sampsonidis, I.; Gika, H. G.; Raikos, N.; Theodoridis, G. A.; Electrophoresis 2015, 36, 2215. [Crossref]
  » Crossref
• ²⁸
  Michopoulos, F.; Whalley, N.; Theodoridis, G.; Wilson, I. D.; Dunkley, T. P. J.; Critchlow, S. E.; J. Chromatogr. A 2014, 1349, 60. [Crossref]
  » Crossref
• ²⁹
  Gika, H. G.; Theodoridis, G. A.; Vrhovsek, U.; Mattivi, F.; J. Chromatogr. A 2012, 1259, 121. [Crossref]
  » Crossref
• ³⁰
  Zhang, W.; Hu, X.; Zhou, W.; Tam, K. Y. ; J. Proteome. Res 2018, 17, 3012. [ Crossref]
  » Crossref
• ³¹
  Dong, X.-M.; Wu, P.; Cheng, L.-H.; Shou, L.; Dong, H.; Chen, X.-Y.; Gao, H.-J.; Chen, J.-X.; Xiang, F.; Zhang, Q.; Zhang, D.-H.; Zhou, J.-L.; Xie, T.; J. Chromatogr. A 2022, 1686, 463654. [Crossref]
  » Crossref
• ³²
  Barbas, C.; Moraes, E. P.; Villaseñor, A.; J. Pharm. Biomed. Anal 2011, 55, 823. [Crossref]
  » Crossref
• ³³
  Galli, V. ; Barbas, C.; J. Chromatogr. A 2004, 1032, 299. [Crossref]
  » Crossref
• ³⁴
  Saavedra, L.; Barbas, C.; Electrophoresis 2003, 24, 2235. [Crossref]
  » Crossref
• ³⁵
  de Valme García, M.; Campoy, C. J.; Barroso, C.; Eur. Food Res. Technol 2001, 213, 381. [Crossref]
  » Crossref
• ³⁶
  Moreno, M. V. G.; Jurado, C. J.; Barroso, C. G.; Chromatographia 2003, 57, 185. [Crossref]
  » Crossref
• ³⁷
  Gao, P.; Shi, C.; Tian, J.; Shi, X.; Yuan, K.; Lu, X.; Xu, G.; J. Pharm. Biomed. Anal 2007, 44, 180. [Crossref]
  » Crossref
• ³⁸
  AOAC International; Appendix F: Guidelines for Standard Method Performance Requirements; AOAC: Gaithersburg, 2016. [Link] accessed in October 2023
  » Link
• ³⁹
  R Core Team; RStudio, 4.2.2; R Foundation for Statistical Computing, Austria, 2022.
• ⁴⁰
  Kassambara, A.; Mundt, F.; factoextra, 1.0.7.; CRAN, Austria, 2022.
• ⁴¹
  Hirokawa, T.; Nishino, M.; Aoki, N.; Kiso, Y.; Sawamoto, Y.; Yagi, T.; Akiyama, J.; J. Chromatogr 1983, 271, D1. [Crossref]
  » Crossref
• ⁴²
  Gebauer, P.; Beckers, J. L.; Boček, P.; Electrophoresis 2002, 23, 1779. [Crossref]
  » Crossref
• ⁴³
  Doble, P.; Macka, M.; Haddad, P. R.; TrAC, Trends Anal. Chem 2000, 19, 10. [Crossref]
  » Crossref
• ⁴⁴
  Santoro, C.; Garcia, M. J. S.; Walter, X. A.; You, J.; Theodosiou, P.; Gajda, I.; Obata, O.; Winfield, J.; Greenman, J.; Ieropoulos, I.; ChemElectroChem 2020, 7, 1312. [Crossref]
  » Crossref
• ⁴⁵
  Tůma, P.; Šustková-Fišerová, M.; Opekar, F.; Pavlíček, V. ; Málková, K.; J. Chromatogr. A 2013, 1303, 94. [Crossref]
  » Crossref
• ⁴⁶
  Augusto, F.; de Andrade, J. C.; Custodio, R.; Chemkeys 2000, 3, 1. [Crossref]
  » Crossref
• ⁴⁷
  Pimentel, M. F.; de Barros Neto, B.; Quim. Nova 1996, 19, 268. [Link] accessed in October 2023
  » Link
• ⁴⁸
  Al Kadhi, O.; Melchini, A.; Mithen, R.; Saha, S.; J. Anal. Methods Chem 2017, 2017, 5391832. [Crossref]
  » Crossref
• ⁴⁹
  Yamamoto, T.; Sato, K.; Yamaguchi, M.; Mitamura, K.; Taga, A.; Biochem. Biophys. Res. Commun 2021, 584, 53. [Crossref]
  » Crossref
• ⁵⁰
  Li, M.; Wang, X.; Aa, J.; Qin, W.; Zha, W.; Ge, Y. ; Liu, L.; Zheng, T.; Cao, B.; Shi, J.; Zhao, C.; Wang, X.; Yu, X.; Wang, G.; Liu, Z.; Am. J. Physiol.: Renal Physiol 2013, 304, F1317. [Crossref]
  » Crossref
• ⁵¹
  Khaskhali, M. H.; Bhanger, M. I.; Khand, F. D.; J. Chromatogr. B: Biomed. Sci. Appl 1996, 675, 147. [Crossref]
  » Crossref

Edited by

Editor handled this article: Andrea R. Chaves (Associate)

Publication Dates

• Publication in this collection
  01 Mar 2024
• Date of issue
  2024

History

• Received
  07 July 2023
• Accepted
  17 Oct 2023