Introduction

Dopamine is a neurotransmitter within the central nervous system that plays a crucial role in the regulation of numerous physiological processes1. Dopaminergic signaling dysfunction can result in a variety of developmental problems, such as autism spectrum disorder (ASD), attention deficit/hyperactivity disorder (ADHD), and schizophrenia2. Dopamine is effectively metabolized into the inactive metabolite homovanillic acid (HVA) by a sequence of events catalyzed mostly by the enzyme’s monoamine oxidase and catechol-O-methyltransferase3. The quantification of HVA in biological fluids such as cerebrospinal fluid, plasma, and urine offers valuable insights into dopamine metabolism and has recently been utilized as a biomarker for the identification of various pathological conditions, including neuroblastoma, dementia, and ASD4.

Several investigations were undertaken to investigate the association between HVA concentrations in various bodily fluids and ASD. Khan et al. (2022) employed gas chromatography-mass spectrometry (GC-MS) to assess and compare the organic acid profiles in the urine of children diagnosed with ASD against normal peers5. The findings revealed statistically significant differences in HVA levels between the two groups, implying potential dopaminergic irregularities in the ASD cohort under investigation. Kałuzna-Czaplińska et al. (2010) devised a GC-MS method to analyze HVA in urine samples obtained from 20 ASD and 36 normal children6. Autistic children have greater levels of HVA in their urine than healthy children. Gevi et al. (2020) innovated a liquid chromatography-mass spectrometry (LC-MS) technique to investigate urinary metabolomic profiles in a cohort of 40 ASD children alongside 40 controls7. The data indicated a heightened level of HVA in the urine of the ASD children relative to the control group. Conversely, Narayan et al. (1993) devised an LC-amperometric approach for quantifying HVA levels in cerebrospinal fluid from 17 autistic children and 15 healthy children, with the findings demonstrating no significant differences between the groups8.

Recently, enzymatic biocatalytic oxidation assays have gained considerable traction in the field of analytical chemistry due to their environmentally benign characteristics and selectivity9. Horseradish peroxidase (HRP) readily interacts with H2O2, resulting in the formation of the HRP-H2O2 complex, which possesses the capability to oxidize a diverse array of hydrogen donors. HVA is a nonfluorescent compound that, upon undergoing oxidation within the HRP-H2O2 system, is transformed into a highly fluorescent dimer10,11.

The aim of the present study was to determine and compare the serum levels of HVA in autistic and healthy children using a validated bioanalytical spectrofluorimetric methodology. The analytical method relies on conversion of HVA into a highly fluorescent dimer using HRP-H2O2 catalytic system. The method was validated using ICH M10 guidelines and the standard addition method (SAM) for determining the endogenous analytes was utilized. The findings of the study indicated that autistic children exhibit significantly elevated serum HVA levels compared to their healthy counterparts, confirming the link between dopaminergic signaling dysfunction and autism.

Experimental

Instrumentation

All fluorescence measurements were carried out with an FP-6200 spectrofluorometer (Jasco, Japan) and a 1 cm quartz cuvette. The apparatus used a 150-watt Xenon lamp as an excitation light source, and the slit widths for both monochromators were set to 10 nm.

Materials and solutions

HVA and HRP were procured from Sigma-Aldrich (USA). Monobasic potassium phosphate, sodium hydroxide, and H2O2 were sourced from ADWIC (Egypt). Solutions of HRP (2 U/mL), and H2O2 (0.05 mol/L) were prepared individually by dissolving the requisite amounts in distilled water. The stock solution of HVA (100 µg/mL) was formulated by dissolving 10 mg of HVA in 60 mL of distilled water within a 100 mL volumetric flask and subsequently adjusting the volume with additional distilled water. Moreover, three working solutions of HVA were generated from the stock solution at concentrations of 50, 100, and 400 ng/mL through dilution with distilled water.

Study samples

This study was approved by the Ethical Committee of the Faculty of Medicine, Al-Azhar University, Damietta, Egypt (approval code: DFM-IRB 000122367-24-11-008). All procedures adhered to the principles outlined in the Declaration of Helsinki. Written informed consent was obtained from all participants before their enrollment in the trial. Blood samples were collected for the determination of serum HVA levels between the hours of 7:00 and 8:00 am following an overnight fasting period from two age-matched groups of children. Group I comprises 24 children diagnosed with ASD, whereas Group II consists of 15 healthy developing children. None of the participants had received pharmacological treatments that could potentially confound the assessment of HVA. Baseline demographic characteristics, including mean age ± standard deviation (SD) and sex distribution, are presented in Table 1. No statistically significant difference in age was found between the two groups (P = 0.63), confirming appropriate age matching. The normality of serum HVA concentrations was assessed using the Shapiro–Wilk test, which showed no significant deviation from normality (P > 0.05); therefore, an independent samples t-test was used for group comparisons. The blood samples were permitted to undergo coagulation prior to the centrifugation process. The resultant serum was subsequently collected and preserved at a temperature of -20 °C.

Table 1 Baseline demographic characteristics of study participants.
Full size table

The SAM calibration graphs

Each serum sample was divided into four aliquots of equal volume (1 mL) which were utilized to construct its respective calibration curve. All aliquots, with the exception of one, were individually spiked with 1 mL of the HVA working solutions. The spiked solutions were designated as SAM spiking solutions, whereas the non-spiked solution was referred to as the blank solution. The solutions were treated with 1 mL of 0.1 M NaCl and 4 mL of acetonitrile for protein precipitation, vortexed for 1 min, then centrifuged for 30 min. The resulting supernatants evaporated, and the residues were dissolved with an appropriate proportion of water before being transferred to a series of 10-mL amber volumetric flasks. To each flask, 1 mL of HRP (2 U/mL) solution, 1 mL of H2O2 (0.05 mol/L) solution, and 1 mL of phosphate buffer solution (pH 7.8) were added. The solutions were incubated for 25 min in a dark place, after which the flasks were adjusted to the desired volume with distilled water. The fluorescence intensity (FI) of the solutions was measured at 430 nm following excitation at 335 nm. The SAM calibration curve for each serum sample was constructed by plotting the FI values against the concentrations of the four solutions (0 ng/mL for blank solution and 5, 10, and 40 ng/mL for SAM spiking solutions). Subsequently, the endogenous HVA concentration for each serum sample was determined as the negative x-intercept of the SAM calibration curve generated for that particular sample.

Results and discussion

Method development and optimization

In this study, HVA was oxidized into a highly fluorescent dimer via catalytic activity of HRP in the presence of H2O2. The dimeric entity of HVA was synthesized through the establishment of a covalent bond, which occurs by the cleavage of the methoxy functional group from the carbon atom at the 3′ position of one HVA molecule, facilitating a bond with the carbon atom at the 5′ position of an adjacent HVA monomer, as shown in Fig. 1. The resultant HVA dimer exhibited a strong emission peak at 430 nm following excitation at 335 nm, as illustrated in Fig. 2.

Fig. 1
Fig. 1
Full size image

Proposed catalytic oxidation pathway of HVA by HRP-H2O2 system.

Fig. 2
Fig. 2
Full size image

Excitation (Ex) and emission (Em) spectra of the HVA dimer.

The factors that affected fluorescence intensity of HVA dimer was studied and optimized. These factors involved HRP concentration, H2O2 concentration, buffer pH, and incubation time. To determine the optimal concentration of HRP, a range of concentrations (spanning from 0.5 to 4 U/mL) was analyzed. As depicted in Fig. 3, the fluorescence intensity demonstrated an increase corresponding to the rise in HRP concentration up to 2 U/mL, beyond which a gradual decrease was noted. Thus, the optimal concentration of HRP was established as 2 U/mL. In order to ascertain the optimal concentration of H2O2, various concentrations of H2O2 (ranging from 0.01 to 0.1 mol/L) were investigated. As shown in Fig. 4, the fluorescence intensity increased with increasing H2O2 concentration until it reached 0.05 mol/L, after which a gradual decline was documented. Therefore, the ideal concentration of H2O2 was recognized as 0.05 mol/mL. To identify the optimal pH for the conversion of HVA into HVA dimer, a variety of buffer solutions encompassing the entire pH spectrum were analyzed. It was determined that a phosphate buffer with a pH of 7.8 yielded the most favorable results. Additionally, the incubation duration necessary for the conversion of HVA into HVA dimer was examined over a timeframe of 0 to 40 min. It was observed that the maximum fluorescence intensity occurred following an incubation period of 25 min.

Fig. 3
Fig. 3
Full size image

Effect of the HRP concentration on the fluorescence intensity of HVA dimer.

Fig. 4
Fig. 4
Full size image

Effect of the H2O2 concentration on the fluorescence intensity of HVA dimer.

Bioanalytical method validation

The method was validated using ICH M10 guidelines and Coglianese et al. applications12,13.

The SAM calibration graphs

The calibration graphs of the study samples were constructed by plotting the FI values against the concentrations of the four calibration solutions (0 ng/mL for blank solution and 5, 10, and 40 ng/mL for SAM spiking solutions), as seen in Fig. 5. The regression equation (y = ax + b) corresponding to each calibration graph was established, wherein the intercept (b) signifies the FI value of the blank solution, which remains greater than zero owing to the existence of an unknown endogenous concentration of HVA within the real serum sample. When y is equal to zero, the negative x-intercept on the calibration graph denotes the concentration of endogenous HVA present in the blank solution as follows.

Fig. 5
Fig. 5
Full size image

Example of SAM calibration curve for HVA determination in real serum sample (dilution factor, 1:10 v/v).

$$\:\varvec{x}=\:-\frac{\varvec{b}}{\varvec{a}}$$
(1)

Additionally, the endogenous HVA concentrations in each SAM spiking solution (xn) was also calculated. Initially, the intercept for each SAM spiking solution was derived utilizing the subsequent Eq. 

$$\:{\varvec{b}}_{\varvec{n}}=\:\left(\varvec{a}\:{\varvec{x}}_{+\varvec{n}}\right)-\:{\varvec{y}}_{\varvec{n}}$$
(2)

In this context, (bn) represents the intercept, (x+ n) represents the nominal HVA concentration spiking into each SAM spiking solution, and (yn) represents the fluorescence intensity (FI) of each SAM spiking solution. Then the endogenous HVA concentration (xn) for each SAM spiking solution was calculated using Eq. 1. The final endogenous HVA concentration for each study sample was established as the mean of the endogenous concentrations evaluated in the blank solution (x) and SAM spiking solutions (xn).

Accuracy and precision

The parameters of accuracy (as percentage recovery, %R) and precision (as percent coefficient of variation, %CV) were assessed through the quantification of the HVA concentration (C+ n) spiked into each SAM spiking solution. One of the SAM calibration graphs was assessed three times in one day for intra-day calculations and three times across three days for inter-day calculations. The HVA concentration that was spiked into each SAM spiking solution was subsequently determined utilizing the following equation:

$$\:{\varvec{C}}_{+\varvec{n}\:}=\left(\raisebox{1ex}{${\varvec{y}}_{\varvec{n}}$}\!\left/\:\!\raisebox{-1ex}{$\varvec{a}$}\right.\right)\:-\:{\varvec{x}}_{\varvec{n}}$$
(3)

Then the %R was calculated as the following:

$$\:\varvec{\%}{\varvec{R}}_{+\varvec{n}}=\:\left(\raisebox{1ex}{${\varvec{C}}_{+\varvec{n}}$}\!\left/\:\!\raisebox{-1ex}{${\varvec{x}}_{+\varvec{n}}$}\right.\right)\varvec{*}100\:$$
(4)

The degree of accuracy was deemed satisfactory when the computed %R fell within ± 15% of the nominal HVA concentration, whereas the degree of precision was regarded as satisfactory when the computed %CV remained below 15%. All of the accuracy and precision results in Table 2 were within acceptable ranges, confirming the accuracy and precision of the approach used.

Table 2 The outcomes of accuracy and precision.
Full size table

Lower limit of quantification (LLOQ)

In the absence of a blank matrix, determining the lower limit of quantification (LLOQ) presents considerable challenges. This study investigates the instrumental LLOQ (iLLOQ) in water as a neat blank and the method LLOQ (mLLOQ) in real serum sample. HVA samples were prepared at concentrations from 0.1 to 5 ng/mL in water and analyzed in triplicate to determine the iLLOQ. The iLLOQ was defined as the lowest concentration, yielding a signal-to-noise ratio of ≥ 5. In this study, the iLLOQ was established at 0.3 ng/mL. The mLLOQ was defined as the minimum HVA concentration detectable in serum with acceptable accuracy and precision. The serum sample for mLLOQ assessment was obtained from a composite pool of low HVA serum samples. The pooled sample was divided into five 1 mL aliquots. Four aliquots were spiked with HVA concentrations of 0.5, 1, 3, and 5 ng/mL, while one remained as a blank. The standard addition method was applied, and accuracy and precision were calculated for three determinations (n = 3) for each spiked concentration. Accuracy and precision values in Table 3 indicated that 1 ng/mL was the lowest added concentration in authentic serum samples that achieved acceptable accuracy and precision.

Table 3 Determination of the mLLOQ (n = 3).
Full size table

Determination of the serum HVA levels in ASD and healthy children

The spectrofluorimetric approach was successfully employed to estimate serum HVA levels in both ASD and healthy children. The data presented in Table 4 elucidated that the concentration of HVA was significantly elevated in the serum of children with ASD (mean value of 89 µg/L, n = 24) in comparison to that of healthy children (mean value of 56 µg/L, n = 15), P < 0.01. The serum HVA concentrations observed in this study (56–89 µg/L) were higher than reference values reported in studies employing GC–MS or HPLC, which typically range from 10 to 15 ng/mL for older children. This discrepancy can be attributed to methodological differences, as the spectrofluorimetric method measures native fluorescence within the total serum matrix, possibly including protein-bound or conjugated HVA forms that may not be extracted or quantified by chromatographic assays. Furthermore, variations in sample preparation, calibration matrices, and age-related metabolic differences may influence baseline HVA levels. Therefore, the higher apparent concentrations reported here are consistent with matrix effects inherent to fluorescence-based quantification rather than reflecting true physiological elevation.

Table 4 Serum homovanillic acid (HVA) levels in children with ASD and healthy Controls.
Full size table

Conclusion

In this study, serum levels of HVA in ASD and healthy children were measured using a spectrofluorimetric method. The approach is based on the conversion of HVA into a highly fluorescent dimer via an HRP–H₂O₂ catalyst system. The method was validated using ICH M10 recommendations, and endogenous HVA concentrations were calculated using the standard addition method. The study found that ASD children have significantly higher serum HVA levels than their healthy peers, supporting an association between increased dopaminergic activity and autism spectrum disorder.