Condensation of delta-1-piperideine-6-carboxylate with ortho-aminobenzaldehyde allows its simple, fast, and inexpensive quantification in the urine of patients with antiquitin deficiency

Abstract

Antiquitin (ATQ) deficiency leads to tissue, plasma, and urinary accumulation of alpha-aminoadipic semialdehyde (AASA) and its Schiff base delta-1-piperideine-6-carboxylate (P6C). Although genetic testing of ALDH7A1 is the most definitive diagnostic method, quantifications of pathognomonic metabolites are important for the diagnosis and evaluation of therapeutic and dietary interventions. Current metabolite quantification methods use laborious, technically highly complex, and expensive liquid chromatography-tandem mass spectro-metry, which is available only in selected laboratories worldwide. Incubation of ortho-aminobenzaldehyde (oABA) with P6C leads to the formation of a triple aromatic ring structure with characteristic absorption and fluorescence properties. The mean concentration of P6C in nine urine samples from seven ATQ-deficient patients under standard treatment protocols was statistically highly significantly different (P < .001) compared to the mean of 74 healthy controls aged between 2 months and 57 years. Using this limited data set the specificity and sensitivity is 100% for all tested age groups using a P6C cut-off of 2.11 μmol/mmol creatinine, which represents the 99% prediction interval of the P6C concentrations in 17 control urine samples from children below 6 years of age. Plasma P6C concentrations were only elevated in one ATQ subject, possibly because P6C is trapped by pyridoxal-5-phosphate (PLP) blocking fusing with oABA. Nevertheless, both urine and plasma samples were amenable to the quantification of exogenous P6C with high response rates. The P6C quantification method using fusion of oABA with P6C is fast, simple, and inexpensive and might be readily implemented into routine clinical diagnostic laboratories for the early diagnosis of neonatal pyridoxine-dependent epilepsy.

Keywords: antiquitin deficiency; delta-1-piperideine-6-carboxylate; ortho-aminobenzaldehyde; pyridoxal-5-phosphate; pyridoxine-dependent epilepsy; α-aminoadipic semialdehyde.

Figure 1

P6C concentrations can be precisely quantified in the urine samples of healthy volunteers and ATQ patients with high response rates of exogenously spiked P6C. A, Representative absorption scans of urine from a healthy volunteer (HV) after adding 0.5% ethanol (oABA matrix) or oABA and different concentrations of P6C; crosses (x) 0.5% ethanol; white squares () no P6C spiking; black triangles () 3.1 μM; white circles () 6.3 μM; black circles () 12.5 μM and white triangles () 25 μM P6C spiking; each symbol represents the mean of duplicates; B, These curves were generated by subtracting the 0.5% ethanol control from the curves shown in (A) representing specific oABA/P6C signal; C, The resulting standard curve from the data shown in (B) using absorption data at 460 nm; the regression line is y = 0.00366x + 0.0044; R = 0.998; D, Percent P6C response after exogenous P6C spiking presenting urinary data from 8 HVs (Mean_HV; two independent experiments in duplicate), nine individual samples from seven individual ATQ subjects and the means of the samples from Graz (Mean_G; three independent experiments in duplicate) and Zurich (Mean_Z; two independent experiments in duplicate); the means ± SE of the means (SEMs) are shown; ATQ, antiquitin; oABA, ortho‐aminobenzaldehyde; P6C, delta‐1‐piperideine‐6‐carboxylate

Figure 2

Urinary delta‐1‐piperideine‐6‐carboxylate (P6C) concentrations of nine samples from seven different antiquitin (ATQ) subjects are highly elevated compared to controls from 74 healthy volunteers of different ages. A, Six and three ATQ urine samples from Zurich and Graz, respectively, were measured 4‐ (Zurich) and 5‐ (Graz) times in duplicate in independent experiments; all 40 measurements are shown as black circles (); All 118 measurements from 74 urine samples from healthy volunteers (HV) of different ages are shown as white circles (); the solid and dashed lines represents the mean (2.2 μM) and the 99% prediction interval (PI; mean + 2.648*SD*sqrt(1 + 1/n); 5.71), respectively, from 74 HV urine samples; B, The data from (A) were converted to μmol P6C/mmol creatinine and the mean of all control samples (0.45) is shown as solid line. The dashed line at 1.33 represents the 99% PI (mean + 2.648*SD*sqrt[1 + 1/n]) of all urine control samples; the dotted line at 2.11 represents the 99% PI (mean + 2.921*SD*sqrt[1 + 1/n]) of only the control cohort samples from children below 6 years of age (n = 17)

Figure 3

Examples of representative absorptions scans used for the quantification of endogenous and exogenous P6C in urine samples of ATQ and control subjects. Each symbol represents the mean of duplicates; G193 (A); Z1_1 (B) and Z3 (C) are data from 3 ATQ subjects and 3 healthy volunteers are shown in (D). In A‐C on the left side white squares () represent urine samples incubated with 0.5% ethanol; black circles () with oABA and black squares () with oABA, and 10 μM P6C; in (D) white and black symbols represent incubation with ethanol and oABA, respectively; in A‐C on the right side white squares () represent the specific endogenous P6C signal after subtracting the 0.5% ethanol from the oABA data; black circles () represent the specific signal after subtracting the oABA from the spiked P6C representing exogenous P6C and black squares () represent the signal after subtracting the ethanol from the exogenous P6C data representing both endogenous and exogenous P6C; in (D) the symbols represent endogenous P6C of three controls. ATQ, antiquitin; oABA, ortho‐aminobenzaldehyde; P6C, delta‐1‐piperideine‐6‐carboxylate