Quantitation of plasma thiamine, related metabolites and plasma protein oxidative damage markers in children with autism spectrum disorder and healthy controls

Abstract

Aims/hypothesis: To assess thiamine and related metabolite status by analysis of plasma and urine in autistic children and healthy controls, correlations to clinical characteristics and link to plasma protein markers of oxidative damage.

Methods: 27 children with autism (21 males and 6 females) and 21 (15 males and 6 females) age-matched healthy control children were recruited. The concentration of thiamine and related phosphorylated metabolites in plasma and urine and plasma protein content of dityrosine, N-formylkynurenine and 3-nitrotyrosine was determined.

Results: Plasma thiamine and thiamine monophosphate concentrations were similar in both study groups (median [lower–upper quartile]): autistic children – 6.60 nM (4.48–8.91) and 7.00 nM (5.51–8.55), and healthy controls – 6.82 nM (4.47–7.02) and 6.82 nM (5.84–8.91), respectively. Thiamine pyrophosphate (TPP) was decreased 24% in autistic children compared to healthy controls: 6.82 nM (5.81–8.52) versus 9.00 nM (8.41–10.71), p < .01. Urinary excretion of thiamine and fractional renal clearance of thiamine did not change between the groups. No correlation was observed between clinical markers and the plasma and urine thiamine concentration. Plasma protein dityrosine content was increased 88% in ASD. Other oxidative markers were unchanged.

Conclusions/interpretation: Autistic children had normal plasma and urinary thiamine levels whereas plasma TPP concentration was decreased. The latter may be linked to abnormal tissue handling and/or absorption from gut microbiota of TPP which warrants further investigation. Increased plasma protein dityrosine may reflect increased dual oxidase activity in response to change in mucosal immunity and host–microbe homeostasis.

Keywords:

• Autism
• thiamine
• thiamine pyrophosphate
• dityrosine

Introduction

Autism spectrum disorders (ASD) are defined [1] as developmental disorders involving impaired mental skills, and affecting social interactions and a wide range of other disabilities, such as speech disturbances, repetitive and/or compulsive behaviors, hyperactivity, anxiety, difficulty to adapt to new environments, with or without cognitive impairment. The ASD definition now includes Asperger syndrome, classic autism, childhood disintegrative disorder and pervasive developmental disorders not otherwise specified. In a recent review, Bourgeron [2] discussed the wider notion of “ESSENCE”, or “Early Symptomatic Syndromes Eliciting Neurodevelopmental Clinical Examinations”, to refer to children presenting in clinical settings with impairing child symptoms before age 3–5 years in the fields of (1) general development, (2) communication and language, (3) social interrelatedness, (4) motor coordination, (5) attention, (6) activity, (7) behavior, (8) mood, and/or (9) sleep.

Genetic causes of ASD are clearly recognized in about 30–35% of cases, while for the remaining 65–70% of patients it is generally agreed that ASD results from the combination of environmental factors with multiple de novo mutations, CNV and rare genetic variants, each possibly lending to additive effects. In turn, the role played by environmental factors may be reflected in epigenetic modifications [3]. All this complexity is reflected in the high heterogeneity of the clinical presentation.

Phenotypic differences between ASD and typically developing (TD) children not involving the core autistic symptoms may help to identify metabolic impairments and pathway alterations shared by ASD patients despite their clinical heterogeneity. Thus, a number of candidate ASD biomarkers have emerged, which may not only help in diagnosis and in prognosis but point as well to the recognition of previously underscored mechanisms [4–6]. This is the case, for instance, of oxidative stress, which emerged as a widespread feature in ASD children [7,8] and led to the suggestion that inflammatory activity might be enhanced in these patients [9].

The essential micronutrient thiamine has an important role in maintaining a normal metabolic profile by conversion to thiamine pyrophosphate (TPP), which is a cofactor for pyruvate dehydrogenase, α-ketoglutarate dehydrogenase and transketolase. Transketolase has a pivotal role in glycolytic metabolism in sustaining both the oxidative and reductive pentosephosphate pathways. Maintenance of NADPH production in the oxidative branch sustains reductase activities for antioxidant capacity and metabolism of carbonyl compounds formed from lipid peroxidation and other sources. In a pilot intervention study of a synthetic thiamine derivative, thiamine tetrahydrofurfuryl disulfide, clinical benefit was claimed [10]. The involvement of thiamine homeostasis in the etiology of ASD has been recently discussed within the role played herein by oxidative metabolism, heavy metal deposition and cellular immunity [11].

The aim of this study is to characterize the status of thiamine and related metabolites in children with ASD and age-matched healthy children as nornal controls to assess if there is any disturbance in thiamine metabolsim that may contribute to a susceptibility to oxidative stress.

Materials and methods

Subjects

A total of 48 children were recruited. Of these, 27 had a diagnosis of ASD (21 male and 6 female) and 21 were classified as TD children (15 male and 6 female). The subject age of the two study groups was not significantly different. Subject age was: ASD group, 7.4 years ±2.0 years, range 5–12 years and TD group, 8.3 ± 2.1 years, range 5–12 years. All ASD subjects received a diagnosis of autism by the Child Neurological and Psychiatric Unit of the Bellaria Hospital of Bologna (IRCCS Institute of Neurological Sciences), according to the Diagnostic and Statistical Manual of Mental Disorders V (DSM IV TR [1]) criteria, Autism Diagnostic Observation Schedule (ADOS) [10] and Childhood Autism Rating Scale (CARS) [12]. Developmental and cognitive levels were assessed by Psychoeducational Profile-3 (PEP-3) [13] and Leiter International Performance Scale–Revised (Leiter-R) [14]. For both ASD and TD subjects, exclusion criteria were: presence of inflammatory or infective disease and taking food supplement at the time of study. No subject underwent any surgery intervention in the 4 months prior to blood and urine collection. None of the ASD subjects had active epilepsy at the time of blood and urine sampling. Subjects with medical and neurological comorbidity were excluded, assessed by electroencephalography (recorded during awake and sleep), cerebral magnetic resonance imaging, standard clinical and neurological examination, neurometabolic and genetic investigations (including 550 band karyotype, and molecular assay for Fragile X and MECP2. TD children were recruited in the local community, with no sign of cognitive, learning and psychiatric involvement. They were attending mainstream school and had not been subjected to stressful events. Dietary habits, as assessed by a Food Questionnaire, showed that both patients and controls were on a typical Mediterranean diet. Demographic and clinical features of ASD are summarized in Table 1. Blood was withdrawn in the morning from fasting children. Spot urine samples were the first ones in the morning. Whole blood and urine samples were collected form children with written informed consent of a parent. Blood samples were collected using ethylenediaminetetraacetic acid (EDTA) as anticoagulant. Plasma and blood cells were separated immediately by centrifugation (2000 × g, 10 min) and plasma samples stored at −80 °C until analysis and transferred between collaborating laboratories on dry ice. All subjects were recruited at the Child Neurological and Psychiatric Unit of the Bellaria Hospital of Bologna, Bologna, Italy. The study was approved by the Unified Ethics Committee of Bologna, Imola and Ferrara (CE BIF); project number 13062. The experiments conformed to the principles set out in the World Medical Association Declaration of Helsinki.

Table 1. Demographic and clinical features of the autistic children group.

No.	Age (months)	Gender	ADOS score	CARS total score	CARS activity level item score (hyperactivity)	CARS body use item score (stereotypes)	CARS total number of items whose score was ≥3	Autism severity level	Cognitive/developmental impairment (level)
1	72	M	22	41.0	2.0	3.0	9	Severe	Severe
2	67	M	14	34.0	2.0	2.0	2	Mild	Mild
3	66	M	18	40.5	2.5	3.5	9	Severe	Moderate
4	65	F	17	31.5	2.0	2.0	1	Mild	Mild
5	102	M	22	36.5	2.0	2.5	6	Mild	Mild
6	104	M	19	44.5	3.0	3.0	11	Severe	Moderate
7	81	M	19	36.5	2.5	2.0	7	Mild	Normal IQ
8	66	F	15	41.5	2.5	2.0	9	Severe	Borderline IQ
9	143	M	22	44.5	2.5	3.0	11	Severe	Severe
10	110	F	15	47.5	3.5	3.0	13	Severe	Moderate
11	144	M	20	39.0	3.0	3.5	8	Severe	Severe
12	74	M	22	42.5	3.0	2.5	10	Severe	Severe
13	80	M	21	40.5	2.5	3.0	9	Severe	Severe
14	79	M	19	37.0	2.5	3.0	7	Severe	Moderate
15	66	M	22	40.5	3.0	3.0	9	Severe	Moderate
16	68	M	20.0	41.5	2.5	3.0	11	Severe	Severe
17	94	M	21.0	46.0	3.5	4.0	11	Severe	Severe
18	68	F	20.0	43.5	2.5	3.0	10	Severe	Normal IQ
19	93	M	20.0	48.5	3.0	4.0	12	Severe	Severe
20	82	M	19.0	42.0	3.0	3.0	9	Severe	Moderate
21	115	M	19.0	39.5	3.0	3.0	7	Severe	Severe
22	74	M	19.0	41.0	2.5	3.0	9	Severe	Severe
23	99	M	16.0	35.0	2.5	2.0	3	Moderate	Moderate
24	85	M	22.0	41.0	2.5	2.0	7	Severe	Moderate
25	74	F	17.0	38.0	2.0	2.5	7	Severe	Severe
26	124	F	21.0	33.5	2.0	2.0	4	Moderate	Normal IQ
27	128	M	19.0	39.0	2.0	2.5	7	Moderate	Normal IQ

Table 2. Thiamine-related variables.

Variable	Healthy control (n = 21)	ASD (n = 27)
Plasma thiamine (nM)	6.82 (4.47–7.02)	6.60 (4.48–8.91)
Plasma TMP (nM)	6.82 (5.84–8.91)	7.00 (5.51–8.55)
Plasma TPP (nM)	9.00 (8.41–10.71)	6.82 (5.81–8.52)**
Urinary thiamine  (nmol/mg creatinine)	1.20 (0.84–1.59)	1.84 (1.23–2.86)
FEthiamine (%)	70.8 (36.3–91.8)	63.4 (43.2–112.8)

Data are median (lower–upper quartile).

Significance: **p < .01 with respect to healthy controls.

Assay of thiamine and phosphorylated metabolites

Plasma and urinary thiamine, thiamine monophosphate (TMP), and TPP were determined by high performance liquid chromatography (HPLC) with fluorimetric detection after pre-column derivatization to the respective thiochromes by potassium ferricyanide under alkaline conditions [15]. HPLC was performed on Dionex Ultimate 3000 HPLC system (Thermo Scientific, Hemel Hempstead, UK). Briefly, plasma or urine, diluted 5-fold in water (50 μl), were mixed with 20% (w/v) trichloroacetic acid (TCA) for de-proteinization; internal standard, chloroethylthiamine (1 μM, 10 μl), was then added and mixed again. Samples were kept on ice for 10 min and then the precipitate was sedimented by centrifugation (6000 × g, 4 °C, 10 min). The supernatant was removed and adjusted to pH 4.5 by addition of 2 M sodium acetate and spin-filtered (0.2 μm, 4000 × g, 4 °C, 10 min). The filtrate (40 μl) was analyzed by HPLC. Chromatographic conditions were optimized for thiochrome recovery. The column used was a 3 × 150 mm C18 column with 3.5 μm particle size with a 3 × 20 mm guard column (Xbridge; Waters, Elstree, UK). Mobile phases: A – 10 mM K₂HPO₄/KH₂PO₄ in water at pH 8.4; B – 10 mM K₂HPO₄/KH₂PO₄ in 50% methanol at pH 8.4; and C – 30% isopropanol in water with 0.1% trifluoroacetic acid. The flow rate was 0.5 ml/min. The elution profile was: 0 min, 95% A + 5% B; 0–20 min, a linear gradient of 5–100% B; 20–32 min 100% C (column washing); and 32–48.5 min 95% A + 5% B (re-equilibration). Samples were derivatized with a NaOH-K₃[Fe(CN)₆] solution prepared and mixed with samples immediately prior to injection. The autosampler was programmed to mix 40 μl 15% NaOH and 10 μl 1% K₃[Fe(CN)₆], then to add 10 μl of this mixture to 40 μl of the sample; finally, 25 μl of the derivatized sample were injected onto the column for each analysis. Thiochromes formed by derivatization were detected by fluorescence spectrophotometry at excitation 365 nm and emission 439 nm. The retention times, limits of detection, interbatch coefficient of variation (CV) values and recoveries for these metabolites were: thiamine 13.1 min, 36 fmol, 1.1 and 97%; TMP 6.0 min, 52 fmol, 2.1 and 92%; and TPP 4.5 min, 51 fmol, 2.9 and 94%. Sample storage studies indicated that analyte content was stable for plasma and urine stored at −80 °C for at least 6 weeks and for plasma, urine and related de-proteinized extracts for 8 h. Stock solutions of thiamine, TMP and TPP were calibrated by spectrophotometry assuming molar extinction coefficients of ɛ₂₃₃=14.2, ɛ₂₄₇=15.3 and ɛ₂₄₇=13.0 mM⁻¹ cm⁻¹, respectively [16]. Shewhart analysis assessing the stability of analyte estimates in samples every day over a period of 10 consecutive days indicated the analysis had acceptable quality control (all estimates within mean ±2 SD) [17]. The urinary excretion of thiamine is given in nmol thiamine per mg creatinine. The fractional excretion of thiamine (FE_thiamine) were deduced as follows: FE_thiamine (%) = 100 × ([Thiamine]_urine/[Thiamine]_plasma)/([Creatinine]_urine/[Creatinine]_plasma).

Assay of markers of oxidation and nitration damage in plasma protein

Markers of oxidative and nitration damage to plasma protein were analyzed by stable isotopic dilution analysis liquid chromatography-tandem mass spectrometry (LC-MS/MS). Plasma protein is washed by diafiltration by 4 cycles of dilution with water and concentration, exhaustive enzymatic hydrolysis and quantitation of oxidized and nitrated amino acids, as described [18]. Analytes determined were: o,o’-dityrosine (DT) – a marker of tyrosine residue oxidation; N-formylkynurenine (NFK) – a marker of tryptophan residue oxidation; and 3-nitrotyrosine (3-NT) – a marker of tyrosine nitration. Estimates were normalized to related unmodified amino acid residues, tyrosine and tryptophan residues, and are presented as mmol analyte/tyr or trp, as appropriate.

Statistical analysis

Data are presented as mean ± SD for parametric distributions and median (lower–upper quartile) for non-parametric distributions. Significance was evaluated by Student’s t-test or by Mann–Whitney U test for parametrically or non-parametrically distributed data, respectively. Correlation analysis was performed by the Spearman’s method. Data were analyzed using SPSS, version 22.0 (IBM Corporation, Armonk, NY).

Results

The plasma concentration (median [lower–upper quartile]) of thiamine and related metabolites in plasma were: thiamine, 6.82 (4.47–7.02) nM; TMP, 6.82 (5.84–8.91) nM; and TPP, 9.0 (8.41–10.71) nM. The plasma concentration of thiamine and TMP were not changed in subjects with ASD whereas the plasma concentration of TPP was decreased by 24% (p < .01). The fractional excretion of thiamine was unchanged in subjects with ASD. No correlations were found between the concentration of thiamine and related metabolites and clinical parameters (Table 2).

Assay of plasma protein markers of oxidative damage revealed an 88% increase of dityrosine residue content in subjects with ASD compared to healthy controls. There was no change in NFK and 3-NT residue contents, suggesting only selected oxidative damage and no increase in protein nitration is present in ASD. In correlation analysis, plasma protein NFK residue content in healthy and ASD subjects combined correlated positively with plasma thiamine concentration (r = 0.38, p < .01) and TMP (r = 0.30, p < .05). In healthy controls, there was no correlation of plasma protein oxidative and nitration damage markers with thiamine-related variables. In subjects with ASD, the correlation of plasma protein NFK residue content with plasma thiamine concentration was maintained (r = 0.52, p < .01) – Table 3.

Table 3. Plasma protein oxidative damage markers.

Oxidation markers	Healthy controls (n = 21)	ASD (n = 27)	p Value
DT (mmol/mol tyr)	0.025 (0.019–0.031)	0.047 (0.035–0.094)	***
NFK (mmol/mol trp)	15.7 (14.0–16.8)	15.0 (14.0–16.1)	NS
3-NT (mmol/mol tyr)	0.0056 (0.0045–0.0070)	0.0053 (0.0045–0.0064)	NS

Data are median (lower–upper quartile).

Significance: ***p < .001 with respect to plasma levels of healthy controls; Mann–Whitney U test.

NS: not significant.

Discussion

No deficiency of thiamine and TMP in plasma of children with ASD was found. A criterion for thiamine deficiency is a urinary excretion of ≤0.20 μmol/24 h for an adult human subject [19] which is equal to a urinary excretion rate of ca. 0.13 nmol per mg creatinine. Spot urinary thiamine/creatinine ratio correlate strongly with 24 h urine excretion [20]. By comparison with urinary excretion of thiamine reported in the present work, it can be concluded that both healthy controls and subjects with ASD are highly replete with thiamine, i.e. that the examined children have sufficient thiamine by this criterion. Urinary thiamine excretion correlates strongly with dietary intake of thiamine [21] and hence both study groups were likely taking in similar amounts of thiamine in their diets.

Subjects with ASD had decreased plasma TPP compared to healthy controls. This may reflect a disturbance in inter-tissue exchange of TPP or of metabolites using similar transporter systems. TPP is formed from thiamine by thiamine pyrophosphokinase inside human cells. TPP permeates the plasma membrane of cells via the reduced folate carrier-1 (RFC-1) transporter [22]. A further TPP-specific transporter in colonic epithelial cells has recently been characterized, which absorbs TPP produced by and released from gut bacteria [23]. Recent genomic studies of the human microbiome suggested thiamine is produced by 40–65% of the 256 human gut microbes [24]. Decreased plasma TPP in children with ASD may reflect decreased release of TPP from cells and/or impaired uptake of TPP from the gastrointestinal tract. The latter provides a link of thiamine metabolism to the gut microbiome, which has recently been considered as potentailly influential in ASD [25]. The significance of the decrease in plasma TPP in children with ASD is unknown and deserves further investgation.

DT residue content of plasma protein was increased in subjects with ASD whereas NFK and 3-NT residue contents remained unchanged. DT residue formation occurs by reaction of tyrosine residues in proteins with reactive oxygen species (ROS) or with dual oxidase (DUOX) [26]. The selective increase in DT may suggest a role of increased DUOX activity in subjects with ASD. DUOX has an important role in gut mucosal immunity, host–microbe homeostasis and signaling for neutrophil recruitment into allergic airways [27,28]. The plasma protein content of NFK residues in ASD has not been reported previously. There was a genetic association of ASD with tryptophan 2,3-dioxygenase (TDO) which forms NFK from tryptophan in the kynurenine pathway of tryptophan metabolism [29]. TDO oxidizes free tryptophan to NFK; formation of NFK residues in proteins is rather mediated by reaction of ROS with tryptophan residues [30]. The positive correlation of plasma protein NFK residue content with thiamine and TMP may reflect an association with decreased ability to metabolize thiamine further to TPP and increased ability to metabolize TPP to TMP, and hence a tendency toward decreased TPP or decreased thiamine functionality in ASD. There was no increase in plasma protein content of 3-NT in subjects with ASD. This is similar to the finding of a previous report [31].

Features of thiamine metabolism and oxidative damage that may be linked to the microbiome, mucosal immunity and host–microbe are interesting outcomes from this study and deserve further investigation.

Disclosure statement

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this article.

Funding

This work received funding from Fondazione Del Monte di Bologna e Ravenna, Italy, from Fondazione Augusta Pini and Istituto del Buon Pastore ONLUS, Bologna, Italy, from Fondazione Nando Peretti, Rome, Italy, and from a legacy of late Ms. Maria Luisa Cimadori, a member of ANGSA (Associazione Nazionale Genitori Soggetti Autistici).