Decreased phenol sulfotransferase activities associated with hyperserotonemia in autism spectrum disorders

Abstract

Hyperserotonemia is the most replicated biochemical abnormality associated with autism spectrum disorders (ASD). However, previous studies of serotonin synthesis, catabolism, and transport have not elucidated the mechanisms underlying this hyperserotonemia. Here we investigated serotonin sulfation by phenol sulfotransferases (PST) in blood samples from 97 individuals with ASD and their first-degree relatives (138 parents and 56 siblings), compared with 106 controls. We report a deficient activity of both PST isoforms (M and P) in platelets from individuals with ASD (35% and 78% of patients, respectively), confirmed in autoptic tissues (9 pineal gland samples from individuals with ASD-an important source of serotonin). Platelet PST-M deficiency was strongly associated with hyperserotonemia in individuals with ASD. We then explore genetic or pharmacologic modulation of PST activities in mice: variations of PST activities were associated with marked variations of blood serotonin, demonstrating the influence of the sulfation pathway on serotonemia. We also conducted in 1645 individuals an extensive study of SULT1A genes, encoding PST and mapping at highly polymorphic 16p11.2 locus, which did not reveal an association between copy number or single nucleotide variations and PST activity, blood serotonin or the risk of ASD. In contrast, our broader assessment of sulfation metabolism in ASD showed impairments of other sulfation-related markers, including inorganic sulfate, heparan-sulfate, and heparin sulfate-sulfotransferase. Our study proposes for the first time a compelling mechanism for hyperserotonemia, in a context of global impairment of sulfation metabolism in ASD.

Fig. 1. Serotonin metabolism.

Serotonin is synthesized from tryptophan (red steps) by tryptophan hydroxylases (TPH1, non-neuronal, and TPH2, neuronal) and aromatic amino-acids decarboxylase (AADC). Of the two catabolic pathways (blue steps), the major one sequentially involves monoamine oxidase A (MAO-A) and aldehyde dehydrogenase (ALDH) or reductase (ALDR). Sulfoconjugation by phenol sulfotransferases (PST) is a quantitatively minor pathway. Alternatively, tryptophan can enter the kynurenine pathway after oxidation by indoleamine-2,3-dioxygenase (IDO) or tryptophan 2,3-dioxygenase (TDO), or serotonin be converted into melatonin by arylalkylamine N-acetyltransferase (AANAT) and acetylserotonin methyltransferase (ASMT).

Fig. 2. Phenol sulfotransferase activities in blood, post-mortem pineal glands, and ileal samples from patients with ASD.

a–c Blood samples were taken in the morning from 97 unrelated individuals with ASD, their first-degree relatives (138 parents and 56 unaffected sibs), and 106 controls. a Platelet PST-M activities. b Platelet PST-P activities. c, d Pineal gland samples were obtained from 9 patients with ASD and 22 age- and sex-matched controls. e, f Ileal samples were obtained from 13 patients with ASD and 11 matched controls. Boxes indicate medians and quartiles. Dashed lines indicate the threshold of the 5th percentiles of the control group. Groups were compared using the Wilcoxon two-sample test.

Fig. 3. Relationship between PST activity and whole blood serotonin.

a Whole blood serotonin. b, c Correlation between whole-blood serotonin concentration and platelet PST-M activity in b patients with ASD and c their relatives and controls (Pearson’s test after log transformation). d–f Platelet PST-M and PST-P activities and blood serotonin according to intellectual disability defined as verbal and performance IQ < 70. g Blood serotonin measured in wild type FVB/N mice (WT), in mice knockout for sult1a1 gene (sult1a1−), in mice transgenic for human SULT1A (tg hSULT1A1/1A2) and in wild-type FVB/N mice treated with 17α-ethinylestradiol (WT FVB/N + EE2), n = 5 in each group. Boxes indicate medians and quartiles. Groups were compared using the Wilcoxon two-sample test.

Fig. 4. Sulfation metabolism in ASD.

a Plasma inorganic sulfate concentrations. b Plasma HS concentrations. c Pineal HS contents. d Pineal heparan sulfate sulfotransferase activities, stratified by time of death (left panel) and pooled for comparison (right panel). e Ileal heparan sulfate contents. f Ileal heparan sulfate sulfotransferase activities. Boxes indicate medians and quartiles. Groups were compared using the Wilcoxon two-sample test.