Disruption of melatonin synthesis is associated with impaired 14-3-3 and miR-451 levels in patients with autism spectrum disorders

Autism spectrum disorders (ASD) are characterized by a wide genetic and clinical heterogeneity. However, some biochemical impairments, including decreased melatonin (crucial for circadian regulation) and elevated platelet N-acetylserotonin (the precursor of melatonin) have been reported as very frequent features in individuals with ASD. To address the mechanisms of these dysfunctions, we investigated melatonin synthesis in post-mortem pineal glands - the main source of melatonin (9 patients and 22 controls) - and gut samples - the main source of serotonin (11 patients and 13 controls), and in blood platelets from 239 individuals with ASD, their first-degree relatives and 278 controls. Our results elucidate the enzymatic mechanism for melatonin deficit in ASD, involving a reduction of both enzyme activities contributing to melatonin synthesis (AANAT and ASMT), observed in the pineal gland as well as in gut and platelets of patients. Further investigations suggest new, post-translational (reduced levels of 14-3-3 proteins which regulate AANAT and ASMT activities) and post-transcriptional (increased levels of miR-451, targeting 14-3-3ζ) mechanisms to these impairments. This study thus gives insights into the pathophysiological pathways involved in ASD.

acetylserotonin. One mL of chloroform was added. The mixture was vortexed twice for 15 s to extract the product, [ 3 H]-melatonin, into the chloroform. The organic layer was washed once with borate buffer, and an aliquot (500 µL) was dried and its radioactivity measured.
In order to verify that the activities measured in platelets reflect authentic AANAT and ASMT and not non-specific actions of other N-acetyltransferases and methyltransferases, si-RNA specific for each enzyme (SR300008 and SR300317, Origene; 1 nM final concentrations after transfection by the siTran 1.0 as recommended by Origene) were used.
Total RNA was extracted from plasma samples (haemolysed samples were avoided since haemolysis was reported to alter miR plasma content 9 ) and pineal glands using the mirVana PARIS kit (Ambion, Thermo Fisher Scientific, St-Quentin en Yvelines, France). Spiked-in synthetic C. elegans miRNA controls (cel-miR39, 54 and 238; Qiagen, Courtaboeuf, France) were added to samples for correction of extraction efficiency. After DNAse treatment, RNAs were reverse transcribed with the miScript reverse transcription kit (Qiagen). cDNA was diluted 10-fold before quantitative PCR with the miScript SYBR Green PCR kit (Qiagen). A control without reverse transcriptase and a control without RNA were added to each PCR plate to ensure the absence of contaminating DNA and to check for non-specific amplification, respectively. Expression values were normalized using the mean Ct of the spiked-in controls and calculated with the 2 -Δ Ct formula. Additional technical details have previously been published 10 .

Sequencing of AANAT and ASMT
Primers and PCR conditions for exon amplification and sequencing are listed in Supplementary Table 4. Non-synonymous mutations were confirmed by sequencing an independent PCR product. The segregation of mutations within families was determined when DNA was available for relatives. The nomenclature of genetic variations was determined according to the following reference protein sequences in Ensembl database: ENSP00000370627 (345 amino acids) for ASMT, and ENSP00000376282 (207 amino acids) for AANAT. Exon 6 of ASMT is a LINE insertion and yields an inactive form of the protein, thus rare variants located in exon 6 were not considered as coding mutations.
The functional impact of each non-synonymous coding mutation was assessed both in silico using Polyphen2 algorithm (http://genetics.bwh.harvard.edu/pph2/) and in vitro by targeted mutagenesis, as described previously 11 . A pool-analysis of coding mutation frequencies was performed including our results and studies published and referenced in PubMed, in which a direct sequencing of all exons of ASMT or AANAT was performed in individuals with ASD and/or in controls.

Primers and conditions for ASMT CNV breakpoint determination (GenomeWalker experiment)
The partial duplication of ASMT gene was first screened by MLPA as previously described 13 . Four independent DNA samples positive for MLPA screening were then subjected to a genome walker experiment for CNV mapping, using Genome Walker Universal kit (Clontech, USA) according to the manufacturer's instructions. Briefly, DNAs were submitted to digestion by restriction enzymes (EcoRV, PvuII, StuI, SspI; one reaction for each enzyme and each DNA), followed by ligation of adapters.
Nested PCRs were performed on the ligation products with the provided adaptor primers, together with gene-specific primers. (d) platelet ASMT data previously published 14 , (e) independent platelet ASMT data collected for this study.
The control groups were compared to other groups using the Wilcoxon two-sample test.
Boxes indicate medians and quartiles.