Untargeted metabolomic, and proteomic analysis identifies metabolic biomarkers and pathway alterations in individuals with 22q11.2 deletion syndrome

Abstract

Introduction: The chromosome 22q11.2 deletion syndrome (22q11.2DS) is characterized by a well-defined microdeletion and is associated with a wide range of brain-related phenotypes including schizophrenia spectrum disorders (SCZ), autism spectrum disorders (ASD), anxiety disorders and attention deficit disorders (ADHD). The typically deleted region in 22q11.2DS contains multiple genes which haploinsufficiency has the potential of altering the protein and the metabolic profiles.

Objectives: Alteration in metabolic processes and downstream protein pathways during the early brain development may help to explain the increased prevalence of the observed neurodevelopmental phenotypes in 22q11.2DS. However, relatively little is known about the correlation of dysregulated protein/metabolite expression and neurobehavioral impairments in individuals who developed them over time.

Methods: In this study, we performed untargeted metabolic and proteomic analysis in plasma samples derived from 30 subjects including 16 participants with 22q11.2DS and 14 healthy controls (TD) enrolled in a longitudinal study, aiming to identify a metabolic and protein signature informing about the underlying mechanisms involved in disease development and progression. The metabolic and proteomic profiles were also compared between the participants with 22q11.2DS with and without various comorbidities, such as medical involvement, psychiatric conditions, and autism spectrum disorder (ASD) to detect potential changes among multiple specimens, collected overtime, with the aim to understand the basic underlying mechanisms involved in disease development and progression.

Results: We observed a large number of statistically significant differences in metabolites between the two groups. Among them, the levels of taurine and arachidonic acid were significantly lower in 22q11.2DS compared to the TD group. In addition, we identified 16 proteins that showed significant changes in expression levels (adjusted P < 0.05) in 22q11.2DS as compared to TD, including those involved in 70 pathways such as gene expression, the PI3K-Akt signaling pathway and the complement system. Within participants with 22q11.2DS, no significant changes in those with and without medical or psychiatric conditions were observed.

Conclusion: To our knowledge, this is the first report on plasma metabolic and proteomic profiling and on the identification of unique biomarkers in 22q11.2DS. These findings may suggest the potential role of the identified metabolites and proteins as biomarkers for the onset of comorbid conditions in 22q11.2DS. Ultimately, the altered protein pathways in 22q11.2DS may provide insights of the biological mechanisms underlying the neurodevelopmental phenotype and may provide missing molecular outcome measures in future clinical trials to assess early-diagnosis treatment and the efficacy of response to targeted treatment.

Keywords: 22q11.2 deletion syndrome; APS; AS; Biomarker; Metabolomics; Pathways; Proteomics.

Fig. 1

a Multiomics analysis with the Omni-MS workflow. Plasma samples are randomized and aliquoted in for preparation, undergoing solvent denaturation, metal chelation, and trypsin digestion. The undigested debris is solvent precipitated and clarified with centrifugation. The multiomic extract is injected into the LC–MS instrument, separated using reverse phase and HILIC chromatography, ionized by ESI, and data collected by high resolution MS1 scans and data-dependent MS2 scans. The ddMS2 data is used to determine identifications, which are in turn matched between runs to generate label-free quantification between samples. b Multiomic LC–MS separation. The figure shows the total ion chromatogram (TIC) visualization for a representative plasma sample preparation containing polar metabolites, tryptic peptides, and lipids after analysis via the RP-HILIC LC–MS system. The y-axis shows the relative total ion intensity of eluting components (metabolites, tryptic peptides, and lipids) as a function of retention time (x-axis)

Fig. 2

Differential metabolite levels between 22q11.2DS and the groups. Volcano plot of differential expression results comparing metabolite expression in the 22q11.2DS and the TD subjects at baseline. The x-axis shows the log2 fold change for 22q11.2DS/TD and the y-axis shows − log10 (raw P value). The name of the 8 most significant metabolites is indicated in blue and these plots are generated using R software

Fig. 3

Taurine and arachidonic acid levels between TD and 22q11.2DS groups. Box plots showing decreased levels of taurine (a) and of arachidonic acid (b) in 22q11.2DS as compared to TD. The heavy line in each box represents the median, the lower and upper box edges represent the 25th and 75th percentiles, respectively, and the lower and upper whiskers represent the smallest and largest observations, respectively

Fig. 4

Differential proteins levels between 22q11.2DS and the TD groups. a The volcano plot was generated using R software. Volcano plot of differential expression results comparing protein expression in 22q11.2DS and TD subjects at baseline. The x-axis shows the log2 fold change for 22q11.2DS /TD and the y-axis shows − log10 (raw P value). The name and formula of the 5 most significantly proteins are shown in blue. b Heatmap of proteins, lipids, and metabolites that are differentially expressed (adjusted P < 0.05) between 22q11.2DS and TD subjects at baseline. Rows are sorted based on the hierarchical clustering dendrogram shown on the left-hand side and columns are sorted by subject diagnosis

Fig. 5

Altered protein pathways observed between the 22q11.2DS and the TD groups. Barplot of top 10 most significantly enriched KEGG pathways (a) and Reactome pathways, (b) from enrichment analysis of differential expression of proteins between 22q and TD at baseline. The length of the bars shows − log10 (P value) from the enrichment analysis and the color of the bars shows the number of proteins included in the DE analysis (regardless of significance) in the indicated pathway

Fig. 6

Altered pathways between 22q11.2DS and TD including representative genes in the complement cascade, coagulation and Ca2 + signaling pathways, other psychiatric and autism altered protein datasets, and some of their known. The figure shows selected enriched genes identified by multiple pathway databases, including Reactome, Bioplanet and Wikipathways. Enrichment analysis was generated using the Epistemic AI platform