A Non-targeted Proteomics Newborn Screening Platform for Inborn Errors of Immunity

Abstract

Purpose: Newborn screening using dried blood spot (DBS) samples for the targeted measurement of metabolites and nucleic acids has made a substantial contribution to public healthcare by facilitating the detection of neonates with genetic disorders. Here, we investigated the applicability of non-targeted quantitative proteomics analysis to newborn screening for inborn errors of immunity (IEIs).

Methods: DBS samples from 40 healthy newborns and eight healthy adults were subjected to non-targeted proteomics analysis using liquid chromatography-mass spectrometry after removal of the hydrophilic fraction. Subsequently, DBS samples from 43 IEI patients were analyzed to determine whether patients can be identified by reduced expression of disease-associated proteins.

Results: DBS protein profiling allowed monitoring of levels of proteins encoded by 2912 genes, including 1110 listed in the Online Mendelian Inheritance in Man database, in healthy newborn samples, and was useful in identifying patients with IEIs by detecting reduced levels of disease causative proteins and their interacting proteins, as well as cell-phenotypical alterations.

Conclusion: Our results indicate that non-targeted quantitative protein profiling of DBS samples can be used to identify patients with IEIs and develop a novel newborn screening platform for genetic disorders.

Keywords: Dried blood spot; Newborn screening; Non-targeted proteomics.

Fig. 1

Proteomics analysis distinguishes healthy newborn dried blood spot (DBS) samples from those of adults. A Clustered heatmap of Pearson correlation coefficient values between all healthy newborn and adult DBS sample pairs. Dark red denotes higher correlation and dark blue denotes lower correlation (Pearson r = 0.82–0.96). B Principal component analysis of healthy newborn and adult DBS sample pairs. Red denotes newborn (NB) and blue denotes adult DBS samples

Fig. 2

Proteins differentially expressed between healthy newborn and adult samples. A Heatmap of hierarchical clustering of all differentially expressed proteins (DEPs) detected by comparison between healthy newborn and adult samples. Red, upregulated DEPs; blue, downregulated DEPs. B Volcano plot showing DEPs in dried blood spot samples from healthy newborns compared with those from healthy adults. C Protein–protein interaction network analysis illustrating the top 100 upregulated and downregulated proteins (q-value < 0.05), contrasting healthy newborns and adults; network generated in Cytoscape using the STRING database. Node size corresponds to P value, while colors (red and blue) signify DEPs; red and blue circles represent the 'Hemoglobin complex' and 'Complement and coagulation' Gene Ontology (GO) terms, respectively. D GO enrichment analysis of DEPs. Bar charts showing the top 10 GO terms for biological processes upregulated (left) and downregulated (right) in newborn samples relative to adult samples

Fig. 3

Quantification of proteins responsible for various genetic disorders in neonatal dried blood spot samples. A Venn diagram of proteins identified in DBS samples. OMIM, Online Mendelian Inheritance in Man. B Levels of 124 proteins listed in the 2022 International Union of Immunological Societies classifications of inborn errors of immunity in DBS samples from healthy newborns (blue) and healthy adults (red)

Fig. 4

Expression of disease-responsible proteins in samples from patients with inborn errors of immunity. Proteomics evaluation of disease-responsible proteins in samples from healthy newborns (HNB), healthy adults (HA), and patients with IEIs, including: familial hemophagocytic lymphohistiocytosis (FHL) type 2, 3, and 5 (FHL2: n = 2, FHL3: n = 8, and FLH5: n = 1, respectively); p91-phox, p22-phox, and p47-phox deficiencies (n = 6, 1, and 1, respectively); Hermansky-Pudlak syndrome type 2 (HPS2: n = 3); Wiskott-Aldrich syndrome (WAS: n = 5); X-linked agammaglobulinemia (XLA: n = 7); Chédiak-Higashi syndrome (CHS: n = 5); adenosine deaminase deficiency (ADA: n = 2); common γ chain deficiency (IL2RG: n = 2). All data were standardized relative to the expression of β-actin in the same sample. Levels of (A) Munc13-4, (B) Munc18-2, (C) Syntaxin-11, (D) perforin, (E) p91-phox, (F) p22-phox, (G) p47-phox, (H) WASP, (I) BTK, (J) LYST, (K) AP3B1, and (L) ADA in samples from HNB, HA, and patients with IEIs. Open symbols, samples obtained from pre-symptomatic newborn patients with IEIs; half-closed symbols, results under the limit of detection plotted at approximate lower detection limits. Statistical differences among groups were analyzed using one-way ANOVA on ranks (the Kruskal–Wallis test), followed by Dunn’s multiple comparisons. Data represent the mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001

Fig. 5

Identification of marker proteins associated with molecular phenotypes of inborn errors of immunity (IEIs). Gene ontology enrichment analysis of differentially expressed proteins (DEPs) (left) and volcano plots showing DEPs (right) in dried blood spot (DBS) samples from patients with (A) Hermansky-Pudlak syndrome type 2 (HPS2: n = 3), (B) Chédiak-Higashi syndrome (CHS: n = 5), (D) SCID: n = 4, and (F) Wiskott-Aldrich syndrome (WAS: n = 5) relative to samples from healthy newborns (HN) (left) and healthy adults (HA) (right). Levels of (C) cathepsin G, myeloperoxidase, and ELANE, (E) CD3E, and (G) ITGA2B (CD41) and Rab27B in DBS samples from HNB, HA, and patients with IEIs. Statistical differences among groups were analyzed using one-way ANOVA on ranks (the Kruskal–Wallis test), followed by Dunn’s multiple comparisons. Data represent the mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. #, Top Gene Ontology terms commonly altered in patient samples compared to both newborn and adult control groups. Red dots in volcano plots indicate proteins shared in biological pathways commonly altered in patient samples relative to those from HN and HA

Fig. 6

Comparison of proteomics data from patients with inborn errors of immunity with those from healthy newborns and adults. Heatmap showing hierarchical clustering of the levels of 42 proteins that are disease-responsive or associated with molecular phenotypes in dried blood spot samples from healthy newborns, healthy adults, and patients with inborn errors of immunity. Red, relatively higher expression; blue, relatively lower expression