Clinical and molecular overlap between nucleotide excision repair (NER) disorders and DYRK1A haploinsufficiency syndrome

Abstract

Nucleotide excision repair (NER) disorders are genetic conditions caused by defects in the pathway responsible for repairing DNA lesions due to UV radiation. These defects lead to a variety of heterogeneous disorders, including Cockayne syndrome (CS) and trichothiodystrophy (TTD). In this study, we report 11 patients initially suspected of having CS or TTD who were ultimately diagnosed with DYRK1A haploinsufficiency syndrome using high-throughput sequencing. Comparing clinical presentations, we observed that DYRK1A symptoms overlapped with CS, with shared features such as intellectual disability and microcephaly, systematically present in both disorders and other common symptoms including feeding difficulties, abnormal brain imaging, ataxic gait, hypertonia, and deep-set eyes. However, distinctive features of DYRK1A syndrome, such as severely impaired language, febrile seizures, and autistic behavior or anxiety, helped differentiate it from CS, which typically manifests with severe growth delay, bilateral cataracts, and pigmentary retinopathy. Among the cohort, three patients carried novel DYRK1A variants, including two truncating and one in-frame variant p.Val237_Leu241delinsGlu whose pathogenicity have been confirmed through functional analysis of DYRK1A protein. While previous research has implicated DYRK1A in DNA repair, with DYRK1A being one of the most downregulated genes in CS cells, our study found that DYRK1A patient-derived cell lines did not exhibit NER defects and did not share the CS transcriptomic signature. These findings suggest that if clinical symptoms overlap stems from common molecular disruptions, DYRK1A is involved downstream of the CS genes. This research highlights the importance of considering DYRK1A haploinsufficiency syndrome in the differential diagnoses for NER disorders.

Keywords: Cockayne syndrome; DYRK1A gene; ERCC6/CSB gene; ERCC8/CSA gene; nucleotide excision repair (NER); trichothiodystrophy.

Figure 1

Effect of the p.(Val237_Leu231delinsGlu) (Val237delins); Val237delins) variant on DYRK1A protein level, cellular localization and autophosphorylation. (A) Level of mutant DYRK1A proteins expressed in HeLa, HEK293 and COS cells transiently transfected with DYRK1A constructs containing wild-type (WT), Trp345Ter or Val237delins cDNA sequences. Protein levels were normalized on the level of GFP protein (expressed from a cotransfected pEGFP plasmid). Quantifications were performed on a total of n = 9 series of cells (n = 3 Hela cells, n = 3 HEK293 and n = 3 COS-7 cells) using ImageJ software. One-way ANOVA with multiple comparison test was performed to compare the level of mutant DYRK1A proteins to the level of WT DYRK1A protein, applying Bonferroni’s correction: ns: not significant; ***p < 0.001; error bars represent SEM, standard error of the mean. (B) Cellular localization of DYRK1A WT and mutant proteins observed in HeLa cells after overexpression of FLAG-tagged wild-type and mutant DYRK1A constructs (n = 3 series of Hela cells; 50 cells minimum counted per series; the total number of cells counted per condition is indicated in the graph; scale bars of illustrative images correspond to 10 μm). Three types of cellular localization of DYRK1A protein have been observed: DYRK1A located mostly in the nucleus (blue on the graph), in both nucleus and cytoplasm (orange), or mostly in the cytoplasm (purple). Scale bars: 50 μm. Chi-square test was performed to compare localization of mutants and wild-type DYRK1A proteins, ns: not significant; ***p < 0.001; error bars represent SEM, standard error of the mean. Immunofluorescence images illustrate the data (DAPI-labeled nucleus, flag-labeled DYRK1A and merge). (C) Measure of DYRK1A autophosphorylation on its Tyr321 tested in HEK293 cells (n = 3) transfected with constructs containing WT and Val237delins cDNA sequences, as well two other variants, Ala341Ser which is a benign variant from gnomAD not supposed to affect DYRK1A kinase activity and Ser311Phe which is a kinase-dead variant. An immunoprecipitation with anti-DYRK1A antibody was followed by an immunoblot using anti-phospho-HIPK2 as described in Widowati et al. and Courraud et al. DYRK1A phospho-Tyr321 levels were normalized on DYRK1A total level. The percentage of phosphorylated form for the different mutants compared to the WT proteins is indicated under the blot. One-way ANOVA test was performed to compare mutants to wild-type DYRK1A proteins. ns, not significant; ***p < 0.001; error bars represent SEM, standard error of the mean.

Figure 2

Molecular characterization of DYRK1A patients cell lines compared to Cockayne syndrome patients cell line. (A) Recovery of RNA synthesis (RRS) measured 23 h after UV irradiation in fibroblasts from patients 1, 2, 4 and 5, compared to control cell line (individual non affected with CS) and CSA-mutated cell line (CS). Error bars represent SEM, standard error of the mean. (B) Unscheduled DNA synthesis (UDS) measured 23 h after UV irradiation in fibroblasts from patients 1, 2, 4 and 5, compared to control cell line (individual non affected with CS) and XPF-mutated cell line (XP). A.U.: arbitrary units. (C–E) Study of CS transcriptomic signature and ATF3 expression in DYRK1A patients 1, 2, 4 and 5 compared to controls, CSA and CSB patients. Expression level of CDK5RAP2, NIPBL, DYRK1A, NRG1and ATF3 genes was studied by quantitative RT-PCR in the different cell lines 24 h hours after treatment with 20 J/m² UV-C. Gene expression at 24 h was normalized to 0 h and to GAPDH expression. Error bars represent standard deviation. T-test was performed to compare the expression level of each gene in CSA, CSB and DYRK1A patients to the expression level in control cell lines, applying Welch’s correction: ns: not significant; *p < 0.005;***p < 0.001.