The GATAD2A protein is part of the NuRD complex, which is a multiprotein complex that is intrinsically involved in gene regulation, cell-cycle progression, genome integrity, and cellular differentiation. One basic role GATAD2A has (similar to GATAD2B) is as a bridge between NuRD’s CHD paralogs (where chromatin remodeling activity resides) and the Histone Deacetylase Core (via NuRD’s MBD paralogs). Disruption of this bridging role likely disrupts NuRD function and decreases NuRD activity.
GARND is due to this decreased NuRD activity altering gene expression and regulation. As GATAD2A is likely the predominant GATAD2 paralog in early development, this may lead to severe disruptions in embryogenesis that may interfere with development. Whether this causes embryonic demise is unclear, although the extremely rare incidence of GARND may lend credence to this concept.
GATAD2A is also present during neurogenesis and corticogenesis and although it is not the predominant GATAD2 paralog making up NuRD complexes during those periods (compared to GATAD2B), it likely also has an effect on brain development as seen in patients with global delays and structural brain anomalies.
Some GARND children have some phenotypic findings that are similar to another NuRDopathy, GAND, which is associated with GATAD2A’sparalog, GATAD2B. GATAD2A and GATAD2B proteins are mutually exclusive paralogs in NuRD complexes forming different NuRD subtypes (GATAD2A-NuRD and GATAD2B-NuRD, respectively).
Whether these GATAD2 paralogs have redundant and/or exclusive functions is under study, but their phenotypic similarities indicate there are likely some overlapping roles.
The GATAD2Avariants in the reported cohort consisted of two missense (c.1259 G>A, p.C420Y; c.626C>T, p.T209I), another variant is likely a splice site as it is on the last nucleotide of an exon (c.1205G>T, G402V); one was nonsense (c.673C>T, p.R225*) and one was frameshift (c.1877delT, p.I627Tfs) at the extreme C terminus.
Some of these variants correspond to pathogenic variants seen in GATAD2Band GAND. The missense variant (c.1259 G>A, p.C420Y) is at a homologous site in a CR2 zinc finger seen in several patients with GAND and did not interact with CHD proteins. Another variant (c.1205G>T, G402V) is also at a homologous site of GATAD2Bpathogenic variants as well. The nonsense variant (c.673C>T, p.R225*) likely causes nonsense-mediated decay (NMD), similar to the many nonsense/frameshift variants reported in GAND. Interestingly, the frameshift (c.1877delT, p.I627Tfs) is in the final exon and likely does not cause NMD, further investigations found a conserved identical motif in both GATAD2Aand GATAD2Bin this region. Interestingly, the affected child was the most distinct phenotypically of this diverse group. This conserved identity likely indicates this region is important in GATAD2 paralog function and/or interactions with other proteins.
All of this information indicate that although these variants produce diverse phenotypes, these variants are likely pathogenic and involved in GARND. Future research dissecting the regions of GATAD2Aand its clinical counterpart are ongoing.
GATAD2A