The UBE3B Gene and Protein
The UBE3Bgene (OMIM *608047) is located on chromosome 12q24.11, spans 28 exons, and encodes a 1,068 amino acid E3 ubiquitin-protein ligase. The protein contains an N-terminal IQ motif (approximately residues 28–50), a short calmodulin-binding motif thought to mediate protein-protein interactions, and a C-terminal HECT domain (approximately residues 690–1,068) that mediates the catalytic transfer of ubiquitin from an E2-conjugating enzyme to substrate proteins. Alternative splicing of UBE3Bgenerates at least two transcripts: one encoding the full HECT domain, and a second, shorter isoform that lacks it.
Type of Mutations
Reported pathogenic variants in UBE3Bare diverse and include:
• Nonsense (stop-gain) variants
• Frameshift variants
• Canonical splice-site variants
• Missense variants, predominantly affecting the HECT domain
Mutational Spectrum: High Allelic Heterogeneity, with One Notable Exception
Unlike some recessive neurodevelopmental genes where a single ancestral mutation accounts for most cases, UBE3Bshows considerable allelic heterogeneity; overall, different families tend to carry their own private mutations, scattered across the coding sequence. The great majority of variants – whether truncating or missense – map to, or upstream of, the HECT domain, and are predicted to abolish the catalytic ubiquitin ligase activity of the protein.
A clear genotype-phenotype trend has emerged: patients with protein-truncating variants, or with missense substitutions located within the HECT domain itself, tend to show the full, severe phenotype. In contrast, the rare missense variants identified outside the HECT and IQ domains have been associated with comparatively milder craniofacial dysmorphism, although global developmental impairment remains present.
That said, one specific nonsense variant stands out: c.1076G>A (p.W359*) was identified by Zaki and colleagues (2020) in 3 of 9 unrelated Egyptian patients from 2 of 6 families (Families 1 and 6). The largest single Egyptian cohort reported to date. All three patients shared additional synonymous, missense, and intronic variants flanking UBE3B, consistent with a shared haplotype rather than two independent mutational events. The authors explicitly raised this as evidence of a likely founder mutation in this population, originating from the same governorate in Upper Egypt. Whether this represents a single shared ancestral haplotype has not yet been formally tested with broader haplotype analysis across additional families, but its recurrence in this cohort is clinically useful: it may be reasonable to test for this specific variant as a first step in Egyptian patients, particularly from Upper Egypt, with a compatible phenotype.
Two other recurrent variants merit mention: c.61G>T (p.E21*) has been reported in unrelated families of Mexican and Hispanic-American origin, and c.1166G>A (p.W389*) has been identified independently in two unrelated Italian patients, raising the possibility of further population-specific founder effects or mutational hotspots. However, neither has been formally confirmed with haplotype analysis.
Summary of reported UBE3B variants by publication
The table below lists the distinct pathogenic variants reported across the literature reviewed for this page, to give an overview of the mutational spectrum (variants reported in more than one unrelated family are marked with an asterisk):
Basel-Vanagaite et al., 2012 (4 patients, 3 families): c.1741+2T>C (splice); c.545-2A>G (splice); c.2223_2224delAG, p.R741Sfs*3 (frameshift); c.2180A>C, p.Q727P (missense)
Flex et al., 2013 (2 patients): c.556C>T, p.R186* (nonsense); c.1166G>A, p.W389* (nonsense)*
Basel-Vanagaite et al., 2014 (6 patients, 5 families):c.61G>T, p.E21* (nonsense)*; c.1445T>A p.L482H + c.1616T>C p.L539P (compound missense, same patient); c.2990G>C, p.R997P (missense)*; c.2335G>A, p.G779R (missense); c.2098C>T, p.Q700* (nonsense) + c.2990G>C (compound het.)
Pedurupillay et al., 2015 (2 sisters): c.1A>G, p.M1? (initiation codon); c.1773delC, p.F591fs (frameshift)
Yilmaz et al., 2018 (4 patients): c.1166G>A, p.W389* (nonsense)* + c.2254-2A>T (splice, compound het.); c.1856+1G>T (splice); c.886C>T, p.R296* (nonsense) + c.1856+1G>A (splice, compound het.); c.2624dupA, p.N875Kfs*876 (frameshift) + c.3014A>C, p.E1005P (missense, compound het.)
Galarreta et al., 2019 (5 patients, 3 families):c.2569-1G>C (splice); c.61G>T, p.E21* (nonsense)*, found in 3 of the 5 patients in this report; c.518C>A, p.S173* (nonsense)
Zaki et al., 2020 (9 patients, 6 families):c.1076G>A, p.W359* (nonsense)* (the candidate Egyptian founder variant, found in 3 patients/2 families); c.1629dupT, p.D544* (nonsense); c.1282+1G>A (splice); c.3015+1G>A (splice); c.2364G>C, p.E788D (missense)
Fida et al., 2023 (1 patient):c.469C>T, p.R157* (nonsense)
Abdelfattah & Abu Saleh, 2026 (1 patient):c.1741+2T>C (splice), same variant as Basel-Vanagaite et al. 2012, first comprehensive clinical description of this specific change
Across these reports, roughly 25 distinct pathogenic variants have been described, spanning nonsense, frameshift, splice-site, initiation-codon, and missense categories, with the great majority clustering at or upstream of the HECT domain.
Pathophysiologic Mechanism
Loss of UBE3B ubiquitin ligase activity is thought to disrupt the timely degradation or trafficking of as-yet incompletely defined substrate proteins during brain and craniofacial development. Ube3b knockout mice recapitulate several key aspects of the human disorder, including reduced body and brain , reduced grip strength, mild hearing impairment, and reduced cholesterol and cholesterol-precursor (lathosterol) levels, supporting hypocholesterolemia as a genuine downstream consequence of UBE3B deficiency rather than a dietary artifact. The probable Caenorhabditis elegans ortholog, oxi-1, has been shown to participate in ubiquitin-proteasome-mediated protein turnover, particularly under conditions of oxidative stress, suggesting a conserved protective function for this branch of the ubiquitin pathway across species.
Diagnostic Testing
Diagnosis is established through identification of biallelic (homozygous or compound heterozygous) pathogenic variants in UBE3Bby Sanger sequencing of the coding exons, targeted gene panels, or, increasingly, whole-exome or whole-genome sequencing – the latter now responsible for the majority of recently reported diagnoses. Unlike some other genes in this series, no pseudogene or large genomic rearrangement complicating UBE3Btesting has been reported. In Egyptian patients, particularly those from Upper Egypt, targeted testing for the recurrent variant c.1076G>A (p.W359*) may be a reasonable first-line approach.
Figure 3: Schematic of the UBE3Bgene structure (28 exons) and protein domains (IQ motif, HECT domain), with mutation positions plotted and the Egyptian founder variant c.1076G>A (p.W359*) highlighted.
