ANKRD11 variants: KBG syndrome and beyond

Mutations affecting the transcriptional regulator Ankyrin Repeat Domain 11 (ANKRD11) are mainly associated with the multisystem developmental disorder known as KBG syndrome, but have also been identified in individuals with Cornelia de Lange syndrome (CdLS) and other developmental disorders caused by variants affecting different chromatin regulators. The extensive functional overlap of these proteins results in shared phenotypical features, which complicate the assessment of the clinical diagnosis. Additionally, re‐evaluation of individuals at a later age occasionally reveals that the initial phenotype has evolved toward clinical features more reminiscent of a developmental disorder different from the one that was initially diagnosed. For this reason, variants in ANKRD11 can be ascribed to a broader class of disorders that fall within the category of the so‐called chromatinopathies. In this work, we report on the clinical characterization of 23 individuals with variants in ANKRD11. The subjects present primarily with developmental delay, intellectual disability and dysmorphic features, and all but two received an initial clinical diagnosis of either KBG syndrome or CdLS. The number and the severity of the clinical signs are overlapping but variable and result in a broad spectrum of phenotypes, which could be partially accounted for by the presence of additional molecular diagnoses and distinct pathogenic mechanisms.


| INTRODUCTION
Transcriptional regulators are key players in numerous biological processes. Ankyrin Repeat Domain 11 (ANKRD11) is an important coregulator able to induce changes in gene expression by recruiting chromatin remodelers to target genes upon interaction with specific transcriptional repressors or activators. 1,2 The corresponding gene (OMIM *611192) is located at 16q24.3 and encodes a 298 kDa protein of 2663 amino acids containing five ankyrin repeats (amino acids 162-284), two repression domains (amino acids 318-611 and 2369-2663) and one activation domain (amino acids 1851-2145). 1 Due to its unique structure, ANKRD11 is believed to mediate both transcriptional activation and repression. 1,3 ANKRD11 is best characterized for its function as a co-regulator in the developing brain, where it plays a critical role for the proliferation of neural progenitors, for the genesis and positioning of newborn neurons, 4 for neuronal plasticity 5 and for dendritic differentiation. 6 ANKRD11 was first associated with human disease when deletions at 16q24.3 were identified in individuals with autism spectrum disorder (ASD). 7 Two years later, Willemsen and colleagues provided evidence for a novel microdeletion syndrome by describing four patients characterized by ASD, variable levels of intellectual disability and dysmorphic features carrying interstitial deletions at 16q24.3. 8 Subsequent reports of individuals with intellectual disability, facial dysmorphism, and ASD allowed the narrowing of the minimal common region of overlap of this 16q24.3 microdeletion syndrome to ANKRD11 only, suggesting a role of ANKRD11 in neurodevelopment. 9,10 The first point mutations in ANKRD11 were identified in seven individuals with KBG syndrome (KBGS, OMIM #148050). 5 This is a rare disorder named after the initials of the first three affected individuals and characterized by intellectual disability, global developmental delay, short stature, skeletal anomalies, distinctive facial features, and macrodontia of the upper central incisors. 11 Since the first description by Sirmaci and colleagues, 5 additional individuals with KBGS have been reported to carry point mutations, duplications or microdeletions involving ANKRD11, thus pointing to ANKRD11 as the main gene responsible for this syndrome. [12][13][14][15][16][17][18][19][20][21][22][23][24][25][26][27] Importantly, a marked interfamilial and intra-familial phenotypical variability has been reported in association with KBGS, indicating variable expressivity and penetrance. 11,20 With the falling cost and increasing accessibility of next generation sequencing technologies and microarrays, variants in ANKRD11 have also been reported in association with neurodevelopmental syndromes other than KBGS. Specifically, an individual with an initial clinical diagnosis of Silver-Russell syndrome was found to harbor a 348 kb microdeletion at 16q24.3 encompassing ANKRD11 and SPG7. 28 Point mutations in ANKRD11 were also identified in subjects with phenotypes reminiscent of Cornelia de Lange syndrome (CdLS) (OMIM #122470). [29][30][31][32] Loss-of-function variants in ANKRD11 were similarly described in association with Coffin-Siris syndrome (CSS) (OMIM #135900). 23 Importantly, CdLS and CSS clinically overlap to some extent with KBGS. The shared clinical features include a variable degree of developmental delay and intellectual disability, growth retardation, limb anomalies and characteristic facial dysmorphism. 23,29,30 These findings suggest that variants in ANKRD11 are not necessarily associated with KBGS only, but that they are rather linked to a larger spectrum of neurodevelopmental syndromes.
Accordingly, ANKRD11 has been described as one of the most frequently mutated genes in individuals with severe developmental disorders. 33,34 In this work we discuss the clinical and molecular findings of 23 individuals with variants in ANKRD11 and describe a wide spectrum of phenotypes associated with mutations in this gene.

| Cohort
Individuals herein described were recruited thanks to a large international cooperation that includes Germany, Italy, Ireland, Colombia, Canada, and the United States.
Procedures including subjects were initially approved by the Ethical Committee of the University of Lübeck (approval number for human studies HL07-158) and the Ethical Committees of the respective institutions. All procedures were performed in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent was obtained from all individuals included in this study. An additional informed consent was collected for the publication of subjects' photographs.
Individuals were analyzed by means of exome sequencing, gene panels or microarrays at their respective institutions. Referring physicians provided detailed developmental, neurological, and behavioral history of the subjects. Variants were described on the ANKRD11 NM_013275.6 RefSeq transcript using HGVS recommendations. 35 All variants have been submitted to the ClinVar database and have been assigned the following accession numbers: SCV001478030-SCV001478045.

| Facial dysmorphology novel analysis
The Facial Dysmorphology Novel Analysis (FDNA Inc., Boston, MA) technology combines facial recognition software with biological knowledge. This technology allows to detect dysmorphic features and recognizable patterns of facial malformations from 2D facial photographs. Face2Gene (FDNA) was used as a tool for computer analysis of subjects' photographs (https://face2gene.com). 36 3 | RESULTS

| Individuals
The group of individuals described herein is composed of 11 females and 12 males, with an age range extending from four to 23 years. The median age of the initial clinical diagnosis was 6 years and 5 months, whereas the median age of the last clinical examination was 9 years and 3 months. Phenotypical appearance of the individuals can be found in Figure 1.
Thirteen individuals received a clinical diagnosis of KBGS (Individuals 1, 8, 10, 11, 14, 15, 17, 18, 19, 20, 21, 22, and 23). Three individuals received an initial clinical diagnosis of CdLS or atypical CdLS that was reconfirmed at a later re-evaluation (Individuals 2, 5, and 6), while five individuals were diagnosed as CdLS during early childhood but were reclassified as KBGS after a re-evaluation later in life ( Individuals 3,4,7,9,and 12). Two individuals presented with nonspecific syndromic intellectual disability and developmental delay (Individuals 13 and 16). A schematic representation of the diagnostic evaluation of the individuals of the present series is provided in the Figure S1.

| Clinical features
An overview of the clinical features of each individual is listed in Table 1.
Milestones in motor and verbal development were found to be delayed for all individuals but one: sitting independently was achieved at a median age of 12 months and walking independently at 24 months. The median age of pronouncing the first words was 24 months. Individuals 11, 13, and 21 are currently still non-verbal at an age of 6, 4, and 15 years, respectively.
Intellectual disability and behavioral problems were also detected in the large majority of the individuals (85% and 68% of subjects, respectively). The level of intellectual disability could be assessed in four individuals and appeared moderate in one and mild in three subjects. Behavioral problems ranged from shyness or inability to recognize and respect personal boundaries to aggressiveness, autistic features and attention deficit hyperactivity disorder.

| Clinical re-evaluation and age-dependent phenotypical evolution
Five individuals of the present cohort were diagnosed as CdLS during early childhood but were reclassified as KBGS after phenotypical re-evaluation (Subjects 3, 4, 7, 9, and 12). Table 2 provides an

| Molecular findings
Seventeen distinct ANKRD11 variants were identified in our cohort com-   20 Importantly, the possibility of multiple molecular diagnoses should also be taken into account for an appropriate evaluation of these phenotypes, as they can influence the phenotypical appearance. 45 The majority of our individuals was analyzed by targeted gene panel and we are therefore unable to exclude the presence of additional variants.
Variants in ANKRD11 have been mainly described in association with KBGS. 5,19,20,27 Accordingly, based on a recent review, 171 out of 228 individuals with ANKRD11 variants reported in 38 different studies were formally diagnosed as KBGS. 34  The CSS-associated SWItch/Sucrose Non-Fermentable (SWI/SNF) complex is well known for chromatin remodeling. 47 Notably, the SWI/SNF complex is known to directly interact with the cohesin loader. In yeast, the SWI/SNF complex recruits the cohesin loader to nucleosome-free regions that the cohesin loader subsequently helps to maintain. 51 A direct interaction between ANKRD11 and these two protein complexes has instead not been reported yet. The substantial functional overlap that characterizes these proteins as well as the direct physical interactions that have been reported for some of them could be found accountable for the shared clinical features observed in individuals with different neurodevelopmental disorders. The increasing accessibility of next generation sequencing technologies will allow the identification of additional variants in ANKRD11 in individuals with clinical diagnoses different from KBGS. Correspondingly, several variants in ANKRD11 were identified in individuals with severe undiagnosed developmental disorders and/or intellectual disability. 33,34,52 Also in this cohort, variants in ANKRD11 have been reported in KBGS subjects but also in individuals with nonspecific intellectual disability or with CdLS. For this reason, variants in this gene should be ascribed to a more ample group of neurodevelopmental disorders that fall within the category of chromatinopathies rather than to KBGS per se. 53,54 In line with the recently proposed dyadic approach for the description of diagnostic entities, 55 the disease phenotypes observed in association with variants in ANKRD11 could be defined as "ANKRD11-related chromatinopathies" or "ANKRD11-related neurodevelopmental disorders." The growing number of variants in ANKRD11 and the varying severity of behavioral and developmental phenotypes associated with these variants make the understanding of the causative mechanisms particularly important. Since the levels of ANKRD11 are tightly regulated during the cell cycle, the most likely pathogenic mechanism is haploinsufficiency. 18 However, analysis of ANKRD11 transcript levels in cell lines of patients has revealed escape from nonsense-mediated mRNA decay to some extent. 15,18 Furthermore, some variants might also lead to a dominant negative pathogenic mechanism due to a lack of proteasome-mediated degradation of the truncated protein. This proposed mechanism depends on the dimerization between the Nterminal repression domains of the wild type and mutant ANKRD11 when the degradation of the mutant protein is impaired by the disruption of the C-terminal degradation signals. 18 The potential that a greater understanding of the molecular mechanism may lead to eventual therapeutic insights represents an exciting prospect.