|Home | About | Journals | Submit | Contact Us | Français|
Genitopatellar syndrome (GPS) and Say-Barber-Biesecker-Young-Simpson syndrome (SBBYSS or Ohdo syndrome) have both recently been shown to be caused by distinct mutations in the histone acetyltransferase KAT6B (a.k.a. MYST4/MORF). All variants are de novo dominant mutations that lead to protein truncation. Mutations leading to GPS occur in the proximal portion of the last exon and lead to the expression of a protein without an activation domain. Mutations leading to SBBYSS occur either throughout the gene, leading to nonsense-mediated decay, or more distally in the last exon. Features present only in GPS are contractures, anomalies of the spine, ribs and pelvis, renal cysts, hydronephrosis and agenesis of the corpus callosum. Features present only in SBBYSS include long thumbs and long great toes and lacrimal duct abnormalities. Several features occur in both, such as intellectual disability, congenital heart defects, genital and patellar anomalies. We propose that haploinsufficiency or loss of a function mediated by the C-terminal domain causes the common features, whereas gain-of-function activities would explain the features unique to GPS. Further molecular studies and the compilation of mutations in a database for genotype-phenotype correlations (www.LOVD.nl/KAT6B) might help tease out answers to these questions and understand the developmental programs dysregulated by the different truncations.
Whole exome sequencing recently facilitated the identification of mutations leading to the truncation of KAT6B (OMIM 605880), a histone acetyltransferase, in Genitopatellar syndrome (GPS) and SBBYSS (Campeau, et al., 2012; Clayton-Smith, et al., 2011; Simpson, et al., 2012). Genitopatellar syndrome (GPS) was first described in 1988 (Goldblatt, et al., 1988), and then delineated in 2000 (Cormier-Daire, et al., 2000). Characteristic findings are hypoplastic or absent patellae, flexion deformities, genital anomalies, microcephaly with agenesis of the corpus callosum and intellectual disability.
In 1986, Ohdo described two sibs and their cousin with blepharophimosis, ptosis, congenital heart defects, intellectual disability and hypoplastic teeth (Ohdo, et al., 1986). The Say-Barber-Biesecker-Young-Simpson (SBBYS) variant of Ohdo syndrome was later described by different physicians (Biesecker, 1991; Say and Barber, 1987; Young and Simpson, 1987) and is characterized by blepharophimosis, ptosis, an immobile mask-like face, a bulbous nasal tip, hypotonia, feeding problems, long thumbs and great toes, and dislocated or hypoplastic patellae. Congenital heart defects, cryptorchidism, dental anomalies and thyroid anomalies are often noted. The subjects in the original family described by Ohdo, in contrast to the later defined variant, had different facial features, proteinuria, no skeletal anomalies and a different mode of inheritance, and it is thus likely that they had a different condition from the SBBYS variant of Ohdo syndrome. An excellent classification and overview of blepharophimosis syndromes has been compiled (Verloes, et al., 2006) as well as a detailed analysis of the clinical features of the SBBYS variant of Ohdo syndrome (Day, et al., 2008).
KAT6B (a.k.a MYST4 and MORF, MOZ related factor) was first cloned by searching collections of human expressed sequence tags for genes encoding MYST domain-containing proteins (Champagne, et al., 1999). MOZ (monocytic leukemia zinc-finger protein, a.k.a. MYST3 or more recently KAT6A, OMIM 601408) was itself first identified as a fusion protein with CBP deriving from translocations in acute myeloid leukemia (AML) (Borrow, et al., 1996), simultaneously to the identification of similar proteins with silencing roles in yeast (Reifsnyder, et al., 1996). MOZ has a highly conserved acetyltransferase domain shared with the yeast YBF2, SAS2 and the human TIP60 (KAT5, OMIM 601409), thus the term MYST-domain. KAT6B has the same highly conserved acetyltransferase domain and has been identified to also fuse with CBP (encoded by CREBBP, OMIM 600140) following translocations in acute myeloid leukemia and myelodysplastic syndrome, as reviewed in (Yang and Ullah, 2007). It is also disrupted by chromosomal translocations in multiple cases of uterine leiomyomata (Kojima, et al., 2003; Moore, et al., 2004; Murati, et al., 2004). KAT6B has been shown to interact with RUNX2 (OMIM 600211) and BRPF1 (OMIM 602410) in a tetrameric complex with ING5 (OMIM 608525) and MEAF6 (OMIM 611001) (Ullah, et al., 2008; Yang and Ullah, 2007). KAT6B was also co-immunoprecipitated with a PPAR-alpha interacting cofactor complex (Surapureddi, et al., 2002), and a yeast two-hybrid screen identified Atrophin-1 as a binding partner (Lim, et al., 2006). Mice hypomorphic for Myst4 have a short stature, an absence of fusion of the tibia and ﬁbula, microcephaly with neurogenesis defects, short palpebral fissures, low set ears and malocclusion (Kraft, et al., 2011; Merson, et al., 2006; Thomas and Voss, 2004; Thomas, et al., 2000). Despite these findings, the precise roles of KAT6B in regulating gene transcription during development have still to be defined. A better understanding of the phenotype resulting from KAT6B mutations may lead to insights into the molecular roles of KAT6B.
In this work, we describe the establishment of a mutation database, review genotype-phenotype associations, and explore the effect of mutations on protein function for KAT6B-related conditions.
Using the Leiden Open source Variation Database (LOVD) package hosted on the Leiden server, we have created a database for KAT6B (RefSeq NM_012330.2) which catalogs all known disease causing mutations (Fokkema, et al., 2011). The database can be publicly accessed at www.LOVD.nl/KAT6B. We have also added a column to the standard mutation database format to allow for detailed phenotype information to be entered (see Figure 1). Along with subjects with GPS and Ohdo (SBBYS) syndrome, we have also included a subject with a “Noonan-like” phenotype resulting from a translocation interrupting KAT6B in an early intron (Kraft, et al., 2011). Data on clinical features and mutations was drawn from recently published articles (Campeau, et al., 2012; Clayton-Smith, et al., 2011; Simpson, et al., 2012). A snapshot from this database can be seen in Figure 1.
The distribution of the mutations for both conditions is shown in Table 1 and illustrated in Figure 2. Mutations causing GPS are seen in a cluster of residues encoded in exon 18 (the last exon); none of the GPS mutations have been shown to undergo-nonsense mediated decay and consequently truncated proteins are produced from these alleles. In contrast, mutations leading to SBBYSS occur seemingly randomly throughout the gene, leading to nonsense-mediated decay, or occur more distally in the last exon than the GPS-associated cluster.
The c.3769_3772delTCTA mutation recurring in 4 subjects with GPS occurs in a direct TCTA repeat, with the deletion of one TCTA repeat (see Figure 3 for diagrams). In SBBYSS, the recurring c.5201_5210dup mutation occurs in a palindromic repeat (TCGACGTCGT precedes the palindrome ACGACGTCGT which is duplicated by the mutation). Template-primer slippage would be the most likely mechanisms for these repeat mutations (Kunkel and Bebenek, 2000). Another unusual mutation is the c.4360_4368delinsAAAAACCAAAA mutation in GPS. Given the homology of the resulting DNA sequence to a region in an intron of RALGAPA2 (AAGAAGTACTGAAAAACCAAAAGA), it is possible that this mutation arises through a gene conversion event between KAT6B, located on chromosome 10, and RALGAPA2, located on chromosome 20. The homology domain is however relatively short compared to most gene conversion events (Chen, et al., 2007).
With the use of the LOVD database, we have performed a detailed comparison of the clinical features, organized by syndrome, resulting from the various mutations in KAT6B (see table 2).
Subjects with GPS have flexion contractures at the hips and knees and can also have club feet. A minority have radiological anomalies in the pelvis, spine and ribs. However, subjects with the same mutation do not consistently have these findings. In contrast, subjects with SBBYSS do not have skeletal anomalies but most have long thumbs and great toes, which are not seen in GPS. Interestingly, these features are only seen with distal truncations in KAT6B. Patellar anomalies are seen in both conditions, albeit more frequently in GPS.
Subjects with either syndrome have severe developmental delay and intellectual disability. Some have hypotonia at birth. Most subjects with GPS have microcephaly, whereas in SBBYSS many subjects have a smaller than average head circumference but not frank microcephaly. A thin or absent corpus callosum is seen in all subjects with GPS whereas this anomaly has not been observed in SBBYSS.
Anal anomalies such as anal atresia or stenosis, rectal duplication and an anteriorly positioned anus are occasionally seen in GPS but not in SBBYSS. Most females with GPS have clitoromegaly and/or hypoplasia of the labia (minora or majora). Cryptorchidism is seen in both conditions but scrotal hypoplasia has only been reported in GPS.
Hydronephrosis is seen in a majority of subjects with GPS and multiple renal cysts are seen in a minority. In a subject with SBBYSS whose mutation status is unknown, vesicoureteric reflux was reported (Day, et al., 2008), but otherwise subjects with SBBYSS do not have renal anomalies. Congenital heart defects are noted in about 50% of subjects with either syndrome, the most frequent defects being atrial septal defects, ventricular septal defects, and a patent foramen ovale.
See figure 4 for a correlation of facial features with the location of the mutations. Subjects with SBBYSS have a very distinctive facial appearance with a mask-like facies, blepharophimosis and ptosis. Several subjects with Ohdo also have lacrimal duct abnormalities. Interestingly, a subject with GPS with a more distal mutation of KAT6B that overlaps with the region associated with SBBYSS has some degree of blepharophimosis and ptosis. Subjects with either condition can have prominent cheeks, and a nose with either a bulbous end or a broad or prominent base. Several subjects with either condition have micro/retrognathia or prognathism. Bitemporal narrowing and prominent eyes is noted in the GPS subjects with the p.Lys1258Glyfs*13 mutation (BCM4, KCL2 and KCL3).
Feeding difficulties are seen in both conditions. Contributing factors may be hypotonia in both syndromes, laryngomalacia in a minority of GPS subjects and cleft palate in some individuals with SBBYSS. Respiratory difficulties were noted in several infants with GPS; again, the hypotonia and laryngomalacia might have been contributing factors. Dental anomalies, thyroid anomalies, and hearing loss are seen in both syndromes, but more frequently in SBBYSS. Rarer features present in GPS are discussed in (Campeau, et al., 2012) and (Penttinen, et al., 2009), whereas those present in SBBYSS are discussed in (Day, et al., 2008).
To help decide when to investigate KAT6B in subjects who present with features of GPS syndrome, we propose the following criteria:
Criteria for the clinical diagnosis of Ohdo/SBBYS syndrome were proposed by White et al. (White, et al., 2003). These criteria included a mandatory presentation of blepharophimosis, ptosis, and intellectual disability. Supporting features included depressed nasal bridge, hypoplastic teeth, deafness, undescended testes and hypotonia. While these criteria are useful for a clinical diagnosis, this strict rubric may result in false negatives if used for genetic testing, due to variability in presentation. We feel the criteria to prompt molecular testing for KAT6B should be broader. We propose criteria similar to GPS that employ two major criteria or one major criterion and two minor criteria.
Mutations proximal to the last exon, leading to reductions in protein levels (haploinsufficiency) lead to an SBBYSS phenotype. Distal mutations in the last exon, which also give SBBYSS phenotype, may thus lead to a similar phenotype by a loss-of-function mechanism. Yet, some of the facial and digit features of SBBYSS are only seen in individuals with the most distal mutations.
We thus hypothesize that features which overlap both GPS and SBBYSS such as developmental delay/ intellectual disability, patellar anomalies, heart defects, genital anomalies, and dental anomalies are likely to result of either haploinsufficiency or a the loss of a function normally mediated by the C-terminal region of the acidic domain (Fig 1). However, there are some unique features of GPS independent of SBBYSS. GPS mutations, which all lead to expression of a truncated protein (Campeau, et al., 2012; Simpson, et al., 2012), may account for the features of GPS that are distinct from those given by the haploinsufficiency mutations. These mutations may indicate that a truncated protein of a given length may gain a new function. We hypothesize that a gain-of-function may be caused by an altered binding affinity or dysregulated interactions with natural binding partners of KAT6B. In SBBYSS, the specific phenotypic consequences associated with longer but still truncated proteins imply that these proteins lack this new function.
Documentation of further genotypes and phenotypes into the LOVD database (discussed above) will likely provide a useful foundation from which to conduct future genotype-phenotype studies.
Philippe Campeau is funded in part by the CIHR Clinician-Scientist training award.
Grant numbers: NIH P01 HD070394 and NIH P01 HD22657 to BL
COMPETING INTERESTS, FUNDING
There are no competing interests.