Molecular mechanisms of hearing loss in Nager syndrome

Abstract

Nager syndrome is a rare human developmental disorder characterized by hypoplastic neural crest-derived craniofacial bones and limb defects. Mutations in SF3B4 gene, which encodes a component of the spliceosome, are a major cause for Nager. A review of the literature indicates that 45% of confirmed cases are also affected by conductive, sensorineural or mixed hearing loss. Conductive hearing loss is due to defective middle ear ossicles, which are neural crest derived, while sensorineural hearing loss typically results from defective inner ear or vestibulocochlear nerve, which are both derived from the otic placode. Animal model of Nager syndrome indicates that upon Sf3b4 knockdown cranial neural crest progenitors are depleted, which may account for the conductive hearing loss in these patients. To determine whether Sf3b4 plays a role in otic placode formation we analyzed the impact of Sf3b4 knockdown on otic development. Sf3b4-depleted Xenopus embryos exhibited reduced expression of several pan-placodal genes six1, dmrta1 and foxi4.1. We confirmed the dependence of placode genes expression on Sf3b4 function in animal cap explants expressing noggin, a BMP antagonist critical to induce placode fate in the ectoderm. Later in development, Sf3b4 morphant embryos had reduced expression of pax8, tbx2, otx2, bmp4 and wnt3a at the otic vesicle stage, and altered otic vesicle development. We propose that in addition to the neural crest, Sf3b4 is required for otic development, which may account for sensorineural hearing loss in Nager syndrome.

Introduction

Nager syndrome (OMIM#154400) is a rare congenital disorder with a little more than 100 cases described in the literature. It is characterized by craniofacial defects including downward slanting of the palpebral fissures, midface retrusion, micrognathia, cleft palate and limb deformities, typically hypoplasia or absence of the thumbs (Wieczorek, 2013; Trainor and Andrews, 2013). Most cases are sporadic, however autosomal dominant and recessive forms of the disease have been reported (Chemke et al. 1988; Nur et al., 2013). Nager syndrome is caused by mutations in the SF3B4 (Splicing factor 3b, subunit 4) gene in 63% of the cases (Bernier et al., 2012; Czeschik et al., 2013; Petit et al., 2013). Rodriguez syndrome (OMIM#201170) is also due to mutations in SF3B4 (Drivas et al., 2019). Rodriguez and Nager syndrome patients have similar craniofacial features, however in Rodriguez syndrome the defects are typically more severe and also include postaxial as well as lower limb defects (Rodriguez et al., 1990). SF3B4 gene encodes SF3B4, a protein component of the spliceosome, the machinery that remove introns from transcribed pre-mRNA (Will and Lührmann, 2011). The mechanisms by which SF3B4 mutations result in these restricted craniofacial and limb defects is not well understood (Beauchamp et al., 2020; Griffin and Saint-Jeannet, 2020).

Nager syndrome patients often have abnormalities of the external ear, auditory canal and middle ear ossicles, which contribute to conductive hearing loss (Petit et al., 2013; Czeschik et al., 2013; Cassina et al., 2017; Likar et al., 2018, Danziger et al, 1990; Aylsworth et al., 1991; Hermann et al., 2005; Davies and Johnson, 2011; Lin, 2012). Interestingly a small number of Nager syndrome patients have been diagnosed with strictly sensorineural (Chemke et al., 1988; Mishra et al., 1999; Battaglia and Magit, 2000; Kumar et al., 2015) or mixed conductive/sensorineural (Czeschik et al., 2013; Hermann et al., 2005) hearing loss suggesting that the etiology of deafness is heterogeneous in this disease. While conductive hearing loss is usually attributed to abnormal development of the first and second pharyngeal arches neural crest-derived middle ear ossicles (reviewed in Minoux and Rijli, 2010), sensorineural hearing loss results from defects to the inner ear or the vestibulocochlear nerve, two structures derived from thickenings of the embryonic head ectoderm, the otic placodes, with a minor contribution from the neural crest (reviewed in Streit, 2001; Whitfield, 2015; Ritter and Martin, 2018).

We have previously shown in an animal model of Nager syndrome that Sf3b4 is required for neural crest formation, and that Sf3b4-depleted Xenopus embryos develop hypoplastic neural crest-derived craniofacial cartilages at the tadpole stage, reminiscent of the craniofacial defects seen in Nager syndrome patients (Devotta et al., 2016). Here we analyze the impact of Sf3b4 knockdown on cranial placode gene expression and otic vesicle development as a possible cause of sensorineural hearing loss in Nager syndrome patients.

Materials and Methods

Xenopus embryos, microinjections and animal cap explants

Xenopus laevis embryos were raised in 0.1X NAM (Normal Amphibian Medium; Slack & Forman, 1980) and staged according to Nieuwkoop and Faber (Nieuwkoop and Faber, 1967). All experiments were conducted in accordance with the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. All procedures were approved by the Institutional Animal Care and Use Committee of New York University (animal protocol #IA16–00052). Beta-galactosidase and noggin mRNAs were synthesized in vitro using the Message Machine Kit (Ambion; Austin, TX). Standard control (CoMO) and Sf3b4 (SF3B4MO; GCCATAACCTGTGAGGAAAAAGAGC; Devotta et al., 2016) morpholino antisense oligonucleotides (MO) were purchased from GeneTools (Philomath, OR). All MOs were injected in one blastomere at 2-cell stage and embryos were analyzed by in situ hybridization (ISH) at the neurula (NF stage 15) or tailbud (NF stage 35) stages. Embryos were co-injected with 500 pg of beta-galactosidase mRNA to identify the injected side. For animal cap explants, both blastomeres at the 2-cell stage were injected in the animal pole region with noggin mRNA (400 pg) alone or in combination with SF3B4MO (10 ng), animal cap explants were dissected at the late blastula stage (NF stage 9) and cultured for 8 hours in 0.5X NAM.

Whole-mount in situ hybridization and histology

Embryos at the appropriate stage were fixed in 4% paraformaldehyde (PFA) and stained for Red-Gal (Research Organics; Cleveland, OH) to visualize the lineage tracer (β-gal mRNA) and processed for ISH. Antisense digoxygenin-labeled probes (Genius kit; Roche, Indianapolis IN) were synthesized using template cDNA encoding six1 (Pandur and Moody, 2000), dmrta1 (Huang et al., 2005), foxi4.1 (Pohl et al., 2002), pax8 (Heller and Brandli, 1999), tbx2 (Hayata et al., 1999), wnt3a (Wolda et al., 1993), bmp4 (Jones et al., 1992), otx2 (Pannese et al., 1995) and ebf2 (Burns and Vetter, 2002). Whole-mount ISH was performed as described (Harland, 1991; Saint-Jeannet, 2017) with minor modifications. For histology, tbx2 stained embryos were embedded in Paraplast+ (Sigma; St Louis, MO), sectioned (12 μm) on an Olympus rotary microtome (Olympus; Waltham, MA), counter stained with Eosin and mounted in Permount (Fisher Scientific; Waltham, MA). The size of the otic vesicle (μm) was determined based on the number of sections (12 μm) required to cut the entire otic vesicle on control side vs. injected side.

qRT-PCR analysis

Total RNAs were extracted from 12 animal cap explants using the RNeasy Micro Kit (Qiagen; Valencia, CA). The RNA samples were digested with RNase-free DNase I before RT-PCR. The amount of RNA isolated was quantified by measuring the optical density using a Nanodrop spectrophotometer (Nanodrop Technologies; Wilmington, DE). The RT-PCR reaction mixture consisted of 10 μl of SYBR Green RT-PCR Master mix (Applied Biosystems; Foster City, CA), 500nM of forward and reverse primers, 1ng of template RNA in total volumes of 10 μl. The reaction was performed on a QuantStudio 3 Real-Time PCR System (Applied Biosystems; Foster City, CA) using primer sets to detect six1, eya1, sox2, krt12.4 and odc (Hong and Saint-Jeannet, 2006).

Statistical Method

Each experiment was performed on at least three different batches of embryos obtained from different females (biological replicates). Gene expression on the Sf3b4MO-injected side was compared to the uninjected side for each embryo, and to CoMO-injected embryos. The data from all biological replicates were pooled for statistical analysis. Significance testing for gene expression by ISH was performed using the Chi-squared test, and for gene expression by qRT-PCR using the Student’s t-test. A p-value of <0.05 was considered significant and analyses were performed using GraphPad Prism 8 (GraphPad Software, San Diego, CA).

Results

Nager syndrome patients are affected by hearing loss

A survey of the scientific literature of confirmed Nager and Rodriguez syndrome patients indicates that approximately 45% of all patients (63/140) exhibit hearing loss in addition of the classic features of the disease, craniofacial and limb defects. That is 54% of patients (36/66) with confirmed SF3B4 mutations (Table 1), and 36% of patients (27/74) who have not been directly linked to a mutation in this gene (Table 2). These numbers are presumably underestimated since information on the hearing status is unknown in 41% (27/66) and 44% (33/74) of the cases with or without SF3B4 mutation, respectively. Among the 63 patients with reported hearing loss (Table 1 and Table 2), the type of hearing loss is categorized as conductive (38%), sensorineural (6%) or mixed (6%), while for the remainder of the patients (49%) the nature of the hearing loss has not been fully characterized. These observations suggest that the cause of deafness is heterogeneous in these patients as conductive hearing loss is typically attributed to abnormal development of the neural crest-derived elements of first and second pharyngeal arches, while sensorineural hearing loss results from defect to the inner ear or the associated vestibulocochlear nerve, which are both derived from the otic placode.

Table 1:

Incidence of hearing loss in Nager and Rodriguez syndrome patients with SF3B4 mutation/deletion.

Nager syndrome cases with SF3B4 mutation/deletion	Number of patients	Hearing status unknown	Normal hearing	Uncategorized hearing loss	Conductive hearing loss	Sensorineural hearing loss
Rodriguez et al., 1989	1**	1	-	-	-	-
Waggoner et al., 1999	1	1	-	-	-	-
Wessels et al., 2002	1**	1	-	-	-	-
Sermer et al. 2007	1**	1	-	-	-	-
Bernier et al., 2012	25	6	-	19	-	-
Czeschik et al., 2013	7	2	1	-	2	2*
Petit et al., 2013	13	4	2	1	6	-
Castori et al., 2014	1	1	-	-	-	-
McPherson et al., 2014	1	1	-	-	-	-
Bellanger et al., 2015	1	-	-	1	-	-
Lund et al., 2016	1	1	-	-	-	-
Marques et al., 2016	3	3	-	-	-	-
Cassina et al., 2017	1	-	-	-	1	-
Denu and Burkard, 2017	1	-	-	1	-	-
Drivas et al., 2018	1	-	-	1	-	-
Likar et al. 2018	1	-	-	-	1	-
Hayata et al., 2019	1	-	-	1	-	-
Zerounian et al., 2019	1	1	-	-	-	-
Jourdain et al. 2019	1	1	-	-	-	-
Drozniewska et al., 2020	1	1	-	-	-	-
Zhao and Yang, 2020	1	1	-	-	-	-
Bukowska-Olech et al. 2020	1	1	-	-	-	-
TOTAL	66	27	3	24	10	2

^*Indicate patients affected by both conductive and sensorineural (mixed) hearing loss.

^**Patient sequenced by Irving et al., (2016).

Table 2:

Incidence of hearing loss in Nager and Rodriguez syndrome patients with unconfirmed SF3B4 mutation.

Nager syndrome cases with unconfirmed SF3B4 mutation	Number of patients	Hearing status unknown	Normal hearing	Uncategorized hearing loss	Conductive hearing loss	Sensorineural hearing loss
Kawira et al., 1984	2	2	-	-	-	-
Hecht et al., 1987	1	1	-	-	-	-
Chemke et al., 1988	2	1	-	-	-	1
Goldstein and Mirkin, 1988	1	1	-	-	-	-
Rodriguez et al., 1989	2	2	-	-	-	-
Danziger et al, 1990	1	-	-	-	1	-
Aylsworth et al., 1991	5	2	-	1	2	-
Zori et al., 1992	1	1	-	-	-	-
Mishra et al., 1999	1	-	-	-	-	1
Battaglia and Magit, 2000	3	1	1	-	-	1
Groeper et al., 2002	1	-	-	1	-	-
Kavadia et al., 2003	1	-	-	-	-	-
Herrmann et al., 2005	11	1	-	-	8	2*
Dimitrov et al., 2005	1**	1	-	-	-	-
Miyawaki et al., 2009	1	1	-	-	-	-
Davies and Johnson, 2011	1	-	-	-	1	-
Lin, 2012	1	-	-	-	1	-
Abdollahi Fakhim et al., 2012	1	-	-	1	-	-
Bernier et al., 2012	16	16	-	-	-	-
Czeschik et al., 2013	5	5	-	-	-	-
Petit et al., 2013	5	1	1	2	1	-
Nur et al., 2013	1	1	-	-	-	-
Malik et al., 2014	1	-	-	1	-	-
Bozathoglu and Muenevveroglu, 2015	1	-	-	1	-	-
Kumar et al., 2015	1	-	-	-	-	1
Rosa et al., 2015	1	1	-	-	-	-
Ural and Ceylander, 2016	1	1	-	-	-	-
Wu et al., 2017	2	2	-	-	-	-
Tay et al., 2017	2	2	-	-	-	-
Rai et al., 2017	1	-	1	-	-	-
TOTAL	74	33	3	7	14	6

^*Indicate patients affected by both conductive and sensorineural (mixed) hearing loss.

^**Patient sequenced by Irving et al., (2016).

Sf3b4 depletion alters cranial placode gene expression

Cranial placode progenitors arise from a common domain anteriorly, known as the pre-placodal region (PPR). Over time this domain is further subdivided to give rise to the specialized paired sensory organs, the adenohypophysis and cranial ganglia (Schlosser, 2010; Grocott et al., 2012; Saint-Jeannet and Moody, 2014). To evaluate a potential role of Sf3b4 in cranial placode formation we used a well-characterized morpholino antisense oligonucleotide (Sf3b4MO; Devotta et al., 2016) to knockdown Sf3b4 function in Xenopus embryos. Unilateral injection of increasing doses of Sf3b4MO (2 ng to 20 ng) resulted in a marked reduction of six1 and dmrta1 expression, two genes expressed in the PPR at the neurula stage (Fig 1a). This effect was dose-dependent (Fig 1b), and we selected the dose of 10 ng for further analyses.

Figure 1: Sf3b4 knockdown affects six1 and dmrta1 expression at the neurula stage.

(a) Unilateral injection of increasing doses of Sf3b4MO (2 ng-20 ng) interfere with six1 and dmrta1 expression. RNA encoding the lineage tracer β-galactosidase was co-injected with Sf3b4MO to identify the injected side (red staining, right side in all panels). The arrows point to the affected placode areas. Anterior views, dorsal to top. (b) Quantification of six1 and dmrta1 ISH results. The numbers on the top of each bar indicate the number of embryos analyzed from three independent experiments.

We expanded our analysis to include a larger repertoire of genes, including foxi4.1, a factor broadly expressed at the PPR, and pax8, which is primarily restricted to the otic placode, the precursor of the inner ear (Schlosser and Ahrens 2004). Both genes were downregulated in a majority of Sf3b4-depleted embryos, while injection of a control MO (CoMO) at the same concentration had no impact on the expression of these genes (Fig 2a, b).

Figure 2: Sf3b4 knockdown affects a broad range of placode genes.

(a) Unilateral injection of Sf3b4MO (10 ng) interfere with six1, dmrta1, foxi4.1 and pax8 expression. RNA encoding the lineage tracer β-galactosidase was co-injected with Sf3b4MO to identify the injected side (red staining, right side in all panels). The arrows point to the affected placode areas. pax8 is also expressed posteriorly in the prospective pronephros (asterisks). Anterior views, dorsal to top. (b) Quantification of the ISH results. The numbers on the top of each bar indicate the number of embryos analyzed from at least three independent experiments. Injection of a CoMO at the same concentration (10 ng) had no impact of the expression of these genes. *** p<0.0001, Chi-squared test.

Sf3b4 is required for placode gene induction by noggin

To confirm Sf3b4 requirement for cranial placode gene expression we used an animal cap explant assay and qRT-PCR (Fig 3a). In this preparation attenuation of bone morphogenetic protein (BMP) signaling promotes cranial placode and neural plate fates (Brugmann and al., 2004; Hong and Saint-Jeannet, 2007). In these explants, placode (six1 and eya1) and neural plate (sox2) gene induction by the BMP antagonist Noggin was significantly repressed by co-injection of Sf3b4MO (Fig 3b). Sf3b4-depletion in noggin-injected explants was associated with an increase in krt12.4 expression (Fig 3b), characteristic of the default epidermal fate of these explants. Altogether these results indicate that in the embryos and in animal cap explants Sf3b4 is required for pan-placodal and neural plate gene expression.

Figure 3: Placode genes induction by noggin require Sf3b4.

(a) Animal cap explants dissected from blastula stage embryos (NF stage 9) injected at the 2-cell stage with noggin mRNAs alone or in combination with Sf3b4MO were cultured for 8 hours. (b) qRT-PCR analyses indicate that Sf3b4MO blocks placode (six1 and eya1) and neural plate (sox2) gene induction by Noggin in animal cap explants, and restores epidermal fate (krt12.4). Values are normalized to odc and presented as mean ± s.e.m.; (*) p < 0.05 and (**) p < 0.001 (Student’s t-test), from three independent samples.

Sf3b4 depletion alters otic vesicle gene expression and development

To further characterize the phenotype of Sf3b4-depleted embryos we have analyzed the expression of pax8, tbx2, wnt3a, bmp4 and otx2 at the otic vesicle stage (NF stage 35). pax8 and and tbx2 are broadly expressed throughout the otic vesicle, otx2 and wnt3a are confined ventrally and dorsally, respectively, while bmp4 is restricted to the anterior and posterior domains of the otic vesicle (Fig 4a; Saint-Germain et al., 2004). Embryos were injected with Sf3b4MO (10 ng) and analyzed by ISH at stage 35. All five genes were greatly reduced upon Sf3b4-depletion in the majority of the embryos examined (Fig 4b,c), suggesting that otic vesicle development was defective. Interestingly, the penetrance of the phenotype was higher at this stage than observed at the neurula stage. One interpretation is that subtle differences in gene expression levels that could not be fully assessed visually at an early stage may still have long-term consequences on otic vesicle development.

Figure 4: Sf3b4 knockdown affects otic vesicle gene expression.

(a) Schematic representation of the expression pattern of pax8, tbx2, wnt3a, bmp4, and otx2 in the epithelium of the otic vesicle of NF stage 35 embryos. A, anterior; P posterior; D, dorsal; V, ventral. (b) In embryos injected with 10 ng of Sf3b4MO the regional otic expression of pax8, tbx2, wnt3a, bmp4, and otx2 is disrupted. RNA encoding the lineage tracer β-galactosidase was co-injected with Sf3b4MO to identify the injected side (red staining). The black arrows point to the otic expression of each gene on the control side and the white arrows to the corresponding affected areas on the Sf3b4MO-injected side. For orientation, the position of the eye is indicated (asterisks). Lateral views, anterior to left, dorsal to top. (c) Quantification of the ISH results. The numbers on the top of each bar indicate the number of embryos analyzed from three independent experiments.

We next analyzed by histology the morphology of the otic vesicle of these animals, by comparing the size of the otic vesicles on the Sf3b4MO-injected versus control side of tbx2 stained embryos. We found that the majority of the embryos had an overall reduction of the otic vesicle size on the injected side by approximately 40% (Fig 5a,b). These results point to Sf3b4 as an important regulator of otic vesicle development.

Figure 5: Sf3b4 knockdown affects otic vesicle development.

(a) Serial sections (#1-anterior to #14-posterior) through a representative tbx2-stained embryo (NF stage 35) injected with 10 ng of Sf3b4MO. On the injected side (left side; double arrows), the size of the otic vesicle is reduced as compared to the control side (right side; single arrows). (b) The size of the otic vesicle (μm) was determined based on the number of sections (12 μm) required to cut the entire otic vesicle on control vs. injected sides. On average this represents 11.33 sections or 136 μm (control) vs. 6.77 sections or 81.33 μm (Sf3b4MO-injected). * p < 0.05 (Unpaired t-test with Welch’s correction), n=9.

Sf3b4 depletion affects olfactory placode gene expression

While information on olfactory defects remains largely undocumented in Nager syndrome patients we expanded our analysis to evaluate the impact of Sf3b4 knockdown on genes expressed in the developing olfactory epithelium namely dmrta1, ebf2 and six1. Interestingly, the expression of these genes was also disrupted in morphant embryos (Fig 6a,b) consistent with a potential role of Sf3b4 in the development of other cranial placode derivatives in addition to the prospective inner ear.

Figure 6: Sf3b4 knockdown affects olfactory gene expression.

(a) Unilateral injection of Sf3b4MO (10 ng) interferes with dmrta1, ebf2 and six1 expression. RNA encoding the lineage tracer β-galactosidase was co-injected with Sf3b4MO to identify the injected side (red staining, right side in all panels). The arrows point to reduce gene expression in the olfactory epithelium. (b) Quantification of the ISH results. The numbers on the top of each bar indicate the number of embryos analyzed from two independent experiments. Injection of a CoMO at the same concentration (10 ng) had no impact of the expression of these genes. *** p<0.0001, Chi-squared test.

Discussion

Here we present evidence that the splicing factor Sf3b4 is required for otic placode gene expression and development in Xenopus embryos, this is in addition to its well-documented role in neural crest formation (Devotta et al, 2016), and we propose that this activity may account for sensorineural deafness in Nager syndrome patients carrying mutations in SF3B4. Our results also suggest that Sf3b4 may have a broader function in the formation of cranial placodes and their derivatives, as several pan-placodal as well as olfactory placode genes are also disrupted upon Sf3b4 knockdown.

The primary cause of Nager syndrome is haploinsufficiency of SF3B4 (Bernier et al., 2012; Czeschik et al., 2013; Petit et al., 2014). This syndrome is characterized by limb defects and mandibulofacial dysostosis (MFD). MFD is a condition affecting neural crest-derived craniofacial skeletal elements arising from the first and second pharyngeal arches (Passos-Bueno et al., 2009; Wieczoreck, 2013). In the first pharyngeal arch, the maxillary prominence forms the maxillary, zygomatic and temporal bones, while the mandibular prominence forms the middle ear bones, malleus and incus, and the mandible. The second pharyngeal arch gives rise to the third middle ear ossicle, stapes, styloid process and the lesser horn of the hyoid. All these structures are defective to varying degrees in Nager syndrome, and a subset of these patients is affected by hearing loss (Table 1 and Table 2). The most common form of deafness among these patients is conductive hearing loss, which is typically due to defect to the middle ear ossicles preventing the proper transmission of the sound to the inner ear.

A mouse model of MFD indicates that this condition is caused by an early depletion of neural crest progenitors through mechanisms involving increased apoptosis and reduced proliferation (Dixon et al., 2006; Jones and et al., 2008). Similarly, we have previously shown that Sf3b4-depleted Xenopus embryos show increased apoptosis in the ectoderm, leading to a reduction in neural crest progenitors, and subsequent development of hypoplastic craniofacial cartilages (Devotta et al., 2016). It is therefore likely that the depletion of neural crest progenitors is also the underlying cause for conductive hearing loss in Nager syndrome patients since the middle ear bones are neural crest derived.

Because a number of Nager syndrome patients also present with sensorineural hearing loss (Table 1 and Table 2), which is typically due to defect to the inner ear or the associated vestibulocochlear nerve, we posited that other ectoderm derivatives, in addition to the neural crest, might be affected by a similar process upon Sf3b4 knockdown. Indeed we found a significant decrease in the expression of several otic-specific genes associated with a reduction in the size of the otic vesicle in Sf3b4 depleted Xenopus embryos. The requirement of Sf3b4 for placode gene expression was also confirmed in an animal cap assay. Therefore we propose that defective otic placode formation and inner ear development is the likely culprit in Nager syndrome patients suffering from sensorineural hearing loss. Future studies will determine the long-term consequences of Sf3b4 knockdown and the extent to which the functional components of the inner ear are affected.

Nager syndrome belongs to a category of diseases known as spliceosomopathies resulting from mutations in genes encoding components of the spliceosome (reviewed in Beauchamp et al., 2020; Griffin and Saint-Jeannet, 2020). Interestingly, mutations in other splicing factors encoding genes such as EFTUD2, SNRPB and TXNL4A also cause MFD in three related syndromes, mandibulofacial dysostosis with microcephaly (MFDM; OMIM#610536; Lines et al., 2012), cerebrocostomandibular syndrome (CCMS; OMIM#117650; Lynch et al., 2014) and Burn-McKeown Syndrome (BMKS; OMIM#608572; Wieczorek et al., 2014), respectively (reviewed in Lehalle et al., 2015; Beauchamp et al., 2020). A number of MFDM, CCMS and BMKS patients have also been diagnosed with hearing loss, and a subset of these patients had confirmed sensorineural hearing loss at least in the case of MFDM (Lines et al., 2012; Gordon et al., 2014). This suggest that the etiology of deafness is heterogeneous across multiple craniofacial spliceosomopathies, and that mutations in proteins of the spliceosome affect at least two major embryonic cell populations the neural crest and cranial placodes. While the spliceosome is presumably active in all cells of the organism, the cell type specificity of these spliceosomopathies is still poorly understood (Griffin and Saint-Jeannet, 2020).

Sf3b4-depleted Xenopus embryo is the only animal model available so far to investigate the pathogenesis of Nager syndrome (Devotta et al., 2016). In the mouse, homozygous Sf3b4 mutation is embryonic lethal, and heterozygous mutants have a relatively mild phenotype characterized by anterior-posterior patterning defects of the axial skeleton (Yamada et al., 2020). The lack of a distinctive craniofacial phenotype in these animals could be explained by compensation by another component of the Sf3b complex. Typically a high level of redundancy is observed in murine gene families, but not as much in Xenopus. Alternatively it is also possible that reducing Sf3b4 by half its normal levels may not be sufficient to affect neural crest formation in mice.

Highlights.

• Nager syndrome is a human craniofacial disorder due to mutation in SF3B4.

• A subset of Nager syndrome patients is also affected by hearing loss.

• Sf3b4-depleted Xenopus embryos have altered otic gene expression.

• This effect may account for conductive hearing loss in human patients.