Microdeletion syndromes are a group of disorders characterised by the deletion of a chromosomal segment spanning multiple disease genes, each potentially contributing to the phenotype independently. Microdeletion syndromes are often characterised by a complex clinical and behavioural phenotype resulting from the imbalance of normal dosage of genes located in that particular chromosomal segment.
NF1 microdeletion syndrome is caused by heterozygous deletions involving the NF1 gene and, in the most common 1.4 Mb deletion, other 14 genes. A more severe clinical phenotype has often been reported in NF1 patients carrying the microdeletion compared to patients with intragenic NF1 mutations. By reviewing the phenotype of 92 patients with NF1 microdeletion, we found that CVMs occurred at a significantly higher incidence in this patient population as compared to NF1 patients with intragenic mutations, suggesting that the whole-gene deletion segment encompasses important genes involved in heart development. Subsequent expression studies indicated three possible candidate genes for CVMs that warranted further studies: ADAP2 (formerly known as CENTA2), SUZ12 and UTP6 (previously called C17ORF40).
Here, we analysed the spatio-temporal expression profile of the above mentioned genes during mouse embryonic and fetal development. Based on this analysis, Adap2 seems to be expressed in heart starting from 9 dpc, during key phases of cardiac development, that is when the heart tube is elongating and looping, and atrial and ventricular septa, as well as AV valves, are forming. Moreover, Adap2 expression in heart continues even in the later stages of development, at least until 15.5 dpc. Of note, Adap2 expression is not restricted to a particular cardiac compartment or structure, but rather seems to localise in atria and ventricles. Suz12 was also detected in heart during mouse development, but its expression seems to be restricted to a short period around 10.5 dpc and to the heart atria. Differently, Utp6 showed no expression in the developing heart at all.
Since the expression of ADAP2 mouse ortholog in heart during fundamental stages of cardiac morphogenesis was suggestive of a role in heart development, we studied the possible role of ADAP2 in heart development by employing zebrafish as a model system. Over the recent years, zebrafish has proven to be a valid model for studying cardiovascular development. Despite its apparent simplicity, the zebrafish heart shares common structural, developmental and genetics features with avian and mammalian heart. Additionally, because of their small size, embryos receive enough oxygen by passive diffusion from external medium to survive and continue to develop in a relatively normal fashion for several days even in the complete absence of blood circulation, allowing a detailed phenotypic analysis of animals with severe cardiovascular defects that would be lethal in other organisms.
The functional inactivation of adap2, the ADAP2 zebrafish ortholog, obtained by the injection of two MO oligos targeting different adap2 mRNA regions (translation start site and 5'-UTR), caused the same circulatory defects, proving the specificity of the phenotypes. We also designed a splice-blocking MO, which was predicted to cause exon 2 skipping, and to produce an altered form of adap2 transcript with the generation of a premature stop codon. However, the injection of this MO at different doses did not cause any evident phenotypic defects. RT-PCR analysis, performed to test the efficacy of the splice-blocking MO, showed that only a fraction of adap2 mRNA was incorrectly spliced. Consequently, we reason that the partial expression of the wild-type protein could be enough to prevent the occurrence of the phenotypic defects. This evidence, along with the presence in the embryo of the maternal transcript, which is targeted only by translation-blocking MOs, might explain the absence of alterations following the injection of this MO.
Our molecular results suggested adap2 involvement in the cardiac jogging process, the morphogenetic process in which the heart cone is displaced to the left with respect to the anterior-posterior axis, which is one of the first evident breaks in left-right symmetry of the primitive zebrafish heart tube. Moreover, adap2 also appeared fundamental for the subsequent D-looping process, the bend of the heart tube to the right, which, by 36 hpf leads to the typical S-shaped heart, with the ventricle positioned on the right of the atrium. This was supported by the high number of adap2 morphants, which at 2 dpf, when D-looping is normally completed, showed a linear heart, a reduced loop or a reversed loop, all defects ascribable to alterations of the heart bending taking place during the D-looping process.
Functional inactivation of adap2 also evidenced its important role during AV valve morphogenesis, since the earliest stages of endocardial cushion formation. Our results strongly suggest that a defective valvulogenesis results in impaired cardiac functionality, therefore, AV valve morphological alterations are most likely accounting for the in vivo blood circulation defects displayed by adap2 morphants. Valve defects, including mitral valve prolapse, pulmonary valve stenosis and aortic valve anomalies, constitute a significant proportion of CVMs observed in patients with NF1 microdeletion syndrome. Taking into account our findings on ADAP2 role in valve morphogenesis, a correlation between ADAP2 haploinsufficiency and the onset of valvular defects in NF1-microdeleted patients can be hypothesised. Additionally, the detachment between endocardium and myocardium observed in adap2 morphants, particularly in the atrial chamber, could be caused by increased amounts of the extracellular matrix (cardiac jelly) juxtaposed between the two cardiac layers. Normal valve development involves multiple signalling pathways and extracellular matrix components take part in this process. Interestingly, dysregulation of components of the extracellular matrix seems to have a role in the myxomatous degeneration, the leaflet thickening and redundancy, typical of valvular abnormalities, such as mitral valve prolapse.
ADAP2 is known to regulate microtubule stability and the activity of ARF6, a GTPase involved in cellular motility, adhesion and polarity by regulating cytoskeleton remodelling and cortical actin formation. The alteration of these functions might impair adhesion and migration properties of AV valve cells, explaining their disorganisation and the irregular valve architecture observed in adap2 morphants.
During the early phases of valve morphogenesis, the myocardial component of AV junction is fundamental for the signalling events leading endocardial cells to begin the formation of cushions, which will be later remodelled to create flap-like valvular structures. The marked alteration of bmp4 myocardial expression in adap2 morphants suggests a compromised signalling from myocardium to endocardium, which might result in the structural valve defects observed at 5 dpf.
Overall, our study points to ADAP2 as a gene involved in heart development, and as a plausible candidate gene for the occurrence of CVMs in NF1-microdeleted patients and in the general population, constituting an advance towards a better comprehension of the complex phenotypic spectrum of the syndrome, as well as of the genetic basis of CVMs.