Neuronal migration genes and a familial translocation t(3;17): which genes are implicated in the phenotype?

Background: While Miller-Dieker syndrome critical region deletions are well known delineated anomalies, submicroscopic duplications in this region have recently emerged as a new distinctive syndrome. So far, only few cases have been described overlapping 17p13.3 duplications. Methods: In this study, we report on clinical and cytogenetic characterization of two new cases involving 17p13.3 and 3p26 chromosomal regions in two sisters with familial history of lissencephaly. Fluorescent In Situ Hybridization and array Comparative Genomic Hybridization were performed. Results: A deletion including the critical region of the Miller-Dieker syndrome of at least 2,9 Mb and a duplication of at least 3,6 Mb on the short arm of chromosome 3 were highlighted in one case. The opposite rearrangements, duplication 17p13.3 and deletion 3p were seen in the second case. This double chromosome aberration is the result of an adjacent 1:1 meiotic segregation of a maternal reciprocal translocation t(3;17)(p26.2;p13.3). Conclusions: 17p13.3 and 3p26 deletions have a clear range of phenotypic features while duplications still have uncertain clinical significance. However, we could suggest that regardless of the type of the rearrangement, the gene dosage and interactions of CNTN4, CNTN6 and CHL1 in the 3p26 and PAFAH1B1, YWHAE in 17p13.3 could result in different clinical spectrums.


6].
In this study, we report a familial translocation (3;17) leading to two different cytogenetic rearrangement resulting in a duplication/deletion of the 17p13.3 critical region for MDS including PAFAH1B1 and YWHAE genes and 3p26 region including CNTN4, CNTN6, CRBN and a part of CHL1.
Duplication and deletion of the same chromosomal region resulted in a distinct phenotypic feature in the offspring.

Patient1 (the proband)
A 2-year-old girl referred for the cytogenetic exploration of a family history of lissencephaly (FIG. 1), is the second child of a healthy consanguineous Tunisian couple. The patient's weight at birth was 3,500 g (+0,6SD). She measured 52 cm (+1,05DS) and had a head circumference of 35 cm (+0,4SD). At 2 years of age, her height and head circumference were 88 cm (+0,9SD) and 45 cm (-2,5SD), respectively. At physical examination, she had psychomotor development delay and abnormal behavior including aggressiveness, anger and agitation. Furthermore, she had craniofacial dysmorphic features (FIG. 2) including a long face, high forehead, down-slanting palpebral fissures, epicanthus, wide nose, long philtrum, thin upper lip, large and high implanted ears and pointed chin with micrognathia. In addition, she showed arachnodactyly. Her cerebral magnetic resonance imaging (MRI) was performed at two years and five months of age, and corpus callosum hypoplasia was detected.

Patient 2
The patient presented at 4 months for exploration of growth retardation, axial hypotonia, seizure and dysmorphic features (FIG. 2) including high forehead, wide nose, low implanted ears and lissencephaly at MRI. She died 10 months later. Her brother (II1) (FIG. 1) suffering from type 1 lissencephaly, died also at an early age of life.

Karyotype
Metaphase chromosome preparations were obtained by phytohemagglutinin (PHA) stimulated lymphocyte culture according to standard procedures. Chromosome analysis was carried out applying R-banding at a 500-band level according to ISCN 2016 [7]in the patient, parents and sister.

Fluorescent in situ Hybridization (FISH)
FISH was performed on blood lymphocytes blocked on metaphases of the patient, those of her sister and those of her mother, according to the standard protocol. Two probes screening the chromosome 17 short arm and the chromosome 3 short arm were used: commercial probes; Miller-Dieker/Lissencephaly region probe set: LISI (Red) and RARA (Green) (Vysis) (Abbott Laboratories, IL, USA) and Totel Vysion Multicolor DNA Probe Mixture 3 (Vysis®, Downers Grove, Illinois, USA)containing 3ptel (Green), 3qtel (Red), 22q (Orange and Green) and LSI BCR (22q11) (Aqua).
The hybridized chromosomal spreads were analyzed using a fluorescent microscope equipped with appropriate filters and Cytovision FISH system image capture software (Zeiss Axioskop 2 plus). Slides were scored on the basis of the number of probe signals for each metaphase. For each target area ten hybridized metaphases were analyzed.

Array CGH
Oligonucleotide array CGH was performed using the Agilent Human Genome CGH Microarray Kit 44K®. This microarray consisted of more than 44,000 oligonucleotide probes that spanned both coding and non-coding regions. The coverage of the human genome was made with an average spatial resolution of 75,000 pair bases.
The patient's DNA as well as a reference DNA was fragmented by heat at 95°C for 20 minutes. Each fragmented DNA product was labeled by random priming using either ULS5 or ULS3. After columnpurification, probes were denaturized and pre-annealed with 5 μg of human Cot-1 DNA, 10 μl of CGH Blocking agent and 55 μl of hybridization buffer. Hybridization was performed at 65 °C during 24 h.
The microarray was washed, scanned and analyzed with Agilent Feature Extraction® 9.1 software.
Results were interpreted with DNA analytics® 4.5 software. Only imbalances involving three or more adjacent probes were held. The identification of probes with a significant gain or loss was based on the log 2 ratio plot deviation from 0 with cutoff values of 0.5 to 1, and -0.5 to -1, respectively.

Results
The conventional cytogenetic analysis did not reveal any chromosomal anomalies in the two sisters and parents' karyotypes.

Discussion
Adjacent 1 segregation of the translocation t(3;17) in the mother led to two different chromosome imbalances in the children. The first type adjacent 1 gave rise to a derivative 3 chromosome (der3) in patient 1 that resulted in partial monosomy 3p and a partial trisomy 17p. While the second adjacent 1 type led to a derivative 17 (der17) in patient 2 that resulted in partial monosomy 17p and a partial trisomy 3p. 17p13.3 deletion encompassed PAFAH1B and YWHAE genes.
While deletions of 17p13.3 are associated with well-known phenotype ranging from Miller Dieker syndrome [8] to partial callosal and milder phenotype [9], duplications of the same chromosomal region still need further clinical and molecular characterization.
So far, to the best of our knowledge, only 13 patients having large 17p13.3 duplications, including the entire MDS comprising both PAFAH1B1 and YWHAE genes have been reported [10-11-2-12-13-14-15-16] (FIG. 6). Interestingly, all submicroscopic 17p13.3 duplications reported to date, including the present case did not share any recurrent breakpoints and have varying sizes. It has also been reported that these duplications might be the result of parental translocations involving chromosome 19 [13], chromosome 10 [14] and chromosome 5 [17] but it has never involved the 3p26 region. The proximal short arm of chromosome 17 is distinctly prone to cryptic rearrangements due to the presence of extensive repetitive sequences [2]. Furthermore, this MDS telomeric critical region is estimated to at least 400kb including eight genes in addition to PAFAH1B1gene [18]. Due to the variability of the involved genes, 17p13.3 duplications have been divided into two classes with distinct phenotypic features [2]. While, Class I duplications involve only YWHAE gene including autistic manifestations, speech, motor delay and dysmorphic facial features, Class II duplications include necessary PAFAH1B1gene and may contain also YWHAE and CRK genes [2]. The phenotypic features in these cases show moderate to mild developmental and psychomotor delay [2].
Nevertheless, when all the three genes, YWHAE, CRK and PAFAH1B1 are duplicated, the phenotype seems to be more severe [10].
Here, our proband shared clinical and dysmorphic features described in patients with duplication of the complete MDS region such as abnormal behavior (Table 1).
We reviewed an exhaustive list for the selection of thirteen cases of 17p13.3 trisomic (Table 1) who showed common dysmorphic features including a high forehead, a small mouth, and a triangular chin.
Some of these features were absent in our patient. In addition, our patient presented arachnodactyly, which is rarely described in patients with partial trisomy of 17p13.3 [10-2-12-17]. By means of complementary cytogenetic techniques, the chromosomal rearrangements were estimated to at least 3,6 Kb on chromosome 3p26.2 and 2,9 Mb on chromosome 17p13.3. The most frequent phenotypic features associated with partial trisomy 17p13.3 were correlated with duplication of the PAFAH1B1 and YWAHE genes that were located in the MDS region. It was hypothesized that the duplication of YWHAE might have an effect on neuronal network development and maturation, and was related to mild development delay and facial dysmorphisms while the duplication ofPAFAH1B1 that lead to its overexpression, was associated with moderate to severe development delay and structural brain abnormalities . Brain-imaging analysis was performed in seven of the eleven reported patients and only four showed structural brain abnormalities (Table 1). Corpus Callosum hypoplasia or agenesis represented the main brain abnormality being frequently described [10-14-11-15].
Likewise, our patient presented corpus callosum hypoplasia. Curiously, patients having the smallest and the largest duplications of the entire MDS region reported so far have presented normal Magnetic Resonance Imaging (MRI) (P1/ [11]; P1/ [16]). This suggests that this heterogeneity depends on the size of the duplication and the involved genes as well as on the involvement of other gene interactions and modifier genes. Indeed, it has been proven that transgenic mice with increased lis1 expression in the developing brain revealed abnormalities in the neuroepithelium such as the thinning of the ventricular zone, and the ectopic positioning of mitotic cells [10]. Furthermore, lis1 overexpression affected both radial and tangential migration. In fact, in this condition, migration delay in both trajectories was observed: radial migration at E13.5 and tangential migration at E12.5 rather than E14.5 [10]. However, subtelomeric neuronal migration defects are not expected to be detected by MRI scans [10]. Consequently, we can postulate that the overexpression of LIS1 gene could explain the phenotype of our patient particularly corpus callosum hypoplasia. It has been demonstrated that both CNTN4 and CNTN6 genes encode a neural adhesion molecule that is part of the immunoglobulin superfamily [24][25]. In fact, the CNTN6 gene plays a crucial role in the development, maintenance, and plasticity of functional neuronal networks in the central nervous system. It has been shown that Cntn6 deficiency in mice causes profoundmotor coordination abnormalities and learning difficulties [26]. Owing to its function, we suggest that CNTN6 gene could be responsible for the observed psychomotor development retardation in the current case. On the other hand, CNTN4, an important gene for brain development, is known to be involved in axon growth, guidance, and fasciculation [27-28-29-30]. In addition, it probably contributes to the behavioral abnormalities in our patient showing aggressiveness, anger and agitation. In fact, knockout mice of homologous neuronal adhesion molecules showed morphological, neurological and behavioral abnormalities [31].
The deletion included also CRBN gene encoding a protein of the ubiquitin proteasome pathway, which seemed to play a crucial role in brain development [32] (FIG. 7). In fact, CRBN protein is part of DCX protein ligase complex involved in the regulation of the surface expression of certain types of ion channels in neuronal memory synapses. Furthermore, the 3p26 deletion disrupted a more distal gene: CHL1 (FIG. 7).The latter encodes a protein member of the L1 family of neural cell adhesion molecules We reviewed six previously reported cases having 3p deletion, compared them to the present case report, and noted that the most frequent features are microcephaly, corpus callosum hypoplasia and facial dysmorphia (Table 2). Conversely, some studies reported cases with 3p deletion and normal phenotypes [45-46-20]. In other studies, the authors have even hypothesized that the distal 3p deletion is probably associated with normal intelligence and normal physical features .
Interestingly, both 3p deletion and17p duplication could share the same network in neuronal migration since both anomalies lead to corpus callosum hypoplasia and pachygyria. So far, both genes duplicated in 17p especially PAFAH1B1 and genes deleted in 3p especially CNTN6 and CRBN affected the process of cortical development by alteration of the stabilization of microtubules, the axon growth and the axon guidance [48-26-49].
Neuronal migration is a complex process that involves several actors and factors [50][51]. The most critical step responsible for a normal brain development is the cell migration from the ventricular zone into the cortical plate [52].
Mutations and chromosomal aberrations can alter the chromosome 3D organization. This alteration could play a more important role than we believe it does in chromosomal interactions and transcriptional regulation of genes. In fact, it has been shown that the chromatin 3D modification could disturb the topologically associating domains (TADs) and consequently the regulation of gene expression [53-54-55]. Such alteration could explain the phenotypic variability in human disease ranging from milder phenotype to microdeletion/microduplication syndrome.

Conclusions
The variability of genes, which are mapped in the involved regions (3p and 17p), and the description of the clinical characteristics of our patient contribute to the confirmation and further delineation of the associated characteristics to the partial trisomy of 17p13.3 encompassing the entire MDS critical region as well as the partial monosomy of chromosome 3p26.2. Various genes and structural chromosomal anomalies have been discovered involved in this process. However, the exact molecular basis of brain malformations still needs further studies. Written informed consent to participate in this study was obtained from the parents.

Consent to publish
Written informed consent was obtained from the parents for photo and clinical data publication.

Availability of data and materials
All data generated or studied during this study are included in the published article which is available upon request from the corresponding author.

Competing interests
All the authors have no competing interests.

Funding
No financial or non-financial benefits have been received or will be received from any party related directly or indirectly to the subject of this article.
Authors' contributions SMZ contributed to conception and design. MHA, SD and HH contributed to all experimental work, analysis and interpretation of data. KBH and AM referred patients to our department. SMZ and SD were responsible for the consultation. SMZ and AS were responsible for overall supervision. MHA drafted the manuscript, which was revised by SMZ. All authors read and approved the final manuscript.