Mutations in the Abnormal Spindle-like Microcephaly-associated (ASPM) gene are the most common cause of primary microcephaly, a rare condition characterized by a severe reduction of brain size at birth. Several studies allowed to identify ASPM as a centrosome and mitotic spindle protein that regulates cell division and spindle orientation. However, little is known about ASPM molecular mechanisms, especially in human neural cells relevant to the disease. In order to decipher the molecular mechanisms of action of ASPM in human corticogenesis, we used co-immunoprecipitation (coIP) followed by mass spectrometry to identify the interactors of ASPM in human HEK cells and human cortical progenitors differentiated from pluripotent stem cells engineered to tag the endogenous ASPM protein. We thus identified and validated 14 ASPM interactors of which 12 are newly reported, and seven are found specifically in neural cells, including the important spindle pole regulator Nuclear mitotic apparatus (NUMA). We then characterized the expression and localization of the identified proteins in human cortical progenitors differentiated from control and isogenic ASPM mutant cells. This revealed that many of the identified proteins are selectively located at the spindle pole, and that this selective localization is disrupted in mutant cells for several of the interactors, in particular the MAP7 domain-containing protein 1 (MAP7D1) and DnaJ homolog subfamily B member 6 (DNAJB6). Our data uncover some of the complex ASPM interactome relevant and specific to human brain development and microcephaly, and suggest that ASPM acts as a major molecular hub at the centrosome and mitotic spindle to control the patterns of cell division of cortical progenitors. / Doctorat en Sciences biomédicales et pharmaceutiques (Médecine) / info:eu-repo/semantics/nonPublished

Congenital microcephaly (conMiC) is a manifestation of severely disrupted prenatal brain development, caused by genetic defects, toxins, severe maternal malnutrition, or infection. The Zika virus outbreak and the devastating impact of Zika infection on the fetal brain have focused much attention on the cellular and molecular pathophysiology of conMiC. Mendelian conMiC disorders offer a unique opportunity for understanding gene and protein networks that direct cellular processes essential for prenatal brain development. Using OMIM and literature searches, I analyzed 68 conMiC disorders and their 65 corresponding genes. ConMiC-disorder phenotypes were characterized by analyzing the co-occurrence of ID, retinal abnormalities, seizures, and short stature. Short stature co-occurred with 70% of conMiC disorders, while seizures and retinopathy co-occurred with 68% and 37%, respectively. In 53% of conMiC disorders, seizures and short stature overlapped, while all features overlapped in 22% of conMiC disorders; only 7% of conMiC disorders lacked one of these co-occurring features. This shows conMiC genes are rarely specialized for brain growth, with generalized functions in overall body growth, retinal development, and/or regulation of neural activity. ConMiC-gene transcript accumulation in the brain is typically greatest during the prenatal period, and then declines postnatally, suggesting active transcriptional repression. Nonetheless, in neurons and glia of the adult brain, 44 conMiC genes had confirmed persistent protein accumulation. Experimental evidence indicates transcription in neural progenitor cells (NPCs) for at least 82% of conMiC genes. The spatiotemporal expression patterns of conMiC genes tend to align well with their biological functions and corresponding mutant phenotypes. Nearly 60% of conMiC gene products have functions in the cell cycle and/or DNA repair. Most conMiC disorders are caused by recessive, loss-of-function mutations. There are direct binding and regulatory interactions amongst many conMiC genes, which interact in larger networks and shared pathways. Depletion of single conMiC gene products can affect the transcript and/or protein levels of other conMiC gene products, which could have a “domino effect”, and disrupt entire networks important for brain development. Further evidence for this model is that 22 conMiC genes are consistently dysregulated in Zika-infected developing human brain tissue. Due to the complexity of conMiC genes and their interactions, there are many unique challenges to developing treatments for conMiC, particularly conMiC caused by maternal Zika-virus infection. However, insights to treatment strategies could be gained by using human genetics to find potential modifiers, screening for drugs that can normalize disrupted cell cycle and DNA-repair processes, or can stabilize protein complexes that are disrupted due to a conMiC gene mutation.

Autosomal recessive primary microcephaly (MCPH) is defined by congenital microcephaly and associated mental retardation with head circumference of the affected individual at least 3 standard deviations below age- and sex-means. It is a disorder of abnormal fetal brain growth which is a consequence of impaired neurogenesis. It is genetically heterogeneous with seven known loci and genes for all the seven loci have been identified: MCPH-1-MCPH1, MCPH2-WDR62, MCPH3-CDK5RAP2, MCPH4-CEP152, MCPH5-ASPM, MCPH6-CENPJ, and MCPH7-STIL. All the seven MCPH proteins localize at the centrosome. Apart from MCPH, many other proteins associated with the phenotype microcephaly have been localized to the centrosome or linked to it functionally. For example, Microcephalic osteodysplastic primordial dwarfism type II protein PCNT and Seckel syndrome protein ATR are also centrosomal proteins. All of the above findings show the importance of centrosomal proteins as the key players in neurogenesis and brain development. However, the exact mechanism as to how the loss-of-function of these proteins leads to microcephaly remains to be elucidated. The study of MCPH genes can also provide insights into the basics of neurogenesis that lead to a normal brain size. The most common cause of MCPH is mutations in the ASPM (abnormal spindle-like, microcephaly-associated protein) gene. The main aim of this study was to gain insight into the function of ASPM using the yeast two-hybrid technique. The main findings of the study are listed below. To find novel interacting proteins for SPM, a GAL4 based yeast two-hybrid system was used. The 3,477 amino acid long ASPM was divided into eight different baits and each bait was individually used for screening a human fetal brain cDNA library cloned in the pACT2 vector. To generate baits, the different regions were amplified from human fetal brain cDNA and cloned in-frame with the GAL4-DNA binding domain in the pGBKT7 vector. Screening with a C-terminus ASPM bait (pGBKT7-CTR) identified Angelman syndrome protein ubiquitin protein ligase E3A (UBE3A) as an ASPM interactor. A region of UBE3A from amino acids 639-875 was found to interact with ASPM. The identification of UBE3A as an ASPM interacting partner was interesting as more than 80% of Angleman syndrome patients are reported to have microcephaly. Screening with the baits pGBKT7-1.4 kb ASPM and pGBKT7-2.1 kb ASPM harboring parts of IQ domain identified calmodulin as an ASPM interating partner. The full length calmodulin was found to interact with the IQ domain of ASPM. The interactions identified in the yeast two-hybrid assay were confirmed in vivo by co-immunoprecipitation studies. For this, a rabbit polyclonal anti-ASPM antibody was raised against the N-terminal region of ASPM (from amino acids 544-1059). The specificity of the antibody was tested by Western blot analysis and immunofluorescence microscopy. ASPM antibody recognized the 410 KDa fulllength ASPM protein in lysates from human fetal tissues and different cell lines. Immunofluorescence analysis in HEK293 cells with the antibody revealed centrosomal staining of ASPM throughout mitosis and midbody staining in cytokinesis, as reported previously. Using antibodies against ASPM and UBE3A and human fetal kidney lysate, ASPM and UBE3A interaction was confirmed in vivo by co-immunoprecipitation. The interaction between ASPM and calmodulin was confirmed similarly. The relevance of the interaction between ASPM and UBE3A was pursued further Like ASPM, UBE3A localized to the centrosome throughout mitotic progression. ASPM levels were found to be unaffected upon overexpression of UBE3A in HEK293 cells, indicating that ASPM is not degraded by a UBE3A-dependent proteasomal pathway or the degradation may be spatial-temporal control. Further, immunofluorescence analysis of UBE3A overexpressing HEK293 cells revealed that UBE3A does not affect either the ASPM localization or its protein level at the centrosome. Synchronization of HEK293 cells in different cell cycle phases revealed that UBE3A is a cell cycle dependent protein and its level peaks in mitosis To explore the functional role of UBE3A’s increased level in mitosis, UBE3A was depleted in HEK293 cells with a shRNA construct and stable clones were generated. HEK293- UBE3A shRNA knockdown cells were examined for normal mitotic progession and spindle defects. There was a 3.81- to 5.52-fold increase in the frequency of anaphase/telophase cells with missegregated chromosomes in UBE3A knockdown clones as compared to scrambled clones. Hence, we identified a definitive role of UBE3A in chromosome segregation. Defective chromosome segregation has been reported in many studies associated with microcephaly-related proteins. Interestingly, chromosome malfunctioning has also been reported in Drosophilia asp mutants (ASPM orthologue) and Celegans aspm-1 knockdown cells. Therefore, the loss of both ASPM and UBE3A leading to chromosome segregation defects reveals the existence of a molecular pathway common to both ASPM and UBE3A As a consequence of chromosome missegregation, UBE3A knockdown cells were found to undergo abnormal cytokinesis and apoptosis. The percentage of apoptotic cells in UBE3A knockdown clones was 1.25- to 3.04-fold higher as compared to scrambled clones. Interestingly, an extensive apoptosis has been found in the neural folds of MCPH7 gene STIL null mice embryos. Thus, the present study links Angleman syndrome protein UBE3A to ASPM, centrosome and mitosis for the first time.

Autosomal recessive primary microcephaly (MCPH) is a congenital neurodevelopmental disorder characterised by a reduced occipital-frontal head circumference (OFC) of less than -3 SDs below the population mean for age and sex. It is a genetically heterogeneous disorder caused by mutations in one of the following 10 MCPH genes: MCPH1 (microcephalin 1), WDR62 (WD repeat domain 62), CDK5RAP2 (cyclin-dependent kinase 5 regulatory associated protein 2), CASC5 (cancer susceptibility candidate 5), CEP152 (centrosomal protein 152 kDa), ASPM (asp [abnormal spindle] homolog, microcephaly associated [Drosophila]), CENPJ (centromeric protein J), STIL (SCL/TAL1-interrupting locus), CEP135 (centrosomal protein 135 kDa) and CEP63 (centrosomal protein 135 kDa). The MCPH1 (microcephalin 1) gene is located on chromosome 8p23.1. Microsatellite analysis has previously shown LOH at the markers D8S518 and D8S277 flanking the MCPH1 locus in 1/21 oral tumors. Furthermore, LOH at the markers D8S1742 and D8S277 flanking the MCPH1 locus has also been observed in 2/32 hepatocellular carcinomas. MCPH1 has been found to be mutated in breast and endometrial cancers. Additionally, it was found to be downregulated at the transcript level in 19/30 ovarian cancer tissues and the protein level in 93/319 breast cancer tissues. Decreased MCPH1 protein levels are associated with triple negative breast cancers and a lower transcript level of MCPH1 correlates with lesser time for metastasis to occur in breast cancer patients. Interestingly, MCPH1 knockout mice in a null TP53 background show susceptibility to cancer.So far, studies have indicated that MCPH1 is a DNA repair protein. MCPH1 is required for the formation of DNA repair foci, chromatin relaxation, HR and NHEJ. It regulates G1/S and G2/M cell cycle checkpoints. Also, depletion of MCPH1 leads to genomic instability and centrosome amplification. Hence, the defect in the function of MCPH1 can lead to plethora of anomalies including cancer. Based on these observations, we hypothesized that MCPH1 may also function as a tumor suppressor (TS) gene, in addition to its role in the brain development. The purpose of this study was to test if MCPH1 also functions as a TS gene using different approaches in OSCC (oral squamous cell carcinoma). OSCC is the sixth most common type of cancer. It includes the cancer of the lips, anterior 2/3rd of the tongue, buccal mucosa, floor of the mouth, retromolar trigone and gingiva. Despite the advances in the treatment of oral cancer, the five-yr survival rate has not increased. Hence, the effective treatment of OSCC requires the identification of molecular targets to design appropriate therapeutic strategies. LOH, mutations and promoter methylation in tumors are the hallmarks of TS genes. In order to ascertain the TS roles of MCPH1, we carried out LOH analysis in 81 matched blood/normal and tumor oral tissues using D8S1819, D8S277 and D8S1798 markers flanking the MCPH1 locus. The results showed LOH at one or more markers in 14/71 (19.72%) informative samples across the tumor stages from T1 to T4. The entire coding region and the exon-intron junctions of the MCPH1 gene were sequenced for mutations in 15 OSCC samples and 5 cancer cell lines (viz., A549, HeLa, KB, SCC084 and SCC131). In total, three mutations namely c.1561G>T(p.Glu521X), c.321delA(p.Lys107fsX39) and c.1402delA(p.Thr468fsX32) were identified. The expression of MCPH1 was analysed at both the transcript and protein levels by real-time quantitative RT-PCR and immunohistochemistry, respectively, in OSCC samples. MCPH1 was downregulated in 51.22% (21/41) of OSCC samples at the transcript level. The MCPH1 protein was downregulated in 76% (19/25) of the OSCC samples. In order to elucidate if the MCPH1 promoter was methylated in OSCC tissues, we retrieved the MCPH1 promoter from the database TRED (Transcriptional Regulatory Element Database). The promoter was analysed for the presence of CpG islands using the CpG Plot/CpG Report program. Two CpG islands (CpGI and CpGII) were identified within the MCPH1 promoter. Both the CpG islands were analysed for methylation in 40 OSCC samples by COBRA (Combined Bisulfite Restriction Analysis). CpGI showed no methylation in 40 OSCC samples. However, CpGII showed methylation in 4/40 (10%) OSCC samples and the methylation was absent in their corresponding normal oral tissues. To analyse the methylation of the MCPH1 promoter in cancer cell lines, HeLa, KB, SCC084 and SCC131 cells were treated with 5’-2-deoxy azacytidine (AZA), a methyltrasferase inhibitor. HeLa and KB cells did not show any change in the MCPH1 transcript level after the AZA treatment. However, SCC084 and SCC131 cells showed upregulation of MCPH1 after the treatment, suggesting methylation of the MCPH1 promoter. To validate these observations, we examined the methylation status of both the CpG islands in these cell lines. We found methylation of CpGII only in SCC084 cells. HeLa, KB and SCC131 cells showed no methylation of CpGI and CpGII. The results obtained by COBRA in these cell lines were further confirmed by bisulfite sequencing of CpGI and CpGII islands. Further, the upregulation of MCPH1 after azacytidine treatment in SCC131 cells can be attributed to a promoter independent mechanism or due to methylation of the CpG sites not examined by us. To elucidate the biological effects of MCPH1 in a cancer cell line, we generated stable clones overexpressing MCPH1 in KB cells. The results showed that MCPH1 overexpression decreased cellular proliferation, cell invasion, anchorage-independent growth in soft-agar and tumor growth in nude mice. Further, MCPH1 overexpression lead to apoptosis. A low frequency of LOH, mutations and promoter methylation suggested that they might not be the major mechanisms of downregulation of MCPH1 in OSCC. We then speculated that MCPH1 could be regulated by miRNAs. We therefore used five miRNA target prediction softwares to identify miRNAs targeting MCPH1. The programs identified two binding sites for miR-27a within the 5.4 kb region of the 3’-UTR of MCPH1. The luciferase assay showed that both the seed regions of MCPH1 were binding to miR-27a. In addition, transient transfection of the premiR-27a construct in KB cells decreased the protein level of MCPH1. Additionally, in a small panel of 10 OSCC samples, there was a negative correlation between the levels of miR-27a and MCPH1. To the best of our knowledge, this is the first report showing any miRNA regulating the MCPH1 gene. It is important to note that tumor suppressors can serve as potential biomarkers with prognostic value. Hence, we analysed the correlation of the expression levels of MCPH1 with clinico-pathological parameters such as TNM, gender, age and site of the cancer by Fischer’s exact test. No statistical correlation was observed between the transcript or protein levels with any of the clinico-pathological parameters. In summary, the results of the present study have suggested that the primary microcephaly gene MCPH1 shows several hallmarks of TS genes and functions as a tumor suppressor in OSCC, in addition to its role in brain development. We have for the first time shown that miR-27a targets MCPH1 and regulates its level. It is interesting to note that none of the other 10 MCPH genes have been shown to be regulated by any miRNA yet. Our study will be useful in designing novel therapeutic methods for the treatment of OSCC either by overexpression of MCPH1 or reducing the level of miR-27a by an antagomir.