WO2009001392A1 - Gene codifying for ambra 1 protein which regulates autophagv and the development of the central nervous system - Google Patents

Abstract

It has now been found and constitutes the object of the present invention a gene which codifies for a protein endowed with a regulatory function toward autophagy with reference to the cerebrospinal apparatus and to related affections, in particular those which show themselves at a prenatal stage.

Description

"Gene codifying for Ambra 1 protein which regulates autophagv and the development of the Central Nervous System."

Technical Field

The present invention relates to a gene codifying for Ambra 1 protein which regulates autophagy and the development of the Central Nervous System.

Prior Art

Autophagy is a self-degradative process of cellular components which is involved in their basal turnover in response to nutrient starvation and to organelle damage in a wide range of eukaryotes (Lum, J. J., DeBerardinis, R. J. & Thompson, C. B. Nat Rev MoI Cell Biol (2005) 6, 439-48; Levine, B. & Yuan, J. J Clin Invest (2005) 115, 2679-88; Levine, B. & Klionsky, D. J. Dev Cell (2004) 6, 463-77). Autophagy is a mechanism which is very well regulated at the gene level, as demonstrated by the identification of several yeasts genes which control this phenomenon, which are collectively called Atg genes (Autophagy Controlling Genes) (Huang, W. P. & Klionsky, D. J. Cell Struct Funct (2002) 27, 409-20).

Recent studies showed not only that the function of Atg genes has been conserved during Vertebrates evolution, but also shed light on the involvement of the autophagic apparatus in many aspects of tissue omeostasis (Levine, B. & Yuan, J. J Clin Invest (2005) 115, 2679-88; Levine, B. & Klionsky, D. J. Dev Cell (2004) 6, 463-77).

Actually, autophagy has a protective role against the onset of neuro- degeneration (Hara, T. et al. Nature (2006) 441, 885-9; Komatsu, M. et al. Nature (2006) 441, 880-4) and in cancer cell growth (Edinger, A. L. & Thompson, C. B. Cancer Cell (2003) 4, 422-4; Liang, X. H. et al. Nature (1999) 402, 672-6; Yue, Z., Jin, S., Yang, C, Levine, A. J. & Heintz, N. Proc Natl Acad Sci U S A (2003) 100, 15077-82). In addition, experimental work in many in vitro systems showed the existence of a complex net of interactions between autophagy and apoptosis, in which failure in the regulation of the apoptotic machinery induces changes in the autophagic programme and vice versa (Shimizu, S. et al. Nat Cell Biol (2004) 6, 1221-8; Boya, P. et al. MoI Cell Biol

(2005) 25, 1025-40; Lum, J. J. et al. Cell (2005) 120, 237-48; Yousefi, S. et al. Nat Cell Biol (2006) 8, 1124-32).

During the occurence of the autophagic phenomenon, portions of cytoplasm are sequestered by double-membraned vesicles, the autophago- somes, and degraded after fusion with lysosomes for subsequent recycling (Klionsky, D. J. J Cell Sci (2005) 118, 7-18). In vertebrates, this process acts as a pro-survival or pro-death mechanism in different physiological and pathological conditions, such as neurodegeneration and cancer (Levine, B. & Yuan, J. J Clin Invest (2005) 115, 2679- 88;Hara, T. et al. Nature (2006) 441, 885-9; Komatsu, M. et al. Nature

(2006) 441, 880-4; 7; Edinger, A. L. & Thompson, C. B. Cancer Cell (2003) 4, 422-4.). However, the possible roles of autophagy during embryonic development are still largely uncharacterised (Levine, B. & Klionsky, D. J. Dev Cell (2004) 6, 463-77).

Beclin 1/ATG6 is the key regulator in autophagosome formation which has been more thoroughly investigated.

Its deletion causes a precocious embryonic lethality and seems to be involved in the genesis of tumors as a suppressor gene (Liang, X. H. et al. Nature (1999) 402, 672-6; Yue, Z., Jin, S., Yang, C, Levine, A. J. & Heintz, N. Proc Natl Acad Sci U S A (2003) 100, 15077-82).

A role for autophagy in vertebrates development has been suggested for some time now, based on the morphological features of autophagy in embryogenesis (Levine, B. & Klionsky, D. J. Dev Cell (2004) 6, 463- 77; Clarke, P. G. Anat Embryol (Berl) (1990) 181, 195-213; Baehrecke, E. H. Nat Rev MoI Cell Biol (2002) 3, 779-87; Lockshin, R. A. & Zakeri, Z. Int J Biochem Cell Biol (2004) 36, 2405-19). Such a role is supported by mutagenesis experiments involving some Atg genes in mice.

Embryos mutant for Beclin 1, i.e. the mammalian gene ortholog for ATG6 in yeasts, die early during embryogenesis and show a developmental delay, whereas atg5- or α£g7-deficient mice survive embryo- genesis but suffer from nutrient and energy insufficiency soon after birth (Kuma, A. et al. Nature (2004) 432, 1032-6; Komatsu, M. et al. J Cell Biol (2005) 169, 425-34). In addition, atg5'- embryos were shown to have deficient autophagy-dependent clearance of apoptotic cell corpses during development (Qu, X. et al. Cell (2007) 128, 931-46).

The morphogenesis of the nervous system is also based on a complex interplay of interactions between proliferation and differentiation control and apoptosis (Copp, A.J. J. Anat. (2005) 207, 623-35).

Autophagy is also most likely involved in the development of some degenerative diseases of the nervous system such as Alzheimer's disease, Huntington disease and Parkinson disease (Rubinsztein D. C. et al, Autophagy (2005) 1, 11-22; Fortun J. et al. Neurobiol Dis. (2007) 25, 252-265).

However, it is not yet known how autophagy can be regulated during development, and whether vertebrate specific factors beside the Atgs play a role in this context.

It is then very important to identify further regulatory factors of autophagy which have been conserved throughout higher Eucariotes evolution, with the aim to apply appropriate screening, prevention and therapy procedures in those pathologies in which this physiologic cell process is of fundamental clinical and diagnostic importance. Brief description of the invention

It has been now found and is object of the present invention a gene which codifies for a protein endowed with a regulatory activity in autophagy, with reference to the cerebro-spinal apparatus and its diseases, particularly those which show themselves during the prenatal stage.

Further objects will become evident from the detailed description and from the enclosed claims.

Brief description of the figures

Figure 1: Neural tube defects in Ambra 1 mutant embryos.

a, Detection of wild-type (+) and Ambra 1 gene trap transcripts (gt) from total E 14.5 embryos analysed by Northern blotting. Size marker: 28S ribosomal RNA. b, Immunoblot analysis of E14.5 embryo brain extracts using antibodies against Ambra 1 (left) and βgal (right). Specific bands (arrows) and unspecific signals (asterisk) are indicated. MW, Molecular weights. The fusion protein can be revealed only by means anti-β gal antibody, c-f, Expression of Ambra 1 in the mouse embryonic nervous system, βgal-staining on whole-mount Ambra 1^+/^ mouse embryos at E8.5 (c), Ell.5 (d) and on cross section of Ell.5 spinal cord (e). Matching expression of the endogeneous gene was revealed by whole-mount Ambra 1 mRNA in situ hybridisation analysis in E 11.5 wild type embryos (f). g-1, Electron scanning microscopy analysis of Ell.5 (g) and E12.5 (h) Ambra l^^t/gt embryos. Note the failure of the neural tube closure, the extensive midbrain/hindbrain exencephaly (Ex) with a closed telencephalon (T) and the lumbosacral spinal bifida (Sb). E14.5 wt (i) and Ambra U^υ^ (j) embryos are characterized by prominent exencephaly (Ex). Histological analysis of E12.5 wt (k) and Ambra l%^tlst (1) embryos on cross sections. Note the absence of a normal ventricular system, the extensive overgrowth of the proliferative neuroepithelium in the diencephalon (Di) and spinal cord (Sc), and the enlarged V-th ganglia (VG) in the Ambra l&et embryo, m-p, TUNEL staining of E10.5 brain (m, n) and E9.5 spinal cord (o, p) in wt (m, o) and Ambra 1&& embryo (n, p) sections, q-t, Analysis of cell proliferation in wt (q, s) and Ambra 1&& embryos (r, t) on transverse sections of E8.5 cephalic neural folds in prospective hindbrain region (mytoses: arrows in q-r) and on sagittal sections of ElO.5 forebrain (BrdU uptake, s-t). Scale bars: 500 μm (c, g- h, k), 1 mm (d), 150 μm (e), 2 mm (i), 400 μm (m, s), 78 μm (o), 50 μm (q). Fb, forebrain; Hb, hindbrain; Mb, midbrain.

Figure 2: Ambra 1 is a novel Beclin 1 interacting protein.

a, Ambra 1-Beclin 1 interaction by yeast two-hybrid assay. Yeast cells were co-transfected with the indicated plasmids and plated in medium with or without histidine (H) and adenine (A). G4, Gal4; DBD, DNA binding domain; AD, Activation Domain; LAM, Lamin. b, Ambra 1- Beclin 1 interaction in mammalian cells. 2F cells were co-infected with retroviral vectors encoding Beclin 1 and the indicated myc-tagged Ambra 1 proteins, or myc-tagged βgal as a negative control. Protein extracts were immunoprecipitated by using an anti-myc-tag antibody (IP myc). Purified complexes and corresponding total extracts were analysed by western blot using anti-myc (WB myc, upper panels) or anti-Beclin 1 (WB Beclin 1, lower panels) antibodies. When multiple bands are present, the upper band has the expected MW. A scheme of the various Ambra 1 mutants is reported, c, Ambra 1 is in a complex which includes Beclin 1 and Vps34. Protein extracts from 2F cells co- expressing Beclin 1 and Ambra 1 full-length protein (FL) or βgal both myc-tagged were immunoprecipitated by using an anti-myc-tag antibody (IP myc). Purified complexes and corresponding total extracts were analysed using anti-Beclin 1 (WB Beclin 1) and anti-Vps34 (WB Vps34) antibodies, d, Ambra 1-Beclin 1 interaction in developing brain tissues. Protein extracts from two E14.5 wt brains (A and B) were immunoprecipitated by using an anti-Ambra 1 antibody (Ambra 1) or the pre-immune serum (Pre). Purified complexes and corresponding total extracts were analysed using anti-Ambra 1 (WB Ambra 1) and anti-Beclin 1 (WB Beclin 1) antibodies. The asterisk indicates the position of co-migrating immunoglobulins, e, Beclin 1 and Ambra 1 co- localise in mammalian cells. 2F cells co-expressing Beclin 1 and myc- tagged Ambra 1 were stained for myc and Beclin 1. Scale bars: 8 μm.

Figure 3: Ambra 1 regulates autophagv.

a, Ambra 1 down-regulation in 2F cells using specific siRNA oligos (No.l and No.2). Ambra 1 mRNA and protein levels were analysed by quantitative PCR (left) and Western blotting (right), respectively. Control: unrelated oligo. Tubulin: protein loading control, b, Rapa- mycin-induced autophagy requires Ambra 1. After Ambra 1 down- regulation, 2F cells were treated with rapamycin and the occurrence of autophagy was analysed 48 h later by appearance of GFP-LC3 punctate staining or by LC3-I to LC3-II conversion (Western blot in the insert). Representative results are accompanied by a graph reporting the data from three experiments. Scale bar: 20 μm. c, Nutrient- starvation-induced autophagy requires Ambra 1. After Ambra 1 down- regulation, 2F cells were starved for 4 h and analysed for appearance of GFP-LC3 punctate staining, d, Ambra 1 overexpression increases basal and rapamycin-induced autophagy. 2F cells were transduced with Ambra 1 full length (FL), Ambra 1 fragments (Fl-3), Beclin 1 and βgal (negative control) encoding retroviruses, stimulated with rapamycin or left untreated and analysed by appearance of GFP-LC3 punctate staining, e, Beclin 1 -induced autophagy requires Ambra 1. After Ambra 1 down-regulation, GFP-LC3 expressing 2F cells were transduced with a Beclin 1-encoding retrovirus. Occurrence of autophagy was analysed 48 h later by appearance of GFP-LC3 punctate staining, f, Ambra 1 down-regulation reduces the amount of Vps34 associated to Beclin 1 during autophagy. 48 h after Ambra 1 down-regulation with Oligo 2, 2F cells were either starved or cultured in standard medium for an additional 4 h. Protein extracts were immunoprecipitated (IP) by using an anti-Beclin 1 antibody. Purified complexes and corresponding total extracts were analysed by Western blotting with anti-Vps34 (left panels) and anti-Beclin 1 (right panels) antibodies. The graph indicates the signal intensity of Beclin 1-associated Vps34 as determined by densitometry. R.L., Relative Levels. Values in (a-f) represent the mean ± s.d. of three experiments.

Figure 4: Autophagy is impaired in Ambra 1 mutant embryos.

a, Alteration of both LC3 conversion and translocation in Ambra l^^ embryos. Left panel: immunoblot analysis of LC3 in E14.5 embryos. LC3-I to LC3-II conversion is reduced in mutant extracts (LC3-II/LC3-I ratio, as measured by densitometry, from left to right: 5%, 9.8%, 2.5%, 4.6%). H, head; B, body. Right panels: sections from E10.5 GFP-LC3; Ambra 1^+/+ and GFP-LC3; Ambra l^'s¹ neuroepithelium were compared so as to evaluate GFP-LC3 subcellular distribution. Clusters of GFP-LC3 punctate structures (arrows) are indicated in wt sections. Images were optimized by deconvolution. Scale bar: 10 μm. b, Ubiquitin positive cells in Ambra le^l/s^t developing brains. Sagittal sections from E 11.5 embryos were stained using an anti-ubiquitin antibody. Ubiquitin accumulates in numerous mutant neurepithelial cells (upper panels, scale bar: 25 μm; lower panels, scale bar: 10 μm). Ambra 1 mutant protein is visualised in the left panels by β-galacto- sidase activity (light blue staining). Counterstaining: toluidine blue, c, Rapamycin-induced autophagy is impaired in Ambra l^¹¹^ MEFs. Cells dissected from embryos with different genotypes (A-D) were treated with rapamycin and the occurrence of autophagy was analysed 48 h later by the appearance of GFP-LC3 punctate staining. Representative results are shown in the upper panels, while the lower panel shows the data from three experiments. Scale bar: 20 μm. d, Ultrastructural analysis by means of electron microscopy of ultra-thin sections from wild type and Ambra I^ε^t cultured MEFs. The arrows indicate autophagic vacuoles. Scale bar: 5 μm. Quantification of the mean number of autophagic vacuoles per cell in wild type and Ambra l^^ cultured MEFs is shown in the graph. Values in (c, d) represent the mean ± s.d. of three experiments. Figure 5: Plasmid construct.

Supplementary Figure Sl: Ambra 1 gene characterization.

a, Chromosomal localization of the Ambra 1 genomic locus by fluorescent in situ hybrydization (FISH) mapping using a Ambra 1- specific probe (2.7 Kb Ambra 1 cDNA fragment). The left panel shows FISH signals on a mouse chromosome; the right panel shows the same mitotic figure stained with DAPI to identify mouse chromosome 2, El- E3, syntenic to human chromosome 11, Ilpll.2-pl2. b, Schematic diagram of the gene trap vector insertion site within the murine Ambra 1 locus. The exon sizes are indicated in base pairs (bp) and the intron sizes are indicated in kilobases (Kb), c, Multiple sequence alignment of human (Hs, Homo sapiens), mouse (Mm, Mus musculus), rat (Rn, Rattus norυegicus) and zebrafish (Dr, Danio rerio) Ambra 1 proteins were originated in the alignment program ClustalW. The sequence termed 'Hs Ambra 1' was obtained from a human brain cDNA bank as described in Methods. Identical residues in all sequences in the alignment are indicated by '*'. Conservative and semi-conservative substitutions are indicated by ':' and '.', respectively. The WD-40 repeats-region is highlighted. Note that different isoforms due to alternative RNA splicing are present in Hs and Mm. d, A list of the disease-linked microsatellite markers, including the Ambra 1 locus (Ilpll.2-pl2), is shown. Interestingly, in Jouberts syndrome 2 (Jbts2/CORS2, MIM No.608091), a large part of chromosome 11 encompassing the Ambra 1 locus, is entirely deleted. Jbts2 is a complex syndrome characterized by the presence of a midbrain/hindbrain malformation which includes cerebellar vermis hypoplasia and thickened, elongated and maloriented cerebellar peduncles ⁶. Potocki- Shaffer syndrome (PSS, MIM No.601224) is a contiguous gene syndrome characterized by skull ossification defects and mental retardation. The latter aspect has been associated to two markers adjacent to the Ambra 1 locus ⁷. The same markers have also been linked to melanoma ⁸. In red are indicated three markers within the Ambra 1 transcriptional unit . Supplementary Figure S2: Expression of Ambra 1 in embrvogenesis and postnatal brain.

a, Whole mount βgal staining of a heterozygous Ambra 1 mouse embryo at E8.5. The embryo is shown with the embryonic annexes (right panel) and, after isolation, from a frontal view and a lateral view (middle and right panels, respectively). Staining of the neural plate is evident in all images. Ea, extraembryonic annexes; Fb, forebrain; Mb, hindbrain; Hb, hindbrain. Scale bars: 1 mm (left) 250 μm (middle) 500 μm (right), b, Expression of the endogenous Ambra 1 gene was revealed by RNA in situ hybridization analysis on sagittal sections from E 10.5 wild type embryos, c, Matching expression of the mutant allele in E10,5 hterozygous littermates was revealed by whole-mount βgal staining. Scale bar: 0.8 mm. Arrows indicate the ventralmost staining in the spinal cord, d, Whole mount βgal staining of an isolated heterozygous Ambra 1 mouse embryo at E14.5. Note the strong Ambra 1 expression in the developing nervous system, in the body mesenchyme and in the heart and tongue. H, heart; Tn, tongue; K, kidney; L, liver. Scale bar: 2 mm. e-i, Whole mount βgal staining on sagittal (e) and anterior-to-posterior cross sections (f-g) of adult heterozygous Ambra 1 brains. In the adult brain Ambra 1 is expressed at a very high level in the Ob, Cx, Hi/Dg, Cpu, Acu, BlA, Hp and GrL of cerebellum. Differentiated nuclei in diencephalons (e, g), midbrain and hindbrain (e, h) show moderate (ZI) or even low expression of Ambra-1. Ace, accumbens nucleus; BlA, basolateral complex of amygdala; Cb, cerebellum; Cpu, caudate putamen nucleus (striatum); Cx, cortex; Dg, dentate girus; GrL, granular layer of cerebellum; Hi, hippocampus; Hp, hypothalamus; Ob, olfactory bulb; ZI, zona incerta. Scale bars: 1 mm.

Supplementary Figure S3: Analysis of cell death in Ambra Je^e* mutant embryos.

a-c, Embryo E8.5 transverse semi-thin sections stained with methylene blue. The staining revealed a limited amount of pyknotic nuclei (arrows) throughout the neural fold neuroepithelium of both wt (a) and Ambra 1&& mutant embryos (b). Quantification of pyknotic nuclei from three independent analyses is shown in the graph (c). No significant differences were observed between wt and mutant embryos (Student-s t test), d-o, TUNEL test of E9.0-E10.5 embryos showed increased TUNEL-positive cells in Ambra 1&& mutant neuroepithelium (right: e, h, j, n) compared with wt (left: d, g, i, m). Images were taken with a fluorescence microscope, converted into greyscale and inverted. TUNEL-positive cells are shown as black dots. As regards the E9.5 head sections, (g-h) are medial sections, whereas (i-j) are more lateral sections, (m-n) show transverse sections through the ElO.5 rostral spinal cord. Representative images of ElO.5 embryonic heads and E9.5 caudal spinal cords are shown in Figure lm-p. The proportion of TUNEL-positive cells from all sections was calculated on serial sections of three E9.0 (f), E9.5 (k), and ElO.5 embryonic heads (1), and rostral or caudal sections through the spinal cord at E9.5-E10.5 (o). Enhanced apoptotic cell death is detected in the mutant neural tube at these stages, in particular within the fore- and hindbrain regions. Fb, forebrain; Mb, midbrain; Hb, hindbrain; Dor, dorsal spinal cord. Scale bars are 50 μm in (a), 250 μm in (d, g) and 200 μm in (m).

Supplementary Figure S4: The molecular patterning of differentiation is slightly affected in the Ambra j^gttø embryonic spinal cord.

a, Immunohistochemical analysis of cross sections from E9.5 wild type (+/+) and Ambra le^t/gt (gt/gt) spinal cord, showing a displaced expression of the signalling morphogen sonic hedgehog (Shh). Scale bar: 78 μm. b, mRNA in situ hybridization analysis on cross sections from ElO.5 wild type (+/+) and Ambra 1&& (gt/gt) spinal cord, showing a moderately decreased expression of the differentiation markers neurogenin-2 (Ngn2) and Pax2 in the mutant. Scale bar: 150 μm. Sc, spinal cord; Not, notochord. Supplementary Figure S5: Quantification of cell proliferation in Ambra jgt/gt mutant embryos.

Proportion of proliferating cells from all neuroepithelial cells was calculated by counting BrdU-positive cells (a) or mitotic figures (b) on serial sections covering different brain regions at E8.5, E9.5 or E 10.5 (three embryos for each genotype at each stage), detected as described in Figure 1. The lower panels of (a) show representative transverse sections from E8.5 neural folds and sagittal sections from E9.5 and ElO.5 forebrains. BrdU positive cells (green) and nuclei of all neuroepithelial cells (red) are shown. The right panels of (b) show methylene-blue-stained representative transverse sections from E8.5 neural fold. Mitoses are indicated by arrows. The calculated mean values are given with standard deviation (±s.d.). *: P < 0.05. The difference between wild-type and Ambra 1 mutant values was analyzed using the two-tailed Student's t-test. Scale bars: 50 μm.

Supplementary Figure S6: Analysis of the effect of Ambra 1 modulation on Beclin l/Vps34-dependent autophagy.

a, Analysis of GFP-LC3 punctate staining after Ambra 1 down- regulation by RNA interference in rapamycin-treated 2F cells. A higher magnification of the images presented in Figure 3b is shown to highlight the staining. Scale bar: 20 μm. b, Rapamycin-treated cells upon Ambra 1 RNA interference, as in Figure 3b, were analysed for AVOs' formation by FACS measurement of Acridine Orange staining. Representative results (upper panels) are accompanied by a graph (lower panel) reporting the data from three experiments. The percent values indicated refer to each upper left quadrant, c, 2F cells transduced with Ambra 1 full length (FL), mutant Ambra 1 F2, F3 or Beclin 1 encoding retroviruses, as in Figure 3d, were analysed for AVOs' formation by FACS measurement of acridine orange staining. The values represent the mean ± s.d. of three experiments, d, Knockdown of Ambra 1 reduces the amount of Vps34 associated to Beclin 1 during rapamycin- induced autophagy. 24 h after transfection with the Ambra 1- specific siRNA Oligo 2 and a non-specific siRNA oligo (Control siRNA), 2F cells were treated with rapamycin for an additional 24 h. Protein extracts were then subjected to immunoprecipitation (IP) by using an anti- Beclin 1 antibody. Purified complexes and corresponding total extracts were finally analysed by Western blotting with anti-Vps34 (left panels) and anti- Beclin 1 (right panels) specific antibodies. The experiment is representative of three independent analyses with similar results, e-f, Decrease of Vps34 activity by Ambra 1 downregulation. 24 h after transfection with the Ambra 1-specific siRNA Oligo 2 or a non-specific siRNA oligo (Control siRNA) together with GFP-p40(phox)PX vector, HEK293 cells were starved for an additional 4 h. Purified complexes and corresponding total extracts were analysed as in (d) for Beclin 1/Vps34 complex formation (e). In parallel, GFP-p40(phox)PX was used as a non-invasive probe to measure PtdIns-3-P levels and distribution, since p40(phox) PX domain specifically binds to phosphatidylinositol 3- phosphate (PtdIns-3-P), the product of Vps34 enzymatic activity (f). The appearance of large PtdIns-3-P-rich vesicles during autophagy was detected using a confocal microscope (upper panels) and was quantified as the mean ± s.d. of three experiments (lower panel). The right panels show representative cells at higher magnification. A reduction in size and number of PtdIns-3-P-rich vesicles is detectable in cells where Ambra 1 has been down-regulated. Scale bar, 20 μm.

Supplementary Figure S7: Analysis of Ambra 1 role in cell proliferation.

a, Knockdown of Ambra 1 increases the proliferation rate of 2F cells. After transfection with siRNA oligo specific for Ambra 1, Beclin 1 or a non-specific siRNA oligo, cells were pulsed with BrdU for 30 min followed by BrdU-PI FACS analysis of the cell cycle distribution. Representative results are shown in the left panels, while the right panel reports the data from three experiments, b, Overexpression of Ambra 1 decreases the proliferation rate of 2F cells. 24 h after infection with βgal- or Ambra 1-expressing viruses, cells were pulsed with BrdU for 30 min followed by BrdU-PI FACS analysis of the cell cycle distribution (left panel, data from three experiments). Effective knockdown of Beclin 1 was confirmed by western blot analysis (right panel). Values represent the mean ± s.d. of three experiments.

Supplementary Figure S8: Analysis of autophagy in mouse embryonic fibroblasts (MEFs) from wt and Ambra !&& embryos.

a, The mTOR phosphorylation pathway is equally affected by rapa- mycin in Ambra 1&& _an(j _wt MEFs. Cells were treated for 24 or 48 h with rapamycin and phosphorylation of p70S6K, a target of mTOR, was analysed by immunoblot. Tubulin was used as a protein loading control, b, Rapamycin-induced autophagy is impaired in Ambra ls^t/s^t MEFs. Cells were treated with rapamycin, and the occurrence of autophagy was analysed 48 h later by FACS measurement of acridine orange staining. Representative results are shown in the upper panels, while the lower panel reports the data from three experiments, c, Nutrient-starvation-induced autophagy is impaired in Ambra 1 mutant MEFs. Cells infected with a GFP-LC3-encoding retrovirus were starved for 4 h and analysed for appearance of GFP-LC3 punctate staining. Results are the mean of three experiments, d-e, the Ambra 1 gene trap allele carries a loss-of-function mutation. HEK293 cells were trans- fected with constructs expressing the murine Ambra 1 wild type protein (Ambra 1 wt) or the mutant protein present in the Ambra l%^υ& mice (Ambra 1 gt). (d) Protein extracts from transfected cells were subjected to immunoprecipitation by using a mouse monoclonal anti- Beclin 1 antibody (IP Beclin 1). Purified complexes were revealed by western blot analysis using anti- Ambra 1 (WB Ambra 1, upper panels) or goat polyclonal anti- Beclin 1 (WB Beclin 1, lower panels) antibodies. Western blots of total extracts probed with the same antibodies are shown on the right side of each panel. Moreover, 24 h after transfection, the cells were treated with rapamycin and the occurrence of autophagy was analysed 24 h later by FACS measurement of acridine orange staining (e). Values in (b-c, e) represent the mean ± s.d. of three experiments. Supplementary Table Sl Statistics of neural tube defects in Ambra 1 mutant embryos.

Embryos from 30 litters generated by crossing of Ambra l^+/& parents were analyzed at stages E10.5-E14.5 in two different backgrounds, outbred mixed (NMRI; CDl x NMRI) and inbred C57B1/6. The occurrence of NTDs (exencephaly and/or spina bifida) is indicated.

Detailed description of the invention

As result of a large scale mutagenesis approach based on gene trapping in mouse (Stoykova, A., Chowdhury, K., Bonaldo, P., Torres, M. & Gruss, P. Deυ Dyn (1998) 212, 198-213), we isolated a novel gene involved in the development of the nervous system. This gene encodes a 1300 aa long WD-40 protein, highly conserved in vertebrates (see Fig. S4). This protein was called Ambra 1 (Activating molecule in Beclin 1- regulated autophagy) because of its roles as described below. In embryos homozygous for the "gene-£rapped, gt" allele {Ambra l&tf), a fusion transcript was revealed ("trapped" gene-lacZ) and its corresponding protein, whereas its protein was absent.

Since βgal activity in a "gene trap" line mimics the expression of the tagged endogenous gene (Skarnes, W.C., Moss, J. E., Hurtley, S.M. & Beddington R.S. Proc Natl Acad Sci U S A (1995) 92, 6592-6), an Ambra 1 expression analysis was performed after whole-mount βgal staining in heterozygous embryos throughout development and in postnatal brain (see Fig. 1).

At embryonic day 8.5 (E8.5) strong staining was detected throughout the neuroepithelium (Fig. lc-e and Supplementary Fig. S2a). At E 11.5 (Fig. ld-e) a robust expression was seen in the ventralmost part of the spinal cord, the encephalic vesicles, the neural retina, the limbs and the dorsal root ganglia. mRNA in situ hybridisation using an Ambra 1 probe confirmed the expression pattern (Fig. If and Supplementary Fig. S2b compared with βgal staining in Fig. S2c). At later develop- mental stages, the βgal staining becomes abundant in the entire developing nervous system as well as in other tissues (Supplementary Fig. S2d); a strong expression was observed in the cortex, hippocampus and striatum of postnatal brain (Supplementary Fig. S2 e-i).

Significantly, we found that homozygosity of Ambra 1 mutation causes embryonic lethality in outbred or inbred genetic backgrounds (Supplementary Table Sl). The majority of Ambra U^υ& embryos at stages E10-E14.5 exhibit neural tube defects (NTDs) detected as midbrain/hindbrain exencephaly and/or spina bifida (Fig. lg-j). Histological analysis revealed that in the mutant embryonic forebrain the vesicles are closed but displaced, the proliferative neuroepithelium shows an extensive overgrowth and the spinal cord is enlarged (Fig. Ik-I). Next, we performed a detailed analysis of cell death, differ- rentiation and growth in Ambra H^υ& embryos. An excess of apoptosis was present in selected areas of the mutant nervous system from E9.0 onwards (Fig. lm-p, Supplementary Fig. S3). Then, we observed that patterning regulators, such as Shh, FGF8, EnI, En2, Gli3, Pax2, Pax3, Pax5, Pax6, Ngn2, Dlxl,2, and Mashl, are normally present in the developing mutant brain and spinal cord, although a few of them are displaced or decreased (Supplementary Fig. S4). Finally, BrdU-uptake experiments and measurement of mitoses revealed a significant increase of proliferating cells in Ambra l^^ E8.5 neural folds (Fig. lq-r and Supplementary Fig. S5a-b). However, from E9.5 onwards the percentage of BrdU-positive cells out of total number of cells was similar between the mutant and the wild-type neuroepithelium, indicating that the hyperproliferation phenotype appears at the onset of neurulation, (Fig. ls-t and Supplementary Fig. S5a). Taken together, these results imply that Ambra 1 is necessary to control cell proliferation and guarantee cell survival during nervous system development.

In order to define the biological processes regulated by Ambra 1, it was decided to identify its molecular interactors by using a yeast two- hybrid approach. A cDNA encoding the first 667 aa of the human Ambra 1 homologue (Supplementary Fig. SIc) was used to screen a human brain cDNA expression library. One of the isolated clones (Fig. 2a) encodes the Atg protein Beclin 1, a component of the class III PI3K/Vps34 complex regulating autophagosome formation in mammals(Yue, Z., Jin, S., Yang, C, Levine, A. J. & Heintz, N. Proc Natl Acad Sci U S A (2003) 100, 15077-82; Kihara, A., Kabeya, Y., Ohsumi, Y. & Yoshimori, T. EMBO Rep (2001) 2, 330-5). Beclin 1- Ambra 1 interaction was confirmed in co-transduced human 2FTGH (2F) fibroblasts by co-immunoprecipitation (coIP) assays (Fig. 2b-c). In order to map the region responsible for Beclin 1 binding, we tested different Ambra 1 mutant constructs in coIP with Beclin 1. A central region of the protein (F2) is necessary and sufficient for an effective interaction with Beclin 1, whereas Ambra 1 amino-terminus (Fl) shows a minimal binding capability and its carboxy-terminus (F3) does not interact at all (Fig. 2b). Beclin 1-associated kinase Vps34 co- immunoprecipitates with Ambra 1, suggesting that Beclin 1, Vps34 and Ambra 1 are components of a multiprotein complex (Fig. 2c). Beclin 1-Ambra 1 interaction was also confirmed in developing brain samples, by co-immunoprecipitating the endogenous proteins (Fig. 2d). Consistent with their interaction, Beclin 1 and Ambra 1 showed a vesicular-like staining which mostly co-localised in 2F cells (Fig. 2e).

The Beclin 1-Ambra 1 interaction prompted us to test for a possible role of Ambra 1 in autophagy regulation. The occurrence of autophagy was analysed in 2F cells by assessing both conversion of LC3-I to LC3- II and its translocation to autophagic structures, two sequential steps in autophagosome formation (Kabeya, Y. et al. Embo J (2000) 19, 5720- 8). First, we tested the cell response to two autophagic stimuli, the mTOR-inhibitor rapamycin and nutrient deprivation, upon reduction of Ambra 1 levels by RNA interference. Down-regulation of Ambra 1 (Fig. 3a) resulted in a remarkable decrease in autophagy (Fig. 3b-c and Supplementary Fig. S6a). Since Beclin 1 overexpression has been shown to induce autophagy per se, we tested the effect of Ambra 1 overexpression on autophagy either at basal level or in rapamycin- treated 2F cells. A significant increase of both basal and rapamycin- induced autophagy was observed (Fig. 3d). Furthermore, it was checked whether Beclin 1 interaction was required for Ambra 1 -induced autophagy by using Ambra 1 Fl, F2 and F3 constructs (see Fig. 2b). F2 was able to induce autophagy at levels similar to the full length Ambra 1, F3 had no detectable effect (Fig. 3d), while Fl showed an intermediate effect, probably due to its residual Beclin 1-binding capability. The effects of Ambra 1 dysregulation on the autophagy process were confirmed by measurement of the formation of acidic vesicular organelles (AVOs) (Paglin, S. et al. Cancer Res. (2001) 61, 439-44) (suppl. S6b-c). Also, it was observed that Beclin 1-mediated induction of autophagy was drastically reduced upon Ambra 1 downregulation (Fig. 3e). It was shown that Ambra 1 downregulation leads to a reduced Beclin 1 capability to interact with its associated kinase Vps34, and to a decrease in Vps34 activity in cells induced to autophagy (Fig. 3f and Supplementary Fig. S6d-f). This finding suggests a role for Ambra 1 in favouring Beclin 1-Vps34 functional interaction. Taken together, these results show that Ambra 1 is a key factor in autophagy regulation and is required for Beclin 1 activity.

The autophagy-promoting activity of Beclin 1 has been associated with inhibition of cell proliferation (Liang, X. H. et al. Nature (1999) 402, 672-6). Given the observed hyperproliferative phenotype in Ambra 1 mutant embryos (see Fig. lq-t and Supplementary Fig. S5), the critical Ambra 1 dosage for cell growth control in vitro was studied. Ambra 1 down-regulation or its overexpression resulted in a significant increase or decrease of cell proliferation rate, respectively (Supplementary Fig. S7a-b). Moreover, inhibition of proliferation by Ambra 1 is dependent on Beclin 1, since this effect is abolished when Beclin 1 is down- regulated (Supplementary Fig. S7c). These observations support the view that dysregulation of Beclin 1-dependent autophagy is linked to abnormal cell proliferation.

Based on these in vitro results, we set out to establish whether autophagy defects were detectable in Ambra 1 mutants. Conversion of LC3-I to LC3-II was dramatically reduced in Ambra 1&& E14.5 embryos (Fig.4a, left panel). LC3 translocation was analysed by crossing Ambra l^+l& mice with transgenic mice ubiquitously expressing GFP-LC3 (Mizushima, N., Yamamoto, A., Matsui, M., Yoshimori, T. & Ohsumi, Y. MoI. Biol. Cell (2004) 15, 1101-11). Fluorescence microscopy examination of E 10.5 ne m^yoepithelium showed clusters of GFP-LC3 dots in wt embryos while a diffuse signal was detectable in Ambra l^^ littermates (Fig. 4a, right panels). Next, since autophagy impairment in the adult nervous system is accompanied by accumulation of ubiquitinated proteins (Hara, T. et al., Nature (2006).441, 885-9; Komatsu, M. et al., Nature (2006) 441, 880-4), we checked for ubiquitin expression in developing brains. Numerous ubiquitin-positive cells were present in the mutant neuroepithelium, with a predominant nuclear staining (Fig. 4b). Furthermore, autophagy defects were confirmed in Ambra is*^/gt mouse embryonic fibroblasts, as revealed by GFP-LC3 translocation to autophagosomal membranes, ultrastructural analysis and measurement of AVOs' formation (Fig, 4c-d, Supplementary Fig. S8a-c). Finally, a reduced activity of the Ambra 1 protein present in the Ambra ls^fs^t mice, in terms of capability to both interact with Beclin 1 and induce autophagy, was confirmed by transfection experiments using a construct mimicking the mutant fusion mRNA (Supplementary Fig. S8d-e). Thus, it is possible to conclude that the Ambra 1 gene trap allele carries a loss-of-function mutation which causes a severe defect in Ambra -l^tf- embryos.

It was therefore shown that deficiency of Ambra 1, a novel Beclin 1 partner mainly expressed in neural tube during development, impairs autophagy and results in severe neural tube disorders. Besides Ambra 1 putative role in human congenital brain malformation, its expression profile in adult brain compartments affected in neurodegenerative disorders (Supplementary Fig. S2), suggests a role for Ambra 1 in these diseases, as already shown for other autophagy regulators (Hara, T. et al., Nature (2006).441, 885-9; Komatsu, M. et al, Nature (2006) 441, 880-4). Our finding that the level of ubiquitinated proteins is dramatically enhanced in Ambra 1 mutant neuroepithelium, implies a role for auto- phagy-dependent protein turn-over in the control of neural development. A tuned balance of degradation of key neurodevelopmental regulators could be hypothesised as playing a role in the determination of cell fate and, when altered, may lead per se to the observed neural tube defects. As an alternative, such a phenotype could be the consequence of disturbed cell proliferation and apoptosis, caused by autophagy dysregulation. In fact, Ambra 1 deficiency during embryogenesis leads to an excess of cell proliferation at early stages followed by increased apoptosis in the neuroepithelium. Genetic evidence that autophagy controls cell proliferation was provided by experiments on dysregulation of Beclin 1 and its interactor UVRAG (Liang, X. H. et al., Nature (1999) 402, 672-6; Yue, Z., Jin, S., Yang, C, Levine, A. J. & Heintz, N. Proc Natl Acad Sci U S A (2003) 100, 15077- 82; Qu, X. et al. J Clin Invest (2003) 112, 1809-20; Liang, C. et al. Nat Cell Biol (2006) 8, 688-98).

The results suggest that the excess of cell proliferation in the Ambra 1 mutant is directly associated with Beclin 1 dysregulation. However, we cannot rule out that neuroepithelial cells deficient in autophagy undergo apoptosis in a cell-autonomous fashion, while the surrounding cells, lacking proper signalling (i.e., Shh-mediated mitogenic induction), overproliferate.

Chromosomal localization

For chromosomal localization of the gene trap insertion site, chromosomal slides were prepared from mouse spleen lymphocytes and hybridized with a biotinylated 2.7 Kb cDNA insert as described by Heng, H.H. and Tsui, L.C. Chromosoma (1993) 102, 325-332. Characterization of the Ambra 1& mutation

Several subsequent rounds of 5'RT-PCR and cDNA library screenings were performed to obtain the 4635 bp of compiled cDNA sequence. In detail, the acronym RACE (Rapid Amplification of cDNA Ends; 5' rapid amplification of cDNA ends) indicates a technique which is useful when extension of the known sequence portion of 5' terminus of a messenger is needed using a PCR. After PCR (Chowdhury, K., et al. Nucleic Acids Res. (1997) 25, 1531-1536), the cDNA template obtained from ES cells mRNA was amplified by two subsequent rounds of nested PCR. Primers were derived from the lacZ sequence. The products PCR were then blotted and hybridized with a probe specific for the vector splice-acceptor site.

Once isolated, the corresponding fragment was used as a probe to screen an embryonic mouse cDNA library. Upon identification of the relative cDNA in the databases, the insertion site was localized in the largest 50.36 Kb intron located between the 100 and 113 bp coding exons as shown in Supplementary Fig. SIb. The vector insertion disrupted the protein at amino acid position 841 (sequence Mm BAE33303.1), thus deleting 459 amino acids from the carboxy terminal end. For Ambra 1 mRNA analysis, PoIy(A)+ RNA was prepared from embryos and 2 μg were analysed by Northern blotting by standard procedure.

The construct Ambra 1^, encoding the Ambra 1-lacZ fusion protein present in the gene trap line was generated as follows. Briefly, a construct containing the sequences encoding the N-terminal part of the murine Ambra 1 protein as present in the gene trap line (PCR form cDNA clone IMAGE:5400613) was inserted in frame within a lacZ-pA vector. Next, the Ambra 1-lacZ-pA fragment (6.1 Kb) was isolated by Smal - Notl digestion and cloned into the pCMV Sport 6 (RZPD) vector in the transcription sense orientation digested in turn with the same restriction enzymes. In order to check whether the Ambra lε* construct was able to generate the expected Ambra 1-lacZ fusion protein in an eukaryotic cell line, we transfected it into the HEK293 cell line using the calcium phosphate method. Cells were fixed and stained for βgal activity 24 h after transfection. Blue staining was already visible after 2 h. Positive and negative control plasmids were also included in the assay, producing the expected results. The full length murine Ambra 1 clone (Ambra l^wt), used as a positive control, contained the entire coding sequence inserted into the pCMV Sport 6 vector in sense orientation.

Determination of Ambra 1 mutant embryos genotype

In order to genotype embryos, a Southern blot analysis was performed after digestion of genomic DNA with BamΗI restriction enzyme. The blots were incubated with radioactive probes specific for LacZ and for the housekeeping Fkh5 gene. Densitometric comparison between the two signals was performed in order to discriminate between heterozygous and homozygous embryos.

Early stage embryos were genotyped by RT-PCR using the following pairs of primers: δ'-AACGCATTTATACCCAGTCCA-S' (primer A) and δ'-ACCATAACGTATCGGCCCATC-S' (primer B) mapping upstream and downstream of the gene trap insertion site respectively, and primer A together with 5'-CCC AGT CAC GAC GTT GTA AAA-3' (primer C), the latter one mapping on the LacZ reporter sequence.

Histological and Immunocvtochemical Analyses

Whole-mount βgal staining of embryos at different embryonic stages was performed as described (Stoykova, A., Chowdhury, K., Bonaldo, P., Torres, M. & Gruss, P. Dev Dyn (1998) 212, 198-213.

In brief, once removed from the uterus, embryos were washed in PBS and fixed in 4% paraformaldehyde (PFA) for 30 minutes at 4⁰C. After washing they were incubated at 30⁰C in a staining solution containing X-gal up to 15 h. The stained embryos were then cut with a vibratome into 60 μm-thick sections and examined under a Zeiss Axioplan microscope.

For GFP-LC3 detection of Ambra U^;GFP-LC3 embryos, 10 μm-thick cryostat sections were prepared and analyzed under an Olympus 1X70 microscope, using softWoRx and DeltaVision Imaging Workstation (Applied Precision).

Sectioning specimens were impregnated with Paraffin wax (Paraplast Plus), embedded, transversally sectioned at 10 μm, and counterstained with cresyl-violet.

For immunohistochemical analysis, the Ambra lst^/gt _ancι _wiid type embryos at E9-9.5 were fixed in 4% PFA, washed and incubated in 25% sucrose before cryo-embedding. Sagittal and coronal 10 μm-thick sections were blocked in 5% horse serum and then incubated overnight with the primary antibodies. Fluorescent secondary antibodies (Molecular Probes) were used for the signal detection. Cellular nuclei were counterstained with DAPI or Propidium Iodide (Molecular Probes). Sections were then examined under an inverted fluorescence microscope (Nikon, Leica). For ubiquitin detection, a bio tiny lated goat anti-mouse IgG was used as a secondary antibody followed by incubation with horseradish peroxidase-conjugated streptavidin (Biogenex). The immunore action product was revealed using 3-amino- 9 ethyl-carbazole as chromogenic substrate and 0.01% H2O2 (Biogenex). In cell death experiments, apoptotic cells in embryonic sections were detected by DNA fragmentation (TUNEL) assay using a fluorescein based detection kit (Promega). Analysis of cell proliferation was performed following intraperitoneal injection of BrdU (Sigma-Aldrich) using an anti-BrdU antibody (Amersham). For confocal analysis of Beclin 1-Ambra 1 co-localisation and GFP-LC3-transduction experiments, cells were cultured on coverslips and fixed with 4% PFA in PBS followed by permeabilization with 0.1% Triton in PBS. Primary antibodies were incubated for 1 h at room temperature and visualised by means of Cy3 or Cy2 conjugated secondary antibodies (Jackson ImmunoResearch). Coverslips were mounted in Vectashield (Vector Laboratories) and examined under a confocal microscope (Leica TCS SP2).

Antibodies

Primary antibodies used in this study were: rabbit polyclonal anti-Myc Tag antibody (Upstate Biotechnology), goat polyclonal anti- Beclin 1 (Santa Cruz), mouse monoclonal anti- Beclin 1 (Becton-Dickinson), rabbit polyclonal anti-Vps34 (Invitrogen), rabbit polyclonal anti- phosphop70S6K (Cell Signalling), mouse monoclonal anti-tubulin (Sigma- Aldrich), rabbit polyclonal anti-LC3 (MBL), mouse monoclonal anti β-galactosidase (Promega), rabbit polyclonal anti-Shh (Santa Cruz), mouse monoclonal anti-ubiquitin (1B3, MBL). The antibody against Ambra 1 was raised by immunising rabbits with a peptide located at the amino-terminal portion of Ambra 1 protein (WEGKRVELPDSPRSC) (Sigma Genosys) and affinity -purified using the immunising peptide.

Yeast Two-Hybrid Screening

pGBKT7-Ambra 1 was generated by cloning the N-term region (nt: 1- 2001) of Ambra 1 into the EcoRI and BgIII sites of pGBKT7 (Clontech). A human brain cDNA library cloned in pACT2 (Clontech) was screened by cotransformation with the pGBKT7-Ambra 1 into AH 109 yeast strain. Positive clones were selected based on their growth on Trp, Leu, Ade and His dropout media (Clontech) containing 5mM 3-amino-l,2,4- triazole (3AT) (Sigma-aldrich). Recovery of the plasmids and μ- galactosidase assay were performed following the manufacturer's directions (Matchmaker Two-Hybrid System Protocol, Clontech). Cell Culture

The human fibrosarcoma 2FTGH cell line, kindly provided by S. Pellegrini, and the human embryonic kidney HEK293 cells, kindly provided by M. Pando, were cultured in Dulbecco's modified Eagle's medium (DMEM, Sigma-Aldrich) supplemented with 10% fetal calf serum (FCS, Sigma-Aldrich), L-glutamine, 1% penicillin/streptomycin solution at 37°C under 5% CO2. For autophagy induction, cells were treated with 2μM Rapamycin (Sigma-Aldrich) or cultured in Earle's balanced salt solution (EBSS). Murine embryonic fibroblast (MEFs) primary cells were prepared from E13.5 embryos, cultured in DMEM supplemented with 20% FCS and utilized for experiments between the second and the seventh passages.

RNA interference

siRNA oligoribonucleotides corresponding to the human Ambra 1 and Beclin 1 cDNA sequences were purchased from Dharmacon (Ambra 1) and Invitrogen

(Beclin 1, BECNl Stealth Select RNAi; No. HSS 112742). Ambra 1 siRNA 1 : δ'-AGAACTGCAAGATCTACAA-S' Ambra 1 siRNA 2 : δ'-GGCCTATGGTACTAACAAA-S'

2x105 cells/well were transfected with 100 pmol siRNA in 6 well plates by Lipofectamine 2000 (Invitrogen) as indicated by the supplier. Trans- fection was repeated on two consecutive days to increase transfection efficiency. 24 h after transfection, cells were trypsinized, plated at 5xlO⁴ cells/well in 6 well plates and treated with rapamycin (Sigma- Aldrich). RNA decrease was checked by real time PCR and western blot 48 h after transfection. Cell proliferation analysis

In order to measure percentage of cells entering in S phase, BrdU was added to a final concentration of 20 μM and incubated at 37°C for 20 minutes. Cells were detached using trypsin, washed with PBS and fixed using cold 70% ethanol. DNA was denaturated by adding 4N HCl for 1 h at room temperature, followed by neutralization with 0,5M Na2B₄O₇ pH 8,5. Cells were then incubated with anti-BrdU Alexa Fluor 488 (0,2 μg/ml, Molecular Probes) in PBS + 0,5% Tween 20 for 1 h at 37°C. Finally, cells were washed with PBS, resuspended in PBS + 5 μg/ml Propidium Iodide and analysed using a FACScan flow cytometer (Becton Dickinson).

cDNA cloning and retroviral vectors

For retroviral expression, all constructs were cloned in a modified version of pLPCX vector (Clontech) in which the Sail restriction site within the puromycin resistance gene was mutated by PCR in order to render the Sail site unique in the multiple cloning site (MCS). An additional CMV promoter was cloned in front of the 5' LTR region in order to increase viral production ⁴ (Ranga, U. et al. J Virol (1997) 71, 7020-7029).

Human Ambra 1 cDNA (Fig. SIc) was obtained by PCR amplification from a human brain cDNA library (Clontech) and cloned in EcoRI and Notl sites of pLPCX. The various pCLPCX-Ambra 1 mutants (NT, CT, Fl, F2, F3, F4, F5) were generated by PCR amplification using specific primers. Myc-tag Ambra 1 fusion proteins were obtained by inserting 5 copies of the myc epitope in the HindϊlI-EcoRI restriction sites of pLPCX. Human Beclin 1 cDNA was obtained by PCR amplification from a HeLa cDNA library (Clontech) and cloned into the EcoRI and Sail sites of pCLPCX. Human LC3 cDNA was obtained by PCR amplification from a HeLa cDNA library (Clontech) and inserted into the EcoRI and Sail sites of pEGFP/C2 (Clontech). GFP-LC3 fusion cDNA was then excised from the pEGFP plasmid by cutting with Nhel and Sail restriction enzymes and inserted into pLPCX digested with EcoRI and Sail, with Nheϊ and EcoRI DNA ends blunted to allow the ligation. pLPCX-lacZ-Myc was obtained by digesting the pcDNA 3.1 myc-His/lacZ plasmid (Invitrogen) with Hindlll-Pmel and inserted into pLPCX digested with Hindlll-Notl (blunted). Human p40(phox)PX cDNA (encoding aa 13-140 of human p40-phox protein) was obtained by PCR amplification from a HeLa cells cDNA library (Clontech), inserted into the EcoRI and Sail sites of pEGFP/C2 (Clontech) and transfected into the HEK293 cell line using the calcium phosphate method. The sequences of all PCR amplified cDNAs were verified by DNA sequencing analysis.

Retrovirus Generation and Infection

Fifteen μg of the retroviral vectors were co-transfected with 5 μg of an expression plasmid for the vesicular stomatitis virus G protein (Somia, N. V. et al. Proc Natl Acad Sci U S A (1999) 96, 12667-72) by using the calcium phosphate method. 48 h later, the supernatant containing the retroviral particles was recovered and supplemented with polybrene (4 μg /mL). 2FTGH or MEF cells were infected by incubation with retroviral containing supernantant for 6-8h.

Real Time PCR

RNA was prepared with Trizol reagent (Invitrogen). cDNA synthesis was generated using the reverse transcription kit (Promega) according to manufacturer recommendations. Real time PCR reactions were performed with the LightCycler (Roche). The LightCycler FastStart DNA Master SYBR Green I (Roche) was used to produce fluorescent- labeled PCR products during each cycling of the amplification reaction. Primer sets for all amplicons were designed using the Primer-Express 1.0 software system. Ambra 1 forward: δ'-AACCCTCCACTGCGAGTTGA-S' Ambra 1 reverse: δ'-TCTACCTGTTCCGTGGTTCTCC-S' L34 forward: δ'-GTCCCGAACCCCTGGTAATAGA-S' L34 reverse: δ'-GGCCCTGCTGACATGTTTCTT-S'

The result of the fluorescent PCR was expressed as the threshold cycle (CT). The ΔCT is the difference between the CT for a specific mRNA and the CT for a reference mRNA, L34. To determine relative mRNA levels, 2 was raised to the power of Δ CT (the difference between the ΔCT from treated cells and the CT from untreated cells). L34 mRNA level was used as an internal control because this gene was shown to be stable with cell induction, β-actin and GAPDH level were used as additional controls to confirm significant decreases.

Immunoprecipitation and Western Blot Assays

In immunoprecipitation experiments, cells or tissues were lysed in HEMG buffer (25 mM Hepes [pH 8.0], 100 mM NaCl, 0.5% Nonidet P- 40, 0.1 mM EDTA 10% glycerol) plus protease and phosphatase inhibitors (Protease inhibitor cocktail, ImM Sodium Fluoride, ImM Sodium orthovanadate, ImM Sodium molibdate; Sigma- Aldrich). Lysates (l-2mg) were then incubated with rotation at 4° C for 30 min. Following a centrifugation at 4° C for 10 min at 1300Ox g to remove insoluble debris, equal amounts of protein were incubated with 30 μl of monoclonal anti-cMyc antibody conjugated with protein A agarose beads (BD Biosciences) with rotation at 4° C for 4 hr, or 2μg of Beclin 1 or anti-Ambra 1 antibodies overnight at 4° C, followed by 60 min incubation with 20 μl of protein A/G sepharose beads (Amersham Bioscience). The beads were finally collected by centrifugation and washed four times with the HEMG buffer. Proteins bound to the beads were eluited with 50 μl of SDS-PAGE sample buffer and heated to 70° C for 10 min. Proteins were separated on NuPAGE Bis-Tris gel (Invitrogen) and electroblotted onto nitrocellulose membranes (Protran, Schleicher & Schuell). Blots were incubated with primary antibodies in 5% non-fat dry milk in PBS plus 0,1% Tween20 overnight at 4°C. Detection was achieved using horseradish peroxidase -conjugate secondary antibody (Jackson Laboratory) and visualized with ECL plus (Amersham Bioscience).

Embryos from stage Ell-11.5 to E14-14.5 were subjected to mechanical lysis in 50 mM Tris HCl pH 7.5, 32OmM Sucrose, 50 mM NaCl, 1% Triton X-100 and protease inhibitors. Solubilized proteins were quantified by a Biorad protein assay and denatured by adding a concentrated boiling Laemmli buffer. 30-50 μg of proteins were analyzed by SDS-PAGE and immunoblotting.

Statistical analysis

Microsoft Excel was used for statistical analysis. Statistical signify - cance was determined using the Student's t-test. A P value of equal to or less than 0.05 was considered significant.

Claims

1. Isolated nucleic acid codifying for a human wild type Ambra 1 polypeptide chain or a modified sort of said polypeptide, which is functionally equivalent or associated with the predisposition to pathologies of the cerebrospinal apparatus.

2. Isolated nucleic acid codifying for a human wild type Ambra 1 polypeptide chain, mutated by insertion, deletion, nonsense mutation, substitution of one or more nucleotides.

3. Isolated nucleic acid according to claims 1 or 2 wherein said nucleic acid is DNA.

4. Isolated nucleic acid according to claims 1 or 2 wherein said nucleic acid is RNA.

5. Isolated nucleic acid according to claim 3, wherein said nucleic acid is a cDNA.

6. Isolated nucleic acid according to claim 3, wherein said nucleic acid is a genomic cDNA.

7. Isolated nucleic acid according to claim 1, wherein the wild type sequence of Ambra 1 has a amino acid sequence substantially similar to the sequence described as SEQ. ID NO 2.

8. Isolated nucleic acid according to claim 1, wherein said nucleic acid comprises a nucleic acid having a sequence substantially similar to the sequence described as SEQ. ID NO 2.

9. Vector comprising the isolated nucleic acid according to claim 1- 8, operatively linked to a promoter for RNA transcription.

10. Vector according to claim 9, wherein the promoter sequence comprises a bacterial, yeast, insect, or mammalian promoter.

11. Vector according to claim 10, further comprising a plasmid, a cosmid, an artificial yeast chromosome (YAC), bacteriophage or eukaryotic viral DNA.

12. Vector according to claim 11, wherein in the retroviral modified vector pLPCX the cDNA of Ambra 1 having a sequence according to claim 1 is inserted between the EcoRI site and the Notl site.

13. Expression system for the manufacturing of a polypeptide comprising a vector according to claim 9 in a suitable host.

14. Expression system for the vector according to claim 13, wherein the suitable host comprises a eukaryotic or prokaryotic cell.

15. Expression system for the vector according to claim 14, wherein the prokaryotic cell comprises a bacterial cell.

16. Expression system for the vector according to claim 14, wherein the eukaryotic cell comprises a yeast, insect, plant, or mammalian cell.

17. Method for the manufacture of a polypeptide comprising growing the expression system according to claim 13, in conditions which allow the production of the polypeptide and the recovery of the polypeptide thus produced.

18. Method for the manufacture of a purified polypeptide comprising the steps of:

(a) introducing the vector according to claims 9-12 in a suitable host cell;

(b) growing the resulting host cell in such a way as to produce the polypeptide;

(c) recovery of the polypeptide produced during step (b); (d) purifying the polypeptide thus recovered.

19. Method according to claim 18, wherein the vector comprises a plasmid, a cosmid, an artificial yeast chromosome (YAC), a bacteriophage or an eukaryotic viral DNA.

20. Method according to claim 18, wherein the suitable host cell comprises a yeast, insect, plant or mammalian cell.

21. Wild-type isolated or recombinant Ambra 1 protein.

22. Polynucleotide comprising the SEQ ID NO 1.

23. Polynucleotide comprising the SEQ ID NO 2.

24. Oligonucleotide of at least 15 nucleotides in length able to specifically hybridize to one sequence only within a nucleic acid codifying for a wild type Ambra 1 protein without hybridizing to any nucleotide sequence within a nucleic acid codifying for a mutant human Ambra 1 protein.

25. Oligonucleotide of at least 15 nucleotides in length able to specifically hybridize to one nucleotide sequence only within a nucleic acid codifying for a mutant Ambra 1 protein without hybridizing to any nucleotide sequence within a nucleic acid codifying for a wild type Ambra 1 protein.

26. Oligonucleotide according to claim 24 or claim 25, wherein the nucleic acid is DNA.

27. Oligonucleotide according to claim 24 or claim 25, wherein the nucleic acid is RNA.

28. Oligonucleotide of the sequence Ambra 1 siRNA 1: 5'- AGAACTGCAAGATCTACAA-3' able to interfere with the expression of human Ambra 1.

29. Oligonucleotide of the sequence Ambra 1 siRNA 2: 5'- GGCCTATGGTACTAACAAA-3' able to interfere with the expression of human Ambra 1.

30. Peptide chosen from among the sequences shown in Figure Sl and in the sequence listing.

31. Monoclonal or polyclonal antibody against Ambra 1 protein.

32. Polyclonal antibody against Ambra 1 protein obtained by immunizing a suitable animal with the peptide of claim 30.

33. Method for determining if a subject suffers from a disease of the cerebrospinal apparatus or is predisposed to be affected by a disease of the cerebrospinal apparatus comprising:

(a) drawing a blood sample from a subject;

(b) testing whether the nucleic acid in said sample is or is not derived from a nucleic acid codifying for a human mutant of Ambra 1 such as to cause a predisposition to a cerebrospinal disease.

34. Method according to claim 33, wherein the sample of nucleic acid comprises an mRNA corresponding to a transcript of the DNA codifying for a mutant Ambra 1 protein and wherein the testing of step (b) comprises:

(i) contacting the mRNA with the oligonucleotide of claims 24-25 in such conditions as to allow binding of mRNA to the oligonucleotide to create a complex;

(ii) isolating the complex thus formed; and

(iii) identifying the mRNA in the complex thus formed to determine whether the mRNA, or its derivative, is a nucleic acid codifying for a human mutant of Ambra 1 protein.

35. Method according to claim 33 wherein the testing of step (b) comprises:

(i) amplifying the nucleic acid present in the sample of step (a) and; (ii) testing the presence of the human mutant of Ambra 1 in the amplified nucleic acid.

36. Method according to claims 33-35, wherein the nucleic acid is labelled with a marker chosen from among a radioactive isotope, a fluorophore or an enzyme.

37. Method according to claims 33-36, wherein the nucleic acid is a sample of genomic DNA.

38. Method according to claims 33-37 wherein the sample is blood, tissue or serum.

39. Method according to claims 33-38, wherein the sample is a fetal sample.

40. Ambra 1 protein or DNA, fractions or mutants thereof, for use in the pharmaceutical field.

41. Ambra 1 protein or DNA, fractions or mutants thereof, for use in the diagnosis, prevention and treatment of cerebrospinal apparatus diseases.

42. Pharmaceutical composition comprising an Ambra 1 protein or DNA, fractions or mutants thereof.

43. Pharmaceutical composition comprising an Ambra 1 protein or DNA, fractions or mutants thereof, for use in the diagnosis, prevention and treatment of cerebrospinal apparatus diseases.

44. Pharmaceutical composition according to claims 42-43, further comprising a pharmaceutically acceptable carrier.

45. Pharmaceutical composition according to claims 42-44, in formulations for topical, oral, aerosol, subcutaneous, infusional , intralesional, intramuscular, intravenous, and intraperitoneal use.

46. Use of Ambra 1, fractions or mutants thereof, for the diagnosis, prevention and treatment of the cerebrospinal apparatus.

47. Use of a monoclonal or polyclonal antibody to detect the presence of Ambra 1 protein or mutants thereof in biological samples.

48. Diagnostic kit comprising the Ambra 1 protein or DNA, fractions or mutants thereof, or the antibody according to claims 32-33 in combination with reagents and laboratory consumables, and instruct- tions to carry out the analytical method.