Barbara Tappino

Identification and characterization of 15 novel GALC gene mutations causing Krabbe disease.

The characterization of the underlying GALC gene lesions was performed in 30 unrelated patients affected by Krabbe disease, an autosomal recessive leukodystrophy caused by the deficiency of lysosomal enzyme galactocerebrosidase. The GALC mutational spectrum comprised 33 distinct mutant (including 15 previously unreported) alleles. With the exception of 4 novel missense mutations that replaced evolutionarily highly conserved residues (p.P318R, p.G323R, p.I384T, p.Y490N), most of the newly described lesions altered mRNA processing. These included 7 frameshift mutations (c.61delG, c.408delA, c.521delA, c.1171_1175delCATTCinsA, c.1405_1407delCTCinsT, c.302_308dupAAATAGG, c.1819_1826dupGTTACAGG), 3 nonsense mutations (p.R69X, p.K88X, p.R127X) one of which (p.K88X) mediated the skipping of exon 2, and a splicing mutation (c.1489+1G>A) which induced the partial skipping of exon 13. In addition, 6 previously unreported GALC polymorphisms were identified. The functional significance of the novel GALC missense mutations and polymorphisms was investigated using the MutPred analysis tool. This study, reporting one of the largest genotype‐phenotype analyses of the GALC gene so far performed in a European Krabbe disease cohort, revealed that the Italian GALC mutational profile differs significantly from other populations of European origin. This is due in part to a GALC missense substitution (p.G553R) that occurs at high frequency on a common founder haplotype background in patients originating from the Naples region.

Molecular analysis of NPC1 and NPC2 gene in 34 Niemann–Pick C Italian Patients: identification and structural modeling of novel mutations

Niemann–Pick C, the autosomal recessive neuro-visceral disease resulting from a failure of cholesterol trafficking within the endosomal–lysosomal pathway, is due to mutations in NPC1 or NPC2 genes. We characterized 34 unrelated patients including 32 patients with mutations in NPC1 gene and two patients in NPC2 gene. Overall, 33 distinct genotypes were encountered. Among the 21 unpublished NPC1 alleles, 15 were due to point mutations resulting in 13 codon replacements (p.C100S, p.P237L, p.R389L, p.L472H, p.Y634C, p.S636F, p.V780G, p.Q921P, p.Y1019C, p.R1077Q, p.L1102F, p.A1187V, and p.L1191F) and in two premature stop codons (p.R934X and p.Q447X); a new mutant carried two in cis mutations, p.[L648H;M1142T] and four other NPC1 alleles were small deletions/insertions leading both to frame shifts and premature protein truncations (p.C31WfsX26, p.F284LfsX26, p.E1188fsX54, and p.T1205NfsX53). Finally, the new intronic c.464-2A>C change at the 3′ acceptor splice site of intron 4 affected NPC1 messenger RNA processing. We also found a new NPC2 mutant caused by a change of the first codon (p.M1L). The novel missense mutations were further investigated by two bioinformatics approaches. Panther proein classification system computationally predicted the detrimental effect of all new missense mutations occurring at evolutionary conserved positions. The other bioinformatics approach was based on prediction of structural alterations induced by missense mutations on the NPC1 atomic models. The in silico analysis predicted protein malfunctioning and/or local folding alteration for most missense mutations. Moreover, the effects of the missense mutations (p.Y634C, p.S636F, p.L648H, and p.V780G) affecting the sterol-sensing domain (SSD) were evaluated by docking simulation between the atomic coordinates of SSD model and cholesterol.

Molecular Characterization of 22 Novel UDP-N-Acetylglucosamine-1-Phosphate Transferase α- and β-Subunit (GNPTAB) Gene Mutations Causing Mucolipidosis Types IIα/β and IIIα/β in 46 Patients

Mutational analysis of the GNPTAB gene was performed in 46 apparently unrelated patients with mucolipidosis IIα/β or IIIα/β, characterized by the mistargeting of multiple lysosomal enzymes as a consequence of a UDP‐GlcNAc‐1‐phosphotransferase defect. The GNPTAB mutational spectrum comprised 25 distinct mutant alleles, 22 of which were novel, including 3 nonsense mutations (p.Q314X, p.R375X, p.Q507X), 5 missense mutations (p.I403T, p.C442Y, p.C461G, p.Q926P, p.L1001P), 6 microduplications (c.749dupA, c.857dupA, c.1191_1194dupGCTG, c.1206dupT, c.1331dupG, c.2220_2221dupGA) and 8 microdeletions (c.755_759delCCTCT, c.1399delG, c.1959_1962delTAGT, c.1965delC, c.2550_2554delGAAAA, c.3443_3446delTTTG, c.3487_3490delACAG, c.3523_3529delATGTTCC). All micro‐duplications/deletions were predicted to result in the premature termination of translation. A novel exonic SNP (c.303G>A; E101E) was identified which is predicted to create an SFRS1 (SF2/ASF) binding site that may be of potential functional/clinical relevance. This study of mutations in the GNPTAB gene, the largest yet reported, extends our knowledge of the mutational heterogeneity evident in MLIIα/β/MLIIIα/β.

Identification and molecular characterization of six novel mutations in the UDP-N-acetylglucosamine-1-phosphotransferase gamma subunit ( GNPTG) gene in patients with mucolipidosis III gamma

Mucolipidosis type III (MLIII) is an autosomal recessive disorder affecting lysosomal hydrolase trafficking. In a study of 10 patients from seven families with a clinical phenotype and enzymatic diagnosis of MLIII, six novel GNPTG gene mutations were identified. These included missense (p.T286M) and nonsense (p.W111X) mutations and a transition in the obligate AG‐dinucleotide of the intron 8 acceptor splice site (c.610–2A>G). Three microdeletions were also identified, two of which (c.611delG and c.640_667del28) were located within the coding region whereas one (c.609+28_610‐16del) was located entirely within intron 8. RT‐PCR analysis of the c.610–2A>G transition demonstrated that the change altered splicing, leading to the production of two distinct aberrantly spliced forms, viz. the skipping of exon 9 (p.G204_K247del) or the retention of introns 8 and 9 (p.G204VfsX28). RT‐PCR analysis, performed on a patient homozygous for the intronic deletion (c.609+28_610‐16del), failed to detect any GNPTG RNA transcripts. To determine whether c.609+28_610‐16del allele‐derived transcripts were subject to nonsense‐mediated mRNA decay (NMD), patient fibroblasts were incubated with the protein synthesis inhibitor anisomycin. An RT‐PCR fragment retaining 43 bp of intron 8 was consistently detected suggesting that the 33‐bp genomic deletion had elicited NMD. Quantitative real‐time PCR and GNPTG western blot analysis confirmed that the homozygous microdeletion p.G204VfsX17 had elicited NMD resulting in failure to synthesize GNPTG protein. Analysis of the sequences surrounding the microdeletion breakpoints revealed either intrinsic repetitivity of the deleted region or short direct repeats adjacent to the breakpoint junctions. This is consistent with these repeats having mediated the microdeletions via replication slippage and supports the view that the mutational spectrum of the GNPTG gene is strongly influenced by the properties of the local DNA sequence environment. Hum Mutat 30:1–7, 2009.

Enigmatic In Vivo iduronate-2-sulfatase (IDS) mutant transcript correction to wild-type in Hunter syndrome

Sequence analysis of the X‐linked iduronate‐2‐sulfatase (IDS) gene in two Hunter syndrome patients revealed a lack of concordance between IDS genomic DNA and cDNA. These individuals were found to be hemizygous respectively for a nonsense mutation [c.22C>T;p.R8X] and a frameshift micro‐insertion [c.10insT;p.P4Sfs] in their genomic DNA. However, both wild‐type and mutant IDS sequences were evident upon cDNA analysis. Similar discrepant results were also obtained in a third unrelated patient carrying the same p.R8X mutation. Since both p.R8X mutations were inherited from carrier mothers, somatic mosaicism could be excluded. Although the presence of wild‐type IDSmRNA‐transcripts was confirmed in all three patients by restriction enzyme digestion, clone sequencing, pyrosequencing and single nucleotide primer extension (SNuPE), no wild‐type IDS genomic sequence was detectable. The relative abundance of wild‐type and mutation‐bearing IDS‐transcripts in different tissues was quantified by SNuPE. Although IDS transcript levels, as measured by real‐time PCR, were reduced (51‐71% normal) in these patients, some wild‐type IDS protein was detectable by western blotting. Various possible explanations for these unprecedented findings (e.g. accidental contamination, artefactual in vitro nucleotide misincorporation, malsegregation of an extra maternal X‐chromosome) were explored and experimentally excluded. PCR‐based discriminant assay and segregation analysis of a linked IDS polymorphism (rs1141608) also served to exclude the presence of IDS cDNA derived from the maternal wild‐type chromosome. Although it remains to be formally demonstrated by direct experimentation, the intriguing possibility arises that we have observed the in vivo correction of heritable gene lesions at the RNA level operating via a correction mechanism akin to RNA‐editing.

Restoration of the Normal Splicing Pattern of the PLP1 Gene by Means of an Antisense Oligonucleotide Directed against an Exonic Mutation

An exonic missense mutation, c.436C>G, in the PLP1 gene of a patient affected by the hypomyelinating leukodystrophy, Pelizaeus–Merzbacher disease, has previously been found to be responsible for the alteration of the canonical alternative splicing profile of the PLP1 gene leading to the loss of the longer PLP isoform. Here we show that the presence of the c.436C>G mutation served to introduce regulatory motifs that appear to be responsible for the perturbed splicing pattern that led to loss of the major PLP transcript. With the aim of disrupting the interaction between the PLP1 splicing regulatory motifs and their cognate splicing factors, we designed an antisense oligonucleotide-based in vitro correction protocol that successfully restored PLP transcript production in oligodendrocyte precursor cells.

Concentration-Dependent Metabolic Effects of Metformin in Healthy and Fanconi Anemia Lymphoblast Cells

Metformin (MET) is the drug of choice for patients with type 2 diabetes and has been proposed for use in cancer therapy and for treating other metabolic diseases. More than 14,000 studies have been published addressing the cellular mechanisms affected by MET. However, several in vitro studies have used concentrations of the drug 10–100‐fold higher than the plasmatic concentration measured in patients. Here, we evaluated the biochemical, metabolic, and morphologic effects of various concentrations of MET. Moreover, we tested the effect of MET on Fanconi Anemia (FA) cells, a DNA repair genetic disease with defects in energetic and glucose metabolism, as well as on human promyelocytic leukemia (HL60) cell lines. We found that the response of wild‐type cells to MET is concentration dependent. Low concentrations (15 and 150 µM) increase both oxidative phosphorylation and the oxidative stress response, acting on the AMPK/Sirt1 pathway, while the high concentration (1.5 mM) inhibits the respiratory chain, alters cell morphology, becoming toxic to the cells. In FA cells, MET was unable to correct the energetic/respiratory defect and did not improve the response to oxidative stress and DNA damage. By contrast, HL60 cells appear sensitive also at 150 μM. Our findings underline the importance of the MET concentration in evaluating the effect of this drug on cell metabolism and demonstrate that data obtained from in vitro experiments, that have used high concentrations of MET, cannot be readily translated into improving our understanding of the cellular effects of metformin when used in the clinical setting.