Search

A correction to this article has been published and is linked from the HTML and PDF versions of this paper. The error has been fixed in the paper.

Introduction: Autism Spectrum Disorder (ASD) is a neurodevelopmental condition characterized by communication impairments, limited social interaction, restricted interests, repetitive behaviors and stereotypies. It manifests within the first 3 years of age and lasts for a lifetime with dramatic personal, familial and social consequences. ASD affects much more males than females (male:female=5:1) and its prevalence, that is continuously increasing, interests 2.24% of children and 1% of the general population [1,3]. Although genetics play a key role in ASD, its etiology is complex. Since most of individuals with ASD suffer from additional comorbidities including gastrointestinal disorders, intestinal permeability, inflammation and allergies, a gene-environment interaction has been proposed as ASD triggering [3]. Dysbiosis has been frequently associated with many neurological disorders including ASD. Thanks to the last advances in metagenomics, much progress has been made in the knowledge of gut microbiota profile and role, especially the prokaryotic one, but literature lacks of studies about eukaryotic colonizer of human intestine and their role in human health. Aims Here, employing different bioinformatics tools, we propose a metagenomics pilot study to define the prokaryotic and eukaryotic gut microbiota of children with ASD and neurotypical controls. The aims are to test different metagenomics pipelines and set up the more performing bioinformatics conditions to identify ASD microbial biomarkers useful for patient stratification and personalized treatments. Materials and Methods: We isolated DNA from stools collected from 6 children with ASD (5 males and 1 female) and 6 neurotypical controls matching for age and sex. Both 16S and 18S were amplified for each DNA and Illumina libraries prepared. NGS was performed by Illunima MiSeq platform coupled with Flowcell V3 2X300 and forward and reverse reading, reaching about 22Milion of sequences. As for bioinformatics analysis, three different software with several pipelines were applied: the automatic pipeline of SILVAngs analysis platform (https://ngs.arb-silva.de/silvangs/) [4], the MiSeq SOP pipeline of Mothur (https://mothur.org/) [5], and two pipelines of Qiime2 (https://qiime2.org/) [6]. The latter includes the Dada2 pipeline and the Deblur one. All the analyses were performed against SILVAv132 database [7], the only one that includes both 16s and 18S reference database. Results: The results obtained from the four tools are superimposable, also at different level of taxonomy detail. Restricting to Bacteria results, the identified genera are comparable with literature data [8]. As for Fungi, this first round of analysis doesn't return relevant differences between patients and controls. Further studies are needed to set up a pipeline specifically for the 18S datasets. Acknowledgements: EU project GEMMA (grant agreement No 825033), EPTRI and CNRBiOmics. Istituto San Vincenzo Erba and Albese, Italy

Autism spectrum disorder (ASD) is a complex neurodevelopmental condition, prevalence is 1:68. Although genetics plays a key role in ASD etiology, only 30% of patients exhibit ASD associated genetic variants that are detectable by CGH, GWAS and/or exome NGS approach and hundreds of genes have been involved. Here, we tested the hypothesis that variants in non-coding and/or NGS unmappable regions could be involved in ASD and detected by long-reads whole-genome sequencing (LR-WGS). This approach allows to probe previously unreadable genomic tracts and discover disease-associated phased loci across very long haplotype blocks. Thus, we performed LR-WGS, by 10XGenomics technology, on genomic DNA isolated from 10 children with ASD (7 males and 3 females, including 2 couples of male siblings and 1 couple of male-female siblings) followed by bioinformatics analysis performed by 10X-Long Ranger pipeline. All the sequences passed the QC test, and the longest phase block reached 9M bp in length. We first looked at large structural variants (SVs) detectable only by long-read approach. We found a total of 204 hetero- or homo-zygous large SVs (deletions, duplications, inversions and breakends), up to 30 per subject and affecting 52 distinct genomic regions of 10^5 bp in average within almost all chromosomes. Interestingly, each SV affected from 1 up to 10 subjects and 51 out of 53 SVs involved genes that have never been associated with autism (not listed in SFARI database). All genes were submitted to DAVID 6.8 and METASCAPE gene enrichment tools evidencing some pathways including that of immune response and olfactory transduction. These SVs will be validated on an independent set of DNA samples previously collected from children with ASD and healthy controls. Resulting data will be integrated with clinical data to find genotype-phenotype associations. Small variants will be analyzed and validated, too. In conclusion, LR-WGS successfully discovered 51 new gene variants in 10 patients with ASD that, if confirmed, could explain some pathogenetic mechanisms of the disorder and represent possible diagnostic biomarkers for patient stratification and personalized medicine approaches.

HCV is an important cause of hepatocellular carcinoma (HCC). HCV NS5A domain-1 interacts with cellular proteins inducing pro-oncogenic pathways. Thus, we explore genetic variations in NS5A domain-1 and their association with HCC, by analyzing 188 NS5A sequences from HCV genotype-1b infected DAA-naïve cirrhotic patients: 34 with HCC and 154 without HCC. Specific NS5A mutations significantly correlate with HCC: S3T (8.8% vs. 1.3%, p = 0.01), T122M (8.8% vs. 0.0%, p &lt, 0.001), M133I (20.6% vs. 3.9%, p &lt, 0.001), and Q181E (11.8% vs. 0.6%, p &lt, 0.001). By multivariable analysis, the presence of &gt, 1 of them independently correlates with HCC (OR (95%CI): 21.8 (5.7–82.3), p &lt, 0.001). Focusing on HCC-group, the presence of these mutations correlates with higher viremia (median (IQR): 5.7 (5.4–6.2) log IU/mL vs. 5.3 (4.4–5.6) log IU/mL, p = 0.02) and lower ALT (35 (30–71) vs. 83 (48–108) U/L, p = 0.004), suggesting a role in enhancing viral fitness without affecting necroinflammation. Notably, these mutations reside in NS5A regions known to interact with cellular proteins crucial for cell-cycle regulation (p53, p85-PIK3, and β-catenin), and introduce additional phosphorylation sites, a phenomenon known to ameliorate NS5A interaction with cellular proteins. Overall, these results provide a focus for further investigations on molecular bases of HCV-mediated oncogenesis. The role of theseNS5A domain-1 mutations in triggering pro-oncogenic stimuli that can persist also despite achievement of sustained virological response deserves further investigation.

CXCR3 is a G-protein coupled receptor expressed principally on leukocytes, monocytes and epithelial cells; it is involved in leukocyte traffic, integrin activation, cytoskeletal changes and chemotactic migration, by binding to its classical ligands, CXCL-9/10/11 (1). Three splicing variants of CXCR3 are known: CXCR3a, the most common isoform, consisting of 368 amino acid residues; CXCR3b resulting from an alternative splicing of the CXCR3 mRNA with a 52 aa extended N-terminal domain when compared to the isoform a, while CXCR3-alt is a significantly truncated variant not involved in classical ligand binding (1). CXCL10, the interferon-?-inducible protein (IP-10) belongs to the CXC family of chemokines and acts as an immunoinflammatory mediator, inhibits angiogenesis and displays antitumor properties (2). Several studies indicated that the CXCR3 N-terminal domain plays a key role in determining binding affinity, receptor selectivity, and also in regulating allosteric signalling through the receptor (3). Moreover, tyrosine sulfation in chemokine receptors is emerging as a post-translational modification that contributes substantially to ligand binding (2). Tyr 27 and 29, two of the N- terminal tyrosine can be sulfonated. Finally, Kleist and co-workers hypothesize a "two-step" model, where receptor binding and activation can be dissociated (4). In the first step, the chemokine binds to the N-term domain of the receptor. In the second, residues on IP10 N-terminal bind to the binding cavity on the TM domain of the receptor and induce the allosteric communication to the cytosol. In this work, we analyse by a supervised molecular dynamics (SuMD) (5) simulation, the binding mechanism of IP10 on CXCR3. We explored the binding site in order to evaluate the residue- residue interactions between the chemockine and its receptor. Starting from the same equilibrated system, 3 runs of supervised MD were performed using the same conditions. A residue selection from the CXCR3 N-term domain and all the Extra-Cellular Loop (ECL) were considered for the binding interface, instead the entire chemokine was treated as ligand in order to perform the supervision on the distance between the receptor (CXCR3) and the ligand (CXCL10). They resulted with consistent similarity in both ligand-receptor contacts and interaction energies. Simulations didn't had identical durations (30.8 ns for simulation 1, 31.2 ns for simulation 2 and 36.6 ns for simulation 3), but all of them reached the ligand correctly positioned in the binding site in 15ns. Simulations 2 and 3 presented overlapping situations of decreasing energies related to minor R-L distances and of contacts, involving both residues from the CXCR3-N- term and the Extra-cellular ?-sheet, which has been described as an important player for the receptor activation. The next step will be the evaluation of the effect of tyrosine sulfation (Y27 and Y29) on binding mode and affinity.

Cystic fibrosis (CF) is mainly caused by the mutation F508del of the cystic fibrosis transmembrane conductance regulator (CFTR) that is thus retained in the endoplasmic reticulum and degraded. New drugs able to rescue F508del-CFTR trafficking and activity are eagerly awaited, a goal that requires the availability of computational and experimental models closely resembling the F508del-CFTR structure and environment in vivo. Here we describe the development of a biosensor based on F508del-CFTR in a lipid environment that proved to be endowed with a wider analytical potential in respect to the previous CFTR-based biosensors. Integrated with an appropriate computational model of the whole human F508del-CFTR in lipid environment and CFTR stability and functional assays, the new biosensor allowed the identification and characterization at the molecular level of the binding modes of some known F508del-CFTR-rescuing drugs and of a new aminoarylthiazole-Lumacaftor/Tezacaftor hybrid derivative endowed with promising F508del-CFTR-binding and rescuing activity.

OBJECTIVES: This study aims to evaluate the reliability and clinical utility of NS3 sequencing in hepatitis C virus (HCV) 1-infected patients who were candidates to start a PI-containing regimen. METHODS: NS3 protease sequencing was performed by in-house-developed HCV-1 subtype-specific protocols. Phylogenetic analysis was used to test sequencing reliability and concordance with previous genotype/subtype assignment by commercial genotyping assays. RESULTS: Five hundred and sixty-seven HCV plasma samples with quantifiable HCV-RNA from 326 HCV-infected patients were collected between 2011 and 2014. Overall, the success rate of NS3 sequencing was 88.9%. The success rate between the two subtype protocols (HCV-1a/HCV-1b) was similarly high for samples with HCV-RNA >3 log IU/mL (>92% success rate), while it was slightly lower for HCV-1a samples with HCV-RNA ≤3 log IU/mL compared with HCV-1b samples. Phylogenetic analysis confirmed the genotype/subtype given by commercial genotyping assays in 92.9% (303/326) of cases analysed. In the remaining 23 cases (7.1%), 1 was HCV-1g (previously defined as subtype 1a), 1 was HCV-4d (previously defined as genotype 1b) and 1 was HCV-1b (previously defined as genotype 2a/2c). In the other cases, NS3 sequencing precisely resolved the either previous undetermined/discordant subtype 1 or double genotype/subtype assignment by commercial genotyping assays. Resistance-associated variants (RAVs) to PI were detected in 31.0% of samples. This prevalence changed according to PI experience (17.1% in PI-naive patients versus 79.2% in boceprevir/telaprevir/simeprevir-failing patients). Among 96 patients with available virological outcome following boceprevir/telaprevir treatment, a trend of association between baseline NS3 RAVs and virological failure was observed (particularly for HCV-1a-infected patients: 3/21 failing patients versus 0/22 achieving sustained virological response; P = 0.11). CONCLUSIONS: HCV-NS3 sequencing provides reliable results and at the same time gives two clinically relevant pieces of information: a correct subtype/genotype assignment and the detection of variants that may interfere with the efficacy of PI.

Background & Aims: Despite the excellent efficacy of direct-acting antivirals (DAA) reported in clinical trials, virological failures can occur, often associated with the development of resistance-associated substitutions (RASs). This study aimed to characterize the presence of clinically relevant RASs to all classes in real-life DAA failures. Methods: Of the 200 virological failures that were analyzed in 197 DAA-treated patients, 89 with pegylated-interferon+ribavirin (PegIFN+RBV) and 111 without (HCV-1a/1b/1g/2/3/4=58/83/1/6/24/25; 56.8% treatment experienced; 65.5% cirrhotic) were observed. Sanger sequencing of NS3/NS5A/NS5B was performed by home-made protocols, at failure (N= 200) and whenever possible at baseline (N= 70). Results: The majority of the virological failures were relapsers (57.0%), 22.5% breakthroughs, 20.5% non-responders. RAS prevalence varied according to IFN/RBV use, DAA class, failure type and HCV genotype/subtype. It was 73.0% in IFN group vs 49.5% in IFN free, with the highest prevalence of NS5A-RASs (96.1%), compared to NS3-RASs (75.9% with IFN, 70.5% without) and NS5B-RASs (66.6% with IFN, 20.4% without, in sofosbuvir failures). In the IFN-free group, RASs were higher in breakthrough/non-responders than in relapsers (90.5% vs 40.0%, P= 2 DAA classes showed multiclass resistance, including 11/11 NS3+NS5A failures. Furthermore, 20.0% of patients had baseline-RASs, which were always confirmed at failure. Conclusions: In our failure setting, RAS prevalence was remarkably high in all genes, with a partial exception for NS5B, whose limited resistance is still higher than previously reported. This multiclass resistance advocates for HCV resistance testing at failure, in all three genes for the best second-line therapeutic tailoring.

Gene-environment interaction, through abnormal intestinal adsorption, has been proposed as possible mechanism for autism pathogenesis in those patients lacking of causative genetic variants. Haptoglobin (HP) is a haemoglobin binding and acute-phase plasma protein, encoded by two co-dominant alleles, HP-1 and HP-2, producing pre-HP-1 and pre-HP-2 proteins that mature in HP-1 and HP-2, respectively. Due to a 1.7Kb copy number variation in the HP gene, the HP-2 allele has two extra exons with respect to HP-1 (Fig. 1). Thus the HP protein is a dimer in homozygous subjects for HP-1 allele and is multimer in homozygous HP-2. Endogenous pre-HP-2 protein deregulates intestinal tight-junctions through EGFR and PAR2 activation, increases intestinal permeability and has been associated with autoimmune and inflammatory diseases as well as with psychiatric conditions (Fasano, 2011; Sturgeon and Fasano, 2016). Since the association between HP alleles and autism has just been investigated in a very small sample size of patients and controls (Rose et al., 2018), we genotyped, by PCR analysis, HP in a cohort of Italian patients with autism (n=406) and in controls (n=367). The aim was to evaluate the possible association of HP-2 and autism spectrum disorder (ASD). Contrary to what we expected, HP-1 allele distribution was different between patients and controls (36.3% and 29.4%, respectively) and significantly associated with autism (P=0.0041). Since a subgroup of patients and controls have already been genotyped by Illumina Human Omni-15-8 v.1.0 and Affy-6.0 chips, respectively, we are trying to impute HP alleles from flanking SNP haplotypes. HP alleles will therefore be predicted in publicly available large cohorts of patients with autism.

Gluten is a complex molecule made of gliadin and glutenins. During digestion, gliadin is reduced into small peptides of about 20 amino acids enriched in glutamine and prolines. In the intestinal epithelium the CXCR3 (CX-Chemokine Receptor Type 3, a G-Protein Coupled Receptor - GPCR) binds two of the gliadin peptides (pep 111-130 QQQQQQQQQQQQILQQILQQ and pep 151-170 QVLQQSTYQLLQELCCQHLW). Moreover, it is involved in celiac disease (CD) development inducing an increase of intestinal permeability, resulting in zonulin release, which is induced and activated through the MyD88-pathway. Two of the three splicing variants of CXCR3 are involved in CD: CXCR3-A and CXCR3-B. Furthermore, it is known that gliadin peptides are modified by tissue transglutaminase (tTG) converting specific glutamine (Q) residues into glutamic acid (E): pep 111-130E QQQQQQQQQQQQILQQILQE and pep 151-170E QVLQESTYQLLQELCCQHLW. Evaluate differences between deamidated and non-modified gliadin peptides bound to both CXCR3 a and b isoforms The selected conformations resulting from the Autodock CG simulations (Table 1) correspond to the complexes with the lowest binding energy if considering peptide 111-130; instead, for peptide 151-170, the chosen conformations result the second in terms of binding energy and the first in terms of cluster numerosity, since the lowest binding energy conformation displays a backward orientation of the peptide. HADDOCK run (Table 2) returned up to 14 clusters for each protein-peptide docking simulation, displaying different score ranges between deamidated and normal peptides. Instead, regarding energy values they display overlapping results except for Cxcr3b-pep111 complexes. The model with the lowest score and/or the lowest binding energy (VdW, Electrostatic and Desolvation) from each run was selected for MD simulations. MD trajectories analysis suggested either a change in CXCR3 TM-helices or in C- ter flexibility when comparing non-modified peptides to deamidated ones. Therefore, what we expect to understand is the effect of the gliadin deamidation, first on binding to CXCR3, secondly on the signal transmission to the cytoplasmic domain. Accordingly, an H-bond analysis was performed on MD trajectories. Results displayed an uninterrupted H-bond network from the binding cavity to the C-ter helix in the CXCR3a/111-130 (Fig. 2a) and CXCR3b/111-130E (Fig. 2b) complexes. But, the two complexes differ for the receptor residues involved in peptide binding (in blue in Fig.2), since the deamidated peptide binds charged residues, while the non-modified interacts with non-charged residues. Likewise, also the resulting Hbond networks involve different residues. The other CXCR3 complexes display several H-bond network stops before the C-ter. Haddock docking simulations did not allow a discrimination between non-modified and deamidated peptides while Autodock CG application displays a preference for CXCR3 a or b isoforms by deamidated peptides 111-130 and 151-170, respectively. H-bond analysis let us evaluate the effects of deamidation on the binding mode and on the affinity with CXCR3 isoforms, concluding that deamidation actively influences the binding to the receptor. Furthermore, we did not observe any substantial differences in signal transduction for peptide 151-170 (Fig. 2c), while peptide 111-130 displayed two distinct networks for each isoform (Fig. 2a and 2b). Given that CXCR3 is a GPCR, so the C-ter is involved in the activation of the G-coupled protein, these results suggest pep 111-130 as favoured ligand than pep 151-170, both for CXCR3 a and b.

Spinal cord injury (SCI) counts about 17,000 new cases each year in the United States. Since axons lose the competence to regenerate in adult mammals, SCI can lead to permanent neurological damages with dramatic personal, social and economic impacts. The long-term deficit of SCI results first from the type of insult and then from the secondary phase that includes many pathophysiological events. Among these, inflammation and epigenetic factors play a crucial role in the recovery of neuron connections. Variations in epigenetic and immune contribution, in turn, depend on age and health status as well as to microbiota profile of individuals at the time of, or consequent to SCI. Indeed, gut microorganisms, highly influence the immune system, but also produce bioactive substances such as folates, butyrate and acetate that participate to the epigenetic processes. Interestingly, variations in the profile of microbiota and bioactive substances have been described in patients with neurologic intestine because of SCI. Recently, different approaches, including stem cell therapy, use of biomaterial and laser therapy, have been proposed for neuronal regeneration after SCI but they are still far from resolutive interventions. As the complexity of pathophysiological processes of the secondary phase can deeply condition the success of regeneration, we provide a landscape of microbiota-epigenetic-immune modulation of neurological recovery predisposition or prevention. We discuss most of data about epigenetics (microRNAs, circulating microRNAs and chromatin remodelling) after SCI in animal models as well as microbiome profile of patients with SCI. We also propose a bioinformatics approach to compare "comics" data (gut microbiome, circulating microRNAs and inflammatory profiles) of patiens with SCI before and after laser therapy to evaluate and improve laser induced neuroregeneration. Acknowledgements: Flagship InterOmics (PB05).

Motivation NGS has revolutionized the genomic research. Technological advances have reduced the sequencing costs while have notably increased the sequence throughput. Typically, artifacts of different nature that arise during the sequencing process affect NGS data. As these artifacts may influence the downstream analyses, data quality control (QC) becomes mandatory. Different QC tools have been proposed. Without claiming to be exhaustive, let us cite some of the most popular tools, i.e., FASTQC [1], FASTX-Toolkit [2], NGS QC Toolkit [3], PRINSEQ [4]. FASTQC is a widely adopted tool for quality assessment. It provides a set of analyses to assess the raw data according to multiple aspects. It also provides a GUI to report all analyses highlighting problems in the data. FASTX-Toolkit is a collection of command line tools for FASTA/FASTQ file processing including some quality statistics. It also provides tools for quality filtering/trimming. NGS QC Toolkit is a standalone tool for QC that includes tools for sequence trimming and statistics calculation. It is also equipped with modules to generate statistics in graphical format. PRINSEQ is a web-based and standalone tool that can be used to generate summary statistics of sequence and quality data and to filter and trim sequences. It should be pointed out that the massive amounts of generated sequences make QC computationally intensive. Despite that, only some tools implement a multicore processing strategy to deal with that computational challenge. In our opinion, even though these tools implement very useful features, their implementation does not permit to efficiently analyze large amounts of NGS data. We deem that QC tasks can be efficiently parallelized on manycore architectures as GPUs. GPUs are devices equipped with hundreds of cores able to handle thousands of threads simultaneously, so that a very high level of parallelism can be reached. In this work, we present G-FastQC (GPU Fast Quality Control) a GPU-based tool for quality assessment, filtering, and trimming of NGS data. Methods G-FastQC has been devised to be massively parallelized on NVIDIA GPUs. It supports single- and paired-end libraries generated with Illumina platforms. G-FastQC implements a set of analyses to perform QC checks on raw data. These analyses allow to assess the data according to aspects related to the quality scores and content of the sequences. In particular, G-FastQC allows to calculate the: average quality values across all bases at each position; quality score distribution over all sequences; GC content across the whole length of each sequence; GC content across all bases; sequence content across all bases at each position.To help users to analyze the data quality, G-FastQC has been integrated with an interactive web-based interface built using Shiny[5] that allows to plot graphs of the performed analyses.G-FastQC also supports quality filtering and trimming. As for quality filtering, G-FastQC allows to filter sequences based on the amount of both low quality and N nucleotides as well as to filter sequences based on the GC content. It can also be used to mask the nucleotides with a quality score lower than a given threshold. As for trimming, G-FastQC implements an operator based on a sliding window approach. It analyzes the amount of low quality nucleotides in a sliding window. When the amount of these nucleotides is higher than a given threshold, G-FastQC trims all nucleotides from the start of the window to the 3'. Trimmed sequences with length lower than a given threshold can be automatically discarded. The same sliding window approach has been used to implement an operator aimed at masking these nucleotides rather than trimming them. It should be pointed out, that unlike the above-mentioned tools, G-FastQC has been designed to perform analysis of several datasets in a single run. Results Experiments carried out on a 12 cores Intel Xeon CPU E5-2667 2.90 GHz and an NVIDIA Tesla k20c show that G-FastQC outperforms notably the other tools in terms of computing time. For instance, in the task of filtering low-quality sequences, it has been 21.4x/28.3x faster than FASTX-Toolkit/NGS QC Toolkit analyzing a dataset consisting of 50M of 100 bp reads. Similar results have been obtained comparing the performance of G-FastQC with those of the other tools to generate the quality reports, and in the tasks of filtering/trimming the raw data. References 1. http://www.bioinformatics.abraham.ac.uk/projects/fastqc/ 2. http://hannonlab.cshl.edu/fastx_toolkit 3. Patel RK, Jain M (2012) NGS QC Toolkit: a toolkit for quality control of next generation sequencing data, PloS one, 7(2), e30619. 4. Schmieder R, Robert E (2011) Quality control and preprocessing of metagenomic datasets. Bioinformatics, 27(6), 863-864. 5. https://shiny.rstudio.com

Tissues and organs are generated during development and repaired over the lifetime by stem cells with extensive self-renewal and differentiation potential. The characteristics and the identity of the stem and differentiated cells are determined by specific transcription factors that act together with chromatin regulators to stabilize expression patterns that maintain the cell identities. Disruption of the chromatin state or changes in the expression levels of chromatin regulators is associated with cellular reprogramming, disease and oncogenesis. While a large number of chromatin regulators have been identified, the epigenomic processes by which stem cells differentiate remains largely unknown for some somatic stem cell types. Three dimensional organoid cultures generated from patient derived single cells with stem cells properties, allow for investigating regulators of the chromatin state and gene expression patterns in mammary gland normal or tumor development. In this work by using cell lineage tracing and cell tracking, I characterized homogenous populations of cells derived from single mammary cells with organoid formation capacity. My findings suggest that chromatin changes in the histone state of mammary cells both initiate and stabilize gene expression patterns before the establishment of lineage specifying transcription factors.