Skewed X-chromosome Inactivation in Women with Idiopathic Intellectual Disability is Indicative of Pathogenic Variants

Intellectual disability (ID) is an early onset impairment in cognitive functioning and adaptive behavior, affecting approximately 1% of the population worldwide. Extreme skewing of X-chromosome inactivation (XCI) can be associated with ID phenotypes caused by pathogenic variants in the X chromosome. We analyzed the XCI pattern in blood samples of 194 women with idiopathic ID, using the androgen receptor gene (AR) methylation assay. Among the 136 patients who were informative, 11 (8%) presented with extreme or total XCI skewing (≥ 90%), which was significantly higher than expected by chance. Whole-exome data obtained from these 11 patients revealed the presence of dominant pathogenic variants in eight of them, all sporadic cases, resulting in a molecular diagnostic rate of 73% (8/11 patients). All variants were mapped to ID-related genes with dominant phenotypes: four variants in the X-linked genes DDX3X (an XCI escape gene; two cases), WDR45, and PDHA1, and four variants in the autosomal genes KCNB1, CTNNB1, YY1, and ANKRD11. Three of the autosomal genes had no obvious correlation with the observed XCI skewing. However, YY1 is a known transcriptional repressor that acts in the binding of the XIST long noncoding RNA on the inactive X chromosome, providing a mechanistic link between the pathogenic variant and the detected skewed XCI in the carrier. These data confirm that extreme XCI skewing in females with ID is highly indicative of causative X-linked pathogenic variants, and point to the possibility of identifying causative variants in autosomal genes with a XCI role.

Abstract B41: Genomic landscape of somatic mutations in osteosarcomas

Pediatric cancers are among the leading causes of childhood mortality, with osteosarcomas being the most prevalent bone tumors, characterized by an aggressive clinical course. However, knowledge about the molecular basis of osteosarcomas is still limited, hampering advances in diagnosis and treatment. In this context, we characterized the genomic landscape of somatic mutations in 28 osteosarcomas (24 pediatric patients and 4 adults) aiming to delineate the panel of prevalent mutations in Brazilian patients. Genomic libraries of these osteosarcoma samples were constructed using the TruSight One panel (Illumina), covering 4,813 clinically relevant genes, including known cancer genes. Generated data at high coverage (median >100x depth coverage) were annotated based on public databases containing population variant frequencies (including the Brazilian ABRAOM), as well as clinical information. For further analysis, we filtered only coding nonsynonymous variants absent from both population databases and an additional pool of 20 nonrelated germline samples. In total, 728 variants were identified mapping in 606 genes, with an average rate of 41 mutations per tumor. The panel of detected variants was composed by 93 loss-of-function mutations (38 frameshift, 25 splicing, and 30 premature stop codon) and 635 missense (43 inframe insertions/deletions and 592 single-nucleotide variations), some of them already reported in cancer databases (27 in COSMIC and 39 in ICGC). From 606 genes, 25 were previously associated with osteosarcomas, and they are related to apoptosis, cell growth and differentiation, DNA repair, transcriptional regulation, and tumor-suppression mechanisms. The most frequent mutations were detected in RB1 and TP53; other recurrently mutated genes were mostly linked to biologic pathways potentially related to the disease onset, such as hormonal response (AR), transcriptional regulation and muscle differentiation (FRG1), DNA repair (HERC2), cell signaling (KIR2DL4), and cell proliferation/differentiation (PTPRQ). Additionally, the tumors presented a highly complex pattern of copy number alterations, some of them resembling chromothripsis. This is the first assessment of a Brazilian cohort of osteosarcomas, and functional studies will be further applied to better understand the role of the disclosed mutations for tumor biology.

Citation Format: Sara F. Pires, Silvia S. Costa, Daniel O. Vidal, André van Helvoort Lengert, Érica Boldrini, Sandra R.M. da Silva, Carla Rosenberg, Luiz F. Lopes, Mariana Maschietto, Ana Krepischi. Genomic landscape of somatic mutations in osteosarcomas [abstract]. In: Proceedings of the AACR Special Conference on the Advances in Pediatric Cancer Research; 2019 Sep 17-20; Montreal, QC, Canada. Philadelphia (PA): AACR; Cancer Res 2020;80(14 Suppl):Abstract nr B41.

Cardiac protein expression patterns are associated with distinct inborn exercise capacity in non-selectively bred rats

In the present study, we successfully demonstrated for the first time the existence of cardiac proteomic differences between non-selectively bred rats with distinct intrinsic exercise capacities. A proteomic approach based on two-dimensional gel electrophoresis coupled to mass spectrometry was used to study the left ventricle (LV) tissue proteome of rats with distinct intrinsic exercise capacity. Low running performance (LRP) and high running performance (HRP) rats were categorized by a treadmill exercise test, according to distance run to exhaustion. The running capacity of HRPs was 3.5-fold greater than LRPs. Protein profiling revealed 29 differences between HRP and LRP rats (15 proteins were identified). We detected alterations in components involved in metabolism, antioxidant and stress response, microfibrillar and cytoskeletal proteins. Contractile proteins were upregulated in the LVs of HRP rats (α-myosin heavy chain-6, myosin light chain-1 and creatine kinase), whereas the LVs of LRP rats exhibited upregulation in proteins associated with stress response (aldehyde dehydrogenase 2, α-crystallin B chain and HSPβ-2). In addition, the cytoskeletal proteins desmin and α-actin were upregulated in LRPs. Taken together, our results suggest that the increased contractile protein levels in HRP rats partly accounted for their improved exercise capacity, and that proteins considered risk factors to the development of cardiovascular disease were expressed in higher amounts in LRP animals.

A proteomic road to acquire an accurate serological diagnosis for human tegumentary leishmaniasis

Diagnostic tools are important for clinical management and epidemiological evaluation of Tegumentary (TL) and Visceral (VL) Leishmaniasis. Serology is not frequently used for the diagnosis of the TL form because low antibody titers and cross-reaction with VL. Therefore, it is crucial to identify specific and immunogenic antigens from species associated with the TL form. Here we employed a proteomic approach coupled to an in silico analysis and identified the most abundant and immunogenic proteins from Leishmania amazonensis, Leishmania braziliensis and Leishmania infantum. Of 16 species specific proteins, nine were from the species causative of the TL form (L. amazonensis and L. braziliensis). In silico analysis revealed 18 B-cell epitopes with 0% similarity to Trypanosoma cruzi orthologs and, therefore, less likely to crossreact with sera of patients with Chagas disease. Two proteins reacted exclusively with serum from TL patients and presented several B-cell epitopes without similarity to T. cruzi orthologs: the hypothetical protein GI 134063939 and the metallo-peptidase Clan MA(E)-Family M3. The immunoassay revealed nine peptides with strong reactivity to sera from TL patients. These proteins and peptides may be good candidates to improve the specificity and sensibility of serological tests aiming to diagnose the TL of this neglected human disease.

BIOLOGICAL SIGNIFICANCE

As no gold-standard test for tegumentary leishmaniasis (TL) exists, a combination of different diagnostic techniques is often necessary to obtain precise results. Thus, the identification of species-specific, highly immunogenic and abundant proteins that stimulate the humoral immune response in the host should help in the development of serological tests for human TL. Herein we searched for these potential antigens in Leishmania species related to American Leishmaniasis (L. amazonensis, L. braziliensis and L. infantum). To this end, we employed an immunoproteomic approach using proteins from these Leishmania species and sera from TL and Visceral Leishmaniasis (VL) patients. Our study unveils specific proteins and peptides that may represent antigens that will help the efforts to improve the accuracy of serological tests to diagnose the TL form.

Immunoproteomic and bioinformatic approaches to identify secreted Leishmania amazonensis, L. braziliensis, and L. infantum proteins with specific reactivity using canine serum

Leishmania spp have a wide range of hosts, and each host can harbor several Leishmania species. Dogs, for example, are frequently infected by Leishmania infantum, where they constitute its main reservoir, but they also serve as hosts for L. braziliensis and L. amazonensis. Serological tests for antibody detection are valuable tools for diagnosis of L. infantum infection due to the high levels of antibodies induced, unlike what is observed in L. amazonensis and L. braziliensis infections. Likewise, serology-based antigen-detection can be useful as an approach to diagnose any Leishmania species infection using different corporal fluid samples. Immunogenic and secreted proteins constitute powerful targets for diagnostic methods in antigen detection. As such, we performed immunoproteomic (2-DE, western blot and mass spectrometry) and bioinformatic screening to search for reactive and secreted proteins from L. amazonensis, L. braziliensis, and L. infantum. Twenty-eight non-redundant proteins were identified, among which, six were reactive only in L. amazonensis extracts, 10 in L. braziliensis extracts, and seven in L. infantum extracts. After bioinformatic analysis, seven proteins were predicted to be secreted, two of which were reactive only in L. amazonensis extracts (52kDa PDI and the glucose-regulated protein 78), one in L. braziliensis extracts (pyruvate dehydrogenase E1 beta subunit) and three in L. infantum extracts (two conserved hypothetical proteins and elongation factor 1-beta). We propose that proteins can be suitable targets for diagnostic methods based on antigen detection.

Cell culture and animal infection with distinct Trypanosoma cruzi strains expressing red and green fluorescent proteins

Different strains of Trypanosoma cruzi were transfected with an expression vector that allows the integration of green fluorescent protein (GFP) and red fluorescent protein (RFP) genes into the beta-tubulin locus by homologous recombination. The sites of integration of the GFP and RFP markers were determined by pulse-field gel electrophoresis and Southern blot analyses. Cloned cell lines selected from transfected epimastigote populations maintained high levels of fluorescent protein expression even after 6 months of in vitro culture of epimastigotes in the absence of drug selection. Fluorescent trypomastigotes and amastigotes were observed within Vero cells in culture as well as in hearts and diaphragms of infected mice. The infectivity of the GFP- and RFP-expressing parasites in tissue culture cells was comparable to wild type populations. Furthermore, GFP- and RFP-expressing parasites were able to produce similar levels of parasitemia in mice compared with wild type parasites. Cell cultures infected simultaneously with two cloned cell lines from the same parasite strain, each one expressing a distinct fluorescent marker, showed that at least two different parasites are able to infect the same cell. Double-infected cells were also detected when GFP- and RFP-expressing parasites were derived from strains belonging to two distinct T. cruzi lineages. These results show the usefulness of parasites expressing GFP and RFP for the study of various aspects of T. cruzi infection including the mechanisms of cell invasion, genetic exchange among parasites and the differential tissue distribution in animal models of Chagas disease.