The detection of germline mutations in BRCA1 and BRCA2 is essential to the formulation of clinical management strategies, and in Brazil, there is limited access to these services, mainly due to the costs/availability of genetic testing. Aiming at the identification of recurrent mutations that could be included in a low-cost mutation panel, used as a first screening approach, we compiled the testing reports of 649 probands with pathogenic/likely pathogenic variants referred to 28 public and private health care centers distributed across 11 Brazilian States. Overall, 126 and 103 distinct mutations were identified in BRCA1 and BRCA2, respectively. Twenty-six novel variants were reported from both genes, and BRCA2 showed higher mutational heterogeneity. Some recurrent mutations were reported exclusively in certain geographic regions, suggesting a founder effect. Our findings confirm that there is significant molecular heterogeneity in these genes among Brazilian carriers, while also suggesting that this heterogeneity precludes the use of screening protocols that include recurrent mutation testing only. This is the first study to show that profiles of recurrent mutations may be unique to different Brazilian regions. These data should be explored in larger regional cohorts to determine if screening with a panel of recurrent mutations would be effective.

BackgroundCurrently, chromosomal microarrays (CMA) are recommended as first-tier test in the investigation of developmental disorders to examine copy number variations. The modern platforms also include probes for single nucleotide polymorphisms (SNPs) that detect homozygous regions in the genome, such as long contiguous stretches of homozygosity (LCSH) also named runs of homozygosity (ROH). LCHS are chromosomal segments resulting from complete or segmental chromosomal homozygosity, which may be indicative of uniparental disomy (UPD), consanguinity, as well as replicative DNA repair events, however also are common findings in normal populations. Knowing common LCSH of a population, which probably represent ancestral haplotypes of low-recombination regions in the genome, facilitates the interpretation of LCSH found in patients, allowing to prioritize those with possible clinical significance. However, population records of ancestral haplotype derived LCSH by SNP arrays are still scarce, particularly for countries such as Brazil where even for the clinic, microarrays that include SNPs are difficult to request due to their high cost.MethodsIn this study, we evaluate the frequencies and implications of LCSH detected by Affymetrix CytoScan® HD or 750 K platforms in 430 patients with neurodevelopmental disorders in southern Brazil. LCSH were analyzed in the context of pathogenic significance and also explored to identify ancestral haplotype derived LCSH. The criteria for considering a region as LCSH was homozygosis ≥3 Mbp on an autosome.ResultsIn 95% of the patients, at least one LCSH was detected, a total of 1478 LCSH in 407 patients. In 2.6%, the findings were suggestive of UPD. For about 8.5% LCSH suggest offspring from first to fifth grade, more likely to have a clinical impact. Considering recurrent LCSH found at a frequency of 5% or more, we outline 11 regions as potentially representing ancestral haplotypes in our population. The region most involved with homozygosity was 16p11.2p11.1 (49%), followed by 1q21.2q21.3 (21%), 11p11.2p11.12 (19%), 3p21.31p21.2 (16%), 15q15 1q33p32.3 (12%), 2q11.1q12.1 (9%), 1p33p32.3 (6%), 20q11.21q11.23 (6%), 10q22.1q23.31 (5%), 6p22.2p22 (5%), and 7q11.22q11.23 (5%).ConclusionsIn this work, we show the importance and usefulness of interpreting LCSH in the results of CMA wich incorporate SNPs.

Chromosomal microarray (CMA) is now recommended as first tier for the evaluation in individuals with unexplained neurodevelopmental disorders (ND). However, in developing countries such as Brazil, classical cytogenetic tests are still the most used in clinical practice, as reflected by the scarcity of publications of microarray investigation in larger cohorts. This is a retrospective study which analyses the reading files of CMA and available clinical data from 420 patients from the south of Brazil, mostly children, with neurodevelopmental disorders requested by medical geneticists and neurologists for diagnostic purpose. Previous karyotyping was reported for 138 and includes 17 with abnormal results. The platforms used for CMA were CYTOSCAN 750K (75%) and CYTOSCAN HD (25%). The sex ratio of the patients was 1.625 males :1 female and the mean age was 9.5 years. A total of 96 pathogenic copy number variations (CNVs), 58 deletions and 38 duplications, were found in 18% of the patients and in all chromosomes, except chromosome 11. For 12% of the patients only variants of uncertain clinical significance were found. No clinically relevant CNV was found in 70%. The main referrals for chromosomal microarrays (CMA) were developmental delay (DD), intellectual disability (ID), facial dysmorphism and autism spectrum disorder (ASD). DD/ID were present in 80%, facial dysmorphism in 52% and ASD in 32%. Some phenotypes in this population could be predictive of a higher probability to carry a pathogenic CNV, as follows: dysmorphic facial features (p-value = < 0.0001, OR = 0.32), obesity (p-value = 0.006, OR = 0.20), short stature (p-value = 0.032, OR = 0.44), genitourinary anomalies (p-value = 0.032, OR = 0.63) and ASD (p-value = 0.039, OR = 1.94). The diagnostic rate for CMA in this study was 18%. We present the largest report of CMA data in a cohort with ND in Brazil. We characterize the rare CNVs found together with the main phenotypes presented by each patient, list phenotypes which could predict a higher diagnostic probability by CMA in patients with a neurodevelopmental disorder and show how CMA and classical karyotyping results are complementary.

BackgroundBiotinidase deficiency (BD) is an inborn error of metabolism in which some genetic variants correlate with the level of enzyme activity. Biotinidase activity, however, may be artifactually low due to enzyme lability, premature birth, and jaundice; this hinders both phenotypic classification and the decision to implement therapy. This study sought to characterize the clinical and genetic profile of a sample of Brazilian patients exhibiting reduced biotinidase activity.MethodsThis observational, multicenter study used a convenience sampling strategy, with sequencing of exons 2, 3, and 4 of the BTD gene.ResultsThe sample comprised 38 individuals with biochemical phenotypes defined a priori on the basis of biotinidase activity in serum/plasma (2 with profound deficiency, 9 with partial deficiency, 15 heterozygous, 1 borderline between partial deficiency and heterozygosity, 2 borderline between heterozygous and normal) or dried blood spot sample (n = 9, all with unspecified deficiency). Most patients were from Southern Brazil (n = 29/38) and were identified by neonatal screening (n = 33/38). Parental consanguinity was reported in two cases. The most commonly found genetic variants were c.1330G > C (p.D444H), c.755A > G (p.D252G), and c.[511G > A;1330G > C] (p.[A171T;D444H]), with allele frequencies of 50%, 9.4%, and 5.4% respectively. Three novel pathogenic variants were identified (c.119 T > C or p.L40P, c.479G > A or p.C160Y, and c.664G > A or p.D222N). Twenty-nine patients had two pathogenic variants detected (with cis/trans status ascertained in 26/29), six had only one variant, and three had no pathogenic variants detected. Genotyping confirmed the original phenotypic classification based on enzyme activity in 16/26 cases. Three polymorphic variants were identified in control individuals, of which two were nonpathogenic (c.1171C > T or p.P391S and c.1413 T > C or p.C471C, with a frequency of 1.5% and 5.5% respectively) and one pathogenic (c.1330G > C, frequency 4%).ConclusionsOur findings suggest that partial BD is the most common form of BD in Brazil, and expand current knowledge on the allelic heterogeneity of this condition.

BackgroundClassical homocystinuria (HCU) is a monogenic disease caused by the deficient activity of cystathionine β‐synthase (CβS). The objective of this study was to identify the CBS mutations in Brazilian patients with HCU.Methods gDNA samples were obtained for 35 patients (30 families) with biochemically confirmed diagnosis of HCU. All exons and exon‐intron boundaries of CBS gene were sequenced. Gene expression analysis by qRT‐PCR was performed in six patients. Novel missense point mutations were expressed in E. coli by site‐directed mutagenesis.ResultsParental consanguinity was reported in 16 families, and pyridoxine responsiveness in five (15%) patients. Among individuals from the same family, all presented the same phenotype. Both pathogenic mutations were identified in 29/30 patients. Twenty‐one different mutations were detected in nine exons and three introns; being six common mutations. Most prevalent were p.Ile278Thr (18.2%), p.Trp323Ter (11.3%), p.Thr191Met (11.3%), and c.828+1G>A (11.3%). Eight novel mutations were found [c.2T>C, c.209+1delG, c.284T>C, c.329A>T, c.444delG, c.864_868delGAG c.989_991delAGG, and c.1223+5G>T]. Enzyme activity in E. coli‐expressed mutations was 1.5% for c.329A>T and 17.5% for c.284T>C. qRT‐PCR analysis revealed reduced gene expression in all evaluated genotypes: [c.209+1delG; c.572C>T]; [c.2T>C; c.828+1G>A]; [c.828+1G>A; c.1126G>A]; [c.833T>C; c.989_991delAGG]; [c.1058C>T; c.146C>T]; and [c.444delG; c.444delG]. The expected phenotype according to the genotype (pyridoxine responsiveness) matched in all cases.ConclusionsMost patients studied were pyridoxine nonresponsive and presented early manifestations, suggesting severe phenotypes. Many private mutations were observed, but the four most prevalent mutations together accounted for over 50% of mutated alleles. A good genotype–phenotype relationship was observed within families and for the four most common mutations.