Targeted Next-Generation Resequencing of F5 Gene Identifies Novel Multiple Variants Pattern in Severe Hereditary Factor V Deficiency

The present study investigated the genetic defects underlying severe Factor V deficiency in a 26-year-old Columbian (South America) female and her immediate family (both parents and newborn child) by next generation sequencing (NGS) of the entire F5 gene locus. Five mutations in the coding sequence of F5, including three missense single-nucleotide variants (R2102H, R513K, D107H) and two synonymous variants (A135A , S184S), were identified and confirmed by the Sanger sequencing in the investigated proband (homozygote for all detected mutations), her parents, and her newborn child (all heterozygotic carriers for identified mutations). Each of the three missense variants was previously associated with separate phenotypes, including Factor V deficiency (R2102H), thrombosis (R513K) and frequent miscarriages (D107H). In addition, at least 75 additional single-nucleotide variants (including six novels) were identified in untranslated region of F5.


Introduction
Coagulation Factor V is a large 330-kD glycoprotein which consists of 2224 amino acid residues including a 28-residue leader peptide, which is structurally and functionally homologous to coagulation Factor VIII [1,2]. The human Factor V gene (official name 5) maps to chromosome 1q23 and contains 25 exons (8). Factor V deficiency is a rare autosomal recessive disorder (incidence < 1 in 1 million), characterized by low levels of antigen and activity [3]. At present, more than one hundred deficiency-causing mutations in the 5 locus have been described, and although most of them are private, a few are common, being found in several individuals of both European, Middle-Eastern and Asiatic descents. In the present study, we used for the first time DNA next-generation sequence (NGS) analysis to detect mutation pattern in the entire 5 locus of a 26-year-old Hispanic parturient with severe Factor V deficiency, as well as in her asymptomatic parents and newborn baby. The relationship between combinations of mutations and clinical phenotypes was evaluated.

Case Presentation
The study protocol was approved by IRB at PSU Hershey Medical College. It was performed in adherence to the tenets of the declaration of Helsinki. Written informed consent was obtained from all participants. The investigated patient was 26-year-old Hispanic (born in Columbia, South America) parturient (G4P1) with several bleeding episodes before and during present pregnancy, multiple fresh frozen plasma (FFP) transfusions, and a history of three miscarriages in the past. The patient was previously diagnosed clinically to have severe Factor V deficiency on the basis of several previous bleeding episodes and laboratory studies demonstrating coagulopathy with moderate to severe decrease in Factor V activity. The remaining past medical history was unremarkable. The family history revealed that both biological parents had no history of bleeding or other coagulation symptoms and had reported normal Factor V activity. In addition, she has two siblings, apparently without any clinical signs of coagulation disorders. In the course of the current pregnancy the patient delivered 2 Case Reports in Genetics the healthy male newborn, who has not displayed, at the time of this analysis, any signs of coagulation abnormalities, besides decreased (36%) level of Factor V. For the purpose of this investigation, we collected the samples of blood (proband and newborn son) and saliva (both biological parents and proband's husband) for DNA analysis (Figure 1).

Mutation Analysis
Genomic DNA (gDNA) was extracted from venous EDTAwhole blood sample (proband) or cord blood (newborn child) employing membrane ultrafiltration method (FujiFilm Life Sciences distributed by Autogene, Holliston, MA, USA), according to the manufacturer recommendations. The saliva samples were collected from both parents and proband/s husband into Oragene container (DNA Genotek, Canada) and extracted according to the manufacturer recommendations. Subsequently two gDNA samples from the proband were submitted to Otogenetics Corporation (Norcross, GA, USA) for target capture and sequencing. Briefly, gDNA was subjected to agarose gel and optical density ratio tests to confirm the purity and concentration prior to Covaris (Covaris, Inc., Woburn, MA, USA) fragmentation. Fragmented gDNAs were tested for size distribution and concentration using an Agilent Bioanalyzer 2100 (Agilent Technologies, Santa Clara, CA, USA) and Nanodrop (Thermo Fisher Scientific, Wilmington DE, USA). Illumina libraries were made from qualified fragmented gDNA using NEBNext reagents (New England Biolabs, Ipswich, MA, USA) and the resulting libraries were subjected to exome enrichment using custom probes targeting 75 kb target on chromosome 1 (169, 481, 192-169, 555, 469 by GRCh 37, Hg19). The resultant libraries were tested for enrichment by qPCR and for size distribution and concentration by an Agilent Bioanalyzer 2100. The samples were then sequenced on an Illumina HiSeq2000 (Illumina, San Diego, CA, USA) which generated pairedend reads of 90 or 100 nucleotides. Data was analyzed for data quality, exome coverage, and exome-wide SNP/InDel using the platform provided by DNAnexus (DNAnexus, Inc, Mountain View, CA, USA). The detected polymorphisms in the coding sequence of the 5 locus were subsequently verified using classical Sanger sequencing using exon primers described by van Wijk et al. [4]. This analysis was performed by direct DNA sequencing using ABI Prism 3100 genetic analyzer (Applied Biosystems, Foster City, CA, USA).
By NGS, we generated about 549 million bases of sequence as pair-end 90 or 100 nucleotide reads, 86% of which was able to align to human reference sequence. A total of 23% of these sequences mapped to the targeted region corresponding to 75 kb sequence of the 5 locus (NM 0001304.4), with 540-fold mean coverage (Table 1). At this depth of coverage, more than 95% of the target bases were covered to pass quality control filtering based on the PHRED score threshold of calling variants (PHRED > 30). Eighty high-confidence variants were annotated in the target region (3 nonsynonymous single-nucleotide variants and 2 synonymous singlenucleotide variants in the coding region of the target (Table 2), as well as 75 single-nucleotide variants in the Open symbols indicate no carrier of investigated mutations. Question marks indicate that genotyping has not been performed.
noncoding sequence of the target ( Table 3). The missense variants were located in the exon 3 (D107H), 10 (R513K), and 23 (R2102H) of the 5 locus (Figure 1), and the investigated patient was a homozygous carrier for all these variants. These variants were additionally confirmed by the Sanger sequencing and showed perfect match in two duplicate samples. The additional verification of the DNA samples from the proband's newborn son and our patient's biological parents performed using the Sanger sequencing revealed that all of them were heterozygous carriers for all 3 missense variants ( Figure 2). No presence of these variants was shown in the husband of the patient (i.e., biological father of the newborn child).

Discussion
The current study presents several novel findings: (i) it employs NGS approach for molecular diagnosis of FV deficiency, including sequencing of both translated and untranslated regions of 5 locus; (ii) it diagnoses the inherited FV deficiency in the S. American Caucasian family of Hispanic ethnicity; and (iii) it depicts detailed genetic evidence for multiple, coinherited mutations in the 5 locus which may be responsible for opposite phenotype characteristics (i.e., increasing and decreasing thrombosis process).
The results of the current study indicate that the investigated proband was a homozygotic carrier of three separate missense mutations in the coding sequence of 5. Surprisingly, however, only one of the 3 detected missense mutations (R2102H) was described previously (   the Factor V [5]. This substitution results in the replacement of an arginine (R) by a histidine (H) in amino acid position 2102 located in the C2-domain of Factor V and several lines of evidence reported previously support the notion that this sequence variant is causative for Factor V deficiency phenotype. Interestingly the remaining two detected missense mutations were reported to be associated mostly with either thrombosis [6][7][8] or increased risk of preterm delivery (D107H) [9]. It is noteworthy that two missense point mutations which were detected in the current proband were previously described in different populations (R2102H in Tunisian population and R513K in Thai, Chinese and Sub-Saharan populations). Although Factor V deficiency and its causative mutations were reported previously from European Caucasians, it is, to our knowledge, the first detailed description of causative mutation in 5 locus in a family from South America and/or Hispanic ethnicity.
In addition, we established that the investigated proband was a homozygous carrier for two synonymous singlenucleotide variants: A135A (rs6029) and S184S (rs6022) previously reported in the online SNP Medline database. There is no information about the potential phenotypic significance of these mutations. In addition to the previously described single-point mutations in the coding region of 5 gene, the analysis of the NGS data from the proband established a presence of additional 75 polymorphisms in the untranslated sequence of the 5 locus (3 in 5 -UTR part, 71 in the intronic part, and 1 in the 3 -UTR part of the gene) ( Table 3). Six of these variants represent newly discovered variants, not presented previously in the SNP Medline database. The types of detected variants include single-nucleotide variants (4 insertions and 4 deletions), as well as 3 additional short indels. The phenotypic significance of these polymorphisms for the Factor V function remains unknown. The most recently (Oct 2012) accessed Human Genome Mutation database (www.hgmd.org) lists 145 missense mutations in the 5 locus associated with altered function of Factor V. From this list, 94 mutations represent single-point mutations, from which approximately 80 have been strongly associated with Factor V deficiency. In this respect, the present study does not add new mutations to this list but confirms the fact of previously described coinheritance of several separate mutations in the 5 locus [10,11], and more importantly provides an example of co-inheritance of 5 mutations with presumably opposite phenotypic coagulation characteristics. Similar situation (i.e., coinheritance of polymorphic variants from which one is associated with decreased Factor V activity and other with increased thrombosis) was described previously for much more frequent prothrombotic Leiden mutations, or more recently, promoter −426 G/A polymorphism [12,13]. Our finding confirms that other prothrombotic mutations in the 5 gene locus may be independently inherited in one patient.
The present study exemplifies the use of NGS approach for detailed diagnosis of the specific clinical pathology and identification of causative mutations for rare genetic disorder. This approach has been recently advocated for both diagnosis and therapy [14]. The obtained genetic data for this cases were successfully used clinically for the peripartum anesthetic management of the previously described patient Table 3: Summary of observed variants in noncoding sequences of F5 gene in investigated proband.