Genetic analysis using long-read sequencing to overcome the difficulties in VWF gene

Abstract

Background: Genetic defects in von Willebrand factor (VWF) can lead to von Willebrand disease (VWD). Identifying causative or modifier variants of VWF is crucial for the diagnosis, classification, and clinical management of VWF disorders. However, owing to the length (178 kb) and complexity of VWF and the presence of the pseudogene VWFP1, Sanger sequencing or short-read next-generation sequencing is often challenging.

Objectives: This study aimed to establish a long-read sequencing method using Oxford nanopore technology (ONT) to overcome difficulties associated with VWF gene analysis.

Methods: Genetic analyses were established using genomic DNA from a healthy donor and validated using 3 VWF disorder patient samples. Long-range (∼15 kb) polymerase chain reaction was optimized to obtain 21 amplicons covering the entire VWF gene, avoiding unwanted amplification due to repetitive sequences and VWFP1. ONT nanopore sequencing data were analyzed using software programs, including Clair3, Longshot, and Sniffles. The identified candidate variants were verified by several approaches such as Sanger sequencing and haplotyping.

Results: The entire VWF gene was successfully read using ONT nanopore sequencing, with >200 variants called in each patient sample. A rare missense variant, p.(Gln2442His) and a rare 2599 bp deletion were identified in patients 2 and 3, respectively. However, the deletion was confirmed as long-range polymerase chain reaction artifacts, which warrant attention when using this method.

Conclusion: This study presents an optimal solution using ONT nanopore sequencing to identify variants in VWF, which may improve the diagnosis of VWF disorders.

Keywords: nanopore sequencing; polymerase chain reaction; von Willebrand disease; von Willebrand factor.

Figure 1

Long-read sequencing design and results. (A) Schematic representation of the advantages of long-read sequencing for VWF genetic analysis and the objective of this study. The red lines above the VWF gene indicate polymerase chain reaction (PCR) amplicons generated from DNA samples. (B) Design of long-range PCR covering the entire VWF gene. PCR2, PCR3, PCR15, and PCR16 (yellow) represent regions where amplification was unsuccessful with the initial primer pair. PCR11-PCR14 (red) target the pseudogene homology region, while PCR21 (green) spans a short overlapping region between PCR5 and PCR6. (C) Representative image of 1.2% agarose gel electrophoresis of PCR amplicons generated from a healthy donor DNA sample. M, DNA ladders of 1.5, 2, 3, 4, 5, 6, 8, and 10 kb. (D) ONT nanopore sequencing reads from healthy donor samples visualized using Integrative Genomics Viewer software. The numbers indicate the positions of exons 1, 10, 20, 23, 28, 34, 40, and 52. ONT, Oxford nanopore technology.

Figure 2

Workflow of ONT nanopore sequencing data analysis in this study. Each step of the workflow is annotated with the corresponding algorithms and databases used, shown in red. BAM, binary alignment map, a comprehensive raw data of genome sequencing; INDEL, insertion–deletion variant; FASTQ, format containing nucleotide sequences and the corresponding quality scores; ONT, Oxford nanopore technology; SNV, single-nucleotide variant; SV, structural variant; VCF, variant call format.

Figure 3

Validation of candidate variants by Sanger sequencing. (A) Confirmation of the heterozygous missense SNV p.(Gln2442His) identified in patient 2. The 412-bp polymerase chain reaction (PCR) products using genomic DNA (gDNA) from patient 2 (lane 1) and using a long-range PCR amplicon from the patient’s gDNA (lane 2) were directly sequenced using the Sanger method. (B) Confirmation of the heterozygous 2599 bp deletion g.6087520_6090118del identified in patient 3. PCR amplification of the region containing the deletion produced the 2969-bp and 370-bp bands using patient 3 gDNA (lane 1), but only the 2969-bp band using healthy donor gDNA (lane 2). Direct Sanger sequencing of the patient’s PCR products indicated the heterogeneous sequences from the deleted site. (C) Identification of heterozygous SNVs within the g.6087520_6090118del region. Three PCR products (441, 556, and 631 bp) using a long-range PCR amplicon from gDNA of patient 3 carries 5 heterozygous SNVs. m, DNA ladders every 100 bp; M, DNA ladders of 0.5, 1, 1.5, 2, 3, 4, 5, 6, 8, and 10 kb. SNV, single-nucleotide variant.

Figure 4

Haplotyping analysis of the polymerase chain reaction (PCR) amplicon carrying g.6087520_6090118del. (A) Schematic representation of the expected haplotyping result of a PCR amplicon with a true 2599-bp deletion under normal circumstances. (B) Fifty reads generated from PCR21 amplicon of patient 3, covering the 2599-bp deletion, were extracted from the raw ONT data. All reads were sorted into 3 types according to the heterozygous SNVs identified in this region and their length. The haplotype of 26 deleted reads (9342 bp) matches that of 11 normal-length reads (11,941 bp). Heterozygous SNVs identified in patient 3 but not in the healthy donor or patients 1 and 2 are shown in light blue. SNV, single-nucleotide variant.

Figure 5

Polymerase chain reaction (PCR) artifacts due to slipped-strand structure. (A) Specific nucleotide sequences with 2 copies of repeats and 1 copy of a nonrepetitive spacer form a slipped-strand structure. (B) Eight SNVs in the g.6087416-6087659 region on 1 allele of patient 3 caused a slipped-strand structure, which tended to produce more deletion amplicons in long-range PCR (∼70% of total amplicons). In contrast, alleles with reference sequences in the region are less likely to have a slipped-strand structure due to incomplete repeat sequences, resulting in few deletion reads (<5% of total amplicons). (C) The first PCR was performed with gDNA from patient 3 as shown in Figure 3B. The second PCR was performed with the same primers using the normal-length band (2969 bp) excised from the agarose gel as a template. The second PCR also produced a 370-bp band, indicating that this phenomenon was a PCR artifact. gDNA, genomic DNA; SNV, single-nucleotide variant.