Whole-exome sequencing applications in prenatal diagnosis of fetal bowel dilatation

2.1 Participants and clinical data

This study reviewed 28 cases which were diagnosed with fetal bowel dilatation by prenatal ultrasonographic examination at Tongji Hospital affiliated to Tongji Medical College of Huazhong University of Science and Technology. Criterion for fetal bowel dilatation was fluid-filled intestinal loops at least 15 mm in length or 7 mm in diameter [6,7].

All participants received routine obstetric examinations, including ultrasound nuchal translucency screening at 11⁺⁰–13⁺⁶ weeks of gestation, non-invasive prenatal testing (NIPT) at 12–16 weeks of gestation, and ultrasonographic examination at 18–24 weeks of gestation. All cases were tested by karyotyping and CNV-seq. Cases confirmed with aneuploidy or pCNVs were not tested in the following WES analysis.

1. Informed consent: Informed consent has been obtained from all individuals included in this study.

2. Ethical approval: The research related to human use has been complied with all the relevant national regulations, institutional policies and in accordance with the tenets of the Helsinki Declaration, and has been approved by the authors’ institutional review board or equivalent committee (IRB ID: TJ-IRB20220104).

2.2 Genetic testing and laboratory evaluation

Amniocentesis was performed and 20 mL of amniotic fluid was sampled for karyotyping. The amniotic fluid was placed in an amniotic fluid culture bottle and grown in a carbon dioxide incubator. After 8 days of incubation, adherent cells were examined and treated with colchicine for 3–4 h. Next, cells were collected according to the standard laboratory procedures for G-band staining and karyotyping.

CNV-seq and WES were serviced by Berry Genomics. Trio-WES was performed to identify gene variations for the family trios. In brief, 1 μg of genomic DNA was extracted from 200 μL peripheral blood, using a Qiagen DNA Blood Midi/Mini kit (Qiagen GmbH, Hilden, Germany) following the manufacturer’s protocol. Fifty nanogram of DNA was fragmented to around 200 bp by enzyme treatment. The DNA fragments were end repaired by adding one A base at the 3′-end. Then, the DNA fragments were ligated with barcoded sequencing adaptors, and fragments at about 320 bp were collected by XP beads. After polymerase chain reaction (PCR) amplification, the DNA fragments were hybridized and captured by NanoWES according to the manufacturer’s protocol. The hybridized DNA products were collected by elution and subjected to PCR amplification and purification. Next, the libraries were quantified by quantitative real-time PCR (qPCR) and size distributions were determined using NanoWES (Berry Genomics, China). Novaseq6000 platform (Illumina, San Diego, USA), with 150 bp pair-end sequencing mode, was used for the family genomic DNA sequencing. Raw image files were processed using CASAVA v1.82 for base calling and generating raw data.

The sequencing reads were aligned to the human reference genome (hg19/GRCh37) using Burrows–Wheeler Aligner tool. PCR duplicates were removed by using Picard v1.57 (http://picard.sourceforge.net/). Verita Trekker® Variants Detection System by Berry Genomics and the third-party software GATK (https://software.broadinstitute.org/gatk/) were employed for variant calling. Variant annotation and interpretation were performed by ANNOVAR and the Enliven^® Variants Annotation Interpretation System referred to multiple databases including:

1. human population databases, such as gnomAD (http://gnomad.broadinstitute.org/), the 1000 Genome Project (http://browser.1000genomes.org), Berrybig data population database, dbSNP (http://www.ncbi.nlm.nih.gov/snp), etc.

2. in silico prediction algorithms, such as SIFT (http://sift.jcvi.org), FATHMM (http://fathmm.biocompute.org.uk), Mutation Assessor (http://mutationassessor.org), CADD (http://cadd.gs.washington.edu), SPIDEX [8], etc.

3. disease and phenotype databases, such as OMIM (http://www.omim.org), ClinVar (http://www.ncbi.nlm.nih.gov/clinvar), Human Gene Mutation Database (HGMD) (http://www.hgmd.org), HPO (https://hpo.jax.org/app/), etc.

The variants are classified into five categories, including pathogenic, likely pathogenic, uncertain significance, likely benign, and benign according to the American College of Medical Genetics and Genomics (ACMG) guidelines for interpretation of genetic variants [9]. Variants with minor allele frequencies <1% in exonic region or with splicing impact were assessed according to the identified ACMG category, inheritance model, available report on the pathogenicity, and recorded clinical synopsis [9–12].

3.1 Study cohort

A total of 28 cases were included in this study, with 54% (N = 15) male fetuses and 45% (N = 13) female fetuses determined by karyotyping. The mean age of pregnant women in this cohort was 29 ± 4 years, and the mean gestational age was 27 weeks, with a range from 21 to 34 weeks. One case had a previous pregnancy with fetal bowel dilatation, while in the other 27 cases fetal bowel dilatation occurred for the first time.

Of the 28 cases, 57% (16/28) pregnancies resulted in live births, 39% (11/28) ended in pregnancy termination, and 4% (1/28) led to fetal demise. Karyotype analysis and CNV-seq were performed in all 28 cases. Twenty-five cases that showed negative results in karyotype analysis and CNV-seq were taken for WES analysis. There were three cases detected with pathogenic variants in karyotyping analysis and CNV-seq, and excluded from WES.

3.2 Detection of genetic variants in cases with different risks of aneuploidy

All the 28 pregnant women were examined with NIPT during pregnancy, of which 26 cases were found carrying low risk of aneuploidy and 2 cases had trisomy 21 at high risk. Two high-risk aneuploidy cases were detected with the presence of the “double-bubble” sign, indicating the duodenal atresia. There was one case out of the 26 low-risk aneuploidy cases detected with pCNVs. Twenty-five cases that showed negative results in karyotype analysis and CNV-seq were taken for WES, and genetic variants were detected in two cases. Differently, both karyotype analysis and CNV-seq suggested Down’s syndrome in two high-risk aneuploidy cases. Overall, the detection in cases with low risk of aneuploidy was 11.54% (3/26), while the detection in cases with high risk of aneuploidy was 100% (2/2) (Table 1).

3.3 Detection of genetic variants in cases with other ultrasound abnormalities

Some cases with fetal bowel dilatation were also found to have other ultrasound abnormalities during the prenatal ultrasound examination, such as polyhydramnios, intestinal malrotation, and fetal echogenic bowel. There were 10 out of the 26 low-risk aneuploidy cases that showed isolated fetal bowel dilatation and had normal genetic testing results, while the remaining 16 cases with other ultrasound abnormalities were detected for genetic variants at a rate of 18.75% (3/16) (Table 2).

3.4 Genetic variant detection in cases at low risk of aneuploidy using WES

Among the 26 cases with low risk of aneuploidy, the application of karyotype analysis, CNV-seq, and WES were found in three cases carrying genetic variants. The karyotype analysis showed normal results. The CNV-seq detected microduplication of chromosome 15q26.1-q26.3 in one case, covering the entire region of 15q26 overgrowth syndrome. The phenotype of this case was typical. This case finally ended in pregnancy termination.

WES revealed genetic variants in two cases with fetal bowel dilatation. One case that resulted in fetal demise carried a heterozygous variant of glial cell-derived neurotrophic factor (GDNF) gene c.329G > A, inherited from his father. This mutation has not been reported previously and it can cause amino acid changes negatively affecting the bio-function of the protein. It is known that GDNF gene mutation can cause an autosomal dominant (AD) Hirschsprung’s disease 3. This disease is characterized by low penetrance [13]. The mother of the case reported that her last pregnancy happened 2 years ago and revealed bowel dilatation in ultrasound examination. The newborn showed non-defecation with vomiting yellow-green secretions after birth, and eventually died 1 month later. No genetic examination was conducted in her last pregnancy. According to the medical history, it was speculated that this mutation led to fetal bowel dilatation in this case.

Another case that had pregnancy termination was identified with a compound heterozygous variant in the solute carrier family 26 member 3 gene (SLC26A3), suggesting a c.1427del variant from the father and a c.269_270dup variant from the mother. The gene mutation can cause autosomal recessive (AR) inherited congenital chloride diarrhea 1. The c.1427del variant was not seen in current genomic databases. The mutation could swift gene open reading frame leading to damage protein function, indicating as strong pathogenicity PVS1 and moderate pathogenicity PM2. The mutation was listed in the Shenzhou Genome database, the human Exon Database (ExAC), the thousand genomes of the reference population (1,000 G), and the Population Genome Mutation Frequency Database (gnomAD). The moderate pathogenicity PM3 is suggested to constitute a complex heterozygote with the mutation site of c.269_270dup. The c.269_270dup variant was also determined as pathogenic (PVS1 + PM2 + PM3) and previously reported as likely pathogenic in the ClinVar database and disease-causing mutation in HGMD [14]. Both variants could lead to protein dysfunction and are considered pathogenic according to ACMG guidelines (Table 3) [9].

Overall, the detection rate of genetic variants was 3.85% (1/26) by CNV-seq and 7.69% (2/26) by WES (Table 4).

3.5 Detection rate of different prenatal diagnosis methods in cases with other ultrasound abnormalities

Among the 16 cases with other ultrasound abnormalities, 5 cases had polyhydramnios, and two of them were detected with pathogenic variants by WES, at a detection rate of 40% (2/5). In other 11 cases, only 1 case was detected with pCNVs by CNV-seq, at a detection rate of 9.09% (1/11) (Table 5).