Discussion
The molecular diagnosis of rare congenital disorders has changed from
family studies by linkage analysis followed by positional cloning to
direct whole exome or whole genome next generation sequencing. Because
every person in the world carries thousands of gene mutations, a
diagnosis can only be made when a mutation is consistent with both the
genetic mode of inheritance and the functional changes leading to the
disorder. In these “experiments of one”, reliance of previously
published cases of similar clinical features and gene mutations makes it
challenging to diagnose cases where mutations are found in genes the
functions of which are little known. Similarly, it is difficult to make
a diagnosis when a deleterious mutation is found in only one copy of a
gene that causes an autosomal recessive disorder.
This study illustrates a way to establish a molecular diagnosis when
there are issues with the mode of inheritance or functional consequences
of a candidate gene mutation. By conducting additional experiments
beyond whole exome or whole genome sequencing, one can determine the
functional consequence of a mutation and explain the inheritance pattern
of the condition. In our case, while the BCAP31 gene is well
established as one associated with DDCH and our patient has clinical
features of DDCH, the particular heterozygous de novo mutation in
our patient has not been reported before. Furthermore, her case is not
consistent with the X-linked recessive mode of inheritance.
We conducted 3 sets of experiments to characterize our case. First, we
exhaustively searched for mutations in the maternally inheritedBCAP31 gene by linked-read whole genome sequencing and optical
mapping to determine that the maternally inherited BCAP31 region
harbors no deleterious mutations. Second, we conducted an XCI assay to
confirm the hypothesis that preferential X inactivation of the
maternally inherited X chromosome bearing the wild-type BCAP31gene created a setting where the patient expresses predominately the
mutated BCAP31 gene. Third, we conducted RNA studies to show that
our patient expresses predominately the mutated gene, resulting in a
frameshifted transcript that skipped over exon 2.
XCI status has been used to explain the wide spectrum of phenotypes
observed (from asymptomatic to severe) in some X-linked diseases in
women, depending on the degree of silencing of the normal allele
(Echevarria et al., 2016; Elstein, Schachamorov, Beeri, & Altarescu,
2012; Fahim et al., 2019; Juchniewicz et al., 2018). For example, the
severity scores and progression phenomena in X-linked Fabry disease
caused by α-Gal deficiency had significant correlation with XCI ratio in
female patients (Echevarria et al., 2016; Elstein et al., 2012;
Juchniewicz et al., 2018). Given the recent observation that skewed XCI
is common in the general female population (Shvetsova et al., 2019),
female patients with clinical features of an X-linked recessive disorder
and one copy of the mutated gene should be screened for skewed XCI.
BCAP31 has 2 protein isoforms translated from four transcripts that
differ in the 5’ UTR in exon 1. Three transcripts (NM_001139441.1,
NM_001256447.2, NM_005745.7) utilize the start codon in exon 2 and
encode a canonical BCAP31 isoform 1 (UniPortKB ID: P51572-1). The
NM_001139457.2 transcript utilizes the start codon in exon 1 and
encodes a BCAP31 isoform 2 (UniPortKB ID: P51572-2). Exon 2 is conserved
in all four transcripts. In our patient, the de novo mutation at
the splicing donor site disrupted the mRNA structure by splicing out the
entire exon 2, resulting in a frameshift transcript utilizing a new
start codon site in exon 3. Consequently, the mutant protein is
shortened and has a completely different structure, essentially
rendering our patient a loss-of-function BCAP31 null. Full genome
analysis that combines linked-read sequencing and optical mapping for
comprehensive single nucleotide and structural variant identification
with phasing allows one to characterize a de novo heterozygous
deleterious mutation in a female patient with BCAP31-related syndrome.
After determining that the de novo mutation arose in the paternal
X chromosome and that the maternal X chromosome is preferentially
inactivation, we conclude that the BCAP31 mutation is the cause
of the patient’s disorder because RNA studies confirms that her WBCs
expresses only the mutant transcript. According to the ACMG guideline,
this variant is deemed “pathogenic” (PS2, PS3, PM2, PP3, PP4) and
deleterious. Our study illustrates that the combination of full genome
analysis (for comprehensive variant detection and phasing), RNA studies
(for alternative splicing and transcript analysis), and XCI studies for
female patients with features of X-linked recessive disorders can lead
to molecular diagnosis of difficult cases of monogenic disorders.