Premature Ovarian Insufficiency in CLPB Deficiency: Transcriptomic, Proteomic and Phenotypic Insights

Abstract

Context: Premature ovarian insufficiency (POI) is a common form of female infertility that usually presents as an isolated condition but can be part of various genetic syndromes. Early diagnosis and treatment of POI can minimize comorbidity and improve health outcomes.

Objective: We aimed to determine the genetic cause of syndromic POI, intellectual disability, neutropenia, and cataracts.

Methods: We performed whole-exome sequencing (WES) followed by functional validation via RT-PCR, RNAseq, and quantitative proteomics, as well as clinical update of previously reported patients with variants in the caseinolytic peptidase B (CLPB) gene.

Results: We identified causative variants in CLPB, encoding a mitochondrial disaggregase. Variants in this gene are known to cause an autosomal recessive syndrome involving 3-methylglutaconic aciduria, neurological dysfunction, cataracts, and neutropenia that is often fatal in childhood; however, there is likely a reporting bias toward severe cases. Using RNAseq and quantitative proteomics we validated causation and gained insight into genotype:phenotype correlation. Clinical follow-up of patients with CLPB deficiency who survived to adulthood identified POI and infertility as a common postpubertal ailment.

Conclusion: A novel splicing variant is associated with CLPB deficiency in an individual who survived to adulthood. POI is a common feature of postpubertal female individuals with CLPB deficiency. Patients with CLPB deficiency should be referred to pediatric gynecologists/endocrinologists for prompt POI diagnosis and hormone replacement therapy to minimize associated comorbidities.

Keywords: CLPB; genetics; infertility; mitochondria; neutropenia; premature ovarian insufficiency; primary mitochondrial disease.

Figure 1.

Whole-exome sequencing and variant filtration identifies compound heterozygous CLPB variants. (A) Variant filtration table showing results for standard gene-centric and variant-centric analysis. Initial data analysis identified only 1 CLPB variant, with the second variant identified by relaxation of filters. (B) IGV-snapshot of WES data indicates the c.1257+5G>A variant is maternally inherited and the c.1249C>T, p.Arg417Ter variant is paternally inherited.

Figure 2.

The novel c.1257G>A variant disrupts CLPB mRNA splicing. (A) RT-PCR of RNA extracted from patient lymphoblasts shows alternative splicing of CLPB mRNA with multiple splice species visible by gel electrophoresis. (B) Schematic representation of the alternative splicing observed in (A), with 3 major splice products corresponding to (1) use of a cryptic splice site and leading to intron retention, (2) residual wild-type splicing and (3) skipping of exon 11. (C) Sequencing of RT-PCR products demonstrates the major splice species lacks exon 11 (45 bp). (D) Sashimi plot generated from RNAseq data from patient lymphoblasts demonstrates skipping of exon 11, residual wild-type splicing and variable intron retention due to inconsistent splicing. (E) RNA seq reads over exon 11 with clean splicing seen in two controls. The majority of mRNA products with intron retention contain the c.1257+5G>A, whereas the majority of mRNA species containing exon 11, harbor the c.1249C>T, p.Arg417Ter variant.

Figure 3.

Proteomics of patient lymphoblasts demonstrate CLPB deficiency and its biological consequences (A) Relative peptide abundance across CLPB protein in the patient compared to controls from whole-cell lymphoblast quantitative proteomics. The mean difference of log₂ MS2 intensities between the patient to controls (n = 2) was plotted for individual peptides across the most common CLPB protein (isoform 2, NP_001245321.1) using loess smoothed curve along with the 95% confidence interval. Missing values in the patient for a single peptide (*) were imputed with the lowest CLPB peptide level detected. The CLPB patient variants were adapted to CLPB isoform 2. (B) Mitochondria were isolated from control and patient lymphoblasts and analyzed by SDS-PAGE on low percentage (10%) acrylamide gel to resolve alternative processing of CLPB protein. Filled arrow indicates mature CLPB bands, asterisk indicates nonspecific banding. (C) The relative level of mature CLPB was quantified and normalized with respect to SDHA loading control, and graphed as average ± SD, (n = 3). Statistical significance was determined by two-sided t test: * P < 0.05. (D) Whole cell dDIA proteomics [LEFT] and filtered mitochondrial component [RIGHT] volcano plots showing relative levels of whole cell/mitochondrial proteins within patient compared to control lymphoblasts. Black dots beyond the vertical lines represent proteins with significant fold change in the patient cell line compared to controls, with significance denoted by P value following a two-sample Student t test (P value <0.05, fold change ±1.5). n = 3 patient replicates and n = 2 independent control replicates. (E) Relative complex abundance (RCA) of OXPHOS complexes in lymphoblasts from CLPB patient relative to controls (n = 2). The middle bar represents the mean complex abundance, upper and lower bars represent 95% confidence interval of the mean. * = P ≤ 0.05, ns = not significant calculated from a paired t test.