An anti-PD-1 antisense oligonucleotide promotes the expression of soluble PD-1 by blocking the interaction between SRSF3 and an exonic splicing enhancer of PD-1 exon 3

PD-1 is a key immune checkpoint molecule. Anti-PD-1 immunotherapy is encouraging in cancer treatment. However, it still needs to be improved. PD-1 has at least five isoforms generated by alternative splicing. An isoform without exon 3 encoding soluble PD-1 (sPD-1) can activate anti-tumor immunity by inhibiting the interaction between cellular surface full-length PD-1 (flPD-1) and PD-L1. However, the regulatory mechanism of exon 3 splicing remains largely unknown. Here, we screened the exon 3 sequence by mutation and searched corresponding splicing factors by SpliceAid database and pulldown assay. The alternative splicing of PD-1 exon 3 was analyzed by RT-PCR. The expression levels of flPD-1 and sPD-1 were analyzed by Western blot, flow cytometry, and ELISA. We discovered that an exonic splicing enhancer (ESE) of exon 3 is essential for its inclusion. Moreover, SRSF3 can bind to this ESE and enhance exon 3 inclusion and flPD-1 expression. We designed and screened out an antisense oligonucleotide (ASO) targeting PD-1 to block the interaction between SRSF3 and ESE, and significantly increase exon 3 skipping and sPD-1 expression, which was verified in various tumor cells in addition to oral cancer cells. Altogether, our results uncovered the regulatory mechanism of human PD-1 exon 3 splicing and sPD-1 expression and further designed a novel anti-PD-1 ASO, which are useful for developing a new method of anti-cancer immunotherapy.

Eleven novel SLC12A1 variants and an exonic mutation cause exon skipping in Bartter syndrome type I

INTRODUCTION

Bartter syndrome type I (BS1) has been rarely reported in large groups. On the other hand, the phenomenon of exon skipping, in which exonic mutations result in abnormal splicing, has been reported to be associated with various diseases. Specifically, mutations that result in the disruption of exonic splicing enhancers (ESEs) and/or the creation of exonic splicing silencers (ESSs) can promote exon skipping. However, the aberrant exon skipping caused by an exonic variant in such splicing regulatory elements (SREs) sequences has never been reported in the causal gene of SLC12A1 in BS1.

METHODS

We analyze the variants in nine Chinese families with BS1, including eight with antenatal BS (aBS) and one presenting as classical BS (cBS), by next-generation sequencing. Then we used bioinformatics programs to analyze all these variants found in this study and identify candidate mutations that may induce exon skipping. Furthermore, the effects of identified variants were classified according to the 2015 American College of Medical Genetics and Genomics (ACMG) standards and guidelines.

RESULTS

Fifteen different variants of SLC12A1 gene were identified, including 11 novel ones. Two of the nine probands were homozygotes, the rest seven ones were compound heterozygotes. One candidate variant (c.1435C>G), not only significantly reduced ESEs scores but also markedly increased ESSs scores, were further investigated by mini-gene splicing assay, and found this single-nucleotide substitution causes abnormal splicing in vitro (exclusion of exon 11). Finally, among 15 variants, 9, 3, and 3 were classified as "pathogenic variants", "likely pathogenic variants", "variants with uncertain significance", respectively.

CONCLUSION

These data would enrich the human gene mutation database (HGMD) and would provide valuable references to the genetic counseling and diagnosis of BS1 for Chinese population. Additionally, our results suggest that aberrant exon skipping is one previously unrecognized mechanism by which an exonic variant in SLC12A1 can lead to BS1.

Exonic CLDN16 mutations associated with familial hypomagnesemia with hypercalciuria and nephrocalcinosis can induce deleterious mRNA alterations

BACKGROUND

Familial hypomagnesaemia with hypercalciuria and nephrocalcinosis type 1 is an autosomal recessive disease characterized by excessive renal magnesium and calcium excretion, bilateral nephrocalcinosis, and progressive chronic renal failure. This rare disease is caused by mutations in CLDN16 that encodes claudin-16, a tight-junction protein involved in paracellular reabsorption of magnesium and calcium in the renal tubule. Most of these variants are located in exons and have been classified as missense mutations. The functional consequences of some of these claudin-16 mutant proteins have been analysed after heterologous expression showing indeed a significant loss of function compared to the wild-type claudin-16. We hypothesize that a number of CLDN16 exonic mutations can be responsible for the disease phenotype by disrupting the pre-mRNA splicing process.

METHODS

We selected 12 previously described presumed CLDN16 missense mutations and analysed their potential effect on pre-mRNA splicing using a minigene assay.

RESULTS

Our results indicate that five of these mutations induce significant splicing alterations. Mutations c.453G > T and c.446G > T seem to inactivate exonic splicing enhancers and promote the use of an internal cryptic acceptor splice site resulting in inclusion of a truncated exon 3 in the mature mRNA. Mutation c.571G > A affects an exonic splicing enhancer resulting in partial skipping of exon 3. Mutations c.593G > C and c.593G > A disturb the acceptor splice site of intron 3 and cause complete exon 4 skipping.

CONCLUSIONS

To our knowledge, this is the first report of CLDN16 exonic mutations producing alterations in splicing. We suggest that in the absence of patients RNA samples, splicing functional assays with minigenes could be valuable for evaluating the effect of exonic CLDN16 mutations on pre-mRNA splicing.

Progressive myoclonus epilepsy in Gaucher Disease due to a new Gly-Gly mutation causing loss of an Exonic Splicing Enhancer

BACKGROUND

Patients with Gaucher Disease (GD) exhibit three phenotypes, including type 1 (non-neuronopathic), type 2 (acute neuronopathic), and type 3 (subacute neuronopathic).

AIM

Identifying which GBA changes represent benign polymorphisms and which may result in disease-causing mutations is essential for diagnosis and genotype/phenotype correlations but is often challenging.

RESULTS

Here, we describe a patient with type 3 GD, presenting with drug-resistant epilepsy, who bears a set of GBA polymorphic variants including the novel c.363A > G (Gly82Gly) synonymous mutation. In silico predictions, mRNA and functional studies revealed that the new Gly82Gly mutation causes skipping of GBA exon 4, leading to a severe reduction of the wild type GBA mRNA. This is the first report of a synonymous change causing GD through loss of an exonic splicing enhancer sequence. The synonymous mutation is in trans with the Asn188Ser missense mutation, thus making the Asn188Ser responsible for the patient's phenotype and strengthening the association of Asn188Ser with the particular neurological phenotype of type 3 GD.

CONCLUSION

We strengthen the association of Asn188Ser with the type 3 GD phenotype and progressive myoclonus epilepsy. Our data confirm that in silico predictions and mRNA analysis are mandatory in discriminating pathological mutations from the background of harmless polymorphisms, especially synonymous changes.

First molecular diagnosis of Donohue syndrome in Africa: novel unusual insertion/deletion mutation in the INSR gene

Donohue syndrome (DS) is a very rare autosomal recessive disease affecting less than one in a million life births. It represents the most severe form of insulin resistance due to mutations involving the insulin receptor (IR) gene "INSR". DS is characterized by pre- and postnatal growth retardation with failure-to-thrive, lipoatrophy, acanthosis nigricans, hypertrichosis, and dysmorphic features. An exhaustive INSR gene sequencing was performed after PCR amplification of coding exons and introns boundaries. Bioinformatic tools, including ESEfinder, MFOLD and Proter software were also used to predict the impact of INSR mutation on INSR on gene expression as well as on the protein structure and function. The results have shown a novel unusual c.3003_3012delinsGGAAG (p.S1001_D1004delinsRE) insertion/deletion (indel) mutation within the exon 16 in the three patients, which represent the fourth indel mutation within the INSR gene. The mutation modifies the secondary structure of DNA and RNA, as well as the composition of exonic splicing enhancers of exon 16. Moreover, despite the conservation of the secondary structure of the IR, the p.S1001_D1004delinsRE in-frame mutation is accompanied by the loss of four amino acids replaced by two residues of different nature and hydrophobicity level in the juxtamembrane domain of the receptor. The results have confirmed the role of the juxtamembrane domain of IR involved in a crucial interaction of the IR with cellular effectors essentially the IR substrate 1 (IRS-1), the SHC and the Nck proteins that ensure the signal mediated by the insulin transduction pathway in target cells. Our findings have also proven the genotype/phenotype correlation between INSR mutation and DS phenotype severity.

A synonymous polymorphic variation in ACADM exon 11 affects splicing efficiency and may affect fatty acid oxidation

In recent studies combining genome-wide association and tandem-MS based metabolic profiling, a single-nucleotide polymorphism (SNP), rs211718C>T, located far upstream of the MCAD gene (ACADM) was found to be associated with serum concentrations of medium-chain acylcarnitines indicating improved beta-oxidation of medium-chain fatty acids. We examined the functional basis for this association and identified linkage between rs211718 and the intragenic synonymous polymorphic variant c.1161A>G in ACADM exon 11 (rs1061337). Employing minigene studies we show that the c.1161A allele is associated with exon 11 missplicing, and that the c.1161G allele corrects this missplicing. This may result in production of more full length MCAD protein from the c.1161G allele. Our analysis suggests that the improved splicing of the c.1161G allele is due to changes in the relative binding of splicing regulatory proteins SRSF1 and hnRNP A1. Using publicly available pre-aligned RNA-seq data, we find that the ACADM c.1161G allele is expressed at significantly higher levels than the c.1161A allele across different tissues. This supports that c.1161A>G is a functional SNP, which leads to higher MCAD expression, perhaps due to improved splicing. This study is a proof of principle that synonymous SNPs are not neutral. By changing the binding sites for splicing regulatory proteins they can have significant effects on pre-mRNA splicing and thus protein function. In addition, this study shows that for a sequence variation to have an effect, it might need to change the balance in the relative binding of positive and negative splicing factors.