1 INTRODUCTION

Spliceosomopathies encompass a group of diseases due to mutations in genes encoding either major or minor spliceosomes.^{1, 2} The spliceosomes are large ribonucleoprotein complexes that are required for removal of introns from pre-messenger RNA (pre-mRNA) and for the ligation of exons to generate mature mRNAs.³ The major spliceosome (MS) is composed of five small nuclear RNAs (snRNA) known as U1, U2, U4, U5, and U6, while the minor spliceosome was known as U12 snRNA. Recent data provided evidences that disrupted RNA splicing integrity can lead to mis-splicing of pre-mRNA. Defects in splicing usually tend to be tissue specific and they have been associated with several genetic disorders including retinitis pigmentosa, hearing defects and spinal muscular atrophy and microcephaly osteodysplastic dwarfism type I.²

Peptidyl-Prolyl Isomerase Like-1 (PPIL1) is encoding an active PPIase substrate that is incorporated within the MS and is essential in regulating the splicing of short and high GC-content introns.¹ The knock in mice showed microcephaly with remarkably reduced cerebral and cerebellar size that recapitulated the phenotype of reported patients.¹

Here, our work refines and expands the clinical and neuro-imaging findings of PPIL1 by reporting nine new cases and describes p.Ala99Thr as a founder variant among Egyptian patients. We also suggest that this variant is responsible for a severe phenotype showing microlissencephaly in addition to the pontocerebellar malformation.

2 MATERIALS AND METHODS

3 RESULTS

3.3 Imaging findings

Brain imaging of patients demonstrated profound microcephaly with variable degrees of severity as regard to the hypoplasia of frontal lobe, diffuse posterior predominant undersulcation, agenesis of the corpus callosum, colpocephaly, pontocerebellar hypoplasia, abnormal hippocampus and Dandy-Walker malformations (Figure 1 and Supplementary Figure 2).

3.6 Molecular results

Exome sequencing revealed a previously reported missense variant in the PPIL1 gene in all our patients, c.295G>A (p.Ala99Thr). According to the ACMG classifications, it is classified as a pathogenic variant. The c.295G>A (p.Ala99Thr) variant was found in the homozygous form in the patients and the in heterozygous state in their respective parents (Figure 2). Thorough investigations of the patients' exome data did not reveal any other candidate variants that can be related to the described manifestations or phenotype.

4 DISCUSSION

Recently, affected splicing integrity of major spliceosomes was established as a new mechanism for congenital microcephaly in association with pontocerebellar hypoplasia and caused by biallelic variants in PPIL1.¹ Our study presents nine new patients with congenital microcephaly, simplified gyral pattern, pontocerebellar hypoplasia, delayed myelination, agenesis/hypogenesis of corpus callosum that are in line with the previous report.¹ Unexpectedly, all our patients (who originate from different governorates in Egypt) had the previously reported homozygous p.Ala99Thr variant of PPIL1 gene. Of note, one of the two Egyptian families reported by Chai et al. had p.Ala99Thr while the other had p.Arg131Gln.¹ Currently, all the patients with p.Ala99Thr are of Egyptian origin and this variant is the most common pathogenic variant (>50% of cases) in PPIL1. Generally, the p.Ala99Thr is an extremely rare variant with only four heterozygous individuals in gnomAD (Allele frequency = 0.00001417). The presence of a founder effect was highly suspected which was confirmed by the presence of a single shared homozygous haplotype of 8 Mb in all patients (Supplementary Table 1). Therefore, p.Ala99Thr variant represents an Egyptian founder variant and targeted sequencing of this variant should be considered first in Egyptian patients.

Ten pathogenic variants in PPIL1 gene have been reported including seven missense, one nonsense, one inframe duplication and one splice site variant. All except for the p.Arg131Gln are located on the enzymatic face of the protein. It was observed that RNA level was slightly higher in p.Ala99Thr when compared with others possibly due to compensatory upregulation.¹ Therefore, the p.Ala99Thr is considered a hypomorphic variant.¹

Besides all our patients displayed common craniofacial features such as bitemporal narrowing, upturned nostrils, tented upper lip, long philtrum, large ears (although nonspecific), they all shared recognizable neuroimaging findings but with variable severity regarding the involvement of the posterior fossa and the gyral pattern (Table 1). Patient 1 showed remarkable paucity of the gyri especially posterior with decreased cortical thickness, small interhemispheric cyst, and complete agenesis of corpus callosum. He also exhibited inferior vermis hypoplasia with enlargement of the fourth ventricle and posterior fossa, obtuse fastigial recess and unpaired caudal vermis lobule giving the tail sign pointing to the diagnosis of Dandy-Walker malformation.⁴ In keeping with this, Patients 2 and 3 from our cohort exhibited the same malformations. As such, we guess that PPIL1 has a major role in the hindbrain development.

In Ppil1^A99T/A99T mouse model, the neuronal progenitors showed postmitotic apoptosis leading to intrauterine small head circumference with both reduced cerebral and cerebellar size¹ unlike primary microcephaly that affects mitosis and survival of neural progenitors or shows premature neurogenesis.⁵ Although, congenital microcephaly is not a constant feature in pontocerebellar hypoplasia spectrum but it was observed in our cohort and those reported previously.¹ In addition, the simplified gyral pattern observed in PPIL1 mutated patients differs markedly from the atrophic changes seen in the different types of the pontocerebellar hypoplasia spectrum which would argue against placing them in the pontocerebellar hypoplasia spectrum (OMIM 619301) as PCH14. On the contrary, the absence of abnormal gyral pattern in some patients reported by Chai et al. (Families 5, 7, and 8) are against categorizing them in the microlissencephaly spectrum.¹ Our patients were initially categorized as microlissencephaly prior to the genetic analysis because they are combining the features of severe congenital microcephaly and the abnormal gyral pattern.^{6, 7} We suggest to use the term PPIL1-related spectrum.

It was postulated that the clinical severity is correlated with the degree of impairment of protein stability or function and this was evidenced by the observation of severe phenotype of compound heterozygous Ppil1^A99T/fs in mutant embryos when compared with Ppil1^A99T/A99T.¹ Our patients showed a more severe neuroradiologic phenotype when compared with those of Chai et al. (Patients 6, 8, and 9) who had duplication, splicing and nonsense variants respectively.¹ Interestingly, the Egyptian patient who had p.Ala99Thr showed a similar phenotype to our patients. We are not sure if the p.Ala99Thr is associated with this severe phenotype, this will be an interesting consideration for future studies. Nevertheless, the reason for the variable severity among our patients with p.Ala99Thr is also unclear, perhaps other factors are involved, such as modifying genes as it is well known that PPIL1 gene binds to regulatory elements of genes involved in neurodevelopment and activates gene transcription. As such, genotype/phenotype correlation in PPIL1 remains elusive.

While all our patients developed myoclonic seizures shortly after birth, two of the patients reported by Chai et al. did not develop seizures. This could be explained by the early death (died at the age of 2 months).¹ The prenatal and postnatal growth retardation found in our patients were not that severe as those described with primordial dwarfism and microlissencephaly caused by RNU4ATAC that affect the minor spliceosome.⁸ Moreover, three of our patients (33.3%) with p.Ala99Thr PPIL1 variant did not show either intrauterine (Patients 2 and 9) or postnatal growth retardation (Patients 1 and 2).

Interesting is the occurrence of pancytopenia in four of our patients (44.4%) and bone marrow failure in one of them (11.1%) was reminiscent of Fanconi anemia syndrome.⁹ Moreover, Chai et al. reported three patients with hematopoietic disorders including two patients with neutropenia (p.Phe82Ser, p.Arg131Gln, and p.Tyr78Cys) and one patient with thrombocytopenia (p.Thr107Ala).¹ Unlike Fanconi syndrome, the age of onset of bone marrow failure was very early in life. Nevertheless, hematopoietic disorders are not observed in every case with PPIL1 variant. This could be attributed to the genetic background effects. The early lethality of most of our patients precluded the extensive hematologic assessment. Hematologic assessment should be investigated in any patient with PPIL1 variant as the potential role of PPIL1 on the bone marrow needs to be further clarified.

In conclusion, we propose that PPIL1 variants are associated with a complex and heterogeneous phenotype. Further, we highlight the recognizable brain imaging anomalies for p.Ala99Thr. This recognizable pattern provides useful elements for the differential diagnosis of patients with pontocerebellar hypoplasia. Additional data will be needed to assess the exact prevalence of hematopoietic disorders and/or bone marrow failure among patients and the precise role of the PPIL1 gene in hematopoietic disorders.

AUTHOR CONTRIBUTIONS

Ghada Abdel-Salam, involved in clinical evaluation and follow-up of the patient and interpreted the brain MRI. Mohamed Abdel-Hamid involved in next Generation sequencing, data analysis and interpretation and Sanger confirmation. Both participated equally in writing and approved the final version.

CONFLICT OF INTEREST STATEMENT

The authors have no relevant financial or non-financial interests to disclose.

ETHICS STATEMENT

Written informed consent for participation and publication was received from all individuals and adherent to the Declaration of Helsinki, and the study was approved by the Medical Research Ethics Committee of the NRC (Approval number: 19259).