Phenotypic spectrum of RNU4ATAC-related spliceosomopathies: four novel cases and integrated reevaluation of previously reported patients

Abstract

Background

Homozygous or compound heterozygous variants in RNU4ATAC, which transcribes a non-coding RNA component of the minor spliceosome, have been associated with a spectrum of disorders, collectively known as RNU4ATAC-related spliceosomeopathies. The phenotypic spectrum of RNU4ATAC-related disease is characterized by dysmorphic features, growth delay, neurological and skeletal features, whose severity ranges from the microcephalic osteodysplastic primordial dwarfism type 1 (MOPD1) to the milder Roifman syndrome.

Objectives

To characterize the clinical spectrum and evaluate long-term outcomes of RNU4ATAC-related diseases.

Methods

We evaluated the phenotypic features of four novel patients with deleterious RNU4ATAC variants, diagnosed by means of whole genome sequencing. Same features were evaluated in previously published cases, identified by literature research on PubMed.

Results

We identified four novel cases with deleterious compound heterozygous variants in RNU4ATAC, which were not restricted to the 5’ stem-loop, including three adult patients. Reported cases expand the clinical spectrum of RNU4ATAC-related disorders, highlighting renal disease, autoimmunity and systemic inflammation as possibly more frequent yet previously under-recognized features. Immunological investigations reveal enhanced HLA-DR and PD-1 expression in T cells from tested patients, suggesting T cell activation and exhaustion. Reevaluation of all previously published cases confirms the strong correlation of RNU4ATAC variants located exclusively at the 5’ stem-loop with severe lethal disease falling under MOPD1. Genotypes carrying at least one variant that spares the 5′ stem-loop are associated with a milder phenotype and later onset.

Conclusion

Homozygous or compound heterozygous RNU4ATAC variants affecting the 5’ stem-loop region are associated with severe phenotypes and adverse disease courses. In contrast, genotypes sparing the critical 5′ stem-loop region of RNU4ATAC can cause a complex phenotype that is not necessarily dominated by dysmorphic features or growth failure, but rather by immunodeficiency and immune dysregulation.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13023-026-04300-x.

Introduction

RNU4ATAC transcribes the small nuclear RNA (snRNA) U4atac, a core component of the minor or U12-dependent spliceosome, which removes minor introns, residing in approximately 750 human genes [1–3]. The 3’ stem loop of U4atac is followed by a single strand region serving as the binding site of Sm proteins, required for the maturation and stability of U4atac [4–6]. The stem I and stem II elements of U4atac pair with U6atac and are separated by the 5’ stem-loop. Although U4atac itself is not catalytically active, it base-pairs U6atac, thereby stabilizing and enabling U6atac to execute its catalytic function. Biallelic variants in RNU4ATAC cause heterogenous disorders, including the microcephalic osteodysplastic primordial dwarfism type I (MOPD1; OMIM #210710), also named Taybi-Linder syndrome, the Roifman (RS; OMIM #616651) and the Lowry-Wood syndrome (LWS; OMIM #226960) [7–10]. All three RNU4ATAC-related spliceosomopathies – collectively referred to as RNU4ATAC-opathies – share a common core phenotype characterized by dysmorphic facial features, growth retardation, and neurological and skeletal involvement of variable severity [11]. In contrast to MOPD1, LWS and RS are associated with a different clinical spectrum, including disease-specific dysmorphic features and survival extending into late childhood or adulthood. Immunodeficiency, retinal anomalies and variable organ-related manifestations have been rather associated with RS and have likewise been reported in LWS.

The majority of RNU4ATAC variants associated with MOPD1 are located exclusively at the 5’ stem loop, whereas variants associated with RS are typically compound heterozygous, with one variant at the 5’ stem loop or the Sm protein binding element and the second variant at the stem II region [12]. The higher pathogenic potential of variants located at the 5’ stem loop region, reflected by the profound disease severity in patients with variants located exclusively at the 5’ stem loop, may be explained through the severely impaired binding of U4atac to key protein constituents of the minor spliceosome (i.e. NHP2L1 and PRPF31) [9, 13, 14].

Defective minor intron splicing as a consequence of deleterious biallelic variants in RNU4ATAC results in intron retention and elevated alternative splicing, affecting the tissue-specific distribution of mRNA isoforms [15, 16]. The latter can impair the function of minor intron-containing genes and their targets. Olthof et al. detected alternatively spliced minor intron-containing genes in peripheral immune cells from patients with RS or LWS, which suggests the pathogenic effect of aberrations in minor intron splicing in the immune system [15]. Likewise, the neurological, retinal, skeletal and other ciliopathy-consistent manifestations may result from the marked enrichment of cilium-related genes within the subset of genes carrying minor introns [17]. With respect to RS-associated immunodeficiency, altered MAPK1 splicing, leading to impaired BAFF–MAPK1–MAPK3 signaling that normally promotes survival beyond the transitional B-cell phase, has been proposed to account for the aberrant B-cell maturation [9, 16–17].

To date, RNU4ATAC-opathies have been commonly reported in young children, most frequently diagnosed with MOPD1. The clinical spectrum and outcome of patients with milder disease remains barely characterized. To this end, we evaluated the phenotypic features of four novel patients with deleterious RNU4ATAC variants. Our cases, especially adult ones, highlight immune dysregulation in the form of autoimmunity and systemic inflammation as later-onset feature of RNU4ATAC-related spliceosomopathies. Review of previously reported cases suggests the phenotypic continuum of RNU4ATAC-opathies with two clearly opposing ends of severity.

Materials and methods

Patient cohort and collection of clinical data

This retrospective cohort study included patients, visiting the outpatient clinics of the Hannover Medical School or the Dresden University Hospital. Available clinical and laboratory data recorded until 06/2025. All patients and parents provided written informed consent and the study was approved by the Ethics Committee of the Hannover Medical School (approval No. 11223_BO_K_2024). Clinical data were obtained from patients’ medical files as described previously [18]. Recorded phenotypic features included body measurements (weight, height, head circumference) and features of cranial and facial dysmorphisms. All diagnoses, imaging studies and laboratory findings, including immunological investigations and clinical chemistry results (including C-reactive protein (CRP), liver enzymes, renal function markers and urine investigations) were reviewed retrospectively.

Genetic analysis

Sample preparation, whole genome sequencing (WGS) analysis and the assessment of rare single nucleotide variants (SNV) were performed as described previously [19]. Briefly, genomic DNA was extracted from peripheral whole blood samples collected in tubes with ethylenediaminetetraacetic acid (EDTA) and sequenced on Illumina NovaSeq 6000 using the Lotus DNA Library Prep Kit (IDT, Leuven, Belgium) or the DNA PCR-Free Prep Kit (Illumina, United States). 150-bp paired-end reads were generated, aligned to the human reference genome (UCSC Genome Browser build GRCh38/hg38) and analyzed with the next-generation sequencing (NGS) data analysis pipeline (megSAP v.0.2-8-ga9d80c2 or TruSight^™ Software Suite (illumine)). SNVs were filtered for genes associated with inborn error of immunity (IEI) and classified following the American College of Medical Genetics (ACMG) Standards and Guidelines [20]. Follow-up of SNVs was performed using databases Exome Aggregation Consortium19 (ExAC), the 1000 Genomes Project and the genome Aggregation Database (gnomAD) for filtering for rare variants (≤ 1% minor allele frequency, MAF). The Leiden Open Variation Database (LOVD), ClinVar and prediction tools including REVEL, phyloP, SIFT, PolyPhen2, FATHMM, and CADD were used for evaluation of pathogenicity.

Systematic review of published cases with RNU4ATAC-related spliceosomopathies

Cases reported until 06/2025 were reviewed using PubMed search term: “Roifman syndrome” OR “MOPD1” OR “Lowry-Wood syndrome”, following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (Suppl. Figure 1) [21]. Cases without documented biallelic variants as well as embryonic and fetal cases were not considered. All available genotypic, phenotypic and immunological data were abstracted and summarized together with the data from the here presented cases. Evaluated data included patient demographic data, the diagnosis of RS, LWS or MOPD1, reported intrauterine growth retardation (IUGR), small for gestational age (SGA), postnatal growth delay, dysmorphic features (facial dysmorphism, dysmorphic digits (including brachydactyly, syndactyly, clinodactyly), dysmorphic limbs), cardiac abnormalities, infectious manifestations and results of immunological investigations (immunoglobulin levels, vaccine responses, B cell and T cell counts). Patients diagnosed with MOPD1 were considered to present with cranial dysmorphism in the form of microcephaly, along with prenatal and postnatal growth retardation, unless otherwise specified in the original case reports. Reported hypogammaglobulinemia, dysgammaglobulinemia, failed vaccine response or reported severe or frequent infections were collectively evaluated as the presence of “clinical or immunological evidence of immunodeficiency”. Reported joint contractures, dislocations, scoliosis, kyphosis and epiphyseal or spondyloepiphyseal dysplasia were also collectively evaluated as “skeletal abnormalities”. Patients reported with vermian hypoplasia were documented to display cerebellar hypoplasia. Reported intracranial cysts, vermian hypoplasia or dysplasia, cerebellar hypoplasia, gyration abnormalities or dysgenesis of the corpus callosum were collectively evaluated as “brain malformation”. Patients diagnosed with atopic dermatitis were documented with eczema. In case of reported ichthyosis, we also documented dry skin. Chilblain, eczema or dry skin, were also collectively evaluated as “skin abnormalities”.

Statistical analysis

For statistical calculation we used GraphPad prism 9 (GraphPad, La Jolla, USA). Descriptive statistics are reported as median and interquartile range (IQR) in case of continuous variables and as counts and percentages for dichotomous variables. Categorical variables were compared by the Yate’ s continuity corrected chi-squared test, which was employed to compare patients with 5’ stem loop variants exclusively with the rest of documented patients. Differences in lymphocyte counts between examined patients and matched healthy blood donors were evaluated with the Mann-Whitney test. Probabilities of survival were compared with the Gehan-Breslow-Wilcoxon test.

Results

Case reports

P1 is a 42-year-old female with a history of recurrent sinusitis and otitis media since approximately the age of 30 years, leading to frequent antibiotic treatment (Fig. 1; Table 1). Her medical history included congenital asplenia, atopic dermatitis presenting already in early childhood and Hashimoto’s thyroiditis, leading to treatment with levothyroxine. Shortly before her first presentation at our immunology outpatient clinic, increased serum creatinine and non-nephrotic proteinuria associating with elevated perinuclear antineutrophil cytoplasmic antibodies (ANCA) against myeloperoxidase (MPO), led to treatment with high-dose prednisolone. Prednisolone treatment improved serum creatine and reduced proteinuria. Suspecting an ANCA-positive vasculitis, a kidney biopsy was recommended; however, the patient declined the procedure. A computed tomography scan of the chest revealed no lesions suggestive of ANCA-positive vasculitis, interstitial lung disease or bronchiectasis. Recurrent upper respiratory tract infection led to immunological investigations prior to the introduction of prednisolone treatment, revealing reduced IgA and IgG4 levels. During high-dose steroid treatment, total IgG levels were also transiently low and recovered after tapering of prednisolone. Given diagnosed dysgammaglobulinemia and recurrent infections, we introduced a prophylactic treatment with azithromycin, which reduced the frequency of infections. Genetic testing by means of whole genome sequencing identified compound heterozygous variants in RNU4ATAC gene (Table 1) and led to the diagnosis of an RNU4ATAC-related spliceosomopathy.

Fig. 1.

Features of patients with RNU4ATAC-related spliceosomopathy. P1 with prominent nose, high nasal root and broad nasal bridge (A). Sparse scalp hair at frontal hairline in case of P2 (B). Computed tomography scan from P2 showing cylindrical bronchiectasis of left upper lobe (marked with “*”, C). Kidney biopsy from P3 (right) revealing severe transmural and cuff-shaped hyalinosis of small preglomerular vascular branches with loss of smooth muscle cells (marked with an arrow, D) (Jones’ methenamine combined with hematoxylin and eosin stain; bar represents 20 μm). Electron microscopy showing small segmental glomerular sclerosis (marked with “*”, E) and 30% loss of podocyte foot processes (marked with arrowheads, F) (electron microscopy after contrasting with osmium tetroxide and uranyl acetate in E and F; bars represent 2,500 nm in E, and 1,000 nm in F)

Table 1.

Phenotypic features of four novel patients with biallelic deleterious variants in RNU4ATAC

Pat. ID	RNU4ATAC variants	Sex	Age*	Growth failure	Cranial/Facial dysmorphism	Other Syndromic features	Skeletal features	Infections	Neurological features	Ophalomological features	Endocrinological features	Integumentary features	Kidney disease	Other features
1	n. 8 C > T; n. 51G > A	F	43.2	no	prominent nose, high nasal root, broad nasal bridge, long palpebral fissures	broad thumbs, thick fingers	scoliosis, osteoarthritis of the spine	recurrent URTI, recurrent bronchitis	no	myopia (OD -1.25 D/ OS -8.25 D)	Hashimoto’s thyroiditis leading to hypothyroidism	atopic dermatitis, rosacea, sparse scalp hair at frontal hairline	ANCA associated vasculitis-like syndromes with chronic kidney disease, CKD G3aA3, creatinine 105 µmol/l (eGFR 54 ml/min/1.73m2), S-cystatin C 1.29 mg/l, (cystatin C clearance 55 ml/min/1.73 m2); urine findings in spot urine: UPCR ratio 406 mg/g creatinine; UACR ratio 275 mg/g creatinine; urine sediment with erythrocyturia & leukocyturia	congenital asplenia, elevated liver enzymes (both ALT & gGT)
2	n. 8 C > T; n. 37G > A	F	32.3	no	microcephaly, high nasal root, broad nasal bridge, deep set eyes, long palpebral fissures, thin upper lip	deep set eyes	scoliosis	recurrrent URTI & LRTI (including otitis media, bronchitis and pneumonia), bronchiectasis Helicobacter pylori–related gastritis, cylindrical bronchiectasis in the right upper lobe, middle lobe & lingula, status post resection of the left upper lobe of the lungs, recurrent herpes labialis, common warts at fingers, herpes zoster	no	myopia, astigmatism	no	sparse scalp hair at frontal hairline, laterally thinned eyebrows	CKD G2A2 creatinine 89 µmol/l (eGFR 85 ml/min/1.73m2), S-cystatin C 1,16 mg/l, (cystatin C clearance 66 ml/min/1.73 m2), urine findings in spot urine: UPCR ratio 386 mg/g creatinine, UACR ratio 47 mg/g creatinine; urine sediment unremarkable	no
3	n. 13 C > T; n. 117 A > G	F	37.6	SGA (likely also IUGR), dwarfism	sloping forehead, high nasal root, broad nasal bridge, dysplastic ears, long midface	bilateral 5th finger clinodactyly, brachydactyly	spondyloepimetaphyseal dysplasia, reduced bone age, genua valga, short metacarpals	recurrrent URTI & LRTI (including pneumonia, bronchitis & sinusitis), bronchiectasis at middle lobe, Salmonella enteritis, cutaneous abscesses, severe COVID-19 (leading to hospitalization), vulvar cancer (pT1a pN0 L0 V0 Pn0 R0), associated with HPV 44, 45 and 54 (treated with vulvectomy), thrush (single episode)	delayed motor milestones	retinitis pigmentosa, intermittent divergent strabismus	IGF1 deficiency, thyreoiditis	dry skin, transient ichthyosis, lichen ruber, sparse scalp hair, laterally thinned eyebrows	nephrotic range proteinuria, CKD G4A2, creatinine 213 µmol/l (eGFR 25 ml/min/1.73m2), S-cystatin C 3.34 mg/l, (cystatin C clearance 16 ml/min/1.73m2). progressing to CKD; kidney biopsy: Severe transmural and cuff-shaped hyalinosis of small preglomerular vascular branches with loss of smooth muscle cells. Electron microscopy reveals a small segmental glomerular sclerosis and 30% loss of podocyte foot processes. urine findings in spot urine: UPCR ratio 227 mg/g creatinine; UACR ratio 52 mg/g creatinine; urine sediment with leukocyturia	gout, hyperuricemia, cholestatic hepatopathy, splenomegaly
4	n. 13 C > T; n. 51G > A	M	8.0	IUGR and postnatal growth retardation	microcephaly, dolichocephaly, long philtrum, micrognathia, upslanting palpebral fissures, large fleshy & upturned nose with hypoplastic alae nasi, low-set & posteriorly rotated ears	bilateral 5th finger clinodactyly, brachydactyly, bilateral transverse palmar creases	genua valga	recurrent mild URTI (rhinitis, cough, otitis media), chronic otitis media, conjunctivitis	unilateral (left) polymicrogyria, right hemiparesis, muscular hypotonia at neonatal age, delayed motor milestones, intellectual disability, mental retardation	no	no	dry and irritable skin, atopic dermatitis, long prominent eyelashes	no	patent foramen ovale, conductive hearing loss, micropenis, cryptorchidism leading to orchidopexy and funicolysis

ALT, Alanine transaminase; ANCA, anti-neutrophil cytoplasmic antibody; CKD, chronic kidney disease; COVID-19, coronavirus disease 2019; D, diopters; F, female; gGT, gamma-glutamyltransferase; HPV, human papillomavirus; IGF1, insulin-like growth factor 1; IUGR, intrauterine growth retardation; LRTI, lower respiratory tract infections; M, male; S, serum; UPCR, urine protein creatinine ratio; SGA, small for gestational age; UACR, urine albumin creatinine ratio; URTI, upper respiratory tract infections

*at reporting

P2 is a 33-year-old female with a history of recurrent respiratory tract infections, including bronchitis and otitis media since the age of 1–2 years (Fig. 1; Table 1). She first developed pneumonia at the age of 12 years. Since then, she suffered from recurrent pneumonias and at the age of 18, she was diagnosed with bilateral cylindrical bronchiectasis (Fig. 1), leading to resection of the upper left lobe of the lung. At the age of 17, she was diagnosed with herpes zoster for the first time. Furthermore, she had a history of recurrent episodes of herpes labialis and refractory confluent common warts affecting the digits at her extremities. Immunological investigations revealed low lgG2 and IgG3 antibodies and led to the diagnosis of selective subclass deficiency and the introduction of immunoglobulin replacement therapy, which the patient tolerated well in its subcutaneous form. Immunoglobulin replacement therapy reduced infection rate. Similar to P1, this patient displayed mildly increased serum creatinine and non-nephrotic proteinuria.

P3 is a 37-year-old female who was born with low birth weight, corresponding to the 6th percentile (Table 1). At the age of 10 weeks, she was diagnosed with a bilateral pneumonia. During early childhood she experienced recurrent episodes of pyelonephritis, attributed to grade 2 vesicoureteral reflux. At the age of 7 years, recurrent episodes of pneumonia and bronchitis led to immunological investigations and the diagnosis of humoral immunodeficiency with low IgM and IgG4 as well as markedly reduced isoagglutinin levels. Consequently, an immunoglobulin replacement treatment was introduced, leading to reduced infection frequency. Ophthalmologic assessment at 3.5 years raised suspicion of cone-rod dystrophy. Worsening visual acuity led to follow-up ophthalmologic investigations at the age of 12 years, whose findings were consistent with a retinitis pigmentosa. Dwarfism, dysmorphic features, including brachydactyly and clinodactyly as well as features of skeletal dysplasia were consistent with a syndromic disorder. At 19 years of age, she developed mild proteinuria, which progressed over the next five years to nephrotic-range proteinuria. This prompted a kidney biopsy, revealing severe segmental hyalinosis of the smaller preglomerular vascular branches, with partly transmural and cuff-shaped lesions and segmentally marked loss of smooth muscle cells (Fig. 1). Electron microscopy demonstrated a small segmental sclerosis and 30% loss of podocyte foot processes (Fig. 1D and E). These findings excluded an immune-mediated glomerulonephritis, did not fulfil the criteria of podocytopathy and did not correspond to any known glomerulopathy. An ultrasound of the kidney at the age 32, showed increased echogenicity. The right kidney was smaller than the left kidney but no vesicoureteral reflux, cystic or dysplastic kidneys were detectable. The patient was treated with an angiotensin-converting enzyme inhibitor, resulting in a mild reduction in proteinuria. Her nephropathy subsequently progressed slowly to chronic kidney disease. The combination of immunodeficiency, nephropathy, and skeletal abnormalities suggested immuno-osseous dysplasia, which could not be confirmed by whole exome sequencing (WES). Consistent with three other cases reported here, the diagnosis of RNU4ATAC-related spliceosomopathy was ultimately established using whole genome sequencing (WGS) (Table 1).

Finally, P4 was an 8-year-old male patient, born hypotrophic at 35 weeks of gestation. At birth, he presented with primary microcephaly and generalized hypotonia. Over the course of early infancy, he developed progressive failure to thrive and short stature. A combined motor and speech developmental delay became evident from 2 to 3 months of age, leading to placement in a special-needs school at the age of 7 years. Neurological examination revealed decreased spontaneous movement of the right extremities with spastic paresis, which showed slight improvement over time. Since his infancy, the patient experienced recurrent respiratory tract infections, averaging 1–2 episodes per month, along with repeated purulent otitis media. At the age of 5, he underwent otolaryngologic surgery, including paracentesis and tympanostomy tube insertion. Remarkably, no hospitalizations for severe infections have occurred. All recommended vaccinations were administered without complications. Following vaccination, the patient exhibited protective serum antibody levels to tetanus and measles. Neuroimaging revealed temporo-occipital polymicrogyria, white matter thinning and consequent enlargement of the left lateral ventricle. Electroencephalogram recordings revealed intermittent epileptiform discharges over the left cerebral hemisphere and the right frontal region, without clinical seizure activity. Echocardiography and renal ultrasonography were unremarkable and serum renal function parameters remained within the normal range so far. The combination of neurodevelopmental delay, failure to thrive and short stature led to genetic testing by means of WGS at the age of six years, revealing two pathogenic RNU4ATAC variants leading to the diagnosis of an RNU4ATAC-related spliceosomopathy. Follow-up neurological and immunological counselling was established. This patient did not receive any disease-related long-term treatments so far.

Summary of clinical findings and laboratory results

Considering the current clinical perception of RNU4ATAC-related spliceosomopathies, patients P3 and P4, were displaying obvious dysmorphic features and growth abnormalities, i.e. dwarfism and history of IUGR that represent rather typical features of TALS. In contrast, in case of the other two patients (P1 and P2), genetic diagnosis was unexpected given their very mild rather infection-immunodeficiency-dominated phenotype and the lack of apparent dysmorphic features. Furthermore, both those patients demonstrated no evidence of cognitive impairment and maintained normal social and professional functioning. All three adult patients (P1-P3) presented with chronic kidney disease (CKD) associating with proteinuria (Table 1). The presence of ANCA-MPO autoantibodies in P1, along with the improvement in creatinine levels and urine parameters under steroid treatment indicated autoimmunity as a likely cause of renal disease. In P3, however, the absence of hemodynamic, drug-related or other cause of a podocytopathy may suggest a directly RNU4ATAC-related etiology of CKD. Further, in all three patients with kidney disease, WGS excluded a monogenic CKD etiology [22] other than the pathogenic RNUA4ATAC variants. Clinical findings from all four cases are summarized in Table 1.

Reassessment of laboratory findings, revealed persistently elevated levels of acute phase reactants, including C-reactive protein (CRP). Elevated CRP levels were as well documented also during infection-free intervals in all four patients suggesting chronic systemic inflammation. Noteworthy, among the documented laboratory findings, P1 and P3 showed elevation of liver enzymes, especially of gamma-glutamyl transferase (gGT). The etiology of elevated liver enzymes remained unclear despite standard work-up for viral hepatitis and autoimmune hepatopathies.

Immunological investigations

Immunological investigations in the pediatric patient (P4) revealed a normal distribution of T-cell subsets. At the age of 6 years, B-cell analysis showed a mild expansion of transitional B cells and CD21^low B cells, along with a slightly reduced class-switched IgG⁺ memory B-cell compartment, consistent with a mild B-cell maturation defect. In the adult patients (P1 to P3), immunophenotyping of peripheral lymphocytes was performed uniformly in the same laboratory, following the identical protocol also for parallel-tested age-matched healthy blood donors (HD), who served as the reference group in the absence of established age-related values (Fig. 2). In the context of lymphopenia (Fig. 2A and B), T cells, particularly both CD4⁺ and CD8⁺ T cells displayed higher levels of programmed death-1 (PD-1) and HLA-DR. These findings are indicative of immune exhaustion and T cell activation, respectively (Fig. 2C and L) [23]. The percentages of the rest of studied T cell subsets, including T follicular helper cells, naive and memory CD4⁺ T cells, exhibited no significant differences between the patient and HD groups. Considering the persistently elevated acute phase reactants, increased expression of HLA-DR and PD-1 may be the consequence of active systemic inflammation. With respect to B cells, patients with RNU4ATAC-related disease exhibited reduced total counts (Fig. 2M and R). Compared to HD, studied patients showed a decrease in the percentage of class-switched memory cells, while transitional B cells were elevated. The latter observation would be consistent with previous reports evaluating B cell subsets in patients with RS and suggests an aberration in the post-transitional B cell differentiation [13, 24]. In case of NK cells subsets, similar to a previous report [9], we observed a skewing towards CD56^bright cells with significant elevation of the CD56^brightCD16⁻ subset (Fig. 2S and W). Given the reduced cytotoxic potential of this NK cell subset, their elevated counts could be relevant for the susceptibility of studied patients to viral infections. On the other hand, within NK cells, the CD56^brightCD16⁻ subset has a rather immunoregulatory function [25] and their elevation in RNU4ATAC-related disease could play a compensatory role in the context of the above-mentioned systemic inflammation.

Fig. 2.

Immunophenotyping of peripheral lymphocytes from three patients with RNU4ATAC-related spliceosomopathy. Lymphocyte counts (A & B), T cell (C-L), B cell (M-R) and NK cell subset counts (S-W). Graphs summarize findings from different healthy blood donors (HD) or patients with RNU4ATAC-related spliceosomopathy and indicate the mean ± SEM (differences in lymphocyte counts were evaluated with the Mann-Whitney test; *P < 0.05; **P < 0.01)

Systematic review of published cases with RNU4ATAC-related spliceosomopathies

A total of 98 patients with RNU4ATAC-opathies were analyzed, including 94 previously published cases that were re-evaluated (Suppl. Table 1) [7–10, 13, 16–17, 26–46] and the four novel cases described above. A fatal outcome was reported in 34 patients (34.7%) (Fig. 3A). The cause of death was specified in only 13 cases and was commonly infection-related, including gastroenteritis (5/13), respiratory tract infection (2/13) and encephalitis (2/13). The majority of reported patients were pediatric cases (Table 2). Only 16/98 (16.3%) were adults, with P1 (43.2 years) being the oldest reported case. The most frequently reported features included growth disorders, microcephaly, craniofacial and digital dysmorphism, brain malformations, skeletal and cutaneous abnormalities, each observed in more than half of the previously described patients (Table 2). More than half of the patients were diagnosed with MOPD1/TALS (55/98, 56.1%), 29/98 (29.6%) with RS and only 5/98 (5.1%) with LWS, while in nine cases (9/98, 9.2%), no clinical diagnosis was reported (Fig. 3B). Identified RNU4ATAC variants spanned across all regions of U4atac, except for the distal part of the 3’ stem-loop (Fig. 3C and D). 81/98 patients (82.7%) had at least one RNU4ATAC variant in the 5’ stem-loop, while 50/98 (51%) harbored two variants located in the 5’ stem-loop, either homozygous (46/50) or compound heterozygous (4/50). The second most prevalent subgroup of patients with respect to genotype, displayed compound heterozygous variants, with one variant at the 5′ stem-loop and the second at the stem II (22/98, 22.4%).

Fig. 3.

Overview of characteristics and RNU4ATAC variants in 98 individuals with RNU4ATAC-opathies. Vital status (A), clinical diagnosis (B) as well as localization of deleterious RNU4ATAC variants (C). Further, schematic representation of U4atac and summary of reported variants across its regions. (adapted from references 9 & 38, U4atac region definition was based on references 4 & 6) (D). The here reported novel variant, identified in P3 is highlighted in green. (LWS, Lowry-Wood syndrome; MOPD1, microcephalic osteodysplastic primordial dwarfism type 1; N, total number; RS, Roifman syndrome; TALS, Taybi-Linder syndrome)

Table 2.

Characteristics of 98 patients with RNU4ATAC-related spliceosomopathies

Reported as deceased, no. (%)	34 (34.7)
Male sex, no. (%)	44 (44.9)
Median age, years (IQR)	4 (1.15–15.5)
Clinical and/or immunological evidence of immunodeficiency, no. (%)	44 (44.9)
Hypo-/dysgammaglobulinemia, no. (%)	28 (28.6)
Low B cells, no. (%)	22 (22.4)
Severe/recurrent infections, no. (%)	39 (39.8)
URTI, no. (%)	24 (24.5)
LRTI, no. (%)	22 (22.4)
CNS infections, no. (%)	5 (5.1)
GI infections, no. (%)	3 (3.1)
Prenatal growth delay, no. (%)	87 (88.8)
Postnatal growth delay, no. (%)	73 (74.5)
Cranial dysmorphism, no. (%)	82 (83.7)
Microcephaly, no. (%)	80 (81.6)
Trigonocephaly, no. (%)	2 (2)
Facial dysmorphism, no. (%)	68 (69.4)
Limb dysmorphism, no. (%)	22 (22.4)
Digit dysmorphism, no. (%)	50 (51)
Brachydactyly, no. (%)	36 (36.7)
Clinodactyly, no. (%)	16 (16.3)
Syndactyly, no. (%)	5 (5.1)
Retinal dystrophy, no. (%)	24 (24.5)
Strabismus, no. (%)	6 (6.1)
Cataract, no. (%)	4 (4.1)
Optic nerve atrophy, no. (%)	2 (2)
Brain malformation, no. (%)	53 (54.1)
Gyration abnormalities, no. (%)	22 (22.4)
Dysgenesis of corpus callosum, no. (%)	34 (34.7)
Cerebellar hypoplasia, no. (%)	16 (16.3)
Vermian hypoplasia, no. (%)	9 (9.2)
Intracranial cysts, no. (%)	11 (11.2)
Sensineuronal hearing loss, no. (%)	7 (7.1)
Skeletal abnormality/ies, no. (%)	81 (82.7)
Epiphyseal-/spondyloepiphyseal dysplasia, no. (%)	32 (32.7)
Joint contracture/dislocation, no. (%)	20 (20.4)
Scoliosis/kyphosis, no. (%)	11 (11.2)
Agenesis of 12th ribs, no. (%)	5 (5.1)
Skin abnormalities, no. (%)	50 (51)
Dry skin, no. (%)	30 (30.6)
Eczema, no. (%)	11 (11.2)
Chilblain-like, no. (%)	3 (3.1)
Hypoplastic/dysplastic nails, no. (%)	8 (8.2)
Cardiac septal defect, no. (%)	19 (19.4)
Cardiomyopathy, no (%)	1 (1)
Endocrinological features, no. (%)	17 (17.3)
Microdontia or enamel defects, no. (%)	11 (11.2)
Autoimmune/autoinflammatory features, no. (%)	7 (7.1)
Hepato-/splenomegaly, no. (%)	6 (6.1)
Hepatopathy, no. (%)	5 (5.1)
Renal disease, no. (%)	5 (5.1)
Asthma, no. (%)	5 (5.1)
Cryptorchidism, no (%)*	8 (18.2)
Micropenis, no (%)*	6 (13.6)

GI, gastrointestinal; CNS, central nervous system; IQR, interquartile Range; LRTI, lower respiratory tract infections; no., number; URTI, upper respiratory tract infections

*Values represent percentages relative to the male patient subgroup

Except for six cases, MOPD1 was diagnosed in patients carrying variants exclusively within the 5′ stem-loop (Table 3, Fig. 4). Of these six patients, two harbored a variant located at the 5’ stem-loop, while the second variant was a structural-disrupting insertion or deletion that would affect the 5’ stem-loop (suppl. Table 1; n.40 C > T; n.16_100dup). Furthermore, three of these six patients, harbored the n.111G > A variant with either a 5’ stem-loop-located variant (n.30G > A; n.111G > A) or a deletion (n.108_126del; n.111G > A). The strong association of MOPD1 and lethal outcomes (Fig. 5) with variants located in the biologically critical 5′ stem-loop region suggests that more severe RNU4ATAC-related phenotypes result from a more profound impairment of U4atac function. Additionally, variants that are exclusively located within the 5’ stem-loop are strongly associated with early childhood mortality, craniofacial dysmorphism and brain malformations, i.e. typical clinical features of MOPD1 (Table 3).

Table 3.

Frequency of reported clinical features in patients with RNU4ATAC variants exclusively inside versus also outside the 5’ stem-loop

	Variants at 5’ stem-loop exclusively (N = 50)	Variant outside the 5’ stem-loop (N = 48)	OR	95% CI	p value1 (summary)
Reported as deceased, no. (%)	33 (66)	1 (2.08)	91.24	13.84–949.1	< 0.0001 (****)
Median age, years (IQR)	1.04 (0.68–2.46)	12.88 (3.87-21)	n.a.	n.a.	< 0.0001 (****)
Not specified RNU4ATAC-related disorder, no. (%)	1 (2)	8 (16.67)	0.1	0.01–0.74	0.0305 (*)
RS, no. (%)	0 (0)	29 (60.42)	0	0-0.05	< 0.0001 (****)
LWS, no. (%)	0 (0)	5 (10.42)	0	0-0.63	0.0596 (ns)
MOPD1/TALS, no. (%)	49 (98)	6 (12.5)	343	47.4–3516	< 0.0001 (****)
Clinical and/or immunological evidence of immunodeficiency, no. (%)	8 (16)	36 (75)	0.06	0.02–0.18	< 0.0001 (****)
Hypo-/dysgammaglobulinemia, no. (%)	1 (2)	27 (56.25)	0.02	0-0.1	< 0.0001 (****)
Low B cells, no. (%)	0 (0)	22 (45.83)	0	0-0.1	< 0.0001 (****)
Severe/recurrent infections, no. (%)	7 (14)	32 (66.67)	0.08	0.03–0.23	< 0.0001 (****)
URTI, no. (%)	1 (2)	23 (47.92)	0.02	0-0.14	< 0.0001 (****)
LRTI, no. (%)	2 (4)	20 (41.67)	0.06	0.01–0.24	< 0.0001 (****)
GII, no. (%)	3 (6)	0 (0)	+inf.	0.85-+inf.	0.2555 (ns)
Encephalitis, no. (%)	4 (8)	0 (0)	+inf.	0.96-+inf.	0.1362 (ns)
Meningitis, no. (%)	0 (0)	1 (2.08)	0	0-8.64	0.9836 (ns)
Viral infections, no. (%)	2 (4)	10 (20.83)	0.16	0.03–0.65	0.0255 (*)
Fungal infections, no. (%)	0 (0)	3 (6.25)	0	0-1.09	0.2267 (ns)
Prenatal growth delay, no. (%)	50 (100)	37 (77.08)	+inf.	4.09-+inf.	0.0011 (**)
Postnatal growth delay, no. (%)	49 (98)	24 (50)	49	7.56–515.2	< 0.0001 (****)
Cranial dysmorphism, no. (%)	50 (100)	32 (66.7)	+inf.	5.99-+inf.	< 0.0001 (****)
Microcephaly, no. (%)	49 (98)	31 (64.58)	26.87	4-287	< 0.0001 (****)
Facial dysmorphism, no. (%)	33 (66)	1 (2.08)	91.24	13.84–949.1	< 0.0001 (****)
Limb dysmorphism, no. (%)	12 (24)	10 (20.83)	1.2	0.46–3.02	0.8939 (ns)
Digit dysmorphism, no. (%)	17 (34)	33 (68.75)	0.23	0.11–0.56	0.0012 (**)
Brachydactyly, no. (%)	9 (18)	27 (56.25)	0.17	0.07–0.42	0.0002 (***)
Clinodactyly, no. (%)	2 (4)	14 (29.17)	0.1	0.02–0.44	0.002 (**)
Syndactyly, no. (%)	2 (4)	3 (6.25)	0.63	0.11–3.19	0.9626 (ns)
Retinal dystrophy, no. (%)	4 (8)	20 (41.67)	0.12	0.04–0.38	0.0003 (***)
Strabismus, no. (%)	0 (0)	6 (12.5)	0	0-0.66	0.0309 (*)
Cataract, no. (%)	1 (2)	3 (6.25)	0.31	0.02–2.14	0.5807 (ns)
Optic nerve atrophy, no. (%)	0 (0)	2 (4.17)	0	0-2.06	0.457 (ns)
Brain malformation, no. (%)	37 (74)	16 (33.33)	5.69	2.27–13.34	0.0001 (****)
Gyration abnormalities, no. (%)	19 (38)	3 (6.25)	9.19	2.52–30.77	0.0004 (***)
Dysgenesis of corpus callosum, no. (%)	26 (52)	8 (16.67)	5.42	2.13–13.84	0.0005 (***)
Cerebellar hypoplasia, no. (%)	9 (18)	7 (14.58)	1.29	0.44–3.46	0.8539 (ns)
Vermian hypoplasia, no. (%)	5 (10)	4 (8.33)	1.22	0.33–4.19	0.9488 (ns)
Intracranial cysts, no. (%)	8 (16)	3 (6.25)	2.86	0.78–10.39	0.2269 (ns)
Sensineuronal hearing loss, no. (%)	1 (2)	6 (12.5)	0.14	0.01–0.96	0.1041 (ns)
Skeletal abnormality/ies, no. (%)	43 (86)	38 (79.17)	1.62	0.59–4.24	0.5312 (ns)
Epiphyseal-/spondyloepiphyseal dysplasia, no. (%)	3 (6)	29 (60.42)	0.04	0.01–0.15	0.0001 (****)
Joint contracture/dislocation, no. (%)	16 (32)	4 (8.33)	5.18	1.58–15.07	0.0079 (**)
Scoliosis/kyphosis, no. (%)	3 (6)	8 (16.67)	0.32	0.09–1.17	0.1763 (ns)
Agenesis of 12th ribs, no. (%)	2 (4)	3 (6.25)	0.63	0.11–3.19	0.9626 (ns)
Skin abnormalities, no. (%)	27 (54)	23 (47.92)	1.28	0.56–2.74	0.6891 (ns)
Dry skin, no. (%)	17 (34)	13 (27.08)	1.39	0.58–3.13	0.6007 (ns)
Eczema, no. (%)	1 (2)	10 (20.83)	0.08	0.01–0.51	0.0085 (**)
Chilblain-like, no. (%)	1 (2)	2 (4.17)	0.47	0.03–4.17	0.9714 (ns)
Hypoplastic/dysplastic nails, no. (%)	6 (12)	2 (4.17)	3.14	0.73–15.77	0.2952 (ns)
Cardiac septal defect, no. (%)	9 (18)	10 (20.83)	0.83	0.32–2.12	0.9211 (ns)
Endocrinological features, no. (%)	1 (2)	16 (33.33)	0.04	0-0.28	0.0001 (***)
Microdontia or enamel defects, no. (%)	5 (10)	6 (12.5)	0.78	0.24–2.98	0.9427 (ns)
Autoimmune/autoinflammatory features, no. (%)	0 (0)	7 (14.58)	0	0-0.5	0.016 (*)
Hepato-/splenomegaly, no. (%)	0 (0)	7 (14.58)	0	0-0.5	0.016 (*)
Hepatopathy, no. (%)	0 (0)	5 (10.42)	0	0-0.63	0.0596
Renal disease, no. (%)	2 (4)	3 (6.25)	0.63	0.11–3.19	0.9626

CI, confidence interval; GII, gastrointestinal infections; inf., infinity; LRTI, lower respiratory tract infections; LWS, Lowry-Wood syndrome; MOPD1, microcephalic osteodysplastic primordial dwarfism type 1; N, total number; n.a., not applicable; n.s. not significant; no., number; OR, odds ratio; RS, Roifman syndrome; TALS, Taybi-Linder syndrome; URTI, upper respiratory tract infections

¹p < 0.05 *; p < 0.01 **; p < 0.001 ***; p < 0.0001 ****; ns, non-significant

Fig. 4.

Correlation of genotype with clinical features in 98 patients with RNU4ATAC-related spliceosomopathy. Each panel shows the prevalence of documented manifestations, including vital status (A), clinical diagnosis (B), immunodeficiency (C), growth failure and syndromic features (D), ophthalmologic abnormalities (E), neurological findings (F), skeletal abnormalities (G), integumentary manifestations (H) and other features (I), across genotype-defined patient subgroups. (GII, gastrointestinal infections; LRTI, lower respiratory tract infections; LWS, Lowry-Wood syndrome; MOPD1, microcephalic osteodysplastic primordial dwarfism type 1; n, total number; RS, Roifman syndrome; TALS, Taybi-Linder syndrome; URTI, upper respiratory tract infections)

Fig. 5.

Kaplan-Meier survival curves comparing patients with RNU4ATAC variants located exclusively within the 5′ stem-loop versus those with at least one variant outside this region (probabilities of survival were compared with the Gehan-Breslow-Wilcoxon test; ****P < 0.0001)

Discussion

Here we report four new cases with RNU4ATAC-related disease, including three adult patients, two of whom displayed a relatively late disease onset. The genetic diagnosis in these latter two patients was particularly unexpected given the absence of profound dysmorphic features, skeletal anomalies or neurodevelopmental delay. This observation suggests that, despite its classification as an IEI with dysmorphic features [47], genetic testing for RNU4ATAC variants should also be considered in patients with immunodeficiency, even in the absence of dysmorphic or skeletal features that would be typical for MOPD1 or RS. Nephropathy and hepatopathy, which we observed in three patients, have only rarely been described in individuals with RNU4ATAC-related disease. To our knowledge, persistent systemic inflammation, which was present in all four cases reported here, has not been yet described. Together with the unusually late disease onset in two of our patients, these findings broaden the phenotypic spectrum of RNU4ATAC-related disease.

One of the patients described here (P3) exhibited a combination of features, including immunodeficiency, renal disease and skeletal dysplasia that led to the clinical diagnosis of immuno-osseous dysplasia. However, WES failed to identify the causative variants since it focuses on protein-coding sequences. The diagnosis was ultimately achieved through WGS, highlighting its additive diagnostic value for IEIs and syndromic disorders involving non-coding RNA genes. Routine WGS and consideration of RNU4ATAC even in adult patients without typical features of MOPD1 or RS, would be expected to increase the number of patients with RNU4ATAC-related spliceosomopathies, enabling a more objective evaluation of the phenotypic spectrum of RNU4ATAC-related disease.

The re-evaluation of all published cases with RNU4ATAC-opathies confirmed a strong genotype-phenotype correlation: homozygous or compound heterozygous variants within the 5′ stem-loop region are associated with severe disease characterized by extreme dysmorphic features, primordial dwarfism, brain anomalies, and early mortality (i.e., MOPD1). In contrast, patients with at least one variant outside the 5′ stem–loop typically present with milder disease, including variable growth retardation, developmental delay, spondyloepiphyseal dysplasia, retinal dystrophy, and immunodeficiency – features more characteristic of RS. Nevertheless, some patients diagnosed with MOPD1 have been found to carry only one variant in the critical 5′ stem-loop region. In most of these cases, the variant outside the 5′ stem-loop was a deletion or duplication, expected to affect the 5′ stem-loop as well. The aforementioned genotype-phenotype correlation along with the distinct severity and adverse outcome of patients with variants exclusively within the 5’ stem-loop, underscore the clinical relevance of specific syndromes such as MOPD1 or RS over a general molecular term for all forms of RNU4ATAC-related disease. Yet, milder or infection-only cases (such as P2) with a rather incomplete RS, suggest the need for defining novel RNU4ATAC-related syndromes (Fig. 6).

Fig. 6.

Perception of RNU4ATAC-related spliceosomopathies as a phenotypic continuum that should be expanded to encompass milder presentations and later-onset features

The relative absence of reported immunodeficiency or retinal dystrophy in patients with variants exclusively located at the 5′ stem-loop, who were diagnosed with MOPD1, likely does not reflect an alternative pathophysiology, but rather a bias toward reporting the more prominent or congenital features, such as the dysmorphic features. Further, the early mortality associated with MOPD1 may preclude the onset of clinically evident immunodeficiency or other features with a relatively delayed onset, such as autoimmunity, retinopathy, renal disease or bronchiectasis.

The here presented cases suggest CKD as a relatively common feature of RNU4ATAC-opathies. Further, renal histopathological findings related to deleterious RNU4ATAC variants have been reported previously. Among previously reported cases, specific renal manifestations included congenital anomalies of the kidney and urinary tract (CAKUT) such as cystic or dysplastic kidneys, renal tubular dysfunction, specifically renal tubular acidosis and infection-associated chronic kidney disease, particularly renal atrophy, potentially due to recurrent infections or chronic kidney disease [39, 48]. All three patients presented with proteinuria, whereas none exhibited renal tubular acidosis. The differential diagnosis of proteinuria in this context is broad. In P1, it included renal involvement of ANCA-associated vasculitis. In case of P3, renal histopathology did not correspond to any known glomerulopathy and did not meet the criteria for podocytopathy. Further characterization of the renal phenotype is required to better understand the mechanisms causing CKD in these patients and determine whether they follow a consistent pattern.

Although immunodeficiency is less well characterized in severe RNU4ATAC-related phenotypes (MOPD1 or LWS), reported infectious complications in MOPD1, including severe viral infections with fatal outcomes, suggest a profound immunodeficiency. This contrasts with most RS cases or those with at least one deleterious RNU4ATAC variant outside the 5′ stem-loop region, where reported immunological findings and infections are rather consistent with antibody deficiency. Commonly reported immunological abnormalities in RS include hypogammaglobulinemia or dysgammaglobulinemia and the loss of antibody responses to vaccinations. The latter findings commonly associated with reduced total B cell counts and expanded transitional B-cell populations, a finding replicated in all four patients reported here. However, severe infectious complications leading to bronchiectasis (P2 and P3) or the presence of confluent warts (P2) are rather suggestive of a broader immunodeficiency.

Persistent systemic inflammation was reported in all four patients. The latter, together with the finding of increased PD-1 expression on T cells is consistent with T cell exhaustion, which may contribute to both cellular and humoral immunodeficiency in RNU4ATAC-related disease. The lack of clinical evidence for infection at testing suggests inflammation and immune exhaustion in the context of a disease-intrinsic immune dysregulation that drives T cell hyperactivation. The finding of elevated HLA-DR expression on T cells in all tested patients would be also consistent with chronic activation.

The broad clinical spectrum of RNU4ATAC-related disorders highlights the need to screen for diverse manifestations, including CNS abnormalities, neurodevelopmental delay and cardiological, dermatological as well as nephrological complications, at diagnosis and during follow-up. Accordingly, patient management requires interdisciplinary care involving multiple specialties. As demonstrated by the cases presented here, nephrological, pulmonary or rheumatological manifestations may be the initial or predominant features. Consequently, raising physician awareness of RNU4ATAC-related disorders is essential for improving diagnosis. Beyond their rarity and low awareness among clinicians, the absence or subtlety of dysmorphic or skeletal features and the requirement for WGS to detect non-coding variants, represent additional obstacles to diagnosing RNU4ATAC-related disorders.

Supplementary Information

Below is the link to the electronic supplementary material.

Authors’ contributions

Conception and design of the work: AB, GS, SL, SvH. Data acquisition, analysis, interpretation: AB, GS, SL, SvH. Patient recruitment: AB, FR, GS, JHB, JK, JP, SL, SH, SvH. First manuscript draft: AB, GS, SL, SvH. Funding: GS, KNS, NDD, TW. All authors revised and approved the manuscript.

Funding

Open Access funding enabled and organized by Projekt DEAL. This project was funded by the Rosemarie-Germscheid Foundation. It was additionally supported by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany’s Excellence Strategy – EXC 2155 “RESIST” – Project ID 39087428.

Data availability

Data are available for formal research purposes only upon request to the corresponding author.

Declarations

Ethics approval and consent to participate

This study was conducted in accordance with the principles of the Declaration of Helsinki. Informed consent for genetic analyses was obtained within the framework of clinical care and included approval for research use. Ethical approval was obtained/exempted according to the requirements of the local institutional review board (approval 11223_BO_K_2024).

Consent for publication

Written informed consent for publication of clinical data was obtained from the patients/legal guardians. Additional consent was obtained for the publication of patient facial images.

Competing interests

The authors declare that they have no competing interests.