A novel Bi-Allelic pathogenic MCOLN1 variant underlying mucolipidosis type IV in an Iranian family: clinical, genetic, and molecular dynamics-based structural analysis Authors: Mohaddese Mohsenipour, Parham Nejati, Teymoor Khosravi, Elham Alimoradi, Mohammad Salehi, Morteza Oladnabi, Reza Alibakhshi Published: 2025 ABSTRACT BackgroundMucolipidosis type IV (MLIV) is a rare autosomal recessive lysosomal storage disorder due to biallelic pathogenic variants in the MCOLN1 gene. Its main impact is on the central nervous system, leading to severe psychomotor delays, progressive visual impairment, and characteristic brain abnormalities.MethodsA 12-year-old male from a consanguineous Iranian family underwent clinical and imaging evaluations for suspected MLIV. Exome sequencing identified the causative variant, confirmed by Sanger co-segregation analysis, in silico tools assessed pathogenicity, protein stability, and structural impact, followed by 3D modeling (I-TASSER) and protein interaction analysis (STRING). Molecular dynamics simulations were performed with GROMACS 2020.4 employing the GROMOS96 43a1 force field to compare wild-type and mutant structures, evaluating key parameters, including root mean square deviation (RMSD), radius of gyration (Rg), hydrogen bond profiles, and solvent-accessible surface area (SASA), were analyzed, and results which were visualized using GraphPad Prism.ResultsExome sequencing revealed a previously unreported homozygous nonsense variant in MCOLN1 (NM_020533.3: c.1384G > T; p.Glu462). This variant introduces a premature termination codon predicted to yield a truncated protein if translated; however, it is likely subject to nonsense-mediated mRNA decay, leading to transcript degradation and consequent loss of functional protein. Sanger sequencing confirmed the variant and its co-segregation within the family, with both parents heterozygous carriers and the patient homozygous. Bioinformatic analysis classified the variant as likely pathogenic, with high deleteriousness scores. Structural modeling indicated disruption of a helical domain. STRING analysis demonstrated strong functional associations between MCOLN1 and its paralogs MCOLN2 and MCOLN3, supporting its biological relevance. This variant expands the known spectrum of genetic causes of MLIV.ConclusionWe report the first Iranian case of MLIV due to a novel homozygous nonsense variant in MCOLN1 (c.1384G > T; p.Glu462*). These findings expand the spectrum of MLIV, underscore phenotypic variability and the value of population-specific genetic data in rare disease diagnostics, and support the inclusion of this variant in targeted diagnostic panels for Iranian patients.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12920-025-02271-9. Introduction Mucolipidosis type IV (MLIV; OMIM: 252650 [1]) is a rare autosomal recessive condition that belongs to the spectrum of lysosomal storage diseases. It is recognized as a neurodegenerative condition that predominantly impacts the central nervous system (CNS) by disrupting the maturation of neuronal networks, leading to hallmark features such as neurodevelopmental delay and progressive visual impairment [2]. The classic phenotype of MLIV, caused by a deficiency of functional TRPML1 protein, is marked by hypomyelinating leukodystrophy accompanied by abnormal iron deposition within the brain. Clinically, this presents with severe psychomotor developmental delays followed by progressive neurological deterioration, degeneration of the cerebellum, and damage to neuronal axons [3, 4]. The earliest ocular manifestations—bilateral corneal clouding and strabismus—typically appear within the first year of life. These are followed by retinal dystrophy, which rapidly progresses during adolescence and frequently culminates in complete vision loss by early adulthood [5]. Spasticity, diffuse hyperreflexia, and reduced voluntary muscle strength usually emerge within the first five years and progressively worsen. Achlorhydria, along with hypergastrinemia and anemia due to iron deficiency, is also frequently reported [6]. Characteristic magnetic resonance imaging (MRI) features observed in children often involve abnormalities such as underdevelopment or significant thinning of the corpus callosum, diminished subcortical white matter associated with dysmyelination, presence of variable T2 hyperintense lesions, abnormal iron (ferritin) accumulation or decreased signal intensity within the basal ganglia, cerebellar volume loss, alongside relative preservation of the cortical gray matter [7, 8]. The gene responsible for MLIV, MCOLN1 (OMIM: 605248 [9]), is situated on chromosome 19p13.2–13.3. Spanning 14 exons, it contains an open reading frame (ORF) of 1740 nucleotides and codes for mucolipin-1 (TRPML1), a protein consisting of 580 amino acids [10]. Various variant types have been reported in affected individuals. MLIV is more prevalent among individuals of Ashkenazi Jewish descent, particularly those with North Polish–South Lithuanian ancestry, with an estimated prevalence of 1 in 40,000 and a carrier frequency of 1 in 100 [11]. A defective TRPML1 protein is the underlying cause of MLIV. TRPML1 is an intracellular ion channel localized to late endosomes and lysosomes and belongs to the mucolipin subfamily within the Transient Receptor Potential (TRP) channel family [12]. Unlike other mucolipidosis subtypes, which result from metabolic enzyme deficiencies or dysfunctions, MLIV arises from defects in the sorting and trafficking of macromolecules within the late endocytic pathway [13]. Studies of membrane marker trafficking have shown that MLIV cells exhibit excessive transport from late endosomes to lysosomes, with a pronounced delay in retrograde trafficking to the Golgi apparatus compared to control cells [14]. Despite regular lysosomal degradation activity, variability has been observed in the composition of the accumulated materials [15]. Elevated intralysosomal pH in MLIV cells further supports a disruption in the function or organization of the late endosome–lysosome system [16]. Clinical diagnosis of MLIV is challenging due to significant phenotypic heterogeneity, the absence of disease-specific neurological findings, and considerable overlap with other neurogenetic disorders. Thus, the identification of MCOLN1 variants is essential for accurate diagnosis. In this study, we describe a previously unreported variant in the MCOLN1 gene discovered by exome sequencing in an Iranian patient affected by MLIV. This investigation aimed to clarify the genetic cause of the disorder within a consanguineous family involving first-cousin parents. Materials and methods Clinical evaluation of the affected individual This study involved a 12-year-old boy born to consanguineous parents (first cousins). The patient presented with clinical symptoms that prompted a referral to a medical geneticist and a specialist physician for further evaluation. A detailed medical and family history was collected from the patient and his parents to exclude non-genetic causes. Additional clinical assessments, including physical examination, laboratory tests, and brain MRI, were performed. Exome sequencing was conducted on the patient’s genomic DNA to determine the genetic basis of the disorder. Informed consent for participation and genetic testing was provided by the patient’s parents, as legal guardians, prior to enrollment in the study. Extraction and assessment of genomic DNA quality Peripheral blood samples (5 mL) were drawn from the patient and both parents. Genomic DNA was isolated from 1 µL of whole blood using the AddBio DNA Extraction Kit (catalog number: 10023). The quality and quantity of the isolated DNA were evaluated using a NanoDrop spectrophotometer (Thermo Scientific, USA). DNA quality and integrity were further evaluated by electrophoresis on a 1% agarose gel. Exome sequencing and variant detection Exome sequencing was carried out on the patient’s DNA utilizing the Illumina NovaSeq 6000 platform, producing 150-base pair paired-end reads. The sequencing targeted an average coverage depth of 100×, ensuring that 90–95% of the bases were covered at a minimum depth of 20×. For library preparation, genomic DNA was fragmented, labeled with unique barcodes, and subjected to hybridization using the Agilent SureSelect Human All Exon V7 Plus probe set. Following sequencing, paired-end reads were mapped to the NCBI human reference genome (GRCh37/hg19) utilizing the Burrows-Wheeler Aligner (BWA) [17]. Identification of variants, encompassing single-nucleotide variants (SNVs) and small insertions or deletions, was conducted using the Genome Analysis Toolkit (GATK) [18]. Variant prioritization was performed using a comprehensive bioinformatics pipeline. Variants with a minor allele frequency (MAF) ≤ 1% were excluded based on population databases, including gnomAD, 1000 Genomes, and ExAC. Functional annotation was conducted with ANNOVAR, and pathogenicity was evaluated using in silico tools (MutationTaster, CADD, REVEL, SIFT, PROVEAN, and Dynamut). Conservation metrics and gene-level constraint scores were also assessed. Priority was given to predicted loss-of-function (LoF) variants such as nonsense, frameshift, or canonical splice-site changes, in genes with established phenotypic relevance. A novel MCOLN1 nonsense variant was identified, absent from all major population databases and not reported in the homozygous state. Given its predicted LoF impact and the known pathogenic mechanism of MCOLN1, the variant was classified as likely pathogenic according to ACMG/AMP guidelines, supported by automated interpretation tools (InterVar and VarSome). Co-segregation analysis for variant validation A co-segregation analysis was conducted to assess intergenerational transmission and to evaluate the potential pathogenicity of the identified MCOLN1 variant within the affected family. Co-segregation of the variant with the disease phenotype supports its potential role in disease causation. To validate the variant identified through exome sequencing, conventional Sanger sequencing was performed. Prior to sequencing, the genomic region harboring the variant was amplified by polymerase chain reaction (PCR) using a primer pair specifically designed to target the variant locus within the MCOLN1 gene: forward primer MCOLN1-F, 5′- TAGCCATTTGCATGGGAC − 3′ (Tm: 58.4 °C, 18 bp, GC content: 50.0%), and reverse primer MCOLN1-R, 5′- AGGAGTAAAGGTAGAGCTG − 3′ (Tm: 56.3 °C, 19 bp, GC content: 47.4%). After amplification, Sanger sequencing was conducted on the proband and both parents to verify the variant’s presence and assess its zygosity. Sequence chromatograms were analyzed using CodonCode Aligner software to verify the segregation pattern. In Silico analysis of MCOLN1 variants The pathogenicity of the identified variant, along with reported MCOLN1 variants retrieved from the ACMG database and literature, was evaluated using a range of in silico bioinformatic tools, including CADD [19], MutationTaster [20], PolyPhen-2 [21], I-Mutant [22], MUpro [23], and FATHMM [24]. Three-dimensional structural modeling of the mutant protein I-TASSER [25] (Iterative Threading ASSEmbly Refinement) is a commonly utilized computational tool that facilitates the prediction of protein structure and function based on amino acid sequences. The approach constructs full-length atomic models by integrating diverse sequence alignment strategies and iterative structural modeling, followed by atomic-level refinement to improve model accuracy. Functional annotation is achieved by comparing the predicted 3D models to known protein structures and functions in curated databases, utilizing both sequence- and structure-based alignments. In this study, I-TASSER was employed to generate 3D structural models of both the wild-type and mutant forms of the MCOLN1 protein. The resulting models were subsequently visualized and analyzed using UCSF Chimera [26], a software platform designed for interactive molecular graphics and structural analysis. Network analysis of protein-protein interactions STRING [27] (Search Tool for the Retrieval of Interacting Genes/Proteins) is a pre-computed database that enables the investigation and evaluation of established and predicted protein–protein interactions (PPIs). These interactions are categorized into three main types: physical (direct), functional (indirect), and computationally predicted interactions. Data supporting these interactions are derived from multiple sources, including genomic context, high-throughput experimental results, co-expression analyses, and text mining. To assess the reliability of each interaction, STRING implements a confidence scoring system ranging from 0 to 1, with higher scores indicating stronger supporting evidence. This system enables users to evaluate and prioritize predicted interactions based on the robustness of the underlying data. In this study, PPIs analysis for MCOLN1 was performed using the STRING database to identify potential functional partners and gain insights into its biological role. Molecular dynamics simulation analysis Molecular dynamics (MD) simulations were executed using GROMACS v.2020.4, implementing the GROMOS96 43a1 force field to evaluate the impact of the MCOLN1:c.1384G > T (p.Glu462*) mutation on the conformational stability of wild-type (WT) and mutant TRPML1 protein structures. Each system was enclosed in a triclinic computational box and solvated with the simple point charge (SPC) water model. To replicate physiological ionic strength, systems were balanced with 0.15 M NaCl. Initial energy minimization was achieved using the steepest descent method for a maximum of 5,000 iterations to alleviate steric conflicts and refine the starting configuration. Equilibration proceeded in two sequential phases. The first phase, thermal equilibration, was performed under the NVT ensemble (constant particle number, volume, and temperature), stabilizing the system at 298 K with a V-rescale thermostats, permitting temperature variations between 150 K and 300 K. The second phase, pressure equilibration, utilized the NPT ensemble, maintaining a pressure of 1.0 bar via the Parrinello-Rahman barostat. Following equilibration, a 50-nanosecond production run was executed employing the leap-frog integrator with a 2-femtosecond time step, yielding approximately 1,000 trajectory frames for subsequent analysis. Structural dynamics of WT and mutant TRPML1 proteins were assessed using GROMACS-integrated tools. Key structural metrics, including root mean square deviation (RMSD), radius of gyration (Rg), intramolecular hydrogen bond counts, and solvent-accessible surface area (SASA), were calculated to evaluate protein behavior over the 10–50 ns simulation period. Statistical comparisons between WT and mutant proteins for these metrics were conducted using the nonparametric Wilcoxon signed-rank test in GraphPad Prism 9, with a significance threshold of p < 0.05. Data visualizations and statistical analyses were generated using GraphPad Prism 9. Result Clinical findings A 12-year-old male patient of Kurdish ethnicity from Kermanshah, Iran, born to first-cousin consanguineous parents, was the subject of this study (Fig. 1a). His clinical presentation included physical and motor impairments, intellectual disability (ID), strabismus, corneal opacity, and speech difficulties. Fig. 1Pedigree of a consanguineous Iranian family showing a male patient (IV-2) affected by MLIV due to a novel homozygous MCOLN1 mutation (a). Cranial MRI findings in a patient with MLIV. MRI reveals cerebral and cerebellar atrophy, bilateral ventriculomegaly, and periventricular white matter volume loss (b) Pedigree of a consanguineous Iranian family showing a male patient (IV-2) affected by MLIV due to a novel homozygous MCOLN1 mutation (a). Cranial MRI findings in a patient with MLIV. MRI reveals cerebral and cerebellar atrophy, bilateral ventriculomegaly, and periventricular white matter volume loss (b) According to his medical history, the pacormal height and weight. However, by 21 months of age, he was unable to sit independently and could speak only a few words. At the time of clinical evaluation, his head circumference measured 46 cm, and his weight was 10.7 kg. Laboratory investigations revealed elevated total cholesterol at 308 mg/dL (reference range: 150–220 mg/dL) and mildly elevated liver enzymes—SGOT (AST) at 65 U/L and SGPT (ALT) at 54 U/L (reference range: 5–40 U/L for both). Serum iron was 68 µg/dL (reference range: 50–170 µg/dL), within normal limits, while ferritin was elevated at 233 ng/mL (reference range: 12–150 ng/mL) and total iron-binding capacity (TIBC) was decreased at 221 µg/dL (reference range: 250–400 µg/dL). The ceruloplasmin at 23 mg/dL (normal ≥ 15 mg/dL) was within normal limits. Serological testing revealed a positive Rubella IgG level of 15 IU/mL (reference > 10 IU/mL), consistent with prior exposure or immunity. CMV IgG was elevated at 50 IU/mL (positive threshold > 14 IU/mL), indicating previous cytomegalovirus infection or immune status. An abdominal ultrasound demonstrated a normal-sized liver (73 mm) with homogeneous parenchymal echotexture and no cystic or space-occupying lesions. The gallbladder was normal in size and wall thickness without evidence of calculi or intraluminal masses. Intrahepatic and extrahepatic bile ducts were of normal caliber, and portal vein anatomy and flow were unremarkable. His cranial MRI revealed widened subarachnoid spaces and prominent cortical sulci, most notably in the frontal and peritrigonal regions, consistent with diffuse cerebral atrophy. Additionally, bilateral ventriculomegaly was observed. The periventricular white matter exhibited reduced volume and decreased signal intensity. The cerebellar hemispheres and vermis also demonstrated diminished volume, indicative of cerebellar atrophy (Fig. 1b). The patient’s CT scan revealed ground-glass opacities in the paracardiac region, suggestive of pneumonic changes. This finding may be secondary to aspiration due to the patient’s swallowing difficulties and persistent drooling, which are rare manifestations of MLIV. The patient has been under medical supervision since early childhood. Over time, he developed a broad spectrum of neurological and physical symptoms, including delayed motor and language development, pyramidal and extrapyramidal dysfunction, reduced voluntary muscle activation, achilles tendon, soleus muscle shortening, tremor, ataxia, jerky nystagmus, myopathic facial appearance with gradual reduction of facial movements, persistent drooling, and difficulties with chewing and swallowing. Genetic findings and variant identification To investigate the underlying cause of the affected individual’s clinical manifestations, exome sequencing was performed. The analysis revealed a novel nonsense variant in the MCOLN1 gene (NM_020533.3: c.1384G > T; p.Glu462*), which is associated with MLIV. This single-nucleotide transversion (G > T) at position 1384 results in the substitution of glutamic acid (Glu) at codon 462 with a premature termination codon (Ter), which, if translated, would lead to early truncation of the protein (Fig. 2). The resulting truncated protein is predicted to be non-functional. Based on the ACMG criteria, the c.1384G > T variant is classified as likely pathogenic. Fig. 2Graphical representation of the c.1384G > T mutation in the MCOLN1 gene, illustrating its genomic location from the chromosomal to nucleotide level and its resulting impact on the protein sequence Graphical representation of the c.1384G > T mutation in the MCOLN1 gene, illustrating its genomic location from the chromosomal to nucleotide level and its resulting impact on the protein sequence Co-segregation analysis of the identified variant Co-segregation analysis was conducted to confirm the mode of inheritance, provide supporting evidence for pathogenicity, and validate the exome sequencing results. Sanger sequencing was conducted on blood samples from the patient and his parents. The analysis confirmed that the c.1384G > T variant co-segregated with the disease phenotype: both parents were heterozygous carriers, while the affected patient was homozygous for the variant (Fig. 3). Fig. 3Sanger sequencing chromatogram of proband (IV-2 mutant) and his parents (III-6 and III-7 carrier) Sanger sequencing chromatogram of proband (IV-2 mutant) and his parents (III-6 and III-7 carrier) In Silico analysis of pathogenicity, protein stability, and evolutionary conservation The c.1384G > T variant detected in this investigation is previously unreported and absent from established databases. Its potential pathogenicity and impact on the protein’s structure and function were assessed using a range of in silico prediction tools. In total, 67 MCOLN1 variants have been reported in the Human Gene Mutation Database (HGMD), including 34 missense/nonsense variants, 11 splice-site variants, 10 small deletions, 5 small insertions, 4 gross deletions, 1 gross insertion, and 2 complex rearrangements. Of these, 44 variants are publicly accessible and are listed in Supplementary Material 1. Table 1 summarizes the results of ACMG classification, CADD, MutationTaster, PolyPhen-2, FATHMM, I-Mutant, and MUpro analyses for the novel c.1384G > T variant, as well as for previously reported missense and nonsense variants retrieved from HGMD and the literature. Fig. 4a illustrates the spectrum of MCOLN1 variants retrieved from the ClinVar database, most of which are classified as benign or likely benign copy number variants (CNVs). Table 1An overview of miss/nonsense mutations reported in the MCOLN1 geneVariantProtein effectType of VariantExon/IntronPhenotypeACMG predCADD scoreMutation taster predFATHM predI-Mutant predMupro predPOLY PHEN 2 HVAR predStudyc.1384G > Tp.Glu462*nonsenseE12MLIVLikely Pathogenic0.269disease causingN/AN/AN/AN/APresent studyc.195G > Cp.Lys65ArgmissenseE2Intellectual DeficiencyVUS7.103disease causingTOLERATEDIncrease StabilityDECREASE stabilityprobably damagingCheng et al. (2020)c.235 C > Tp.Gln79*nonsenseE2MLIVLikely Pathogenic5.066disease causingN/AN/AN/AN/ABach et al. (2005)c.304 C > Tp.Arg102*nonsenseE3MLIVPathogenic0.194disease causingN/AN/AN/AN/ASun et al. (2000)c.317T > Cp.Leu106PromissenseE3MLIVVUS1.756disease causingTOLERATEDDecrease StabilityDECREASE stabilityprobably damagingAltrarescu et al. (2002)c.362 C > Tp.Thr121MetmissenseE3Oromandibular Dystonia and ParkinsonismLikely Pathogenic4.135benignTOLERATEDIncrease StabilityINCREASE stabilityprobably damagingGhasemi et al. (2024)c.378 C > Gp.Tyr126*nonsenseE3Intellectual disabilityPathogenic0.716disease causingN/AN/AN/AN/AStojanovic et al. (2019)c.410T > Cp.Leu137PromissenseE4MLIVVUS1.527disease causingTOLERATEDDecrease StabilityDECREASE stabilityprobably damagingArai-Ichinoi et al. (2016)c.497G > Tp.Cys166PhemissenseE4MLIVVUS1.784disease causingTOLERATEDDecrease StabilityDECREASE stabilityprobably damagingWakabayashi et al. (2011)c.514 C > Tp.Arg172*nonsenseE4MLIVPathogenic0.953disease causingN/AN/AN/AN/ASun et al. (2000)c.551T > Cp.Ile184ThrmissenseE4Generalized dystonia, ataxia and bradykinesiaVUS2.962disease causingTOLERATEDDecrease StabilityDECREASE stabilityprobably damagingManzoor et al. (2018)c.576 C > Ap.Cys192*nonsenseE4Corneal DystrophyPathogenic0.986disease causingN/AN/AN/AN/APatterson et al. (2023)c.694 A > Cp.Thr232PromissenseE6MLIVPathogenic4.761disease causingTOLERATEDDecrease StabilityDECREASE stabilityprobably damagingBargal et al. (2001)c.776T > Cp. Leu259PromissenseE6Corneal DystrophyVUS4.696disease causingTOLERATEDDecrease StabilityDECREASE stabilityprobably damagingPatterson et al. (2023)c.782 C > Tp.Thr261MetmissenseE6Developmental DelayBenign0.054disease causingDAMAGINGDecrease StabilityDECREASE stabilitybenignMeloche et al. (2019)c.832 C > Tp.Gln278*nonsenseE7Cerebellar ataxia, hyperreflexia, external ophthalmoparesis, bilateral corneal clouding, and abnormal behaviourPathogenic1.269disease causingN/AN/AN/AN/AKremer et al. (2017)c.871 C > Tp.Gln291*nonsenseE7Corneal DystrophyPathogenic3.868disease causingN/AN/AN/AN/APatterson et al. (2023)c.964 C > Tp.Arg322*nonsenseE8MLIVPathogenic0.647disease causingN/AN/AN/AN/ABargal et al. (2000)c.1009 C > Tp.Gln337*nonsenseE9Corneal DystrophyLikely Pathogenic3.599disease causingN/AN/AN/AN/APatterson et al. (2023)c.1084G > Tp.Asp362TyrmissenseE9MLIVPathogenic0.412disease causingDAMAGINGDecrease StabilityDECREASE stabilityprobably damagingSun et al. (2000)c.1061G > Ap.Try354*nonsenseE9Intellectual DeficiencyLikely Pathogenic1.795disease causingN/AN/AN/AN/ACheng et al. (2020)c.1207 C > Tp.Arg403CysmissenseE10MLIVPathogenic2.076disease causingTOLERATEDDecrease StabilityDECREASE stabilityprobably damagingGoldin et al. (2004)c.1292G > Ap.Cys431TyrmissenseE11MLIVVUS1.02disease causingTOLERATEDDecrease StabilityDECREASE stabilityprobably damagingSaijo et al. (2016)c.1307 A > Gp.Tyr436CysmissenseE11MLIVLikely Pathogenic0.594disease causingTOLERATEDDecrease StabilityDECREASE stabilityprobably damagingAlbakheet et al. (2013)c.1331G > Ap.Trp444*nonsenseE11Corneal DystrophyLikely Pathogenic1.137disease causingN/AN/AN/AN/APatterson et al. (2023)c.1336G > Tp.Val446LeumissenseE11MLIVLikely Pathogenic0.878disease causingTOLERATEDDecrease StabilityDECREASE stabilitypossibly damagingSun et al. (2000)c.1340T > Cp.Leu447PromissenseE11MLIVVUS4.987disease causingTOLERATEDDecrease StabilityDECREASE stabilityprobably damagingAltrarescu et al. (2002)c.1367 C > Tp.Ser456LeumissenseE12MLIVVUS3.445disease causingTOLERATEDDecrease StabilityINCREASE stabilityprobably damagingTüysüz et al. (2009)c.1388G > Ap.Cys463TyrmissenseE12MLIVVUS2.639disease causingTOLERATEDDecrease StabilityDECREASE stabilityprobably damagingWakabayashi et al. (2011)c.1395 C > Gp.Phe465LeumissenseE12MLIVVUS1.543disease causingTOLERATEDDecrease StabilityDECREASE stabilityprobably damagingBargal et al. (2001) Table 1An overview of miss/nonsense mutations reported in the MCOLN1 geneVariantProtein effectType of VariantExon/IntronPhenotypeACMG predCADD scoreMutation taster predFATHM predI-Mutant predMupro predPOLY PHEN 2 HVAR predStudyc.1384G > Tp.Glu462*nonsenseE12MLIVLikely Pathogenic0.269disease causingN/AN/AN/AN/APresent studyc.195G > Cp.Lys65ArgmissenseE2Intellectual DeficiencyVUS7.103disease causingTOLERATEDIncrease StabilityDECREASE stabilityprobably damagingCheng et al. (2020)c.235 C > Tp.Gln79*nonsenseE2MLIVLikely Pathogenic5.066disease causingN/AN/AN/AN/ABach et al. (2005)c.304 C > Tp.Arg102*nonsenseE3MLIVPathogenic0.194disease causingN/AN/AN/AN/ASun et al. (2000)c.317T > Cp.Leu106PromissenseE3MLIVVUS1.756disease causingTOLERATEDDecrease StabilityDECREASE stabilityprobably damagingAltrarescu et al. (2002)c.362 C > Tp.Thr121MetmissenseE3Oromandibular Dystonia and ParkinsonismLikely Pathogenic4.135benignTOLERATEDIncrease StabilityINCREASE stabilityprobably damagingGhasemi et al. (2024)c.378 C > Gp.Tyr126*nonsenseE3Intellectual disabilityPathogenic0.716disease causingN/AN/AN/AN/AStojanovic et al. (2019)c.410T > Cp.Leu137PromissenseE4MLIVVUS1.527disease causingTOLERATEDDecrease StabilityDECREASE stabilityprobably damagingArai-Ichinoi et al. (2016)c.497G > Tp.Cys166PhemissenseE4MLIVVUS1.784disease causingTOLERATEDDecrease StabilityDECREASE stabilityprobably damagingWakabayashi et al. (2011)c.514 C > Tp.Arg172*nonsenseE4MLIVPathogenic0.953disease causingN/AN/AN/AN/ASun et al. (2000)c.551T > Cp.Ile184ThrmissenseE4Generalized dystonia, ataxia and bradykinesiaVUS2.962disease causingTOLERATEDDecrease StabilityDECREASE stabilityprobably damagingManzoor et al. (2018)c.576 C > Ap.Cys192*nonsenseE4Corneal DystrophyPathogenic0.986disease causingN/AN/AN/AN/APatterson et al. (2023)c.694 A > Cp.Thr232PromissenseE6MLIVPathogenic4.761disease causingTOLERATEDDecrease StabilityDECREASE stabilityprobably damagingBargal et al. (2001)c.776T > Cp. Leu259PromissenseE6Corneal DystrophyVUS4.696disease causingTOLERATEDDecrease StabilityDECREASE stabilityprobably damagingPatterson et al. (2023)c.782 C > Tp.Thr261MetmissenseE6Developmental DelayBenign0.054disease causingDAMAGINGDecrease StabilityDECREASE stabilitybenignMeloche et al. (2019)c.832 C > Tp.Gln278*nonsenseE7Cerebellar ataxia, hyperreflexia, external ophthalmoparesis, bilateral corneal clouding, and abnormal behaviourPathogenic1.269disease causingN/AN/AN/AN/AKremer et al. (2017)c.871 C > Tp.Gln291*nonsenseE7Corneal DystrophyPathogenic3.868disease causingN/AN/AN/AN/APatterson et al. (2023)c.964 C > Tp.Arg322*nonsenseE8MLIVPathogenic0.647disease causingN/AN/AN/AN/ABargal et al. (2000)c.1009 C > Tp.Gln337*nonsenseE9Corneal DystrophyLikely Pathogenic3.599disease causingN/AN/AN/AN/APatterson et al. (2023)c.1084G > Tp.Asp362TyrmissenseE9MLIVPathogenic0.412disease causingDAMAGINGDecrease StabilityDECREASE stabilityprobably damagingSun et al. (2000)c.1061G > Ap.Try354*nonsenseE9Intellectual DeficiencyLikely Pathogenic1.795disease causingN/AN/AN/AN/ACheng et al. (2020)c.1207 C > Tp.Arg403CysmissenseE10MLIVPathogenic2.076disease causingTOLERATEDDecrease StabilityDECREASE stabilityprobably damagingGoldin et al. (2004)c.1292G > Ap.Cys431TyrmissenseE11MLIVVUS1.02disease causingTOLERATEDDecrease StabilityDECREASE stabilityprobably damagingSaijo et al. (2016)c.1307 A > Gp.Tyr436CysmissenseE11MLIVLikely Pathogenic0.594disease causingTOLERATEDDecrease StabilityDECREASE stabilityprobably damagingAlbakheet et al. (2013)c.1331G > Ap.Trp444*nonsenseE11Corneal DystrophyLikely Pathogenic1.137disease causingN/AN/AN/AN/APatterson et al. (2023)c.1336G > Tp.Val446LeumissenseE11MLIVLikely Pathogenic0.878disease causingTOLERATEDDecrease StabilityDECREASE stabilitypossibly damagingSun et al. (2000)c.1340T > Cp.Leu447PromissenseE11MLIVVUS4.987disease causingTOLERATEDDecrease StabilityDECREASE stabilityprobably damagingAltrarescu et al. (2002)c.1367 C > Tp.Ser456LeumissenseE12MLIVVUS3.445disease causingTOLERATEDDecrease StabilityINCREASE stabilityprobably damagingTüysüz et al. (2009)c.1388G > Ap.Cys463TyrmissenseE12MLIVVUS2.639disease causingTOLERATEDDecrease StabilityDECREASE stabilityprobably damagingWakabayashi et al. (2011)c.1395 C > Gp.Phe465LeumissenseE12MLIVVUS1.543disease causingTOLERATEDDecrease StabilityDECREASE stabilityprobably damagingBargal et al. (2001) An overview of miss/nonsense mutations reported in the MCOLN1 gene Fig. 4A graphical overview of MCOLN1 variants categorized by type and clinical significance, as reported in the ClinVar database (a) and a visual summary of MCOLN1 variant mutations derived from the HGMB database (b) A graphical overview of MCOLN1 variants categorized by type and clinical significance, as reported in the ClinVar database (a) and a visual summary of MCOLN1 variant mutations derived from the HGMB database (b) The ConSurf tool was used to evaluate the evolutionary conservation of amino acid residues within the MCOLN1 protein, aiming to identify structurally and functionally important regions by integrating conservation data with solvent accessibility. The deleted region of the mutant MCOLN1 protein (p.Glu462*) encompasses a substantial portion of the polycystin cation channel domain (residues 385–524) and includes the selectivity filter motif (residues 469–474). ConSurf analysis revealed that 29.66% of the deleted amino acids have a conservation score of 9 (on a scale of 1 to 9, with 9 indicating the highest conservation), 16.10% have a score of 8, and 5.93% have a score of 7. These highly conserved residues span the entire polycystin cation channel domain and the selectivity filter motif, underscoring their functional importance (Fig. 5). Fig. 5The evolutionary conservation of the deleted region in the MCOLN1 was analyzed through ConSurf. a Amino acid sequence of the deleted region, with conservation scores indicated for each residue. b 3D structure of the MCOLN1 protein colored according to residue conservation scores. c Isolated deleted region colored according to conservation scores The evolutionary conservation of the deleted region in the MCOLN1 was analyzed through ConSurf. a Amino acid sequence of the deleted region, with conservation scores indicated for each residue. b 3D structure of the MCOLN1 protein colored according to residue conservation scores. c Isolated deleted region colored according to conservation scores Predicted three-dimensional structures of MCOLN1 I-TASSER was employed to model the three-dimensional (3D) structures of both the wild-type and mutant forms of the MCOLN1 protein. The analysis indicated that the affected residue, glutamic acid at position 462 (p.Glu462*), is situated within a helical domain of the protein. This prediction was supported by a confidence score of 9, reflecting high model reliability. Fig. 6 presents the predicted tertiary structures of the wild-type and mutant MCOLN1 proteins. Fig. 6The three-dimensional structures of the wild-type and mutant MCOLN1 proteins. a Wild-type model of MCOLN1 protein. b Mutant p.Glu462* model of MCOLN1 protein. c Wild-type 3D model of PKD_channel domain of MCOLN1 protein. d Mutant type (p.Glu462*) 3D model PKD_channel domain of MCOLN1 protein. e superimpose model of wild (blue) and mutant type and mutant types (purple) The three-dimensional structures of the wild-type and mutant MCOLN1 proteins. a Wild-type model of MCOLN1 protein. b Mutant p.Glu462* model of MCOLN1 protein. c Wild-type 3D model of PKD_channel domain of MCOLN1 protein. d Mutant type (p.Glu462*) 3D model PKD_channel domain of MCOLN1 protein. e superimpose model of wild (blue) and mutant type and mutant types (purple) Protein interaction network involving MCOLN1 The STRING database compiles data from various sources to predict PPIs and infer functional relationships between genes. In this study, STRING was employed to construct the interaction network of the MCOLN1 protein. Fig. 7 displays the first-shell interactors of MCOLN1, which include MCOLN2, MCOLN3, TFEB, FRMD6, PNPLA6, PKD2, PDCD6, NEU3, TPCN2, and TRPM4. The second-shell interactors include TPCN1, TRPM2, TRPV2, TRPC1, TRPC3, TRPC5, TRPV5, TRPV6, TRPV3, and TRPV4. These proteins are predicted to have the strongest functional or physical associations with MCOLN1. Notably, MCOLN2 exhibited the highest confidence score (0.984), indicating robust interaction evidence (Table 2). The interactions between MCOLN1 and its paralogs, MCOLN2 and MCOLN3, are supported by multiple lines of evidence, such as shared genomic context, inferred gene fusions, patterns of simultaneous presence across species, correlated expression profiles, sequence similarity, and information extracted from published literature. Fig. 7Protein–protein interaction network of the MCOLN1 gene based on STRING database analysis, showing direct interactors as colored nodes and indirect (second-shell) interactors as white nodes Protein–protein interaction network of the MCOLN1 gene based on STRING database analysis, showing direct interactors as colored nodes and indirect (second-shell) interactors as white nodes Table 2The table shows the interaction strength between MCOLN1 and various genes, with scores ranging from 0 to 1GeneScoreMCOLN20.984MCOLN30.976TFEB0.860FRMD60.826PNPLA60.781PKD20.780PDCD60.777NEU30.772TPCN20.768TRPM40.760 Table 2The table shows the interaction strength between MCOLN1 and various genes, with scores ranging from 0 to 1GeneScoreMCOLN20.984MCOLN30.976TFEB0.860FRMD60.826PNPLA60.781PKD20.780PDCD60.777NEU30.772TPCN20.768TRPM40.760 The table shows the interaction strength between MCOLN1 and various genes, with scores ranging from 0 to 1 Molecular dynamics findings To evaluate the structural impact of the MCOLN1: c.1384G > T (p.Glu462*) nonsense variant, MD simulations were conducted to compare the WT TRPML1 protein with its truncated mutant counterpart over a 50 ns simulation period. Four key structural parameters, such as Rg, intramolecular hydrogen bonds, RMSD, and SASA, were analyzed in the stable 10–50 ns timeframe to assess differences in structural stability and compactness. Statistical significance was evaluated using the Wilcoxon signed-rank test in GraphPad Prism 9. Radius of gyration (Rg) Analysis of Rg (Fig. 8b) revealed an initial decrease in both WT and mutant proteins, indicative of structural relaxation during the early simulation stages. However, the mutant protein consistently exhibited a lower Rg throughout the 10–50 ns period, suggesting a more compact conformation. The median Rg of the mutant (difference: −0.0305 nm) was significantly lower than that of the WT (p < 0.0001, Wilcoxon signed-rank test), likely due to the loss of C-terminal domains contributing to the extended architecture of the native protein. Fig. 8Predicted 3D structures of the wild-type and mutant MCOLN1 proteins generated from MD simulations. a. Comparison of Rg between WT and mutant TRPML1 proteins over the 10–50 ns simulation period. The mutant exhibited significantly lower Rg (p < 0.0001, Wilcoxon signed-rank test). b. Comparison of intramolecular hydrogen bonds between WT and mutant TRPML1 proteins over the 10–50 ns simulation period. The mutant showed significantly fewer hydrogen bonds (p < 0.0001, Wilcoxon signed-rank test) (c). Comparison of RMSD between WT and mutant TRPML1 proteins over the 10–50 ns simulation period. The mutant exhibited significantly higher RMSD (p < 0.0001, Wilcoxon signed-rank test) (d). Comparison of SASA between WT and mutant TRPML1 proteins over the 10–50 ns simulation period. The mutant showed significantly lower SASA (p < 0.0001, Wilcoxon signed-rank test) (e) Predicted 3D structures of the wild-type and mutant MCOLN1 proteins generated from MD simulations. a. Comparison of Rg between WT and mutant TRPML1 proteins over the 10–50 ns simulation period. The mutant exhibited significantly lower Rg (p < 0.0001, Wilcoxon signed-rank test). b. Comparison of intramolecular hydrogen bonds between WT and mutant TRPML1 proteins over the 10–50 ns simulation period. The mutant showed significantly fewer hydrogen bonds (p < 0.0001, Wilcoxon signed-rank test) (c). Comparison of RMSD between WT and mutant TRPML1 proteins over the 10–50 ns simulation period. The mutant exhibited significantly higher RMSD (p < 0.0001, Wilcoxon signed-rank test) (d). Comparison of SASA between WT and mutant TRPML1 proteins over the 10–50 ns simulation period. The mutant showed significantly lower SASA (p < 0.0001, Wilcoxon signed-rank test) (e) Intramolecular hydrogen bonds The number of intramolecular hydrogen bonds (Fig. 8c) was significantly reduced in the mutant compared to the WT over the 10–50 ns period. The WT protein stabilized with approximately 420–450 hydrogen bonds, while the mutant maintained a lower range of 330–360 bonds (median difference: −81 bonds, p < 0.0001, Wilcoxon signed-rank test). This reduction reflects the loss of stabilizing interactions due to truncation of critical regions for tertiary structure maintenance. Root mean square deviation (RMSD) RMSD analysis (Fig. 8d) indicated greater conformational deviation in the mutant compared to the WT. Over the 10–50 ns period, the median RMSD of the mutant (1.45 nm) was significantly higher than that of the WT (1.20 nm) (median difference: 0.0813 nm, p < 0.0001, Wilcoxon signed-rank test), suggesting increased structural fluctuations and reduced stability in the truncated form. Solvent-accessible surface area (SASA) The SASA of the mutant protein (85–90 nm²) was consistently lower than that of the WT (100–110 nm²) over the 10–50 ns period (Fig. 8e). The median SASA difference (−21.35 nm²) was statistically significant (p < 0.0001, Wilcoxon signed-rank test), indicating a more globular conformation in the mutant, likely as a mechanism to shield hydrophobic residues, a common feature of misfolded or destabilized proteins. Collectively, these MD simulation results demonstrate that the p.Glu462* variant in MCOLN1 induces marked structural alterations in TRPML1, characterized by increased compactness (lower Rg and SASA), reduced hydrogen bonding, greater conformational flexibility (higher RMSD), and decreased solvent exposure. These changes, all statistically significant (p < 0.0001), suggest that the nonsense mutation compromises the structural integrity of TRPML1, likely contributing to its impaired function in disease pathology. Discussion Mucolipidosis type IV, a rare autosomal recessive lysosomal storage disease, results from pathogenic variants in the MCOLN1 gene, which codes for the transient receptor potential cation channel mucolipin-1 (TRPML1) [28]. Although MLIV occurs across all ethnicities, it has been most frequently reported in Ashkenazi Jewish populations, where two founder variants, c.416–2 A >G and c.1_788del, account for approximately 95% of cases [29]. In contrast, broader variant heterogeneity has been observed among other populations. For instance, the c.1367 C >T variant has been described in a Turkish patient, while a Japanese patient carried compound heterozygous variants (c.936_938del and c.1503dupC) [30, 31]. These findings have expanded our understanding of the MCOLN1 pathogenic variant spectrum and the clinical heterogeneity of MLIV. In this study, we report the first Iranian MLIV patient with a homozygous nonsense variant in MCOLN1, c.1384G >T (p.Glu462*), thereby contributing to the expanding global landscape of this gene. Several nonsense variants in MCOLN1 have been documented in diverse ethnic groups. Among these are c.304 C >T (p.Arg102Ter), c.378 C >G (p.Tyr126Ter), and c.514 C >T (p.Arg172Ter), identified in both Ashkenazi and non-Ashkenazi Jewish populations, all resulting in truncated TRPML1 proteins [10, 29, 32]. Within the Iranian population, the frequency of MLIV is unknown. However, according to the Iranome database [33], only intronic, non-coding exons, and splice site variants have been reported to date; no nonsense variants have been previously identified. One earlier Iranian case reported a novel missense variant (c.362 C >T, p.Thr121Met) in a consanguineous family, leading to a substitution of threonine with methionine and subsequent MCOLN1 dysfunction [34]. In contrast, our patient, also born to first-cousin consanguineous parents, carries a novel homozygous nonsense variant (c.1384G >T), marking the first such case in the Iranian population. This variant introduces a premature stop codon at position 462, which, if translated, would produce a truncated and likely nonfunctional TRPML1 protein (p.Glu462*). However, the transcript is predicted to be targeted by nonsense-mediated mRNA decay, resulting in reduced or absent production of the truncated protein. While MLIV is classically defined by hallmark clinical features, its phenotypic expression varies considerably between individuals, often depending on the specific MCOLN1 variant. For example, a 68-year-old Japanese patient carrying the c.1292G >A variant presented with atypical features such as late-onset clinical manifestations, jaw-opening dystonia, chronic kidney disease, and vacuolated lymphocytes but notably lacked seizures or anemia [35]. A Turkish patient with the c.1367 C >T variant exhibited distinct traits including facial dysmorphisms (e.g., micrognathia, high-arched palate, and gingival hypertrophy), bilateral clinodactyly, and bluish corneal discoloration [30]. Interestingly, in another case, a patient with the c.1503dupC variant exhibited no ophthalmological abnormalities, which are usually hallmark features of MLIV [31]. A review of the literature and the OMIM database indicates that rare phenotypic findings in MLIV include delayed-onset corneal opacity, absence of corneal opacity despite severe myopia, capsular lens opacities, mild neurological involvement with relatively preserved psychomotor development, microcephaly, dysplastic splenium with absent rostrum, subtle coarse facial features, gastric parietal cell dysfunction leading to hypergastrinemia, and partial ganglioside sialidase deficiency. Consistent with previous reports of phenotypic variability, our patient exhibited both classic manifestations, such as developmental delay and strabismus, and atypical features, including truncal involvement, tremor, progressive reduction of facial movements, persistent drooling, and shortening of the Achilles tendon and soleus muscle. These latter features have rarely been documented in MLIV and underscore the importance of comprehensive clinical evaluation, as some aspects of the phenotype may be underrecognized in the current literature. Laboratory investigations revealed hypercholesterolemia, likely due to impaired lysosomal lipid trafficking and reduced cholesterol clearance from late endosomes and lysosomes [36]. Mild elevations in SGOT and SGPT were consistent with low-grade hepatocellular injury, likely stemming from lysosomal accumulation in hepatocytes, although abdominal ultrasound showed normal liver morphology and echotexture [37]. Despite the fact that mucolipidosis is typically associated with impaired iron availability due to lysosomal dysfunction, our patient demonstrated a paradoxically normal serum iron level. Further laboratory evaluation revealed elevated ferritin and decreased total iron-binding capacity, a pattern consistent with functional iron deficiency rather than actual iron depletion. This profile likely reflects iron sequestration within reticuloendothelial cells and the influence of chronic inflammation, similar to what is observed in anemia of chronic disease. Therefore, normal serum iron levels in this context may be misleading and do not necessarily indicate sufficient bioavailable iron [38, 39]. Additionally, decreased MCV and MCH may indicate chronic illness, malnutrition, or gastrointestinal malabsorption. At the same time a reduced MPV may reflect impaired megakaryocyte maturation due to lysosomal dysfunction, although its direct association with MLIV remains unclear [40]. The age at symptom onset in MLIV is also variable. Most cases exhibit early-onset symptoms, typically between 1.5 months and 8 years (mean: 7.25 months) [8], although later-onset cases have also been reported. Ghasemi et al. noted a mean age at onset of 8.25 ± 2.86 years in a separate cohort. In our case, the patient showed developmental delay by 21 months, was unable to sit independently, and was limited to a vocabulary of only two words, consistent with early-onset MLIV. The presence of a truncating variant like c.1384G >T may correlate with more severe, early-manifesting forms of the disease. The clinical variability and age of onset in MLIV are often dictated by the type and position of MCOLN1 variants [6]. TRPML1 contains a polycystin cation channel domain (residues 385–524), including a crucial selectivity filter motif (residues 469–474). The c.1384G >T variant introduces a premature stop codon at residue 462, truncating the protein before the selectivity filter is formed. This truncation is expected to abolish ion selectivity and disrupt the function of the TRPML1 channel. Dysfunctional TRPML1 impairs macromolecule sorting in the late endocytic pathway, a key pathological mechanism in MLIV. In silico analyses supported the pathogenicity of the novel c.1384G >T variant. Evaluation using ACMG guidelines, CADD, MutationTaster, PolyPhen-2, FATHMM, I-Mutant, and MUpro consistently classified this variant as pathogenic. Comparative analysis with 34 previously reported missense and nonsense variants from the HGMD database and the literature further substantiated the variant’s deleterious nature. Conservation analysis using ConSurf and Clustal Omega analysis confirmed that the affected glutamic acid residue at position 462 is highly conserved across species. A maximal conservation score and its exposure on the protein surface indicate its functional importance, potentially in enzymatic activity or protein-protein interactions. The evolutionary pressure to preserve this residue underscores the detrimental impact of its loss. Our findings provide strong evidence that the c.1384G >T variant impairs MCOLN1 protein function and contributes to the MLIV phenotype. MD simulation analysis of the MCOLN1: p.Glu462* mutant revealed key structural perturbations consistent with a destabilizing effect on TRPML1. The observed reduction in both the Rg and SASA suggests increased protein compactness, likely resulting from the loss of C-terminal domains. A substantial decrease in the number of intramolecular hydrogen bonds, along with elevated RMSD values, further indicates compromised structural stability and enhanced conformational flexibility in the mutant. Together, these alterations support a pathogenic mechanism in which truncation disrupts proper protein folding and compromises structural integrity, potentially impairing TRPML1 function and contributing to the molecular pathology of MLIV. This study highlights the importance of identifying population-specific MCOLN1 variants to inform genetic counseling and diagnostic strategies, particularly in regions with high consanguinity. The c.1384G > T variant may warrant inclusion in targeted sequencing panels for Iranian patients with suspected lysosomal storage disorders. Furthermore, functional studies to validate the variant’s impact on TRPML1 channel activity could provide insights into therapeutic targets, given the current lack of disease-modifying treatments for MLIV. In conclusion, our report of a novel nonsense variant in an Iranian MLIV patient advances our understanding of the genetic and clinical spectrum of this rare disorder and it lays the groundwork for improved diagnosis and management in underrepresented populations. Conclusions In conclusion, we report a novel homozygous nonsense variant, c.1384G > T, in the MCOLN1 gene of an Iranian patient with MLIV. This case expands the known variant spectrum and clinical variability of MLIV and provides new insights into the molecular underpinnings of this disorder within underrepresented populations. Supplementary Information Supplementary Material 1. Supplementary Material 1.