This is a research article published in Clinical & Molecular Biomedicine that focuses on the molecular consequences of missense mutations in the GUSB gene.

The work is framed around MPS VII, also called Sly syndrome, and combines variant collection, pathogenicity prediction, structural stability assessment, conservation analysis, and aggregation-propensity analysis.

1. Introduction1 𝛽-glucuronidase (GUSB) is a crucial enzyme widely used in 2 molecular biology and genetic engineering projects. (Yang et al.,3 2017). 3 The housekeeping enzyme GUSB plays an active role in 4 proteoglycan synthesis in lysosomes and is expressed in most5 tissues. It contributes significantly to the gradual decline in 6 dermatan and keratan sulfates by enhancing the disintegration7 of GAGs during the fifth cycle. GUSB promotes the dispersion of8 active or inert compounds from glucuronides, thereby modifying9 the manner in which prodrugs respond and operate (Naz et al.,10 2013).11 GUSB deficiency leads to mucopolysaccharidosis type VII,12 resulting in brain lysosomal storage (Kong et al., 2022). Humans13 have been shown to exhibit 11 distinct types (MPS I to IIIA, B,14 C, and IV to IX), which are divided according to the deficient15 enzyme (Hytönen et al., 2012). There are seven different types16 of Mucopolysaccharidosis (MPS), which are categorized based17 on defects in a certain number of the eight specific lysosomal18 metabolic enzymes (Khan et al., 2017). Mutations in the IDUA19 gene on chromosome 4p16 cause the first kind of MPS, which 20 is caused by a deficiency of the enzyme a-l-iduronidase. A-l- 21 iduronidase deficiency results in the accumulation of GAGs, such 22 as dermatan and heparan sulphates (DS and HS), in many human 23 tissues, leading to severe organ malfunction (Nagpal et al., 2022). 24 MPS II is linked to lysosomal dysfunction due to a deficiency 25 of the enzyme iduronate 2-sulphatase, which degrades heparan 26 sulphate and dermatan sulphate (DS). A rapidly developing 27 neurodegenerative lysosomal storage disease is MPSIII. It is 28 caused by biallelic variants in a gene encoding an enzyme in 29 the digestive tract that degrades heparan sulfate. Neuronal 30 inflammation and significant activation of astrocytes and 31 microglia are hallmarks of MPS III, a neurological disease 32 (Seker Yilmaz et al., 2021). MPS IVA, additionally referred 33 to as Morquio syndrome type A, constitutes one of the most 34 significant lysosomal illnesses. The lysosomal hydrolase N- 35 acetylglucosamine-6-sulfate sulfatase (GALNS) is absent in this 36 autosomal-recessive genetic disorder (Sawamoto et al., 2020). 37 There is nothing abnormal regarding the nervous system’s 38 functions, but common features include elevated blood and urine 39 Copyright: Distributed under Creative Commons CC-BY 4.0 OPEN ACCESS 1–8 Elucidating impacts of deleterious Missense mutations on the structure and function of 𝛽-glucuronidase: Investigating the molecular basis of pathogenesis of MPSVII potassium (K+) levels, a petite stature, odontoid hypoplasia,40 pectus carinatum, kyphoscoliosis, genu valgum, joint laxity, and41 corneal clouding (Tomatsu et al., 2011).42 Many lysosomal storage disorders, including43 mucopolysaccharidosis type VII (MPS VII), commonly referred44 to as Sly syndrome, are frequently linked to these mutations (?).45 The integrity and activity of 𝛽-glucuronidase are dramatically46 affected by missense, frameshift, and nonsense mutations in the47 GusB gene, according to recent investigations (?). GusB Gene48 encodes a 651-amino-acid-long homotetrameric protein with49 three distinctive domains in every monomer (Florindo et al.,50 2018). One such relationship involves the lysosomal-associated51 membrane protein 2 (LAMP2), which supports the stability and52 appropriate transit of 𝛽-glucuronidase inside lysosomes (Staudt53 et al., 2016). Furthermore, heparan sulfate and chondroitin54 sulfate are broken down by 𝛽-glucuronidase, which works55 in tandem with other enzymes such as arylsulfatase B and56 heparanase (Naz et al., 2013).57 According to Urayama et al. (2004), it is imperative that58 GUSB reach its intended intracellular location via this targeting59 mechanism. The lysosomal route of GAG breakdown is the60 main pathway in which GUSB is involved. An essential stage 61 in the breakdown of GAGs is the cleavage of the 𝛽-D-glucuronic62 acid residues by GUSB. GAG buildup is caused by a deficit in 𝛽-63 glucuronidase activity and results in lysosomal storage disorders64 (Hytönen et al., 2012). GAGs (glycosaminoglycans) are broken65 down by many lysosomal enzymes in an intricate procedure. The66 remaining molecules of 𝛽-D-glucuronic acid from GAGs such as67 heparan sulfate, dermatan sulfate, and chondroitin sulfate are68 particularly susceptible to hydrolysis by GUSB (Awolade et al.,69 2020).70 GUSB utilizes its N-terminus domain to identify and adhere 71 to a GAG chain, its substrate. The catalytic function of the 72 enzyme in question depends on the initial binding (Urayama73 et al., 2004). The site that catalyzes the breakdown of 𝛽-D-74 glucuronic acid residues is located in the catalytic region of GUSB75 (Hytönen et al., 2012). After splitting, the active region of the76 enzyme discharges the resulting molecules, which are shorter77 oligosaccharides, enabling 𝛽-Glucuronidase to interact with a 78 different substrate (Staudt et al., 2016). Supporting autophagy 79 and lysosomal activity requires 𝛽-glucuronidase. It supports the80 prevention of cellular malfunction by facilitating the breakdown81 of autophagic organelles. The amino acids associated with LAMP282 maintain its continued existence within lysosomes and ensure a83 steady supply of functional GUSB for the degradation of lysosomal 84 proteins (Staudt et al., 2016).85 GUSB is involved in the processes of extracellular matrix86 reorganization and cell-mediated cascade modulation, in addition87 to lysosomal breakdown, by controlling the accessibility of88 functional glycosaminoglycan fragments (Wan et al., 2020). By89 interacting with receptors located on the cell surface, like integrins90 and growth factor receptors, GAG fragments can encourage91 cytoskeleton remodeling and improve cell motility (Casale &92 Crane, 2024).93 The residues of breakdown may change the extracellular94 matrix’s mechanical features and the substance, which may95 affect how cells engage with their surroundings (Wang et al.,96 2017). Pattern recognition receptors (PRRs) on immune cell97 populations identify particular glycosaminoglycan components98 as danger-associated molecular patterns (DAMPs), which then,99 in turn, cause inflammatory reactions (Garantziotis & Savani,100 2022). About carcinoma, the breakdown components of101 glycosaminoglycan might affect the activity of the cancerous cells 102 and the stromal cells that envelop them, therefore facilitating the 103 growth, blood vessel development, and territorial expansion of 104 the tumor (Hua et al., 2022). Prospective treatment options for 105 recovering 𝛽-glucuronidase function in MPS VII patients include 106 gene therapy and enzyme replacement therapy (ERT) (Grubb 107 et al., 2010). Furthermore, the use of small-molecule chaperones 108 to regulate or increase the efficiency of mutant 𝛽-glucuronidase 109 has already been investigated (Doherty et al., 2023). 110 Single-nucleotide polymorphisms (SNPs), particularly non- 111 synonymous SNPs (nsSNPs), are major contributors to genetic 112 variability and are frequently implicated in inherited metabolic 113 disorders. Despite numerous reported variants, a systematic 114 approach to distinguish pathogenic mutations from benign 115 polymorphisms remains limited. Given the increasing volume of 116 genomic data generated through high-throughput sequencing, 117 there is a critical need to prioritize functionally significant 118 variants using reliable approaches. Identifying deleterious 119 nsSNPs and understanding their structural and functional 120 consequences can provide valuable insights into disease 121 mechanisms, genotype–phenotype correlations, and molecular 122 instability associated with protein dysfunction. Therefore, this 123 study was undertaken to comprehensively analyze nsSNPs in 124 the GUSB gene using integrative sequence- and structure-based 125 bioinformatics tools. By predicting pathogenicity, assessing 126 structural destabilization, and evaluating aggregation propensity, 127 the work aims to bridge the gap between genetic variation 128 data and mechanistic understanding. Ultimately, such insights 129 may facilitate improved diagnostic interpretation, biomarker 130 identification, and the development of targeted therapeutic 131 interventions.

2.1. Data retrieval and analysis 134 The GUSB sequence was extracted in FASTA format from the 135 UniProt repository (UniProt ID: P08236). A collection of non- 136 synonymous SNPs was generated using information obtained 137 from the PubMed literature search and databases such as dbSNP 138 (Sherry et al., 2001), HGMD (Stenson et al., 2009), ClinVar 139 (Landrum et al., 2014), and Ensembl (Hubbard et al., 2002). The 140 list was cleared to eliminate redundant nsSNPs. The protein data 141 bank provided the crystal structure of the human GUSB gene. Out 142 of the total Data, 1305 Missense variants along with 3’UTR and 5’ 143 UTR were obtained from dbSNP and Ensembl, and are shown in 144 Figure 1. We used a comprehensive computational approach to 145 anticipate harmful mutations in the GUSB at both structural and 146 functional levels, as illustrated in Figure 2. 147

2.2. Sorting Intolerant from Tolerant 148 Sorting Intolerant from Tolerant (SIFT) was used to predict 149 how possible amino acid alterations may affect protein function. 150 SIFT is expanded to include predictions for frame-shifting indels. 151 Human genetic research has extensively used SIFT to assess amino 152 acid substitutions (e.g., cancer, Mendelian disorders, and viral 153 infections). Studies of human diseases and other study subjects 154 are not the only areas in which SIFT is useful. The consequences 155 of missense mutations on model animals such as rats, dogs, and 156 Arabidopsis, as well as agricultural plants, have been studied using 157 SIFT (Sim et al., 2012). If the SIFT score is less than or equal to 158 0.05, then the mutation is not tolerable (Ng & Henikoff, 2003). The 159 SIFT method was utilized to forecast the impact of these nsSNPs 160 on the protein.

2.3. Polymorphism Phenotyping v2 163 Polymorphism Phenotyping v2 (PolyPhen-2), a program and 164 Web server accessible, estimates the potential effects of amino 165 acid alterations on the stability and activity of human proteins. 166 SNPs are functionally annotated; coding SNPs are mapped to 167 genetranscripts;proteinsequenceannotationsareextracted;and 168 conservation profiles are constructed. The likelihood that the 169 missensemutationwouldcauseharmisthencalculatedusinga 170 combinationofallthesecharacteristics(Adzhubeietal.,2013). In additiontoestimatingthelikelihoodthatamutationisharmful171 172 when it is not, PolyPhen-2 also provides predictions of false- 173 positiveandtrue-positiverates. PolyPhen-2computesthenaive 174 Bayesposteriorprobabilitythataparticularmutationisharmful. 175 Thequalitativeassessmentofamutationdetermineswhetherit 176 is benign, perhaps harmful, or likely harmful (Adzhubei et al., 177 2010). 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 2.4. Mutation Assessor Mutation Assessor (http://mutationassessor.org/r3/) is a sequence-based tool that predicts the functional impact of an nsSNPsonaprotein. Thisserverpredictsthefunctionaleffectsof missensepolymorphismsandaminoacidmodificationsobserved in cancer-causing mutations in proteins. Depending on the affectedaminoacidsandthedegreeofevolutionarypreservation in protein homologs, the functional impact is evaluated (?). A sizablecollectionofpolymorphicanddisease-associated(OMIM) variants—60,000intotalhasbeenusedtoverifythetechnique. 2.5. Function Analysis through Hidden Markov Models FunctionAnalysisthroughHiddenMarkovModels(FATHMM) isamethodthathelpspredicthowbothcodingandnon-coding variationsinthehumangenomeaffectfunction. Todetermineif SNPsandothermutationstendtobedamaging,hiddenMarkov models are employed. When it comes to genetic research and personalizedtherapy,FATHMMishighlybeneficial,asithelpsto clarifythesignificanceofgeneticvariants(Shihabetal.,2013). 195 2.6. Mutation Cutoff Scanning Matrix 196 A computer method called Mutation Cutoff Scanning Matrix 197 (mCSM) is used to predict how mutations may affect small- 198 moleculebinding,protein-proteininteractions,protein-nucleic 199 acidinteractions, andproteinstability. mCSMisaninnovative 200 tool for assessing non-synonymous mutations that predicts 201 destabilizing mutations using a graph-based method. If a 202 mutation results in a mCSM score (∆∆𝐺) less than 0, it affects 203 proteinstructure(Choudhuryetal.,2021). Itassistsinforecasting 204 theimpactofmutationsondrugbinding,hencecontributingto 205 thedevelopmentofmorepotentmedications. Ithelpsunderstand 206 how changes in protein interactions and stabilization lead to 207 conditionscausedbymutations. Directsthecreationofproteins 208 foruseinindustryandmedicinethathavetheappropriatestability 209 andinteractioncharacteristics(Piresetal.,2016). 210 2.7. MAESTRO 211 Thefreeenergychangeonproteinunfoldingisestimatedusing212 MAESTRO(https://pbwww.che.sbg.ac.at/maestro/web),amulti- 213 agentstabilitypredictiontool. Itdetermineshowapointmutation214 affectstheprotein’sstabilitybycalculatingthedifferenceinfree 215 energy (∆𝐺) between the mutant and wild-type proteins. The 216 stability of the protein is changed by a mutation if its score is217 less than zero (Laimer et al., 2015). Mutations may be limited 218 to certain amino acid classes, exposed or buried residues, and 219 user-specifiedareas,contingentuponthepurposeoftheresearch 220 (Laimeretal.,2015). 221 2.8. PREMPS 222 PremPScalculateschangesintheunfoldingGibbsfreeenergyto 223 assess the effects of single mutations on protein stability. This224 approachrequiresaprotein’sthree-dimensionalstructure(Chen 225 et al., 2020). The PREMPS software determines the mutation’s 226 correspondingchangeinfreeenergy(∆∆𝐺). PREMPSforecasts 227 the impact of the mutation on protein stability based on the 228 calculated∆∆𝐺. Generally,adestabilizingmutationisindicated 229 byahigh∆∆𝐺value,whereasastabilizingmutationissuggested 230 byanegative∆∆𝐺(Chenetal.,2020). 231 2.9. DynaMut 232 DynaMutforecaststheeffectsofmutationsonproteinstabilityby 233 consideringMDsimulationoutcomes. Theproteinstructureis 234 stabilizedordestabilizedbyamutationbasedonthecalculation 235 ofmetricssuchaschangesinfreeenergy(∆∆𝐺). Thisprediction 236 is essential for evaluating how mutations in biological systems 237 will affect functionality (Rodrigues et al., 2018). PhD-SNP is a 238 tooldesignedtohelpunderstandthegeneticbasisofillnesses. It 239 analyzesSNPsbyassessingtheirpossiblenegativeconsequences 240 usingavarietyofstructuralandevolutionaryfactors. Aprediction241 scorethatindicatestheprobabilityofanSNPbeingdeleteriousis242 producedbyPhD-SNP.Agreaterscoredenotesahigherlikelihood 243 thattheSNPwouldresultinfunctionalalterationsorincreased 244 susceptibilitytoillness,withascorerangefrom0to1(Capriotti 245 etal.,2006). 246 2.10. SNP & GO 247 A bioinformatics technique called SNP & GO (http://snps-a 248 nd-go.biocomp.unibo.it/snps-and-go/) integrates data from 249 both Gene Ontology (GO) and SNP annotations to forecast the 250 functionaleffectsofSNPs. Thistoolfacilitatesunderstandingof 251 genomicdatainrelationtodiseaseandphenotypebyilluminating 252 thepotentialeffectsofgeneticvariantsonproteinfunctionand 253 3 Elucidating impacts of deleterious Missense mutations on the structure and function of 𝛽-glucuronidase: Investigating the molecular basis of pathogenesis of MPSVII 254 biological mechanisms. The algorithms used here categorize 255 SNPsasneutralordeleterious(likelytoimpairproteinfunction) 256 basedontheattributesthatwereretrieved. SNP&GOproduces 257 predictionscoresdeterminedbythemachinelearningmodelthat 258 showhowlikelyitisthatanSNPwillhavenegativeeffects. Higher 259 scoresindicateagreaterlikelihoodoffunctionalimpact,which 260 helpsresearcherschooseSNPsforfurtherstudy(Capriottietal., 261 2013). 262 2.11. MutPred2 263 MutPred2 was created to predict how missense mutations in 264 humanproteinsmightaffectproteinfunction. Itevaluateshow 265 mutationsinfluenceproteinstructure,function,andinteractions 266 by combining diverse features and machine learning methods, 267 deliveringsignificantinsightsintotheirpotentialrolesindisease 268 processes(Choudhuryetal.,2021). 269 2.12. Solubility based on Disorder and Aggregation 270 A technique to determine the aggregation, disorder, helix, and strandpropensitiesresultingfrommutationsiscalledSolubility271 272 based on Disorder and Aggregation (SODA) (http://protein. 273 bio.unipd.it/soda/). The PDB format structure file or the protein sequence may be entered into this program. SODA274 275 uses Fells, ESpritz-NMR, PASTA 2.0, and other resources to 276 forecastmutationsofmanytypes,includinginsertions,deletions, 277 substitutions, and duplications. Based on how differently the 278 WTandmutantproteinsaresoluble,SODAassignsafinalscore 279 (Paladinetal.,2017). 280 2.13. Consurf Analysis 281 Consurf (https://consurf.tau.ac.il/) is a bioinformatics tool 282 designed to find areas in protein sequences that have been 283 conservedacrosstime. Itusesevolutionarylinksamongvarious 284 species and sequence homology to predict the structural and 285 functional significance of amino acid residues. Using multiple 286 sequencealignment,theConSurftoolwasusedtoassessresidue 287 conservationataparticularlocation(Ashkenazyetal.,2016).

The results section combines large-scale variant collection, sequence-based filtering, structure-based destabilization analysis, disease-phenotype prediction, conserved-residue mapping, SODA solubility analysis, and focused structural comparison of specific pathogenic substitutions.

The article emphasizes a stringent multi-tool strategy to identify variants that are consistently deleterious across both functional and structural predictors.

The paper concludes that integrative computational analysis can prioritize disease-relevant GUSB variants by combining sequence-level deleteriousness prediction, structure-based destabilization analysis, pathogenicity scoring, conservation mapping, and solubility analysis.

It proposes that the findings could support future therapeutic strategies such as pharmacological chaperones, solubility-enhancing agents, or other precision approaches aimed at restoring mutant β-glucuronidase function.

The article closes with declarations regarding conflicts of interest, funding, data availability, and use of AI tools.