Background
The protective protein/cathepsin A (PPCA or CTSA) is a multifunctional lysosomal enzyme with distinct protective and catalytic function [
1].
The mature form of PPCA/CTSA, consisting of a 32/20 kDa disulfide-linked two chain product, is found in a high molecular weight, lysosomal multienzyme complex (LMC) together with two other glycosidases, β-galactosidase (GLB1) and N-acetyl alpha neuraminidase 1 (NEU1) [
2‐
4]. A supposed role of N-acetylgalactosamine-sulfate sulfatase (GALNS) in such complex [
5] still lacks interpretation.
Association with PPCA/CTSA in an early biosynthetic compartment ensures the correct lysosomal transport, activation and stability of GLB1 and NEU1 [
6]. This defines the enzyme protective function. On the other hand, studies of the physiological role of PPCA/CTSA as a serine carboxypeptidase/deamidase/esterase demonstrated a role of the enzyme in the inactivation of selected neuropeptides, like substance P, oxytocin and endothelin I [
7]. PPCA/CTSA is also responsible for the proteolytic inactivation of Lysosome-associated membrane protein (LAMP)2a, a lysosomal integral membrane protein involved in chaperone mediated autophagy, thus regulating this lysosomal pathway of protein degradation [
8].
Mutations in the CTSA gene are the cause of the lysosomal storage disease galactosialidosis (GS). Loss of function of PPCA/CTSA results in the secondary combined deficiency of GLB1 and NEU1, which is the biochemical hallmark of the disease. Patients with GS present with a broad spectrum of clinical manifestations, but are usually classified as early infantile, late infantile or juvenile/adult type based on the age of onset and severity of their symptoms.
The early infantile phenotype, the most severe form of the disease, usually presents with hydrops fetalis, cherry red spots, visceromegaly, psychomotor delay, coarse facies, skeletal dysplasia, and early death. Late infantile forms are characterized by corneal clouding, cardiac involvement, visceromegaly and, rarely, psychomotor retardation. Most patients with the milder juvenile/adult form, exhibited myoclonus, ataxia, neurological deterioration, angiokeratoma, and absence of visceromegaly [
9,
10].
It is still unclear whether the catalytic function of PPCA/CTSA contributes to particular clinical signs of GS. In this respect, impaired LAMP2a degradation due to PPCA/CTSA deficiency may be linked to the low weight of affected individuals [
8] and the lack of inactivation of specific bioactive peptides may play a role in the regulation of the blood pressure [
7].
In addition, it has been previously suggested that PPCA/CTSA plays a role in elastic fibers assembly, through its association with the enzymatically inactive, spliced variant of β- galactosidase, known as the elastin binding protein (EBP). Because EBP acts as an intracellular chaperone for tropoelastin, facilitating the trafficking and deposition of elastic fibers [
11], lack of PPCA/CTSA in GS can be accompanied by alterations in elastogenesis, affecting the cardiovascular and respiratory systems [
7,
12,
13]. This is especially true for GS patients with a longer survival, as they need periodic assessment of their pulmonary function and emphysema, linked to a defect in elastic fiber assembly [
10].
Because PPCA/CTSA is present in the LMC, mutations altering the folding of one protein in the complex can influence the other components as well [
14,
15]. Multiple sequence alignments may predict functional sites or domains that may favor intra- or inter-molecular interactions within the LMC. Analogously, structural web applications may predict an amino acid substitution as disease-causing or neutral in humans, and it could suggest the molecular causes of a disease. For instance, gain of helical propensity or loss of a phosphorylation site or disorder to order transitions caused by Molecular Recognition Features (Morf), specific regions of proteins that exhibit molecular binding functions [
16].
A total of 23
CTSA gene mutations have been reported (HGMD professional
https://portal.biobase-international.com/cgi-bin/portal/login.cgi). These include deletions, missense and splicing mutations, but no nonsense mutations. Alternatively spliced transcripts of the
CTSA gene have been described, and at least three
CTSA mRNA sequences have been deposited in the GenBank database (RefSeq,
http://www.ncbi.nlm.nih.gov/gene/5476). In early reports mutations nomenclature was selected according to all
CTSA isoforms, which generated confusion in the way
CTSA mutations were reported. Some mutations were numbered based on a PPCA sequence that was either 46 (missing all the amino acids of the signal peptide) [
17‐
20] or 18 [
13,
21] amino acids shorter than that reported in the HGMD professional database.
In order to assist physicians in the interpretation of detected mutations, we review and discuss previously reported CTSA transcripts, underlying erroneous or current nomenclature. We also present the molecular and clinical assessment of four new observations of the rare infantile form of GS. Computational analyses to predict a role of the new and/or previously reported missense mutations are discussed in order to address the clinical and molecular implications of the CTSA defects in GS patients.
Discussion
GS is a rare lysosomal storage disease with most of the described patients having the juvenile/adult form [
9,
10]. The prevalence of GS is unknown; more than 100 observations have been reported but only 23 mutations have been identified so far, including point-mutations and rearrangements, but no nonsense mutations [
9,
10]. Here we report the first case of GS carrying a
CTSA gene nucleotide change leading a stop codon mutation. The rarity of such mutation type in the GS patient population is likely attributable to the rare incidence of the early infantile form of the disease.
We also describe two new missense mutations: c.347A>G (p.His116Arg), and c.775T>C (p.Cys259Arg). The p.His116Arg mutation was detected in combination with the c.114delG, a deletion found at the homozygous level also in Pt4 and previously reported in an unrelated patient originating from the same area of central Italy [
13]. Since this mutation was not reported in patients with other origin in the literature and due to the small number of previously reported mutations, a founder effect for such one-base deletion can be hypothesized.
As mentioned earlier, three different
CTSA cDNAs transcripts are deposited in the GenBank database, thus the past literature on
CTSA mutations referred randomly to the different isoforms, which has been the source of confusion. For example, the p.Val150Met mutation has been first reported as p.Val104Met [
22] and then reported as p.Val132Met [
23]. We would like to stress the importance of having the
CTSA mutation nomenclature homologated to the HGMD professional guide lines, which choose
CTSA Variant 1 as the reference sequence for mutation nomenclature of the CTSA resulting protein.
To deepen the correlation between mutations and their effects on the CTSA protein structure we performed computational analyses using in silico tools based on phylogenetic alignments and functional/structural predictions. Multiple alignments of related sequences among organisms and structural web applications help to identify regions or domains that are conserved and may indicate functional constraints.
We found computational analyses to be helpful in improving the determination of the pathognomonic effects of newly identified nucleotide variants and genotype-phenotype correlations. This method has been particularly useful in the case of compound heterozygous mutations reported in mild affected patients or when clinical data are insufficient, i.e. if expression studies and/or structural analysis of compound heterozygous mutations are not available. The reverse process: from a described phenotype to computational analysis showed a good correlation between in silico predictions and mutation severity.
We want to emphasize that predictive functional/structural analyses of mutant proteins and phenotypes are closely related, often providing clearcut indications on the specific form of the disease (early infantile, late infantile, juvenile). We found such analyses of use to predict the extent of disease severity related to two mutations p.Tyr267Asn and p.Phe458Val found in juvenile GS patients. In contrast, phylogenetic comparisons in most cases provide indications that are limited to the pathogenic/non-pathogenic effects of mutations without further details. Thus, the combination of multiple computational analyses is an effective strategy.
CTSA binds and regulates GLB1 and NEU1 inside lysosomes [
6]. Structure predictions, identifying mutations that alter Molecular Recognition Features (MoRFs) or result in gain or loss of function in CTSA, could accordingly provide information on the three-dimensional structure of the complex proteins. These findings might provide indications on the pathogenetic effects of mutations and on the interactions between the proteins within the LMC.
Known and predicted protein-protein interactions evaluated by String and GeneMania prediction software evidenced that PPCA/CTSA does not directly interact with GALNS thus corroborating the molecular data showing that the PPCA/CTSA deficit does not affect the GALNS activity in Pt3. However our findings can not exclude the involvement of GALNS in the LMC after the binding with NEU1.
The protective function of PPCA dramatically affects both GLB1 and NEU1, leading to a broad range of clinical manifestations, worsening with age [
2]. PPCA defects causes both glycosidases to malfunction, indeed early infantile forms of GS share clinical signs observed in the infantile forms of both GM1 gangliosidosis and type II sialidosis [
19]. In addition, altered catalytic activity of PPCA could contribute to the variability of symptoms due to the potential esterase/deamidase activity of PPCA in platelets, endothelial cells, heart and kidney [
3]. A cardiovascular role of PPCA could also be linked to the altered function of EBP due to structural mutations of PPCA as its binding partner at the plasma membrane [
11]. Our prediction tools identified only one CTSA amino acid change (p.Val150Met) that could affect its catalytic site. This mutation, detected in combination with the p.Gln406* change, is linked to the early infantile phenotype with fetal hydrops and cardiac involvement.
CTSA gene, spanning about 43.000 bp and containing 15 exons, is about twice the average length of human genes. However, reported CTSA mutations are a very low number. The reason for which a region or site have a higher or lower mutation rate is poorly understood, except in the case of CpG islands where cytosine can become methylated and unstable, leading to a higher rate of mutation [
26].
The majority of
CTSA genetic lesions occur at positions that are evolutionarily conserved [
27], and non-variable sites may indicate protein sequences under more selective constraints [
28]. Thus, the PPCA/CTSA paradigm could be useful both at genetic level, identifying base composition bias around CpG dinucleotides, distribution and rate of single nucleotide polymorphisms, and at functional level, giving structural consequences of mutated amino acids and regions.
Conclusions
In early reports mutations nomenclature was selected according to all CTSA isoforms (three different isoforms), thus generating a lot of confusion. In order to assist physicians in the interpretation of detected mutations, we underline the correct nomenclature for CTSA mutations.
Four cases with the rare infantile form of galactosialidosis are here detailed and three novel nucleotide changes were identified, one of them resulting in a stop codon, a type of mutation identified for the first time in galactosialidosis.
We also present some data on brain magnetic resonance, never detailed so far in galactosialidosis. Likewise, predictive functional/structural analyses of mutant proteins and phenotypes have been shown to be closely related, often giving clearcut indications on the specific form of the disease (early infantile, late infantile, juvenile).
The complexity of the clinical phenotypes in GS reflects the dual functions of PPCA/CTSA (catalytic and regulating/protective) and thus its functional role in both lysosomal and cell membranes. Further three-dimensional studies can provide additional information on functional domains, on protein-protein interactions within the lysosomal and the non-lysosomal complexes and on the onset and progression of symptoms.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
AC, and AM conceived the study, participated in the genetic studies and drafted the manuscript. SC, and RT participated in the genetic studies, sequence alignments and in the in silico analyses. HM and IM have made substantial contribution in the analyses and interpretation of data. MAD, RG and AD participated in the design and coordination of the study. LL and CRP have been involved in revising the manuscript critically. All authors read and approved the final manuscript.