Article Index

Eleonora Palagano, Harry C Blair, Alessandra Pangrazio, Irina Tourkova, Dario Strina, Andrea Angius, Gianmauro Cuccuru, Manuela Oppo, Paolo Uva, Wim Van Hul, Eveline Boudin, Andrea Superti-Furga, Flavio Faletra, Agostino Nocerino, Matteo C Ferrari, Guido Grappiolo, Marta Monari, Alessandro Montanelli, Paolo Vezzoni, Anna Villa, Cristina Sobacchi




Autosomal recessive osteopetrosis (ARO) is a rare genetic bone disease with genotypic and phenotypic heterogeneity, sometimes translating into delayed diagnosis and treatment. In particular, cases of intermediate severity often constitute a diagnostic challenge and represent good candidates for exome sequencing. Here, we describe the tortuous path to identification of the molecular defect in two siblings, in which osteopetrosis diagnosed in early childhood followed a milder course, allowing them to reach the adult age in relatively good conditions with no specific therapy. No clearly pathogenic mutation was identified either with standard amplification and resequencing protocols or with exome sequencing analysis. While evaluating the possible impact of a 3'UTR variant on the TCIRG1 expression, we found a novel single nucleotide change buried in the middle of intron 15 of the TCIRG1 gene, about 150 nucleotides away from the closest canonical splice site. By sequencing a number of independent cDNA clones covering exons 14 to 17, we demonstrated that this mutation reduced splicing efficiency but did not completely abrogate the production of the normal transcript. Prompted by this finding, we sequenced the same genomic region in 33 patients from our unresolved ARO cohort and found three additional novel single nucleotide changes in a similar location and with a predicted disruptive effect on splicing, further confirmed in one of them at the transcript level. Overall, we identified an intronic region in TCIRG1 that seems to be particularly prone to splicing mutations, allowing the production of a small amount of protein sufficient to reduce the severity of the phenotype usually associated with TCIRG1 defects. On this basis, we would recommend including TCIRG1 not only in the molecular work-up of severe infantile osteopetrosis but also in intermediate cases and carefully evaluating the possible effects of intronic changes. © 2015 American Society for Bone and Mineral Research.


Human osteopetrosis is a genetic bone disease characterized by increased bone density owing to failure in bone resorption by osteoclasts. For a long time, two main forms have been distinguished: an autosomal recessive form, also called malignant infantile, because of its early onset and frequent lethal outcome in the absence of a therapy; and an autosomal dominant form, also termed benign, because of its later presentation and milder course. Since 2000, the molecular basis of both forms has largely been unraveled, revealing a great genetic heterogeneity.[1] Subsequently, careful clinical characterization and longer follow-up have identified patients who do not fit within either class, documenting a wider variability of phenotypes, some of which escaped molecular characterization, because a responsible mutation could not be identified. In particular, it has become clear that, besides the paradigmatic recessive malignant and dominant benign forms, an intermediate group exists, in which the defect can be inherited as either a recessive or a dominant trait. These intermediate cases often represent a diagnostic challenge but also constitute the opportunity to learn important lessons on the bone biology. On this regard, we have recently reported on a young osteopetrotic patient in which an incomplete splicing defect in the TCIRG1 gene allowed for the production of the limited amount of wild-type transcript sufficient to dampen the severity of the disease to an almost completely normal lifestyle.[2] However, in that article, we pointed out that a definitive conclusion regarding the prognosis of the pathology was not possible because of the young age of the patient and the extraordinary nature of the mutation.

Exome sequencing is a powerful, high-throughput technique that in few years has greatly improved the discovery of the genetic defect underlying Mendelian disorders.[3] This approach is fundamentally based on the observation that the vast majority of mutations causing inherited diseases are located in coding regions of the genome and, vice versa, a large fraction of rare variants altering a protein structure are predicted to impact on its function. However, current sets of probes for exome capture target not only coding regions but also 5′ and 3′ untranslated regions (5'UTR and 3′UTR, respectively) and stretches of intronic regions, including exon-flanking regions and short introns contained between targeted exons.

Here, we describe the tortuous and, to some extent, serendipitous identification of the molecular defect in the middle of intron 15 of the TCIRG1 gene in two siblings affected by intermediate recessive osteopetrosis and in three additional patients in which a single mutated allele in the same gene had previously been found.

Materials and Methods


Clinical data and specimens, including blood and DNA samples, were collected from patients and their relatives after informed consent. This research complies with the standards established by the Independent Ethical Committee of the Humanitas Clinical and Research Centre.

Exome analysis

Exome capture was performed on 1.5 μg of high-quality genomic DNA from patients 1A and 1B and their parents, and the TruSeq Exome Enrichment Kit (Illumina Inc., San Diego, CA, USA) was used. The enriched library was validated by the Agilent DNA 1000 Kit (Agilent Technologies Inc., Santa Clara, CA, USA) and loaded on the cBot Station (Illumina Inc.) to create clonal clusters on the flow cell. Sequencing was performed on the Hiseq2000 Instrument. Reads extracted with the Illumina tools were aligned to the reference genome hg19 by using BWA-MEM[4] and stored in compressed binary files (BAM). Single nucleotide variations, insertions, and deletions were called using the Genome Analysis Toolkit (GATK).[5] All the analyses were performed using the Orione platform.[6] The variants identified were merged in a VariantCallingFormat (VCF) file, annotated by KGGSeq,[7] prioritized, and filtered according to a standard workflow for exome sequencing.

Molecular studies

The molecular analysis of genes known to be responsible for the different types of the disease (TCIRG1, CLCN7, PLEKHM1, RANKL, RANK, and SNX10) was performed by amplification and direct sequencing of exons and intron-exon boundaries as previously described.[ 8-13] The TCIRG1 intron 15 was amplified using the forward primer 10271F 5′-TGTTCCTCTTCTCCCACAGC-3′ located in exon 15 and the reverse primer 5R 5′-CATCGGAGCTCCAGCCATT-3′ in exon 17; sequencing was performed using the 10271F primer. After mRNA isolation from peripheral blood mononuclear cells lysed in Trizol (Invitrogen, Carlsbad, CA, USA) and reverse transcription using the High Capacity cDNA Reverse Transcription kit (Life Technologies, Carlsbad, CA, USA) according to the manufacturers’ instructions, the effect of the intronic variants on TCIRG1 transcript processing was investigated with the forward primer 9912F 5′-GCCTGGCTGCCAACCACTTG-3′ in exon 14 and the reverse primer 5R in exon 17. For patient (Pt) 1A and Pt 1B, the PCR product was directly cloned in the TOPO TA Cloning plasmid vector (Life Technologies, Carlsbad, CA, USA), according to the manufacturer's instructions, and individual clones were sequenced using the 9912F forward primer. For Pt 2, a PCR product was obtained from the cDNA with the forward primer 5F 5′-CTGGCCCAGCACACGATGCT-3′ in exon 13 and the reverse primer 5R in exon 17; this product was cloned and individual clones were sequenced using primer 5R. For the confirmation of the variant in the LRP5 gene, the relevant genomic region was amplified using primers forward 5′-TGGGAGGAAGGAAGGAATGC-3′ and reverse 5′-TCAGTGGCATGGGGATTAGG-3′, whereas primers forward 5′-GGGAGTGAGCACCGTCTATA-3′ and reverse 5′-CTCCTAGGACTACGCCCAAG-3′ were used for confirmation of the variant in the CHKA gene; in both cases, sequencing was performed using the forward primer.

For all the PCR reactions above, the thermocycling conditions were: initial denaturation step at 94°C for 3 minutes, followed by 34 cycles of denaturation at 94°C for 30 seconds, annealing at 62°C (55°C for the LRP5 variant) for 30 seconds, and amplification at 72°C for 30 seconds.

The effect of the mutations identified in the intron 15 of the TCIRG1 gene was predicted using the software Human Splicing Finder, Version 2.4.1 ([14] The mutation nomenclature conforms to HGVS ([15]

Histological analysis

A bone biopsy specimen of Pt 1A was fixed in 4% paraformaldehyde (PFA), decalcified for 4 hours at 20 °C in 5% HCl (pH ∼1), then rinsed in water, paraffin-embedded, and sectioned at 6 μm. Sections were deparaffinized for hematoxylin-eosin staining and for fluorescent antibody labeling as described.[16] For TCIRG1 labeling, a mouse monoclonal (clone 6H3) antibody was used (Sigma Aldrich Co., St. Louis, MO, USA) at 1:100 dilution. For TRAP, a rabbit polyclonal antibody, C-terminal (Abcam PLC, Cambridge, UK) was used at 1:100 dilution. Briefly, sections were blocked in PBS/2% BSA for 2 hours, reacted overnight with antibodies at indicated concentrations in PBS with 0.01% Tween 20, then labeled for fluorescent analysis using Alexa Fluor 488 Donkey Anti-Mouse IgG (H+L) Antibody (green) or Alexa Fluor 594 Donkey Anti-Rabbit IgG (H+L) Antibody (red) antibodies, at 1:1000, both from Life Technologies, for 1 hour. For nuclear labeling, Hoechst 33342 blue (Invitrogen) was used. A Nikon TE2000 inverted microscope was used for imaging, via 14-bit 2048 × 2048 pixel monochrome CCD and RGB filters to reconstruct color (Spot, Sterling Heights, MI, USA). Green fluorescence (Cy-2) used excitation at 450 to 490 nm, a 510-nm dichroic mirror, and a 500- to 570-nm emission filter. Red fluorescence (Cy3) used excitation at 530 to 560 nm, a 575-nm dichroic mirror, and a 580- to 650-nm emission filter.


Clinical evaluation of patients

Patients 1A and 1B (Pt 1A and Pt 1B) were two affected siblings born from Italian consanguineous parents (third-degree consanguinity). Pt 1A, female, was first evaluated at age 6 years because of visual problems, in particular right amaurosis and divergent strabismus. A radiological investigation of the skull revealed increased bone density, and a skeletal survey confirmed generalized osteopetrosis (Fig. 1A). Computed tomography (CT) scan showed bilateral optic and acoustic foramina restriction; analysis of the visual evoked potentials (VEP) confirmed visual impairment on the right eye, whereas brainstem evoked response audiometry (BERA) gave normal results. At that time, her hematological function was normal despite mild hepatosplenomegaly. On these results, a diagnosis of osteopetrosis with low progression was made. The presence of a similar bone phenotype in her younger brother (Pt 1B) at the age of 2 years (Fig. 1B), together with the absence of pathological findings in their parents as well as their consanguinity, suggested an autosomal recessive pattern of inheritance. In the following years, the clinical history of both patients was mainly characterized by recurrent fractures (8 in Pt 1A, 11 in Pt 1B) at different skeletal segments, particularly the lower limbs (Fig. 1C), and dental problems, with delayed tooth eruption, congenitally missing teeth and, in the elder sibling, dental abscesses, requiring intravenous administration of antibiotics and a surgical intervention. The visual acuity progressively worsened in Pt 1A and was completely lost in the right eye side at the age of 9 years; subsequently, a surgery was performed to preserve the vision of the left eye. In addition, she lost hearing in the left ear starting from age 6 years. Most recent laboratory investigations, performed concomitantly with an orthopedic surgery for total hip replacement after removal of a previous prosthesis, did not show alterations of the evaluated parameters (Table 1). Histological analysis on a bone specimen from Pt 1A demonstrated an incomplete osteopetrosis with some areas displaying mostly retained mineralized cartilage and primary spongiosa, whereas others showing mostly normal lamellar bone in well-formed trabecular bone (Fig. 2). Importantly, several osteoclasts were visualized, thus clearly identifying the disease in this patient as an osteoclast-rich osteopetrosis (Fig. 2). At present, both affected siblings do not receive any specific therapy. Overall, their clinical picture can be considered benign, with Pt 1B displaying no sensorial defects and leading an almost completely normal life at age 32 years.

Fig. 1
Fig. 1.

Representative X-rays of Pt 1A, Pt 1B, and Pt 2, showing a diffuse increase in bone density, the typical harlequin-mask pattern of the sclerotic skull base on the AP cranium, the lack of metaphyseal bone remodeling, and the bone deformities resulting from the remodeling defects and the non-perfect healing and remodeling of fractures. (A) Pt 1A at age 6 years; (B) Pt 1B at age 2 years; (C) Pt 1A at age 35 years; (D) Pt 2 at age 22 years. Overall, the radiographs testify for a severe form osteopetrosis..

Table 1.  Most Recent Laboratory Data in Pt 1A andPt 1B
 Pt 1APt 1BRef. range
Age (years) 37 33  
Erythrocytes (× 106/μL) 3.79 5.10 4.2–5.4
Hemoglobin (g/dL) 11.2 14.5 12–16
Hematocrit (%) 34.5 43.9 37–47
Serum calcium (mmol/L) 2.38 2.48 2.10–2.60
Serum phosphorus (mmol/L) 1.22 1.40 0.80–1.50
Serum iron (μg/dL) 98 123 50–170
Serum ferritin (ng/mL) 23 288 10–120
Serum alkaline phosphatase (U/L) 44 65 40–150
Serum bone ALP isoenzyme (μg/L) 6.2 10.9 6–26
Serum osteocalcin (ng/mL) 21 18 6.5–42.3
Serum PTH (pg/mL) 37.6 27.1 14.5–87.1
Serum calcitonin (pg/mL) 3.8 4 0–5.5
Serum 25(OH)D3 (ng/mL) 10.4 23.2 8.6–54.8
Fig. 2
Fig. 2.

Representative histological sections of a bone specimen from Pt 1A at age 35 years. (A) Visualization in polarized light and hematoxilin-eosin staining. Scale bar = 200 μm. The mature cortical bone shows lamellae in polarized light; the defects within the bone are islands of retained mineralized cartilage. (B) Higher magnification of hematoxilin-eosin staining showing several multinucleated osteoclasts. Scale bar = 100 μm. (C) TRAP fluorescent antibody labeling (red) and phase contrast with the fluorescence image overlaid; nuclei are stained with Hoechst (blue). The inset shows a single osteoclast at twice the magnification. Scale bar = 100 μm; inset = 10 μm. (D) TCIRG1 fluorescent antibody labeling (green) and phase contrast with the fluorescence overlaid (right). Scale bar = 100 μm; inset = 10 μm.

Patient 2 (Pt 2) was the third child of unrelated, healthy Italian parents. The symptoms of the disease appeared at age 4 years, with a generalized increase in bone density (representative, most recent radiological findings in Fig. 1D). He reported 7 nontraumatic fractures and suffered from recurrent osteomyelitis of the jaw. He underwent 3 orthopedic surgeries and several maxilla-facial interventions. No hematological nor sensorial deficits arose, and the patient is alive and reasonably well at age 24 years.

Patient 3 (Pt 3) has been already described (Pt 6 in Mazzolari and colleagues[17]). Briefly, the patient presented with a classical ARO phenotype, with bone defects and secondary hematological and neurological deficits at age 6.5 months. She received hematopoietic stem cell transplantation (HSCT) from a matched unrelated donor (MUD) at age 20.5 months, achieved full engraftment, and was alive and well 7.5 years after transplantation.

Patient 4 (Pt 4) was born from healthy unrelated parents from Southeastern Asia. He presented with bone defects, blindness, and hepatosplenomegaly in neonatal life. He received HSCT but eventually died 4 years after HSCT without engraftment. No further information is available.

Genetic findings

Based on the suspected recessive pattern of inheritance of the disease in Pt 1A and Pt 1B, molecular analysis in this family initially focused on the genes known to cause recessive osteopetrosis (TCIRG1, CLCN7, PLEKHM1, RANKL, RANK, and SNX10); the OSTM1 gene was not included in the analysis because the two affected siblings did not display the primary neurological defects that are a hallmark of this ARO form. However, no molecular defect could be identified; therefore, we performed exome sequencing on both patients and their parents. This approach achieved an 80× mean coverage over the 62 Mb of exomic sequence, with more than 94% of targeted regions covered. The overall transition to transversion rate (Ti/Tv) was 2.4, which is in line with what was expected for exome sequencing. The analysis identified a total of 191,872 variants that were filtered with dbSNP138 and 1000 Genome Project and according to the pattern of inheritance of the disease and to the parental consanguinity (Table 2).

Table 2. Number of Variants/Genes Identified in Pt 1A and Pt 1B Through Exome Sequencing
Total variants 191,872
Shared variants after variant-quality filtering 157,065
Variant with MAF <0.05 19,170
Nonsynonymous/indel/splice sites/UTR variants 1319/98/506/5149
Homozygous variants 32
Novel variants (dbSNP138/1000GP queried) 2
Genes with plausible disease association 1 (LRP5)

Among the homozygous variants, we found a single nucleotide change in exon 22 of the LRP5 gene (Supplementary Fig. 1A) (NG_015835.1:g.138882C>T; NM_002335:c.4574C>T) predicted to cause an amino acid substitution (p.Ala1525Val). Thus far, the so-called “activating mutations” linked with an increased bone density are clustered in the first beta-propeller, at the extracellular N-terminus of the protein, whereas mutations causing low bone density appear to be clustered in the second beta-propeller.[ 18] Interestingly, the variant c.4574C>T lies in a completely different region of the protein, that is the cytoplasmic tail (Supplementary Fig. 1B), known to interact with axin and FRAT, which are key components in the Wnt/β-catenin signaling pathway.[ 19, 20]

So we confirmed this variant by Sanger sequencing and tested its activity in a classical luciferase reporter assay; however, the activation of the Wnt signaling was comparable in the presence of either the wild-type or the mutant LRP5 construct (Supplementary Fig. 1C). In addition, in this experiment, no difference between the wild-type and mutant LRP5 was observed after cotransfection with sclerostin (Supplementary Fig. 1C), whereas the “activating mutations” in LRP5 are known to result in increased Wnt signaling activity owing to reduced inhibition by sclerostin or dickkopf 1. So, these results did not support the hypothesis of a causative role for the variant in the pathogenesis of the disease. Accordingly, at the time when this article was in preparation, an update of SNP databases included this change in the LRP5 gene at the homozygous state.

Therefore, we returned to the list of variants identified by exome sequencing. No additional noteworthy variants in coding regions were present, so we focused our attention on variants in 3′ untranslated regions (3′UTR) because they are known to influence gene expression.[21] A single nucleotide change was annotated in the 3′UTR of the choline kinase alpha (CHKA) gene (NM_001277; *327T>C), at the homozygous state in patients 1A and 1B and at the heterozygous state in their parents. Interestingly, CHKA maps to chromosome 11q13.2 on the reverse strand and immediately downstream to the TCIRG1 gene. Because of the close proximity of the two genes, we asked whether the *327T>C variant might affect the expression of the TCIRG1 gene. Investigation at the cDNA level through semiquantitative RT-PCR did not show reduced expression; however, at least two unexpected bands in the two affected siblings by RT-PCR spanning exons 14 to 17 were observed (Fig. 3A), suggesting a splicing aberration. Therefore, we sequenced the corresponding genomic region, including also the whole of the three introns, and identified a single nucleotide change in the middle of intron 15 (g.10466G>A, c.1884+146G>A), which was not reported in dbSNP and was present at the homozygous state in Pt 1A and Pt 1B and at the heterozygous state in their parents but absent in 100 control chromosomes from a matched population (Fig. 3B and data not shown). Despite this change being far from both the donor of exon 15 and the acceptor of exon 16, in silico analysis suggested a potential detrimental effect on the splicing process, by altering the strength of alternative splice sites compared with the wild-type sequence. As a confirmation of this prediction, when we cloned the cDNA PCR product spanning exons 14 to 17 and sequenced 73 independent clones, we found that few clones displayed the wild-type sequence (4/73 clones), whereas at least 8 diverse aberrant transcripts were produced, the most abundant being clones retaining 144 nucleotides of intron 15 (22/73 clones) and clones in which the first 183 nucleotides of exon 15 were skipped and 144 nucleotides of intron 15 retained (30/73 clones) (Fig. 3C). These data clearly supported the hypothesis of a pathogenetic role of the variant.

Fig. 3
Fig. 3.

(A) Semi-quantitative RT-PCR spanning exons 14 to 17 (upper panel), showing the presence of at least two abnormal bands in the patients (the normal band seen in Pt 2 is from the other mutated allele), or exons 7 to 10 (middle panel), showing only the presence of the expected band in all the samples. (B) Schematic representation of the relevant region of the TCIRG1 gene, highlighting the position of the mutations identified in intron 15; the nucleotide change found in each patient is in bold in the lower line. (C) Structure of the most frequent aberrant transcripts found in Pt 1A, Pt 1B and Pt 2: i) clones in which only the last 31 nucleotides of exon 15 are retained, followed by 143 nucleotides of intron 15, and ii) clones retaining 143 nucleotides of intron 15. The nucleotide change found in each patient is in bold in the lower line, as in panel B.

Prompted by these findings, we sequenced TCIRG1 intron 15 in 26 osteopetrotic patients not yet assigned to any subgroup because of the absence of mutations in known genes, as well as in 7 ARO patients in which we had previously identified only a single mutated TCIRG1 allele[17, 22] (unpublished data). Although no mutation was found in the 26 unclassified patients, in 3 of the 7 patients harboring a single mutated TCIRG1 allele, we found 3 additional, heterozygous single nucleotide changes in intron 15, close to that carried by Pt 1A and Pt 1B, namely g.10462T>A (c.1884+142T>A) in Pt 2 (mutation confirmed at the heterozygous state in Pt 2's healthy mother and in one unaffected brother; data not shown), g.10452T>C (c.1884+132T>C) in Pt 3, and g.10469C>T (c.1884+149C>T) in Pt 4 (Fig. 3B and Table 3. In silico analysis was performed also on these variants, obtaining a similar prediction of potential disruptive effects on the splicing process. A cell sample suitable for expression studies was available only for Pt 2, and we could confirm also in this case by RT-PCR followed by cloning and sequencing, the presence of a residual amount of wild-type transcript (5/75 clones, 6.7%) and the same pattern of aberrant transcripts previously found in Pt 1A and Pt 1B, namely 12/75 clones retaining 143 nucleotides of intron 15, 20/75 clones skipping the first 183 nucleotides of exon 15 and retaining 143 nucleotides of intron 15, 20/75 clones skipping the first 183 nucleotides of exon 15 and retaining 141 nucleotides of intron 15, plus other less frequent species. Sequence analysis of individual clones showed that loss of large part of exon 15 was due to the usage of an internal, cryptic acceptor site, while retention of 143 nucleotides of intron 15 was due to the usage of an intronic, cryptic donor site (Supplementary Fig. 2).

Table 3.  Molecular Findings
PatientGenomic changeaLocationcDNA changebEffect
  • a

    Accession number genomic sequence of the TCIRG1 gene: AF033033.

  • b

    Accession number of the TCIRG1variant1 cDNA: NM_006019; the numbering used starts with nucleotide +1 for the A of the ATG-translation initiation codon.

1A and 1B g.10466G>A Intron 15 c.1887+146G>A Putative aberrant splicing
g.10466G>A Intron 15 c.1887+146G>A Putative aberrant splicing
2 g.9909G>A Exon 14 c.1559G>A p.Trp520X
g.10462T>A Intron 15 c.1887+142T>A Putative aberrant splicing
3 g.11622T>C Exon 19 c.2345T>C p.Met782Thr
g.10452T>C Intron 15 c.1887+132T>C Putative aberrant splicing
4 g.8600T>C Intron 11 c.1305+2T>C Putative aberrant splicing
g.10469C>T Intron 15 c.1887+149C>T Putative aberrant splicing


Fifteen years of genetic dissection of human osteopetrosis have greatly extended our understanding of the physiopathology of this disease and, more in general, of the bone tissue. At present, an exact classification is obtained in most cases of the extreme phenotypes: on one hand, autosomal recessive forms with a very severe and early presentation; on the other, adult, benign forms mildly impacting on the patients' life. This allows the identification of the most appropriate therapy. On the contrary, intermediate, atypical forms are more difficult to solve and manage; therefore, these are often addressed by next-generation sequencing technologies or require an adaptation of commonly used protocols.

We recently reported the identification of an incomplete splicing defect in the TCIRG1 gene leading to benign, recessive osteopetrosis in a young girl.[2] At that time, we thought that this kind of phenotype owing to the TCIRG1 gene was extremely rare because no other similar finding was present in either our cohort of more than 300 patients or in the literature, to the best of our knowledge. However, our present data suggest that this is not the case; in fact, we report here the identification of 4 different single nucleotide changes in intron 15 predicted to impact on the splicing process of the neighboring exons. For 2 of 4 novel mutations, the effect was demonstrated at the cDNA, showing that these genetic defects were hypomorphic and allowed for the production of a small amount of wild-type transcript. In addition, bone histology in Pt 1A allowed us to clearly document osteoclast-rich osteopetrosis, in the presence of areas of well-formed, mature trabecular bone. Interestingly, the mutations herein described are located in close proximity to one another and probably identify a hot spot region for mutation within intron 15. Moreover, they are found in the middle of a 368-nucleotide-long intron, quite far from the canonical splice sites, so our standard molecular analysis of the TCIRG1 gene initially failed to identify them. Due to their localization, we were not able to exactly define the pathogenic mechanism; however, a careful sequence analysis of the alternative transcripts revealed the usage of cryptic splicing sites, namely an acceptor in exon 15 and a donor in intron 15. Deep intronic mutations, leading to the formation of cryptic splice sites which compete with the canonical ones during RNA processing, have been recently reported also in different diseases, such as Stargardt disease, Gorlin syndrome, optic atrophy, Marfan syndrome and others.[23-26] In the field of skeletal diseases, a deep intronic mutation in the TMEM165 gene has been reported to cause the congenital disorder of glycosylation type II with bone dysplasia.[27] At present this type of defects is likely underestimated, since the genetic tests are still mainly focused on coding genomic sequences and transcript analysis is rarely performed; however, these data clearly highlight the importance of the noncoding genome and the need to assess the functional impact of remote intronic variants in potential disease-genes.

The patients described in the present study displayed a different level of severity of the disease. Whereas in Pt 3 and Pt 4 the diagnosis was made in early life and their osteopetrosis was treated with HSCT, in Pt 1A and Pt 2 the bone defects became apparent at school age and the disease progression was slow and overall benign, thus representing typical intermediate cases. Moreover, Pt 1B was diagnosed at an asymptomatic early stage because of the previous identification of his elder affected sister, Pt 1A. In this regard, the difference between these two siblings is interesting; whereas the elder showed important secondary neurological deficits, these were completely absent in her younger brother, who seemed to be more mildly affected. This is reminiscent of what is commonly observed in autosomal dominant osteopetrosis, in which phenotypic variations can be observed in the presence of the very same genetic defect, even within the same family. A role for variants in modifier genes has been often suggested. Intriguingly, we noticed that Pt 1A and Pt 1B carried an almost completely different set of single-nucleotide polymorphisms (SNPs) in the CLCN7 gene (only 2 shared SNPs out of 30 genotyped, data not shown). Among these, Pt 1B, but not his sister, bore 2 SNPs within exon 15 (rs12926089 and rs12926669) at the heterozygous state, which have been associated with low bone mineral density in different populations.[28, 29] Similarly, it could be hypothesized that additional variants in other genes influencing bone metabolism are present in one or the other affected sibling and modulated their phenotype. Moreover, in adolescence and adult life, sexual hormones might have further impacted on the disease manifestation.

In conclusion, our results depict an even wider spectrum of mutations and phenotypes in TCIRG1-dependent osteopetrosis and suggest the analysis of this gene is appropriate not only in the molecular work-up of severe patients but also of intermediate cases. Based on these findings, standard protocols for gene amplification and sequencing, focused on exons and exon-intron boundaries, are likely to be revised. In particular, intron 15 should be included in the routine sequencing of the TCIRG1 gene; more in general, intronic changes in genes associated with osteopetrosis, even deeply embedded in introns, should be considered and their effect carefully evaluated.


We are grateful to the patients and their relatives for their cooperation. This work was partially supported by the Telethon Foundation (grant GGP12178 to CS); by PRIN Project (20102M7T8X_003 to AV); by Giovani Ricercatori from Ministero della Salute (grant GR-2011-02348266 to CS); by Ricerca Finalizzata from Ministero della salute (RF-2009-1499,542 to AV); by the European Community's Seventh Framework Program (FP7/2007-2013, SYBIL Project to AV and WVH); by PNR-CNR Aging Program 2012–2014 to PV and AV; by the Leenaards Foundation Lausanne and the Swiss National Foundation to AS-F; by the National Institutes of Health USA (R01 AR065407-01 to HCB), and by the Department of Veteran's Affairs USA (BX002490-01 to HCB); by the Research Foundation-Flanders (FWO grant G.0197.12N to WVH); and by a TOP grant to WVH. EB holds a postdoctoral fellowship of the Research Foundation-Flanders.

Author's roles: Molecular studies: EP, AP, and DS. Patient evaluation: ASF, FF, AN, MCF, and GG. Laboratory investigation: MM and AM. Exome sequencing and bioinformatics analysis: AA, GC, MO, and PU. Histological analysis: HCB and IT. Luciferase assay on LPR5: WVH and EB. Study design: PV, AV, and CS. Drafting manuscript and revising manuscript content: all authors.


  1. Sobacchi C, Schulz A, Coxon FP, Villa A, Helfrich MH.
    Osteopetrosis: genetics, treatment and new insights into osteoclast function.
    Nat Rev Endocrinol. 2013; 9:522–536.
  2. Sobacchi C, Pangrazio A, López AG, et al.
    As little as needed: the extraordinary case of a mild recessive osteopetrosis due to a novel splicing hypomorphic mutation in the TCIRG1 gene.
    J Bone Miner Res. 2014; 29:1646–1650.
  3. Rabbani B, Tekin M, Mahdieh N.
    The promise of whole-exome sequencing in medical genetics.
    J Hum Genet. 2014; 59:5–15.
  4. Li H, Durbin R.
    Fast and accurate long-read alignment with Burrows-Wheeler transform.
    Bioinformatics. 2010; 26:589–595.
  5. McKenna A, Hanna M, Banks E, et al.
    The genome analysis toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data.
    Genome Res. 2010; 20:1297–303.
  6. Cuccuru G, Orsini M, Pinna A, et al.
    Orione, a web-based framework for NGS analysis in microbiology.
    Bioinformatics. 2014; 30:1928–1929.
  7. Li M-X., Kwan JSH, Bao S-Y, et al.
    Predicting Mendelian disease-causing non-synonymous single nucleotide variants in exome sequencing studies.
    PLoS Genet. 2013; 9:e1003143.
  8. Frattini A, Orchard PJ, Sobacchi C, et al.
    Defects in the TCIRG1-encoded 116kD subunit of the vacuolar proton pump are responsible for a subset of human autosomal recessive osteopetrosis.
    Nat Genet. 2000; 25:343–346.
  9. Frattini A, Pangrazio A, Susani L, et al.
    Chloride channel ClCN7 mutations are responsible for severe recessive, dominant and intermediate osteopetrosis.
    J Bone Miner Res. 2003; 18:1740–1747.
  10. Van Wesenbeeck L, Odgren PR, Coxon FP, et al.
    Involvement of PLEKHM1 in osteoclastic vesicular transport and osteopetrosis in incisors absent rats and humans.
    J Clin Invest. 2007; 117:919–930.
  11. Sobacchi C, Frattini A, Guerrini M, et al.
    Osteoclast-poor human osteopetrosis due to mutations in RANKL.
    Nat Genet. 2007; 39:960–962.
  12. Guerrini MM, Sobacchi C, Cassani B, et al.
    Human osteoclast-poor osteopetrosis with hypogammaglobulinemia due to TNFRSF11A (RANK) mutations.
    Am J Hum Genet. 2008; 83:64–76.
  13. Pangrazio A, Fasth A, Sbardellati A, et al.
    SNX10 mutations define a subgroup of human autosomal recessive osteopetrosis with variable clinical severity.
    J Bone Miner Res. 2013; 28:1041–1049.
  14. Desmet FO, Hamroun D, Lalande M, Collod-Béroud G, Claustres M, Béroud C.
    Human Splicing Finder: an online bioinformatics tool to predict splicing signals.
    Nucl Acids Res. 2009; 37:e67.
  15. Den Dunnen JT, Antonarakis SE.
    Mutation nomenclature extensions and suggestions to describe complex mutations: a discussion.
    Hum Mutat. 2000; 15:7–12.
  16. Robinson LJ, Mancarella S, Songsawad D, et al.
    Gene disruption of the calcium channel Orai1 results in inhibition of osteoclast and osteoblast differentiation and impairs skeletal development.
    Lab Invest. 2012; 92:1071–1083.
  17. Mazzolari E, Forino C, Razza A, Porta F, Villa A, Notarangelo LD.
    A single-center experience in 20 patients with infantile malignant osteopetrosis.
    Am J Hematol. 2009; 84:473–479.
  18. Balemans W, Van Hul W.
    The genetics of low-density lipoprotein receptor-related protein 5 in bone: a story of extremes.
    Endocrinology. 2007; 148:2622–2629.
  19. Zeng X, Tamai K, Doble B, et al.
    A dual-kinase mechanism for Wnt co-receptor phosphorylation and activation.
    Nature. 2005; 438:873–877.
  20. Katoh M, Katoh M.
    WNT signaling pathway and stem cell signaling network.
    Clin Cancer Res. 2007; 13:4042–4045.
  21. Michalova E, Vojtesek B, Hrstka R.
    Impaired pre-mRNA processing and altered architecture of 3' untranslated regions contribute to the development of human disorders.
    Int J Mol Sci. 2013; 14:15681–15694.
  22. Sobacchi C, Frattini A, Orchard P, et al.
    The mutational spectrum of human malignant autosomal recessive osteopetrosis.
    Hum Mol Genet. 2001; 10:1767–1773.
  23. Bax NM, Sangermano R, Roosing S, et al.
    Heterozygous deep-intronic variants and deletions in ABCA4 in persons with retinal dystrophies and one exonic ABCA4 variant.
    Hum Mutat. 2015; 36:43–47.
  24. Bholah Z, Smith MJ, Byers HJ, Miles EK, Evans DG, Newman WG.
    Intronic splicing mutations in PTCH1 cause Gorlin syndrome.
    Fam Cancer. 2014; 13:477–480.
  25. Bonifert T, Karle KN, Tonagel F, et al.
    Pure and syndromic optic atrophy explained by deep intronic OPA1 mutations and an intralocus modifier.
    Brain. 2014; 137:2164–2177.
  26. Gillis E, Kempers M, Salemink S, et al.
    An FBN1 deep intronic mutation in a familial case of Marfan syndrome: an explanation for genetically unsolved cases?
    Hum Mutat. 2014; 35:571–574.
  27. Zeevaert R, de Zegher F, Sturiale L, et al.
    Bone Dysplasia as a Key Feature in Three Patients with a Novel Congenital Disorder of Glycosylation (CDG) Type II Due to a Deep Intronic Splice Mutation in TMEM165.
    JIMD Rep. 2013; 8:145–152.
  28. Pettersson U, Albagha OME, Mirolo M, et al.
    Polymorphisms of the CLCN7 gene are associated with BMD in women.
    J Bone Miner Res. 2005; 20:1960–1967.
  29. Chu K, Snyder R, Econs MJ.
    Disease status in autosomal dominant osteopetrosis type 2 is determined by osteoclastic properties.
    J Bone Miner Res. 2006; 21:1089–1097.
  30. Gong Y, Slee RB, Fukai N, et al.
    LDL receptor related protein 5 (LRP5) affects bone accrual and eye development.
    Cell. 2001; 107:513–523.
  31. Balemans W, Devogelaer JP, Cleiren E, Piters E, Caussin E, Van Hul W.
    Novel LRP5 missense mutation in a patient with a high bone mass phenotype results in decreased DKK1-mediated inhibition of Wnt signaling.
    J Bone Miner Res. 2007; 22:708–716.
Submit to FacebookSubmit to Google PlusSubmit to TwitterSubmit to LinkedIn

With your donation,
You can support our
commitment to research..



GAP II è un sistema per l’archiviazione di dati clinici e chirurgici di pazienti sottoposti ad interventi di chirurgia protesica di anca, ginocchio e spalla.