Paget’s disease of bone (PDB) is a chronic bone disorder and although genetic factors appear to play an important role in its pathogenesis, to date PDB causing mutations were identified only in the Sequestosome 1 (SQSTM1) gene at the PDB3 locus. PDB6 locus, also previously linked to PDB, contains several candidate genes for metabolic bone diseases. We focused our analysis in the most significantly associated variant with PDB, within the Optineurin (OPTN) gene, i.e. the common variant rs1561570. Although it was previously shown to be strongly associated with PDB in several populations, its contribution to PDB pathogenesis remains unclear. In this study we have shown that rs1561570 may contribute to PDB since its T allele results in the loss of a methylation site in patients’ DNA, leading to higher levels of OPTN gene expression and a corresponding increase in protein levels in patients’ osteoclasts. This increase in OPTN expression leads to higher levels of NF-κB translocation into the nucleus and increasing expression of its target genes, which may contribute to the overactivity of osteoclasts observed in PDB. We also reported a tendency for a more severe clinical phenotype in the presence of a haplotype containing the rs1561570 T allele, which appear to be re-enforced with the presence of the SQSTM1/P392L mutation. In conclusion, our work provides novel insight towards understanding the functional effects of this variant, located in OPTN intron 7, and its implication in the contribution to PDB pathogenesis.
Citation: Silva IAL, Conceição N, Gagnon É, Brown JP, Cancela ML, Michou L (2018) Molecular effect of an OPTN common variant associated to Paget's disease of bone. PLoS ONE 13(5): e0197543. https://doi.org/10.1371/journal.pone.0197543
Editor: Sakamuri V. Reddy, Charles P. Darby Children's Research Institute, UNITED STATES
Received: February 10, 2018; Accepted: May 3, 2018; Published: May 21, 2018
Copyright: © 2018 Silva et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: This research was supported in part by national funds from Foundation for Science and Technology (FCT) through project UID/Multi/04326/2013 to MLC and by Canadian Institutes for Health Research (MOP130457 to LM), Canada, CHU de Québec Foundation, Canadian Foundation for Innovation, Fonds de recherche du Québec-santé, Laval University and CHU de Québec-Université Laval Research Centre. The authors would also like to acknowledge the CHU de Québec-Université Laval Research Centre (CHUL) Gene Expression Platform, Quebec, Canada for the quantitative real-time PCR measurements and the Centre of Marine Sciences (CCMAR), Faro, Portugal for the use of the sequencing platform. IS and NC were supported by doctoral and post-doctoral fellowships (SFRH/BD/77227/2011 and SFRH/BPD/111898/2015, respectively) from FCT. LM is supported by a career award from the Fonds de recherche Quebec-Santé (FRQ-S), Quebec, Canada. All the funders cited above had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Paget’s disease of bone (PDB) is a metabolic bone disorder that affects between 1% and 3% of Caucasians over the age of 55 [1–3]. This disease is characterized by focal abnormal bone remodelling, with increased bone resorption and accelerated, excessive and disorganized new bone formation. The pathophysiology of PDB is currently an area of intensive investigation and both genetic and non-genetic factors have been implicated. Fifteen to 40 percent (15–40%) of affected patients have a first-degree relative with PDB and numerous studies have described large families with PDB exhibiting an autosomal dominant mode of inheritance [3–5]. Linkage studies in these families have identified a number of susceptibility loci on chromosomes 6p21 (PDB1) , 18q21.1–22 (PDB2) , 5q35 (PDB3), 5q31 (PDB4) , 2q36 (PDB5) , 10p13 (PDB6)  and 18q23 (PDB7) . These regions contain many genes that could be plausible candidates based on their known functions. A genome-wide scan in British families with PDB has shown a linkage to the 10p13 (PDB6) locus [9,10] and reanalysis of data from this genome-wide scan confirmed this genetic association , namely to the common variant rs1561570 (hg19, chr10:g.13155726 T>C) in the Optineurin (OPTN) gene. Our group has also replicated the strong, statistically significant, genetic association of rs1561570 (p-value = 5.65×10−7) with PDB in the French-Canadian population . OPTN gene has been previously linked to glaucoma  and neurodegenerative diseases, like Alzheimer’s, Parkinson’s or amyotrophic lateral sclerosis . It was also found to be important for several cellular events such as membrane trafficking and maintenance of the Golgi complex, autophagy and protein turnover through the ubiquitin-proteasome pathway [15–18]. OPTN has been characterized as an autophagy adaptor protein, which regulates selective autophagy of ubiquitin-coated cytosolic proteins, organelles and parasites such as Salmonella enterica. This function depends on the phosphorylation of its LC3-interacting motif by TANK-binding kinase 1 (TBK1) [16,18–20]. OPTN can also mediate protein aggregate removal through a ubiquitin-independent mechanism  and, in addition, it seems to be part of the TNFα pathway, contributing to mediate the availability of NF-κB in the cell [22,23]. A possible role for OPTN in bone was recently described by Obaid et al. (2015) who have shown that mice harbouring a loss-of-function mutation in Optn presented an increase in osteoclast activity and in bone turnover, thus indicating that this protein could be acting as a negative regulator of osteoclast differentiation, possibly by inhibiting NF-κB activation . Despite these findings, this animal model did not develop a PDB-like phenotype. Previously, in a different model and in contrast with those results, OPTN was described as being involved in the activation of NF-κB pathway . Indeed, Journo and colleagues (2009) have observed that in human cells, infected with the human T lymphotropic virus type 1, OPTN increased the ubiquitination of Tax1 as well as Tax1-dependent NF-κB signalling, thus promoting NF-κB activation, a result that is in agreement with other studies [26–28]. These discrepancies suggest that the pathologies linked to OPTN cannot be associated exclusively to its loss of function, but more likely to its aberrant expression, resulting from either mutant or gain of OPTN function affecting different pathways in distinct biological contexts, and which are not yet fully understood. Therefore, additional work is required to understand the complexity of the events associated to OPTN function, namely using ex-vivo tools such as PDB patients’ samples. In the present work the effect in osteoclastogenesis of the most significantly PDB-associated OPTN variant (rs1561570) was assessed using patients’ monocytes as study model. These cells are precursors of osteoclasts and can be induced to fuse in order to originate mature osteoclast cells in vitro. The results obtained are expected to contribute to further elucidate the role of OPTN in the pathophysiology of a complex chronic bone disease such as PDB.
Materials and methods
This study was approved by the CHU de Québec-Université Laval Ethics Committee and all participants have signed a consent form before inclusion in the study. We investigated patients with familial form of PDB (one patient per family, to avoid the biais of relatedness in our genetic association analysis), unrelated PDB patients and healthy controls, all from the French-Canadian population. The intra familial cosegregation of the T allele among affected relatives within familial forms was not performed. All patients and healthy donors studied here were non-carrier of the P392L mutation within the SQSTM1 gene (PDB3 locus). For the PDB osteoclastic phenotype characterization and to assess the potential synergistic effect of rs1561570 with SQSTM1/P392L mutation in osteoclastogenesis, we studied a patient heterozygous for rs1561570 (CT genotype) and also carrier of the P392L mutation in SQSTM1. Phenotype assessment, DNA and RNA extraction methods are detailed in the S1 File.
To analyse if rs1561570 was affecting the methylation status at that particular location we used the MethPrimer (http://www.urogene.org/methprimer/), an in silico tool used to predict CpG island by examining the CG dinucleotides and to design specific primers to analyse those CpG islands (S1 Table).
Methylation analysis by bisulfite conversion and sanger sequencing
Sodium bisulfite treatment was performed on 1.5 μg of genomic DNA sample using the EpiMark® Bisulfite Conversion Kit  (New England Biolabs Inc., Canada) following the manufacturers’ standard protocol. Further details are provided in the S1 File.
Cell culture conditions
The U937 (Human monocytes/lymphoma) and U2OS (Human bone osteosarcoma) cell lines were grown in Dulbecco's modified eagle medium (DMEM), supplemented with 10% fetal bovine serum (FBS), 2mM L-glutamine and 1% penicillin/streptomycin (P/S). The T47D (Human breast carcinoma) cell line was grown in RPMI-1640 medium supplemented with 10% FBS, and 1% P/S. The MG63 (Human bone osteosarcoma) cell line was grown in αMEM supplemented with 10% FBS, and 1% P/S. All cell lines were incubated at 37°C in a humidified atmosphere containing 5% CO2. The medium, FBS, antibiotics and glutamine were obtained from Invitrogen, Portugal. Cell lines were obtained from American Type Culture Collection (ATCC).
Demethylating treatment and immunofluorescence
The localization of NF-κB after treatment with 5-Azacitidine (5-Aza, Sigma-Aldrich, Portugal) for 72h in T47D and U937 cell lines was assessed by immunofluorescence. NF-κB localization was also assessed by immunofluorescence in osteoclasts derived from Peripheral Blood Mononuclear Cells (PBMCs) from healthy donors and PDB patients with all studied genotypes. The immunofluorescence protocol is described in the S1 File.
Preparation of human in vitro-differentiated mature osteoclasts
Human mature osteoclasts were differentiated in vitro using mononuclear cells from blood of healthy controls and PDB patients. PBMCs were obtained by density gradient centrifugation using Ficoll-Hypaque. The cells were resuspended (3x106 cells/mL) in OPTI-MEM containing 10% FBS. The cell suspension was added to 6-well plates (9x106 cells/well) and to Lab-Tek 8 well-slides (3-6x105 cells/well). After 24h, the cells were washed thoroughly and lymphocyte-free adherent cells were incubated for three weeks with M-CSF (25 ng/ml, Life technologies, Canada) and RANKL (30 ng/ml, Fitzgerald, Canada).
Osteoclast morphology assessment, TRAP assay and in vitro assessment of bone resorption
Mature osteoclast formation was evaluated by quantification of TRAP-positive multinuclear cells using an acid phosphatase kit (Sigma-Aldrich, Canada), according to the protocols provided by the manufacturer. The bone resorption activity of osteoclasts was tested in vitro, using a previously validated model . The details of the TRAP assay and the in vitro assessment of bone resorption are provided in the S1 File.
Quantitative real-time PCR
In order to test if the most strongly PDB associated OPTN variant had an impact on its gene expression, we performed qPCR as previously described . The details of this analysis are provided in the S1 File.
Western blot analysis
Osteoclasts derived from patients and controls PBMCs were washed once in PBS and lysed using Trizol. The protein concentrations were determined using the Bradford reagent (Bio-Rad, Canada). Proteins were separated by 8% SDS-polyacrylamide gels and transferred onto PVDF membranes (Life technologies, Canada). After transfer, the membranes were blocked in 0.1% TBS/Tween 20 containing 5% nonfat dried milk at room temperature for 1 h. The membranes were then incubated overnight at 4°C with anti-OPTN (1:500, #100000 Cayman, Canada), anti-LC3B (1:1000, #2775 Cell Signaling, Canada) or anti-αTubulin (1:2000, #2144 Cell Signaling, Canada) in 0.1% TBS/Tween 20, followed by extensive washing using 0.1% TBS/Tween 20 and an incubation with HRP-conjugated secondary antibody (1: 2000, Cell signaling, Canada) in 0.1% TBS/Tween 20 during 1h, at room temperature. After extensive washing with 0.1% TBS/Tween 20, specific proteins were detected using a chemiluminescence kit (GE Healthcare, Canada). The densitometric analysis was performed using ImageJ software.
Statistical analysis was performed with GraphPad Prism 5 (GraphPad, La Jolla, CA). Comparisons between two groups were made using a two-tailed Student’s t-test. For comparisons between multiple groups, one-way ANOVA followed by Tukey’s post-hoc test was used. Differences were considered statistically significant when p<0.05.
Rs1561570 increases in vitro osteoclast differentiation and bone resorption activity
To better assess the effect of rs1561570 on osteoclasts formation, we performed a TRAP assay in osteoclasts derived from the PBMCs of patients carrying the rs1561570 C or the T allele. The presence of a haplotype with at least one T allele for rs1561570 increased significantly osteoclast formation, attested by the mean number of multinucleated cells yielded from PBMCs cultures, when compared to healthy controls (Controls CC 10%, Patients CC 50%, Patients CT/TT 80%, p-value = 0.006) (Fig 1A and 1C). The osteoclasts generated displayed a higher mean number of nuclei than the osteoclasts from healthy controls (Controls CC mean = 3, Patients CC mean = 5, Patients CT/TT mean = 10, p-value = 0.002) (Fig 1B and 1C).
(A) The osteoclast differentiation rate (displayed as a ratio between multinucleated osteoclasts (Multi OCL) and TRAP positive (+) cells) was higher in patients with PDB carrying at least one T allele (Patients CT/TT vs Controls CC p-value < 0.01; Patients CT/TT vs Patients CC p-value < 0.05). (B) Osteoclasts from patients carrying at least one T allele contained significantly more nuclei than osteoclasts from healthy controls (Patients CT/TT vs Controls CC p-value < 0.001; Patients CT/TT vs Patients CC p-value < 0.01). (C) The images are representative of the distribution of nuclei number observed in osteoclasts generated from PBMCs derived from healthy controls and PDB patients that were stained for TRAP and counterstained with haematoxylin. (ANOVA, * represents a p-value < 0.05, ** represents a p-value < 0.01, *** represents a p-value < 0.001, **** represents a p-value < 0.0001; non-mutated controls with CC genotype (n = 3), non-mutated patients with CC genotype (n = 5) and patients carrying at least one T allele (n = 3).
Accordingly, bone resorption was also significantly increased in osteoclast cultures generated from patients with a haplotype carrying at least one T allele for rs1561570, resulting in 3-fold increase in the total bone resorption area when compared to cultures prepared from healthy controls (p-value < 0.001) (Fig 2A and 2B) and 1.4-fold increase when compared to cell cultures prepared from PDB patients carrying two C alleles (p-value = 0.05). We also analysed the bone resorption area from osteoclast cultures of one patient carrying a haplotype with one T allele (CT) in combination with the SQSTM1/P392L mutation. Interestingly, there was an increase of the bone resorption area in the patient with one T allele, suggesting that there might be a synergistic effect of OPTN and SQSTM1 genetic variants in PDB (Fig 2B).
(A) Representative images of in vitro bone resorption assays (left) with (B) quantification of the results (right) by using ImageJ. The bone resorption area was higher in patients with PDB carrying at least one T allele (Patients CT/TT vs Controls CC p-value < 0.001; Patients CT/TT vs Patients CC p-value < 0.01). This result was even more evident in one PDB patient carrying one T allele (CT genotype) plus the SQSTM1/P392L mutation (p-value < 0.001). At least three different wells per patient were analysed. (ANOVA, * represents a p-value < 0.05, ** represents a p-value < 0.01, *** represents a p-value < 0.001, **** represents a p-value < 0.0001; non-mutated controls with CC genotype (n = 3), non-mutated patients with CC genotype (n = 5) and patients carrying at least one T allele (n = 3)).
Rs1561570 does not affect autophagy
OPTN has been suggested to be an autophagy adaptor and since impairment in autophagy was reported in PDB, we assessed the autophagy process in patients with a haplotype carrying at least one rs1561570 T allele. By immunofluorescence and western blot analysis we found that LC3BII expression was impaired, however that effect did not seem to be dependent of the presence of a haplotype with the T allele (S1A and S1B Fig), suggesting that the autophagy impairment may be independent from the rs1561570 genotype.
Rs1561570 causes a change in methylation and increases OPTN expression
Since our previous study has shown a strong association between rs1561570 (located in OPTN intron 7) and PDB , we determined the functional effect of rs1561570 in that disease by employing the splice site prediction algorithm, and showed that the allele harbouring a C was likely to create a potential new acceptor site for splicing and disrupt a potential branch point (consensus value for mutant sequence = 69.7%) (S2 Table). To validate this finding, we performed RT-PCR and qPCR using cDNA samples from patients and healthy controls with the three genotypes. Retention of intronic sequences was not observed in the genotypes analysed (results not shown).
We then studied the predicted impact of rs1561570 in the OPTN methylation status using the Methprimer tool. We found that the allele C was predicted to form a CpG dinucleotide, creating a methylation site, which was not predicted in the presence of the T allele (Fig 3A). After bisulfite treatment, we confirmed that all samples with at least one C allele in their rs1561570 genotype were in fact methylated (Fig 3B). To correlate the methylation site created by the C allele with the OPTN transcriptional levels, we performed qPCR (Fig 3C) and western blot (Fig 3D) analysis using lymphocytes from patients with the three genotypes. Samples carrying a haplotype with the T allele (CT or TT) showed levels of OPTN gene and protein expression higher than samples carrying the CC genotype. This result suggested that the methylated C allele is associated with a decrease in OPTN gene expression levels. The protein expression levels were also measured by western blot analysis in PBMC-derived osteoclasts and the results were in accordance with those obtained by qPCR and western blot analysis in lymphocytes (Fig 3E).
(A) In silico prediction using Methprimer tool showing a methylation site only in the presence of the allele C (not the allele T). (B) Sequencing results showing that rs1561570 C allele after bisulfite treatment remains a C (arrow head), while other C nucleotides that are not methylated change to T (asterisk). (C) Analysis of OPTN gene expression in lymphocytes from several healthy controls (CC genotype n = 10, CT/TT genotypes n = 6) and PDB patients (CC genotype n = 10, CT/TT genotypes n = 22) with the three genotypes. The levels of OPTN gene expression were measured by qPCR normalized to PPIB gene expression levels. Values are the mean of two replicates in each experiment and the experiment was repeated at least three independent times. (D) Quantification and (E) representative blots of OPTN protein expression in lymphocytes from non-mutated controls with CC genotype (n = 5), and patients carrying at least one T allele (n = 19). (F) Quantification and (G) representative blots of OPTN protein expression in osteoclasts from non-mutated controls with CC genotype (n = 3), non-mutated patients with CC genotype (n = 5) and patients carrying at least one T allele (n = 3). The levels of OPTN protein expression were measured by western blot analysis and related to levels of α-Tubulin (t-test and ANOVA, * represents a p-value < 0.05). The figures are representative of all the western blot analyses performed (n = 3).
Rs1561570 increases NF-κB translocation into the nucleus
To determine the role of rs1561570 on the osteoclastic phenotype, and since there are previous publications describing OPTN as an important protein in NF-κB pathway , we analysed the NF-κB cellular localization by immunofluorescence in PBMC-derived osteoclasts from PDB patients carrying different genotypes of rs1561570 and from healthy controls. Our results showed NF-κB translocation to the nucleus (showed as a ratio between co-localization of NF-κB and DAPI) is increased in patients with genotype CT and TT (Fig 4A). To confirm that the presence of NF-κB in the nucleus was affecting the expression of its target genes, we measured the expression of NF-κB, IL-6, ELK1, TRAP and NFATc1 expression. Our results showed an increased expression of NF-κB target genes related with osteoclastogenesis, such as IL-6, ELK1, TRAP and NFATc1 (Fig 4B)–thus providing additional evidence that an increase of OPTN expression observed in the presence of the T allele in the OPTN methylation status leads to an increase of NF-κB translocation into the nucleus and an increase of its target genes expression.
(A) NF-κB localization in osteoclasts derived from healthy controls and patients PBCMs showing an increase of the translocation of NF-κB into the nucleus, in patients with genotype CT and TT. This was also confirmed by co-localization ratio between NF-κB and DAPI–the nuclear staining. Five different fields of view were analysed per sample and at least three different samples were analysed per patient. Co-localization coefficient was calculated by Volocity software as the ratio of NF-κB staining related to DAPI. (B) Analysis of NFATc1, IL6, TRAP and ELK1 gene expression in several healthy controls (n = 16) and PDB patients (CC n = 10, CT/TT n = 22) with all genotypes. The levels of NFATc1, IL6, TRAP and ELK1 expression were measured by qPCR related to levels of PPIB. Values are the mean of at least three independent replicates. (t-test, * represents a p-value < 0.05, ** represents a p-value < 0.01).
As a proof of concept, we also analysed the NF-κB cellular localization by immunofluorescence in two different cell lines which carry the rs1561570 C allele–T47D (CT genotype) and U937 (CC genotype)–following a treatment with a demethylating agent, 5-Azacitidine. We were able to confirm that, before the demethylating treatment, NF-κB was localized in the cytoplasm and in the nucleus. However, after OPTN demethylation, NF-κB localization was almost exclusively located in the nucleus (Fig 5A and 5B). To confirm the effect of the demethylating agent we measured OPTN expression and to confirm that the presence of NF-κB in the nucleus was also affecting the expression of its target genes (as in patients carrying the T allele), we measured the expression of NF-κB, IL-6, ELK1, TRAP and NFATc1. Our results showed that there was an increase both in OPTN expression, which was expected after the demethylation treatment, and in the expression of the NF-κB target genes analysed. NF-κB levels were maintained as expected (Fig 5C).
(A) NF-κB localization in U937 cells (CC genotype). (B) NF-κB localization in T47D cells (CT genotype), following 5-Azacitidine (5-Aza) treatment. After OPTN demethylation, NF-κB was translocated exclusively to the nucleus. (C) Analysis of NF-κB target genes expression. The levels of OPTN, NF-κB and NF-κB target genes (IL-6, ELK1, NFATc1) expression were measured by qPCR related to levels of GAPDH gene. Values are the mean of at least three independent replicates. (t-test, * represents a p-value < 0.05, ** represents a p-value < 0.01 *** represents a p-value < 0.001, **** represents a p-value < 0.0001).
In addition we performed the same procedure using two osteosarcoma cell lines–U2OS (CC genotype) and MG63 (TT genotype) and the results showed that the demethylating agent increased OPTN and NF-κB target genes in U2OS while NF-κB expression was maintained (S2A Fig), suggesting that a change in the methylation status due to the effect of the demethylating agent had an effect in OPTN and in NF-κB target genes expression. In contrast, in the MG63 cell line this treatment did not have any effect in the expression of OPTN or NF-κB target genes (S2B Fig), suggesting that the effect in OPTN expression might be specific for the cells carrying one C allele, thus supporting our hypothesis.
Rs1561570 may have an effect on PDB severity in the presence of SQSTM1 mutations
By comparing the clinical phenotype of patients with PDB carriers of the CC genotype of OPTN rs1561570 with patients carrying one or two T alleles, we observed a tendency, although not significant, for an: i) increase of total serum ALP (sALP) levels, ii) increase of mean number of affected bones, and iii) increase of disease extension measured by Renier’s index–related with the number of T alleles in the genotype (S3 Table). These results suggest that rs1561570 could have a dose effect depending on the number of T alleles present in the genotype. In fact, in the presence of the P392L/SQSTM1 mutation and rs1561570 T allele our results showed an increased tendency for i) higher levels of total sALP, ii) higher risk for polyostotic involvement, iii) higher number of affected bones and iv) an increase in the disease extension/Renier’s index (S4 Table) with the increase in the number of T alleles. Altogether, these results may suggest a modifier effect of OPTN common variant in the presence of SQSTM1/P392L mutation in patients with PDB, which could lead to an increase in the disease activity and extent, and therefore should be replicated in independent cohorts with higher number of participants.
In the present study, we investigated the functional role of the most significantly PDB associated genetic variant of OPTN- rs1561570 (C>T). Previously, we reported that the C allele was more frequent in controls in the French-Canadian population (C allele frequency = 52%, T allele frequency = 48%) while the T allele was more frequent in PDB patients (C allele frequency = 36%, T allele frequency = 64%) . Using in silico tools, we found that the rs1561570 T allele was predicted to cause the loss of a methylation site. With bisulfite treatment, we confirmed that the C allele in that position was always methylated. Furthermore, both qPCR and western blot analysis correlated the presence of the C allele with a decrease of OPTN expression levels. The existence of a correlation between the hypermethylation status of a given gene, resulting in a decrease in gene transcription levels affecting a related disease, has been previously reported in the literature [31–33]. Our immunofluorescence analysis also allowed us to confirm that an increase in OPTN levels (due to the demethylation treatment in cells and the presence of at least one T allele in PDB patients—dominant genetic effect) is related to an increase in NF-κB translocation into the nucleus, confirming that a decrease in OPTN methylation (as seen in PDB patients with rs1561570 T allele) increases the translocation of NF-κB into the nucleus, and that is related to an increased expression of NF-κB target genes related with osteoclastogenesis, such as NFATc1 –a master transcription factor of osteoclast differentiation , IL6—an important cytokine for osteoclastogenesis  and ELK1—a relevant transcription factor in mature osteoclasts . We also showed that the autophagy process is impaired in PDB patients and that it is not dependent on the rs1561570 genetic status. Therefore, we can hypothesize that in healthy individuals, the autophagy process works normally both in the presence of the C allele (when the OPTN expression is controlled at basal levels), and in the presence of the T allele (when the OPTN expression is potentiated due to the loss of a methylation site). In the latter, the efficient autophagy process signals the excessive OPTN for degradation by ubiquitination and recycling of the protein, and thus OPTN expression is reduced to a basal level. In both situations, this will result in normal levels of NF-κB in the nucleus and physiological osteoclastogenesis rate. In PDB patients, where the autophagy process is impaired, in the presence of the rs1561570 C allele, the PDB phenotype might be explained by the presence of other factors, such as other genetic variants or environmental triggers that will lead to an increase in osteoclastogenesis. The effect of rs1561570 T allele in PDB patients may therefore be explained by the increase in OPTN expression associated with a decrease in its degradation due to the autophagy defect present in these patients. Consequently, there is an accumulation of OPTN that promotes NF-κB translocation to the nucleus at a much higher rate than normal, thus also promoting osteoclastogenesis and that in addition to the existence of other genetic variants or environmental triggers potentiate the clinical phenotype of PDB in these patients.
Journo and its co-workers had already described the translocation of NF-κB into the nucleus due to an increase in OPTN expression , but upon a virus infection. OPTN is part of a complex, which also includes TAX1 and TAX1BP1, that can activate the IKK complex by ubiquitination, thus leading to the degradation of the NF-κB inhibitor and subsequent release of NF-κB into the nucleus. In the present work, the change in OPTN expression levels caused by one single change in nucleotide may be explained by the alteration of the already reported binding of an enhancer—JUND, a member of the AP1 family that is present in osteoclasts [37–39]—in the genomic region containing the T allele, which is lost in the presence of the C allele, as confirmed by the ENCODE project. The ENCODE project shows regions of transcription factor binding, derived from a large collection of ChIP-seq experiments, together with DNA binding motifs identified within these regions by the ENCODE Factor book repository. Indeed, due to the resulting change in DNA conformation in the presence of the T allele, the loss of a JUND binding site around the rs1561570 region of these patients will result in an alteration in OPTN expression levels, promoting an increase in osteoclastogenesis and the development of PDB. Our results are in contrast with data from Obaid et al. (2015) who, based in the description by Zeller and Westra’s groups (who have shown that rs1561570 T allele was related to lower levels of OPTN expression [40,41]), analysed the effect of a lack of OPTN in the differentiation of osteoclasts using a mouse knock out for OPTN. Those authors concluded that depletion of OPTN in that model increased the differentiation of osteoclasts (p-value = 0.002), suggesting that OPTN might be a negative regulator of osteoclast differentiation . However, and, unlike in humans, these mice did not develop a PDB-like phenotype ,. Interestingly, in the same study, authors also refer that OPTN expression was significantly increased during days 2 to 5 of in vitro osteoclast differentiation using primary bone-marrow-derived macrophages, in which they knocked down OPTN by using a small hairpin RNA. In addition, it was found in another set of in vivo works, using either knock-in mice expressing a point mutation in OPTN and abolishing its polyubiquitin binding capacity , or mice lacking either the entire C-terminal ubiquitin-binding  or the N-terminal TANK-binding kinase 1 (TBK1)-binding domain , that mice had retained normal NF-κB responses. These results argued against the fact that a decrease per se in OPTN or in OPTN function could promote osteoclastogenesis by affecting NF-κB signalling pathways.
As referred in a recent review, OPTN was thought to be important for NF-κB signalling because of its strong homology to the NF-κB essential modulator, NEMO—the core element of the inhibitor of NF-κB kinase (IKK) complex -, that is essential for NF-κB activation . Altogether, the lack of a PDB-like phenotype developed in those studies is not in favour of a lack of OPTN expression being the causing factor. As suggested in the same review, it is more likely that the PDB phenotype is the result of an aberrant expression of OPTN, including a possible gain of function  which is in agreement with our results. Our study is unique since it analyses, using in vitro and ex vivo techniques, the single effect of rs1561570 in OPTN gene and protein expression in several samples of patients with PDB non carriers of a SQSTM1/P392L mutation, and in different cell types (lymphocytes and osteoclasts derived from patients PBMCs), thus providing, at several levels, clear evidences supporting our hypothesis. Our finding that, in one patient carrying the SQSTM1/P392L mutation, there was a tendency for a more severe clinical phenotype in the presence of a haplotype with the rs1561570 T allele and that osteoclasts from that patient had a higher in vitro bone resorption rate, suggests that there might be a synergistic effect but this should be further investigated with additional patients.
Altogether, the gathered evidence obtained from gene to protein analysis, from in vitro to patient samples, from genetic to morphological cellular features, show that OPTN appears to be important for NF-κB translocation into the nucleus, activation of its target genes and consequent increase in osteoclastogenesis, providing relevant information towards a better understanding on how genetic variants in OPTN can contribute to PDB by acting as modifiers and complementing previous data on the potential functional role of variants in OPTN and how they may contribute to affect the physiological regulatory mechanisms of this gene .
In conclusion, we have shown that rs1561570 is responsible for a change in the methylation status of OPTN in patients with PDB and that it has an impact in its expression levels, which may contribute to PDB pathogenesis by affecting NF-κB translocation to the nucleus. Accordingly, to this date, OPTN remains the most likely gene within PDB6 locus to contribute to PDB pathophysiology, possibly in combination with other genetic/environmental factors yet unknown.
S1 Fig. rs1561570 effect in autophagy.
(a) LC3BII foci (red) in PBMC-derived osteoclasts from controls (with genotype CC, n = 3) and PDB patients (non-mutated patients with CC genotype (n = 5) and patients carrying at least one T allele (n = 3)) were analysed by immunofluorescence. At least three different wells per patient were analysed. (b) The levels of LC3BII protein expression were measured by western blot analysis and related to levels of α-Tubulin (left). Quantification of the results was performed by using ImageJ (right). The figures are representative of all the western blot analyses performed. (ANOVA, * represents a p-value < 0.05).
S2 Fig. In vitro effect of OPTN demethylation in expression of NF-κB target genes in U2OS and MG63 cells.
The levels of OPTN, NF-κB and NF-κB target genes (IL-6, ELK1, NFATc1) expression were measured by qPCR related to levels of GAPDH gene in (a) U2OS (rs1561570 CC genotype) and (b) MG63 cells (rs1561570 TT genotype). Values are the mean of at least three independent replicates. (t-test, * represents a p-value < 0.05, ** represents a p-value < 0.01, *** represents a p-value < 0.001, **** represents a p-value < 0.0001).
S2 Table. Splicing effects predicted by human splicing finder.
S3 Table. Comparisons of main clinical characteristics between French-Canadian PDB patients carriers of the rs1561570 TT or TC genotype and the CC genotype carriers, considered as the reference.
All 232 patients were non-carrier of the P392L mutation within the SQSTM1 gene (PDB3 locus).
S4 Table. Comparisons of main clinical characteristics between French-Canadian PDB patients carriers of the rs1561570 TT or TC genotype and the CC genotype carriers, considered as the reference.
All 32 patients were carrier of the P392L mutation within the SQSTM1 gene (PDB3 locus).
The authors would like to acknowledge the CHU de Québec-Université Laval Research Centre (CHUL) Gene Expression Platform, Quebec, Canada for the quantitative real-time PCR measurements and the Centre of Marine Sciences (CCMAR), Faro, Portugal for the use of the sequencing platform.
- 1. Altman RD, Bloch DA, Hochberg MC, Murphy WA. Prevalence of pelvic Paget’s disease of bone in the United States. J Bone Miner Res. 2000;15: 461–5. pmid:10750560
- 2. van Staa TP, Selby P, Leufkens HGM, Lyles K, Sprafka JM, Cooper C. Incidence and natural history of Paget’s disease of bone in England and Wales. J Bone Miner Res. 2002;17: 465–71. pmid:11878305
- 3. Morales Piga AA, Rey Rey JS, Corres Gonzales J, Garcia Sagredo JM, Lopez Abente G. Frequency and Characteristics of Familial Aggregation of Paget ‘ s Disease of Bone. J Bone Miner Res. 1995;10: 663–670. pmid:7610939
- 4. Hocking L, Slee F, Haslam S., Cundy T, Nicholson G, van Hul W, et al. Familial Paget’s disease of bone: patterns of inheritance and frequency of linkage to chromosome 18q. Bone. 2000;26: 577–580. pmid:10831928
- 5. Laurin N, Brown JP, Lemainque A, Duchesne A, Huot D, Lacourcière Y, et al. Paget disease of bone: mapping of two loci at 5q35-qter and 5q31. Am J Hum Genet. 2001;69: 528–43. pmid:11473345
- 6. Fotino M, Haymovits A, Falk CT. Evidence for linkage between HLA and Paget’s disease. Transplant Proc. 1977;9: 1867–8. pmid:146288
- 7. Haslam SI, Van Hul W, Morales-Piga A, Balemans W, San-Millan JL, Nakatsuka K, et al. Paget’s disease of bone: evidence for a susceptibility locus on chromosome 18q and for genetic heterogeneity. J Bone Miner Res. 1998;13: 911–7. pmid:9626621
- 8. Good D, Busfield F, Fletcher BH, Duffy DL, Kesting JB, Andersen J, et al. Linkage of Paget disease of bone to a novel region on human chromosome 18q23. Am J Hum Genet. 2002;70: 517–525. pmid:11742440
- 9. Hocking LJ, Herbert CA, Nicholls RK, Williams F, Bennett ST, Cundy T, et al. Genomewide search in familial Paget disease of bone shows evidence of genetic heterogeneity with candidate loci on chromosomes 2q36, 10p13, and 5q35. Am J Hum Genet. 2001;69: 1055–61. pmid:11555792
- 10. Lucas GJA, Riches PL, Hocking LJ, Cundy T, Nicholson GC, Walsh JP, et al. Identification of a major locus for Paget’s disease on chromosome 10p13 in families of british descent. J Bone Miner Res. 2008;23: 58–63. pmid:17907922
- 11. Albagha OME, Visconti MR, Alonso N, Langston AL, Cundy T, Dargie R, et al. Genome-wide association study identifies variants at CSF1, OPTN and TNFRSF11A as genetic risk factors for Paget’s disease of bone. Nat Genet. 2010;42: 520–4. pmid:20436471
- 12. Michou L, Conceição N, Morissette J, Gagnon E, Miltenberger-Miltenyi G, Siris ES, et al. Genetic association study of UCMA/GRP and OPTN genes (PDB6 locus) with Paget’s disease of bone. Bone. Elsevier Inc.; 2012;51: 720–8.
- 13. Rezaie T, Sarfarazi M. Molecular cloning, genomic structure, and protein characterization of mouse optineurin. Genomics. 2005;85: 131–8. pmid:15607428
- 14. Osawa T, Mizuno Y, Fujita Y, Takatama M, Nakazato Y, Okamoto K. Optineurin in neurodegenerative diseases. Neuropathology. 2011;31: 569–74. pmid:21284751
- 15. Kachaner D, Génin P, Laplantine E, Weil R. Toward an integrative view of Optineurin functions. Cell Cycle. 2012;11: 2808–2818. pmid:22801549
- 16. Shen WC, Li HY, Chen GC, Chern Y, Tu PH. Mutations in the ubiquitin-binding domain of OPTN/optineurin interfere with autophagy- mediated degradation of misfolded proteins by a dominant-negative mechanism. Autophagy. 2015;11: 685–700. pmid:25484089
- 17. Slowicka K, Vereecke L, McGuire C, Sze M, Maelfait J, Kolpe A, et al. Optineurin deficiency in mice is associated with increased sensitivity to Salmonella but does not affect proinflammatory NF-κB signaling. Eur J Immunol. 2016;46: 971–980. pmid:26677802
- 18. Slowicka K, Vereecke L, van Loo G. Cellular Functions of Optineurin in Health and Disease. Trends Immunol. Elsevier Ltd; 2016;37: 621–633. pmid:27480243
- 19. Wong YC, Holzbaur ELF. Optineurin is an autophagy receptor for damaged mitochondria in parkin-mediated mitophagy that is disrupted by an ALS-linked mutation. Proc Natl Acad Sci U S A. 2014;111: E4439–48. pmid:25294927
- 20. Wild P, Farhan H, McEwan DG, Wagner S, Rogov VV, Brady NR, et al. Phosphorylation of the autophagy receptor optineurin restricts Salmonella growth. Science. 2011;333: 228–33. pmid:21617041
- 21. Korac J, Schaeffer V, Kovacevic I, Clement AM, Jungblut B, Behl C, et al. Ubiquitin-independent function of optineurin in autophagic clearance of protein aggregates. J Cell Sci. 2012;126: 580–592. pmid:23178947
- 22. Nagabhushana A, Bansal M, Swarup G. Optineurin is required for CYLD-dependent inhibition of TNFα-induced NF-κB activation. PLoS One. 2011;6: e17477. pmid:21408173
- 23. Swarup G, Vaibhava V, Nagabhushana A. Functional Defects Caused by Glaucoma–Associated Mutations in Optineurin. In: Rumelt S, editor. Glaucoma—Basic and Clinical Aspects. InTech; 2013.
- 24. Obaid R, Wani SE, Azfer A, Hurd T, Jones R, Cohen P, et al. Optineurin Negatively Regulates Osteoclast Differentiation by Modulating NF-κB and Interferon Signaling: Implications for Paget’s Disease. Cell Rep. The Authors; 2015; 1–7.
- 25. Journo C, Filipe J, About F, Chevalier SA, Afonso PV, Brady JN, et al. NRP/Optineurin Cooperates with TAX1BP1 to potentiate the activation of NF-kappaB by human T-lymphotropic virus type 1 tax protein. PLoS Pathog. 2009;5: e1000521. pmid:19609363
- 26. Harhaj EW, Sun SC. IKKgamma serves as a docking subunit of the IkappaB kinase (IKK) and mediates interaction of IKK with the human T-cell leukemia virus Tax protein. J Biol Chem. 1999;274: 22911–4. pmid:10438454
- 27. Xiao G, Cvijic ME, Fong A, Harhaj EW, Uhlik MT, Waterfield M, et al. Retroviral oncoprotein Tax induces processing of NF-kappaB2/p100 in T cells: evidence for the involvement of IKKalpha. EMBO J. 2001;20: 6805–6815. pmid:11726516
- 28. Nasr R, Chiari E, El-Sabban M, Mahieux R, Kfoury Y, Abdulhay M, et al. Tax ubiquitylation and sumoylation control critical cytoplasmic and nuclear steps of NF-kB activation. Blood. 2006;107: 4021–4029. pmid:16424386
- 29. Hernández HG, Tse MY, Pang SC, Arboleda H, Forero DA. Optimizing methodologies for PCR-based DNA methylation analysis. Biotechniques. 2013;55: 181–197. pmid:24107250
- 30. Itzstein C, Espinosa L, Delmas PD, Chenu C. Specific antagonists of NMDA receptors prevent osteoclast sealing zone formation required for bone resorption. Biochem Biophys Res Commun. 2000;268: 201–209. pmid:10652236
- 31. Iwamoto K, Bundo M, Ueda J, Oldham MC, Ukai W, Hashimoto E, et al. Neurons show distinctive DNA methylation profile and higher interindividual variations compared with non-neurons. Genome Res. 2011;21: 688–96. pmid:21467265
- 32. Brenet F, Moh M, Funk P, Feierstein E, Viale AJ, Socci ND, et al. DNA methylation of the first exon is tightly linked to transcriptional silencing. PLoS One. 2011;6: e14524. pmid:21267076
- 33. Hrašovec S, Hauptman N, Glavač D, Jelenc F, Ravnik-Glavač M. TMEM25 is a candidate biomarker methylated and down-regulated in colorectal cancer. Dis Markers. 2013;34: 93–104. pmid:23324576
- 34. Lorenzo J, Horowitz M, Choi Y, Schett G, Takayanagi H. Osteoimmunology—Interactions of the Immune and Skeletal Systems. Choi Y, editor. Springer; 2007.
- 35. Roodman G, Kurihara N, Ohsaki Y, Kukita A, Hosking D, Demulder A, et al. Interleukin 6. A potential autocrine/paracrine factor in Paget’s disease of bone. J Clin Invest. 1992;89: 46–52. pmid:1729280
- 36. Yao GQ, Itokawa T, Paliwal I, Insogna K. CSF-1 induces fos gene transcription and activates the transcription factor Elk-1 in mature osteoclasts. Calcif Tissue Int. 2005;76: 371–378. pmid:15812575
- 37. Inoue D, Santiago P, Horne WC, Baron R. Identification of an osteoclast transcription factor that binds to the human T cell leukemia virus type I-long terminal repeat enhancer element. J Biol Chem. 1997;272: 25386–93. pmid:9312160
- 38. David JP, Rincon M, Neff L, Horne WC, Baron R. Carbonic anhydrase II is an AP-1 target gene in osteoclasts. J Cell Physiol. 2001;188: 89–97. pmid:11382925
- 39. Beranger GE, Momier D, Guigonis J-M, Samson M, Carle GF, Scimeca J-C. Differential binding of poly(ADP-Ribose) polymerase-1 and JunD/Fra2 accounts for RANKL-induced Tcirg1 gene expression during osteoclastogenesis. J Bone Miner Res. 2007;22: 975–983. pmid:17419679
- 40. Zeller T, Wild P, Szymczak S, Rotival M, Schillert A, Castagne R, et al. Genetics and Beyond–The Transcriptome of Human Monocytes and Disease Susceptibility. 2010;5.
- 41. Westra H-J, Peters MJ, Esko T, Yaghootkar H, Schurmann C, Kettunen J, et al. Systematic identification of trans-eQTLs as putative drivers of known disease associations. Nat Genet. 2013;45: 1238–1243. pmid:24013639
- 42. Gleason CE, Ordureau A, Gourlay R, Arthur JSC, Cohen P. Polyubiquitin binding to optineurin is required for optimal activation of TANK-binding kinase 1 and production of interferon β. J Biol Chem. 2011;286: 35663–74. pmid:21862579
- 43. Munitic I, Giardino Torchia ML, Meena NP, Zhu G, Li CC, Ashwell JD. Optineurin Insufficiency Impairs IRF3 but Not NF- B Activation in Immune Cells. J Immunol. 2013;191: 6231–6240. pmid:24244017
- 44. Meena N, Zhu G, Mittelstadt P, Giardino Torchia M, Pourcelot M, Arnoult D, et al. The TBK1-binding domain of optineurin promotes type I interferon responses. FEBS Lett. 2016;590: 1498–1508. pmid:27086836
- 45. Silva IAL, Conceição N, Gagnon E, Caiado H, Brown JP, Gianfrancesco F, et al. Effect of genetic variants of OPTN in the pathophysiology of Paget's disease of bone. Biochim Biophys Acta. 2018;1864: 143–151. pmid:28993189