Abstract
Bone differentiation defects have been recently tied to Wnt signaling alterations occurring in vitro and in vivo Gaucher disease (GD) models. In this work, we provide evidence that the Wnt signaling multi-domain intracellular transducers Dishevelled 1 and 2 (DVL1 and DVL2) may be potential upstream targets of impaired beta glucosidase (GBA1) activity by showing their misexpression in different type 1 GD in vitro models. We also show that in Gba mutant fish a miR-221 upregulation is associated with reduced dvl2 expression levels and that in type I Gaucher patients single-nucleotide variants in the DVL2 3′ untranslated region are related to variable canonical Wnt pathway activity. Thus, we strengthen the recently outlined relation between bone differentiation defects and Wnt/β-catenin dysregulation in type I GD and further propose novel mechanistic insights of the Wnt pathway impairment caused by glucocerebrosidase loss of function.
Introduction
Gaucher disease (GD; OMIM #230800, #230900, #2301000) is the most frequent lysosomal disorder characterized by defective glucocerebrosidase activity and progressive glucosylceramide accumulation (1). Among the three traditionally classified forms, the so-called non-neuronopathic type 1 form (OMIM #230800) is the most frequent, accounting for 90% of all described cases (2). Affected patients suffer from many visceral symptoms and peripheral abnormalities, including hematological manifestations and skeletal defects, which have been commonly attributed to progressive infiltration of macrophage-like cells (Gaucher cells) in the bone marrow (BM). In the past few years, an extensive description of in vitro and in vivo GD experimental models has raised important concerns about the etiology of skeletal alterations in GD patients (3). In particular, more accumulating evidences have disclosed a complex network of cellular pathways and pathogenic mechanisms that may in part explain the extremely heterogeneous range of severity and bone phenotypes (4). Based on our preliminary report and according to the results of an independent study, it has been recently suggested that glucocerebrosidase deficiency may trigger a cascade of molecular defects that may impinge on the canonical Wnt pathway (Wnt/β-catenin) and lead to reduced osteoblast differentiation (5,6).
The Wnt/β-catenin pathway is a complex developmental pathway characterized by the activation of multiple cytoplasmic transducers, following the binding of extracellular soluble factors (Wnt ligands) to target cell surface receptors (Frizzled and LRP5/6). Upon recruitment of members of the Dishevelled (DVLs) family at cell membrane receptors, the so-called destruction complex, composed of different proteins (APC, CK1, AXIN, GSK3β), is dismantled, thus preventing the progressive proteasomal degradation of β-catenin, the major cytoplasmic Wnt signaling transducer. Once relieved from the ‘destruction complex’, β-catenin is able to shuttle into the nucleus and associate with TCF/LEF transcription factors, which activate arrays of genes involved in different cellular processes, including cell proliferation, differentiation and epithelial-to-mesenchymal transition (7,8).
A large number of studies have emphasized the role of the Wnt/β-catenin pathway in the differentiation of osteoblasts from mesenchymal precursors and bone remodeling (9–11). Among these investigations, many of them have collected evidences on the role of extracellular secreted Wnt signaling-antagonizing factors (namely, SCLEROSTIN and DICKKOPF) and target canonical Wnt pathway transducers in the osteogenic process (12–14). Moreover, several miRNAs, which modulate Wnt/β-catenin signaling and promote osteoblast differentiation, have been so far identified, claiming the possibility of using miRNAs as an alternative therapeutic treatment for metabolic bone diseases (15).
In this study, we explored the potential involvement of the master Wnt signaling components, DVL1 and DVL2, in the canonical Wnt/β-catenin dysregulation occurring in different experimental models of GD. By using alternative molecular approaches, we found that in GD fibroblasts, as well in human osteoblasts, in which we chemically inhibited GBA1 activity, a quantitative decrease of DVL proteins occurs. Moreover, we identified reduced levels of EPSIN 1 and DACT-1 proteins, which bind to DVL2, in the same experimental models and described the potential dysregulation of DVL2-targeting miRNAs in the previously established zebrafish model of GD. Finally, we documented single-nucleotide variants (SNVs) in the 3′ untranslated region (UTR) region of DVL2 in GD patients exhibiting unexpected increased Wnt pathway activation, supporting a potential mechanism whereby potential fluctuations in post-transcriptional silencing of DVL2 may drive a differential Wnt pathway transduction. Altogether, this study provides a novel mechanistic explanation of the canonical Wnt pathway dysregulation in GD pathogenesis and suggests the targeting of DVLs as potential candidate for therapeutic approaches.
Results
The upstream canonical Wnt master regulators Dvl1 and Dvl2 are negatively affected in experimental GD models
We and others have previously shown that Wnt/β-catenin pathway activity is significantly reduced in experimental models of GD (5,6,16). To further extend our previous observations, we sought to test whether Dishevelled (DVLs) proteins, which are master Wnt signaling modulators and recruited in sphingolipid-enriched plasma membrane regions (‘lipid rafts’) (17,18), could be affected by glucocerebrosidase loss of function. Towards this aim, we performed multiple western blot analysis in cultured fibroblasts from GD1 patients and healthy control. As shown in Figure 1A and B, we found that both DVL1 and DVL2 protein levels were reduced in GD fibroblasts. While in agreement with these results, we found decreased total beta catenin levels in GD1 fibroblasts, we unexpectedly detected unchanged levels of AXIN1, which constitutes part of the β-catenin destruction complex (Supplementary Material, Fig. S1) (19). However, when assayed by immunoprecipitation, AXIN1 was found to bind more β-catenin in GD1 patients’ extracts rather than control ones, suggesting that in GD1 fibroblasts more AXIN1-dependent degradation of β-catenin may occur (Fig. 1C). We next evaluated by immunohistochemistry whether DVL1 and DVL2 protein distribution could be affected by GBA functional impairment. As shown in Fig. 1D and E, neither DVL1 nor DVL2 cellular distribution appeared significantly changed in GD cells when compared with controls. However, while for DVL2 we could confirm the overall significant decrease in GD1 fibroblasts, we were not able to detect the same degree of DVL1 reduction, when compared with controls.
The Wnt signaling upstream transducers DVL1 and DVL2 are dysregulated in type 1 GD patient fibroblasts. (A) Representative western blot analysis of total cell lysates showing the quantitative decrease of DVL1 and DVL2 in GD1 fibroblasts. (B) Bar graph representing the mean DVL1 and DVL2 protein levels assessed in control and Gaucher fibroblasts by three independent replicates (*P < 0.05, t-test). (C) Representative western blot analysis, showing the increased coimmunoprecipitation of β-CATENIN with AXIN1 occurring in total cell lysates from GD1 fibroblasts, when compared with control extracts. (D) Representative immunofluorescence analysis for LAMP1, DVL1 and DVL2 on control and GD1 fibroblasts. No apparent differences in DVL1 and DVL2 cellular distribution are detected. (E) Bar graphs depict the ImageJ-based quantitative analysis of the mean fluorescent intensity performed on 100 individual cells from control and GD1 fibroblasts for the indicated markers (**P = 0.0067, t-test).
To verify our preliminary data on an alternative experimental model, we chemically abrogated GBA1 activity, using the well-known cyclitol epoxide, conduritol-β-epoxide (CBE), which irreversibly inactivates GBA1 enzymatic activity, in commercially available human fetal osteoblasts (hFOB). As recently outlined in the paper by Kuo and colleagues, minimal working concentrations of CBE may reduce potential off-target effects (20). We, therefore, assessed the effect of CBE in our cellular model by testing a range of CBE concentrations (10–500 μM) and measured GBA activity. As shown in Supplementary Material, Figure S2A, we found that already at 10 μM CBE there was a marked reduction of GBA1 activity (about 80%) and increasing concentrations of CBE did not significantly change the extent of GBA enzymatic inhibition. We next exposed hFOB cells to sustained 10 μM CBE treatment for several days (up to 11 days) and harvested cell lysates for western blot analysis. As shown in Fig. 2A and B, we found that after 4 days of chronic CBE treatment, total beta catenin levels were significantly reduced, together with an almost significant decrease of DVL1 and DVL2 protein levels. The effect of CBE treatment on DVLs was also mirrored at a transcriptional level, although we could detect significant differences only 7 days after CBE treatment (Supplementary Material, Fig. S2B). Prolonged exposure to CBE did not induce the same decrease of Wnt-related target proteins (Supplementary Material, Fig. S3). From these results, we could conclude that DVL1 and DVL2 are negatively affected by GBA loss of function, although in a different experimental model (human osteoblasts) this effect seems transient.
The Wnt signaling upstream transducers DVL1 and DVL2 are negatively affected by transient CBE-mediated glucocerebrosidase inhibition in human osteoblasts. (A) Representative western blot analysis showing the decrease of DVL1 and DVL2 protein levels occurring after a treatment with 10 μM CBE. (B) Bar graphs showing the mean ± SD of three independent western blot replicates for the CBE-treated osteoblasts after 4 days and untreated control. Note that both β-CATENIN and SP7(OSTERIX) are also negatively affected by CBE (*P < 0.05, t-test).
The ubiquitin-binding adaptor protein EPSIN1 binds to DVL2 in human fibroblasts and is downregulated in GD cells
Considering that in human osteoblasts CBE treatment negatively affected DVL mRNA levels with a temporal delay (7 days after initiation of CBE administration) with respect to the decrease in protein levels (4 days after initiation of CBE administration), we suspected that post-transcriptional mechanisms could be responsible for the CBE-induced Wnt pathway dysregulation. EPSIN 1 proteins have been recently identified as binding partners of DVL2 and required for positive regulation of canonical Wnt signaling in colon cancer (21). We, therefore, first tested whether EPSIN1 could bind to DVL1 or DVL2 or both in our GD disease cellular model. As shown in Figure 3A, immunoprecipitation of protein extracts from control and GD fibroblasts demonstrated an interaction of DVL2 and EPSIN1, confirming previous observations. We next evaluated by western blot the levels of EPSIN1 in GD fibroblasts and control fibroblasts with or without a CBE treatment. As shown in Figure 3B–D, we found that EPSIN 1 was reduced in both experimental models, although in GD1 fibroblasts the decrease did not reach statistically significant differences. However, immunohistochemistry analysis for EPSIN1 demonstrated a significant decrease of EPSIN1 immunoreactivity and nuclear localization in GD1 cells, when compared with controls (Fig. 3E). Indeed, both western blots and immunohistochemistry analysis did not show any significant differences between control and CBE-treated osteoblasts. We, therefore, could conclude that a decrease in EPSIN1 protein levels could justify only in GD1 fibroblasts the reduced extent of DVL2 protein accumulation.
The ubiquitin-binding adaptor protein EPSIN 1 binds to DVL2 and is negatively affected by GBA1 loss of function. (A) Representative western blot for EPSIN1 after immunoprecipitation of endogenous DVL1 and DVL2 in control and GD1 fibroblasts. The image is representative of three independent experiments. (B and C) Representative western blot analysis of total cell lysates showing the quantitative decrease of EPSIN 1 in GD1 fibroblasts and fibroblasts treated with CBE. (D) Bar graph showing the quantification of multiple independent western blots for EPSIN for the reported samples. Results are the mean ± SD of four independent replicates (*P < 0.05, t-test). (E) Representative immunofluorescence analysis for EPSIN1 in control and GD1 fibroblasts, showing the decrease of EPSIN 1 immunolabeling and nuclear localization in GD1 cells (white arrowheads point to positive nuclear staining of EPSIN). Cells have been counterstained with DAPI. The bar graphs depict the ImageJ-based quantitative assessments of mean fluorescent intensity and colocalization for 100 individual cells per condition (***P = 0.0010, t-test).
Dapper1 (DACT-1), a molecular interactor of DVL2, is reduced in GD experimental models
Dapper1, originally identified by a yeast two-hybrid screen in Xenopus as an interacting protein of Dvl, has been controversially associated with a positive or negative regulation of DVL2 in mammalian models (22–24). To predict a potential functional relation between DACT-1 and DVL2 in our GD experimental models, we performed repeated western blot analysis in GD fibroblasts and in both human fibroblasts and osteoblasts under CBE treatment. As shown in Figure 4A, we found that DACT-1 was reduced in both GD1 fibroblasts and fibroblasts under CBE treatment. The same result was confirmed by immunohistochemistry in fixed pools of GD1 and control cells (Fig. 4B). Indeed, no apparent differences were seen by western blots in osteoblasts undergoing CBE treatment for several days (Supplementary Material, Fig. S4), although immunofluorescence performed on the same cells and control ones demonstrated a significant decrease of DACT-1 immunopositivity. These results could enable us to conclude that at least in GD1 fibroblasts the decrease of DVL2 could be attributed to reduced DACT-1 levels.
The Dishevelled binding protein DACT-1 is dysregulated by GBA1 loss of function. (A) Representative western blot analysis of total cell lysates showing the quantitative decrease of DACT1 in GD1 fibroblasts and fibroblasts under CBE exposure compared with control cells. The bar graphs indicate the quantitative degree of DACT-1 decrease as assessed by the mean ± SD of four independent western blots. (B) Representative immunofluorescence analysis for DACT-1 on control and GD1 fibroblasts. Cells have been counterstained with DAPI. The bar graph shows the ImageJ-based quantitative decrease of DACT-1 nuclear localization in GD1 cells. (**P = 0.0068, t-test). (C) Representative immunofluorescence analysis for DACT-1 on control and CBE-treated human osteoblasts. An evident decrease of DACT-1 immunolabeling in CBE-treated cells is described by the bar graph. Data are the mean ± SD fluorescent intensity measured by ImageJ in 50 individual cells per condition (***P < 0.0001, t-test).
A screening of Wnt signaling-related miRNA identifies miR-221 as candidate post-transcriptional modulator of DVL2
In search for additional mechanisms, we considered that several studies have reported the role of miRNAs in canonical Wnt signaling modulation and differentiation of mesenchymal precursors into osteoblasts (25). To verify whether the Wnt signaling dysregulation in GD1 could be attributed to impaired miRNAs expression, we selected a group of miRNAs that were previously shown to target key Wnt signaling components (25). By RQ-PCRs, we found that several miRNAs were dysregulated in bone extracts from adult Gba mutant fish when compared with age-matched controls (Fig. 5A). We next tested in the total RNA extracts of the same samples the expression profile of dvl1 and dvl2 and found that only dvl2 was significantly downregulated in Gba mutants bone tissue when compared with that of age-matched controls (Fig. 5B). We, therefore, went back to the group of dysregulated miRNAs and found that miR-221 was predicted to bind to dvl2 3′UTR (26,27). As miR-221 was significantly upregulated, we explored its expression profile in CBE-treated human osteoblasts to evaluate whether acute GBA inhibition could trigger the miR-221 increase. However, while RQ-PCR analysis demonstrated a slight but not significant increase after 4 days of GBA enzymatic inhibition (Fig. 5C), transient overexpression of miR-221 in both fibroblasts and human osteoblasts induced a significant decrease of DVL2 protein levels (Fig. 5D and data not shown). From these data, we could infer that miR-221 dysregulation may represent an additional mechanism responsible for the DVL2 downregulation.
A dysregulation in miRNA expression profile is detected in the skeletal tissue of adult Gba1 mutant fish and associated with reduced Dvl2 mRNA levels. (A) Representative bar graph, showing the expression profile analysis of target Wnt signaling-related miRNAs performed on skeletal tissue extracts of 4-month aged control and Gba mutant fish. A statistically significant decrease of miR-34a, miR-144 and miR-124 and increase of miR-200b and miR-221 are detected in the skeletal tissue of Gba mutants when compared with controls. Results are the mean ± SD of four independent replicates (*P < 0.05; **P < 0.01; ***P < 0.0001, t-test). (B) Representative bar graph, showing the expression profile analysis of Wnt pathway-related targets performed on skeletal tissue extracts of 4-month aged control and Gba mutant fish. A statistically significant decrease of Dvl2 mRNA levels is seen in bone extracts from mutant fish when compared with controls (*P < 0.05, t-test). (C) miR-221 expression levels in RNA extracts from human osteoblasts under chronic CBE exposure for 4 and 7 days. (D) Bar graph showing the quantitative decrease of DVL2 protein levels in miR-221-transfected fibroblasts assessed by multiple western blots. Data are the mean ± SD of four independent replicates (**P = 0.0054, t-test).
SNVs in the 3′-UTR of DVL2 of GD patients are observed and point to an epigenetic modulation of Wnt signaling in GD1-affected patients
According to the identification of dysregulated miRNAs targeting DVL2, we hypothesized that the complex spectrum of phenotypes generally observed in GD patients may be in part attributable to a different extent of miRNA-dependent post-transcriptional DVL2 silencing. We, therefore, selected a cohort of 50 type 1 GD patients (Table 1) and performed a high-throughput screening of single-nucleotide variants (SNVs) in the 3′UTR region of DVL2. As shown in Figure 6A, we found a SNV in the 3′-UTR region of DVL2 in 4 (patient no. 1, 3, 6 and 45 of Table 1) out of 50 GD patients (100 sequenced alleles), while no variants were found in the same sequenced DNA tract from a cohort of 32 healthy individuals (64 sequenced alleles). All identified nucleotide substitutions, except the G>A transition (patient no. 45), were located outside of the miRNA target regions. Indeed, the G>A transition was inside the miR-124 target region and very close to the most distal miR-221 target sequence at the 3′ end of DVL2 UTR region. While none of the isolated SNVs were found associated with decreased miR-221 and DVL2 mRNA levels (Fig. 6B and Supplementary Material, Fig. S2), one out of the four SNVs was detected in a GD1 patient, exhibiting increased β-catenin and DVL2 protein levels, as assessed by western blot analysis (Fig. 6C and D). Due to limited access to clinical data, we were not able to establish a direct correlation between the presence of the variants and the phenotype. However, despite the limited number of analyzed samples, we could infer that the 3′UTR of DVL2 in GD patients is a region subjected to variations that may affect the extent of canonical Wnt pathway dysregulation.
List of GD patients recruited in the DNA variants analysis and their detected mutations in the GBA1 gene
| Patient | Allele-1 | Allele-2 |
|---|---|---|
| 1 | N370S | R131C |
| 2 | N370S | N370S |
| 3 | N370S | c.413 delC |
| 4 | N370S | R131C |
| 5 | nd | nd |
| 6 | N370S | N370S |
| 7 | N370S | nd |
| 8 | L444P | I402T |
| 9 | N370S | S42R |
| 10 | N370S | L444P |
| 11 | N370S | Y492X |
| 12 | N370S | N370S |
| 13 | N370S | c.1577_1578insGCTA |
| 14 | N370S | nd |
| 15 | N370S | RecnciI |
| 16 | N370S | G377S |
| 17 | N370S | H255Q + D409H |
| 18 | L444P | nd |
| 19 | N370S | L444P |
| 20 | N370S | L444P |
| 21 | N370S | L444P |
| 22 | N370S | L444P |
| 23 | N370S | P391L |
| 24 | N370S | W184R |
| 25 | N370S | N370S |
| 26 | N370S | N370S |
| 27 | G377S | G377S |
| 28 | N370S | N370S |
| 28 | N370S | N370S |
| 30 | N370S | N370S |
| 31 | N370S | Y108C |
| 32 | N370S | N370S |
| 33 | N370S | g.4356_7031con J03060.1:g.2544_4568 |
| 34 | N370S | N188K |
| 35 | N370S | IVS2 + 1G>A |
| 36 | N370S | D283N |
| 37 | N370S | Q362X |
| 38 | N370S | N370S |
| 39 | V191E | G202R |
| 40 | N370S | D409H + H255Q |
| 41 | N370S | N370S |
| 42 | L444P | W312S |
| 43 | N370S | N370S |
| 44 | N370S | W184R |
| 45 | N370S | R131C |
| 46 | L444P;A456P;V460V | F216Y |
| 47 | N370S | nd |
| 48 | N370S | Nd |
| 49 | N370S | S366R |
| 50 | 6727(-11delC)-(14 T>A) | R170P |
| Patient | Allele-1 | Allele-2 |
|---|---|---|
| 1 | N370S | R131C |
| 2 | N370S | N370S |
| 3 | N370S | c.413 delC |
| 4 | N370S | R131C |
| 5 | nd | nd |
| 6 | N370S | N370S |
| 7 | N370S | nd |
| 8 | L444P | I402T |
| 9 | N370S | S42R |
| 10 | N370S | L444P |
| 11 | N370S | Y492X |
| 12 | N370S | N370S |
| 13 | N370S | c.1577_1578insGCTA |
| 14 | N370S | nd |
| 15 | N370S | RecnciI |
| 16 | N370S | G377S |
| 17 | N370S | H255Q + D409H |
| 18 | L444P | nd |
| 19 | N370S | L444P |
| 20 | N370S | L444P |
| 21 | N370S | L444P |
| 22 | N370S | L444P |
| 23 | N370S | P391L |
| 24 | N370S | W184R |
| 25 | N370S | N370S |
| 26 | N370S | N370S |
| 27 | G377S | G377S |
| 28 | N370S | N370S |
| 28 | N370S | N370S |
| 30 | N370S | N370S |
| 31 | N370S | Y108C |
| 32 | N370S | N370S |
| 33 | N370S | g.4356_7031con J03060.1:g.2544_4568 |
| 34 | N370S | N188K |
| 35 | N370S | IVS2 + 1G>A |
| 36 | N370S | D283N |
| 37 | N370S | Q362X |
| 38 | N370S | N370S |
| 39 | V191E | G202R |
| 40 | N370S | D409H + H255Q |
| 41 | N370S | N370S |
| 42 | L444P | W312S |
| 43 | N370S | N370S |
| 44 | N370S | W184R |
| 45 | N370S | R131C |
| 46 | L444P;A456P;V460V | F216Y |
| 47 | N370S | nd |
| 48 | N370S | Nd |
| 49 | N370S | S366R |
| 50 | 6727(-11delC)-(14 T>A) | R170P |
Note. nd, undetermined.
List of GD patients recruited in the DNA variants analysis and their detected mutations in the GBA1 gene
| Patient | Allele-1 | Allele-2 |
|---|---|---|
| 1 | N370S | R131C |
| 2 | N370S | N370S |
| 3 | N370S | c.413 delC |
| 4 | N370S | R131C |
| 5 | nd | nd |
| 6 | N370S | N370S |
| 7 | N370S | nd |
| 8 | L444P | I402T |
| 9 | N370S | S42R |
| 10 | N370S | L444P |
| 11 | N370S | Y492X |
| 12 | N370S | N370S |
| 13 | N370S | c.1577_1578insGCTA |
| 14 | N370S | nd |
| 15 | N370S | RecnciI |
| 16 | N370S | G377S |
| 17 | N370S | H255Q + D409H |
| 18 | L444P | nd |
| 19 | N370S | L444P |
| 20 | N370S | L444P |
| 21 | N370S | L444P |
| 22 | N370S | L444P |
| 23 | N370S | P391L |
| 24 | N370S | W184R |
| 25 | N370S | N370S |
| 26 | N370S | N370S |
| 27 | G377S | G377S |
| 28 | N370S | N370S |
| 28 | N370S | N370S |
| 30 | N370S | N370S |
| 31 | N370S | Y108C |
| 32 | N370S | N370S |
| 33 | N370S | g.4356_7031con J03060.1:g.2544_4568 |
| 34 | N370S | N188K |
| 35 | N370S | IVS2 + 1G>A |
| 36 | N370S | D283N |
| 37 | N370S | Q362X |
| 38 | N370S | N370S |
| 39 | V191E | G202R |
| 40 | N370S | D409H + H255Q |
| 41 | N370S | N370S |
| 42 | L444P | W312S |
| 43 | N370S | N370S |
| 44 | N370S | W184R |
| 45 | N370S | R131C |
| 46 | L444P;A456P;V460V | F216Y |
| 47 | N370S | nd |
| 48 | N370S | Nd |
| 49 | N370S | S366R |
| 50 | 6727(-11delC)-(14 T>A) | R170P |
| Patient | Allele-1 | Allele-2 |
|---|---|---|
| 1 | N370S | R131C |
| 2 | N370S | N370S |
| 3 | N370S | c.413 delC |
| 4 | N370S | R131C |
| 5 | nd | nd |
| 6 | N370S | N370S |
| 7 | N370S | nd |
| 8 | L444P | I402T |
| 9 | N370S | S42R |
| 10 | N370S | L444P |
| 11 | N370S | Y492X |
| 12 | N370S | N370S |
| 13 | N370S | c.1577_1578insGCTA |
| 14 | N370S | nd |
| 15 | N370S | RecnciI |
| 16 | N370S | G377S |
| 17 | N370S | H255Q + D409H |
| 18 | L444P | nd |
| 19 | N370S | L444P |
| 20 | N370S | L444P |
| 21 | N370S | L444P |
| 22 | N370S | L444P |
| 23 | N370S | P391L |
| 24 | N370S | W184R |
| 25 | N370S | N370S |
| 26 | N370S | N370S |
| 27 | G377S | G377S |
| 28 | N370S | N370S |
| 28 | N370S | N370S |
| 30 | N370S | N370S |
| 31 | N370S | Y108C |
| 32 | N370S | N370S |
| 33 | N370S | g.4356_7031con J03060.1:g.2544_4568 |
| 34 | N370S | N188K |
| 35 | N370S | IVS2 + 1G>A |
| 36 | N370S | D283N |
| 37 | N370S | Q362X |
| 38 | N370S | N370S |
| 39 | V191E | G202R |
| 40 | N370S | D409H + H255Q |
| 41 | N370S | N370S |
| 42 | L444P | W312S |
| 43 | N370S | N370S |
| 44 | N370S | W184R |
| 45 | N370S | R131C |
| 46 | L444P;A456P;V460V | F216Y |
| 47 | N370S | nd |
| 48 | N370S | Nd |
| 49 | N370S | S366R |
| 50 | 6727(-11delC)-(14 T>A) | R170P |
Note. nd, undetermined.
SNVs in the 3′UTR of DVL2 are detected in GD1 patients and associated with increased Wnt pathway activity. (A) Schematic picture showing the 3′UTR region of human DVL2 with the detected SNVs depicted in red. The miR-221 target sequences are in italic blue, while the miR-124 target sequence is underlined. (B) Bar graph showing the miR-221 expression levels detected in samples from the GD1 patients with SNVs in the 3′UTR of DVL2 (*P < 0.05, t-test). (C) Representative western blot analysis of total cell lysates, showing the increase of DVL2 and total β-CATENIN in the cell lysate from a GD1 patient harboring a G>A variant in the 3′UTR of DVL2. (D) Bar graphs depict the quantitation of three independent western blot replicates for the indicated target proteins in patients’ cell extracts (**P < 0.005, t-test).
Discussion
Unbalanced bone remodeling, due to excessive bone resorption and decreased bone formation, represents a key feature of many human disorders characterized by skeletal abnormalities, including GD. In the past few years, independent investigations have significantly contributed to the understanding of bone pathogenesis in GD-affected patients, leading to a re-examination of the commonly accepted paradigm supporting the primary pathogenic role of macrophage-like cells’ (Gaucher cells) infiltration in the BM (28,29). In agreement with these studies, we previously uncovered a novel pathogenic cascade related to GD, by showing that Wnt/β-catenin alterations occur in a zebrafish model of GD, before the onset of osteoblast-related gene expression dysregulation and bone demineralization (5). However, in this previous investigation, we did not have any mechanistic explanation for the detected decrease of total β-catenin levels. More recently, two independent studies by the same research group, relying on the use of induced pluripotent stem cells (iPSCs) from type 2 and 3 GD patients, discovered the involvement of the Wnt/β-catenin pathway in the reduced differentiation potential of reprogrammed cells into neurons and osteoblasts (6,16). The same authors unequivocally showed that reactivation of the same pathway by recombinant glucocerebrosidase (GCase) or CHIR99021 could rescue the cell differentiation defects.
To shed light on the pathogenic cascade underlying the Wnt/β-catenin impairment, we undertook a systematic search for upstream factors and proteins involved in GD-related Wnt signaling dysregulation. By using different experimental models, we were able to identify the involvement of the two members DVL1 and DVL2, of the DVL family, in the downstream reduced β-catenin accumulation. In particular, we discovered that different post-transcriptional mechanisms may act to finely downregulate DVL2 protein levels, thus making β-catenin more accessible to degradation by the ‘destruction complex’. The identification of DVLs as key upstream factors affected by glucocerebrosidase loss of function emphasizes the potential role of sphingolipids turnover in the Wnt signaling transduction across the cell membrane, in agreement with recent observations (30). In addition, considering that both sphingolipids and DVLs are enriched in lipid rafts (17,18,31,32), it is clear why we also found in our GD experimental models reduced levels of EPSIN 1, which is recruited for clathrin-dependent endocytosis (33) and physically interacts with DVL2 (21). Notably, we found that in control and GD fibroblasts EPSIN1 binds to DVL2, as in colon–rectal cancer cells (21), suggesting a conserved mechanism among different cell populations, which relies on a tight functional interaction between Wnt signaling transduction and clathrin-mediated endocytosis.
Among binding partners of DVL2, we also found a downregulation of DACT-1 (Dapper) in our GD experimental models. DACT-1 is nucleocytoplasmic protein that has been controversially pointed as negative or positive modulator of the canonical Wnt signaling (23,24). While we documented reduced DACT-1 levels and less DACT-1 accumulation in the nucleus of GD fibroblasts, we have not been able to justify in detail the biological meaning of such reduction. We assume that, in line with previous suggestions (24), the role of DACT-1 as Wnt signaling antagonizing factor may be cell population-specific, and therefore, in our biological system, its downregulation may be associated with reduced Wnt pathway activity. Future studies will likely clarify this key point.
In this work, we also unveiled a novel mechanistic aspect in the Wnt signaling defects as a consequence of GBA functional impairment. We identified target Wnt signaling-related miRNAs that were dysregulated in both in vitro and in vivo GD models. Interestingly, some of these miRNAs (miR-34a-5p, miR-144-3p and miR-200a-3p) have been already found downregulated in the brain stem of the neuronopathic GD mouse model (4L;C*) (34). In our target screening, we indeed found an upregulation of miR-221-3p, which was associated with decreased dvl2 mRNA levels in the zebrafish model of GD. Interestingly, miR-221-3p was already known to target DVL2 (27) and its overexpression has been recently found to decrease the osteogenic potential in an in vitro model (35). A second miRNA, miR-124-3p, was found to be downregulated both in the zebrafish model of GD and in CBE-treated osteoblasts (data not shown). Unexpectedly, miR-124-3p is also able to bind to DVL2 3′UTR, thus raising the question as to whether its downregulation may be linked to increased DVL2 protein levels (26). However, it has been recently shown that it also targets DACT1 to activate Wnt/β-catenin signaling during neural differentiation (36). We, therefore, assume that its reduced levels may be responsible for a Wnt pathway activity decrease in agreement with other observations (37).
Given the reduced Wnt pathway activation, the decreased amount of DVL2 protein levels and the dysregulation of two miRNAs targeting DVL2 in our experimental GD models (summarized in the model of Figure 7), we asked whether a potential interference with the dysregulated DVL2/β-catenin axis may in part explain the complex range of GD-related phenotypes. There is in fact a growing body of evidence that SNVs in multiple ‘modifier’ genes may sensitively contribute to GD clinical manifestations (38). In a limited selected cohort of GD1 patients, we performed a blind screening of nucleotide variations in the DVL2 3′UTR, where the target regions for miR-221 and miR-124 have been identified. We found four single-nucleotide variations that were not present in the control population. Three out of four of these variants were present in Ensemble genome database, while the T>G substitution was a discovered new variant. The already-known G>A variant (patient no. 45, L31), which lies inside the miR-124 target region and very close to one of the miR-221 target regions (26), was found to be associated with highly increased β-catenin (almost 10-fold) and DVL2 (almost 3-fold) levels. Indeed the C>A variant (patient no. 3, L47) was associated with slightly increased β-catenin (1.7-fold) and DVL2 (1.5-fold) levels, despite significantly higher miR-221 expression, when compared with controls. As stated above, the lack of detailed clinical data prevented us from establishing a genotype–phenotype correlation. The presence of these SNVs in the DVL2 3′UTR region of GD patients suggests, however, that there may be a population-specific selection of target polymorphisms and genetic variants affecting Wnt signaling transduction, which could modulate the phenotype in patients fully lacking glucocerebrosidase activity. The analysis of a wider population of GD patients, as well the functional characterization of each single variant, will be matter of future studies.
Multiple mechanisms are involved in GD-related Wnt signaling dysregulation. Schematic picture summarizing three concurrent defects responsible for the Wnt signaling downregulation: increased Axin1 binding to β-catenin and reduced DVL protein accumulation (left), decreased EPSIN 1 levels (middle) and miR-221-dependent downregulation of DVL2 mRNA levels (right). A red arrow indicates downregulation, while the green arrow points to an upregulation. NM, nuclear membrane; PM, plasma membrane.
In summary, this study pinpoints the relevance of Wnt signaling activity impairment and its relation with glucocerebrosidase dysfunction, supporting the perspective that manipulation of this pathway may be of therapeutic relevance.
Materials and Methods
Zebrafish
Zebrafish larvae were raised for the first 5 days in Petri dishes and maintained at 28°C during later stages in 5 l tanks filled with fish water at neutral pH, according to standard methods (http://ZFIN.org). All procedures involving fish husbandry and manipulation were evaluated and accepted by the Local Ethical Committee at the University of Padova and National Agency (Italian Ministry of Health, project 78/2013). The mutant sa1621 fish line was obtained by the Zebrafish International Resource Center (ZIRC) at Eugene (OR, USA), and fish were genotyped as previously described (5).
Cell lines
The human fetal osteoblasts line (hFOB 1.19) was purchased from ATCC (ATCC® CRL11372™, Manassas VA, USA). Fibroblasts from type 1 GD patients and healthy relatives, supplied by the Biobank (G.Gaslini), have been obtained for analysis and storage with the patients’ (and/or a family member’s) written informed consent. The consent was sought using a form approved by the local Ethics Committee.
RNA isolation and RT-qPCR
Human osteoblasts were cultured according to manufacturer’s instructions in Ham’s F12 Medium Dulbecco’s Modified Eagle’s medium (ThermoFisher Scientific, Monza, Italy) with 2.5 mm L-glutamine together with 0.3 mg/ml G418 (Sigma, Milan, Italy) and 10% fetal bovine serum (Biowest, Nuaillé, France). Fibroblasts were maintained in RPMI 1640 medium supplemented with 10% fetal bovine serum (Biowest, Nuaillé, France), L-glutamine and penicillin–streptomycin (ThermoFisher Scientific, Monza, Italy). For zebrafish samples, genotyped homozygous mutants and age-matched controls (N = 4 for each condition) were euthanized at 4 months of age, and the vertebral column was manually dissected and placed in liquid nitrogen.
RNA samples (mRNA and miRNAs) from cells and bone tissue were extracted using the microRNA Purification Kit (Norgen Biotek, Canada). RNAs were resuspended in 20 μL of RNAse-free water and then quantified by Nanodrop 1000 (ThermoFisher Scientific, Monza, Italy). A total of 2 μg of RNA were reverse transcribed using a SuperScript III Reverse Transcriptase (ThermoFisher Scientific, Monza, Italy), while for miRNAs, cDNA synthesis was performed using the miRCURY LNA™ RT Kit (Qiagen, Milan, Italy), according to manufacturer’s instructions.
The cDNAs were subsequently subjected to SYBR Green-based real-time PCR using a RotorGene 3000 (Corbett, Concorde, NSW). For target miRNA analysis, miRCURY LNA™ primers (miR-200a-3p, miR-144-3p, miR-34a-5p, miR-26a-5p, miR-200b-3p, miR-221-3p, miR-124-3p) were purchased from Exiqon (Qiagen, Milan, Italy) and amplified using the ExiLENT SYBR® Green master mix (Qiagen, Milan, Italy) and a QuantStudio 6 Flex Real-Time PCR System (ThermoFisher Scientific, Monza, Italy). Primers for mRNA analysis are listed in Table 2. RT-qPCR data were analyzed using a manually set threshold, and the baseline was set automatically to obtain the threshold cycle (Ct) value for each target. GAPDH and miR-26a-5p were employed as endogenous controls to normalize the data for mRNA and miRNA analysis, respectively. Relative gene expression among samples was determined using the comparative Ct method (2 − ΔΔCt). Results are expressed as the mean ± standard deviation (SD) relative expression. All statistical analyses were performed using GraphPad Prism 5.0 software.
List of primers and their relative targets used for the RT-qPCR analysis
| Primer | RefSeq | |
|---|---|---|
| Zebrafish Gene | ||
| dvl1a-for | ACAGAGAGCCACAAAGAGCT | XM_021469998.1 |
| dvl1a-rev | TGTTCCTGTCTCTGTGTCCA | XM_021469998.1 9 |
| dvl1b-for | ACTGAACATCCACCACCACT | XM_005169297.4 |
| dvl1b-rev | GATGACGAAGAACAGGCTCG | XM_005169297.4 |
| dvl2-for | CTCAGATGACAGTGCCAAGC | NM_212648.1 |
| dvl2-rev | GAGGAGGTGAGGGCTGAGA | NM_212648.1 |
| axin1-for | GAGAGACAGCCATGGAGAGG | NM_131503.2 |
| axin1-rev | TGCTCATAGTGTCCCTGCAC | NM_131503.2 |
| axin2-for | GCTCCCTGGAGGAGAGACTT | AB032263 |
| axin2-rev | TCGAAGCAATCCTGGAGACT | AB032263 |
| β-catenin for | GGTGGCAGATCTGGACGGG | NM_001001889.1 |
| β-catenin for | CCACCTGGTCCTCGTCGTTC | NM_001001889.1 |
| gapdh-for | GTGGAGTCTACTGGTGTCTTC | NM_001115114 |
| gapdh-rev | GTGCAGGAGGCATTGCTTACA | NM_001115114 |
| Human Gene | ||
| DVL1-for | CCGACTTCAAGAACGTGCTC | NM_001330311.2 |
| DVL1-rev | GGGAAGCTTGGCATTGTCAT | NM_001330311.2 |
| DVL2-for | ACCTGGTGAAGATCCCTGTC | XM_005256502.2 |
| DVL2-rev | CGGGCGTTGTCATCTGAAAT | XM_005256502.2 |
| SP7(osterix)-for | CAGCTCTCTCCATCTGCCTG | NM_001173467.3 |
| SP7(osterix)-rev | AGAGCCACCAAATTTGCTGC | NM_001173467.3 |
| GAPDH For | CGAGCCACATCGCTCAGAC | NM_002046 |
| GAPDH Rev | ACAATATCCACTTTACCAGAGTTAAAAGC | NM_002046 |
| Primer | RefSeq | |
|---|---|---|
| Zebrafish Gene | ||
| dvl1a-for | ACAGAGAGCCACAAAGAGCT | XM_021469998.1 |
| dvl1a-rev | TGTTCCTGTCTCTGTGTCCA | XM_021469998.1 9 |
| dvl1b-for | ACTGAACATCCACCACCACT | XM_005169297.4 |
| dvl1b-rev | GATGACGAAGAACAGGCTCG | XM_005169297.4 |
| dvl2-for | CTCAGATGACAGTGCCAAGC | NM_212648.1 |
| dvl2-rev | GAGGAGGTGAGGGCTGAGA | NM_212648.1 |
| axin1-for | GAGAGACAGCCATGGAGAGG | NM_131503.2 |
| axin1-rev | TGCTCATAGTGTCCCTGCAC | NM_131503.2 |
| axin2-for | GCTCCCTGGAGGAGAGACTT | AB032263 |
| axin2-rev | TCGAAGCAATCCTGGAGACT | AB032263 |
| β-catenin for | GGTGGCAGATCTGGACGGG | NM_001001889.1 |
| β-catenin for | CCACCTGGTCCTCGTCGTTC | NM_001001889.1 |
| gapdh-for | GTGGAGTCTACTGGTGTCTTC | NM_001115114 |
| gapdh-rev | GTGCAGGAGGCATTGCTTACA | NM_001115114 |
| Human Gene | ||
| DVL1-for | CCGACTTCAAGAACGTGCTC | NM_001330311.2 |
| DVL1-rev | GGGAAGCTTGGCATTGTCAT | NM_001330311.2 |
| DVL2-for | ACCTGGTGAAGATCCCTGTC | XM_005256502.2 |
| DVL2-rev | CGGGCGTTGTCATCTGAAAT | XM_005256502.2 |
| SP7(osterix)-for | CAGCTCTCTCCATCTGCCTG | NM_001173467.3 |
| SP7(osterix)-rev | AGAGCCACCAAATTTGCTGC | NM_001173467.3 |
| GAPDH For | CGAGCCACATCGCTCAGAC | NM_002046 |
| GAPDH Rev | ACAATATCCACTTTACCAGAGTTAAAAGC | NM_002046 |
List of primers and their relative targets used for the RT-qPCR analysis
| Primer | RefSeq | |
|---|---|---|
| Zebrafish Gene | ||
| dvl1a-for | ACAGAGAGCCACAAAGAGCT | XM_021469998.1 |
| dvl1a-rev | TGTTCCTGTCTCTGTGTCCA | XM_021469998.1 9 |
| dvl1b-for | ACTGAACATCCACCACCACT | XM_005169297.4 |
| dvl1b-rev | GATGACGAAGAACAGGCTCG | XM_005169297.4 |
| dvl2-for | CTCAGATGACAGTGCCAAGC | NM_212648.1 |
| dvl2-rev | GAGGAGGTGAGGGCTGAGA | NM_212648.1 |
| axin1-for | GAGAGACAGCCATGGAGAGG | NM_131503.2 |
| axin1-rev | TGCTCATAGTGTCCCTGCAC | NM_131503.2 |
| axin2-for | GCTCCCTGGAGGAGAGACTT | AB032263 |
| axin2-rev | TCGAAGCAATCCTGGAGACT | AB032263 |
| β-catenin for | GGTGGCAGATCTGGACGGG | NM_001001889.1 |
| β-catenin for | CCACCTGGTCCTCGTCGTTC | NM_001001889.1 |
| gapdh-for | GTGGAGTCTACTGGTGTCTTC | NM_001115114 |
| gapdh-rev | GTGCAGGAGGCATTGCTTACA | NM_001115114 |
| Human Gene | ||
| DVL1-for | CCGACTTCAAGAACGTGCTC | NM_001330311.2 |
| DVL1-rev | GGGAAGCTTGGCATTGTCAT | NM_001330311.2 |
| DVL2-for | ACCTGGTGAAGATCCCTGTC | XM_005256502.2 |
| DVL2-rev | CGGGCGTTGTCATCTGAAAT | XM_005256502.2 |
| SP7(osterix)-for | CAGCTCTCTCCATCTGCCTG | NM_001173467.3 |
| SP7(osterix)-rev | AGAGCCACCAAATTTGCTGC | NM_001173467.3 |
| GAPDH For | CGAGCCACATCGCTCAGAC | NM_002046 |
| GAPDH Rev | ACAATATCCACTTTACCAGAGTTAAAAGC | NM_002046 |
| Primer | RefSeq | |
|---|---|---|
| Zebrafish Gene | ||
| dvl1a-for | ACAGAGAGCCACAAAGAGCT | XM_021469998.1 |
| dvl1a-rev | TGTTCCTGTCTCTGTGTCCA | XM_021469998.1 9 |
| dvl1b-for | ACTGAACATCCACCACCACT | XM_005169297.4 |
| dvl1b-rev | GATGACGAAGAACAGGCTCG | XM_005169297.4 |
| dvl2-for | CTCAGATGACAGTGCCAAGC | NM_212648.1 |
| dvl2-rev | GAGGAGGTGAGGGCTGAGA | NM_212648.1 |
| axin1-for | GAGAGACAGCCATGGAGAGG | NM_131503.2 |
| axin1-rev | TGCTCATAGTGTCCCTGCAC | NM_131503.2 |
| axin2-for | GCTCCCTGGAGGAGAGACTT | AB032263 |
| axin2-rev | TCGAAGCAATCCTGGAGACT | AB032263 |
| β-catenin for | GGTGGCAGATCTGGACGGG | NM_001001889.1 |
| β-catenin for | CCACCTGGTCCTCGTCGTTC | NM_001001889.1 |
| gapdh-for | GTGGAGTCTACTGGTGTCTTC | NM_001115114 |
| gapdh-rev | GTGCAGGAGGCATTGCTTACA | NM_001115114 |
| Human Gene | ||
| DVL1-for | CCGACTTCAAGAACGTGCTC | NM_001330311.2 |
| DVL1-rev | GGGAAGCTTGGCATTGTCAT | NM_001330311.2 |
| DVL2-for | ACCTGGTGAAGATCCCTGTC | XM_005256502.2 |
| DVL2-rev | CGGGCGTTGTCATCTGAAAT | XM_005256502.2 |
| SP7(osterix)-for | CAGCTCTCTCCATCTGCCTG | NM_001173467.3 |
| SP7(osterix)-rev | AGAGCCACCAAATTTGCTGC | NM_001173467.3 |
| GAPDH For | CGAGCCACATCGCTCAGAC | NM_002046 |
| GAPDH Rev | ACAATATCCACTTTACCAGAGTTAAAAGC | NM_002046 |
Enzymatic assay
Samples were obtained by homogenizing cells in ice with water (ThermoFisher Scientific, Monza, Italy). Debris was pelleted by centrifugation at 4°C and supernatants were collected and assayed for total protein concentration (mg/ml) by the BCA Protein Assay Kit (ThermoFisher Scientific, Monza, Italy).
Glucocerebrosidase activity was determined by a fluorimetric assay using the substrate Methylumbelliferyl-b-D-glucopyranoside (4-MUG) (Sigma, Milan, Italy). Results were normalized for total protein levels and GBA activity was expressed as 1 nmol of substrate hydrolyzed in 4 h per mg total proteins (nmol/mg/4 h).
Western blot
All procedures were performed, as previously described (5). Briefly, harvested cells were lysed in Tissue Extraction Reagent (ThermoFisher Scientific, Monza, Italy) in the presence of protease inhibitor (ThermoFisher Scientific, Monza, Italy) and phosphatase Inhibitor Cocktails 2 and 3 (Sigma-Aldrich, Milan, Italy), The lysates were centrifuged at 13 000×g for 30 min at 4°C. The supernatant was collected and protein concentration was determined by BCA Protein Assay Kit (ThermoFisher Scientific, Monza, Italy). Extracted proteins were supplemented with 4× sample buffer, heated at 70°C for 10 min and run onto precast SDS-PAGE (ThermoFisher Scientific, Monza, Italy). Proteins were transferred on PVDF membranes (Merk Millipore, Italy) in 25 mm Tris, 192 mm glycine, 20% methanol (v/v), 0.1% SDS. Membranes were incubated overnight with primary antibodies against the following proteins: β-catenin (#2534; Cell signaling Biotechnology, Danvers, USA), β-actin (SC59459; Santa Cruz Biotechnology, Dallas, USA), β-actin (A1978, Sigma-Aldrich-Merk, Milan, Italy), phospho-β-catenin (Ser552) (#9566 Cell signaling Biotechnology, Danvers, USA), Axin 1 (ab55906, Cambridge, UK), DVL 1 (ab106844, Cambridge, UK), DVL2 (#3224, Cell signaling Biotechnology, Danvers, USA), EPSIN (ab75879 Cambridge, UK), DACT-1 (#PA5-23216, ThermoFisher, Milan, Italy), OSTERIX(SP7) (sc-393 325, Santa Cruz Biotechnology, Dallas, USA) and GAPDH (ab-9485, Abcam, Cambridge, UK). All antibodies were used at 1:1000 dilution. Secondary horseradish peroxidase-conjugated goat anti-rabbit and mouse IgG antibodies (Biorad, Milan, Italy) were incubated at room temperature for 1 h at 1:3000 dilution. Chemiluminescent signals were detected incubating the blotted membranes with the Supersignal West Pico Chemiluminescent substrate kit (ThermoFisher Scientific, Monza, Italy) and visualized by Image Quant Las 4000 (GE Healthcare, Milan, Italy) and analyzed by ImageJ software (https://imagej.nih.gov/ij/).
Coimmunoprecipitation of Axin/β-catenin and Epsin/DVLs complexes
Coimmunoprecipitation was performed using Dynabeads (ThermoFisher Scientific, Monza, Italy) by an optimized protocol. A polyclonal anti-Axin1 or anti-DVL1 (2) antibody was incubated with native protein lysates from fibroblasts of a healthy control and GD1 patient o/n at 4°C. Dynabeads were then incubated for 4 h. Pull down of antibody-bound beads was performed by using magnets. Pellets were washed three times with PBS, while recovery of protein complexes was performed adding NuPAGE™ LDS Sample (ThermoFisher Scientific, Monza, Italy) followed by a run on precast SDS-PAGE (ThermoFisher Scientific, Monza, Italy). Blotted PVDF membranes were finally incubated with antibodies against β-CATENIN or EPSIN.
Immunofluorescence on type 1 GD fibroblasts
Twenty thousand fibroblasts per well were seeded on polylysine-coated coverslips in a 24-well plate. After 24 h, cells were fixed with 4% buffered paraformaldehyde. Blocking was achieved with a 2% (w/v) bovine serum albumin in PBS. After few washes in PBS-Tween (0.1% v/v), cells were incubated overnight with primary antibody at 4°C, and after three washes in PBS-Tween (0.1% v/v), a secondary antibody incubation for 1 h was performed. Coverslips were finally mounted with Fluoromount medium (ThermoFisher Scientific, Monza, Italy) on glass slides and observed under C2 confocal microscope (Nikon, Italy). The following antibodies were used: mouse anti-β-CATENIN (Sigma, Milan Italy; 1:100), rabbit anti-DVL1 (Abcam, Milan, Italy; 1:100), rabbit anti-DVL2 (Cell signaling, Milan, Italy; 1:100) and rabbit anti-DACT-1 (Abcam, Milan, Italy; 1:100).
miR-221 overexpression
Transfection of human fibroblasts and osteoblasts with fluorescently labeled hsa-miR-221-3p miRCURY LNA miRNA Mimic and negative control miRNA (Qiagen, Milan Italy) was performed according to manufacturer’s instructions. Briefly, cells were seeded in 24-well plates at 75% confluency. After 12 h, transfection mixtures containing the miRNA mimic or the negative control and Hiperfect Transfection reagent (Qiagen, Milan Italy) were prepared. Complexes were added to seeded cells at 5 and 10 nM final concentration of the miRNAs and incubation with the complexes was kept up to 48 h. Transfection efficiency was monitored every 12 h under fluorescent microscopy (Leica DM IL Led, Milan, Italy). The transfection was stopped and cell harvested for total protein extraction.
Genomic DNA sequencing and variants analysis
Total DNAs were extracted from cultured lymphoblasts using DNA mini kit (Qiagen, Courtaboeuf, France). PCR was performed by using FastStart High Fidelity PCR System (Roche Diagnostics, Mannheim, Germany) and specific primers designed by reference to the genomic sequence (GenBank no. NM_004422). F: 5′ TGG GTC GAC ACG GAG TGA T 3′ and R: 5′ TGC CTC TAT TGC TTT ATT TGG TGT 3′. PCR conditions were 95°C for 2 min, then 35 cycles of 95°C for 30 s, 60°C for 30 s and 72°C for 1 min 20 s, finally 72°C for 5 min. The products from PCR were purified and directly sequenced using an ABI 377 DNA automated sequencer with dye terminator cycle sequencing kits (Applied Biosystems, Foster City, CA).
Acknowledgements
We thank Dr Martina Milanetto and Dr Luigi Pivotti for the care of fish husbandry.
Conflict of Interest Statement
None declared.
Funding
Genzyme Sanofi (Genzyme Generation Program 2013) GZ2014-11252 to E.M; ‘Cinque per mille e Ricerca Corrente, “Ministero della Salute’ (to S.L. and M.F.). Patient samples were obtained from the ‘Cell Line and DNA Biobank from patients affected by Genetic Diseases (Istituto Giannina Gaslini), member of Telethon Network of Genetic Biobanks (project no. GTB12001).
