|Home | About | Journals | Submit | Contact Us | Français|
Bisphosphonates (BPs) are the first-line treatment of bone loss resulting from various pathological conditions. Due to their high affinity to bone they have been used to develop conjugates with pro-anabolic or anti-catabolic drugs. We recently demontrated that hydrogen sulfide (H2S), promotes osteogenesis and inhibits osteoclast differentiation. Here we developed an innovative molecule, named DM-22, obtained from the combination of alendronate (AL) and the H2S-releasing moiety aryl-isothiocyanate. DM-22 and AL were assayed in vitro in the concentration range 1-33μM for effects on viability and function of human osteoclasts (h-OCs) and mesenchymal stromal cells (h-MSCs) undergoing osteogenic differentiation. Amperometric measures revealed that DM-22 releases H2S at a slow rate with a thiol-dependent mechanism. DM-22 significantly inhibited h-OCs differentiation and function, maintaining a residual h-OCs viability even at the high dose of 33μM. Contrary to AL, in h-MSCs DM-22 did not induce cytotoxicity as revealed by LDH assay, significantly stimulated mineralization as measured by Alizarin Red staining and increased mRNA expression of Collagen I as compared to control cultures. In conclusion, DM-22 is a new BP which inhibits h-OCs function and stimulate osteogenic differentiation of h-MSCs, without cytotoxicity. DM-22 is an ideal candidate for a novel family of osteoanabolic drugs.
A progressive decline in bone strength is a common consequence of aging and several bone-wasting diseases in humans. Osteoporosis, the most prevalent cause of bone fragility, affects 1 in 2 women and 1 in 5 men age 50 and above and causes up to 9 million fractures per year worldwide1,2. Among pharmacological therapies, bisphosphonates (BPs) are the first-line and the most prescribed treatment for a number of diseases leading to abnormal bone turnover, including osteoporosis1. Key to the mechanism of action of BPs is the high affinity of these molecules for the mineralized bone matrix, which arises from the P-C-P backbone structure3,4. Once BPs bind hydroxyapatite within the bone matrix, the acidic pH caused by osteoclasts (OCs) resorption induces their dissociation from the mineral surface and the subsequent internalization within the OCs5. Owing to their strong affinity for the bony mineralized matrix, BPs have been broadly developed both as a powerful drugs for bone metabolism and as a carrier to achieve a targeted delivery of bone active molecules6,7. Among BPs, amino BPs (N-BPs) were found to be far more potent than simple BPs at inhibiting bone resorption8. N-BPs inhibit the mevalonate pathway by targeting the farnesyl diphosphate synthase (FPPS)9, leading to the accumulation of unprenilated GTPases in the cytoplasm, which results in toxicity and cell death9,10.
Clinical studies have conclusively showed that long-term use (up to 10 years) of BPs is associated to a good safety profile and to a significant reduction in the risk of vertebral, non-vertebral and hip fractures11,12. However, several studies have associated BPs therapy with a potential risk of osteonecrosis of the jaw13,14 and atypical subtrochanteric femoral fractures; at the cellular level, prolonged exposure to BPs was shown to eventually cause the suppression of osteoblast function both by direct and OC-mediated mechanisms15. The inability to restore the lost bone structure may be the cause of poor bone quality and increased risk of atypical bone fractures in BPs treated patients16. Although the prevalence of side effect from BPs is very low, the impact from media reports has led to a dramatic decrease in the compliance to BPs therapies; for example, the use of alendronate (AL) has declined by over 50% from 2008 to 201217. As a consequence, drugs capable of stimulating bone formation while maintaining anti-resorptive activity can provide strong advantages as compared to current BPs.
A growing body of evidence indicates that hydrogen sulfide (H2S) is a gaseous molecule produced in substantial amounts by mammalian tissues, which exerts a variety of physiological effects in different systems, including bone18. Consistently, the discovery of suitable H2S-donor agents and H2S-releasing multi-target drugs is currently considered a timely and challenging field of research in drug discovery19.
In particular, our group and others recently described that H2S inhibits h-OCs development in vitro 20 and stimulates osteogenic differentiation of human mesenchymal stromal cells (h-MSCs) in vitro and in vivo 18,21. Based on the dual action exerted by H2S, we sought to exploit these findings through the development of an H2S-releasing BPs, prototype of a novel family of hybrid molecules aiming at treating bone loss, by means of a multi-target pharmacological approach. An H2S-releasing BPs could allow a local delivery of H2S at the site where new bone formation is required thereby maximizing the osteogenic properties of H2S.
Ideal H2S-releasing moieties should generate H2S with a relatively slow rate; moreover, they should be stable, allowing an easy chemical manipulation22. In this perspective, we have recently investigated several organic sulfur compounds of either synthetic or natural origin23,24 and identified aryl-isothiocyanate as a suitable moiety, useful for building H2S-releasing hybrid drugs25.
This paper describes an innovative molecule, named DM-22, derived from the widely known N-BPs, AL, hybridized with an aryl-isothiocyanate - based H2S-releasing moiety. We investigated in vitro the potential beneficial effects on bone cells of DM-22 compared to that of the parent drug AL.
Melting points were determined on a Kofler apparatus and are uncorrected. Chemical shifts (δ) are reported in parts per million and are calibrated using residual undeuterated solvent as an internal reference. 1H NMR,31P NMR and13C NMR spectra of all compounds were recorded with a Varian Gemini 200 spectrometer operating at 200MHz or Bruker TopSpin 3.2 spectrometer operating at 400MHz, in a ~2% solution of deuterated water (D2O), unless otherwise stated. Data for1H NMR spectra are reported as follows: chemical shift (δ ppm) (multiplicity, coupling constant (Hz), integration). Multiplicities are reported as follows: s=singlet, d=doublet, t=triplet, q=quartet, m=multiplet, br=broad, or combinations thereof. The 95% purity of tested compounds was confirmed by combustion analysis. ESI-MS/MS experiment was performed in negative ion mode, using a LCQ Advantage ion trap mass spectrometer (ThermoFinnigan, San Jose, CA, USA) equipped with Xcalibur 3.1 software. Merck gel plates (60 F254) were used for analytical TLC. UV light was used to examine the spots. Evaporation was performed in vacuo (rotating evaporator). Sodium sulfate was used as the drying agent. Commercially available chemicals were purchased from Sigma-Aldrich and TCI Chemicals. The synthesis of DM-22 was performed in 3 steps characterized by the preparation of the following intermediate products.
p-nitrobenzoic acid (300mg, 1.64mmol) was treated with thionyl chloride (685mg, 5.76mmol) at 80°C for 12h. The solvent was evaporated to yield a yellow solid which was dissolved in a small amount of THF and added dropwise to a solution of γ-aminobutyric acid (78mg, 0.76mmol) in aqueous NaOH (156mg, 4.40mL). The resulting solution was stirred at room temperature (RT) for 12h; Then, the pH value was adjusted to 2 by the addition of HCl 1N. The precipitate was collected and crystallized from H2O thus yielding a white solid corresponding to the product (1) (197mg, yield 74%): 1H-NMR (CD3OD): δ 1.86–2.00 (m, 2H, CH2); 2.41 (t, 2H, J=7.3Hz, CH2); 3.46 (t, 2H, J=6.9Hz, CH2), 8.02 (d, 2H, J=9.0Hz, Ar); 8.32 (d, 2H, J=9.0Hz, Ar) ppm.
To a solution of 4-(4-nitrobenzamido)butanoic acid (77mg, 0.31mmol) in MeOH (2mL) was added carbon (16mg), a catalytic amount of FeCl3 and hydrazine monohydrate (116mg, 5.18mmol). The resulting mixture was refluxed for 24h. Afterwards, the reaction mixture passed through a pad of celite and the solvent evaporated affording a crude product that was purified by precipitation from EtOH/Et2O (43mg, yield 64%).1H-NMR (CD3OD): δ 1.80–1.94 (m, 2H, CH2); 2.28 (t, 2H, J=7.2Hz, CH2); 3.37 (t, 2H, J=6.8Hz, CH2), 6.66 (d, 2H, J=8.5Hz, Ar); 7.60 (d, 2H, J=8.5Hz, Ar) ppm.
A solution of cathecol borane in THF 1M (0.88mL, 0.88mmol) was added to compound 2 (63mg; 0.28mmol) under N2 atmosphere. The reaction mixture was stirred at RT for 1h. Then, P(OSiMe3)3 (347mg, 1.16mmol) was added to the reaction mixture and stirred at rt for 16h. The solvent was removed in vacuo to give a crude residue, which was triturated with CHCl3. The crude oil was dissolved in a solution of NaHCO3 0.4M (2mL) and thiophosgene (228mg, 1.98mmol) was added to the reaction mixture and stirred at RT for 2h. Finally, the reaction mixture was extracted with dichloromethane (DCM). The combined organic layers were dried over MgSO4, and filtered, and the solvent was removed in vacuo to give a residue, which was triturated with hexane to obtain the final product (20mg, yield 24%).1H-NMR (D2O): δ 1.90–2.15 (m, 4H, CH2); 3.45 (t, 2H, J=6.6Hz, CH2); 7.43 (d, 2H, J=8.8Hz, Ar), 7.80 (d, 2H, J=8.8Hz, Ar) ppm.13C-NMR (D2O): δ 167.70, 129.37, 128.93, 126.39, 124.06, 123.70, 76.66, 40.56, 31.24, 23.52 ppm.31P- NMR (D2O) δ 19.23 ppm; ESI-MS m/z calculated for C12H16N2O8P2S, 410.01; found, 409.45. [M - H+].
The characterization of the potential H2S-generating properties of DM-22 has been carried out by an amperometric approach, through the Apollo-4000 free radical analyzer (WPI) detector and H2S-selective mini-electrodes. 6 different recordings have been carried out at RT. Following the manufacturer’s instructions, a “PBS buffer 10x” was prepared (NaH2PO4•H2O 1.28g, Na2HPO4.12H2O 5.97g, NaCl 43.88g in 500ml H2O) and stocked at 4°C. Immediately before the experiments, the “PBS buffer 10x” was diluted using distilled water (1:10) to obtain the assay buffer and the pH adjusted to 7.4. The H2S-selective mini-electrode was equilibrated in 10ml of the assay buffer, until the recovery of a stable baseline. Then, 100μl of a dimethyl sulfoxide (DMSO) solution of DM-22 was added (the final concentration of DM-22 was 1mM; the final concentration of DMSO in the assay buffer was 1%). The generation of H2S was observed for 20min. Preliminary experiments demonstrated that DMSO 1% did not produce any interference on the amperometric recording. When required by the experimental protocol, L-Cysteine (final concentration 4mM) was added 10min before the addition of DM-22. L-Cysteine alone did not produce any amperometric response. The correct relationship between the amperometric currents (recorded in pA) and the corresponding concentrations of H2S was previously determined by suitable calibration curves, which were obtained by the use of sodium hydrosulfide, an H2S donor, (NaHS; 1-3-5-10μM) at pH 4.0.
In preparing samples for human tissues, all protocol and procedures were in accordance with the ethical standards of the institutional ethical committee of the Rizzoli Orthopedic Institute, who approved the study, and with the 1964 Helsinki Declaration and its later amendments.
Human monocytes (h-monocytes) were isolated from peripheral blood mononuclear cells (PBMCs) of 3 healthy donors, carrying out a gradient separation with Ficoll (Lympholite-H, Cederlane, Burlington, Ontario, Canada) followed by immunomagnetic positive selection of CD11b+ cells (MACS system, Miltenyi Biotech, Calderara di Reno, Italy), according to procedures well established in our laboratory26. Purity of CD11b+ enriched population was assayed by flow cytometry (FACS) analysis using FACS canto II (BD bioscience, San Jose, California, USA).
h-monocytes were seeded in quadruplicate into 96 well-plates at a density of 5×105 cells/cm2. h-OCs were obtained by culturing h-monocytes for 6 days α–MEM medium (Euroclone, Milan, Italy) supplemented with 10% FBS (Lonza, Basel, Switzerland) and 1% penicillin/streptomycin, in the presence of M-CSF (10 ng/ml) and RANKL (75 ng/ml) (Miltenyi Biotech). Cells were cultured in presence or absence of AL and DM-22 (1, 3.3, 10, 33µM); medium and stimuli were replaced three times per week.
h-monocytes were seeded as above and cultured in α–MEM 5% FBS (Thermo Fischer Scientific, Waltham, Massachusetts, U.S.A) depleted of phenol-red for 24–72h in the presence or absence of increasing concentrations of AL and DM-22 (1-3.3-10-33µM). Lactate dehydrogenase (LDH) assays (Cytotoxicity detection kit (LDH), Roche) were performed on supernatants, according to manufacturer instructions, to test the acute toxicity. Colorimetric detection of LDH was performed at 492–620nm on TECAN instrument and cytoxicity calculated with reference to the control (unstimulated samples) and positive control (Triton X-100 treated samples) according to the formula: [(experimental value-unstimulated control value)/(Positive control-unstimulated control value)]*100. According to the previous formula unstimulated control value express a 0% of cytotoxicity while positive control a 100% of cytotoxicity.
To evaluate OC differentiation, tartrate acid phosphatase (TRAP) assay (Acid Phosphatase, Leukocyte (TRAP) Kit, Sigma Aldrich) was performed after 7 days in culture. Mature h-OCs were defined as TRAP positive cells containing at least three nuclei and were manually counted using an inverted microscope; 8 microscope fields at 20X magnification were counted for each duplicate well and the h-OCs count was expressed as average h-OCs number/fields.
For pit assays, h-monocytes were seeded on synthetic hydroxyapatite-coated 16-well slides (Osteologic slides, BD Pharmingen, Franklin Lakes, NJ, USA), which mimics in vitro bone matrix, at the density of 5×105/cm2.
h-OCs were obtained by culturing h-monocytes for 7 days in osteoclastogenic medium in presence or absence of AL and DM-22 (1–3.3–10–33µM). After washing with bleach to eliminate cells, pits formed on the Osteologic slides upon matrix breakdown by h-OCs, were captured by NIS software (Nikon, Firenze, Italy) and Nikon Instruments Europe BV (Amstelveen, the Netherlands) using differential interference contrast system and evaluated.
Bone resident h-MSCs were isolated, after obtaining informed consent by each donor, from the tybial plateau of 6 patients undergoing surgical knee replacement according to procedure well established by our laboratory27. Briefly, bone fragments were mechanically fragmented into small pieces to generate a cell suspension which was subjected to Ficoll-density gradient isolation protocol as previously reported27. Cells were grown and expanded in α–MEM medium supplemented with 15% FBS and 1% penicillin/streptomycin until passage 2.
h-MSCs were seeded into 96 well-plates at a concentration of 3×104 cells/ cm2 in α–MEM 7,5% FBS depleted of phenol-red for 24–72h in the presence or absence of increasing concentrations of AL and DM-22 (1–3.3–10–33µM). Morphological analysis was examined under light microscope. Photos were taken using Nikon Instruments after having performed toluidine blue assay for increasing the sharpness of cells. Cells were fixed in formalin 10% for 20min, toluidine was added for few seconds and then cells were washed with deionized water to eliminate unspecific staining. Afterwards, the positive staining was measured at 560nm on TECAN Infinite® 200 PRO (Tecan Italia S.r.l., Cernusco Sul Naviglio, Italy). LDH assays (Cytotoxicity detection kit (LDH), Roche) were performed as described above.
h-MSCs were seeded in quadruplicates in 96-well plates at 3×104 cells/cm2 and cultured in α-MEM 15% FBS at 37°C, 5% CO2 and 95% O2. Proliferation was tested after 24 and 72h in culture. 18h before each time point, 5 µCi of3H-thymidine (Perkin Elmer, Boston, MA, USA) was added to each well and radioactivity was then measured using a beta-counter (Perkin Elmer).
Cells were harvested and seeded into 12 well-plates at a concentration of 5×104 hMSCs/cm2. Cells were induced toward osteogenic differentiation with α-MEM 20% FBS supplemented with 0,1µM dexamethasone (Sigma Aldrich, St. Louis, MO, USA), 100µM ascorbic acid (Sigma Aldrich) and 10mM β-glicerolphosphate (Sigma Aldrich) in the presence or absence of AL and DM-22 (1–3.3–10–33µM). Cells were cultured for up to 21 days at 37°C, 5% CO2 and 95% O2 and medium was replaced twice per week. Alizarin Red S (AR-S) staining (Sigma Aldrich) was performed as follows. Cells were fixed for 15minutes at RT in formaldehyde (Kaltek, Padova, Italy) 10% phosphate buffered saline (PBS), washed twice with PBS, stained with 40mM AR-S for 20minutes at RT and washed with deionized water to eliminate unspecific staining. A spectrophotometric analysis with TECAN was used to quantify AR-S positive stain. Briefly, absorbance was red at 510nm, as previously detailed27, and data were further elaborated subtracting the levels of background. As a result, readings which do not detect a positive stain gave negative values.
To evaluate gene expression, RNA was isolated with RNApure (Euroclone), and purified from DNA with DNA-free™ Kit (Ambion, Life Technologies; Carlsbad, California, U.S.A) according to manufacturer instructions. RNA concentration was assessed by Nanodrop 2000c (Thermo Scientific, Rockford, IL, USA) and the quality of RNA was analyzed through the 260/280 and 260/230nm absorbance ratio. Only samples showing a 260/280 ratio >1.8 were used for the following transcription step. Reverse transcription (SuperScript® VILO™ cDNA Synthesis Kit; Invitrogen, Life Technologies) was performed utilizing 0,5-1µg of RNA according to manufacturer instruction as follows: 25°C for 10minutes, 42°C for 60minutes, 85°C for 5minutes and 4°C for 30minutes on 2720 Thermal cycler (Applied Biosystem, Life Technologies). Polymerase chain reaction (PCR) (SYBR Premix Ex Taq, TaKaRa Biomedicals, Tokyo, Japan; LightCycler Instrument, Roche) was performed on 20 ng cDNA as following: one cycle at 95°C for 10seconds and 45 cycles at 95°C for 5seconds and at 60°C for 20seconds. The specificity of the PCR products was confirmed by standard melting curve analysis with the following thermal cycling profile: 95°C for 10seconds, 65°C for 15seconds and 95°C in one-degree increments. Relative quantification of PCR products was obtained with the comparative CT method, comparing to the housekeeping mRNA expression of glyceraldehyde-3 phosphate dehydrogenase (GAPDH). All the primers, whom sequences are reported in Table 1, were obtained from Life Technologies.
GraphPad Prism 7 (La Jolla, CA, USA) and IBM SPSS Statistics (New York, USA) software were used for statistical analyses. D’Agostino-Pearson omnibus test was performed to test normality of continuous variables. Data on H2S release by DM-22 were analyzed by two-way ANOVA followed by Bonferroni post-test.
Multiple comparisons were made by one-way ANOVA and Tukey multiple comparison tests or Kruskall-Wallis and Dunn’s multiple comparison test, respectively in parametric or non-parametric data set. Simple comparisons were made by using Unpaired t-test or Mann-Whitney test, respectively in parametric or non-parametric data set. Wilcoxon signed rank test was used to compare each sample with a hypothetical value (1 or 100) for analysis of fold increase. A statistical multivariate analysis, the Generalized Linear Model (GLM), was used to analyze three groups of non-parametric data in the osteogenic gene expression data set.
P values <0.05 were considered statistically significant.
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.
The new hybrid compound was synthesised as reported in Fig. 1. Briefly, the condensation of p-nitrobenzoic acid with the γ-aminobutirric acid in an aqueos solution of NaOH yielded the amide 1 (Fig. 1a). The subsequent reduction of the nitro group with hydrazine monohydrate in the presence of FeCl3 gave the 4-(4-aminobenzamido)butanoic acid 2 (Fig. 1b). The intermediate namely the 1-hydroxy-1,1,-bisphosphonic acid was obtained through an efficient and simple one-pot synthesis starting from the carboxylic acid 1, cathecol-borane and tris(triethylsylilphosphate)28. The final reaction with thiophosgene and NaHCO3 yielded the isothiocianate derivative DM-22 (Fig. 1c). The H2S donor-BP was characterized by 1H, 13C, 31P NMR spectra.
The incubation of DM-22 1mM in aqueous solution (assay buffer, at pH 7.4 and 20°C), in the presence of L-Cysteine 4mM, led to the generation of a time-related increasing concentration of H2S (Fig. 2). The rise of the H2S production reached a steady-state after about 10min and the highest concentration of H2S, recorded after 20min was 42.0±2.1µM. In contrast, in the absence of L-Cysteine, the incubation of DM-22 1mM led to the formation of dramatically lower levels of H2S (1.8±0.1µM, after 20min of incubation). Such a organic thiol-dependent mechanism of H2S release is in agreement with previous observations on aryl-isothiocyanates25, and even on different chemotypes of H2S-donors, such as thioamides23 or organic disulfides29.
Firstly we investigated the safety profile of DM-22 by analyzing its cytotoxicity on mature h-OCs by LDH assay. As shown in Fig. 3A, DM-22 displayed no cytotoxicity throughout the concentration range tested. Negative values of cytotoxicity are explained by the fact that DM-22 stimulation resulted in reading values below the baseline levels generated by control, unstimulated sample (=0). On the contrary, AL consistently revealed a detectable cytotoxicity, which was significantly higher than DM-22 at each concentration tested in our experiments (p<0.05 for 1 and 3.3μM, p<0.01 for 10μM, p<0.0001 for 33μM versus DM-22, Fig. 3A).
Next, we asked whether DM-22 retained the AL-like inhibition on osteoclastogenesis. Analysis of osteoclastogenic differentiation revealed a different pattern of inhibition by DM-22 compared to AL. Representative pictures of TRAP+hOCs at the end of the culture period are shown in Fig. 3B. DM-22 significantly decreased the number of mature hOCs at all the concentrations tested (Fig. 3C; p<0.05 for 1μM, p<0.01 for 3.3 and 10μM, p<0.0001 for 33μM vs control sample). Inhibition of h-OCs differentiation peaked at the high concentration of 33μM, where no cytotoxicity had been shown, and reached nearly 40% inhibition relative to control samples.
As expected, AL dose-dependently decreased the total number of TRAP+h-OCs, resulting into highly significant inhibition at all the concentrations tested (p<0.0001 vs control sample) (Fig. 3C). In particular AL decreased the number of h-OCs by 62% and 77%, respectively, at 1 and 3.3μM and virtually abolished h-OCs differentiation starting from the concentration of 10μM.
These data suggest that DM-22 likely preserve a residual population of h-OCs, important for the mantainment of bone turnover.
Coherently, the ability of mature h-OCs to break down a mineral substrate, as tested in vitro by the ‘pit assay’, was strongly prevented by 1μM of AL treatments and virtually completely inhibited by higher concentrations (Fig. 3D). Conversely, h-OCs function was dose-dependently inhibited by DM-22.
Next we investigated the effect of DM-22 on h-MSCs on viability and proliferation. Staining with the vital dye toluidine blue revealed that treatment with DM-22 and AL at the highest concentration (33μM) triggered a different response on h-MSCs after 72h in culture (Fig. 4A,B). While DM-22 did not alter the intensity of staining compared to control samples (Fig. 4A-a,c; Fig. 4B), AL induced a marked decreased in cellularity (Fig. 4A-a,d); moreover, after AL treatment, h-MSCs showed occasionally picnotic nuclei (Fig. 4A-d) and the intensity of staining was significantly decreased compared to control samples, close to the value induced by treatment with the detergent Triton X-100 (CTRL+, Fig. 4B).
Next we investigated the acute cytotoxicity within a time-range of 72h. Quantification of LDH assay performed after 24h (Fig. 4C) and 72h (Fig. 4D) confirmed that DM-22 was completely devoid of cytoxicity at all the concentration tested. Consistant with data revealed by toluidine blu staining, the highest concentration of AL resulted in significant cytoxicity in h-MSCs after 72h in culture (Fig. 4D; 20%; p<0.0001). Significant cytotoxicity was also induced by 10μM AL, however the percentage of inhibition is negligible compared to that of the highest dose (Fig. 4D; 1%; p<0.05).
Analysis of cell proliferation in h-MSCs also revealed that AL treatment induce a dose-dependent decrease in cellular proliferation, resulting into a significant inhibition at 33μM at 72h in culture (Fig. 4E,F). The highest concentration of AL induced a maximum inhibition of proliferation of nearly 50% relative to control samples after 72h (p<0.01). Although DM-22 also induced a significant inhibition of cell proliferation at the highest concentration (20%, p<0.01), it showed a markedly lower inhibition of cell proliferation compared to AL (Fig. 4F; p<0.05).
To further assess the effect of DM-22 on h-MSCs function, we investigated bone resident h-MSCs during in vitro osteogenic differentiation. Figure 5A shows representative pictures of AR-S staining at the day 21 time point for each treatment. As expected, control cultures of h-MSCs under osteogenic stimuli produce large mineralized nodules (Fig. 5A); interestingly, stimulation with DM-22 resulted in further increase of mineral matrix apposition, which induced a significant increase in mineralization as detected by quantification of nodules positive to AR-S staining at all the concentrations tested (Fig. 5B, p<0.05, p<0.001). A similar treatment with AL induced a marked inhibition of mineralization, which became evident by 1μM and 3.3μM (Fig. 5A), concentration levels that did not previously show cytoxicity (Fig. 4C,D). Consistant with findings showing rising cytoxicity of AL, stimulation with higher concentrations of AL resulted in the complete inhibition of mineral matrix deposition by h-MSCs (Fig. 5A,B; p<0.01, p<0.0001).
To further investigate the nature of DM-22 stimulation of osteogenic differentiation, we evaluated mRNA expression of osteogenic markers during the osteogenic culture (D14) and at the end of the experiment (D21).
Figure 6 summarizes gene expression of ALP (Fig. 6A,B), BSP (Fig. 6C,D), COLL I (Fig. 6E,F) and COLL XV (Fig. 6G,H). The GLM statistical analysis revealed that mRNA expression was affected by DM-22 or AL independent of the dose used; as a consequence, data are represented as box plots showing the overall effect of stimuli on the expression of target genes. h-MSCs stimulated with DM-22 showed significantly higher expression of COLLI at D14 in culture (Fig. 6E, p<0.05) compared to control samples. DM-22 also induced gene expression above that of control samples for ALP at D14 (Fig. 6A), BSP at D21 (Fig. 6B), COLLXV at D21 (Fig. 6H); however, due to the high variability in response, these comparison did not achieve statistical significance in our analysis. When compared to AL-treated h-MSCs, DM-22 induced consistantly higher gene expression level for all the endpoints examined; in particular, mRNA of ALP D14 (Fig. 6A, p<0.05), COLLI D14-D21 (Fig. 6E,F, p<0.05) and COLLXV D14 (Fig. 6G, p<0.05) was significanlty higher in samples treated with DM-22 compared to AL. Moreover, AL suppressed mRNA expression of ALP at D14-D21 (Fig. 6A,B, p<0.05), COLLI at D21 (Fig. 6F, p<0.05) and COLLXV D14 (Fig. 6G, p<0.05) compared to control h-MSCs. Notably, for the same genes at the same time point, DM-22 stimulation did not suppress mRNA expression levels compared to control samples.
H2S has recently shown an unexpected role in bone biology as a pivotal regulator of bone turnover. In the present study, we exploited the properites of H2S into a novel H2S-releasing BPs, named DM-22, obtained by “chemical hybridization” of AL, the lead compound in the N-BPs class, with an aryl-isothiocyanate-based H2S-releasing moiety. We investigated the main pharmacodynamic features of DM-22 in vitro and assessed its efficacy and safety in h-MSCs and h-OCs.
Electrochemical measurements demonstrated that DM-22 released H2S with a long-lasting kinetic, in line with the feature of isothiocyanate-based H2S-donor23,24. In designing this hybrid drug, such a pharmacological property was expected to improve the whole pharmacological profile and to attenuate the possible cytotoxicity of the BPs molecular portion.
DM-22 has been designed to meet the need for molecules showing improved efficacy and safety profile as compared to existing drugs for bone metabolism.
BPs are the first-line treatment for the therapy of several bone-wasting conditions30. However, long-term treatments with BPs are associated with an increased risk of developing adverse events; for example, epidemiological studies have established a compelling association between BPs potency, duration of treatment and the risk for developing osteonecrosis of the jaw31 as well as atypical, low energy fractures32. Long-term therapy with AL has been recently associated with the accumulation of microcracks in bone leading to reduced bone strenght in patients33, consistant with the hypothesis that excessive suppression of bone turnover can deteriorate bone quality16. Moreover, a recent report from the multidisciplinary task force of the American Society for Bone and Mineral Research (ASBMR) highlighted that, although the benefits of BPs treatment by far outweighs the risk of incurring in side effects, over 50% of patients failed to adhere to treatment longer than 12 months due, at least in part, to concerns about side effects.
DM-22 did not show cytoxicity in h-OCs at any of the concentrations tested. However, DM-22 induced a significant reduction of the number of mature h-OCs at all concentrations, although the inhibition was more pronounced at the highest concentration of 33μM. By contrast, AL triggered a much more pronounced inhibition of h-OCs differentiation which was accompanied by an increased cytoxicity; consistant with previous reports34,35, AL completely abolished h-OCs differentiation and their ability to resorb mineral matrix in the concentration range between 1 and 33μM.
Although we did not investigate the mechanism for the reduced anti-resorptive activity of DM-22, it is conceivable that the structural modification induced at the R2 side chain of AL is responsible for such a decreased activity. Indeed, it is known that increasing the distance between the nitrogen and the phosphonate group in N-BP results into a decreased ability to inhibit farnesyl diphosphate synthase (FPPS) in OCs36. Overall, DM-22 showed a moderate ability to act as anti-resorptive agent and mitigated the effects of AL by preserving h-OCs viability even in the upper concentration range tested in our experiments.
At the cellular level, our findings support the hypothesis that AL directly inhibits osteogenic differentiation and cause cytotoxicity in osteoprogenitor cells. Whether AL stimulate or inhibits osteogenic differentiation of MSCs in vitro is a controversial matter in previous reports and appears to be dependent on the protocol used; our data correlate with similar findings obtained using h-MSCs37 and with work by Orriss et al. showing cytotoxic inhibition of osteoblast proliferation and function in rats by N-BPs in the low micromolar concentration range15. Moreover, in broad agreement with our observations, it was reported that AL inhibits bone nodule formation at concentrations 10-fold lower than those required to induce apoptosis in rabbit calvarian osteoblast38. By contrast, AL was shown to stimulate osteogenic differentiation of h-MSCs39 peaking at 10nM, a concentration 100-fold lower than the low-dose used in our experiments. One possible explanation for these diverging effects could be that AL triggers an hormetic dose-response during osteogenic differentiation of h-MSCs.
Compared to its parent molecule, DM-22 showed a substantially improved profile of safety and efficacy on h-MSCs induced to osteogenic differentiation within the same concentration range. DM-22 was devoid of cytotoxicity even at the high dose of 33μM; even though a slight decrease in cell proliferation was induced by DM-22 at the high concentration of 33μM, it did not affect cell viability and the ability of this compound to significantly increase the extent of mineralizing matrix as compared to control h-MSCs.
Semi-quantitative analysis of gene expression in h-MSCs provided further evidence of the potential osteoanabolic role of DM-22. h-MSCs stimulated with DM-22 showed consistantly higher mRNA expression than AL-stimulated samples across all the osteogenic targets and timepoints evaluated in our analysis. In particular the capacity of DM-22 to significantly stimulate COLL I expression marked a striking contrast with the effect induced by AL and may extend the applications of DM-22 to conditions characterized by defective synthesis of collagen.
Based on these data, DM-22 is the first N-BPs to display both anabolic and anti-resorptive functions in bone cells in vitro. Its biological activity on h-MSCs and h-OCs resembles that of H2S-releasing drugs previously tested by our group and others18,20,21. In particular, in vivo studies have ascertained that H2S-donors promote bone formation and prevent the bone wasting effect of estrogen deficiency in mice18, making it DM-22 a suitable candidate to achieve a long sought mitigation of the marked suppression of bone turnover exerted by traditional BPs.
Most BPs are characterized by poor bioavailability and very high affinity for the bone mineral matrix. Even tough further studies to test the in vivo bioavailability of DM-22 should be done in the near future, it is conceivable that the addition of aryl-isothiocianate moiety together with the “phosphonate” head of the molecule would lead to a local accumulation of DM-22 at the mineral bone surface, thus facilitating the local delivery of H2S. This would not only avoid potential systemic cytotoxicity but would locally replenish a molecule that is lowered in the pathological condition of bone loss due to osteoporosis18.
In conclusion, this work described the synthesis of a new H2S-hybrid N-BPs able to induce osteogenic differentiation of h-MSCs and, at the same time, retaining an anti-osteoclastogenic activity in vitro.
To the best of our knowledge, no other BPs elicit this combination of pharmacological effects in h-OCs and h-MSCs. The relevant features of the DM-22 is that it resulted from the winning conjugation of an anti-catabolic agent with a H2S-releasing moiety. The new multitasking N-BPs conjugate DM-22 may therefore pave the way to the development of novel therapies for bone loss providing new agents with a reduced toxicity profile.
Our work marks a significant expansion of H2S-based therapeutics, which stems from the recognition of the important role of H2S as a signaling molecule in several organs and tissues19,40, to the field of bone metabolism. This work will provide novel therapeutical opportunities in the field of bone loss therapies.
This work was financially supported by the following Grants: Grant Ricerca Finalizzata # PE-2011-02348395, from the italian Ministry of Health and Grant 5x1000 from Istituto Ortopedico Rizzoli.
V.C. and F.G. designed the study and conceived the experiments. S.R., L.G., V.a.C., C.F. and M.D. conducted the experiments. S.R., L.G., G.L. and F.G. analyzed the results. F.G., V.C. and S.R. drafted the manuscript. All authors reviewed the manuscript.
The authors declare that they have no competing interests.
Vincenzo Calderone and Francesco Grassi contributed equally to this work.
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.