Thiazide Diuretics Directly Induce Osteoblast Differentiation and Mineralized Nodule Formation by Interacting with a Sodium Chloride Co-Transporter in Bone
Melita M. Dvorak*,,
Cyrille De Joussineau*,
D. Howard Carter,
Trairak Pisitkun,
Mark A. Knepper,
Gerardo Gamba||,
Paul J. Kemp* and
Daniela Riccardi*,¶
* Cardiff University School of Biosciences and ¶ Cardiff Institute of Tissue Engineering and Repair, Cardiff, United Kingdom; Endocrine Research Unit, Department of Medicine, University of California, San Francisco, California; Turner Dental School, University of Manchester, Manchester, United Kingdom; Laboratory of Kidney and Electrolyte Metabolism, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland; and || Molecular Physiology Unit, Instituto Nacional de Ciencias Medicas y Nutricion Salvador Zubiran, Instituto de Investigaciones Biomedicas Universidad Nacional Autonoma de Mexico, Mexico City, Mexico
Correspondence: Dr. Daniela Riccardi, Cardiff University School of Biosciences, Museum Avenue, Cardiff, CF10 3US, UK. Phone: +44-29-20879132; Fax: +44-29-20874116; E-mail: riccardi{at}cardiff.ac.uk
Received for publication March 22, 2007.
Accepted for publication May 22, 2007.
Thiazide diuretics are used worldwide as a first-choice drugfor patients with uncomplicated hypertension. In addition totheir antihypertensive effect, thiazides increase bone mineraldensity and reduce the prevalence of fractures. Traditionally,these effects have been attributed to increased renal calciumreabsorption that occurs secondary to the inhibition of thethiazide-sensitive sodium chloride cotransporter (NCC) in thedistal tubule. The aim of the current study was to determinewhether thiazides exert a direct bone-forming effect independentof their renal action. We found that the osteoblasts of humanand rat bone also express NCC, suggesting that these bone-formingcells may be an additional target for thiazides. In vitro, NCCprotein was virtually absent in proliferating human and fetalrat osteoblasts, whereas its expression dramatically increasedduring differentiation. Thiazides did not affect osteoblastproliferation, but directly stimulated the production of theosteoblast differentiation markers runt-related transcriptionfactor 2 (runx2) and osteopontin. Using overexpression/knockdownstudies in fetal rat calvarial cells, we show that thiazidesincrease the formation of mineralized nodules, but loop diureticsdo not. Overall, our study demonstrates that thiazides directlystimulate osteoblast differentiation and bone mineral formationindependent of their effects in the kidney. Therefore, in additionto their use as antihypertensive drugs, our results suggestthat thiazides may find a role in the prevention and treatmentof osteoporosis.
It has been known for several decades that treatment of hypertensionwith thiazides has the beneficial side effect of strengtheningbone.1–5 To date, this bone-protective effect has beenattributed to thiazides’ acting at the distal nephronto inhibit the Na+-Cl– co-transporter (NCC).6 In favorof this hypothesis are two observations. First, patients withGitelman syndrome and the equivalent murine model, in whichNCC is nonfunctional as a result of a mutation in the SLC12A3gene, exhibit an increased bone mineral density.7,8 Second,patients with pseudohypoaldosteronism type II, who present withan increased NCC activity, exhibit reduced bone mineral density.9The mechanism by which genetic (Gitelman syndrome) or pharmacologic(thiazide treatment) inhibition of NCC results in enhanced bonemineral density has been hypothesized to be due to increasedcirculating serum calcium levels.10,11 This model proposes thatNCC inhibition in the distal tubule evokes hyperpolarization,increased electrical driving force for calcium reabsorption,and a subsequent decrease in urinary calcium loss.6 However,in patients with Gitelman syndrome, as in patients undergoingthiazide therapy, the expected increase in circulating parathyroidhormone levels that should accompany the increase in free ionizedplasma calcium concentration is not seen. In fact, plasma parathyroidhormone levels are either decreased or unchanged, and serumcalcium levels remain essentially normal.8,11 These observationssuggest that the increase in bone mineral density is probablynot directly related to the enhanced renal tubular calcium transportbut to a direct action of thiazides on bone. Potential mechanismsby which thiazides may exert their effect on bone are via aninhibition of osteoclast-mediated bone resorption and/or byan increase in osteoblastic bone formation. Although in vitrostudies indicate that thiazides are capable of reducing osteoclasticactivity independent of NCC,12,13 anabolic effects of thiazideshave not been demonstrated. Because NCC mRNA has been previouslyreported in a bone-derived cell line,14 we hypothesized thatthiazides increase bone mineral density by interacting directlywith an osteoblast NCC protein. Given the potential therapeuticimportance of long-term thiazide treatment in the preventionof age-related osteoporosis, it was the objective of this studyto investigate the effects of thiazides by establishing NCCexpression in human bone and to determine the effects of thiazideson osteoblast proliferation, differentiation, and mineralizednodule formation in osteoblast models.
For the immunologic detection of NCC expression in skeletaltissue, we used freshly frozen rat femur; freshly frozen humanmandible; and EDTA-decalcified, wax-embedded rat femur. At lowmagnification, we observed NCC immunoreactivity in human andrat cortical and trabecular bone, in both undecalcified, frozen(Figure 1, A and B) and decalcified, paraffin wax–embeddedhuman (data not shown) and rat bone (Figure 1C). No nonspecificimmunoreactivity was detected when the primary antibodies wereomitted (Figure 1D). NCC immunoreactivity was also observedin snap-frozen sections of rat femur (Figure 1, E through I).With the use of two different antibodies raised against twodifferent epitopes of either human or rat proteins, NCC-specificimmunoreactivity was localized to cells of the osteoblasticlineage (particularly osteoblasts) in both human and rat bone(Figure 1, A, G, and top of K). NCC protein was also presentin some but not all osteocytes (Figure 1, A, C, and H), perhapsindicating that cells at different stages of differentiationdisplay differential expression of this protein. Occasionally,osteoclasts in the rat cryosections (Figure 1I) but not thosein sections of decalcified bone (data not shown) also expressedNCC; immunostaining in osteoclasts was absent in human freshlyfrozen and paraffin sections (data not shown). Furthermore,no NCC immunoreactivity was detectable in either human or ratcartilage (data not shown). Positive control experiments carriedout on rat kidney cryosections (Figure 1J) confirmed the expressionof NCC exclusively in the distal convoluted tubule,1 thus demonstratingthe specificity of the antibody. To investigate the size ofhuman NCC protein in osteoblasts, we performed Western analysison crude membrane-enriched fractions from human osteoblast-derivedMG63 cells. Human kidney cortex was used as the control tissue.Results showed comparable immunoreactivities of the expectedmolecular weights for the NCC monomer and dimer (Figure 1K,top). Figure 1K also shows that MG63 cells expressed the type1 Na+-K+-2Cl– co-transporter (the ubiquitously expressedform NKCC1; Figure 1K, bottom),1 whereas they lack the kidney-specificisoform of the type 2 Na+-K+-2Cl– co-transporter (NKCC2;data not shown).1 Western analysis was performed on homogenatesof established osteoblast cellular models after they had reachedconfluence. These were rat osteosarcoma UMR-106 cell line, mouseosteoblast-derived 2T3 clonal cell line, and fetal rat calvaria(FRC). To varying degrees, all of these osteoblast-derived linesexpressed immunoreactivity of the expected size for NCC (Figure 2A).After enzymatic deglycosylation with N-glycanase, the molecularweight of the NCC monomeric protein in MG63 cells was reducedfrom approximately 160 kD to approximately 120 kD, a size closeto the unglycosylated core protein (Figure 2B). It is interestingthat Western analysis also showed that NCC protein expressionlevels in FRC cells (Figure 2C) and in MG63 cells (Figure 2D)were virtually undetectable in proliferating cells (up to 10d in culture). In contrast, in postconfluent, differentiatingFRC and MG63 cells, NCC protein expression levels increasedup to 15 d after confluence.
Figure 1. Sodium chloride co-transporter (NCC) is expressed in bone. Using human NCC antibodies, specific immunofluorescence is evident in undecalcified frozen human (A) and rat (B) sections. NCC is expressed in reversal lines (A; arrows), in osteocytes (A; small arrows), and lining osteoblasts (B; arrows). (C) Bright-field photomicrograph of EDTA-decalcified paraffin section of rat femur shows NCC immunoperoxidase activity (brown) in reversal lines (arrows), osteoblasts (arrowheads), and some osteocytes (arrowhead +). A proportion of osteocytes are negative for NCC (arrowhead –). (D) Omission of primary antibodies does not result in staining. (E through I) Photomicrographs of rat femur cryosections show that NCC immunofluorescence is in reversal lines (E and F; arrows), osteoblasts (G; arrows), osteocytes (H; arrows), and some osteoclasts (I; arrows). (J) NCC-specific, distal convoluted tubule staining is present in the rat kidney. Bar = 20 µm throughout. (K, top) NCC immunoreactivity in osteoblast-derived MG63 cells (ob; 30 µg of total homogenate loaded) is comparable to that detected in human kidney cortex sample (kc; 30 µg), with predicted sizes for the NCC monomer (M) and dimer (D) of approximately 160 and 320 kD, respectively. (K, bottom) Expression of the ubiquitously expressed type 1 Na+-K+-2Cl– co-transporter (NKCC1) is also expressed in MG63 cells using an anti-rat NKCC1 antibody (ob, 30 µg of total homogenate loaded). The figure also shows that, as expected, NKCC1 is low abundant in human kidney cortex and outer medulla. om, kidney outer medulla, 10 µg.
Figure 2. NCC is expressed in differentiating osteoblasts. (A) Western analysis of protein homogenates extracted from established models of osteoblasts (human UMR-106, murine 2T3, and fetal rat calvaria [FRC]) reveals that NCC immunoreactivity is of comparable size to that observed in rat kidney. (B) Enzymatic deglycosylation of protein extracts from the human osteoblast-derived cell line MG63 indicates a reduction in the NCC monomer molecular weight from approximately 160 kD to approximately 120 kD. Control protein exposed to 37°C for 60 min without N-glycanase showed presence of the fully glycosylated protein. (C and D) NCC protein expression is absent in preconfluent, proliferating FRC (C) and MG63 cells (D). In both models, NCC expression increases as the differentiation process progresses and peaks at approximately 2 wk after confluence. Note that 0 d after confluence represents the beginning of differentiation.
It is conceivable that the osteoanabolic effects of metolazonecould be ascribed to stimulation of cell proliferation. In linewith the absence of NCC expression in preconfluent, proliferatingosteoblasts, FRC cell proliferation was not affected by metolazone(1 to 100 µM) for up to 2 wk in culture (Figure 3A, threeindependent experiments performed in triplicate). Type I collagenis the first marker of osteoblast differentiation.15 Our datashow that neither metolazone nor chlorothiazide had any effecton the production of type I collagen (Figure 3B, three independentexperiments performed in triplicate). In contrast, metolazoneproduced a concentration-dependent increase in the expressionof later osteoblast differentiation markers runt-related transcriptionfactor 2 (Runx2; Figure 3, C and D, five independent observationsfrom three independent cell isolations) and osteopontin (Figure 3,E and F, five independent observations from three independentcell isolations) in both FRC (Figure 3, C and E) and MG63 (Figure 3,D and F) cells. In addition, chronic metolazone treatment perse increased NCC expression in a dosage-dependent manner (datanot shown).
Figure 3. Metolazone (MET) does not affect proliferation but stimulates the expression of osteoblast differentiation markers. (A) FRC cell proliferation (for up to 2 wk in culture) is not affected by increasing dosages of the thiazide-like metolazone. (B) Consistent with the lack of NCC expression at days 1 to 3 after confluence (see Figure 2, C and D), the activity of collagen 1A (coll1A), the major structural component of the organic matrix and early differentiation marker, is not affected by 48-h treatment with MET (10 µM) or with chlorothiazide (CTZ; 10 µM). (C through F) Western analyses show increased expression of osteoblast markers runt-related transcription factor 2 (Runx2; C [*P < 0.05 at 10 µM, ANOVA, Tukey post hoc test] and D [*P < 0.01, unpaired t test]) and osteopontin (E [*P < 0.05, ANOVA, Tukey post hoc test] and F [*P < 0.05, unpaired t test]) in rat (C and E) and human (D and F) osteoblast models in response to increasing concentrations of metolazone (7 d of treatment). Histograms represent mean band densities and indicate that the levels of expression of Runx2 or osteopontin, normalized for the levels of expression of -actin in the same samples, are significantly increased.
The effect of thiazides and of loop diuretics on mineralizednodule formation was tested in postconfluent FRC cells keptin culture up to 3 wk, and mineralization was visualized usingvon Kossa staining. Figure 4 shows that both metolazone andchlorothiazide treatment of FRC cells induced a dramatic, concentration-dependentincrease in mineralized nodule formation. Figure 4 also showsthat, under the same experimental conditions, the loop diureticbumetanide did not increase mineralization of FRC cells. Theseresults indicate that thiazides increase mineralization andstrongly suggest that this effect is dependent on NCC expression.
Figure 4. Thiazides stimulate mineralized nodule formation by FRC cells. (A) Continuous treatment of FRC cells with MET (1 to 100 µM) and CTZ (1 and 10 µM) for 10 to 21 d after confluence induces dosage-dependent increases in mineralized nodule formation (MET: n = 3; *P < 0.05, **P < 0.01; CTZ: n = 3; *P < 0.05). The loop diuretic bumetanide (BMT; 10 µM) does not significantly affect nodule formation. (B) Representative von Kossa staining of FRC cells (15 d after confluence) treated with MET (1 to 100 µM) shows a concentration-dependent increase in the number of mineralized nodules (top), whereas no effects on FRC cell mineralization are seen in the presence of BMT (10 µM). (Bottom) The number of mineralized nodules is comparable for MET- and CTZ-treated postconfluent FRC cells (representative of three independent cell isolations).
To confirm that NCC is required for the thiazide-evoked mineralization,we genetically manipulated NCC expression in postconfluent FRCcells. First, to rule out nonspecific effects, we transfectedpostconfluent FRC cells with an empty vector (pcDNA3.1) andmeasured mineralization in the presence or absence of concentrationsof metolazone known to evoke statistically significant effectsin nontransfected cells (i.e., 10 µM; see Figure 4). Figure 5Ashows that the effects of metolazone were maintained even afterplasmid delivery. Indeed, 10 µM metolazone increased thenumber of mineralized nodules per well from 285.9 ± 16.3to 470.2 ± 18.6 (n = 9 observations from three independentcell isolations; P < 0.05). In the absence of metolazonetreatment, neither overexpression nor antisense knockdown evokedsignificant changes in mineralized nodule formation (Figure 5B).In contrast, overexpression of NCC increased metolazone-dependentmineralized nodule formation by approximately 50% (49.3 ±13.0%; n = 12 replicates from four independent cell isolations),an effect that was completely prevented by NCC knockdown (–13± 8.7%; n = 12 replicates from four independent cellisolations; P < 0.05).
Figure 5. MET-induced mineralization is mediated by NCC. (A) MET treatment evokes a significant increase in mineralization in postconfluent FRC cells transiently transfected with empty vector (pcDNA3.1; *P < 0.05; n = 9 from three independent cell isolations). (B) MET-dependent (10 µM) mineralization in postconfluent FRC cells is dramatically enhanced by NCC overexpression (sense [S]), and this is significantly blunted by NCC knockdown (antisense [AS]; P < 0.05; n = 12 observations made from four independent cell isolations). Note that NCC overexpression increases mineralization of postconfluent FRC cells only after MET treatment. Data are expressed as change in number of nodules per well, defined as [(NodulesNCC – NodulespcDNA)/NodulespcDNA], where NodulespcDNA is the number of mineralized nodules in cells transfected with empty vector and NodulesNCC is the number of mineralized nodules in cells transfected with NCC sense or antisense, as indicated below bars. (C) Representative von Kossa staining of FRC cells after the treatments indicated below each well.
Long-term thiazide treatment is associated with a reductionin the risk for hip and wrist fractures in postmenopausal womenand elderly men.1–5 This bone-sparing effect is thoughtto occur through blockage of the renal sodium chloride co-transporterNCC and subsequent reduction in urinary calcium excretion. Whetherthiazides directly effect new bone formation independent oftheir renal action has never been demonstrated. Here, we showthat NCC is expressed in freshly frozen and decalcified sectionsof human and rat bone; in cells of the osteoblast lineage, particularlyosteoblasts; and, to a lesser extent, in the osteocytes, inboth rat and human bone.
Immunoblotting performed on crude membrane extracts of freshlyisolated (FRC), osteosarcoma-derived (UMR-106) and virally transformed(2T3) cells confirmed that NCC in osteoblasts is of an equivalentmolecular mass as its renal counterpart. The renal NCC containstwo N-linked glycosylation sites that are important for sensitivityto thiazides.16 Enzymatic deglycosylation of NCC in MG63 cellsdemonstrated the presence of carbohydrate residues in the osteoblastprotein. Taken together, these observations show that NCC isexpressed in bone and suggest that the osteoblast NCC, likethe kidney NCC, may also be a target for thiazide diuretics.Thus, the bone-protective effects of thiazides may be due totheir direct interaction with this protein in the osteoblasts.
Bone formation is characterized by a distinctive sequence ofevents beginning with the commitment of mesenchymal cells toosteoblast lineage, followed by osteoblastic proliferation anddifferentiation. This sequence of events culminates in the formationof mineralized extracellular matrix by terminally differentiatedosteoblasts.15 Previous studies have ruled out an effect ofhydrochlorothiazide on human bone marrow stromal cells, suggestingthat the thiazide-dependent enhanced bone mineral density isnot due to an increase in osteoblast progenitors.17 However,the effects of thiazides on osteoblast proliferation are controversial,with either an increase or no change in proliferation havingbeen reported,13,18 depending on species. Therefore, we investigatedthe effects of thiazides on osteoblast proliferation, differentiation,and mineralization to ascribe a potential role of NCC in eachof these events. To exclude the possibility that the effectsof thiazides could be indirect (i.e., a consequence of theirrenal actions), we carried out these studies in vitro, usingboth primary and established models of osteoblasts of eitherrat or human derivation, namely FRC and MG63 cells, respectively.First, we tested the effects of metolazone and of chlorothiazideon FRC cells and found that neither of these compounds affectedproliferation rates for up to 10 d in culture. Consistent witha lack of effect of thiazides on osteoblast proliferation, NCCprotein expression levels in FRC cells and in MG63 cells werenegligible during the proliferative phase (i.e., up to 10 din culture). At this point, FRC cells had stopped proliferatingand begun their differentiation process, evident from increasedexpression of alkaline phosphatase (data not shown) and aggregationof cells into nodular areas. In postconfluent, differentiatingosteoblasts, NCC protein expression levels in both FRC and MG63cells gradually rose and peaked at approximately 2 wk afterconfluence, indicating that NCC acts as a novel potential osteoblastdifferentiation marker. Having ascertained the presence of NCCin postconfluent, differentiating osteoblasts, we then testedthe effects of thiazides on the expression levels of known osteoblastmarkers. Type I collagen is the major structural component ofthe organic matrix of bone and one of the earliest marker ofosteoblast differentiation.15,19 Our data show that, at a stagewhen NCC protein expression is absent (i.e. 2 d after confluence),neither metolazone nor chlorothiazide had any effect on theproduction of type I collagen. In contrast, metolazone produceda concentration-dependent increase in the expression of Runx2,a master osteoblast-specific transcription regulator, in bothFRC and MG63 cells. The ability of thiazides to regulate theosteoblast differentiation process was tested by measurementof the levels of osteopontin, a soluble, secreted phosphoproteinthat is a component of the bone mineralized extracellular matrix(also called bone sialoprotein I, secreted phosphoprotein, 2ar,and bp69). In both FRC and MG63 cells, metolazone significantlyincreased osteopontin production. These data, taken togetherwith our observations that metolazone was ineffective duringstages when NCC expression was undetectable and upregulatedNCC expression in MG63 cells, suggest that thiazides act directlyon differentiating osteoblasts through NCC.
The ultimate osteoblast differentiation marker is the formationof new bone, and FRC cells in culture form calcified nodulesthat can be visualized by von Kossa staining. Metolazone treatmentof postconfluent FRC cells induced a dramatic, concentration-dependentincrease in mineralized nodule formation. This effect was specificto thiazide diuretics because it could be mimicked by chlorothiazidebut could not be emulated by the loop diuretic bumetanide, eventhough our results show that osteoblast models express NKCC1,the molecular target for loop diuretics.
As proof of concept, we used plasmid delivery of NCC antisenseand sense cDNA in FRC cells and assessed the effects of metolazoneon mineralized nodule formation. Overexpression of NCC resultedin a significant increase in metolazone-induced mineralizednodule formation. This increase was completely prevented byNCC knockdown with the antisense construct. This effect is evenmore striking when one considers that the efficiency of transfectionof the plasmid in primary cells is only approximately 10% (estimatedwith co-transfection with a fluorescence reporter; data notshown). The evidence that the increase in mineralization isobserved only in the presence of thiazides suggests the possibilitythat NCC might act as receptor for thiazides, rather than aco-transporter, although such an interpretation would not explainwhy patients with Gitelman syndrome exhibit an increased bonemineral density.
Finally, it has been suggested that thiazides prevent bone lossbecause they reduce acid production by inhibiting carbonic anhydraseactivity in osteoclasts.20 It is interesting that we demonstrateNCC immunostaining in some osteoclasts of cryoprepared rat femorabut not human bone. Given that osteoclast staining was observedin five different preparations, with the appropriate positiveand negative controls, we believe that NCC immunofluorescencein a subpopulation of osteoclasts is real. This observationopens the possibility that thiazides might affect osteoclasticfunction through NCC in addition to creating alkalinizationof the resorption milieu. The dual action of thiazide drugson both osteoblast and osteoclast function could account forthe observed reduced remodeling in patients taking such drugsin the absence of changes in plasma circulating parathyroidhormone levels.
The main finding of this study is the demonstration that thiazidesdirectly stimulate osteoblast differentiation and mineral productionindependent of their renal action. This effect of thiazidesis concentration dependent, is not mimicked by loop diuretics,is not due to increased osteoblast proliferation, and is enhancedby NCC overexpression. Together with the observations that thiazidetreatment and inactivating mutations of NCC are associated withan increased bone mineral density in humans and in knockoutmurine models, our findings support a pivotal role for the osteoblastNCC in mediating thiazide-induced bone formation. Thiazide diureticsare inexpensive and exhibit a good safety profile. Our findingssuggest that it might be possible to develop osteoblast-specificthiazides as part of osteoporosis prevention and therapeuticprograms.
Animals
Sprague-Dawley rats (Charles River Laboratories, Wilmington,Kent, UK) were killed by cervical dislocation and used in accordanceto the UK Animals Scientific Procedures Act of 1986.
Cell Culture
The human osteoblast cell line MG-63 was cultured as describedpreviously.18 FRC cells were isolated as described previously15;FRC, rat UMR-106, and mouse osteoblast-derived 2T3 cells werecultured as described previously.19 Metolazone (Sigma-Aldrich,Poole, Dorset, UK) was dissolved at 37°C for 2 h in theculture medium before being added to the cells. This procedurewas repeated every 3 d. From confluence onward, the media weresupplemented with ascorbic acid (284 µM for MG-63, UMR-106,and 2T3 and 568 µM for FRC cells; Sigma-Aldrich) and -glycerophosphate(3 mM; Sigma-Aldrich).
Western Blotting
SDS-PAGE immunoblotting of MG-63, UMR-106, 2T3, and FRC cells(whole-cell lysates) and human and rat kidney was performedas described previously,21,22 using the following primary antibodies:Affinity-purified rabbit anti-human NCC polyclonal antibodies(1:1000),22 affinity-purified rabbit anti-rat NCC polyclonalantibodies (1:5000),23 affinity-purified rabbit anti-rat NKCC1(a gift of Dr. R. James Turner, National Institute of Dentaland Craniofacial Research, Bethesda, MD; 1:1000), affinity-purifiedrabbit anti-rat NKCC2 polyclonal antibodies (1:1000),24 mouseanti-human Runx2 mAb (a gift of Dr. Andre von Wijnen, Universityof Massachusetts, Worcester, MA; 1:4000), mouse anti-rat osteopontinmAb (Iowa Hybridoma Bank, Iowa City, IA; 1:4000), and mouseanti–-actin mAb (Abcam, Cambridge, Cambridgeshire, UK;1:10,000). For SDS-PAGE, samples were heated to 60°C for10 to 15 min in a 5x Laemmli sample buffer, in the presenceof dithiothreitol (30 mg/ml) or -mercaptoethanol (143 mM). Proteinswere resolved by SDS-PAGE and transferred onto nitrocellulosemembranes before blocking (30 min; PBS containing 5% semiskimmilk powder or Odyssey blocking buffer; Li-Cor, Lincoln, NE)and antibody incubations (1 to 12 h). Membranes were washedin Tween-Tris–buffered saline (15 mM Tris [pH 8], 150mM NaCl, and 0.1% [vol/vol] Tween 20). Antibody binding wasvisualized by an enhanced chemiluminescence system (AmershamPharmacia Biotech, Little Chalfont, Buckinghamshire, UK). Humankidney samples for immunoblotting were obtained from the unaffectedportion of a kidney that had been resected because of a renaltumor (approved as exempt from review by Office of Human SubjectsResearch).
Immunohistochemistry
Undecalcified, snap-frozen rat femora (2 mo) and human mandibleresections and EDTA-decalcified rat femora were prepared andused for immunofluorescence and immunoperoxidase experimentsas described previously.19 Anti-human and anti-rat NCC polyclonalantibodies (see previous section) were used at 1:10 and 1:100dilutions, respectively. The human mandible tissue was fromneck resections from patients with squamous cell carcinoma invadingbone (ethical approval and signed informed patient consent wereobtained).
Effect of Metolazone on Osteoblast Proliferation
Cells were plated in 12-well plates at a density of 5000 cells/cm2and were treated with metolazone (1 to 100 µM) for upto 2 wk. At each time point, triplicate wells were trypsinizedand the cells were counted with a Coulter counter (Beckman Coulter,High Wycombe, Buckinghamshire, UK).
Effect of Metolazone on Collagen 1A Content
The effects of metolazone on collagen 1A content were quantifiedas measurements of the hydroxyproline content in FRC cells after48 h of treatment. After HCl digestion for 24 h at 110°C,the samples were freeze-dried to remove the acid, diluted indistilled water, and oxidized with chloramine followed by couplingwith dimethylamino benzaldehyde at 70°C for 10 to 20 min.The colored product was measured at 550 nm. Standards of 1 to10 µg/ml hydroxyproline were used to calculate the standardcurve.
Effect of Metolazone on Osteoblast Differentiation
FRC and MG63 cells were cultured in 35-mm dishes (20,000 cells/cm2)and treated with metolazone or chlorothiazide, for 2 d afterconfluence for early differentiation experiments or for 7 to14 d after confluence for mineralization experiments. For immunoblotting,the cells were washed in PBS and lysed in RIPA buffer as describedpreviously.21 Semiquantitative changes in Runx2 and osteopontinimmunoreactivities were normalized for the levels of -actin(mouse monoclonal; Abcam, Cambridge, Cambridgeshire, UK).
Effect of Thiazides on Mineralized Nodule Formation
FRC cells were treated from confluence up to 3 wk. Mineralizednodules were visualized by von Kossa staining, as describedpreviously.19,21 The images of mineralized nodules were capturedon a flatbed scanner, and image analysis software (Scion Image,Frederick, MD) was used to count the number of mineralized nodules.
Statistical Analyses
The statistical significance was assessed by one-way ANOVA withthe Tukey post hoc test or with the unpaired t test, as appropriate.Observations were considered to be statistically significantdifferent at P 0.05.
This work was funded by the Arthritis Research Campaign (R0626to D.R. and D.H.C.) and The Wellcome Trust (CRIG 070159 to D.R.and G.G.).
We thank Prof. V. Duance and Dr. S. Gilbert, Cardiff University,for the help with the collagen I assay; Dr. A. von Wijnen, Universityof Massachusetts, for the gift of the Runx2 antibody; Dr. A.Mee, University of Manchester, UK, for the gift of MG-63 cells;N. Vasquez, National University of Mexico, for technical help;and the Iowa Hybridoma Bank for the osteopontin antibody.
Footnotes
Published online ahead of print. Publication date availableat www.jasn.org.
M.M.D. and C.D.J. contributed equally to this work.
Supplemental information for this article is available onlineat http://www.jasn.org/.
Gamba G: Molecular physiology and pathophysiology of electroneutral cation-chloride cotransporters.
Physiol Rev 85
: 423
–493, 2005[Abstract/Free Full Text]
Wasnich RD, Benfante RJ, Yano K, Heilbrun L, Vogel JM: Thiazide effect on the mineral content of bone.
N Engl J Med 309
: 344
–347, 1983[Abstract]
LaCroix AZ, Wienpahl J, White LR, Wallace RB, Scherr PA, George LK, Cornoni-Huntley J, Ostfeld AM: Thiazide diuretic agents and the incidence of hip fracture.
N Engl J Med 322
: 286
–290, 1990[Abstract]
Jones G, Nguyen T, Sambrook PN, Eisman JA: Thiazide diuretics and fractures: Can meta-analysis help?
J Bone Miner Res 10
: 106
–111, 1995[Medline]
Brickman AS, Massry SG, Coburn JW: Changes in serum and urinary calcium during treatment with hydrochlorothiazide: Studies on mechanisms
J Clin Invest 51
: 945
–954, 1972[Medline]
Gesek FA, Friedman PA: Mechanism of calcium transport stimulated by chlorothiazide in mouse distal convoluted tubule cells.
J Clin Invest 90
: 429
–438, 1992[Medline]
Cruz DN: The renal tubular Na-Cl co-transporter (NCCT): A potential genetic link between blood pressure and bone density?
Nephrol Dial Transplant 16
: 691
–694, 2001[Free Full Text]
Nicolet-Barousse L, Blanchard A, Roux C, Pietri L, Bloch-Faure M, Kolta S, Chappard C, Geoffroy V, Morieux C, Jeunemaitre X, Shull GE, Meneton P, Paillard M, Houillier P, De Vernejoul MC: Inactivation of the Na-Cl co-transporter (NCC) gene is associated with high BMD through both renal and bone mechanisms: Analysis of patients with Gitelman syndrome and Ncc null mice.
J Bone Miner Res 20
: 799
–808, 2005[CrossRef][Medline]
Mayan H, Vered I, Mouallem M, Tzadok-Witkon M, Pauzner R, Farfel Z: Pseudohypoaldosteronism type II: Marked sensitivity to thiazides, hypercalciuria, normomagnesemia, and low bone mineral density.
J Clin Endocrinol Metab 87
: 3248
–3254, 2002[Abstract/Free Full Text]
Kamel HK: Low-dose thiazide and bone density.
Ann Intern Med 136
: 252
–253, 2002[Free Full Text]
Van Den Berg CJ, Tucker RM, Dousa TP: Idiopathic hypercalciuria: Hydrochlorothiazide decrease urinary calcium without altered renal response to parathyroid hormone.
J Clin Endocrinol Metab 55
: 23
–26, 1982[Abstract]
Hall TJ, Schaueblin M: Hydrochlorothiazide inhibits osteoclastic bone resorption in vitro.
Calcif Tissue Int 55
: 266
–268, 1994[CrossRef][Medline]
Lalande A, Roux S, Denne MA, Stanley ER, Schiavi P, Guez D, De Vernejoul MC: Indapamide, a thiazide-like diuretic, decreases bone resorption in vitro.
J Bone Miner Res 16
: 361
–370, 2001[CrossRef][Medline]
Barry EL, Gesek FA, Kaplan MR, Hebert SC, Friedman PA: Expression of the sodium-chloride cotransporter in osteoblast-like cells: Effect of thiazide diuretics.
Am J Physiol 272
: C109
–C116, 1997[Medline]
Aronow MA, Gerstenfeld LC, Owen TA, Tassinari MS, Stein GS, Lian JB: Factors that promote progressive development of the osteoblast phenotype in cultured fetal rat calvaria cells.
J Cell Physiol 143
: 213
–221, 1990[CrossRef][Medline]
Hoover RS, Poch E, Monroy A, Vazquez N, Nishio T, Gamba G, Hebert SC: N-Glycosylation at two sites critically alters thiazide binding and activity of the rat thiazide-sensitive Na(+):Cl(–) cotransporter.
J Am Soc Nephrol 14
: 271
–278, 2003[Abstract/Free Full Text]
Kim CH, Kim SW, Kim GS: Effects of hydrochlorothiazide and furosemide diuretics on human bone marrow stromal osteoprogenitor cells.
Metabolism 49
: 17
–21, 2000[Medline]
Aubin R, Menard P, Lajeunesse D: Selective effect of thiazides on the human osteoblast-like cell line MG-63.
Kidney Int 50
: 1476
–1482, 1996[Medline]
Dvorak MM, Siddiqua A, Ward DT, Carter DH, Dallas SL, Nemeth EF, Riccardi D: Physiological changes in extracellular calcium concentration directly control osteoblast function in the absence of calciotropic hormones.
Proc Natl Acad Sci U S A 101
: 5140
–5145, 2004[Abstract/Free Full Text]
Raisz LG, Simmons HA, Thompson WJ, Shepard KL, Anderson PS, Rodan GA: Effects of a potent carbonic anhydrase inhibitor on bone resorption in organ culture.
Endocrinology 122
: 1083
–1086, 1993
Dallas SL, Park-Snyder S, Miyazono K, Twardzik D, Mundy GR, Bonewald LF: Characterization and autoregulation of latent transforming growth factor beta (TGF beta) complexes in osteoblast-like cell lines. Production of a latent complex lacking the latent TGF beta-binding protein.
J Biol Chem 269
: 6815
–6821, 1994[Abstract/Free Full Text]
Biner HL, Arpin-Bott MP, Loffing J, Wang X, Knepper M, Hebert SC, Kaissling B: Human cortical distal nephron: Distribution of electrolyte and water transport pathways.
J Am Soc Nephrol 13
: 836
–847, 2002[Abstract/Free Full Text]
Kim GH, Masilamani S, Turner R, Mitchell C, Wade JB, Knepper MA: The thiazide-sensitive Na-Cl cotransporter is an aldosterone-induced protein.
Proc Natl Acad Sci U S A 95
: 14552
–14557, 1998[Abstract/Free Full Text]
Kim GH, Ecelbarger CA, Mitchell C, Packer RK, Wade JB, Knepper MA: Vasopressin increases Na-K-2Cl cotransporter expression in thick ascending limb of Henle’s loop.
Am J Physiol 276
: F96
–F103, 1999[Medline]
Top
Abstract
Introduction
-glycerophosphate (3 mM; Sigma-Aldrich). 
0.05. 
