Human mesenchymal stem cells (hMSCs), the precursors of osteoblasts during osteogenesis, play a role in the balance of bone formation and resorption, but their functioning in uremia has not been well defined. To study the effects of the uremic milieu on osteogenic properties, we applied an in vitro assay culturing hMSCs in osteogenic medium supplemented with serum from healthy donors and from uremic patients on hemodialysis. Compared to control, serum from uremic patients induces, in hMSC cultures, a modification of several key regulators of bone remodeling, in particular a reduction of the ratio Receptor Activator of Nuclear factor Kappa B Receptor (RANKL) over osteoprotegerin, indicating an adaptive response of the system to favor osteogenesis over osteoclastosis. However, the levels of osteopontin, osteocalcin, and collagen type I, are increased in cell medium, while BMP-2, and alizarin red staining were decreased, pointing to a reduction of bone formation favoring resorption. Selected uremic toxins, such as p-cresylsulfate, p-cresylglucuronide, parathyroid hormone, indoxyl sulfate, asymmetric dimethylarginine, homocysteine, were able to mimic some of the effects of whole serum from uremic patients. Serum from cinacalcet-treated patients antagonizes these effects. Hydrogen sulfide (H2S) donors as well as hemodialysis treatment are able to induce beneficial effects. In conclusion, bone modifications in uremia are influenced by the capability of the uremic milieu to alter hMSC osteogenic differentiation. Cinacalcet, H2S donors and a hemodialysis session can ameliorate the hampered calcium deposition.
Citation: Lanza D, Perna AF, Oliva A, Vanholder R, Pletinck A, Guastafierro S, et al. (2015) Impact of the Uremic Milieu on the Osteogenic Potential of Mesenchymal Stem Cells. PLoS ONE 10(1): e0116468. https://doi.org/10.1371/journal.pone.0116468
Academic Editor: Giovanni Li Volti, University of Catania, ITALY
Received: October 14, 2014; Accepted: December 9, 2014; Published: January 30, 2015
Copyright: © 2015 Lanza et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Data Availability: All relevant data are within the paper.
Funding: This work was supported by the Italian Society of Nephrology (SIN), http://www.sin-italy.org, GRANT “Ricercando 2011” to AFP. Project name: “Exploring the osteogenic properties of mesenchymal stem cells for bone structural –functional repair in hemodialysis patients: role of uremic toxins.” The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
In chronic kidney disease (CKD) and especially in patients on hemodialysis, chronic kidney disease-mineral and bone disorder (CKD-MBD) frequently affects quality of life, morbidity and mortality . These alterations are to a large extent induced by the determinants of parathyroid hormone (PTH) secretion [2,3], and are also influenced by uremic retention solutes, especially p-cresylsulfate (pCS) and p-cresylglucuronide (pCG), indoxyl sulfate, asymmetric dimethylarginine (ADMA), and homocysteine [4–10].
Bone marrow-derived stem cells are either hematopoietic or non-hematopoietic (mesenchymal). Human bone marrow-derived mesenchymal stem cells (hMSCs) (marrow stromal cells)  can differentiate into osteoblasts, as well as chondrocytes, adipocytes and other cell types [12–14].
During hMSCs osteoblastic differentiation, several molecules, such as osteoprotegerin (OPG), receptor activator of nuclear factor kappa B ligand (RANKL), osteopontin, osteocalcin, collagen type I, bone morphogenic protein-2 (BMP-2), and alkaline phosphatase play a key role in this complex process, resulting in matrix formation and calcium deposition (Fig. 1).
The horizontal line indicates the time period from cell plating until the beginning of calcification. Dashed bar indicates the detection of calcium deposition (ARS assay). Arrows indicate medium change with replacement of fresh serum from uremic patients or from healthy control donors. Boxes highlight specific marker detection or treatments. Early markers: OPG, sRANKL. Late markers: alkaline phosphatase, osteopontin, osteocalcin, collagen type I, BMP-2. Calcium deposition was detected by ARS assay. Differential morphology of hMSC before (fibroblast-like) and at the end of osteogenic differentiation process (spherical cells with calcium deposition) is also shown (panels A and B, respectively).
OPG is a soluble cytokine belonging to the Tumor Necrosis Factor receptor superfamily acting as negative regulator of RANKL, which instead induces osteoclast differentiation in bone marrow through its receptor RANK. OPG acts as a decoy receptor that blocks RANKL, thus preventing RANK activation, osteoclast differentiation and bone resorption. RANKL is also produced by osteoblasts. The ratio between RANKL and OPG is considered to be a key index in bone formation/resorption: in case of high RANKL/OPG concentration ratio, osteoclastosis is favored, and when the opposite is present, osteoblastogenesis is preferred . In hemodialysis patients, serum OPG concentration is higher, and serum RANKL concentration is lower than controls [16,17].
Osteopontin is an acidic phosphoprotein of bone, and produced by fully differentiated osteoblasts, which is involved in regulation of mineralization by acting as an inhibitor of apatite crystal growth, as well as promoting osteoclast function [18,19]. Serum osteopontin concentration is increased in hemodialysis patients [20,21].
Osteocalcin is a protein, which, along with collagen type I, is produced by osteoblasts and together combine extracellularly to form osteoid, the organic substrate upon which mineralization occurs . In CKD-MBD, both markers are correlated with serum PTH levels .
The serum concentration of bone-specific alkaline phosphatase reflects the cellular activity of osteoblasts, and is a useful marker of bone formation, also in CKD-MBD .
In this study, we investigated the effects of serum from uremic patients on hemodialysis on the osteogenic differentiation of hMSCs, which are still largely unknown, through the analysis of the various markers. An in vitro assay system in which hMSCs cultured with osteogenic medium supplemented with healthy or uremic patient serum was applied. The effect of selected uremic solutes were tested, in particular pCS, pCG, PTH, indoxyl sulfate, ADMA, homocysteine, on the above osteogenic differentiation markers.
The effects of the calcimimetic cinacalcet, a drug utilized in the therapy of CKD-MBD, as well as the effects of hydrogen sulfide (H2S) donors were also investigated. H2S, an endogenous vasodilator and antioxidant, which concentration is reduced in uremia , is involved in osteoblast proliferation through enhancement of the transcriptional levels of alkaline phosphatase, osteocalcin, and collagen , as well as OPG/RANKL .
1 Products and reagents
All cell culture materials were purchased from Gibco (Life Technology, Carlsbad CA, USA), and all chemicals, as well as PTH, indoxyl sulfate, ADMA, L-homocysteine and Diallyl disulfide (DADS) were from Sigma Chemical Co. (St. Louis, MO, USA), when not otherwise specified. Diallyl trisulfide (DATS) was from Cayman Chemical Co. (La Jolla, Ca, U.S.A.). PrestoBlue Cell Viability Reagent was purchased from Invitrogen (Life Technology, Carlsbad, CA, USA). pCS and pCG were synthesized at the organic chemistry department of the Ghent University, Belgium. pCS was synthesized according to Feigenbaum and Neuberg as a potassium salt  and pCG was synthesized and purified based on the work of Van der Eycken et al . All reagents used were of analytical grade. Vacutainer tubes were purchased from Becton Dickinson Italia S.p.A.
2 Preparation of hMSCs
The bone marrow aspirate samples were obtained from the right or left posterior iliac crest of 17 patients (9 males and 8 females, aged >18 years) diagnosed with Non-Hodgkin Lymphoma without bone marrow involvement and with normal renal function. The bone marrow aspiration was performed before core biopsy using the same bone marrow core biopsy needle (Biopsybell, Modena, Italy). Written consent was obtained from all patients, and the study received approval from the institutional ethics committee of the Second University of Naples (Protocol No. 283 of 12 July 2013).
HMSC cultures were initiated as described previously . Briefly, the heparinized bone marrow sample was diluted 1:5 with complete culture medium (Opti-MEM containing 10% fetal calf serum, 100 units/ml penicillin, 100 μg/ml streptomycin and 50 μg/ml sodium ascorbate), and incubated at 37°C in a 5% CO2 humidified atmosphere. Although present in bone marrow in an extremely low percentage in proportion to total mononuclear cells, hMSCs can be easily isolated on the basis of their ability to adhere to polystyrene plates, while cells of the hematopoietic lineage remain in suspension and can be removed.
After 48h, medium containing all non-adherent cells was centrifuged for 10min at 1000 rpm; the pellet was discarded, while the supernatant was added to the culture dish, in a 1:1 ratio with the fresh medium. In 3–4 days, several foci of adherent spindle-like cells appeared and reached sub-confluence in 1–2 weeks. The medium was refreshed every 3 days, each time leaving one half of conditioned medium. The cells harvested from each donor were pooled. Afterwards, cells were trypsinized, and their phenotype confirmed by flow cytometry using a wide panel of labeled monoclonal antibodies, as previously reported . Cultures between the second and fourth passage were used in the experiments.
3 Serum samples
A control group of healthy donors and a group of uremic patients on hemodialysis (hemodiafiltration, HDF) were selected. All patients were clinically stable and recruited according to the following inclusion criteria:
- - no clinical evidence of diabetes, lupus erythematosus, viral hepatitis, cancer;
- - maintenance HDF thrice weekly for at least 3 months; polysulfone filters; Kt/V > 1.4.
Some of the HDF patients were treated with cinacalcet (Mimpara, Amgen Dompè, Italy). All previous transplant recipients were excluded.
Blood samples were drawn under fasting conditions and collected in the appropriate Vacutainer tubes for serum. In the patient group, this occurred immediately before the hemodialysis session (mid-week session), and before and after the hemodialysis session (for the experiments assessing the effects of hemodialysis). After blood withdrawal, the sample was centrifuged at 4°C for 15 min at 3000 rpm, serum was collected under sterile laminar flow hood, and stored at-20°C. One aliquot was immediately processed for routine biochemical analysis.
In Table 1, concentrations in control and patient serum of creatinine, urea, calcium, phosphorus, intact PTH (iPTH), and 25-OH vitamin D are depicted.
Prior to each experiment, the required amount of serum from healthy controls (control serum) or serum from uremic patients on hemodialysis (uremic serum) was obtained by respectively pooling together all the various aliquots stored as above, using always the same pool.
4 Cell treatments
The ability of isolated hMSCs to grow in the presence of human serum was tested in preliminary experiments, in order to optimize growth and viability, and these conditions were used in experimental set-up. We initially checked the ability of hMSC to grow in medium supplemented with human serum at different percentage, compared to 10% fetal calf serum as a standard cell culture condition . We detected optimal cell growth using 10% human serum for 3 days, which insured excellent cell viability. Therefore in all our experiments we utilized 10% human serum.
Isolated hMSCs were induced to differentiate into osteoblasts following a standard protocol [38–41]. Cells were seeded at a density of 104 cells/cm2 in 6-well plates and the following day stimulated with either standard growth (non-osteogenic) medium (Opti-MEM) or osteogenic medium (Opti-MEM, plus 0.1 μM dexamethasone, 10 mM sodium β-glycerophosphate, 0.05 mM L-Ascorbic acid and 0.5 mM CaCl2) for 15–21 days. Normal or osteogenic medium were supplemented with 10% control serum, or 10% uremic serum pools, or with 10% control serum plus toxins, at final concentrations comparable to those present in uremia [42–44], i.e.: pCS and pCG at 0.48 mM and 80 μM, respectively; PTH 106 pM (1000 pg/ml); indoxyl sulfate (80 μM); ADMA (1 μM); homocysteine (100 μM). Where indicated, uremic serum was added with DADS or DATS (organosulfur H2S donors derived from garlic), at 200 μM final concentration [45–47]. About 30% of uremic toxins are protein-bound in circulation , while albumin binding capacity (its concentration in the serum being 530–770 μmol/L) highly exceeds in vivo, by orders of magnitude, the concentrations of most of these protein-bound retention solutes. Therefore, we set the experimental conditions so that albumin never became limiting for protein-bound retention solutes utilized. Medium was refreshed every 2–3 days and an aliquot of the supernatant was saved for determination of the relevant dosages (Fig. 1).
An adequate number of culture wells was prepared in order to have, for each set of experiments a sufficient number of parallel samples to test all the indicated markers during differentiation, as indicated in Fig. 1.
Hemofiltration is a hemodialysis technique based on uremic toxin removal obtained through convection. Hemofiltrates from 3 patients were pooled and concentrated by reversed-phase chromatography in the presence of a cationic ion-pair reagent (triethylammonium acetate (TEAA) 40 mM) . Retained substances were eluted by using increasing concentrations of ethanol in water: 20%, 40%, 60%, 80% and 100% (20%= 20% ethanol in water, etc.), . Therefore, the first fraction, obtained with 20% ethanol, contains a higher concentration of hydrophilic uremic toxins, and this concentration decreases with the ethanol percentage increase. The hemofiltrates were dried in SpeedVac Concentrator (Thermo Scientific, Thermo Fisher Scientific Inc.—Boston, MA, USA) to remove ethanol, and resuspended in absolute medium (D-MEM). Various dilutions of the hemofiltrate pool fractions were tested on MSC viability and osteogenic differentiation.
5 Cell viability and osteogenic differentiation markers
5.1 Alkaline Phosphatase and Alizarin Red Staining (ARS).
After 10 days, alkaline phosphatase specific activity was evaluated. Once the medium was removed, wells were rinsed with 20 mM Tris/HCl, pH 7.4, 0.5 M NaCl (TBS) and the cells lysed by a specific buffer (20 mM Tris/HCl, pH 7.4, 0.5 mM NaCl, 0.25% Triton X-100, 0.5 mM phenylmethylsulfonyl fluoride, 0.5 mM DTT). After 30 min on ice, cell lysates were centrifuged at 13000xg for 5 min. Alkaline phosphatase activity was determined on the supernatant by measuring the release of p-nitrophenol from disodium p-nitrophenyl phosphate. The reaction mixture contained 10 mM disodium p-nitrophenyl phosphate, 0.5 mM MgCl2, 0.1 M diethanolamine phosphate buffer pH 10.5, and 10–30 mg of cell lysate in a final volume of 500 μl. After 30 min at 37°C, the reaction was stopped by adding 500 μl of 0.5 M NaOH; p-nitrophenol levels were measured with spectrophotometer at 405 nm wavelength. One unit was defined as the amount of enzyme that hydrolyzes 1 nmol of p-nitrophenyl phosphate x min-1. Specific activity is expressed as nmol/mg x min and is corrected for protein concentration of the cell lysates.
ARS, an anthraquinone derivative, may be used to identify calcium deposition in cell cultures. After 15–21 days of differentiation, cells were fixed with 4% paraformaldehyde for 15–30 min, washed with bidistilled water and stained with ARS at 40 mM, pH 4.1. After 20 min incubation at room temperature with gentle shaking, the unincorporated dye was removed and cells washed with bidistilled water. Calcium deposits were visible as a red staining. For quantification, 800 μl 10% acetic acid was added to each well. Cell monolayers were then scraped and transferred to a 1.5 ml microcentrifuge tube. After vortexing for 30 s, the tube was heated at 85°C for 10 min, transferred on ice for 5 min, and centrifuged at 20000xg for 15 min. Then, 500 μl of the supernatant were removed and transferred to a new tube, and 200 μl of 10% ammonium hydroxide were added to neutralize the acid. Aliquots (150 μl) of the supernatant were read in triplicate with a spectrophotometer set at 405 nm wavelength.
5.2 PrestoBlue viability assay.
PrestoBlue is a resazurin-based solution that functions as a cell viability indicator by using the reducing power of living cells, to quantitatively measure cell proliferation. When added to cells, the PrestoBlue reagent is modified by the reducing environment of the viable cell and turns red, becoming highly fluorescent. Viable cells retain the ability to reduce resazurin into resorufin. Nonviable cells rapidly lose metabolic capacity and thus do not generate a fluorescent signal. This colour is detected using fluorescence or absorbance measurements. Cells were washed with PBS, incubated ≥10min at 37ºC with 500 μl PrestoBlue 1X (in D-MEM without phenol red); then, for each sample, the reagent was removed and saved in a tube, cells were washed 2 times with 1 ml PBS and the supernatant saved in the same tube and then fluorescence was read at 560 nm wavelength [50,51]. The advantage of this assay is that it allows continuous monitoring of cultures, as the culturing process continued after the screening test.
Enzyme concentrations in the treated hMSCs culture medium were determined utilizing the relevant ELISA kits according to the supplier’s protocols. ELISA Kits were: Human Soluble Receptor Activator of Nuclear Factor-KB Ligand (sRANKL Total), human osteocalcin and human Osteoprotegerin (OPG), (BioVendor—BioVendor-Laboratorni medicina a.s., Brno, Czech Republic), Human osteopontin and Human Bone morphogenic protein-2 (BMP-2) (Biorbyt Ltd., Cambridge, UK), Human collagen type I (Uscn Life Science, Houston, USA).
As illustrated in Fig. 1, for the OPG and sRANKL assays (early markers), culture media obtained after 3 days of differentiation were used. For the assays of osteopontin, BMP-2, collagen type I and osteocalcin (late markers), media after 10 days were used.
6 Statistical Analysis
Data are presented as mean (SE). All experiments were done at least in triplicate except when otherwise stated. Within the same assay session, each sample of a set relevant to individual treatments was run in triplicate. Results on the levels of each marker were reported in the relevant units (in the experiments concerning evaluation of uremic serum effects vs. control serum). Considering experiments using uremic toxins, the difference between the measured effects of each toxin vs. control serum was reported as percentage. A paired t Student test was performed . All calculations were performed using the software package GraphPad Prism, Version 5.0 for Windows (GraphPad Software, San Diego, CA, USA). Statistical significance is considered at p<0.05. Grubbs’ method was used to assess outliers.
1 Effects of various treatments on cells viability
We preliminarily checked cell viability under the various cell growth conditions and treatment that we employed, using the Presto Blue test assay performed both at 3 and 10 days of culture. Treatments with serum from healthy donors (control serum) and with serum from uremic patients on hemodialysis (uremic serum) and/or osteogenic medium as well as the addition of uremic toxins to medium did not significantly compromised cell viability compared to standard growth conditions (data not shown); see also ref. [37–41]. Therefore, we could rule out the occurrence of any acute or sub-acute toxic effect of added compounds in our experimental conditions.
Also hemofiltrate fractions were tested on MSCs to study the effects on cell viability. Hemofiltrate fractions were diluted in the cell medium to a concentration of indoxyl sulfate (as a typical uremic toxin) in the range of 12–120 ppm, i.e. above the range of indoxyl sulfate concentration, referred to 10% whole uremic serum (about 2 ppm); see also ref. . Fig. 2 showed that both control and uremic serum provide culture conditions insuring satisfactory cell viability, and addition of hemofiltrate fractions to cell cultures at a final toxin concentration well above that of uremic serum, still insured suitable cell viability for experiments on osteogenic differentiation.
PrestoBlue assay on cells treated with the five different hemofiltration fractions (20%, 40%, 60%, 80% and 100%), at a final concentrations of 120 or 12 ppm, still ensured suitable cell viability, compared to cell treated with serum from healthy donors (control serum; CS) and serum from uremic patients (uremic serum; US), in normal medium (bar plot). Photomicroscopy images of hMSCs treated with CS, US or in the presence of the five indicated hemofiltrate fractions are also shown.
2 Cell culture in uremic medium is associated with an alteration of the sRANKL/OPG ratio concentration
A number of key markers of bone metabolism and osteogenic differentiation were monitored according to Fig. 1, after three days from differentiation induction of hMSCs in the presence of uremic serum or selected uremic toxins compared to controls, at concentrations comparable to those detected in uremia. In Fig. 3A, the effects of uremic serum vs. control serum on OPG release in the medium are displayed. Uremic serum did not significantly influence OPG concentration, although there was a trend towards a decrease. Regarding sRANKL (Fig. 3B), uremic serum decreased levels six-fold compared to control serum. The sRANKL/OPG ratio of uremic serum vs. control serum was therefore also markedly decreased (Fig. 3C).
OPG (panel A), sRANKL (panel B) and their ratio (panel C) were measured in hMSC cell medium at 3 days after induction of osteogenic differentiation in the presence of uremic serum (US), compared to control serum (CS). Another set of experiments was performed by inducing osteogenic differentiation in hMSCs cultured in the presence of control serum added with the indicated uremic toxins (pCS, p-cresylsulfate; pCG, p-cresylglucuronide; PTH, parathyroid hormone; IS, indoxyl sulfate; ADMA, asymmetric dimethylarginine; Hcy, homocysteine), compared to control serum alone. Panel D: OPG; panel E: sRANKL; panel F: sRANKL/OPG ratio. *p<0.05, **p≤0.01, ***p<0.001.
When checking the effects of various uremic toxins on the same markers, pCS, PTH and homocysteine decreased OPG (Fig. 3D). PTH and indoxyl sulfate significantly decreased sRANKL concentration (Fig. 3E). Regarding the sRANKL/OPG ratio (Fig. 3F), PTH and indoxyl sulfate determined an increase.
3 The uremic milieu reduces osteogenic differentiation of hMCSs
Markers of bone differentiation, ten days after the start of osteogenic induction (according to Fig. 1), were monitored in the medium of hMSCs cultured in the presence of uremic serum vs. control serum or, in parallel, samples in which selected uremic toxins had been added to control serum vs. no addition samples.
Uremic serum increased osteopontin (Fig. 4A), osteocalcin (Fig. 4B) and collagen type I (Fig. 4C), in comparison to control serum. This effect was mimicked by ADMA and homocysteine for osteopontin (Fig. 4D), by homocysteine and by PTH for soluble osteocalcin (Fig. 4E). Although a tendency of some tested toxins to influence collagen type I could be detected, no significant changes were observed (Fig. 4F). Alkaline phosphatase expression, a typical late marker of osteogenic differentiation, increased upon induction of osteogenic differentiation both in uremic serum vs. control serum treated hMSC, with no significant difference between the two, nor upon treatment with selected toxins (data not shown).
Osteopontin (OPN, panel A), osteocalcin (OST, panel B) and collagen type I (COL1, panel C) were measured in the medium of hMSCs cultured in the presence of uremic serum (US) vs. control serum (CS). OPN (panel D), OST (panel E) and COL1 (panel F) were also monitored in parallel experiments in which hMSCs were cultured in the presence of control serum added with the indicated uremic toxins (pCS, p-cresylsulfate; pCG, p-cresylglucuronide; PTH, parathyroid hormone; IS, indoxyl sulfate; ADMA, asymmetric dimethylarginine; Hcy, homocysteine) compared to NS as such. *p<0.05.
BMP-2 levels, induced upon osteogenic differentiation, were decreased when hMSCs were treated with uremic serum compared to control serum (Fig. 5A). None of the tested uremic toxins modified BMP-2 (Fig. 5B).
BMP-2 levels in uremic serum (US) treated hMSCs (panel A) or upon treatment with selected toxins added to control serum (CS) (panel B), and ARS in CS or US treated hMSCs (panel C) or upon treatment with selected toxins (pCS, p-cresylsulfate; pCG, p-cresylglucuronide; PTH, parathyroid hormone; IS, indoxyl sulfate; ADMA, asymmetric dimethylarginine; Hcy, homocysteine) added to NS (panel D) are shown. Effects of H2S donors DADS and DATS on ARS in US treated cells (panel E). Examples of ARS stain in hMSCs treated with CS or US, as culture well picture (panel F) and as microscopy images (panel G). *p<0.05.
Calcium deposition, as evaluated by the ARS assay, was monitored during late osteogenic differentiation. ARS stain was decreased in the presence of uremic serum, as shown in a representative experiment in Fig. 5F and 5G (red staining was faded in the uremic serum treated cells with respect to control serum) and shown as average levels in Fig. 5C, as well as levels in the presence of pCS, PTH, indoxyl sulfate, and ADMA added to control serum, compared to control serum per se (Fig. 5D).
H2S donors, DATS and DADS, added to hMSCs, in the presence of uremic serum, significantly increased calcium deposition, monitored as ARS, compared to the H2S donor-untreated cells (Fig. 5E). All other osteogenic parameters were not affected by H2S donors (data not shown).
4 Serum of cinacalcet-treated patients antagonizes the effect of the uremic milieu on hMSC osteogenic differentiation
We performed a set of experiments in which hMSCs were induced to differentiation by osteogenic medium in the presence of a pool of sera from hemodialysis patients in therapy with cinacalcet. Serum from cinacalcet-treated uremic patients induced a decrease of osteopontin, osteocalcin, and collagen type I compared to serum from untreated patients. (Fig. 6A, 6B, and 6C). The other parameters were not modified (data not shown).
Osteopontin (OPN, panel A), osteocalcin (OST, panel B) and collagen type I (COL1, panel C) during hMSC differentiation in the presence of serum from hemodialysis patients treated with cinacalcet (US cinacalcet), compared to serum from non-treated patients (uremic serum, US). Results from control serum (CS) were added for comparison. *p<0.05, ***p<0.001.
5 Effects of hemodialysis
The effect of a hemodialysis session, studied by comparing hMSCs treated with pre-dialysis or post-dialysis serum, was investigated, and we show that a single session did not induce a change in OPG levels (Fig. 7A), while it causes a decrease of sRANKL (Fig. 7B), sRANKL/OPG ratio (Fig. 7C), and osteocalcin (Fig. 7D). All other parameters showed no differences (data not shown).
OPG levels (panel A), sRANKL (panel B) and their relevant ratio (panel C), osteocalcin (OST, panel D) were measured, in the medium, during osteogenic differentiation of hMSC grown in the presence of uremic serum from the same patients before (PRE) and after hemodialysis (POST). Results from control serum (CS) were added for comparison. *p<0.05.
A uremic environment markedly influences hMSCs, playing a central role in the balance between bone formation and resorption. The reduced bone osteogenesis present in CKD can be explained, at least in part, by the alterations induced by the uremic milieu on hMSCs osteogenic differentiation. Cinacalcet and H2S donors, as well as a hemodialysis session, can improve some of these derangements.
HMSCs can tip the balance between bone formation and resorption, by their known capacity to differentiate into osteoblasts, which synthesize the protein components, that allow extracellular osteoid generation, the organic matrix in which mineralization occurs.
Up till now, hMSCs have essentially been used to study calcification occurring in the vascular system, in that the serum from uremic patients (uremic serum) induces a procalcific phenotype, accompanied by matrix remodeling and actual calcification . However, to the best of our knowledge, the effect of uremic milieu on hMSCs differentiation into osteoblasts has never been studied.
Many uremic toxins may be expected to influence hMSCs function and osteogenic differentiation. Several retention compounds in uremia display noxious effects on bone metabolism and function [4–10]. We investigated the effects of the uremic milieu, as well as that of selected uremic toxins (pCS, pCG, PTH, indoxyl sulfate, ADMA, and homocysteine), chosen in consideration of their effects on bone function [4–10].
Compared to serum from healthy control donors (control serum), uremic serum treatment determines several modifications of key regulators of bone remodeling, first of these being a reduction of the ratio sRANKL/OPG, which conforms to what is found in vivo in hemodialysis patients [15–17]. The lower ratio sRANKL/OPG is compatible with an initial compensatory response to the uremic environment of the system to favor osteogenesis over osteoclastosis, acting as a limiting factor to excessive resorption.
However, later in the osteogenic differentiation (see Fig. 1), the levels of osteopontin, osteocalcin, collagen type I, are significantly increased under these conditions, while BMP-2, and ARS (a marker of calcium deposition) are decreased, indicating, in the end, that bone formation is reduced in favor of bone resorption.
Osteopontin, a positive modulator of osteoclastosis to prevent excessive bone formation, is increased in cell culture medium upon treatment with uremic serum, consistent with in vivo data [18, 21]. The increase of osteopontin in vitro during late differentiation opposes the decrease of the sRANKL/OPG ratio (pro-osteogenesis), which is prominent during the first phase of osteogenesis.
Also osteocalcin and collagen type I levels mimic what found in vivo [22,23]. This could indicate that they are secreted, and persist in the supernatant in their soluble form, instead of being incorporated into the matrix, thus explaining the reason why in vitro calcium deposition is reduced.
On the contrary, BMP-2 levels are lower in our experimental setting, while serum levels are usually higher . BMP-2 is a procalcific protein, and the fact that its levels are lower in vitro is in accordance with the hypothesis that the uremic milieu creates an altered microenvironment where bone calcium deposition is reduced. In fact, we also detected lower ARS levels.
All in all, these effects show that, in the balance between bone formation and bone resorption, hMCSs, under uremic medium conditions, uncouple these two processes and are prone towards bone resorption instead of bone formation.
The effects of the selected uremic toxins mimic some of the effects observed in the presence of whole uremic serum. None of the toxins is able to completely replicate the pattern exerted by uremic serum on hMSCs; however, almost all of them reduce ARS levels, and therefore reduce calcification. Regarding sRANKL/OPG, PTH and indoxyl sulfate act in an opposite manner with respect to uremic serum, in that they both increase the ratio. This result is not unexpected for PTH, since this hormone physiologically stimulates bone resorption over deposition. It must be considered that the hMSCs used in our model are exposed to toxins for a limited period of time with respect to cells exposed in vivo to the same toxins in the chronic condition of uremia. In addition, it should be noted that many other compounds present in the uremic milieu deserve being tested, aside from the ones we analyzed, for example those that may responsible for the decrease in BMP-2.
Calcimimetics are agents that allosterically increase the sensitivity of the calcium-sensing receptor in the parathyroid gland to calcium . Cinacalcet, the only currently available calcimimetic, is an emerging option in the treatment of secondary hyperparathyroidism in CKD patients on hemodialysis [57–59]. We therefore expect that the serum microenvironment of CKD patients treated in vivo with cinacalcet could be more similar to a normal condition, with respect to the alterations of bone metabolism, compared to what was detected in cinacalcet-untreated CKD patients. Results from treatment of hMCSs with cinacalcet-treated patients serum actually showed that the uremic microenvironment was favorably influenced by this in vivo treatment, since osteopontin, osteocalcin, and collagen type I behaved more similarly to what was observed in cell cultures grown with control serum. This indicates that therapy with cinacalcet positively modifies the uremic microenvironment with respect to parameters involved in hMSC osteogenic differentiation.
H2S, an endogenous vasodilator and antioxidant, is also involved in bone formation and osteoblast proliferation through enhancement of the transcriptional levels of alkaline phosphatase, osteocalcin, and collagen , as well as RANKL/OPG . Its levels are low in the plasma of uremic patients on hemodialysis , and we have seen in vitro experiments that H2S treatment decreases the inflammation typical of this condition . It has been already demonstrated that H2S regulates bone marrow MSCs in mice  and it has been shown that in H2S-deficient mice, that display an osteoporotic phenotype, restoring H2S via non-toxic donors may provide treatment for diseases such as osteoporosis .
Slow-releasing H2S donors, such as DADS and DATS, incubated in vitro with hMSCs are able to counteract the decrease in ARS levels induced by uremic serum. This finding indicates that these compounds can be of potential therapeutic use, not only with respect to the inflammatory effects of uremia, but also on hMSCs osteogenic differentiation.
Serum from patients after a single hemodialysis session induces a decrease of sRANKL/OPG, because a tendency to a further reduction of sRANKL. Osteocalcin levels were decreased, which is a finding that goes in the opposite direction of what found in cells treated with uremic serum. Therefore, findings are in accordance with the idea that hemodialysis restores the situation in favor of bone formation.
In conclusion, bone modifications in uremia, mediated by secondary hyperparathyroidism and its determinants, can be also influenced by the capability of uremic milieu to alter hMSCs osteogenic potential and differentiation. New therapeutic strategies, such as cinacalcet and H2S donors, can be utilized to intervene on these alterations, which need to be further explored. In this respect, the in vitro system we set up can be considered as a useful way to study how various conditions and treatments may influence the effects of the uremic microenvironment on hMSC osteogenic differentiation.
The authors AF Perna, R Vanholder, V Jankowski, and J Jankowski, participating in this study, are members of the European Uraemic Toxin working group of the European Society of Artificial Organs, endorsed by the European Renal Association—European Dialysis and Transplantation Association.
Conceived and designed the experiments: DL AFP DI. Performed the experiments: DL AO AP SG ADN CV VJ JJ. Analyzed the data: DL RV AFP GC JJ DI. Contributed reagents/materials/analysis tools: DL AFP AO RV AP SG ADN CV GC VJ JJ DI. Wrote the paper: DL RV AFP DI.
- 1. Kidney Disease: Improving Global Outcomes (KDIGO) CKD-MBD Work Group (2009) KDIGO clinical practice guideline for the diagnosis, evaluation, prevention, and treatment of Chronic Kidney Disease-Mineral and Bone Disorder (CKD-MBD). Kidney Int Suppl 113: S1–130.
- 2. Almeras C, Argilés A (2009) The general picture of uremia. Semin Dial 22:329–333. pmid:19708976
- 3. Cunningham J, Locatelli F, Rodriguez M (2011) Secondary hyperparathyroidism: pathogenesis, disease progression, and therapeutic options. Clin J Am Soc Nephrol 6(4): 913. pmid:21454719
- 4. Iwasaki Y, Yamato H, Nii-Kono T, Fujieda A, Uchida M, et al. (2006) Uremic toxin and bone metabolism. J Bone Miner Metab 24(2): 172–175. pmid:16502128
- 5. Tanaka H, Iwasaki Y, Yamato H, Mori Y, Komaba H, et al. (2013) p-Cresyl sulfate induces osteoblast dysfunction through activating JNK and p38 MAPK pathways. Bone 56(2):347–354. pmid:23851293
- 6. Liabeuf S, Glorieux G, Lenglet A, Diouf M, Schepers E, et al. (2013) Does p-cresylglucuronide have the same impact on mortality as other protein-bound uremic toxins? PLoS One 24; 8(6): e67168.
- 7. Xiao ZS, Quarles LD, Chen QQ, Yu YH, Qu HP, et al. (2001) Effect of asymmetric dimethylarginine on osteoblastic differentiation. Kidney Int 60(5): 1699–1704. pmid:11703587
- 8. McLean RR, Jacques PF, Selhub J, Tucker KL, Samelson EJ, et al. (2004) Homocysteine as a predictive factor for hip fracture in older persons. N Engl J Med 350(20): 2042. pmid:15141042
- 9. Yang J, Hu X, Zhang Q, Cao H, Wang J, et al. (2012) Homocysteine level and risk of fracture: A meta-analysis and systematic review. Bone 51(3): 376–382. pmid:22749888
- 10. Nii-Kono T, Iwasaki Y, Uchida M, Fujieda A, Hosokawa A, et al. (2007) Indoxyl sulfate induces skeletal resistance to parathyroid hormone in cultured osteoblastic cells. Kidney Int 71(8): 738. pmid:17264878
- 11. Horwitz EM, Le Blanc K, Dominici M, Mueller I, Slaper-Cortenbach I, et al. (2005) Clarification of the nomenclature for MSC: The International Society for Cellular Therapy position statement. Cytotherapy 7(5): 393–395. pmid:16236628
- 12. Pittenger MF, Mackay AM, Beck SC, Jaiswal RK, Douglas R, et al. (1999) Multilineage potential of adult human mesenchymal stem cells. Science 284(5411): 143–147. pmid:10102814
- 13. Schneider RK, Puellen A, Kramann R, Raupach K, Bornemann J, et al. (2010) The osteogenic differentiation of adult bone marrow and perinatal umbilical mesenchymal stem cells and matrix remodelling in three-dimensional collagen scaffolds. Biomaterials 31:467–480. pmid:19815272
- 14. Rodriguez JP, Gonzalez M, Rios S (2004) Cytoskeletal Organization of Human Mesenchymal Stem Cells (MSC) Changes During Their Osteogenic Differentiation. J Cell Biochem; 93: 721–731.
- 15. Tanaka H, Mine T, Ogasa H, Taguchi T, Liang CT (2011) Expression of RANKL/OPG during bone remodeling in vivo. Biochem Biophys Res Commun 411(4):690–694. pmid:21771583
- 16. Kazama J (2004) Osteoprotegerin and bone mineral metabolism in renal failure. Curr Opin Nephrol Hypertens 13:411–415. pmid:15199291
- 17. Doumouchtsis KK, Kostakis AI, Doumouchtsis SK, Tziamalis MP, Tsigris C, et al. (2007) sRANKL/osteoprotegerin complex and biochemical markers in a cohort of male and female hemodialysis patients. J Endocrinol Invest 30:762–766. pmid:17993768
- 18. Giachelli CM, Steitz S (2000) Osteopontin: a versatile regulator of inflammation and biomineralization. Matrix Biol 19: 615–622. pmid:11102750
- 19. Nemcsik J, Kiss I, Tislér A (2012) Arterial stiffness, vascular calcification and bone metabolism in chronic kidney disease. World J Nephrol 1(1): 25–34. pmid:24175239
- 20. El-Abbadi MM, Pai AS, Leaf EM, Yang HY, Bartley BA, et al. (2009) Phosphate feeding induces arterial medial calcification in uremic mice: role of serum phosphorus, fibroblast growth factor-23, and osteopontin. Kidney Int 75(12): 1297–1307. pmid:19322138
- 21. Nitta K, Ishizuka T, Horita S, Hayashi T, Ajiro A, et al. (2001) Soluble osteopontin and vascular calcification in hemodialysis patients. Nephron 89(4): 455–458. pmid:11721165
- 22. Calvo MS, Eyre DR, Gundberg CM (1996) Molecular basis and clinical application of biological markers of bone turnover. Endocr Rev 17(4): 333–368. pmid:8854049
- 23. Takano M, Okano K, Tsuruta Y, Yamashita T, Echida Y, et al. (2011) Correlation of new bone metabolic markers with conventional biomarkers in hemodialysis patients. Clin Invest Med 34(5): E267. pmid:21968268
- 24. Garrett IR, Chen D, Gutierrez G, Zhao M, Escobedo A, et al. (2003) Selective inhibitors of the osteoblast proteasome stimulate bone formation in vivo and in vitro. J Clin Invest 111(11): 1771–1782. pmid:12782679
- 25. Costa E, Coimbra J, Catarino C, Ribeiro S, Reis F, et al. (2012) Major determinants of BMP-2 serum levels in hemodialysis patients. Ren Fail 34(10): 1355–1358. pmid:23016911
- 26. Dalfino G, Simone S, Porreca S, Cosola C, Balestra C, et al. (2010) Bone morphogenetic protein-2 may represent the molecular link between oxidative stress and vascular stiffness in chronic kidney disease. Atherosclerosis 211(2): 418–423. pmid:20537331
- 27. Ureña P, Hruby M, Ferreira A, Ang KS, de Vernejoul MC (1996) Plasma total versus bone alkaline phosphatase as markers of bone turnover in hemodialysis patients. J Am Soc Nephrol 7(3): 506–512. pmid:8704118
- 28. Puchtler H, Meloan SN, Terry MS (1969) On the history and mechanism of alizarin and alizarin red S stains for calcium. J Histochem Cytochem 17(2): 110–124. pmid:4179464
- 29. Lievremont M, Potus J, Guillou B (1982) Use of alizarin red S for histochemical staining of Ca2+ in the mouse; some parameters of the chemical reaction in vitro. Acta Anat (Basel) 114(3): 268–280.
- 30. Stanford CM, Jacobson PA, Eanes ED, Lembke LA, Midura RJ (1995) Rapidly forming apatitic mineral in an osteoblastic cell line (UMR 106–01 BSP). J Biol Chem 270(16): 9420–9428. pmid:7721867
- 31. Gregory CA, Gunn WG, Peister A, Prockop DJ (2004) An Alizarin red-based assay of mineralization by adherent cells in culture: comparison with cetylpyridinium chloride extraction. Anal Biochem 329(1): 77–84. pmid:15136169
- 32. Perna AF, Luciano MG, Ingrosso D, Raiola I, Pulzella P, et al. (2009) Hydrogen sulphide-generating pathways in haemodialysis patients: a study on relevant metabolites and transcriptional regulation of genes encoding for key enzymes. Nephrol Dial Transplant 24(12): 3756–3763. pmid:19654230
- 33. Xu ZS, Wang XY, Xiao DM, Hu LF, Lu M, et al. (2011) Hydrogen sulfide protects MC3T3-E1 osteoblastic cells against H2O2-induced oxidative damage—implications for the treatment of osteoporosis. Free Radic Biol Med 50(10): 1314–1323. pmid:21354302
- 34. Liao C, Hua Y (2013) Effect of hydrogen sulfide on the expression of osteoprotegerin and receptor activator of NF-kB ligand in human periodontal ligament cells induced by tension-force stimulation. Arch Oral Biol 58:1784–1790. pmid:24200305
- 35. Feigenbaum J, Neuberg C (1941) Simplified method for the preparation of aromatic sulfuric acid esters. J Am Chem Soc 63:3529–3530.
- 36. Van der Eycken E, Terryn N, Goeman JL (2000) Sudan-ß-d-glucuronides and their use for the histochemical localization of ß-glucuronidase activity in transgenic plants. Plant Cell Rep 19:966–970.
- 37. Oliva A, Passaro I, Di Pasquale R, Di Feo A, Criscuolo M, et al. (2005) Ex vivo expansion of bone marrow stromal cells by platelet-rich plasma: a promising strategy in maxillo-facial surgery. Int J Immunopathol Pharmacol 18(3 Suppl):47–53. pmid:16848987
- 38. Borriello A, Caldarelli I, Basile MA, Bencivenga D, Tramontano A, et al. (2011) The tyrosine kinase inhibitor dasatinib induces a marked adipogenic differentiation of human multipotent mesenchymal stromal cells. PLoS One 6(12): e28555. pmid:22164306
- 39. Guida L, Annunziata M, Rocci A, Contaldo M, Rullo R, et al. (2010) Biological response of human bone marrow mesenchymal stem cells to fluoride-modified titanium surfaces. Clin Oral Implants Res 21(11): 1234–1241. pmid:20497444
- 40. Granéli C, Thorfve A, Ruetschi U, Brisby H, Thomsen P, et al. (2014) Novel markers of osteogenic and adipogenic differentiation of human bone marrow stromal cells identified using a quantitative proteomics approach. Stem Cell Res 12(1): 153–165. pmid:24239963
- 41. Cook DA, Fellgett SW, Pownall ME, O’Shea PJ, Genever PG, et al. (2013) Wnt-dependent osteogenic commitment of bone marrow stromal cells using a novel GSK3β inhibitor. Stem Cell Res 12(2): 415–427. pmid:24382458
- 42. Perna AF, Ingrosso D, Satta E, Lombardi C, Galletti P, et al. (2004) Plasma Protein Aspartyl Damage Is Increased in hemodialysis Patients: Studies on Causes and Consequences. J Am Soc Nephrol 15: 2747–2754. pmid:15466280
- 43. Duranton F, Cohen G, De Smet R, Rodriguez M, Jankowski J, et al. (2012) Normal and pathologic concentrations of uremic toxins. J Am Soc Nephrol 23(7): 1258–1270. pmid:22626821
- 44. Meert N, Schepers E, Glorieux G, Van Landschoot M, Goeman JL, et al. (2012) Novel method for simultaneous determination of p-cresylsulphate and p-cresylglucuronide: clinical data and pathophysiological implications. Nephrol Dial Transplant 27(6):2388–2396. pmid:22167586
- 45. Huang YT, Yao CH, Way CL, Lee KV, Tsai CY, et al. (2013) Diallyl trisulfide and diallyl disulfide ameliorate cardiac dysfunction by suppressing apoptotic and enhancing survival pathways in experimental diabetic rats. J Appl Physiol 114(3): 402–410. pmid:23139364
- 46. Kolluru GK, Shen X, Bir SC, Kevil CG (2013) Hydrogen sulfide chemical biology: pathophysiological roles and detection. Nitric Oxide 35: 5–20. pmid:23850632
- 47. Bordia A, Verma SK, Srivastava KC (1998) Effect of garlic (Allium sativum) on blood lipids, blood sugar, fibrinogen and fibrinolytic activity in patients with coronary artery disease. Prostaglandins Leukot Essent Fatty Acids 58(4): 257–263. pmid:9654398
- 48. Vanholder R, De Smet R, Glorieux G, Argilés A, Baurmeister U, et al. (2003) Review on uremic toxins: classification, concentration, and inter- individual variability. Kidney Int 63:1934–1943. pmid:12675874
- 49. Jankowski J, van der Giet M, Jankowski V, Schmidt S, Hemeier M, et al. (2003) Increased plasma phenylacetic acid in patients with end-stage renal failure inhibits iNOS expression. J Clin Invest 112(2):256–64. pmid:12865413
- 50. Lall N, Henley-Smith CJ, De Canha MN, Oosthuizen CB, Berrington D (2013) Viability Reagent, PrestoBlue, in Comparison with Other Available Reagents, Utilized in Cytotoxicity and Antimicrobial Assays. Int J Microbiol 2013:420601. pmid:23653650
- 51. Boncler M, Różalski M, Krajewska U, Podsędek A, Watala C (2014) Comparison of PrestoBlue and MTT assays of cellular viability in the assessment of anti-proliferative effects of plant extracts on human endothelial cells. J Pharmacol Toxicol Methods 69(1): 9–16. pmid:24103906
- 52. Dawson-Saunders B, Trapp RG (1990) Basic and Clinical Biostatistics. Norwalk: Appleton & Lange. 199 p.
- 53. Lesaffer G, De Smet R, Lameire N, Dhondt A, Duym P, et al. (2000) Intradialytic removal of protein-bound uraemic toxins: role of solute characteristics and of dialyser membrane. Nephrol Dial Transplant 15(1):50–57. pmid:10607767
- 54. Kramann R, Counson SK, Neuss S, Kunter U, Bovi M, et al. (2011) Exposure to uremic serum induces a procalcific phenotype in human mesenchymal stem cells. Arterioscler Thromb Vasc Biol 31(9):e45–54. pmid:21680902
- 55. Chen NX, Duan D, O’Neill KD, Wolisi GO, Koczman JJ, et al. (2006) The mechanisms of uremic serum-induced expression of bone matrix proteins in bovine vascular smooth muscle cells. Kidney Int 70(6): 1046–53. pmid:16837922
- 56. Kawata T, Nagano N, Obi M, Miyata S, Koyama C, et al. (2008) Cinacalcet suppresses calcification of the aorta and heart in uremic rats. Kidney Int 74(10): 1270–1277. pmid:18813289
- 57. Block A, Martin KJ, de Francisco AL, Turner SA, Avram MM, et al. (2004) Cinacalcet for secondary hyperparathyroidism in patients receiving hemodialysis. N Engl J Med 350(15):1516–1525. pmid:15071126
- 58. Urena P, Jacobson SH, Zitt E, Vervloet M, Malberti F, et al. (2009) Cinacalcet and achievement of the NKF/K-DOQI recommended target values for bone and mineral metabolism in renal-world clinical practice—the ECHO observational study. Nephrol Dial Transplant 24(9):2852–2859. pmid:19369690
- 59. Li D, Shao L, Zhou H, Jiang W, Zhang W, et al. (2013) The efficacy of cinacalcet combined with conventional therapy on bone and mineral metabolism in hemodialysis patients with secondary hyperparathyroidism: a meta-analysis. Endocrine 43(1):68–77. pmid:22669774
- 60. Perna AF, Sepe I, Lanza D, Capasso R, Zappavigna S, et al. (2013) Hydrogen sulfide reduces cell adhesion and relevant inflammatory triggering by preventing ADAM17-dependent TNF-α activation. J Cell Biochem 114(7): 1536–1548. pmid:23297114
- 61. Liu Y, Yang R, Liu X, Zhou Y, Qu C, et al. (2014) Hydrogen sulfide maintains mesenchymal stem cell function and bone homeostasis via regulation of Ca(2+) channel sulfhydration. Cell Stem Cell 15(1):66–78. pmid:24726192