Homocysteine enhances apoptosis in human bone marrow stromal cells

Bone 39 (2006) 582 – 590 www.elsevier.com/locate/bone

Homocysteine enhances apoptosis in human bone marrow stromal cells Duk Jae Kim a,1 , Jung-Min Koh a,1 , Oksun Lee b , Na Jung Kim b , Young-Sun Lee b , Yang Soon Kim b , Joong-Yeol Park a , Ki-Up Lee a , Ghi Su Kim a,⁎ a

Division of Endocrinology and Metabolism, Asan Medical Center, University of Ulsan College of Medicine, 388-1 Poongnap-Dong, Songpa-Gu, Seoul 138-736, Republic of Korea b Asan Institute for Life Sciences, 138-736 Seoul, Korea Received 14 December 2005; revised 13 March 2006; accepted 15 March 2006 Available online 27 April 2006

Abstract Introduction: High plasma homocysteine (Hcy) levels have been associated with increased risk of fracture. Since Hcy has been shown to induce apoptosis in many cell types, including vascular endothelial cells, we hypothesized that Hcy would have a similar apoptotic effect on osteoblasts, leading to osteoporosis by reducing bone formation. Materials and methods: Using primary human bone marrow stromal cells (hBMSC) and HS-5 cell line (human bone marrow stromal cell line), we investigated the effects of Hcy on these cells by cell viability assay and analysis of cytoplasmic histone-associated DNA fragments. Caspase activity assay, Western blots, and electrophoresis mobility shift assay (EMSA) were performed to find the mechanism of apoptosis. Intracellular reactive oxygen species (ROS) were measured by spectrometry using dichlorofluorescein diacetate, and cellular total glutathione level was determined by a commercially available kit. N-acetylcysteine (NAC) and pyrrolidine dithiocarbamate (PDTC) were used as tools for investigating the role of ROS and nuclear factor-κB (NF-κB), respectively. Results: Hcy induced apoptosis in primary human bone marrow stromal cells and the HS-5 cell line, and this apoptotic effect was caspasedependent. In addition, Hcy increased cytochrome c release into the cytosol, and activated caspase-9 and caspase-3, but not caspase-8, indicating that Hcy induces apoptosis via the mitochondria pathway. Hcy increased ROS, and NAC inhibited the apoptotic effect of Hcy. Western blot and EMSA showed that Hcy activated the NF-κB pathway. PDTC blocked Hcy-induced caspase-3 activation and apoptosis. Conclusion: These results suggest that Hcy induces apoptosis via the ROS-mediated mitochondrial pathway and NF-κB activation in hBMSCs, and that Hcy may contribute to the development of osteoporosis by reducing bone formation. Antioxidants may have a role in preventing bone loss in individuals with hyperhomocysteinemia. © 2006 Elsevier Inc. All rights reserved. Keywords: Homocysteine; Apoptosis; Human bone marrow stromal cells; Reactive oxygen species; NF-κB

Introduction Both cardiovascular disease and osteoporosis are major public health problems leading to significant morbidity and mortality [1,2]. Although these common diseases have been considered unrelated, recent evidence suggests a strong connection between them [3–5]. A high prevalence of atherosclerosis is found in individuals with low bone mineral density (BMD) [6–8], and the ⁎ Corresponding author. Fax: +82 2 3010 6962. E-mail address: [email protected] (G.S. Kim). 1 Contributed equally to this work. 8756-3282/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.bone.2006.03.004

rate of bone loss in elderly women is associated with cardiovascular mortality [9]. Severe hyperhomocysteinemia is found in patients with homocysteinuria, an inherited disorder of amino acid metabolism, and is associated with osteoporosis and cardiovascular disease at a very young age [10]. In individuals without this genetic disorder, modestly elevated plasma homocysteine (Hcy) levels are commonly encountered, particularly in postmenopausal women and the elderly [11–13], and are associated with increased risk of cardiovascular disease and mortality [12–14]. Elevated Hcy levels have also been associated with low BMD and increased risk of fracture in observational and longitudinal studies [15–19], and Hcy-

D.J. Kim et al. / Bone 39 (2006) 582–590

lowering treatments, such as with folate and vitamin B12, reduce the risk of hip fracture [20]. Genetic studies have shown that Hcy is involved in the development of osteoporosis [21,22]. Although it has been suggested that Hcy induces bone fragility by interfering with collagen cross-linking [23,24], the underlying mechanism by which Hcy affects bone metabolism has not been determined. Osteoporosis is characterized by low bone mass resulting from an imbalance between bone resorption by osteoclasts and bone formation by osteoblasts [25]. Therefore, decreased bone formation by osteoblasts may lead to the development of osteoporosis, and the rate of apoptosis is responsible for the regulation of bone formation [26]. Recently, Hcy was shown to attenuate osteocalcin expression in osteoblast-lineage cells, suggesting that Hcy may disturb osteoblastic function [27]. Meanwhile, Hcy has been reported to induce apoptosis in various cell types including vascular endothelial cells [28]. If Hcy has a sim ilar apoptotic effect in osteoblasts and/or their precursor cells, it may increase the risk of fracture and lead to low BMD by reducing bone formation. We therefore assayed the ability of Hcy to induce apoptosis in osteoblast-lineage cells using primary human bone marrow stromal cells (hBMSCs) and the HS-5 cell line. Materials and methods Chemicals and reagents The pan-caspase inhibitor, Z-VAD-fmk, was purchased from Calbiochem (La Jolla, CA, USA); Hcy and N-acetylcysteine (NAC), extracellular signal regulated kinase (ERK) inhibitor (PD98059), c-jun N-terminal kinase (JNK) inhibitor (SP600125), p38 mitogen-activated protein kinase (MAPK) inhibitor (SB203580), and NF-κB inhibitor (pyrrolidine dithiocarbamate, PDTC) were purchased from Sigma-Aldrich (St. Louis, MO, USA); and the fluorescent probe dichlorofluorescein-diacetate (DCF-DA) was purchased from Molecular Probes (Eugene, OR, USA). Antibodies against JNK, p38 MAPK, IκBα, and their phosphorylated forms were from Cell Signaling Technology (Beverly, MA, USA). Antibodies against ERK and cytochrome c were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA).

Cell culture The hBMSC line, HS-5, was purchased from the American Type Culture Collection (ATCC; Manassas, VA, USA) and cultured in Dulbecco's modified Eagle's medium (DMEM; Gibco, Grand Island, NY, USA) at 37°C in a humidified chamber of 5% CO2. Our previous study [29] showed that the HS-5 cell has an osteoblastic phenotype, such as positive alkaline phosphatase (ALP) staining, osteocalcin secretion, and production and secretion of receptor activator of NF-κB ligand. In the present study, we further observed that the cells express the mRNAs of ALP, osteocalcin and osteopontin, using reverse transcription polymerase chain reaction (RT-PCR) (Fig. 1B). Primary hBMSCs were isolated from human ribs as previously described [30] and maintained in αmodified Eagle's medium (α-MEM, Sigma-Aldrich) containing 10% fetal bovine serum (FBS, Gibco) and 1% penicillin/streptomycin (Gibco) at 37°C in a humidified chamber of 5% CO2. All media were changed twice weekly. Upon growth to 80% confluence, the cells were subcultured using 0.01% trypsin/ 0.05% EDTA (Gibco). Human ribs were obtained with the approval of the Institutional Review of Board of Asan Medical Center (Seoul, Korea).

Mineralized nodule formation Primary hBMSCs were cultured in a 6-well plate at a concentration of 1.5×105 cells/well in 2 ml α-MEM supplemented with 10% FBS and 1% penicillin/streptomycin. On day 3 after the start of culture, we supplemented the medium with 50 μg/ ml ascorbic acid (Sigma-Aldrich) and 10 mM β-glycerophosphate (Sigma-Aldrich).

583

The medium was changed every second day. On day 14, the cultures were fixed for 1 h with 70% ethanol and then stained with 40 mM Alizarin red S (Sigma-Aldrich).

RT-PCR for the mRNAs of ALP, osteocalcin and osteopontin Total RNA was purified from HS-5 cells using TRIzol reagent (Invitrogen, Rockville, MD, USA) according to the manufacturer's instruction, and cDNAs were synthesized from 5 μg aliquots of total RNA using the Superscript™ III First-Strand Synthesis System (Invitrogen). All PCR amplifications were performed using 5 units Taq polymerase (Bioneer, Daejeon, Korea), 15 mM MgCl2 and 2.5 mM dNTP mixture (Bioneer) in a final volume of 30 μl. The specific primer pairs for ALP, osteocalcin, and osteopontin were used (Table 1). The amplification protocol consisted of 25–40 cycles of denaturation at 94°C for 1 min, annealing for 1 min at the indicated temperature, and extension at 72°C for 1 min. The PCR products (10 μl) were separated on 1% agarose gels, which were stained with ethidium bromide and visualized under UV light.

Assessment of individual fibroblast colony forming unit (CFU-F) colonies Freshly isolated whole bone marrow cells were obtained from the human rib bones. The cells seeded in a 6-well plate at 2 × 106 cells/well in 2 ml α-MEM containing 10% FBS, 50 μg/ml ascorbic acid, and 10 mM β-glycerophosphate. One-half of the medium was replaced every 5 days. On day 10 of culture, the cells were fixed, stained for ALP, and counterstained with hematoxylin. ALP+ colonies containing ≥20 cells were enumerated.

Cell viability assay Viability of both cell types was determined using a commercially available cytotoxicity assay kit (Cell Counting Kit-8, Dojindo, Kyushu, Japan), according to the manufacturer's instructions. This colorimetric assay is based on the reduction of water soluble tetrazolium salt (WST-8{(2-(2-methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H- tetrazolium, monosodium salt}) by mitochondrial dehydrogenase of viable cells. Briefly, cells (3000 cells/well) were incubated with Hcy (0 μM, 10 μM, 100 μM, and 500 μM) for 72 h, 10 μl of CCK-8 solution was added, and after 2 h incubation, the absorbance at 450 nm was measured using a SpectraMax 340PC384 microplate reader (Molecular Devices, Palo Alto, CA, USA) with a reference wavelength at 650 nm.

Apoptosis assessment Apoptosis of both cell types was determined using a Cell Death Detection ELISAplus kit (Roche Applied Science, Indianapolis, IN, USA), according to the manufacturer's instructions. After incubation with Hcy (0 μM, 10 μM, 100 μM, and 500 μM), the cells were pelleted by centrifugation, and the supernatant was discarded. The cells were treated with 200 μl lysis buffer and centrifuged, and an aliquot of the supernatant was transferred to a streptavidin-coated microplate and exposed to anti-histone antibody (biotin-labeled) and anti-DNA antibody (peroxidase-conjugated) for 2 h at room temperature. The cells were washed with incubation buffer to remove unbound antibodies and cellular components that were not immunoreactive, and antibody-nucleosome complexes bound to the microplate were determined spectrophotometrically using ABTS {2,2′azino-di(3-ethylbenzthiazolin-sulphonate)} substrate at 405 nm.

Caspase activity assay Caspase-9 activity in HS-5 cells was measured using an ApoAlert Caspase Fluorescence Assay Kit (Clontech, CA, USA), and cellular activity of caspase-3 and -8 were measured using a Caspase activity assay kit (Peptron, Daejun, Korea), each according to the manufacturer's instructions. After treating the cells with 100 μM of Hcy for various times, the cells were lysed with lysis buffer and centrifuged, and the supernatants were incubated with 50 μM DEVD-AMC (caspase-3 substrate), IETD-AMC (caspase-8 substrate), or LEHD-AMC (caspase9 substrate) for 1 h. Ac-DEVD-CHO (Peptron) and LEHD-CHO (Clontech) were used for inhibitors of caspase-8 and -9, respectively. The AMC fluorescence was measured with a Gemini XS (Molecular Devices) fluorescence microplate reader, with excitation at 380 nm and emission filter at 440 nm.

584

D.J. Kim et al. / Bone 39 (2006) 582–590

Analysis of cytochrome c release

Measurement of intracellular ROS production

Release of cytochrome c from the mitochondria of HS-5 cells into the cytosol was measured by immunoblotting as described previously with minor modifications [30]. Briefly, following incubation with 100 μM Hcy for various times, the cells were harvested by centrifugation and lysed, and the soluble cytosolic fractions were obtained by centrifugation. Aliquots of the cytosolic fractions containing equal amounts of proteins were separated by 15% SDS-PAGE and cytochrome c release was measured by immunoblotting with specific antibody.

DCF-DA, which becomes fluorescent upon reaction with hydroxyl radicals and hydrogen peroxide, was used to measure intracellular ROS formation in HS5 cells. After incubation of cells with 100 μM Hcy for 6 h, the culture medium was removed, and the cells were washed with HBSS and incubated with DCFDA (2.5 μM) at 37°C for 1 h. After washing and lysis, DCF fluorescence in the supernatant was measured using a Gemini XS fluorescence microplate reader with excitation at 480 nm and emission filter at 530 nm.

A Mineralized Nodule Formation (%of Dish Area)

70

Hcy 10 µM

Control

Hcy 100 µM

60 50 40

**

30 20

**

10

**

0

Hcy 500 µM

10 100 500 CONT

B

1 days CONT

Hcy (µM)

3 days

Hcy

CONT

Hcy

ALP Osteocalcin Osteopontin

150

Density of Osteopontin (% of 1 Day Control)

Density of ALP (% of 1 Day Control)

GAPDH

120 90

**

60 30 0

1 Day 3 Days

Viability Assay (% of Control)

120 100 80 60 40 20 0

CONT Hcy 100 µM

120 90 60

*

0

1 Day 3 Days

*

*

*

**

HS-5

**

**

30

Treatment Time

Treatment Time

C

150

*

CONT Hcy 10 µM Hcy 100 µM Hcy 500 µM

hBMSC

Fig. 1. Homocysteine (Hcy) suppressed bone nodule formation, some osteoblastic differentiation markers and cell viability in hBMSCs. (A) Primary hBMSCs were incubated with or without Hcy at the indicated concentrations for 14 days and stained with Alzarin red S. Shown are representative figure (left panel), and the mean (+SD) density of three independent experiments (right panel). (B) HS-5 cells were cultured with or without 100 μM Hcy for 1 and 3 days, and mRNAs of alkaline phosphatase (ALP), osteocalcin, and osteopontin were measured by reverse transcription-polymerase chain reaction (RT-PCR), as described under “Materials and methods”. Shown are representative blots (upper panel), and the mean (+SD) densitometry of three independent experiments (lower panel). (C) HS-5 cell and primary hBMSCs were incubated with or without Hcy at the indicated concentrations for 72 h, and their viability was determined by a colorimetric CCK-8 assay. The experiment was repeated three times of triplicate measurements. Shown are the mean (+SD) percentages of the untreated control levels. *P b 0.05, **P b 0.01 vs. control.

D.J. Kim et al. / Bone 39 (2006) 582–590 Table 1 Protocols for reverse transcription polymerase chain reactions mRNA

Primer pairs

Annealing temperature (°C)

Alkaline phosphatase

(Sense) 5′-ACGTGGCTAAGA ATGTCATC-3′ (Antisense) 5′-CTGGTAGGCGATGT CCTTA-3′ (Sense) 5′-GGCAGCGAGGTAG TGAAGAG-3′ (Antisense) 5′-CTGGAGAGGAG CAGAACTGG-3′ (Sense) 5′-TTGCTTTTGCCTCCT AGGCA-3′ (Antisense) 5′-GTGAAAACTTCG GTTGCTGG-3′ (Sense) 5′-ACTTTGTCAAGCTC ATTTCC-3′ (Antisense) 5′-TGCAGCGAACT TTATTGATG-3′

58.2

GAPDH

Statistical analysis 61.8

58.2

60.0

Total glutathione assay A commercially available Total Glutathione Assay Kit (Oxford Biomedical Research, Oxford, MI, USA) was used to measure the total glutathione content in cell lysates. Briefly, 1 × 104 HS-5 cells were seeded into each well of a 96-well plate, incubated with 100 μM Hcy for 48 h and washed with PBS, pH 7.2. To each well was added 100 μl 5% metaphosphoric acid (MPA), and the plate was frozen at −80°C and subsequently thawed at 37°C. After two freeze/thaw cycles, 50 μl of each cell lysate was transferred into new microtiter wells and 50 μl of 0.5 mg/ml DTNB (5,5′-dithiobis-2-nitrobenzoic acid) and glutathione oxidoreductase solution was added to each well. The microtiter plate was incubated for 10 min at room temperature, and 50 μl of the reduced form of β-nicotinamide adenine dinucleotide phosphate (β-NADPH2, 0.6 mg/ml) was added to each well. Following incubation at room temperature for 4 min, the absorbance at 405 nm was measured.

Western blot analysis Cell lysates were prepared in non-denaturing lysis buffer (10 mM Tris–HCl, pH7.4, 150 mM NaCl, 1% Triton X-100, 1 mM EDTA, 1 mM EGTA, 0.3 mM phenyl methyl sulfonlylfluoride, 0.2 mM sodium orthovanadate, 0.5% NP-40, 5 U/ ml aprotinin), and the protein concentrations in the lysates were determined using a Protein assay kit (Bio-Rad laboratories, Hercules, CA, USA). For immunoblot analyses, 50 μg of protein per sample was denatured in 5× SDS-PAGE sample buffer and subjected to SDS-PAGE on Tris–glycine gels. The proteins were transferred onto nitrocellulose membranes (Amersham Biosciences, Piscataway, NJ, USA), which were incubated with 8% non-fat milk powder (w/v) in TBS (500 mM Tris–HCl, pH7.4, 1.5 M NaCl, 0.1% Tween 20) for 2 h at 4°C. After blocking, the membranes were incubated overnight at 4°C with primary antibodies. Specific antibody binding was detected with horseradish peroxidase-conjugated secondary antibodies (Santa Cruz Biotechnology) and visualized using enhanced chemiluminescence detection reagent (Amersham Biosciences). Band density was quantified by densitometric analysis using Quantity One software (VersaDoc Model 3000 Imaging System, Bio-Rad Laboratories).

Electrophoretic mobility shift assay (EMSA) of NF-κB Nuclear proteins were isolated from HS-5 cells, and the nuclear extracts (2 mg) were incubated for 30 min at room temperature with 32P end-labeled NFκB-specific oligonucleotide 5′-AGTTGAGGGGACTTTCCCAGGC-3′ (Promega, Madison, WI, USA) in binding buffer. The NF-κB-DNA complex was separated from the free oligonucleotide on 4% native polyacrylamide gel. For competition experiments, a 25-fold excess of unlabeled probe was added prior to

Statistical analysis was performed by nonparametric Kruskal-Wallis test with SPSS 11.5 (SPSS, Chicago, IL, USA). A P value b 0.05 was considered statistically significant.

Results Effects of Hcy on bone nodule formation, cell differentiation and viability Treatment of primary hBMSCs with 10–500 μM Hcy, dose dependently suppressed bone nodule formation (Fig. 1A). At the concentration of 10 and 100 μM, Hcy suppressed the bone nodule formation by 44.5 ± 11.6% and 10.7 ± 19.3%, compared with controls, respectively (both, P b 0.01). We also noted that the treatment of Hcy (100 μM) significantly attenuated the mRNA expressions for ALP and osteopontin in HS-5 cells (Fig. 1B), but not that for osteocalcin, suggesting that Hcy may affect differentiation of hBMSC. In addition, we found that Hcy also affected the cell viability of HS-5 cells and primary hBMSC (Fig. 1C). At a concentration of 10 μM, Hcy

A

250

* Apoptosis Assay (% of Control)

Osteopontin

the addition of the labeled probe. Supershift experiments were performed by adding 2 mg of an antibody raised against either the p50 (Calbiochem) or p65 subunit (Santa Cruz Biotechnology) of NF-κB before the addition of the labeled probe. The gels were then dried, and the radioactive bands were visualized by a PhosphoImager (Bio-Rad Laboratories).

B

*

200 150

*

*

*

*

Control Hcy 10 µM Hcy 100 µM Hcy 500 µM

100 50 0

Apoptosis Assay (% of Control)

Osteocalcin

585

HS-5

hBMSC

250

*

200 150 100 50 0 Control

Hcy

Z-VAD-fmk

Z-VAD-fmk + Hcy

Fig. 2. Homocysteine (Hcy) induces cell death in HS-5 cells and primary hBMSCs. (A) Cells were cultured in the presence of Hcy (0 μM, 10 μM, 100 μM, or 500 μM) for 48 h, and apoptosis was assayed by measurement of histone-associated DNA fragments. (B) After preincubation with 50 μM ZVAD-fmk (a pan-caspase inhibitor) for 2 h, HS-5 cells were treated with 100 μM Hcy for 48 h, and apoptosis was assayed. The experiments were repeated three times of triplicate measurements. Shown are the mean (+SD) percentages of the untreated control levels. *P b 0.01 vs. control.

586

D.J. Kim et al. / Bone 39 (2006) 582–590 Control 200

*

150

*

*

*

100 50 0 0

6

12

24

48

Time (hours) 200 150 100

Apoptosis via mitochondrial pathway

50 0 0

6

12

24

48

Time (hours)

C Caspase-3 Activity (% of Control at 0 hr)

We found that Hcy enhanced apoptosis of both types of cells in a dose-dependent manner (Spearman's correlation coefficient = 0.842, P b 0.01 for HS-5 cell, and 0.767, P = 0.08 for primary hBMSCs) (Fig. 2A). At the concentration of 10 μM, Hcy increased the apoptosis of HS-5 cells and primary hBMSCs by 47% and 41% compared with controls, respectively (both, P b 0.01). Pretreatment with 50 μM of Z-VAD-fmk, a pancaspase inhibitor, almost completely attenuated the Hcyinduced apoptosis of HS-5 cells (P = 0.667, Z-VAD-fmk + Hcy vs. control) (Fig. 2B), indicating that the apoptotic effect of Hcy was caspase-dependent.

200 150

*

*

12

24

*

*

100 50 0 0

6

48

Time (hours)

D 200

*†

*

To determine whether Hcy-induced apoptosis occurred via the mitochondrial pathway or death receptor pathway, we determined the activities of caspases-3, -9, and -8 in HS-5 cells. We found that Hcy significantly increased the activities of caspases-3 and -9 as early as 6 h after incubation (Figs. 3A and C). Compared with its baseline activity, caspase-9 activity was increased by 69% at 24 h and 89% at 48 h after Hcy treatment (both, P b 0.001), whereas caspase-3 activity was increased by 60% at 24 h and 80% at 48 h (both, P b 0.01). In contrast, caspase-8 activity, a key enzyme in the death receptor pathway, was not altered by treatment with

A

Treatment Time (hrs) Control Hcy 100 μM

160 120

0

4

24

0

4

24

80

Cytochrome c

40 0

Hcy 100 µM

-

+

-

+

Ac-DEVD-CHO

-

-

+

+

-

-

LEHD-CHO

-

-

-

-

+

+

-

+

Fig. 3. Homocysteine (Hcy) enhances the activity of caspase-9 and caspase-3 but not caspase-8. Activities of specific caspases were measured in HS-5 cells using commercially available assay kit (A–C). After treatment with or without 100 μM of Hcy for indicated times, the enzyme activity was measured as the fluorescence produced by cleavage of substrate by the respective caspase. (D) HS-5 cells were treated with 50 μM of caspase-8 inhibitor (Ac-DEVD-CHO) and caspase-9 inhibitor (LEHD-CHO) 2 h prior to Hcy 100 μM treatment. After 24 h of incubation, the cells were prepared to measure caspase-3 activity. All experiments were repeated three times of triplicate measurements. Shown are the mean (+SD) percentages of the untreated control levels at 0 h. *P b 0.01 vs. untreated control, †P b 0.01 vs. Ac-DEVD-CHO-treated control.

Actin

B

15

* Relative Density

Caspase-8 Activity (% of Control at 0 hr)

B

Caspase-3 Activity (% of Control)

Effects of Hcy on caspase-dependent apoptosis of hBMSC

Hcy 100 µM

(Folds of Control)

Caspase-9 Activity (% of Control at 0 hr)

A

12 9

*

6 3 0 0 hr

significantly reduced the viability of both types of cells (P = 0.048 and P = 0.037 vs. controls, respectively) (Fig. 1A). In HS5 cells, reduced cell viability was dependent on Hcy concentrations (16% at 10 μM, 33% at 100 μM, and 69% at 500 μM; Spearman's correlation coefficient = −0.914, P b 0.01). In the culture of whole bone marrow cells, Hcy (100 μ) did not affect CFU-F count (data not shown), suggesting that Hcy has no effects on number of fibroblastic progenitor cells.

4 hrs

24 hrs

Treatment Time Fig. 4. Homocysteine (Hcy) induces cytochrome c release into the cytosol. (A) Cells were incubated in the presence or absence of 100 μM Hcy for the indicated times, their cytosolic fractions were isolated, and the cytochrome c content of each was measured by immunoblotting. Shown is the representative blot. (B) Band intensity determined by densitometry and presented as the ratio after Hcy treatment relative to that at baseline. Shown are the means (+SD) of three independent experiments. *P b 0.01 vs. untreated control.

D.J. Kim et al. / Bone 39 (2006) 582–590

100 μM of Hcy (Fig. 3B). In addition, inhibitor for caspase-9 near completely blocked Hcy-stimulated caspase-3 activity, but inhibitor for caspase-8 did not (Fig. 3D). These data suggest that Hcy may induce apoptosis of hBMSC via the mitochondrial pathway. Furthermore, immunoblotting showed that cytochrome c release into the cytosol, the upstream signal of caspase-9, was increased by incubation with 100 μM Hcy for the indicated times (Fig. 4). Cytochrome c release was observed as early as 4 h after treatment with Hcy and markedly increased by 24 h. ROS-mediated apoptosis Following incubation with 100 μM Hcy for 10, 20, and 30 min, intracellular ROS was increased by 1.2-, 1.6-, and 2.4fold, respectively (P b 0.05 vs. untreated controls), but no further increase was observed after 60 min (data not shown). At 30 min of Hcy treatment, the level of intracellular ROS was increased 2.4-fold, but this was completely blocked by pretreatment with 10 mM NAC (Fig. 5A). At the same time, the level of total glutathione was significantly reduced by Hcy treatment, to 44% of control (P b 0.01), and this reduction was also blocked by NAC (Fig. 5B). Furthermore, NAC completely blocked Hcy-induced apoptosis (Figs. 5C and D). NF-κB-mediated apoptosis Western blot, which was performed to investigate the signaling link between oxidative stress and apoptosis in osteoblasts [31], did not show any altered phosphorylation of JNK, ERK, or p38 MAPK, and addition of the specific inhibitors of these

200 150 100 50

CONT Hcy

NAC

Apoptosis Assay (% of Control)

Viability Assay (% of Control)

80 60

*

20 0 CONT Hcy

NAC

NAC + Hcy

80 60

*

40 20

CONT Hcy

D

100

100

0

NAC + Hcy

120

40

In the present study, Hcy attenuated differentiation and increased apoptosis in hBMSC, thereby suppressing bone nodule formation. We focused here on Hcy-induced apoptosis. We noted that Hcy increased oxidative stress in osteoblast-lineage cells, thereby inducing apoptosis by activating the NF-κB and mitochondrial pathway. We also observed Hcy-induced apoptosis in mature mouse calvarial osteoblasts (data not shown), suggesting that the cytotoxicity of Hcy on hBMSCs is not dependent on the degree of its differentiation. To our knowledge, this is the first study to demonstrate a direct pro-apoptotic effect of

120

*

0

C

Discussion

B

300 250

MAPK (PD98509, SP600125, and SB203580, respectively) had no effect on Hcy-induced apoptosis (data not shown). However, Western blot showed that the phosphorylation of IκBα was remarkably increased by 6- to 7-fold shortly after Hcy treatment (Fig. 6A). Hcy increased the DNA binding activity of NF-κB with the strongest binding at 3 h (Fig. 6B, left panel), as verified by competition assays with unlabeled probe (Fig. 6B. lane 2). Supershift assay with anti-p50 or p65 antibodies was performed on the nuclear extract after Hcy treatment for 3 h. The addition of anti-p50 antibody led to a shift of NF-κB band (lane 3). However, the use of anti-p65 antibody reduced the NF-κB DNA binding activity without a prominent shift of band (lane 4). These results indicate that p50 is the major NF-κB subunit, possibly as a homodimer or heterodimer with the other NF-κB subunit. In addition, the specific inhibitor of NF-κB pathway, PDTC, nearly completely blocked caspase-3 activation and the apoptotic effect by Hcy (Fig. 7).

Glutathione Level (% of Control)

Intracellular ROS (% of Control)

A

587

NAC

NAC + Hcy

NAC

NAC + Hcy

250

* 200 150 100 50 0 CONT Hcy

Fig. 5. Homocysteine (Hcy) induces apoptosis by increasing intracellular ROS. HS-5 cells were preincubated in the presence or absence of 10 mM N-actyl cysteine (NAC) for 2 h and then treated with 100 μM Hcy for 30 min. (A) Intracellular ROS measured by DCF-DA (2.5 μM) and (B) total glutathione were determined after Hcy treatment. (C and D) Cell viability and apoptosis were assayed, as described under “Materials and methods”. The experiments were repeated three times of triplicate measurements. Shown are the mean (+SD) percentages of the untreated control levels. *P b 0.01 vs. control.

588

D.J. Kim et al. / Bone 39 (2006) 582–590

A P-IκB IκB

C

0.5

1

2

4

8

12

24

Treatment Time (hrs)

Density of p-IκB (fold of control)

8

* *

*

6

*

*

* *

4 2 0 C

0.5

1

2

4

8

12

24

Treatment Time (hrs)

B Treatment Time (hrs) C 0.5 1

3

6 24

C

NS

1

2

3

4

NS

cells (data not shown), it did not have any toxic effects on MC3T3-E1 cells (data not shown), suggesting that Hcy has cellspecific effects. In addition, glutathione contained in the culture medium in the previous study [27] have attenuated the effect of Hcy on MC3T3-E1 cells. Hcy is an intermediate metabolite formed from the demethylation of methionine and is a highly reactive substance [13]. Hyperhomocysteinemia, which is defined by plasma Hcy concentrations higher than 15 μM, can be classified as moderate (15 to 30 μM), intermediate (30 to 100 μM), or severe (N100 μM) [13], and individuals with plasma Hcy concentrations higher than 15 μM have a hazard ratio for hip fracture of 3.84. However, individuals with plasma Hcy concentrations as low as 11.0 ± 0.6 μM have a hazard ratio for hip fracture of 1.67 compared with individuals with lower Hcy levels (8.5 ± 0.9 μM) [17]. We observed Hcy-induced apoptosis at concentrations as low as 10 μM, suggesting that Hcy may play a role in bone cells of individuals with even normal plasma Hcy levels. Similarly, osteoclast tartrate-resistant acid phosphatase activity was shown to be increased at 10 μM, compared with controls [32]. The majority of osteoblasts are fated to undergo apoptosis and the rate of apoptosis is responsible for the regulation of bone formation [26]. Apoptosis may be initiated by death receptors such as Fas or the TNF receptor. Binding of these receptors by their

Hcy on osteoblast-lineage cells. In addition, we found that NAC, a potent thiol antioxidant, rescued cells from Hcy-induced apoptosis by reducing oxidative stress. We have shown here that Hcy exerted significant toxic effects on the cell viability of primary cultured hBMSCs and HS-5 cells, a human bone marrow stromal cell line. In contrast, Hcy concentrations as high as 500 μM were found to have no effect on the viability of MC3T3-E1 preosteoblastic murine cells [27]. To determine the cause of this discrepancy, we evaluated the effect of Hcy on MC3T3-E1 cells and primary mouse calvarial bone cells. We found that, while Hcy had similar toxic effects on mouse calvarial bone cells as on primary hBMSCs and HS-5

250

*

200 150 100 50 0 Control

B

Hcy

PDTC

PDTC + Hcy

PDTC

PDTC + Hcy

300

*

250

Apoptosis Assay (% of Control)

Fig. 6. Homocysteine (Hcy) activates NF-κB. (A) HS-5 cells were incubated with 100 μM of Hcy for indicated periods. After incubation, cell lysates were subjected to Western blot with antibodies against IκBα and its phosphorylated form. Shown is the representative blot (upper panel). The band intensity of phosphorylated IκBα determined by densitometry is presented. Shown are the mean (+SD) folds of three independent experiments (lower panel). *P b 0.01 vs. control. (B) After incubation of HS-5 cells with 100 μM of Hcy for indicated times, EMSA was performed to determine the DNA binding activity. The binding activity increases with Hcy treatment with strongest binding occurring at 3 h (left panel). Supershift analysis (right panel) at 3 h performed with antip50 (lane 3). The use of anti-p65 did not lead to significant shift of band (lane 4), and competition assay (lane 2) with P32-unlabeled probe verified that the band (arrows) is NF-κB. Shown are the representative figures. NS, non-significant band. C, untreated control. Lane 1, Hcy 100 μM.

Caspase-3 Activity (% of Control)

A

200 150 100 50 0 Control

Hcy

Fig. 7. Homocysteine (Hcy)-induced apoptosis requires NF-κB pathway activation. HS-5 cells were preincubated with or without PDTC (20 μM) for 2 h. After that, cells were cultured with or without 100 μM of Hcy. (A) After Hcy treatment for 24 h, caspase-3 activity was measured as in Fig. 3. (B) After Hcy treatment for 48 h, apoptosis was determined as in Fig. 2. The experiments were repeated three times of triplicate measurements. Shown are the mean (+ SD) percentages of the untreated control levels. *P b 0.01 vs. control.

D.J. Kim et al. / Bone 39 (2006) 582–590

ligands recruits and activates procaspases-8 and induces apoptosis [33]. Alternatively, cellular stress and cytotoxic stimuli may induce apoptosis by altering the permeability of the mitochondrial membrane, resulting in cytochrome c release. In the cytosol, cytochrome c forms apoptosome complexes with apoptotic protease activating factor 1 (Apaf-1) and procaspase-9, leading to activation of caspase-9, the initiator caspase. We found that Hcy significantly enhanced cytochrome c release and activated caspases-9 and -3, but not caspase-8, indicating that Hcy induces apoptosis of osteoblast-lineage cells through the mitochondrial pathways. These results are in agreement with a recent report showing that Hcy induced mitochondrial cytochrome c release and caspase-dependent apoptosis in endothelial cells [28]. We also found that Hcy increased intracellular ROS and decreased cellular total glutathione, and that NAC prevented these effects as well as cell death, indicating that oxidative stress is essential for Hcy-induced apoptosis in osteoblast-lineage cells. Although it is not known how Hcy increases oxidative stress in these cells, Hcy may activate NADPH oxidase, leading to ROS production and the inhibition of glutathione peroxidase, superoxide dismutase, and catalase, all of which are major antioxidant enzymes [34,35]. We showed that Hcy increased IκBα phosphorylation and conversely decreased innate IκBα. Hcy also increased the DNA binding activity of NF-κB in EMSA. An inhibitor of NF-κB, PDTC, nearly completely attenuated caspase-3 activation and apoptosis, suggesting that the apoptotic effect of Hcy requires NF-κB activation. It was shown that in murine osteoblasts, hydrogen peroxide activated NF-κB and increased cell death, while scavenging free radicals inhibited NF-κB activation by hydrogen peroxide [36]. Therefore, it seems likely that Hcy increased ROS and subsequently activated NF-κB. NF-κB typically comprises the p50/p65 dimer together with inhibitor protein, IκBα, and is sequestrated in cytoplasm in resting cells [37]. Phosphorylation of IκBα releases the p50/p65 dimer, which translocates to the nucleus and binds κB elements in promoter regions, leading to target gene activation. Although we showed that the activation of NF-κB pathway is necessary for Hcy-induced apoptosis, how NF-κB associates in the apoptotic pathway remains to be clarified. However, recruitment of NF-κB into mitochondria during apoptosis suggests a possible role of NF-κB in cytochrome c release [38–40]. Indeed, NF-κB activation leads to translocation of bax, which may oligomerize to form a pore in the mitochondrial membrane, and PDTC inhibits the translocation [41]. Hyperhomocysteinemia is commonly found in postmenopausal women and the elderly, individuals in whom osteoporosis and fracture more commonly occur. Hcy is associated with lower BMD and increased fracture rates, and Hcy-lowering treatments with vitamin B12 and folate, were found to reduce the risk of hip fracture [15,20]. A very recent study reported that Hcy stimulated osteoclastic activity and bone resorption in a human peripheral blood mononuclear cell culture system, although they did not study its detailed mechanisms [32]. Their report and ours suggest that Hcy may contribute to the devel opment of osteoporosis by the increased bone resorption as well as the decreased bone formation.

589

In conclusion, we have shown here that Hcy increased oxidative stress in osteoblast-lineage cells, thereby inducing apoptosis via the mitochondrial pathway. This may be a mechanism by which Hcy exerts unfavorable effects on bone. These findings suggest that Hcy reduces bone formation and subsequently increases the risk of fracture. Supplementation with NAC may attenuate this effect, particularly in individuals with hyperhomocysteinemia. Acknowledgments This work was supported by the Korea Health 21 R&D Project from Korean Ministry of Health and Welfare (Project 01-PJ3-PG6-01GN11-0002 and A050491) and 21C Frontier Proteomics Project from Korean Ministry of Science and Technology (FPR05C2-281). References [1] Center JR, Nguyen TV, Schneider D, Sambrook PN, Eisman JA. Mortality after all major types of osteoporotic fracture in men and women: an observational study. Lancet 1999;353:878–82. [2] Browner WS, Pressman AR, Nevitt MC, Cummings SR. Mortality following fractures in older women: the study of osteoporotic fractures. Arch Intern Med 1996;156:1521–5. [3] Abedin M, Tintut Y, Demer LL. Vascular calcification: mechanisms and clinical ramifications. Arterioscler Thromb Vasc Biol 2004;24:1161–70. [4] Nawroth P, Pirzer R, Fohr B, Schilling T, Ziegler R, Bierhaus A, et al. Osteoporosis and cardiovascular disease—Two sides of the same coin? Med Klin 2003;98:437–46. [5] Collin-Osdoby P. Regulation of vascular calcification by osteoclast regulatory factors RANKL and osteoprotegerin. Circ Res 2004;95:1046–57. [6] Jorgensen L, Joakimsen O, Rosvold Berntsen GK, Heuch I, Jacobsen BK. Low bone mineral density is related to echogenic carotid artery plaques: a population-based study. Am J Epidemiol 2004;160:549–56. [7] Montalcini T, Emanuele V, CEravolo R, Gorgone G, Sesti G, Perticone F, et al. Relation of low bone mineral density and carotid atherosclerosis in postmenopausal women. Am J Cardiol 2004;94:266–9. [8] Tanko LB, Bagger YZ, Christiansen C. Low bone mineral density in the hip as a marker of advanced atherosclerosis in elderly women. Calcif Tissue Int 2003;73:15–20. [9] Kado DM, Browner WS, Blackwell T, Gore R, Cummings SR. Rate of bone loss is associated with mortality in older women: a prospective study. J Bone Miner Res 2000;15:1974–80. [10] Yap S, Naughten E. Homocystinuria due to cystathionine β-synthase deficiency in Ireland: 25 years' experience of a newborn screened and treated population with reference to clinical outcome and biochemical control. J Inherit Metab Dis 1998;21:738–47. [11] Hak AE, Polderman KH, Westendorp IC, Jakobs C, Hofman A, Witteman JC, et al. Increased plasma homocysteine after menopause. Atherosclerosis 2000;149:163–8. [12] Aguilar B, Rojas JC, Collados MT. Metabolism of homocysteine and its relationship with cardiovascular disease. J Thromb Thrombolysis 2004;18:75–87. [13] Welch GN, Loscalzo J. Homocysteine and atherothrombosis. N Engl J Med 1998;338:1042–50. [14] Clarke R, Daly L, Robinson K, Naughten E, Cahalane S, Fowler B, et al. Hyperhomocysteinemia: an independent risk factor for vascular disease. N Engl J Med 1991;324:1149–55. [15] Golbahar J, Hamidi A, Aminzadeh MA, Omrani GR. Association of plasma folate, plasma total homocysteine, but not methylenetetrahydrofolate reductase C667T polymorphism, with bone mineral density in postmenopausal Iranian women: a cross-sectional study. Bone 2004;35:760–5.

590

D.J. Kim et al. / Bone 39 (2006) 582–590

[16] van Meurs JB, Dhonukshe-Rutten RA, Pluijm SM, van der Klift M, de Jonge R, Lindemans J, et al. Homocysteine levels and the risk of osteoporotic fracture. N Engl J Med 2004;350:2033–41. [17] McLean RR, Jacques PF, Selhub J, Tucker KL, Samelson EJ, Bore KE, et al. Homocysteine as a predictive factor for hip fracture in older persons. N Engl J Med 2004;350:2042–9. [18] Sato Y, Honda Y, Iwamoto J, Kanoko T, Satoh K. Homocysteine as a predictive factor for hip fracture in stroke patients. Bone 2005;36:721–6. [19] Morries MS, Jacques PF, Selhub J. Relation between homocysteine and Bvitamin status indicators and bone mineral density in older Americans. Bone 2005;37:234–42. [20] Sato Y, Honda Y, Iwamoto J, Kanoko T, Satho K. Effect of folate and mecobalamin on hip fractures in patients with stroke: a randomized controlled trial. JAMA 2005;293:1082–8. [21] Villadsen MM, Bunger MH, Carstens M, Stenkjaer L, Langdahl BL. Methylenetetrahydrofolate reductase (MTHFR) C677T polymorphism is associated with osteoporotic vertebral fractures, but is a weak predictor of BMD. Osteoporos Int 2005;16:411–6. [22] Miyao M, Morita H, Hosoi T, Kurihara H, Inoue S, Hoshino S, et al. Association of methylenetetrahydrofolate reductase (MTHFR) polymorphism with bone mineral density in postmenopausal Japanese women. Calcif Tissue Int 2000;66:190–4. [23] Lubec B, Fang-Kircher S, Lubec T, Blom HJ, Boers GH. Evidence for McKusick's hypothesis of deficient collagen cross-linking in patients with homocystinuria. Biochim Biophys Acta 1996;1315:159–62. [24] Raisz LG. Homocysteine and osteoporotic fractures—Culprit or bystander? N Engl J Med 2004;350:2089–90. [25] Seeman E. Pathogenesis of bone fragility in women and men. Lancet 2002;359:1841–50. [26] Jilka RL, Weinstein RS, Bellido T, Parfitt AM, Manolagas SC. Osteoblast programmed cell death (apoptosis): modulation by growth factors and cytokines. J Bone Miner Res 1998;13:793–802. [27] Sakamoto W, Isomura H, Fujie K, Deyama Y, Kato A, Nishihira J, et al. Homocysteine attenuates the expression of osteocalcin but enhances osteopontin in MC3T3-E1 preosteoblastic cells. Biochim Biophys Acta 2005;1740:12–6. [28] Lee SJ, Kim KM, Namkoong S, Kim CK, Kang YC, Lee H, et al. Nitric oxide inhibition of homocysteine-induced human endothelial cell apoptosis by down-regulation of p53-dependent Noxa expression

[29]

[30]

[31] [32]

[33]

[34]

[35]

[36]

[37] [38]

[39]

[40]

[41]

through the formation of S-nitrosohomocysteine. J Biol Chem 2005;280:5781–8. Byun CH, Koh JM, Kim DK, Park SI, Lee KU, Kim GS. α-Lipoic acid inhibits TNFα-induced apoptosis in human bone marrow stromal cells. J Bone Miner Res 2005;20:1125–35. Kim GS, Hong JS, Kim SW, Koh JM, An CS, Choi JY, et al. Leptin induces apoptosis via ERK/cPLA2/cytochrome c pathway in human bone marrow stromal cells. J Biol Chem 2003;278:21920–9. Martindale J, Holbrook NJ. Cellular response to oxidative stress: signaling for suicide and survival. J Cell Physiol 2002;192:1–15. Herrmann M, Widmann T, Colaianni G, Colucci S, Zallone A, Herrmann W. Increased osteoclast activity in the presence of increased homocysteine concentrations. Clin Chem 2005;51:2348–53. Hock JM, Krishnan V, Onyia JE, Bidwell JP, Milas J, Stanislaus D. Osteoblast apoptosis and bone turnover. J Bone Miner Res 2001;16:975–84. Becker JS, Adler A, Schneeberger A, Huang H, Wang Z, Walsh E, et al. Hyperhomocysteinemia, a cardiac metabolic disease: role of nitric oxide and the p22phox subunit of NADPH oxidase. Circulation 2005;111:2112–8. Chang L, Xu J, Yu F, Zhao J, Tang X, Tang C. Taurine protected myocardial mitochondria injury induced by hyperhomocysteinemia in rats. Amino Acids 2004;27:37–48. Liu AL, Zhang ZM, Zhu BF, Liao ZH, Liu Z. Metallothionein protects bone marrow stromal cells against hydrogen peroxide-induced inhibition of osteoblastic differentiation. Cell Biol Int 2004;28:905–11. Bowie A, O'Neill LA. Oxidative stress and nuclear factor-kB activation. Biochem Pharmacol 2000;59:13–23. Zamora M, Merono C, Vinas O, Mampel T. Recruitment of NF-κB into mitochondria in involved in adenine nucleotide translocase 1 (ANT1)induced apoptosis. J Biol Chem 2004;279:38415–23. Petronilli V, Penzo D, Scorrano L, Bernardi P, Lisa FD. The mitochondrial permeability transition, release of cytochrome c and cell death. J Biol Chem 2001;276:12030–4. Bottero V, Rossi F, Samson M, Mari M, Hofman P, Peyron J. IκBα, the NF-κB inhibitory subunit, interacts with ANT, the mitochondrial ATP/ ADP translocator. J Biol Chem 2001;276:21317–24. Aoki M, Nata T, Morishita R, Matsushita H, Nakagami H, Yamamoto K, et al. Endothelial apoptosis induced by oxidative stress through activation of NF-κB. Hypertension 2001;38:48–55.