Aminobisphosphonates Cause Osteoblast Apoptosis and Inhibit Bone Nodule Formation In Vitro

LABORATORY INVESTIGATIONS

Aminobisphosphonates Cause Osteoblast Apoptosis and InhibitBone Nodule Formation In Vitro

Aymen I. Idris Æ Javier Rojas Æ Iain R. Greig ÆRob J. van’t Hof Æ Stuart H. Ralston

Received: 15 October 2007 / Accepted: 27 December 2007 / Published online: 8 February 2008

� Springer Science+Business Media, LLC 2008

Abstract Bisphosphonates are widely used for the treat-

ment of bone diseases associated with increased

osteoclastic bone resorption. Bisphosphonates are known

to inhibit biochemical markers of bone formation in vivo,

but it is unclear to what extent this is a consequence of

osteoclast inhibition or a direct inhibitory effect on cells of

the osteoblast lineage. In order to investigate this issue, we

studied the effects of various bisphosphonates on osteoblast

growth and differentiation in vitro. The aminobisphosph-

onates pamidronate and alendronate inhibited osteoblast

growth, caused osteoblast apoptosis, and inhibited protein

prenylation in osteoblasts in a dose-dependent manner over

the concentration range 20-100 lM. Further studies

showed that alendronate in a dose of 0.1 mg/kg inhibited

protein prenylation in calvarial osteoblasts in vivo, indi-

cating that alendronate can be taken up by osteoblasts in

sufficient amounts to inhibit protein prenylation at clini-

cally relevant doses. Pamidronate and alendronate inhibited

bone nodule formation at concentrations 10-fold lower than

those required to inhibit osteoblast growth. These effects

were not observed with non-nitrogen-containing

bisphosphonates or with other inhibitors of protein pre-

nylation and were only partially reversed by cotreatment

with a fourfold molar excess of ß-glycerol phosphate. We

conclude that aminobisphosphonates cause osteoblast

apoptosis in vitro at micromolar concentrations and inhibit

osteoblast differentiation at nanomolar concentrations by

mechanisms that are independent of effects on protein

prenylation and may be due in part to inhibition of min-

eralization. While these results need to be interpreted with

caution because of uncertainty about the concentrations of

bisphosphonates that osteoblasts are exposed to in vivo, our

studies clearly demonstrate that bisphosphonates exert

strong inhibitory effects on cells of the osteoblast lineage at

similar concentrations to those that cause osteoclast inhi-

bition. This raises the possibility that inhibition of bone

formation by bisphosphonates may be due in part to a

direct inhibitory effect on cells of the osteoblast lineage.

Keywords Aminobisphosphonate � Osteoblast �Apoptosis � Bone nodule formation � Bisphosphonate

Bisphosphonates are widely used for the prevention and

treatment of osteoporosis and other bone diseases charac-

terized by accelerated osteoclastic bone resorption.

Bisphosphonates such as etidronate and clodronate suppress

bone resorption by incorporating within nonhydrolyzable

ATP analogues that impair oxidative phosphorylation and

lead to osteoclast death [1, 2]. In contrast, nitrogen-con-

taining bisphosphonates such as alendronate, ibandronate,

risedronate, and zoledronate inhibit the enzyme farnesyl

diphosphate synthase and block prenylation of small

GTPases such as Ras, Rac, Rho, and cdc42 [3–5]. This

results in accumulation of constitutively active

A. I. Idris � J. Rojas � R. J. van’t Hof � S. H. Ralston (&)

Rheumatic Diseases Unit, Molecular Medicine Centre,

University of Edinburgh, General Western Hospital,

Edinburgh EH4 2XU, UK

e-mail: [email protected]

A. I. Idris

e-mail: [email protected]

I. R. Greig

Department of Medicine and Therapeutics, Institute of Medical

Sciences, University of Aberdeen, Foresterhill,

Aberdeen AB25 2ZD, UK

123

Calcif Tissue Int (2008) 82:191–201

DOI 10.1007/s00223-008-9104-y

unprenylated GTPases in the cytoplasm of the osteoclast,

which causes inappropriate activation of downstream sig-

naling pathways, leading to disruption of normal osteoclast

function [6]. In addition to their cellular effects, bisphos-

phonates bind avidly to hydroxyapatite crystals. Although

all bisphosphonates share the same phosphorus–carbon–

phosphorus core, different bisphosphonates differ markedly

in their affinity for hydroxyapatite binding, and this is

thought to play a role in determining their duration of action

[7]. While the main effect of bisphosphonates in vivo is on

osteoclast activity, effects on cells of the osteoblast lineage

have also been described. For example, it has been reported

that bisphosphonates at very low concentrations can protect

against apoptosis induced by etoposide and corticosteroids

in osteoblasts and osteocytes through connexin 43 channels

and extracellular signal–regulated kinase (ERK) phos-

phorylation [8, 9]. Other researchers have reported that

bisphosphonates stimulate mineralized bone nodule for-

mation in vitro [10] and promote differentiation of

mesenchymal stem cells into osteoblasts [11]. While these

experiments raise the possibility that bisphosphonates may

have positive effects on osteoblasts and osteocytes as well

as antiresorptive properties, there is accumulating evidence

from studies in humans and experimental animals that bis-

phosphonates actually suppress bone formation in vivo [12]

and impair the anabolic response to parathyroid hormone

[13–18]. In view of this, we have re-evaluated the effects of

bisphosphonates on osteoblast growth, differentiation, and

function in vitro to determine if we could find any evidence

for inhibitory effects on cells of the osteoblast lineage.

Methods

Materials

All bisphosphonates and other reagents were purchased

from Sigma (Poole, UK), unless otherwise stated.

Rabbit Osteoclast Cultures

Bone marrow cells were flushed from the long bones of

newborn rabbits and plated onto six-well tissue culture

plates in 2 mL of a-minimal essential medium (a-MEM;

GIBCO Life Technologies, Paisley, UK) supplemented

with 10 nM 1,25-dihyroxyvitamin D3 (Alexis Biochemi-

cals, Nottingham, UK), 10% fetal calf serum (FCS), and

penicillin/streptomycin at a seeding density of 1 x 106

cells/well. Test substances were added on day 7 of the

culture, and the cultures were terminated on day 9. At the

end of the culture period, osteoclasts were identified by

staining for tartrate-resistant acid phosphatase (TRAP) as

described [19]. Cells that were TRAP-positive with three or

more nuclei were considered to be osteoclasts.

Osteoblast Culture

Primary osteoblasts were isolated from the calvarial bones of

2-day-old mice by sequential collagenase digestion as

described [20] and maintained in a-MEM supplemented with

10% FCS and penicillin at 37�C in 5% CO2. Prior to exper-

imentation, the cells were seeded into 96-well plates at

5 9 103 cells/well, 48-well plates at 12 9 103 cells/well, or

12-well plates at 60 9 103 cells/well in culture medium and

ALN

CLO

ETI

PAM

ALN

CLO

ETI

PAM

tnuoctsalcoetsO

)llew/

CN

M+

PA

RT(

0

40

80

120

160

200

240

280

320b

a

*

***

****

****

*

*

rebmu

NtsalboetsO

)lortnoC

%(

**0

20

40

60

80

100

0.1 1 10 100

****

****

**

**

Bisphosphonate (µM)

0.1 1 10 100


Fig. 1 Nitrogen-containing bisphosphonates inhibit viability of

osteoclasts and osteoblasts in vitro. Mouse osteoblasts (a) and rabbit

osteoclasts (b) were treated with the bisphosphonates indicated at 1–

100 lM for 48 hours. Osteoblast numbers were determined by

Alamar blue assay, and osteoclast numbers were assessed by counting

multinucleated TRAP-positive cells with three or more nuclei.

Osteoblast numbers are expressed as a percentage of the values

found in control cultures. Values in the graphs are means ± SEM and

were obtained from three to five independent experiments. *P \ 0.05,

**P \ 0.01 from control cultures

192 A. Idris et al.: Aminobisphosphonates and Osteoblast Apoptosis

123

left to adhere overnight. Osteoblast viability was determined

by adding 10% (v/v) of Alamar blue reagent to each well. The

cells were incubated for a further 3 hours, and fluorescence

was measured (excitation 530 nm, emission 590 nm) using a

BioTek (Winooski, VT) Synergy HT plate reader. Alkaline

phosphatase (ALP) activity was assessed by a colorimetric

assay, which measures the conversion of p-nitrophenyl

phosphate to p-nitrophenyl in cell lysate. The cell layer was

homogenized in lysis buffer (1 M diethanolamine, 1 mM

MgCl2, 0.05% Triton X-100), and the lysates were stored a -

20�C until use. The cell lysate was mixed with an equal

volume of p-nitrophenol phosphate at a concentration of 20

mM in lysis buffer, and absorbance was measured at 414 nm

at 37�C in relation to a standard curve of p-nitrophenol

known concentrations. ALP activity was normalized to cell

number as determined by the Alamar blue assay.

Detection of Apoptosis

Mouse osteoblasts and rabbit osteoclasts were seeded onto

48-well plates in culture medium. Following treatments,

both adherent and nonadherent cells were collected and

cytospun onto glass slides. Apoptosis was detected by the

characteristic changes in nuclear morphology following

4,6-diamidino-2-phenylinole (DAPI) staining as previously

described [21] and by detection of fragmented DNA by

terminal deoxynucleotidyl transferase–mediated dUTP

nick end labeling (TUNEL) using the ApopTag Plus

Fluorescein In Situ Apoptosis Detection Kit (Intergen,

Purchase, NY) according to manufacturer’s instructions.

Cells with clearly fragmented DNA or positive ApopTag

labeling were counted and expressed as a percentage of

total cell number. Activities of caspases 3 and -7 were

measured using the commercially available Apo-ONE

homogeneous caspase-3/7 assay kit (Intergen). In brief, an

equal amount of the supplied caspase-3/7 substrate/buffer

reagent was added to the cell lysate and incubated for 4

hours. Caspase-3/7 enzymatic activity in the cell lysate was

measured at an excitation wavelength of 485 nm and an

emission of 530 nm on a BioTek FL600 microplate reader.

Data were corrected for background (no substrate or no cell

lysate) and expressed as fold stimulation over the vehicle

control.

vehi

cle

ALN

C

LO

TunelDAPI

ALN

CLO

ETI

PAM

Apo

ptot

ic O

steo

clas

ts (

%)

vehi

cle

ALN

C

LO

TunelDAPI

ALN

CLO

ETI

PAM

0

20

40

60

80

100

0.1 1 10 100

**

**

**

**

**

** **

** *

0

20

40

60

80

100

0.1 1 10 100



**

**

**

**

**

** **

* Apo

ptot

ic O

steo

blas

ts (

%)

a b

c

d

Fig. 2 Nitrogen-containing

bisphosphonates cause

apoptosis in rabbit osteoclasts

and mouse calvarial osteoblasts

in vitro. Mouse osteoblasts (a)

and rabbit osteoclasts (c) were

treated with bisphosphonate at

concentrations of 1–100 lM for

48 hours. Apoptotic cells were

visualized by DAPI and

TUNEL staining, and cells with

fragmented DNA were counted

and expressed as a percentage of

the total cell number. Values in

the graphs are means ± SEM

and were obtained from three to

five independent experiments.

*P \ 0.01, **P \ 0.001 from

control. (b) Representative

photomicrographs of apoptotic

osteoblasts and (d) osteoclasts

visualized by DAPI and

TUNEL staining exposed to

vehicle, clodronate (CLO), or

alendronate (ALN) at 100 lM

A. Idris et al.: Aminobisphosphonates and Osteoblast Apoptosis 193

123

Bone Nodule Assay

Primary calvarial osteoblasts were seeded into six-well

plates at 1 x 105 cells/well in 2 mL of a-MEM supplemented

with 10% FCS, penicillin, streptomycin, 2 mM glutamine,

10 mM ß-glycerol phosphate (bGP), and 50 lg/mL L-

ascorbic acid. The cells were cultured for 10–21 days, with

replacement of the culture medium every 3 days and daily

replacement of L-ascorbic acid. Test substances were added

every 3 days, followed by 1 day in culture medium alone;

and these cycles were repeated throughout the culture per-

iod. At the end of the cultures, the plates were washed five

times with phosphate-buffered saline (PBS) and fixed in

70% cold ethanol for 1 hour, and mineralized nodules were

detected by alizarin red staining. Alizarin red staining was

performed by washing the cultures in deionized water and

immersing them in 40 mM alizarin red S solution (pH 4.2)

for 20 minutes at room temperature on an orbital rotator.

The cultures were then washed four times with deionized

water to remove any unbound alizarin red. The number of

nodules was evaluated and quantified according to the

method described by Stewart et al. [22]. Briefly, cells were

destained for 15 minutes on an orbital rotator in a solution

of 10% (w/v) cetylpyridinium chloride in 10 mM sodium

phosphate (pH 7.0). The absorbance of the extracted stain

was then measured at 562 nm using a BioTek Synergy HT

plate reader and compared to an alizarin red S standard

curve. The values were normalized to cell number as

determined by the Alamar blue assay.

Measurement of Osteocalcin and Collagen

Primary calvarial osteoblasts were plated at a density of

60 9 103 cells into 12-well culture plates for 5 days. The

medium was replaced with a-MEM containing 10 mM bGP

for 24 hours. Then, the cells were treated with vehicle or

drugs in the presence of 10 mM b-GP for 72 hours, and

osteocalcin and collagen contents in culture medium were

measured according to the manufacturers’ instructions,

Actin

Jnk

p38

Osteoblasts OsteoclastskD

60

50

40

30

p-Jnkp-Jnk

Jnk

60

50

40

30

60

50

40

30

60

50

40

30

60

50

40

30

20

Actin

C3i

C3a

60

50

40

30

20C3a

C3i

60

50

40

30

60

50

40

30

kD

60

50

40

30

60

50

40

30

ALN

M C 1 50

CLO ETI PAM

1 501 501 50

ALN

M C 1 50

CLO ETI PAM

1 501 501 50

p-p38

p38

p-p38

ALN

M C 1 50

CLO ETI PAM

1 501 501 50

ALN

M C 1 50

CLO ETI PAM

1 501 501 50

ALN

M C 1 50

CLO ETI PAM

1 501 501 50

ALN

M C 1 50

CLO ETI PAM

1 501 501 50

a

c

e

b

d

f


bisphosphonates activate p38

MAPK, JNK MAPK, and

caspase-3 in osteoclasts and

osteoblasts in vitro. Mouse

calvarial osteoblasts (a, c, e) and

rabbit osteoclasts (b, d, f) were

exposed to vehicle (C) or the

bisphosphonates alendronate

(ALN), clodronate (CLO),

etidronate (ETI), or

pamidronate (PAM) at 1 lM

and 50 lM for 48 hours; and

cell lysates were analyzed by

Western blotting. M, molecular

weight marker (kDa); p38,

native p38; p-p38,

phosphorylated p38; Jnk, native

JNK; p-Jnk, phosphorylated

JNK; C3i, inactive caspase-3;

C3a, active (cleaved) caspase-3.

Results are representative of

three independent experiments


123

using the Osteocalcin ELISA kit (Biomedical Technologies,

Stoughton, MA) or the Collagen Sircol Collagen Assay kit

(Biocolor, Newtonabbey, Northern Ireland), respectively.

Western Blotting

Rabbit osteoclasts and mouse osteoblasts were seeded in 12-

well plates to confluence and treated with test compounds

for the desired period of time. Following the incubation

period, the drug/vehicle-containing medium was removed

and the monolayer rinsed with ice-cold PBS. Adherent cells

were then gently scraped in standard lysis buffer (0.1% [w/

v] SDS, 0.5% [w/v] sodium deoxycholate, 1% Triton X-100,

1 mM ethylenediaminetetraacetic acid [EDTA], and 2% [v/

v] protease inhibitor cocktail). For studies involving

extraction of phosphorylated proteins, 10 mM of sodium

fluoride and 2% (v/v) phosphatase inhibitor cocktail were

added to the standard lysis buffer described above. The

lysate was incubated on ice for 10 minutes and spun at

12,000 x g at 4�C for 10 minutes. The supernatant was

collected and the protein concentration determined using the

bicinchoninic acid assay (Pierce, Rockford, IL). Total pro-

tein (30–50 lg) was resolved by sodium dodecyl sulfate

(SDS)-polyacrylamide gel electrophoresis on 4–12% poly-

acrylamide SDS gels, transferred onto polyvinylidene

difluoride membranes (Bio-Rad, Hemel Hempstead, UK),

and immunoblotted with antibodies according to the man-

ufacturer’s instructions. Intact (p-30) and activated

(cleaved, p-19) caspase–3 levels were detected using a goat

polyclonal anti-caspase-3 antibody (Santa Cruz Biotech-

nology, Santa Cruz, CA) and a rabbit polyclonal anti-

cleaved caspase-3 antibody (Cell Signaling Technology,

Beverly, MA). Native and phosphorylated p38 and Jun N-

terminal kinase (JNK) were detected using antibodies to

ERK and phospho-ERK (Cell Signaling Technology). Un-

prenylated Rap1A was detected using a goat polyclonal

antibody (Santa Cruz Biotechnology) followed by a goat

polyclonal antiactin antibody. Immunocomplexes were

visualized by an enhanced chemiluminescence detection kit

(Pierce, Rockford, IL) using horseradish peroxidase–con-

jugated secondary antibody.

Statistical Analyses

Statistical analyses were performed using SPSS for Win-

dows, version 11 (SPSS, Chicago, IL). Significant

differences between groups were determined by analysis of

variance. If more than two groups were compared, signif-

icant differences were determined using Tukey’s posttest.

All data are presented as means ± standard error of the

mean (SEM) unless stated otherwise. P \ 0.05 was con-

sidered significant.

Results

Aminobisphosphonates Inhibit Osteoblast Growth and

Stimulate Apoptosis

The aminobisphosphonates alendronate and pamidronate

inhibited osteoblast growth in a concentration-dependent

manner at concentrations greater than 1 lM, whereas

etidronate and clodronate showed only slight inhibition at

100 lM (Fig. 1a). The concentration at which different

bisphosphonates inhibited osteoblast growth by 50% (IC50)

was 13.9 ± 0.9 lM for alendronate, 8.4 ± 0.8 lM for

pamidronate, and [ 100 lM for clodronate and etidronate.

We observed no significant effects of any bisphosphonate

on osteoblast growth or apoptosis at concentrations below

100 nM (data not shown). Inhibitory effects of the

12000

VehicleSB203580SP600125SB203580+ SP600125

0

2000

4000

6000

vehicle

Cca 7/3esa

psa

**

****** ** **

****

vitit

y(A

U)

10000

8000

CLO PAMALN

VehicleSB203580SP600125SB203580+ SP600125

**

****

**

20

40

60

80

100

****

**

**

% cit

o tp

op

A oet s

ocl

ats

0CLO PAMvehicle ALN

a

b

Fig. 4 Nitrogen-containing bisphosphonates induce caspase activity

(a) and apoptosis (b) in osteoblasts independent of p38 and JNK

MAPK activation. Mouse osteoblasts were treated with the indicated

agents in the presence and absence of SB203580 (10 lM), SP600125

(10 lM), or the combination of 5 lM of SB203580 and SP600125 for

12 hours (a) or 48 hours. Apoptotic bone cells were visualized by

DAPI staining and TUNEL staining, and cells with clearly frag-

mented DNA were counted and expressed as a percentage of total

DAPI-positive nuclei (**P \ 0.01). Activity of caspase-3 and -7 is

expressed in arbitrary units (AU); **P \ 0.01 from control cultures

and clodronate-treated cultures. ALN, alendronate; CLO, clodronate;

PAM, pamidronate


123

bisphosphonates were also observed in rabbit osteoclasts at

concentrations similar to those that caused osteoblast

inhibition. The IC50 values for osteoclasts were

16.2 ± 2.0 lM for alendronate, 8.4 ± 0.8 lM for

pamidronate, and greater than 100 lM for clodronate and

etidronate (Fig. 1b). All of the bisphosphonates tested

stimulated osteoblast apoptosis in a concentration-depen-

dent manner from 20 lM and above, but the non-nitrogen-

containing bisphosphonates etidronate and clodronate had

relatively weak effects on apoptosis even at 100 lM

(Fig. 2). In contrast, approximately 80% of osteoblasts

were apoptotic when exposed to alendronate and pamidr-

onate at a concentration of 100 lM (Fig. 2a), and similar

effects were observed in osteoclasts (Fig. 2b). Exposure of

osteoclast and osteoblast cultures to alendronate and

pamidronate at a concentration of 50 lM caused phos-

phorylation of p38 mitogen-activated protein kinase

(MAPK) and JNK and activation of caspase-3 (Fig. 3).

Clodronate and etidronate had no effect on p38, JNK, or

caspase activation at the same concentration; and none of

the bisphosphonates tested affected p38, JNK, or caspase at

a concentration of 1 lM (Fig. 3). We studied the effects of

the JNK inhibitor SB203580 and the p38 kinase inhibitor

SP600125 on bisphosphonate-induced apoptosis in osteo-

blasts and osteoclasts since previous workers had reported

that p38 inhibitors enhance aminobisphosphonate-induced

apoptosis in osteoclasts [6]. However, no significant effect

of either agent on apoptosis was observed in either cell type

(Fig. 4).

Aminobisphosphonates Inhibit Protein Prenylation in

Osteoblasts In Vitro and In Vivo

Alendronate and pamidronate resulted in accumulation of

unprenylated Rap1A at concentrations of 50 lM and

above in osteoblasts and osteoclasts, but clodronate and

etidronate did not affect Rap1A prenylation in either cell

type (Fig. 5a–d). Alendronate also inhibited protein pre-

nylation in osteoblasts in vivo since there was

accumulation of unprenylated Rap1A in calvarial osteo-

blasts harvested from mice which had been given a single

injection of alendronate 0.1 mg/kg 24 hours previously

(Fig. 5d).

Rap1A

0

5

10

15

20

25

C A LN CLO ETI PAM

eD

nsyti

)U

A(

ALN

kD

60

50

40

30

20

Actin

0

4

8

12

16

20

C A LN CLO ETI PAM

Dsne

( ytim

m2)

50µM

100µM

M C 50 100 50 100 50 100 50 100

CLO E TI PAM

kD

60

50

40

30

20

Osteoclasts in vitro24 hours

Osteoblasts in vitro24 hours 50µM

100µM

ALN 0.1mg/KgControlM

Osteoblasts ex vivokD60

50

40

30

20

ALN

M C 50 100 50 100 50 100 50 100

CLO E TI PAM

Rap1A

Actin

Rap1A

Actin

a

c

e

b

d


bisphosphonates induce the

accumulation of unprenylated

Rap1A in bone cells in vitro and

in vivo. Western blot analysis

was performed in (a, b) rabbit

osteoclasts or (c, d) osteoblasts

treated with vehicle or the

indicated compound (50 or

100 lM) for 24 hours. (e)

Calvarial osteoblasts were

isolated from mice treated with

a single injection of ALN (0.1

mg/kg). Total cellar protein was

subjected to Western blot

analysis (50 lg/lane) using

rabbit antibodies directed

against beta actin and

unprenylated Rap1A. Identical

experiments were repeated three

times


123

Aminobisphosphonates Inhibit Bone Nodule Formation

In Vitro

Further studies were performed to analyze the effects of

bisphosphonates on bone nodule formation. Although the

bisphosphonates had no significant effect on cell growth or

apoptosis at concentrations of 1 lM and below (Figs. 1 and

2), alendronate and pamidronate strongly inhibited bone

nodule formation at concentrations of 500 nM, 750 nM

(Fig. 6a, b), and 1 lM (not shown). In contrast, etidronate

and clodronate had no inhibitory effect on bone nodule for-

mation at these concentrations or concentrations of up to

1 lM (Fig. 6a, b). No inhibitory effects on ALP activity or

cell number were observed at these concentrations (Fig. 6c,

d). We went on to study the effects of other inhibitors of

protein prenylation on bone nodule formation and attempted

to rescue the inhibitory effect of aminobisphosphonates on

bone nodule formation with farnesyl pyrophosphate (FPP).

Neither mevastatin (1 lM), geranylgeranyltransferase

inhibitor 298 (GGTI, 1 lM) nor farnesyltransferase inhibitor

277 (FTI, 1 lM) inhibited bone nodule formation (Fig. 7a,

b), and FPP was unable to rescue the inhibitory effect of

alendronate or pamidronate on nodule formation (Fig. 7c).

However, all agents tested were found to cause accumulation

of unprenylated Rap1A in bone nodule cultures (Fig. 7d).

Taken together, these data indicate that the inhibitory effect

of aminobisphosphonates on bone nodule formation is

unrelated to effects on cell growth, ALP activity, or inhibi-

tion of protein prenylation.

We went on to study the possibility that the inhibitory

effect of aminobisphosphonates on bone nodule formation

might be mediated by an effect on mineralization, by

adding increased amounts of bGP to the cultures (Fig. 8).

This showed that the inhibitory effect of 250 nM alendr-

onate on bone nodule formation could be partially rescued

by bGP, whereas the inhibitory effect of 500 nM alendr-

onate on nodule formation persisted even in the presence of

20 mM bGP.

PhaseContrast

AlizarinRed

Veh

ALN750nM

CLO750nM

PAM750nM

120

P

140

mun lleC

br

e

2

4

6

8

10

0

20

40

60

80/deR nirazil

Alb aive

llec s 500nM 750nM140

**

****

**

120

100

ETI750nM

0

40

80

LA

100

60

20

14

12

Veh ALN CLOPAM0

ETI

Veh ALN CLOPAM ETI

Veh ALN CLOPAM ETI

500nM 750nM

500nM 750nM

a b

c

d

Fig. 6 Aminobisphosphonates

inhibit bone nodule formation in

vitro. Mouse calvarial

osteoblasts were cultured in

osteogenic medium for up to 21

days and exposed to

bisphosphonates at the

concentrations indicated every

72 hours. Representative

photomicrographs from the

cultures are shown in (a) and

quantitation of nodule formation

in (b) expressed as a percentage

of vehicle-treated cultures. (c)

ALP activity. (d) Cell number

from cultures at 21 days.

Columns in the graphs are

means and bars are SEM.

**P \ 0.01 from vehicle


123

Discussion

Bisphosphonates are highly effective agents for the treat-

ment for bone diseases associated with increased osteoclast

activity, but recent evidence suggests that they also inhibit

bone formation [12] and blunt the anabolic response to

parathyroid hormone [13, 15–17, 23]. The mechanisms by

which bisphosphonates exert these inhibitory effects on

bone formation are poorly understood, and in fact the

negative effects of bisphosphonates in vivo conflict to some

extent with the results of several preclinical studies which

have reported that bisphosphonates can protect against

osteoblast and osteocyte apoptosis and promote osteoblast

differentiation. In this study, therefore, we investigated the

effects of different bisphosphonates on osteoblast growth

and differentiation in vitro over a wide range of concen-

trations and evaluated their effects on the mevalonate

pathway and other signaling pathways in osteoblasts. We

found that alendronate and pamidronate inhibited osteoblast

growth and promoted osteoblast apoptosis at concentrations

of 10 lM and above. Osteoblast growth was completely

arrested at aminobisphosphonate concentrations of 50 lM

Veh Mev 1µM

0

20

40

60

80

100

120

140

)heV

%( deR ni razil

A

Veh MEV GGTIFTI

GGTI 1µMFTI 1µMlcihe

Ve

Veh ALN 1µM PAM 1µM

PP

F 01(µ

)M

kD60

50

40

30

20

Actin

Rap1A

M MEV(1µM)

FTI(1µM)

GGTI(1µM)

Veh ALN(0.5µM)

d

c

a b

Fig. 7 Nitrogen-containing bisphosphonates inhibit bone nodule

formation by mechanisms independent of an effect on protein

prenylation. Mouse calvarial osteoblasts were cultured in osteogenic

medium for 21 days and exposed to test agents at the concentrations

indicated. (a) Representative photomicrographs from cultures

exposed to vehicle (Veh), mevastatin (MEV), FTI, and GGTI at the

concentrations indicated. (b) Quantitation of nodule formation from

these cultures expressed as a percentage of the value in vehicle-

treated cultures. There was no significant difference between the

agents. (c) Representative cultures from cultures treated with

alendronate (ALN) and pamidronate (PAM) in the presence and

absence of FPP at the concentrations indicated are shown (10 lM).

(d) Calvarial osteoblasts were cultured in osteogenic medium in the

presence of vehicle (Veh), alendronate (ALN), Mevastatin (MEV),

farnesyl transferase inhibitor 277 (FTI) or geranylgeranyltransferase

inhibitor 298 (GGTI) at the concentrations indicated for 21 days. Cell

lysates were prepared and analysed by western blotting for unpreny-

lated Rap1A and beta actin using rabbit antibodies directed against

beta actin and unprenylated Rap1A


123

and above, due to the occurrence of osteoblast apoptosis in

about 80% of cells in the cultures. Osteoclast apoptosis and

osteoclast inhibition were observed at similar concentra-

tions to those that caused osteoblast inhibition and

apoptosis. The non-nitrogen-containing bisphosphonates

etidronate and clodronate also caused osteoblast and

osteoclast apoptosis by about 10%, but this was only

observed at high concentrations of 50 and 100 lM. Expo-

sure of osteoblasts and osteoclasts to pamidronate and

alendronate resulted in inhibition of protein prenylation,

activation of caspase-3, and activation of the p38 MAPK

and JNK signaling pathways. These findings are in broad

agreement with the findings of Dunford et al. [6], who

recently reported that aminobisphosphonates cause apop-

tosis and activation of the MAPK and JNK signaling

pathways in osteoclasts and J774 macrophages in vitro.

Unlike Dunford et al., who reported that p38 MAPK

inhibitors partially rescued bisphosphonate-induced apop-

tosis, we found that inhibitors of the MAPK and JNK

pathways had no effect on osteoblast or osteoclast apoptosis

induced by aminobisphosphonates.

The stimulation of osteoblast apoptosis which we

observed with aminobisphosphonates in this study are

consistent with the findings of previous studies which have

shown that aminobisphosphonates such as pamidronate and

alendronate cause apoptosis of various cell types including

Caco2 intestinal epithelial cells [24, 25], myeloma cells

[26], and keratinocytes [27]. Taken together, these obser-

vations show that the proapoptotic effects of

aminobisphosphonates are not specific to osteoclasts but

are observed in many other cell types including osteoblasts.

The studies reported here seem to contrast with those of

Plotkin and colleagues [8, 28], who reported that amin-

obisphosphonates protect against osteoblast and osteocyte

apoptosis at concentrations in the nanomolar to picomolar

range. In these studies, however, the protective effect of

bisphosphonates was observed in cells which had been

induced to undergo apoptosis by exposure to glucocorti-

coids or etoposide. Our results cannot be directly compared

with those of Plotkin et al. since we did not evaluate the

effects of bisphosphonates on experimentally induced

apoptosis, but we observed no significant effects on

osteoblast growth or apoptosis at bisphosphonate concen-

trations below 100 nM. The inhibitory effects on bone

nodule formation which we observed conflict to some

extent with the findings of Giuliani et al. [10], who reported

that etidronate stimulated bone nodule formation over the

concentration range 10-5 to 10-9 M. We observed no

stimulatory effect of etidronate on bone nodule formation

at these concentrations. In agreement with Giuliani et al.,

however, we found that alendronate inhibited bone nodule

formation at concentrations in the micromolar range; but

we found no effect of the aminobisphosphonates on bone

nodule formation at concentrations lower than 500 nM. The

inhibitory effect of the aminobisphosphonates on bone

nodule formation was not related to an effect on cell

growth or ALP activity and was not mediated by inhibition

of protein prenylation since it could not be rescued by

coadministration of FPP and was not reproduced by other

inhibitors of protein prenylation including FTI, GGTI, and

mevastatin. We gained some evidence to suggest that the

inhibitory effect of alendronate on bone nodule formation

was mediated in part by an interaction with hydroxyapatite

since it was partially rescued by high concentrations of

bGP. It is currently unclear whether inhibition of miner-

alization could completely explain the bisphosphonate-

induced inhibition of bone nodule formation since it was

not rescued by a 100-fold excess of bGP. It could be that

prolonged exposure to bisphosphonates in the bone nodule

assays have had adverse effects on osteoblast function that

were not reflected by changes in cell number or ALP

activity, but this remains speculative. Accordingly, further

work will be required to determine the exact mechanism by

which relatively low concentrations of bisphosphonates

inhibit bone nodule formation.

While we observed that aminobisphosphonates promote

osteoblast apoptosis and osteoclast apoptosis at similar

concentrations in vitro, it is likely that osteoclasts may be

exposed to higher concentrations of bisphosphonates than

osteoblasts in vivo. This is because bisphosphonates bind

to bone matrix and preferentially target to surfaces under-

going resorption [29]. There is also evidence that

Fig. 8 bGP partially rescues the inhibitory effect of alendronate on

bone nodule formation. Mouse calvarial osteoblasts were cultured in

osteogenic medium for 21 days and exposed to alendronate (ALN)

and bGP at the concentrations indicated


123

acidification causes release of bisphosphonate from bone

mineral and that components of bone matrix are internal-

ized by the osteoclast during the resorptive process [30],

providing several routes by which bisphosphonate might be

released at relatively high concentrations within the

osteoclast cytoplasm. The mechanisms by which bisphos-

phonates might be taken up by osteoblasts are less clear,

but bisphosphonates can be taken up by any cell by fluid

phase endocytosis; also, studies using radiolabeled bis-

phosphonates have shown that a proportion of the

administered dose is located at bone-forming surfaces [29].

Since bisphosphonates accumulate in bone, it is possible

that with repeated dosing osteoblasts may be exposed to

sufficient bisphosphonate to affect cellular function.

It should be acknowledged, however, that the results of

any in vitro studies with bisphosphonates need to be

interpreted with caution with regard to their relevance in

vivo since there is no information on the concentrations of

bisphosphonate that osteoblasts, osteocytes, or osteoclasts

are exposed to in the bone microenvironment. Notwith-

standing this uncertainty, we found that exposure of mice

to alendronate at a dose which is equivalent to that used

clinically (0.1 mg/kg) caused inhibition of the mevalonate

pathway in osteoblasts in vivo as reflected by accumulation

of unprenylated Rap1A in calvarial osteoblasts. This

demonstrates that bisphosphonates can be taken up by

osteoblasts in vivo at sufficient concentrations to affect

intracellular signaling. Further evidence that amin-

obisphosphonates inhibit osteoblast function in vivo at

clinically relevant doses comes from the observations of

Tobias et al. [12], who found that pamidronate inhibited the

activity of new and preexisting bone-forming surfaces in

ovariectomized rats and from the observations of several

workers who have found that bisphosphonates inhibit the

anabolic effect off parathyroid hormone in humans and

experimental animals [13–18].

In summary, the findings presented in this study clearly

show that bisphosphonates negatively impact osteoblast

growth and differentiation. This raises the possibility that

the negative effects of bisphosphonates on bone formation

in vivo might be due, at least in part, to a direct inhibitory

effect on cells of the osteoblast lineage.

Acknowledgements This work was supported by a grant from the

Arthritis Research Campaign (UK). A. I. is supported by an ECTS-

AMGEN award.

References

1. Frith JC, Monkkonen J, Auriola S, Monkkonen H, Rogers MJ

(2001) The molecular mechanism of action of the antiresorptive

and antiinflammatory drug clodronate: evidence for the formation

in vivo of a metabolite that inhibits bone resorption and causes

osteoclast and macrophage apoptosis. Arthritis Rheum 44:2201–

2210

2. Frith JC, Monkkonen J, Blackburn GM, Russell RG, Rogers MJ

(1997) Clodronate and liposome-encapsulated clodronate are

metabolized to a toxic ATP analog, adenosine 5’-(beta, gamma-

dichloromethylene) triphosphate, by mammalian cells in vitro. J

Bone Miner Res 12:1358–1367

3. van Beek E, Pieterman E, Cohen L, Lowik C, Papapoulos S

(1999) Farnesyl pyrophosphate synthase is the molecular target of

nitrogen- containing bisphosphonates. Biochem Biophys Res

Commun 264:108–111

4. Dunford JE, Thompson K, Coxon FP, Luckman SP, Hahn FM,

Poulter CD, Ebetino FH, Rogers MJ (2001) Structure–activity

relationships for inhibition of farnesyl diphosphate synthase in

vitro and inhibition of bone resorption in vivo by nitrogen-

containing bisphosphonates. J Pharmacol Exp Ther 296:235–

242

5. Luckman SP, Hughes DE, Coxon FP, Graham R, Russell G,

Rogers MJ (1998) Nitrogen-containing bisphosphonates inhibit

the mevalonate pathway and prevent post-translational prenyla-

tion of GTP-binding proteins, including Ras. J Bone Miner Res

13:581–589

6. Dunford JE, Rogers MJ, Ebetino FH, Phipps RJ, Coxon FP

(2006) Inhibition of protein prenylation by bisphosphonates

causes sustained activation of Rac, Cdc42, and Rho GTPases. J


7. Nancollas GH, Tang R, Phipps RJ, Henneman Z, Gulde S, Wu W,

Mangood A, Russell RG, Ebetino FH (2006) Novel insights into

actions of bisphosphonates on bone: differences in interactions

with hydroxyapatite. Bone 38:617–627

8. Plotkin LI, Weinstein RS, Parfitt AM, Roberson PK, Manolagas

SC, Bellido T (1999) Prevention of osteocyte and osteoblast

apoptosis by bisphosphonates and calcitonin. J Clin Invest

104:1363–1374

9. Plotkin LI, Manolagas SC, Bellido T (2006) Dissociation of the

pro-apoptotic effects of bisphosphonates on osteoclasts from their

anti-apoptotic effects on osteoblasts/osteocytes with novel ana-

logs. Bone 39:443–452

10. Giuliani N, Pedrazzoni M, Negri G, Passeri G, Impicciatore M,

Girasole G (1998) Bisphosphonates stimulate formation of

osteoblast precursors and mineralized nodules in murine and

human bone marrow cultures in vitro and promote early os-

teoblastogenesis in young and aged mice in vivo. Bone 22:455–

461

11. Duque G, Rivas D (2007) Alendronate has an anabolic effect on

bone through the differentiation of mesenchymal stem cells. J


12. Tobias JH, Chow JW, Chambers TJ (1993) 3-Amino-1-hy-

droxypropylidine-1-bisphosphonate (AHPrBP) suppresses not

only the induction of new, but also the persistence of existing

bone-forming surfaces in rat cancellous bone. Bone 14:619–

623

13. Delmas PD, Vergnaud P, Arlot ME, Pastoureau P, Meunier PJ,

Nilssen MH (1995) The anabolic effect of human PTH (1–34) on

bone formation is blunted when bone resorption is inhibited by

the bisphosphonate tiludronate—is activated resorption a pre-

requisite for the in vivo effect of PTH on formation in a

remodeling system? Bone 16:603–610

14. Black DM, Bilezikian JP, Ensrud KE, Greenspan SL, Palermo L,

Hue T, Lang TF, McGowan JA, Rosen CJ (2005) One year of

alendronate after one year of parathyroid hormone (1–84) for

osteoporosis. N Engl J Med 353:555–565

15. Ettinger B, San Martin J, Crans G, Pavo I (2004) Differential

effects of teriparatide on BMD after treatment with raloxifene or

alendronate. J Bone Miner Res 19:745–751


123

16. Samadfam R, Xia Q, Goltzman D (2007) Pretreatment with an-

ticatabolic agents blunts but does not eliminate the skeletal

anabolic response to parathyroid hormone in oophorectomized

mice. Endocrinology 148:2778–2787

17. Finkelstein JS, Hayes A, Hunzelman JL, Wyland JJ, Lee H,

Neer RM (2003) The effects of parathyroid hormone, alendro-

nate, or both in men with osteoporosis. N Engl J Med

349:1216–1226

18. Gasser JA, Kneissel M, Thomsen JS, Mosekilde L (2000) PTH

and interactions with bisphosphonates. J Musculoskelet Neuronal

Interact 1:53–56

19. van’t Hof RJ, von Lindern M, Nijweide PJ, Beug H (1997) Stem

cell factor stimulates chicken osteoclast activity in vitro. FASEB

J 11:287–293

20. van’t Hof RJ (2003) Osteoclast formation in the mouse coculture

assay. In: Helfrich MH, Ralston SH (eds) Bone research proto-

cols. Humana Press, Totowa, NJ, pp 145–152

21. van’t Hof RJ, Idris AI, Ridge SA, Dunford J, Greig IR, Ralston

SH (2004) Identification of biphenylcarboxylic acid derivatives

as a novel class of bone resorption inhibitors. J Bone Miner Res

19:1651–1660

22. Stewart TL, Roschger P, Misof BM, Mann V, Fratzl P, Klau-

shofer K, Aspden RM, Ralston SH (2005) Association of COLIA1Sp1 alleles with defective bone nodule formation in vitro and

abnormal bone mineralisation in vivo. Calcif Tissue Int 77:113–

118

23. Black DM, Greenspan SL, Ensrud KE, Palermo L, McGowan JA,

Lang TF, Garnero P, Bouxsein ML, Bilezikian JP, Rosen CJ

(2003) The effects of parathyroid hormone and alendronate alone

or in combination in postmenopausal osteoporosis. N Engl J Med

349:1207–1215

24. Twiss IM, Pas O, Ramp-Koopmanschap W, den Hartigh J,

Vermeij P (1999) The effects of nitrogen-containing bisphos-

phonates on human epithelial (Caco-2) cells, an in vitro model for

intestinal epithelium. J Bone Miner Res 14:784–791

25. Suri S, Monkkonen J, Taskinen M, Pesonen J, Blank MA, Phipps

RJ, Rogers MJ (2001) Nitrogen-containing bisphosphonates

induce apoptosis of Caco-2 cells in vitro by inhibiting the mev-

alonate pathway: a model of bisphosphonate-induced

gastrointestinal toxicity. Bone 29:336–343

26. Shipman CM, Rogers MJ, Apperley JF, Russell RG, Croucher PI

(1997) Bisphosphonates induce apoptosis in human myeloma cell

lines: a novel anti-tumour activity. Br J Haematol 98:665–672

27. Reszka AA, Halasy-Nagy J, Rodan GA (2001) Nitrogen-bis-

phosphonates block retinoblastoma phosphorylation and cell

growth by inhibiting the cholesterol biosynthetic pathway in a

keratinocyte model for esophageal irritation. Mol Pharmacol

59:193–202

28. Plotkin LI, Bellido T (2001) Bisphosphonate-induced, hemi-

channel-mediated, anti-apoptosis through the Src/ERK pathway:

a gap junction-independent action of connexin43. Cell Commun

Adhes 8:377–382

29. Sato M, Grasser W, Endo N, Akins R, Simmons H, Thompson

DD, Golub E, Rodan GA (1991) Bisphosphonate action.

Alendronate localization in rat bone and effects on osteoclast

ultrastructure. J Clin Invest 88:2095–2105

30. Nesbitt SA, Horton MA (1997) Trafficking of matrix collagens

through bone-resorbing osteoclasts. Science 276:266–269


123

Aminobisphosphonates Cause Osteoblast Apoptosis and Inhibit Bone Nodule Formation In Vitro

Documents

Transcript of Aminobisphosphonates Cause Osteoblast Apoptosis and Inhibit Bone Nodule Formation In Vitro