Eur J Vasc Endovasc Surg (2015)

-,

1e9

Bone Like Arterial Calcication in Femoral Atherosclerotic Lesions:

Prevalence and Role of Osteoprotegerin and Pericytes
J.-M. Davaine a,b,c,h, T. Quillard a,h, M. Chatelais a,d, F. Guilbaud
D. Heymann a,d,e, M.-F. Heymann a,g,**, Y. Goufc e,f,*

a,d,e

, R. Brion

a,d,e

, B. Guyomarch

e,f

, M.. Brennan a,

INSERM, UMR 957, Nantes F-44035, France

Service de Chirurgie Vasculaire, Centre Hospitalier Ren-Dubos, Pontoise, France
c
Service de Chirurgie Vasculaire, CHU Piti-Salptrire, Paris, France
d
Universit de Nantes, Nantes Atlantique Universits, Nantes F-44035, France
e
Centre Hospitalier Universitaire, Nantes, France
f
Institut du Thorax, Nantes, France
g
Department of Medical Oncology, University of Shefeld, Shefeld, UK
b

WHAT THIS PAPER ADDS

Arterial calcication is very common in arterial lesions and has a major clinical impact. Though recognized as
being a highly regulated process, most knowledge is based on studies at the coronary or carotid levels. This work
provides detailed analysis of the presence of bone like arterial calcicationdknown as osteoid metaplasia
(OM)dat the femoral level and suggest a high prevalence of OM in femoral atherosclerotic lesions. It further
suggests that vascular pericytes and the osteoprotegerin/receptor activator for the nuclear factor kappa B ligand
(RANKL)/RANK axis are implicated in the regulation of arterial calcication.

Objective/Background: Arterial calcication, a process that mimics bone formation, is an independent risk factor
of cardiovascular morbidity and mortality, and has a signicant impact on surgical and endovascular procedures
and outcomes. Research efforts have focused mainly on the coronary arteries, while data regarding the femoral
territory remain scarce.
Methods: Femoral endarterectomy specimens, clinical data, and plasma from a cohort of patients were collected
prospectively. Histological analysis was performed to characterize the cellular populations present in the
atherosclerotic lesions, and that were potentially involved in the formation of bone like arterial calcication
known as osteoid metaplasia (OM). Enzyme linked immunosorbent assays and cell culture assays were conducted
in order to understand the cellular and molecular mechanisms underlying the formation of OM in the lesions.
Results: Twenty-eight of the 43 femoral plaques (65%) displayed OM. OM included osteoblast and osteoclast like
cells, but very few of the latter exhibited the functional ability to resorb mineral tissue. As in bone,
osteoprotegerin (OPG) was signicantly associated with the presence of OM (p .04). Likewise, a high plasma
OPG/receptor activator for the nuclear factor kappa B ligand (RANKL) ratio was signicantly associated with the
presence of OM (p .03). At the cellular level, there was a greater presence of pericytes in OM compared with
OMe lesions (5.59  1.09 vs. 2.42  0.58, percentage of area staining [region of interest]; p .04); in vitro,
pericytes were able to inhibit the osteoblastic differentiation of human mesenchymal stem cells, suggesting that
they are involved in regulating arterial calcication.
Conclusion: These results suggest that bone like arterial calcication (OM) is highly prevalent at femoral level.
Pericyte cells and the OPG/RANK/RANKL triad seem to be critical to the formation of this ectopic osteoid tissue
and represent interesting potential therapeutic targets to reduce the clinical impact of arterial calcication.
2015 European Society for Vascular Surgery. Published by Elsevier Ltd. All rights reserved.
Article history: Received 12 May 2015, Accepted 5 October 2015, Available online XXX
Keywords: Atherosclerosis, Femoral artery, Osteoprotegerin, Peripheral arterial disease, Vascular calcication,
Vascular pericytes

J.-M.D. and T.Q. contributed equally to this work.

* Corresponding author. Centre Hospitalier Universitaire, Nantes, France.
** Corresponding author. INSERM, UMR 957, Nantes F-44035, France.
E-mail addresses: mariefrancoise.heymann@shefeld.ac.uk (M.-F. Heymann);
yann.gouefc@chu-nantes.fr (Y. Goufc).
1078-5884/ 2015 European Society for Vascular Surgery. Published by
Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.ejvs.2015.10.004

INTRODUCTION
The prevalence of arterial calcication in atherosclerotic
lesions has been reported to be as high as 90% and is
identied as an independent risk factor for cardiovascular
mortality.1,2 Arterial calcication also has an impact on

Please cite this article in press as: Davaine J-M, et al., Bone Like Arterial Calcication in Femoral Atherosclerotic Lesions: Prevalence and Role of
Osteoprotegerin and Pericytes, European Journal of Vascular and Endovascular Surgery (2015), http://dx.doi.org/10.1016/j.ejvs.2015.10.004

plaque stability.3 Microcalcications in the brous cap inuence plaque stability and increase the risk of rupture.4
Conversely, lesions with a high calcic burden are more
stable.5,6 During open surgery, arterial calcication results in
difculty in clamping vessels and in performing arterial
anastomoses. Also, arterial calcication inuences the
technical and clinical success of endovascular repair.7,8
Identifying the mechanisms underlying this process is thus
a priority.
The femoral territory is unique in terms of both its
anatomy and its pathology, and is very challenging for
endovascular procedures when highly calcied, but little is
known about the calcication process at the femoral level.
Arterial calcication results from a highly regulated process sharing many similarities with bone formation.9 Bone
like tissue, characterized by the presence of osteoid matrix,
osteoblast and osteoclast like cells, and the major bone
regulatory cytokines, has been observed in atheromatous
lesions.6 Smooth muscle cells (SMC) are the most
frequently studied,10 but macrophages and pericytes
recently emerged as key players in this process.6,11,12 In
particular, pericytes are mesenchymal precursors directly
involved in the development of arterial calcication.6,12 The
osteoprotegerin (OPG)/receptor activator for the nuclear
factor kappa B (NF-kB) ligand (RANKL)/receptor activator
for NF-kB (RANK) triad, which regulates bone remodeling, is
also involved in the formation of arterial calcication.13 OPG
is associated with calcication of the coronary arteries,
cardiovascular mortality, and peripheral arterial disease
(PAD).14 It was recently reported that the presence of
osteoid tissue, known as osteoid metaplasia (OM), in carotid lesions was associated with the asymptomatic nature
of the plaques. These results also support the role of pericytes and OPG in this process.6
The aim of this study was to provide detailed characteristics of osteoid tissue at the femoral level and to further
assess the involvement of pericytes and the OPG/RANK/
RANKL triad in this process.
METHODS
Patients, biological samples, and imaging data
Atheromatous plaques were harvested from 43 patients
undergoing endarterectomy of the common femoral artery
and of its bifurcation between February 2008 and June
2009. Clinical presentation was reported according to the
Rutherford classication of PAD.15 All patients participating
in the study provided written informed consent. The clinical research protocol was approved by the institutional
medical ethics committee of Nantes University Hospital.
When a computed tomography (CT) scan was available, the
importance of calcication of the lesions was assessed
separately by two authors (J.M.D. and Y.G.) and categorized as non-calcied, mildly calcied, moderately calcied
or highly calcied. Where there was disagreement, the
analysis of a third vascular surgeon was requested. Surgical
endarterectomy was performed together with best medical
treatment.

J.-M. Davaine et al.

OPG and RANKL measurements

For each patient a blood sample was collected on the day
prior to surgery, centrifuged (5 minutes at 1,465 g), and
plasma was then aliquoted and stored at 80  C for subsequent analysis.
OPG and RANKL levels were measured in plasma by
enzyme linked immunosorbent assay (DY805 ELISA kit; R&D
Systems, Minneapolis, MN, USA). Osteoblastic pericyte
conditioned medium was obtained as follows: the pericytes
were cultured for 16e18 days in an osteoblastic inducing
medium. The cells were then washed and left in starvation
medium for 48 hours before the collection of the cell supernatants. The experiments were carried out in duplicate
and repeated three times.
Histological and immunological analyses
The atherosclerotic plaques were xed in 10% formalin
overnight, decalcied in Sakura TDE 30 uid for 24 hours,
and embedded in parafn. Ten serial sections from the
middle of the atherosclerotic lesion were processed. These
sections were stained with (i) hematoxylin and eosin, (ii)
Massons trichrome, and (iii) tartrate resistant acid phosphatase (TRAP). Immunohistochemistry (IHC) was carried
out to localize and semi-quantify endothelial cells with
CD31 antibody (Dako, Glostrup, Denmark); pericytes and
endothelium with CD146 antibody (Abcam, Cambridge, UK);
pericytes with NG2 antibody (Millipore, Billerica, MA, USA);
smooth muscle actin (SMA), a marker for SMC with SMA
antibody (R&D Systems); macrophages with CD68 antibody
(Immunotech, Marseille, France); and OPG (R&D Systems).
TRAP staining was performed by incubating sections in
1 mg/mL naphthol ASeTR phosphate, 60 mM NN dimethylformamide, and 1 mg/mL fast red salt solution (pH 5.2;
SigmaeAldrich, St. Louis, MO, USA) for 1 hour.
For each IHC staining, a negative control was performed
using a similar procedure excluding the primary antibody.
Plaques were analyzed for the presence of calcication,
categorized as sheet like, nodular, and clear center calcications, and osteoid metaplasia. The prevalence of each
type of arterial calcication of this cohort of femoral lesions
has been described in detail previously.16
Images of the whole sections were obtained with the
NanoZoomer digital slide scanner (Hamamatsu Photonics,
Hamamatsu, Japan). For each patient, the whole slide was
scanned and made available for analysis. As a result, the
whole section was quantied using a dedicated Image J
(National Institutes of Health, Bethesda, MD, USA) macro
applied systematically to all samples (selection parameters
were applied to all samples for each staining). Numbers
were expressed as a percentage of the stained area over the
area of the plaque for each patient, and nal results are the
mean of the whole cohort.
Cell culture and differentiation assays
Reagents were obtained from SigmaeAldrich unless
otherwise stated. Human pericytes (C12980; PromoCell,
Heidelberg, Germany) were placed for 18 days in an

Please cite this article in press as: Davaine J-M, et al., Bone Like Arterial Calcication in Femoral Atherosclerotic Lesions: Prevalence and Role of
Osteoprotegerin and Pericytes, European Journal of Vascular and Endovascular Surgery (2015), http://dx.doi.org/10.1016/j.ejvs.2015.10.004

Bone Like Arterial Calcication in Femoral Atherosclerotic Lesions

Figure 1. (A) Representative section of a calcied femoral artery lesion. (B) Magnication of the area of osteoid metaplasia (OM) with a
dense extracellular matrix (*), osteocytes (;), and bone marrow (-) (Massons trichrome). (C) Magnication of a lesion featuring an
osteoclast like giant multi-nucleated cell (;;), in the vicinity of OM (*) (HE staining). (D) Representative section of femoral lesions with
OM showing the absence of tartrate resistant acid phosphatase (TRAP) staining. (E) Representative section of positive TRAP staining (/).

osteogenic culture medium containing vitamin D3, dexamethasone, ascorbic acid, and b-glycerophosphate supplemented with 100 U/mL penicillin and 100 U/mL
streptomycin. To test the hypothesis that pericytes may
inuence the osteochondrogenic differentiation of surrounding cells, the effect of pericyte supernatant on the
osteogenic differentiation of human mesenchymal stem
cells (Hu-MSC) was tested. The supernatant of the pericytes was prepared as follow: pericytes were plated in six
well plates and cultured for 2e3 days in standard pericyte
culture medium until they reached conuence. Then, the
medium was deprived of fetal calf serum (FCS; from 10%
to 1%) and culture was pursued for 48 hours, after which
supernatant was harvested, centrifuged, aliquoted, and
stored at 80  C. Human bone marrow derived mesenchymal stem cells from healthy donors were grown in
Dulbeccos Modied Eagle Medium (DMEM) supplemented with 10% FCS, 1 ng/mL basic Fibroblast Growth
Factor (R&D Systems), 100 U/mL penicillin, 100 U/mL
streptomycin, and 2 mM L-glutamine.17 On day 3 and until
day 28, osteoblastic differentiation was induced. To do so,
vitamin D3, dexamethasone, ascorbic acid, and b-glycerophosphate were added to the previous medium but
without the use of FGF. Mineralization was assessed using
alizarin red staining.18 As inammation has been recognized to potentiate the cell mineralizing process,19 in
certain experiments recombinant interleukin (IL)-6 and
tumor necrosis factor (TNF)-a (R&D Systems) were added

to the medium. SMC were differentiated toward osteoblastic cells by adding 3 mM inorganic phosphate to 3%
Fetal Calf Serum (FCS) containing DMEM culture medium.
Pericyte supernatant or control medium were added (1/4
dilution) when the inorganic phosphate was added. SMC
differentiation was assessed 7 days after treatment, using
alizarin red staining.
Statistics
Throughout the paper, continuous data are expressed in
terms of mean  SD, and the nominal data as n (%).
Continuous variables were compared with the t test and the
nominal variables were analyzed using the chi-square or
Fisher exact test. Statistical analyses were performed using
SSPS 19.0 (IBM, Armonk, NY, USA). A p-value < .05 was
considered statistically signicant.
RESULTS
Typical bone tissue (OM) in atheromatous femoral plaques
Between February 2008 and June 2009, 43 femoral plaques were harvested. Histological analysis revealed the
presence of typical bone like tissue in the plaque in 65%
(28/43) of the cases (Fig. 1A). This osteoid metaplasia
featured a dense lamellar extracellular matrix made of
collagen, into which osteocyte cells were embedded.
Lipidic and vascular structures similar to bone marrow
were also noted in the immediate vicinity (Fig. 1B). In 17/

Please cite this article in press as: Davaine J-M, et al., Bone Like Arterial Calcication in Femoral Atherosclerotic Lesions: Prevalence and Role of
Osteoprotegerin and Pericytes, European Journal of Vascular and Endovascular Surgery (2015), http://dx.doi.org/10.1016/j.ejvs.2015.10.004

J.-M. Davaine et al.

Table 1. Demographic and clinical data of the patients.

Total (n 43)
OM (n 28)
OMe (n 15)
p
Demographics
Age (y)a
69.2  1.5
70.0  1.9
68.0  2.3
.51
Male
37 (86)
25 (89)
12 (80)
.65
Cardiovascular risk factors
26.4  0.6
26.6  0.7
25.9  1.0
.59
BMIa
LDL (g/L)a
0.9  0
0.9  0.1
1.0  0.1
.38
HDL (g/L)a
0.5  0
0.5  0
0.5  0
.30
Total cholesterol (g/L)a
1.8  0.3
1.9  0.5
1.8  0.2
.92
Dyslipidemia
36 (84)
23 (82)
14 (88)
1.00
Smoking
29 (67)
19 (70)
11 (65)
.75
HT
35 (81)
28 (82)
13 (93)
.66
Diabetes
9 (21)
7 (24)
2 (13)
.70
Cardiovascular comorbidities
Creatine clearance (mL/min)a
78.4  3.4
93.1  11.4
82.2  9.2
.52
CAD
22 (51)
16 (55)
8 (50)
.77
CVD
3 (7)
2 (7)
1 (7)
1.00
Cardiovascular treatment
CEI
21 (51)
13 (48)
8 (53)
.76
Sartan
10 (24)
8 (30)
2 (13)
.29
Statin
33 (77)
18 (67)
14 (93)
.07
APA
39 (91)
25 (93)
14 (93)
1.00
Dual APAs
5 (12)
2 (7)
3 (20)
.33
VKA
4 (9)
3 (19)
1 (14)
1.00
Phosphocalcic metabolism
Vitamin D (ng/mL)a
12.9  1.2
12.9  1.5
13.9  1.8
.93
Calcium (mmol/L)a
1.2  0
1.2  0
1.2  0
.10
PTH (pg/mL)a
54.5  3.6
53.0  4.5
57.0  5.8
.58
Note. OM() presence of osteoid metaplasia; OM() absence of osteoid metaplasia; BMI body mass index; LDL low density
lipoprotein; HDL high density lipoprotein; HT hypertension; CAD coronary artery disease; CVD cerebrovascular disease;
CEI conversion enzyme inhibitors; APA antiplatelet agent; VKA vitamin K agent; PTH parathyroid hormone.
a
Continuous data are presented as median and SD and discrete data are presented as n (%).

28 (61%) of the plaques, giant multinucleated cells similar

to osteoclasts could be clearly identied in the area
around the OM (Fig. 1C). However, only a few TRAP cells
were detected (Fig. 1D, E). A CT scan was available for 19/
28 of the OM lesions and 12/15 of the OMe lesions.
Analysis of the calcication of these lesions showed that,
in the OM lesions, 11 (58%) were considered highly
calcied, six (32%) moderately calcied, and two (11%) not
or mildly calcied. In the OMe group, the results were
eight (67%), two (17%), and two (17%), respectively
(p .72).
No signicant difference was noted between the groups of
patients with or without OM in terms of demographic
characteristics, cardiovascular risk factors, cardiovascular
comorbidities, treatment, and phosphocalcic metabolism
blood markers (Table 1). No difference between patients
with OM and OMe lesions was observed in any of the
clinical parameters. Among the OM group, 23 patients
were categorized as Rutherford 2 (claudication with quality
of life impairment and failure of optimal medical treatment),
one was Rutherford 3, two were Rutherford 4, and one
patient was classied as Rutherford 5. In the OMe group, 11
patients were Rutherford 2, one was Rutherford 3, and two
were Rutherford 4 (2.3  0.15 [n 27] in the OM group
vs. 2.31  0.2 [n 14] mean Rutherford index; p .81).

Osteoprotegerin and OM in the femoral plaques

The OPG/RANK/RANKL triad supports bone remodeling in
both physiological and pathological situations. An increased
expression of OPG was observed in the femoral plaques
with OM compared with the femoral plaques without OM
(5.59  1.09 [n 17] vs. 2.42  0.58 [n 11] percentage
of area staining [region of interest; ROI], respectively;
p .04) (Fig. 2A).
OPG/RANKL ratio in the plasma and OM in the plaques
RANKL xation on its receptor RANK, induces the differentiation and activation of osteoclasts capable of bone
resorption. OPG is a decoy receptor for RANKL, preventing
its xation on RANK at the surface of osteoclasts. As a
result, the OPG/RANKL ratio nely tunes bone remodeling
in both physiological and pathological conditions. The ELISA
assay on patients plasma did not show any difference in
OPG levels between the OM and OMe groups of patients
(3.68  0.21 [n 25] vs. 3.2  0.29 [n 15], respectively;
p .18). In contrast, RANKL levels were signicantly lower
(0.25  0.3 [n 26] vs. 0.58  0.17 [n 16], respectively;
p .03) and the OPG/RANKL ratio higher (14.6  2.93
[n 24] vs. 6.58  1.85 [n 15], respectively; p .06) in
the patient group with OM in their femoral lesions (Fig. 2B).

Please cite this article in press as: Davaine J-M, et al., Bone Like Arterial Calcication in Femoral Atherosclerotic Lesions: Prevalence and Role of
Osteoprotegerin and Pericytes, European Journal of Vascular and Endovascular Surgery (2015), http://dx.doi.org/10.1016/j.ejvs.2015.10.004

Bone Like Arterial Calcication in Femoral Atherosclerotic Lesions

Figure 2. (A) Representative sections of osteoprotegerin immunohistochemical staining of femoral lesions presenting with osteoid
metaplasia (OM) and without osteoid metaplasia (OMe) (magnication 40). The corresponding quantication of the whole cohort is
reported on the bar graph (right panel). (B) Enzyme linked immunosorbent assays were used to measure plasma levels of osteoprotegerin
(OPG) and receptor activator for the nuclear factor kappa B ligand (RANKL). No difference was noted between OM and OMe OPG plasma
levels (left panel). In contrast, RANKL levels were lower and the OPG/RANKL ratio was higher in the plasma of the OM group than the
OMe group (central and right panel, respectively). Note. ROI region of interest.

Vascular pericytes in femoral plaques

No difference was found in either SMC (SMA cells)
(11.0  1.6 [n 26] in OM and 15.3  1.7 [n 14] in
OMe, percentage area staining [ROI]; p .10 [Fig. 3A]) or
endothelial cell (CD31 cells) (1.5  0.2 [n 28] in OM
and 1.5  0.4 [n 15] in OMe, percentage area staining
[ROI]; p .97) content between OM and OMe lesions
(Fig. 3B). In contrast, NG2 expression (pericytes) was
increased in OM compared with OMe lesions
(0.87  0.22 [n 12] in OM and 0.21  0.06 [n 14] in
OMe, percentage area staining [ROI]; p .05 [Fig. 3D]).
Using CD146, another marker for neovessels and pericytes,
a difference was also noted, with more pericytes in the
lesions with OM; however, this difference was not statistically signicant (7.2  1.5 [n 28] in OM and 3.7  0.9
[n 11] in OMe, percentage area staining [ROI]; p .11
[Fig. 3C]). Finally, no difference was noted in macrophages
(CD68 cells) (2.7  0.3 [n 25] in OM and 2.4  0.5
[n 13] in OMe ROI; p .59 [Fig. 3E]). As expected,
CD31 staining was observed in the intimal layer of the
vessels, as well as in the inner part of neovessels and
adventitial vasa vasorum. SMA staining was noted mainly in
the media. CD68 staining was observed within the
atherosclerotic lesions. Interestingly, CD146 and NG2
cells were noted in the OM and OMe groups in the
adventitial layer. However, in the OM group, the staining
was more intense and the cells were located in the external
part of the media and around of OM rather than in the
adventitia (Fig. 3C, D). Also, this staining pattern was highly

similar to that of OPG staining. Indeed, OPG staining was

discrete in the adventitia in the OMe group, whereas it was
more intense and located in the media and around the OM
in the OM group.
Pericytes, OPG, and the arterial calcication process
Femoral plaques with OM contained signicantly more OPG
and more pericytes than plaques without OM. It was
observed that the pericyte supernatant had a signicant
and dose dependent inhibiting effect on Hu-MSC osteogenic
differentiation. Of note, this inhibitory effect was abolished
when the supernatant was obtained from TNF-a and IL-6
stimulated pericytes (Fig. 4). As it seems likely that SMC
play a major role in plaque calcication, the effect of pericytes on the SMC present in the vessel wall was assessed. It
was observed that the pericyte supernatant signicantly
inhibited the osteogenic differentiation of SMC
(0.13  0.02, SMC in standard medium; 0.21  0.02, SMC
with phosphate buffer; 0.15  0.02, SMC with phosphate
buffer and pericyte supernatant, optical density; p < .01)
(see Supplementary data).
DISCUSSION
Femoral lesions have a high prevalence of OM: new data
and clinical implications
In this study, most (65%) of the femoral lesions presented
with ectopic bone formation, known as OM. OM has all the
characteristics of fully mature bone tissue, with osteoblasts

Please cite this article in press as: Davaine J-M, et al., Bone Like Arterial Calcication in Femoral Atherosclerotic Lesions: Prevalence and Role of
Osteoprotegerin and Pericytes, European Journal of Vascular and Endovascular Surgery (2015), http://dx.doi.org/10.1016/j.ejvs.2015.10.004

J.-M. Davaine et al.

Figure 3. Immunohistochemical experiments were carried out to identify the cell populations present in the femoral lesions with or
without osteoid metaplasia (OM and OMe, respectively). No difference was noted in terms of (A) smooth muscle actin (SM-a) staining;
(B) CD31 staining; and (E) CD68 staining between the two groups. In contrast, pericytes were more abundant in the OM plaques using (C)
CD146 and (D) NG2 markers.

and osteocytes embedded in a dense extracellular matrix

made of collagen, and even fatty and vascular structures
similar to bone marrow. This result clearly indicates that
mechanisms that mimic physiological bone ossication are
also responsible for the development of arterial calcication. To the authors knowledge, this is the rst cohort of
femoral endarterectomies to be studied in depth for the
presence of OM. The femoral bed seems to be more prone
to developing OM than other vascular territories. Indeed, a
prevalence of 12% (9/73) of OM was previously found in a
cohort of carotid lesions, and trabecular bone is contained
in human vessels and in 10e20% of valves.5 The calcication aspect of the lesions seems to be one of the
determining factors underlying the heterogeneity between
vascular beds. Femoral plaques of this cohort have been
previously analyzed for other types of calcication. Sheet
like calcication was found in 86%, nodular calcication in
84%, and clear-center calcication in 41% of the lesions.20
The presence of severe calcication in femoral arteries has
a signicant clinical impact, preventing the safe performance of femoral endarterectomy, increasing the difculty
of performing femoral anastomoses, or resulting in difculties performing femoral artery puncture or femoral artery endovascular treatment. As OM was frequently
observed in this cohort of femoral lesions, this study
focused on this particular type of ectopic bone tissue. OM

consists of a highly mature form of arterial calcication and

has a signicant clinical impact, on surgical procedures
and, physiologically, on hemodynamic parameters. As a
result, this study focused on this particular type of ectopic
bone tissue.
In carotid arteries, it was recently found that OM was
positively associated with asymptomatic, stable plaques.6
The present results suggest that OM is a critical determinant in the composition of femoral plaques, and clinical
studies need to be conducted to assess its exact clinical
impact. No difference in medication use between OM and
OMe patients was observed.
Arterial calcication: a process similar to bone remodeling
While TRAP negative giant multi-nucleated cells were found
in calcied lesions, only a few of them showed TRAP activity, a hallmark of osteoclasts. As a result, these cells may
be nonspecic phagocytic cells deriving from CD14/
macrophage cells present in the lesions, with some of them
displaying TRAP activity and be considered as osteoclasts.
The lack of osteoclasts in OM lesions suggests an imbalance resulting in excess mineral formation. Under physiological conditions, RANKL is needed for osteoclasts to
mature fully. The increase in OPG/RANKL ratio
associated with the presence of OM may partly explain the

Please cite this article in press as: Davaine J-M, et al., Bone Like Arterial Calcication in Femoral Atherosclerotic Lesions: Prevalence and Role of
Osteoprotegerin and Pericytes, European Journal of Vascular and Endovascular Surgery (2015), http://dx.doi.org/10.1016/j.ejvs.2015.10.004

Bone Like Arterial Calcication in Femoral Atherosclerotic Lesions

Vascular pericytes as key regulators of arterial calcication

Figure 4. Human mesenchymal stem cells were differentiated into

osteoblasts using an osteo-inducing medium. After 21 days, Alizarin red staining was carried out (representative pictures above
bars) and quantication was assessed. Note. CT standard medium without osteo-inducing factors; Osteo osteo-inducing
medium; SNP 1/50 pericyte supernatant was added to the
medium in the proportion of 1/50; SNP 1/25 pericyte supernatant was added to the medium in the proportion of 1/25; SNP 1/
10 pericyte supernatant was added to the medium in the proportion of 1/10; IL6 and TNFa pericyte supernatant was added
to the medium in the proportion of 1/10 with either interleukin-6
(IL6) or tumor necrosis factor-a (TNFa) (50 ng/mL), which reversed
the inhibiting effect of the pericyte supernatant.

low number of osteoclasts observed. Furthermore, these

multi-nucleated cells could also promote plaque mineralization by producing osteogenic signals. Similar uncoupling
between osteoblasts and osteoclasts has been already reported, with osteoclasts exhibiting anabolic activity, even if
their resorbing activity was partly or fully inhibited.20
The implication of OPG in the formation of OM and its
potential as a biomarker for arterial calcication
As with carotid plaques, an increase in OPG expression in
femoral lesions with OM (Fig. 2) was observed. The OPG/
RANKL/RANK triad is an active player in the arterial calcication process, yet its exact role as an inhibitor or inducer of
arterial calcication remains controversial.14,21 The present
results suggest that secretion of OPG into the plaque leads
to the formation of mineralized tissue. In addition to its role
in the plaque micro-environment, OPG also has potential as
biomarker for both calcication burden and the severity of
arterial disease.22,23 Indeed, it was recently found that OPG
levels were signicantly higher in the plasma of patients
presenting atherosclerosis with OM compared with OMe
carotid lesions.6 Interestingly, levels of RANKL were signicantly lower and the OPG/RANKL ratio was signicantly
higher in the plasma of patients presenting with OM in their
femoral plaques (Fig. 2). A high OPG/RANKL ratio may be
responsible for the absence of mature osteoclast cells
observed in the lesions.

The exact nature of the cells responsible for the formation

of arterial calcication has not yet been determined.24 SMC
have been shown to have osteochondrogenic potential.25
However, there is also much evidence supporting a role
for pericytes in this process. Pericytes are present along the
arterial network and have been shown both in vitro and
in vivo to be able to differentiate along osteoblastic lineage.26,27 Here, a series of immunohistochemistry experiments were performed to identify the presence of pericytes
and other cell types (endothelial cells, SMC, and macrophages) present in the arterial wall. No quantitative difference was found in SMC, endothelial cell, and
macrophage content between OM and OMe femoral
plaques. In contrast, a different staining pattern of CD146
and NG2 pericytes was noted between OMe and OM
groups. In the OMe group, the staining was mild and
located in the adventitia. In the OM group, the staining
was more intense and noted in the external part of the
media, as well as around areas of OM (Fig. 3C, D). Also, the
OPG staining was very similar to that of pericytes. Indeed, it
was more intense in the media and around the OM in the
OM group (Fig. 2A). This observation is consistent with
previous results showing that pericytes are very prone to
secrete OPG.6 These data show that pericyte inltration
correlates positively with arterial ectopic bone formation.
One could hypothesize that pericytes may be resident
vascular progenitor cells present in the adventitia, having
the ability, under pathological stress such as atherosclerotic
disease, to migrate in the media and intima of the vessels,
and to inuence surrounding cells, as well as to differentiate into other cell types (SMC, osteoblasts). Supporting
this hypothesis, the role of adventitia as a niche of progenitor cells and its role in arterial disease (remodeling,
atherosclerosis) is increasingly recognized.28 The present
in vitro results are consistent with this hypothesis. Pericyte
supernatant inhibited the osteogenic differentiation of HuMSC and of SMC (Fig. 4; see Supplementary data). Also, it
has previously been noted that pericyte supernatant
inhibited the osteoclastic differentiation of CD14 cells, a
model of preosteoclasts.6 When inammatory stress was
introduced, using IL-6 and TNF-a, to mimic atherosclerotic
disease, the results were different. If the inhibitory effect
on osteoclastic differentiation of CD14 cells was maintained, the inhibitory effect on osteogenic differentiation of
Hu-MSC, however, was abolished. These results suggest
that in the vessel wall, as in bone tissue, mineral formation
is a tightly regulated process. In the context of atherosclerotic disease, chronic inammatory stresses induce an
imbalance in this coupling process with a shift towards
more mineral formation, and a blockade of resorption. The
precise role of pericytes in this process needs further
investigation but one could hypothesize that their recruitment is a defense mechanism against atherosclerosis. In
highly evolved lesions, this defense mechanism is nally
overwhelmed and mineral formation prevails leading to the
formation of OM.

Please cite this article in press as: Davaine J-M, et al., Bone Like Arterial Calcication in Femoral Atherosclerotic Lesions: Prevalence and Role of
Osteoprotegerin and Pericytes, European Journal of Vascular and Endovascular Surgery (2015), http://dx.doi.org/10.1016/j.ejvs.2015.10.004

Again, it is likely that the OPG/RANKL/RANK triad is

strongly associated with the presence of OM, as suggested
by several results: (i) OPG staining co-localized with
CD146 and NG2 staining; (ii) the OPG/RANKL ratio was
signicantly higher in the plasma of patients presenting
with OM in their lesions; (iii) it has previously been found
that pericytes abundantly secreted OPG in comparison with
SMC and endothelial cells, in particular under inammatory
conditions.6 Abundant pericyte mediated OPG secretion
and the subsequent increase in the OPG/RANKL ratio may
contribute to the formation of OM by inhibiting the full
differentiation of osteoclast precursors lacking resorptive
activity, as discussed above.
Study limitations
The main limitation of this work is the small number of
lesions studied, limiting the statistical power of the conclusions. OM is likely to have a clinical impact, but there is
no imaging technique capable of specically identifying OM
pre-operatively, thus limiting its clinical relevance. The patients in the current study presented, in the majority, with
claudication and did not perform a treadmill test as routine
pre-operative evaluation. More data on patients suffering
from critical limb ischemia may be highly informative.
Quantication of the calcication in the femoral track was
based on a semi-quantitative analysis and a precise calcium
score analysis based on the available angio-CT scan could
not be performed. OPG and RANKL plasma levels are
promising potential biomarkers for plaque calcication and
stability. However, the results are based on a single blood
sample harvested the day before surgery. Hence, these results must be conrmed in serial blood sample examinations and correlated with the onset of clinical events.
CONCLUSION
OM has a high prevalence in femoral lesions. The present
results further support that arterial calcication participates
in arterial heterogeneity. The observation of few functional
osteoclast cells in OM lesions represents an interesting
lead for future investigation. Conrming the results at the
carotid level, OPG and vascular pericytes are key players in
the formation and regulation of arterial calcication.
ACKNOWLEDGMENTS
The authors thank Carine Montagne and Annie Guillard for
management of the biocollection.
APPENDIX A. SUPPLEMENTARY DATA
Supplementary data related to this article can be found at
http://dx.doi.org/10.1016/j.ejvs.2015.10.004.
CONFLICT OF INTEREST
None.

J.-M. Davaine et al.

FUNDING
This work was funded by an Allocation Nationale de
Recherche (ANR, ANR-11-BSV1-0036) for physiopathology
and by an inter-regional Programme Hospitalier de
Recherche Clinique (PHRC, API11/N/076).

REFERENCES
1 ORourke RA, Brundage BH, Froelicher VF, Greenland P,
Grundy SM, Hachamovitch R, et al. American College of Cardiology/American Heart Association Expert Consensus document on electron-beam computed tomography for the
diagnosis and prognosis of coronary artery disease. Circulation
2000;102:126e40.
2 Allison MA, Hsi S, Wassel CL, Morgan C, Ix JH, Wright CM, et al.
Calcied atherosclerosis in different vascular beds and the risk
of mortality. Arterioscler Thromb Vasc Biol 2012;32:140e6.
3 Hoshino T, Chow LA, Hsu JJ, Perlowski AA, Abedin M, Tobis J,
et al. Mechanical stress analysis of a rigid inclusion in distensible material: a model of atherosclerotic calcication and
plaque vulnerability. Am J Physiol Heart Circ Physiol 2009;297:
H802e10.
4 Maldonado N, Kelly-Arnold A, Vengrenyuk Y, Laudier D,
Fallon JT, Virmani R, et al. A mechanistic analysis of the role of
microcalcications in atherosclerotic plaque stability: potential
implications for plaque rupture. Am J Physiol Heart Circ Physiol
2012;303:H619e28.
5 Hunt JL, Fairman R, Mitchell ME, Carpenter JP, Golden M,
Khalapyan T, et al. Bone formation in carotid plaques: a clinicopathological study. Stroke 2002;33:1214e9.
6 Davaine JM, Quillard T, Brion R, Laperine O, Guyomarch B,
Merlini T, et al. Osteoprotegerin, pericytes and bone like
vascular calcication are associated with carotid plaque stability. PLoS One 2014;9:e107642.
7 Morlacchi S, Pennati G, Petrini L, Dubini G, Migliavacca F. Inuence of plaque calcications on coronary stent fracture: a
numerical fatigue life analysis including cardiac wall movement. J Biomech 2014;47:899e907.
8 Otsuka F, Nakano M, Sakakura K, Ladich E, Kolodgie FD,
Virmani R. Unique demands of the femoral anatomy and pathology and the need for unique interventions. J Cardiovasc
Surg 2013;54:191e210.
9 Sage AP, Tintut Y, Demer LL. Regulatory mechanisms in vascular
calcication. Nat Rev Cardiol 2010;7:528e36.
10 Mody N, Parhami F, Saraan TA, Demer LL. Oxidative stress
modulates osteoblastic differentiation of vascular and bone
cells. Free Radic Biol Med 2001;31:509e19.
11 New SE, Goettsch C, Aikawa M, Marchini JF, Shibasaki M,
Yabusaki K, et al. Macrophage-derived matrix vesicles: an
alternative novel mechanism for microcalcication in atherosclerotic plaques. Circ Res 2013;113:72e7.
12 Collett GD, Caneld AE. Angiogenesis and pericytes in the
initiation of ectopic calcication. Circ Res 2005;96:930e8.
13 Heymann MF, Herisson F, Davaine JM, Charrier C, Battaglia S,
Passuti N, et al. Role of the OPG/RANK/RANKL triad in calcications of the atheromatous plaques: comparison between
carotid and femoral beds. Cytokine 2012;58:300e6.
14 Venuraju SM, Yerramasu A, Corder R, Lahiri A. Osteoprotegerin as a predictor of coronary artery disease and cardiovascular mortality and morbidity. J Am Coll Cardiol 2010;55:
2049e61.

Please cite this article in press as: Davaine J-M, et al., Bone Like Arterial Calcication in Femoral Atherosclerotic Lesions: Prevalence and Role of
Osteoprotegerin and Pericytes, European Journal of Vascular and Endovascular Surgery (2015), http://dx.doi.org/10.1016/j.ejvs.2015.10.004

Bone Like Arterial Calcication in Femoral Atherosclerotic Lesions

15 Norgren L, Hiatt WR, Dormandy JA, Nehler MR, Harris KA,

Fowkes FG, et al. Inter-Society Consensus for the Management
of Peripheral Arterial Disease (TASC II). Eur J Vasc Endovasc
Surg 2007;33(Suppl. 1):S1e75.
16 Herisson F, Heymann MF, Chetiveaux M, Charrier C, Battaglia S,
Pilet P, et al. Carotid and femoral atherosclerotic plaques show
different morphology. Atherosclerosis 2011;216:348e54.
17 Tarte K, Gaillard J, Lataillade JJ, Fouillard L, Becker M,
Mossafa H, et al. Clinical-grade production of human mesenchymal stromal cells: occurrence of aneuploidy without transformation. Blood 2010;115:1549e53.
18 Kaden JJ, Bickelhaupt S, Grobholz R, Haase KK, Sarikoc A, Kilic R,
et al. Receptor activator of nuclear factor kappaB ligand and
osteoprotegerin regulate aortic valve calcication. J Mol Cell
Cardiol 2004;36:57e66.
19 Agharazii M, St-Louis R, Gautier-Bastien A, Ung RV, Mokas S,
Lariviere R, et al. Inammatory cytokines and reactive oxygen
species as mediators of chronic kidney disease-related vascular
calcication. Am J Hypertens 2015;28:746e55.
20 Karsdal MA, Martin TJ, Bollerslev J, Christiansen C, Henriksen K.
Are nonresorbing osteoclasts sources of bone anabolic activity?
J Bone Miner Res 2007;22:487e94.
21 Baudhuin M, Duplomb L, Teletchea S, Lamoureux F, RuizVelasco C, Maillasson M, et al. Osteoprotegerin: multiple
partners for multiple functions. Cytokine Growth Factor Rev
2013;24:401e9.

22 Semb AG, Ueland T, Aukrust P, Wareham NJ, Luben R,

Gullestad L, et al. Osteoprotegerin and soluble receptor activator of nuclear factor-kappaB ligand and risk for coronary
events: a nested case-control approach in the prospective
EPIC-Norfolk population study 1993e2003. Arterioscler Thromb
Vasc Biol 2009;29:975e80.
23 Kiechl S, Schett G, Schwaiger J, Seppi K, Eder P, Egger G, et al.
Soluble receptor activator of nuclear factor-kappa B ligand and
risk for cardiovascular disease. Circulation 2007;116:385e91.
24 Lanzer P, Boehm M, Sorribas V, Thiriet M, Janzen J, Zeller T,
et al. Medial vascular calcication revisited: review and perspectives. Eur Heart J 2014;35:1515e25.
25 Tintut Y, Parhami F, Bostrom K, Jackson SM, Demer LL. cAMP
stimulates osteoblast like differentiation of calcifying vascular
cells. Potential signaling pathway for vascular calcication.
J Biol Chem 1998;273:7547e53.
26 Andreeva ER, Pugach IM, Gordon D, Orekhov AN. Continuous
subendothelial network formed by pericyte-like cells in human
vascular bed. Tissue Cell 1998;30:127e35.
27 Doherty MJ, Ashton BA, Walsh S, Beresford JN, Grant ME,
Caneld AE. Vascular pericytes express osteogenic potential
in vitro and in vivo. J Bone Miner Res 1998;13:828e38.
28 Majesky MW, Dong XR, Hoglund V, Daum G, Mahoney Jr WM.
The adventitia: a progenitor cell niche for the vessel wall. Cells
Tissues Organs 2012;195:73e81.

Please cite this article in press as: Davaine J-M, et al., Bone Like Arterial Calcication in Femoral Atherosclerotic Lesions: Prevalence and Role of
Osteoprotegerin and Pericytes, European Journal of Vascular and Endovascular Surgery (2015), http://dx.doi.org/10.1016/j.ejvs.2015.10.004