Increased proportion of hypermineralized osteocyte lacunae in osteoporotic and osteoarthritic human trabecular bone: Implications for bone remodeling

Bone 50 (2012) 688–694

Contents lists available at SciVerse ScienceDirect

Bone journal homepage: www.elsevier.com/locate/bone

Original Full Length Article

Increased proportion of hypermineralized osteocyte lacunae in osteoporotic and osteoarthritic human trabecular bone: Implications for bone remodeling Vincent T. Carpentier a, b, Jinquan Wong a, Youwen Yeap a, Cheryl Gan a, Peter Sutton-Smith a, c, Arash Badiei a, Nicola L. Fazzalari a, c, Julia S. Kuliwaba a, c,⁎ a b c

Bone and Joint Research Laboratory, Surgical Pathology, SA Pathology and Hanson Institute, Adelaide, Australia Faculty of Medicine, University of Paris Diderot-Paris VII, Paris, France Discipline of Anatomy and Pathology, School of Medical Sciences, The University of Adelaide, Adelaide, Australia

a r t i c l e

i n f o

Article history: Received 16 August 2011 Revised 25 November 2011 Accepted 26 November 2011 Available online 7 December 2011 Edited by: David Burr Keywords: Hypermineralized osteocyte lacunae Trabecular bone Osteoporosis Osteoarthritis Quantitative backscattered electron imaging Energy dispersive X-ray spectrometry

a b s t r a c t Hypermineralized osteocyte lacunae (micropetrosis) have received little research attention. While they are a known aspect of the aging human skeleton, no data are available for pathological bone. In this study, intertrochanteric trabecular bone cores were obtained from patients at surgery for osteoporotic (OP) femoral neck fracture (10F, 4M, 65–94 years), for hip osteoarthritis (OA; 7F, 8M, 62–87 years), and femora at autopsy (CTL; 5F, 11M, 60–84 years). Vertebral trabecular bone cores were also obtained from the vertebra of autopsy cases (CVB; 3F, 6M, 53–83 years). Specimens were resin-embedded, polished, and carbon coated for quantitative backscattered electron imaging (qBEI), energy dispersive X-ray (EDX) spectrometry, and imaging analysis. Bone mineralization (Wt %Ca) was not different between OP, OA, and CTL; but was greater in femoral CTL than in CVB. The percent of hypermineralized osteocyte lacunae relative to the total number (HL/TL) was greater in OP and OA than in CTL. However, relative to bone mineral area, OP was characterised by increased hypermineralized osteocyte lacunar number density (Hd.Lc.Dn), whereas OA was characterised by decreased osteocyte lacunar number density (Lc.Dn) and total osteocyte lacunar number density (Tt.Lc.Dn). Lc.Dn was higher in CVB than in femoral CTL. The calcium–phosphorus ratio (RCa/P) was not different between hypermineralized osteocyte lacunae and bone matrix in each group. In addition, this study focused on the phenomenon of osteocyte lacunae hypermineralization using qBEI. Seven morphological types of osteocyte lacunae hypermineralization were described according to the presence of one or several hypermineralized spherites, associated or not with a hypermineralized lacunar ring. This study has described, for the first time, the morphology of hypermineralized osteocyte lacunae in OP and OA human bone. Further studies are suggested to investigate the functional influence of hypermineralized osteocyte lacunae on bone remodeling and bone biomechanical properties. © 2011 Elsevier Inc. All rights reserved.

Introduction Osteocytes are the most numerous cells in the bone [1]. Osteocytes, which arise from osteoblasts after cell differentiation, are cells trapped in the mineral bone matrix [2]. They are connected to each other as well as with osteoblasts by dendritic cell processes forming an important connected cellular network (CCN). The CCN is separated from the mineral compartment, called the lacuno-canalicular network (LCN), by an extracellular matrix (ECM) and by an extracellular

⁎ Corresponding author at: Directorate of Surgical Pathology, SA Pathology and Hanson Institute, Frome Road, Adelaide 5000, Australia. Fax: +61 8 8222 3293. E-mail addresses: [email protected] (V.T. Carpentier), [email protected] (J. Wong), [email protected] (Y. Yeap), [email protected] (C. Gan), [email protected] (P. Sutton-Smith), [email protected] (A. Badiei), [email protected] (N.L. Fazzalari), [email protected] (J.S. Kuliwaba). 8756-3282/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.bone.2011.11.021

fluid (ECF). The ECF translates endogenous and exogenous mechanobiological, biochemical, and electromechanical signals [3–5]. Consequently, the integrity of these two networks, by their extensive surface area and their extensive molecular exchanges, appears to be fundamental to bone homeostasis and the quality of bone [3,6]. Despite the recent dramatic increased research interest in osteocytes and the LCN, the mechanisms of their multiple functions in maintaining a healthy skeleton are still unclear [7]. Osteocytes, in association with the ECF, play a key role in the mechanosensitivity of the human skeleton [5,8]. Mechanical unloading and loading of bone modify the regulation of osteocyte genes and molecule expression [9–12]. Their mechanosensitive behaviour initiates bone remodeling, which can also be initiated by osteocyte death and the absence of sclerostin [13,14]. Osteocytes are also strongly implicated in the regulation of phosphate metabolism and operate as an endocrine system targeting distant organs through molecules such as the dentin matrix acidic phosphoprotein 1 (Dmp1), the fibroblast growth factor 23

V.T. Carpentier et al. / Bone 50 (2012) 688–694

(FGF23), the matrix extracellular phosphoglycoprotein (MEPE), and the phosphate regulating endopeptidase homolog X-linked (Phex) [15,16]. Nevertheless, one of their characteristics, the phenomenon of hypermineralization of osteocyte lacunae (also called micropetrosis [17]) has received little research attention. It was initially described as osteocyte lacunae totally filled by mineral substance. Therefore, it was assumed that the resident osteocyte was not viable when associated with hypermineralized osteocyte lacunae [17,18]. Strikingly, a review of the literature reveals inconsistent data. In fact, we do not know whether the phenomenon of hypermineralization of osteocyte lacunae is an active cellular phenomenon, whether it occurs after or during osteocyte death. Although some studies reported the increase of hypermineralized osteocyte lacunae with age [17,19], no studies investigated the possible modifications of this phenomenon under pathological conditions such as osteoporosis and osteoarthritis. Furthermore, even though some studies have described certain types of intermediate hypermineralized osteocyte lacunae between unmineralized osteocyte lacunae and hypermineralized osteocyte lacunae [20,21], they did not investigate in detail the range of different types of intermediate hypermineralized osteocyte lacunae involved in the phenomenon of hypermineralization of osteocyte lacunae. This study investigates hypermineralized osteocyte lacunae using quantitative backscattered electron imaging (qBEI) and energy dispersive X-ray (EDX) spectrometry with three objectives: 1) identify any difference in pathological versus control human trabecular bone in terms of their number density and mineral content; 2) identify any difference in human femoral versus vertebral trabecular bone in terms of their number density and mineral content; 3) describe the different types of intermediate hypermineralized osteocyte lacunae involved in the phenomenon of hypermineralization of osteocyte lacunae. Materials and methods Human trabecular bone specimens This study investigated fifty four Caucasian human subjects who were biopsied at surgery or autopsy, in the intertrochanteric region of the proximal femur (IT), and at autopsy only, the central region of the second lumbar corpus vertebrae (CVB) was biopsied. Fourteen subjects (ten females, aged 74 to 91 years; and four males, aged 65 to 94 years; mean age of 83 years), sampled from the IT region during hip hemi-arthroplasty for a primary and minimaltrauma subcapital femoral fracture, were defined as osteoporotic bone (OP). Fifteen subjects (seven females, aged 66 to 87 years; and eight males, aged 62 to 85 years; mean age of 75 years), sampled from the IT region during a hip arthroplasty for primary osteoarthritis, were defined as osteoarthritic bone (OA). At surgery, all cases were at advanced stages of OA, Collins grade III or IV [22]. Sixteen subjects (five females, aged 68 to 81 years; and eleven males, aged 60 to 84 years; mean age of 74 years), sampled from the IT region at autopsy, were defined as control bone (CTL). Nine subjects (three females, aged 56 to 87 years; and six males, aged 53 to 83 years; mean age of 71 years), sampled from the CVB at autopsy, were defined as control vertebral bone. All of these post-mortem cases were known to have had no bone-related chronic disease and were admitted to hospital less than three days before death. All of the subjects from the four groups had no history of any medication that may have affected osteocyte death. Trabecular bone core biopsies were removed from all subjects with a 10 mm diameter tube saw in order to obtain approximately 1 cm 3 bone specimens. Specimens were processed and embedded in methyl-methacrylate, blocks faced to expose the bone, polished, and then carbon coated for qBEI, EDX spectrometry, and image analysis. All specimens were collected after obtaining an informed consent,

689

with approval from the Royal Adelaide Hospital Research Ethics Committee. Quantitative backscattered electron imaging A Philips XL20 tungsten filament scanning electron microscope (SEM) and a Philips XL30 field emission scanning electron microscope (FESEM) installed with a two-sector backscattered electron detector were used for qBEI. The XL20 SEM was used to measure the bone tissue mineralization at 200×, the osteocyte lacunar area, and the osteocyte lacunar number densities. It was also used to observe the phenomenon of osteocyte lacunae hypermineralization at a magnification of 400×. The accelerating voltage was set at 15 kV and the working distance at 15 mm. The microscope was calibrated using a carbon standard (spectrographically pure carbon, Bal-Tec, Liechtenstein) for the grey level 25, and an aluminium standard (Specpure aluminium, Johnson Matthey Chemicals, London, UK) for the grey level 225. For each subject, 10 random images at 200× magnification and containing at least 25% bone in the field of view were collected. After acquisition of two images, the two standards were re-imaged to track beam current changes during the period of image collection. By fitting a 5th order polynomial equation to the standard grey level changes over time, the bone images were corrected for minor beam current variation. The linear relationship between average atomic number of carbon, aluminium, and hydroxyapatite was controlled by imaging a block containing known hydroxyapatite (IG-Pore, ApaTech Ltd, UK) under calibrated conditions. An equation to convert the grey levels to the mean value of the weight percent of calcium (Wt %Ca) was used for each subject with a proprietary routine written in Matlab® software (The Mathworks Inc. Natick, MA, USA) [23,24]. Bone specimens analysed by qBEI reveal black or bright white osteocyte lacunae (Fig. 1). Unmineralized black osteocyte lacunae are related to lacunae with and without an osteocyte cell being present. In contrast, bright white lacunae are related to hypermineralized osteocyte lacunae [20]. For each subject, the mean value of the unmineralized black osteocyte lacunar area (Lc.Ar, in μm²), the mean value of the unmineralized black osteocyte lacunar number density (Lc.N/B.Ar = Lc.Dn, in #/mm²), the mean value of the hypermineralized osteocyte number lacunar density (Hd.Lc.N/B.Ar = Hd.Lc.Dn, in #/mm²), the mean value of the total osteocyte lacunar number density (Tt.Lc.N/B.Ar = Tt.LcDn, in #/mm²), and the mean value of the percentage of hypermineralized osteocyte lacunae (Hd.Lc.N/ Tt.Lc.N*100 = HL/TL, in %) were calculated with a semi-automated proprietary routine written in Matlab with the ten images collected for the Wt %Ca determination [25].

Fig. 1. A control 83 years old female subject, 400× magnification qBEI image. We can observe a hypermineralized osteocyte lacuna (A), an intermediate hypermineralized osteocyte lacuna (B), unmineralized black osteocyte lacuna (C), and hypermineralized microcrack (D).

690

V.T. Carpentier et al. / Bone 50 (2012) 688–694

The examination of bone images acquired by qBEI reveals different types of intermediate hypermineralized osteocyte lacunae between unmineralized black osteocyte lacunae and hypermineralized osteocyte lacunae (Fig. 1). Eleven randomly selected subjects from the CTL group were used. For each of these subjects, 10 random images at 400× magnification and containing at least 25% bone in the field of view were collected with the XL20 SEM. The different types of intermediate hypermineralized osteocyte lacunae were defined under a magnification of ×400 according to the relative proportion of mineral content within the lacunae (low mineral content: lacuna filled with less than 49% mineral; high mineral content: lacuna filled with more than 50% mineral), the number of hypermineralized spherites, the presence or absence of contact between hypermineralized spherites and the osteocyte lacunar wall, and the presence or absence of a hypermineralized lacunar ring (Table 1). Some of these intermediate hypermineralized osteocyte lacunae were also observed with the XL30 FESEM at a magnification of 3500× in order to underline the description realised at 400×. The XL20 SEM was used as described above; whereas for the XL30 FESEM, the settings are described below.

Whitney U-test was used. The Pearson or the Spearman tests were also used to evaluate the correlation of each analysed parameter with age within each group. The RCa/P ratio values of each analysis were used for comparison between hypermineralized osteocyte lacunae and bone matrix in each group. Identically as for paired group comparisons, if the distribution of the analysed ratio was parametric for the hypermineralized osteocyte lacunae and bone matrix, the Student's t-test or the Welsh test was used depending on the result of the Fisher's F-test for the equality of two variances. If the distribution of the analysed data was non-parametric for one of two localisations, the MannWhitney U-test was used. The statistical analyses were performed using SAS/STAT® software (SAS Institute Inc., Cary, NC, USA). The critical p-value of significance was set at 0.05 and data were expressed as mean (standard deviation) or as median [quartiles]. Results Pathological versus control bone

Energy dispersive X-ray spectrometry The XL30 FESEM, fitted with an energy dispersive X-ray microprobe (E.D.S system, EDAX), was used to investigate the mineral content of hypermineralized osteocyte lacunae in comparison to the bone matrix in each group [19]. The accelerating voltage was set at 10 kV, and the working distance at 10 mm. In each group, three representative subjects were selected randomly. For each specimen, three spicules of bone were selected, and for each spicule, three pairs of analyses were performed within different remodeling packets. Each pair of analyses included an analysis of one hypermineralized osteocyte lacunae and an analysis of the bone matrix within the same remodeling packet. The analyses of the bone matrix were 15 μm distant from their respective osteocyte lacunar analysis. The microprobe was calibrated with a take-off angle of 38° and an acquisition time of 100 s. The relative weight of calcium (rWt %Ca) and phosphorus (rWt %P) was quantified with the correction of atomic number, absorption, and fluorescence (ZAF) effects for each analysis [26,27]. Then, the value of the calcium-phosphorus ratio (rWt %Ca/ rWt %P = RCa/P) was determined. Statistical analysis Group comparisons were realised using the mean value of Wt %Ca, Lc.Ar, Lc.Dn, Hd.Lc.Dn, Tt.Lc.Dn, and HL/TL of each subject. The Shapiro–Wilk test was used to test for the normality of the data distribution. If the distribution of the analysed data was parametric for the OP, OA, and CTL groups, the ANOVA test and the Tukey's post-hoc test were used for comparison. Whereas, if the distribution of the analysed data was non-parametric for at least one of the three groups, data were transformed using log10 before undertaking the analysis outlined above. Similarly, if the distribution of the analysed data was parametric for the CTL and CVB groups, the Student's t-test or the Welsh test was used depending on the result of the Fisher's F-test for the equality of two variances. If the distribution of the analysed data was non-parametric for one of two groups, the Mann–

The OP and OA groups had a lower Wt %Ca than the CTL group, but they were not significantly different (Fig. 2). The Lc.Ar was similar between the OP, OA, and CTL groups (Fig. 2). Subsequently, comparison tests relating to osteocyte lacunae number densities between the OP, OA, and CTL groups were undertaken. The percent of hypermineralized osteocyte lacunae relative to the total number (HL/TL) was greater in the OP and OA groups than in the CTL group (Table 2). However, relative to bone mineral area, the OP group was characterised by increased hypermineralized osteocyte lacunar number density (Hd.Lc.Dn) (Table 2), whereas the OA group was characterised by decreased osteocyte lacunar number density (Lc.Dn) and total osteocyte lacunar number density (Tt.Lc.Dn) (Fig. 2). Wt %Ca, Lc.Ar, Lc.Dn, Hd.Lc.Dn, TL.N/B.Ar, and HL/TL were not correlated with age in the OP, OA and CTL groups. The RCa/P ratio was not different between hypermineralized osteocyte lacunae and the bone matrix in the OP, OA and CTL groups (Table 3). Vertebral versus femoral bone Wt %Ca was greater in the CTL group than in the CVB group (Fig. 3). The Lc.Ar was similar between the CTL and CVB groups (Fig. 3). Subsequently, comparison tests relating to osteocyte lacunae number densities between the CTL and CVB groups were undertaken. Lc.Dn was higher in the CVB group than in the femoral CTL group (Fig. 3). No difference was found between the CVB and CTL groups concerning Hd.Lc.Dn and HL/TL (Table 4). Wt %Ca, Lc.Ar, Lc.Dn, Hd.Lc.Dn, Tt.Lc.Dn, and HL/TL were not correlated with age in the CTL and CVB groups. The RCa/P ratio was not different between hypermineralized osteocyte lacunae and the bone matrix in the CTL and CVB groups (Table 3). Phenomenon of hypermineralization of osteocyte lacunae Specimen observations, in the four study groups, by qBEI at a magnification of 400× reveal different stages of intermediate hypermineralized

Table 1 Characterisation of the different types of osteocyte lacunae. Four criteria were used: the number of hypermineralized spherites, the presence or absence of contact between hypermineralized spherites and the osteocyte lacunar wall, the presence or absence of a hypermineralized lacunar ring, and the relative proportion of mineral content within the lacuna. N/A: not applicable.

Number of spherites Spherites contact Hypermineralized lacunar ring Relative proportion of mineral content

Type I

Type II

Type III

Type IV

Type V

Type VI

Type VII

Absent N/A Absent 0%

One Free in the lacuna Absent 1% to 49%

Several Free in the lacuna Absent 1% to 49%

Several Wall-connected Absent 1% to 49%

Several Wall-connected Present 50% to 99%

Absent N/A Present 50% to 99%

Absent N/A Absent 100%

V.T. Carpentier et al. / Bone 50 (2012) 688–694

691

Fig. 2. Box plots depicting OP, OA, and CTL group comparisons of: (A) weight percent of calcium (Wt %Ca), no significant difference; (B) unmineralized black osteocyte lacunar area (Lc.Ar, in μm²), no significant difference; (C) unmineralized black osteocyte lacunar density (Lc.Dn, in #/mm²), OA less than CTL; and (D) total osteocyte lacunar density (Tt.Lc.Dn, in #/mm²), OA less than CTL.(C) *Statistical significance between the OA and CTL groups at α = 0.01 (p b 0.01).(D) *Statistical significance between the OA and CTL groups at α = 0.02 (p b 0.02).

osteocyte lacunae between unmineralized black osteocyte lacunae and hypermineralized osteocyte lacunae. For characterisation of the different morphological types of intermediate hypermineralized osteocyte lacunae, fifty seven intermediate hypermineralized osteocyte lacunae were identified in the one hundred and ten images for the eleven randomly selected CTL cases, which correspond to approximately 1% of the total number of osteocyte lacunae. These lacunar observations seem to describe a sequential phenomenon which leads to a hypermineralized osteocyte lacuna. This phenomenon of osteocyte lacunae hypermineralization can be divided into different types of mineralization characterised by specific morphological features of osteocyte lacunae (Fig. 4). Seven different types were defined according to the relative proportion of mineral content within the lacunae (low mineral content: lacuna filled with less than 49% mineral; high mineral content: lacuna filled with more than 50% mineral), the number of hypermineralized spherites, the presence or absence of contact between hypermineralized spherites and the osteocyte lacunar wall, and the presence or absence of a hypermineralized lacunar ring (Table 1). The unmineralized black Table 2 Comparisons of hypermineralized osteocyte lacunae density transformed using log10 (log10(Hd.Lc.Dn)), and percentage of hypermineralized osteocyte lacunae transformed using log10 (log10(HL/TL)) between the OP, OA and CTL groups. Data are expressed as mean (standard deviation). OP Log10 (Hd.Lc.Dn) Log10 (HL/TL) a b c

OA a

1.48 (0.20) 1.03 (0.23)b

Between OP and CTL: p b 0.02 (α = 0.02). Between OP and CTL: p b 0.01 (α = 0.02). Between OA and CTL: p b 0.01 (α = 0.02).

1.40 (0.31) 1.02 (0.34)c

osteocyte lacunae (type I osteocyte lacunae) are osteocyte lacunae which appear entirely black under qBEI. A black osteocyte lacunae within which a single hypermineralized spherite is detectable corresponds to type II osteocyte lacunae. The average value of spherite radii is 1 μm. Type III osteocyte lacunae are characterised by the presence of several hypermineralized spherites within the black osteocyte lacunar area. Osteocyte lacunae with wall-connected hypermineralized spherites are type IV osteocyte lacunae. Types II to IV osteocyte lacunae are intermediate hypermineralized osteocyte lacunae with a low mineral content. Osteocyte lacunae with a hypermineralized lacunar ring associated with hypermineralized spherites but only partially filled are type V osteocyte lacunae. The hypermineralized lacunar ring appears to originate from the type IV osteocyte lacunae as a result of the fusion of hypermineralized spherites on the osteocyte lacunar wall. The type VI osteocyte lacunae are osteocyte lacunae partially filled by hypermineralized substance and without the presence of hypermineralized spherites. Their filling seems to correspond to the thickening of the hypermineralized lacunar ring. Types V and VI osteocyte lacunae are intermediate hypermineralized osteocyte lacunae with a high mineral content. Finally, type VII Table 3 Comparisons of the RCa/P ratio between the bone matrix and the hypermineralized osteocyte lacunae in the OP, OA, CTL and CVB groups. Data are expressed as mean (standard deviation). NS: No significant difference.

CTL a

1.06 (0.58) 0.60 (0.53)b,c

OP OA CTL CVB

Bone matrix

Hypermineralized osteocyte lacunae

Percent difference

p value

2.01 1.96 1.65 1.78

2.01 1.99 1.66 1.74

+ 0.10% + 1.53% + 0.42% − 2.14%

NS NS NS NS

(0.12) (0.12) (0.18) (0.27)

(0.08) (0.12) (0.22) (0.29)

692

V.T. Carpentier et al. / Bone 50 (2012) 688–694

Fig. 3. Box plots depicting CTL and CVB group comparisons of: (A) weight percent of calcium (Wt %Ca), CTL greater than CVB; (B) unmineralized black osteocyte lacunar area (Lc.Ar, in μm²), no significant difference; (C) unmineralized black osteocyte lacunar density (Lc.Dn, in #/mm²), CTL less than CVB; and (D) total osteocyte lacunar density (Tt.Lc.Dn, in #/mm²), no significant difference.(A) *Statistical significance between the CTL and CVB groups (p b 0.01).(C) *Statistical significance between the CTL and CVB groups (p b 0.04).

osteocyte lacunae are osteocyte lacunae totally filled by hypermineralized substance (Fig. 4, Table 1). An identical morphological sequence of the phenomenon can be observed at a magnification of 3500× (Fig. 4). Moreover, images of the type VII osteocyte lacunae support that the phenomenon of osteocyte lacunae hypermineralization leads to an osteocyte lacuna completely filled by hypermineralized substance.

Discussion The large number of images collected for the fifty four subjects within this study enabled us to document the qualitative morphology and number density of hypermineralized osteocyte lacunae. As hypermineralized osteocyte lacunae are not visible with standard histologic techniques, they have received little attention from researchers. Moreover, the majority of studies investigating osteocyte lacunae have not included the analysis of hypermineralized osteocyte lacunae.

Table 4 Comparisons of hypermineralized osteocyte lacunae density (Hd.Lc.Dn), and percentage of hypermineralized osteocyte lacunae (HL/TL) between the CTL and CVB groups. Data are expressed as median [quartiles]. NS: No significant difference.

Hd.Lc.Dn HL/TL

CTL

CVB

p value

11.70 [5.11; 34.86] 4.05 [1.38; 14.22]

8.68 [2.28; 14.80] 2.30 [0.64; 3.76]

NS NS

This study presents comparisons between pathological and control bone regarding the histomorphometric and mineralized parameters of the trabecular bone of the intertrochanteric region of the proximal femur. We did not find any significant difference between the OP, OA and CTL groups concerning bone tissue mineralization (Wt %Ca). However, previous studies of human bone tissue from OP subjects have reported a decrease in Wt %Ca [24,28,29]. In OA, observations in the hard tissue include increased bone hypomineralization [30]. The main interest of this study concerns the hypermineralized osteocyte lacunae. Hypermineralized osteocyte lacunae are a common element of healthy and pathological human trabecular bone [17,31,32]. Even though their presence is not specific to pathological bone, their increased number density is related to pathological conditions. The significant increase of HL/TL in OP and OA highlights a new important aspect of pathological bone. Interestingly, our results, in regard to hypermineralized osteocyte lacunae, reiterate the fact that OP, OA, and the control bone are different to each other. The OP bone is characterised by an increased Hd.Lc.Dn relative to the CTL group; whereas the OA bone is characterised by a decreased Lc.Dn and Tt.Lc.Dn relative to the CTL group. Unlike empty osteocyte lacunae, hypermineralized osteocyte lacunae do not permit ECF circulation within the LCN. We do not yet know definitively whether their presence is detrimental to the quality of bone. The impact of hypermineralized osteocyte lacunae on the quality of the CCN has to be investigated in terms of its mechanosensory role in bone adaptation, its endocrine activity and its ability to detect and repair microdamage. One study has revealed that the presence of hypermineralized osteocyte lacunae does not appear to influence the microhardness of bone [18]. However, further mechanical analyses of bone to investigate the influence of hypermineralized osteocyte lacunae on fracture toughness are recommended. We reported also no difference between the OP and CTL groups relating to Lc.Dn and Tt.Lc.Dn. The osteocyte lacunae number density distribution within OP bone tissue is unclear and a review of the literature indicates that the estimated variation of osteocyte lacunae number density is not consistent [33–37]. In contrast, the decreased Lc.Dn and Tt.Lc.Dn in the OA group is consistent with a previous study [33]. This decrease suggests an alteration of the osteoblast to osteocyte differentiation in OA and the need for further studies of the significance of this observation. Analysis of the mineral content of hypermineralized osteocyte lacunae, compared to the mineral content of the bone matrix, does not reveal any difference of the RCa/P within the OP, OA and CTL groups. These data suggest that the modification of the phenomenon of hypermineralization of osteocyte lacunae in pathological conditions is not RCa/P dependent but dependent on the mechanism of osteocyte lacuna hypermineralization that influences the numerical density of hypermineralized lacunae. Our data support a previous study reporting an increase of the RCa/P in OP [38]. However, a study with larger groups is necessary to confirm the increase of RCa/P in OA reported by our data and a previous study [39]. In addition, this study presents a comparison between vertebral bone and femoral bone. Differences between sites were related to a decreased Wt %Ca and increased Lc.Dn in the CVB group. These modifications are in association with a high remodeling rate within the vertebral bone [40–42]. Differently to Lc.Dn, Hd.Lc.Dn is similar between the two studied sites. These data suggest that hypermineralized osteocyte lacunae seem to be more uniformly shared out between the different skeletal sites than unmineralized osteocyte lacunae. No age correlation was found for Wt %Ca, Lc.Ar, Lc.Dn, Hd.Lc.Dn, Tt.Lc.Dn, and HL/TL in the four study groups. Previous studies have reported a trend of age related decrease in Lc.Dn [19,34,43] and increase in Hd.Lc.Dn in aged human bone [17,19]. As our subjects were fifty six years old and over, we did not find these correlations in the four groups. Hypermineralized osteocyte lacunae are absent

V.T. Carpentier et al. / Bone 50 (2012) 688–694

693

Fig. 4. Images from CTL subjects which describe the phenomenon of osteocyte hypermineralization using qBEI. The last six types of osteocyte lacunae are illustrated: a type II osteocyte lacuna at a magnification of 400× (A), a type III osteocyte lacuna at a magnification of 400× (B), a type IV osteocyte lacuna at a magnification of 400× (C), a type V osteocyte lacuna at a magnification of 400× (D), a type VI osteocyte lacuna at a magnification of 400× (E), a type VII osteocyte lacuna at a magnification of 400× (F) a type III osteocyte lacuna at a magnification of 3500× (G), a type V osteocyte lacuna at a magnification of 3500× (H), and a type VII osteocyte lacuna at a magnification of 3500× (I).

within the bone of new-born, and their number density appears to increase with advancing age [17]. This study presents different osteocyte lacunar density parameters (Lc.Dn, Hd.Lc.Dn, Tt.Lc.Dn and HL/TL) for human trabecular bone. But, one of its limitations is the absence of data related to bone remodeling activity. This study does not investigate the relative importance of the local bone remodeling activity on the different hypermineralized osteocyte lacunae parameters (Hd.Lc.Dn and HL/TL). Even though the Wt %Ca can provide information on the overall tissue remodeling rate [41,42], it does not impart specific information on the remodeling rate of the studied bone spicule or remodeling packet. A few published papers have initiated a research interest in the nature of the phenomenon of the osteocyte lacunae hypermineralization [19,44–46]. Our study suggests that hypermineralized osteocyte lacunae could be the result of a specific phenomenon. Seven types of hypermineralized osteocyte lacunae were defined according to the presence of one or more hypermineralized spherites, associated or not with a hypermineralized lacunar ring. However, our study is limited by the low number density of intermediate hypermineralized osteocyte lacunae. Moreover, the use of two-dimensional static imaging does not provide enough information to extrapolate to a possible three-dimensional dynamic phenomenon. Even though this study illustrates the potential different stages of the phenomenon of hypermineralization of osteocyte lacunae, it does not have sufficient accuracy to confirm the association between the different types of observed intermediate hypermineralized osteocyte lacunae, and between the hypermineralized spherites and the hypermineralized lacunar ring. We do not know the mechanism of osteocyte lacuna hypermineralization. Earlier studies hypothesised that hypermineralized

osteocyte lacunae may result from the inhibition of the osteocytic remodeling of the perilacunar matrix, or the accumulation of apoptotic bodies [19,45,46]. Furthermore, we do not know whether the phenomenon of osteocyte hypermineralization is an active cellular phenomenon, whether it occurs after or during osteocyte death. One study has reported that the colony stimulating factor-1 (CSF-1) appears to modulate the quantity of hypermineralized osteocyte lacunae [47]. Another study has reported that intermediate hypermineralized osteocyte lacunae are associated with the appearance of an apoptotic cell containing a condensed chromatin-like organisation under transmission electron imaging [44]. We report the following: 1) that the osteocyte hypermineralization could be a multi-staged phenomenon; 2) that hypermineralized spherites look like the “mineralized spheres” generated by osteoid-osteocytes in culture [48,49]. These findings remind us that a direct link between the phenomenon of hypermineralization of osteocyte lacunae and osteocyte activity or osteocyte death is possible. Being present in the CTL and CVB groups, hypermineralized osteocyte lacunae are a generalised feature of bone physiology. But, their possible role is unclear given the impact of their relative increasing number in pathology is unknown. In fact, the potential functional alteration of the CCN and the LCN by an increase of hypermineralized occlusions within the LCN may be detrimental to the maintenance of functionally adequate bone quality. The alteration of these two networks will likely have a strong impact on the circulation of the ECF in the LCN, resulting in a decreased capacity of the ECF to translate endogenous and exogenous molecular signals. This may result in a decreased bone capacity to sense loading, to detect microdamage, to control bone remodeling, and to act as an endocrine system.

694

V.T. Carpentier et al. / Bone 50 (2012) 688–694

Conclusions Finally, these data reveal a previously unrecognised, potentially disruptive feature of osteocyte functionality in bone physiology and pathology. We have reported increased hypermineralized osteocyte lacunae percent in OP and OA within human trabecular bone tissue. This increase of HL/TL is associated with increased Hd.Lc.Dn in OP and a decreased Lc.Dn and Tt.Lc.Dn in OA. We have also described seven types of osteocyte lacunae involved in the phenomenon of osteocyte hypermineralization depending on the presence of hypermineralized spherites and a hypermineralized lacuna ring. Further studies are suggested to investigate the functional influence of hypermineralized osteocyte lacunae on bone remodeling and bone biomechanical properties. Funding source This study was supported by the National Health and Medical Research Council (NHMRC) of Australia. Acknowledgments We kindly thank the Orthopaedic Surgeons and Nursing Staff of The Department of Orthopaedics and Trauma, Royal Adelaide Hospital, and the Mortuary Staff of Surgical Pathology, SA Pathology, for support and cooperation in the collection of bone tissue specimens. References [1] Marotti G. The structure of bone tissues and the cellular control of their deposition. Ital J Anat Embryol 1996;101:25–79. [2] Dallas SL, Bonewald LF. Dynamics of the transition from osteoblast to osteocyte. Ann N Y Acad Sci 2010;1192:437–43. [3] Schneider P, Meier M, Wepf R, Muller R. Towards quantitative 3D imaging of the osteocyte lacuno-canalicular network. Bone 2010;47:848–58. [4] Weinbaum S, Cowin SC, Zeng Y. A model for the excitation of osteocytes by mechanical loading-induced bone fluid shear stresses. J Biomech 1994;27:339–60. [5] Wang L, Wang Y, Han Y, Henderson SC, Majeska RJ, Weinbaum S, et al. In situ measurement of solute transport in the bone lacunar–canalicular system. Proc Natl Acad Sci U S A 2005;102:11911–6. [6] Burger EH, Klein-Nulend J. Mechanotransduction in bone—role of the lacunocanalicular network. FASEB J 1999;13:S101–12 (Suppl.). [7] Bonewald LF. The amazing osteocyte. J Bone Miner Res 2011;26:229–38. [8] Price C, Zhou X, Li W, Wang L. Real-time measurement of solute transport within the lacunar–canalicular system of mechanically loaded bone: direct evidence for load-induced fluid flow. J Bone Miner Res 2011;26:277–85. [9] Skerry TM, Bitensky L, Chayen J, Lanyon LE. Early strain-related changes in enzyme activity in osteocytes following bone loading in vivo. J Bone Miner Res 1989;4:783–8. [10] Gluhak-Heinrich J, Ye L, Bonewald LF, Feng JQ, MacDougall M, Harris SE, et al. Mechanical loading stimulates dentin matrix protein 1 (DMP1) expression in osteocytes in vivo. J Bone Miner Res 2003;18:807–17. [11] Robling AG, Niziolek PJ, Baldridge LA, Condon KW, Allen MR, Alam I, et al. Mechanical stimulation of bone in vivo reduces osteocyte expression of Sost/sclerostin. J Biol Chem 2008;283:5866–75. [12] Bonewald LF, Johnson ML. Osteocytes, mechanosensing and Wnt signaling. Bone 2008;42:606–15. [13] Lin C, Jiang X, Dai Z, Guo X, Weng T, Wang J, et al. Sclerostin mediates bone response to mechanical unloading through antagonizing Wnt/beta-catenin signaling. J Bone Miner Res 2009;24:1651–61. [14] Seeman E. Osteocytes—martyrs for integrity of bone strength. Osteoporos Int 2006;17:1443–8. [15] Bonewald LF. Osteocytes as dynamic multifunctional cells. Ann N Y Acad Sci 2007;1116:281–90. [16] Feng JQ, Ward LM, Liu S, Lu Y, Xie Y, Yuan B, et al. Loss of DMP1 causes rickets and osteomalacia and identifies a role for osteocytes in mineral metabolism. Nat Genet 2006;38:1310–5. [17] Frost HM. Micropetrosis. J Bone Joint Surg Am 1960;42-A:144–50. [18] Remaggi F, Ferretti M, Cane V, Zaffe D. Histomorphological and chemico-physical analyses of the mineral matrix of micropetrotic human bone. Ann Anat 1996;178:223–7. [19] Busse B, Djonic D, Milovanovic P, Hahn M, Puschel K, Ritchie RO, et al. Decrease in the osteocyte lacunar density accompanied by hypermineralized lacunar occlusion reveals failure and delay of remodeling in aged human bone. Aging Cell 2010;9:1065–75.

[20] Kingsmill VJ, Boyde A. Mineralisation density of human mandibular bone: quantitative backscattered electron image analysis. J Anat 1998;192(Pt 2):245–56. [21] Boyde A. The real response of bone to exercise. J Anat 2003;203:173–89. [22] Collins D. The pathology of articular and spinal diseases. London: Edward Arnold and Co.; 1949. [23] Roschger P, Fratzl P, Eschberger J, Klaushofer K. Validation of quantitative backscattered electron imaging for the measurement of mineral density distribution in human bone biopsies. Bone 1998;23:319–26. [24] Sutton-Smith P, Beard H, Fazzalari N. Quantitative backscattered electron imaging of bone in proximal femur fragility fracture and medical illness. J Microsc 2008;229:60–6. [25] Qiu S, Phipps RJ, Ebetino FH, Palnitkar S, Sudhaker Rao D. Effect of risedronate on osteocyte viability and bone turnover in paired iliac bone biopsies from early postmenopausal women. Calcif Tissue Int 2010;87:392–7. [26] Roschger P, Plenk Jr H, Klaushofer K, Eschberger J. A new scanning electron microscopy approach to the quantification of bone mineral distribution: backscattered electron image grey-levels correlated to calcium K alpha-line intensities. Scanning Microsc 1995;9:75–86 discussion 86–8. [27] Bloebaum RD, Holmes JL, Skedros JG. Mineral content changes in bone associated with damage induced by the electron beam. Scanning 2005;27:240–8. [28] Roschger P, Rinnerthaler S, Yates J, Rodan GA, Fratzl P, Klaushofer K. Alendronate increases degree and uniformity of mineralization in cancellous bone and decreases the porosity in cortical bone of osteoporotic women. Bone 2001;29: 185–91. [29] Loveridge N, Power J, Reeve J, Boyde A. Bone mineralization density and femoral neck fragility. Bone 2004;35:929–41. [30] Li B, Aspden RM. Composition and mechanical properties of cancellous bone from the femoral head of patients with osteoporosis or osteoarthritis. J Bone Miner Res 1997;12:641–51. [31] Allen MR, Burr DB. The pathogenesis of bisphosphonate-related osteonecrosis of the jaw: so many hypotheses, so few data. J Oral Maxillofac Surg 2009;67:61–70. [32] Jones SJ, Glorieux FH, Travers R, Boyde A. The microscopic structure of bone in normal children and patients with osteogenesis imperfecta: a survey using backscattered electron imaging. Calcif Tissue Int 1999;64:8–17. [33] Jordan GR, Loveridge N, Power J, Clarke MT, Parker M, Reeve J. The ratio of osteocytic incorporation to bone matrix formation in femoral neck cancellous bone: an enhanced osteoblast work rate in the vicinity of hip osteoarthritis. Calcif Tissue Int 2003;72:190–6. [34] Mullender MG, van der Meer DD, Huiskes R, Lips P. Osteocyte density changes in aging and osteoporosis. Bone 1996;18:109–13. [35] Mullender MG, Tan SD, Vico L, Alexandre C, Klein-Nulend J. Differences in osteocyte density and bone histomorphometry between men and women and between healthy and osteoporotic subjects. Calcif Tissue Int 2005;77:291–6. [36] Power J, Noble BS, Loveridge N, Bell KL, Rushton N, Reeve J. Osteocyte lacunar occupancy in the femoral neck cortex: an association with cortical remodeling in hip fracture cases and controls. Calcif Tissue Int 2001;69:13–9. [37] Qiu S, Rao DS, Palnitkar S, Parfitt AM. Reduced iliac cancellous osteocyte density in patients with osteoporotic vertebral fracture. J Bone Miner Res 2003;18:1657–63. [38] Zoehrer R, Perilli E, Kuliwaba JS, Shapter JG, Fazzalari NL, Voelcker NH. Human bone material characterization: integrated imaging surface investigation of male fragility fractures. Osteoporos Int 2011, doi:10.1007/s00198-011-1688-9. [39] Akesson K, Grynpas MD, Hancock RG, Odselius R, Obrant KJ. Energy-dispersive Xray microanalysis of the bone mineral content in human trabecular bone: a comparison with ICPES and neutron activation analysis. Calcif Tissue Int 1994;55:236–9. [40] Diab T, Allen MR, Burr DB. Alendronate treatment results in similar levels of trabecular bone remodeling in the femoral neck and vertebra. Osteoporos Int 2009;20:647–52. [41] Boivin G, Farlay D, Bala Y, Doublier A, Meunier PJ, Delmas PD. Influence of remodeling on the mineralization of bone tissue. Osteoporos Int 2009;20:1023–6. [42] Ruffoni D, Fratzl P, Roschger P, Klaushofer K, Weinkamer R. The bone mineralization density distribution as a fingerprint of the mineralization process. Bone 2007;40:1308–19. [43] Qiu S, Rao DS, Palnitkar S, Parfitt AM. Age and distance from the surface but not menopause reduce osteocyte density in human cancellous bone. Bone 2002;31: 313–8. [44] Bell LS, Kayser M, Jones C. The mineralized osteocyte: a living fossil. Am J Phys Anthropol 2008;137:449–56. [45] Kogianni G, Mann V, Noble BS. Apoptotic bodies convey activity capable of initiating osteoclastogenesis and localized bone destruction. J Bone Miner Res 2008;23: 915–27. [46] Noble BS. The osteocyte lineage. Arch Biochem Biophys 2008;473:106–11. [47] Harris SE, Macdougall M, Horn D, Woodruff K, Zimmer SN, Rebel VI, et al. Meox2Cre-mediated disruption of CSF-1 leads to osteopetrosis and osteocyte defects. Bone 2012;50:42–53. [48] Barragan-Adjemian C, Nicolella D, Dusevich V, Dallas MR, Eick JD, Bonewald LF. Mechanism by which MLO-A5 late osteoblasts/early osteocytes mineralize in culture: similarities with mineralization of lamellar bone. Calcif Tissue Int 2006;79: 340–53. [49] Kato Y, Boskey A, Spevak L, Dallas M, Hori M, Bonewald LF. Establishment of an osteoid preosteocyte-like cell MLO-A5 that spontaneously mineralizes in culture. J Bone Miner Res 2001;16:1622–33.