Aluminum and Iron Can Be Deposited in the Calcified Matrix of Bone

JIB-09809; No of Pages 6 Journal of Inorganic Biochemistry xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Journal of Inorganic Biochemistry

journal homepage: www.elsevier.com/locate/jinorgbio

Aluminum and iron can be deposited in the calciﬁed matrix of bone exostoses

Daniel Chappard a,b,⁎, Guillaume Mabilleau a,b, Didier Moukoko c, Nicolas Henric c, Vincent Steiger d, Patrick Le Nay d, Jean-Marie Frin c, Charlotte De Bodman c a GEROM Groupe Etudes Remodelage Osseux et bioMatériaux — LHEA, IRIS-IBS Institut de Biologie en Santé, CHU d'Angers, LUNAM Université,49933, ANGERS Cedex, France b SCIAM, Service Commun d'Imagerie et Analyses Microscopiques, IRIS-IBS Institut de Biologie en Santé, CHU d'Angers, LUNAM Université, 49933 Angers, Cedex, France c Chirurgie Pédiatrique Orthopédique, CHU d'Angers, 49933, Angers, Cedex, France d Département de Chirurgie Osseuse, CHU d'Angers, 49933, Angers, Cedex, France. article info abstract

Article history: Exostosis (or osteochondroma) is the most common benign bone tumor encountered in children and adults. Ex- Received 14 April 2015 ostoses may occur as solitary or multiple tumors (in the autosomal syndromes of hereditary multiple exostoses). Received in revised form 2 September 2015 Exostoses are composed of cortical and medullary bone covered by an overlying hyaline cartilage cap. We have Accepted 14 September 2015 searched iron (Fe) and aluminum (Al) in the matrix of cortical and trabecular bone of 30 patients with exostosis. Available online xxxx Al3+ and Fe3+ are two cations which can substitute calcium in the hydroxyapatite crystals of the bone matrix. The bone samples were removed surgically and were studied undecalcified. Perls' Prussian blue staining (for Keywords: 3+ Aluminum Fe) and solochrome azurine B (for Al) were used on the histological sections of the tumors. Al was detected 3+ Bone matrix histochemically in 21/30 patients as linear bands deposited by the osteoblasts. Fe was detected in 10 out of Exostosis these 21 patients as linear bands in the same locations. Fe3+ and Al3+ were not identified in the bone matrix Hydroxyapatite of a control group of 20 osteoporotic patients. Energy X-ray Dispersive Spectrometry failed to identify Fe and Mineralization Al in bone of these tumors due to the low sensitivity of the method. Wavelength Dispersive Spectrometry iden- Iron tified them but the concentrations were very low. Histochemistry appears a very sensitive method for Fe3+ and Al3+ in bone.The presence of these two metals in the exostoses advocates for a disturbed metabolism of osteoblasts which can deposit these metals into the bone matrix, similar to which is observed in case of hemochromatosis with Fe3+. © 2015 Elsevier Inc. All rights reserved.

1. Introduction umbrella term that describes a set of characteristics that influences bone strength and explain the interrelationships of these characteris- Exostosis (or osteochondroma) is the most frequent benign bone tics” [4]. The different determinants of bone strength have been tumor encountered in children or adults. Exostosis may occur as solitary reviewed elsewhere together with the various methods available to an- or multiple tumors in the case of an autosomal genetic disorder (the alyze them [5]. Histochemistry is a powerful tool to characterize the Multiple Hereditary Exostoses — MHE). MHE affects 1/50,000 people mineralization of the bone matrix and to identify the presence of certain and is caused by a mutation in the Golgi-associated heparin-sulfate metal ions abnormally present in it where they can alter bone quality. polymerases EXT1 or EXT2 [1]. However, the pathophysiology of isolat- The aim of the present study was to analyze the calcified bone ma- ed exostosis is unknown but it is likely that a dysregulation of trix in a series of human exostoses in search of two metal ions (alumi- osteoprogenitor cells (chondrocytes and osteoblasts) leads to these num and iron) known to interfere with calcification [6,7] and to alter bony proliferations [2,3]. Exostoses are cartilage-capped tumors which osteoblast functions [8,9]. Histochemistry and X-ray-based spectroscop- can be sessile or pedunculated. The cartilage covers a shell of cortical ic methods coupled with Scanning Electron Microscopy (SEM) were bone on which a network of trabecular bone is internally appended. used. The center of the exostosis is in continuity with the medullary canal of the bone and contains bone marrow with more or less hematopoietic 2. Patients and methods cells [3]. The last decade has witnessed a considerable interest in the quality 2.1. Participants and histological analysis of the bone matrix in metabolic bone diseases. Bone quality is “an Between 2010 and 2015, 30 patients were operated on the orthope- dic or pediatric surgery department for one or more exostoses. Tumors ⁎ Corresponding author at: GEROM — LHEA, IRIS-IBS Institut de Biologie en Santé, LUNAM Université Nantes Angers Le Mans, CHU d'Angers, 49933 ANGERS Cedex, France. were sent to the bone histopathology unit where they were processed E-mail address: [email protected] (D. Chappard). undecalcified after embedding in poly (methylmethacrylate). Sections

Please cite this article as: D. Chappard, et al., Aluminum and iron can be deposited in the calcified matrix of bone exostoses, J. Inorg. Biochem. (2015), http://dx.doi.org/10.1016/j.jinorgbio.2015.09.008 2 D. Chappard et al. / Journal of Inorganic Biochemistry xxx (2015) xxx–xxx were cut dry on a heavy-duty microtome equipped with tungsten car- patient with six analyzed exostoses and two patients with two ex- bide knives. They were stained by Goldner's trichrome (for identifica- ostoses removed during the same surgical intervention (Table 1). tion of osteoid and calcified matrix) [10]. Histochemical identification The total number of exostoses was 38 in this series of 30 patients. of osteoclasts (bone resorbing cells) was done by the tartrate resistant The tumors were either sessile or pedunculated and the X-ray as- acid phosphatase method. The Perls' and solochrome azurine B stainings pect of these exostoses is illustrated in Fig. 1.ThemicroCTaspect were used for the identification of Fe3+ and Al3+, respectively [11,12]. is depicted in Fig. 2 and in the video appearing as supplementary These histochemical reactions were done in cleaned glass vials and the material. technicians never used metallic forceps during the staining to avoid con- The histopathological aspect was consistent with classical de- tamination. Some samples were analyzed by microcomputed tomogra- scriptions of a cartilage covering cap, more or less developed which phy to have a 3D evaluation of these tumors. Morphometric analysis surmounts the bony part of the tumor (Fig. 3). Cartilaginous columns was done but only the osteoid seam thickness will be considered here mimicking a growth plate were observed but the number of trabec- as a parameter characterizing a mineralization defect (O.Th in μm, normal ula was usually reduced, giving broad trabeculae with a central b15 μm). core made of calcified cartilage. Foci of calcified cartilage were sometimes observed in the discontinuous cortical shell made of Haversian 2.2. Control cases systems and lamellar bone. The trabeculae were most often composed of lamellar bone excepted in the youngest children and in In our bone laboratory, the histochemical staining of iron and alumi- two cases of sub-ungueal exostoses. Trabecular and cortical bones num is done on all bone samples since a decade. We chosed the 20 most were composed of calcified bone matrix and the closest areas to the recent bone biopsies refered for osteoporosis as a control group. cartilage had a high bone remodeling level. Numerous foci of osteoclasts were observed in these areas. Numerous osteoid seams were 2.3. Metal atom characterization by spectroscopy observed, but O.Th was never increased (excepted when non- lamellar woven bone was present). Poly (methylmethacrylate) blocks containing the bone samples The mean age of the control patients with osteoporosis was (mean were polished with 0.5 μm diamond particles, carbon-coated and ob- age 54.5 ± 16.8 y; range 32–73 y.). There were 12 females and 8 served by SEM (EVO LS10, Carl Zeiss Ltd., Nanterre, France) equipped males. The final diagnosis was: glucocorticoid induced osteoporosis with an Energy Dispersive X-ray spectrometer (EDS – INCA, Oxford, (N = 5), mastocytosis (N = 3); post-menopausal osteoporosis (N = UK). The EDS analysis was performed with the Inca system fitted with 4) and idiopathic male osteoporosis (N = 8). aX-max20mm2 silicon drift detector (Oxford Instruments, High Wycombe, UK). Prior to quantitative analysis, samples were polished 3.2. Histochemical analysis with diamond particles (1-μm thickness) to reach a surface roughness less than 10 nm. The system was calibrated with pure cobalt (Micro- The most striking fact was the presence of aluminum and iron in Analysis consultants Ltd., St. Ives, UK) and quantitative analysis was per- the calcified bone matrix or in the calcified cartilaginous matrix. formed with an accelerating voltage of 20 keV, a measured probe cur- rent of 500 pA and a working distance of 8.5 mm. During EDX analysis, the specimen is bombarded with a focalized electron beam inside the SEM. The electrons collide with the atoms' own electrons, Table 1 ejecting some of them out of their orbit. A position vacated by an ejected Clinical description of cases. MHE: patients with a Multiple Hereditary Exostoses disease. For the semi-quantitative score: − is for an absence of staining, +for the presence of a lim- electron in an inner shell is then replaced by a higher-energy electron ited number of stained bands; ++ for numerous stained bands; +++ very high amount from an outer orbit. To do so, this transferring outer electron liberates of metal bands in the bone matrix. some of its energy by emitting an X-ray. The atom of every element re- Case Gender Age Localization Aluminum Iron leases X-rays with unique amounts of energy during the transferring process. In our system, a minimum of 200,000 coups were recorded 1 f 22 Right distal tibia +++ +++ −− and the local atomic concentration was calculated with the semiQuant 2 m 54 Right scapula 3 f 51 Right femur ++ + algorithm using the XPP matrix correction. The minimum detectable 4 f 31 Left tibia −− limit of this setup is of 0.05%–0.1%; (atomic %), that means that an ele- 5 m 45 Right lesser trochanter −− ment with a concentration below 1/2000 atoms will not be detected 6 f 27 Right 2nd metatarsal (MHE) −− [13]. 7 m 8 Left humerus + − 8 m 22 Right distal tibia + − The blocks containing the highest amount of aluminum and iron 9 f 14 Right distal tibia −− were also examined on a Wavelength Dispersive X-ray spectrometer 10 m 31 Right peritrochanter ++ + (WDS, Inca Ware 500, Oxford Instrument, UK) installed on a Merlin 11 m 14 Left humerus +++ +++ SEM (Carl Zeiss Ltd) equipped with a field emission gun. Analyses 12 m 3 Right scapula −− −− were also performed at 20 kV. During WDS analysis, the specimen is 13 m 6 Sub-ungueal 3rd metatarsal bone 14 m 12 Left tibia +++ +++ also bombarded with a focalized electron beam inside the SEM to iden- 15 m 11 Tibia (2 exostoses) +++ +++ tify the elemental constituents comprising the sample. This results in X- 16 m 13 Right humerus +++ − ray emission in the same way from the bombarded atoms. The wave- 17 f 15 Left scapula +++ +++ length of the X-rays diffracted into the detector are selected by varying 18 m 16 Right + left scapulae +++ +++ 19 m 80 Right tibia ++ − the position of an analyzing crystal. Unlike EDS, WDS reads or counts 20 m 17 6 exostoses (MHE) + − only the X-rays of a single wavelength at time and does not produce a 21 f 19 Left femur +++ + spectrum of wavelengths simultaneously. 22 f 16 Right femur+ left tibia (MHE) +++ + 23 m 16 Right toe + − − 3. Results 24 m 8 Left humerus ++ 25 m 14 Right humerus + − 26 m 40 Left ulna ++ − 3.1. Histopathology 27 f 15 Left humerus −− 28 f 16 Right femur right tibia + − The mean age of the patients with exostosis was 23.5 ± 18.4 yr. 29 m 7 Left iliac bone + − 30 f 63 1st right metatarsal bone −− (extremes 3 and 80 yr.). There were three patients with MHE; one

Fig. 2. MicroCT images of a sessile exostosis. A) The external view is composed of a cortical bone shell; the upper part corresponds to the calciﬁed cartilaginous part of the exostosis. B) The internal view shows the presence of a trabecular network.

3.3. Spectroscopic analysis

SEM–EDS failed to identify Al and Fe in all patients analyzed. SEM– WDS could identify the presence of Al and Fe and the concentration ranged between 0.032–0.054% (atomic %) for Fe and 0.030–0.034% for aluminum (Fig. 5).

Fig. 1. A) X-ray of a pedunculated exostosis of the left proximal humerus; B) X-ray of symetric bilateral exostoses in a child with Multiple Hereditary Exostoses — MHE.

Al3+ was detected in the form of deep blue bands and Perls' staining identified Fe3+ as light blue bands (Fig. 4). Iron was never observed in the cartilaginous matrix. Al3+ was detected in 21 patients and iron was found in exactly the same areas in 10 of these patients. Al3+ and Fe3+ were present in a various number of bands but it was likely that the number of Al3+ bands was always superior to Fe3+ bands. Bands were usually observed in the trabeculae and sometimes in the cortical shell. The bands corresponded to the cement and arrest lines, which indicate a pause or a recovery in osteoblast activity when they elaborate bone structure units (BSU). In Fig. 3. Histological analysis. A) Transverse section of a pedunculated exostosis showing the MHE patients, Al3+ and Fe3+ were similar in the different observed external cartilage cap (c) covered by the perichondrium (pc); the fenestrated cortical shell (cs). In the center, the trabecular network with the bone marrow can be evidenced. The exostoses. In the control group of osteoporotic patients, histochem- arrowheads indicate the layer of calcified cartilage reproducing a growth plate. 3+ 3+ ical analysis failed to identify Al and Fe in the trabecular and B) Histochemical identification of osteoclasts at the junction calcified cartilage/trabecular cortical bone. bone. Osteoclasts are brown cells, calcified bone is in blue, cartilage is unstained.

Fig. 4. Histochemical identiﬁcation of aluminum (upper raw A to C) detected by the solochrome azurine B method and iron (lower raw D to F) identiﬁed by Perls' Prussian blue. A and D correspond to a patient with a low Al3+ content and no Fe3+; B and E are from a patient with a mid Al3+ and Fe3+; C and F are from a patient with a high Al3+ and Fe3+ content. The bars indicate 100 μm.

4. Discussion bone at the implantation base of the tumor never contained bands of metal in their constituting BSU. In one case Al3+ deposits were observed The pathophysiology of MHE is related to mutations in the EXT1 or in the calcified cartilage. From a crystallographic point of view, Al or Fe EXT2 heparan sulfate polymerases which control the development can substitute a Ca atom in the large channel of the hydroxyapatite crys- and normal functions of the perichondrium [14,15]. However, in solitary tal in position (6 h) [21,22]. In this channel centered by a hydroxide exostoses, the mechanisms are less understood but abnormalities in the group, the six calcium atoms are surrounded by the phosphate groups 3− osteochondral tissues of bones have been recognized very early [16].Ex- and when Al is adsorbed, it is linked to the PO4 groups [23]. ternal irradiation can disrupt the architecture of the perichondral ring of Perls' staining is a worldwide admitted histochemical stain for iron. Ranvier (an area encircling the growth plate) and provokes an exostosis The method works on soft and hard tissues. In the presence of ferrocy- [17,18]. Osteomyelitis and trauma have also been implicated [19,20].In anide ions, Fe3+ is precipitated as a highly water-insoluble blue com- the present study, the development of exostosis fulfills the classical plex made of potassium ferric ferrocyanide, also coined Prussian blue. criteria in the children. In two adult cases, an exostosis was developed Fe2+ is not detected by the method. In bone, it is preferable to avoid after femoral amputation or knee surgery. the use a counterstain dye if one wants to clearly identify the metal The dysfunction of the osteochondral cells leads to an accumulation bands in the bone matrix [24–27]. Al and Fe are deposited in the cement of aluminum and iron in about 2/3 of cases. The amount of metal pres- or arrest lines when a BSU is formed, either in cortical and/or trabecular ent in the bone matrix can only be semi-quantitatively observed and the bone. These lines are known to contain specific proteins (osteopontin score used here is the reflect of the amount of bands presents in the ex- and osteocalcin) that can bind metal ions. It has been shown by a very ostosis. Al3+ and Fe3+ bands were observed in cortical bone and mainly sensitive method (micro X-ray fluorescence analysis with a synchro- in the trabecular envelope of the exostosis. Al3+ can be present alone in tron) that Zn and Pb ions present in the interstitial fluid can accumulate 11 cases and Fe3+ bands are observed in the same locations. Metal ions in the cement lines by uptake in the hydroxyapatite crystal and attach- are mainly observed in the trabeculae close to the calcified cartilage and ment to these proteins [28]. Our histochemical staining methods for

Fig. 5. WDS analysis of the trabecular bone matrix of an exostosis rich en Fe and Al. The method is quantitative (after normalization of the data) but do not show the location of these metals. A) an area imaged and analyzed; B) identiﬁcation of Fe and Al peaks during measurement.

Please cite this article as: D. Chappard, et al., Aluminum and iron can be deposited in the calcified matrix of bone exostoses, J. Inorg. Biochem. (2015), http://dx.doi.org/10.1016/j.jinorgbio.2015.09.008 D. Chappard et al. / Journal of Inorganic Biochemistry xxx (2015) xxx–xxx 5 these two metals also evidenced that Al3+ and Fe3+ form bands corre- co-localization of Fe3+ and Al3+ has also been reported in the brain sponding to these specificarea. of Alzheimer's disease [62,63]. Histochemical identification of aluminum in bone was extensively In bone, low doses of aluminum have been identified as a cause of studied in the 1970′ when it was found that patients with renal failure bone loss in laboratory animals [8] and humans [64]. High doses causes developed encephalopathy and osteomalacia [29,30]. The role of alumi- osteomalacia. Aluminum reduces osteoblastic activity and alters bone num in the pathogenesis of both diseases was evidenced and histo- quality by interfering with calcification and altering the bone levels of chemical identification of Al3+ in the bone matrix was proposed on important other trace elements such as Zn, Cu, Mn Se, B, Sr… [9]. undecalcified bone sections [31]. Aluminum was most often stained by the aluminon technique using aurine tricarboxylic stain [24,32].The 5. Conclusion method was found more sensitive than atomic absorption spectrometry [33]. Atomic absorption spectrometry was not available in the present Al and Fe can bind to the phosphate groups of hydroxyapatite crys- study; however, a previous report from our group has shown that Al tals when an atom of Ca2+ is replaced by a Fe3+ or an Al3+ atom. Histo- (and other metals) had were present at very low concentrations in the chemistry using Perls' and solochrome azurine B methods (for resp. bone matrix and did not increased significantly during aging [34]. Sev- Fe3+ and Al3+) are very sensitive techniques and they evidenced a eral authors have described that solochrome azurine B (also termed modification of bone quality in this series of exostoses mostly due to Mordant blue B or chrome azurol B - CI 43830) [35] gives better results these epigenetic factors. The localization of these two cations occurs in and identifies a larger number of bands in the bone matrix [11,36–38]. the same locations (cement and arrest lines of the bone matrix) in The method was compared with atomic absorption spectrometry and about 2/3 of cases for Al3+ and 1/3 for Fe3+. The mechanism of the de- was found well correlated. However, this physical method is destructive position of these two ions is not fully understood but a dysregulation of 3+ for the samples although it provides a global estimation of the Al con- osteogenic cells (chondrocytes and osteoblasts) in these benign tumors 3+ centration in μg/g of bone. In the present study, we tried to identify Fe seems to be the cause. and Al3+ by spectroscopic methods [39,40]. Previous studies have – found SEM EDS suitable to identify Fe and Al only in in vitro models Acknowledgments of calcification in presence of these metal ions [6,7]. The method was also successfully used with other metals (Cr, Co, Ni, Sr) studied in the Authors thank Mr. Yann Borjon-Piron and Mr. Frédéric Christien, same conditions [41,42]. It should be noted that, in in vitro experiments, Polytech-nantes, Institut des Matériaux de Nantes site Chantrerie for the concentration in the hydroxyapatite crystals is higher than 0.05% kindly providing WDS analyses on our samples. This work was made (atomic %) that is in agreement with the detection limit of the method possible by grants from the French Ministry of Research, in the Bioregate litterature on semi-quantitative analysis by EDS [13,43,44].Inthe program. They also thank Mrs. Nadine Gaborit for her skillful assistance present series of exostoses, the method failed to identify Al and Fe in with microCT and histotechnology and Mrs. Laurence Lechat for secre- – fi our samples. On the other hand, SEM WDS con rmed the presence of tarial assistance. these two atoms and showed that the local concentrations were very low in the patients. WDS is known to be a very sensitive method to Appendix A. Supplementary data identify traces of metals [40]. Histochemistry is confirmed as a very sensitive method for Al3+ and Fe3+ in the bone matrix allowing their Supplementary data to this article can be found online at http://dx. precise localization in these tumors. The histochemical reaction was doi.org/10.1016/j.jinorgbio.2015.09.008. negative in all patients of the control group. The challenge to face is to understand why these two metals can be References deposited in the bone matrix by osteoblasts. Fe3+ is histochemically identified in bone in cases of iron-related diseases such as hemochro- [1] F.M. Vanhoenacker, W. Van Hul, W. Wuyts, P.J. Willems, A.M. De Schepper, Heredi- matosis [45,46], β-thalassemia or drepanocytosis [47–49]. The mecha- tary multiple exostoses: from genetics to clinical syndrome and complications, Eur. – nisms are not clearly understood and the role of proteins such as J. Radiol. 40 (2001) 208 217. [2] N.M. Billah, M. Idrissi, R.I. Kaitouni, H. Faraj, M. El Yaacoubi, S. Bouklata, Imagerie des ferroportin and other transporters have been recently reviewed [50]. exostoses solitaires, Feuill. Radiol. 53 (2013) 11–20. Clearly, it is possible that these proteins can also serve as transporters [3] J.W. Milgram, The origins of osteochondromas and enchondromas ahistopathologic – for other metallic cations. Co-localization of Al3+ and Fe3+ in bone study, Clin. Orthop. Relat. Res. 174 (1983) 264 284. [4] D. Felsenberg, S. Boonen, The bone quality framework: determinants of bone has been reported in a very limited number of studies [37,51]. strength and their interrelationships, and implications for osteoporosis manage- The most intriguing problem is to ascertain the origin of aluminum. ment, Clin. Ther. 27 (2005) 1–11. In the last decades, aluminum has become omnipresent in modern life, [5] D. Chappard, M.F. Baslé, E. Legrand, M. Audran, New laboratory tools in the assess- ment of bone quality, Osteoporos. Int. 22 (2011) 2225–2240. alimentation and environment [52]. Aluminum is present in foods as [6] C.N. Degeratu, G. Mabilleau, C. Cincu, D. Chappard, Aluminum inhibits the growth of the additive E 173 [53] and has been detected in baby milks [54] and hydroxyapatite crystals developed on a biomimetic methacrylic polymer, J. Trace tea [55]. In addition, aluminum, which is one of the most abundant me- Elem. Med. Biol. 27 (2013) 346–351. [7] P. Guggenbuhl, R. Filmon, G. Mabilleau, M.F. Baslé, D. Chappard, Iron inhibits hy- tallic elements of the Earth's crust, is also used to make household prod- droxyapatite crystal growth in vitro, Metab. Clin. Exp. 57 (2008) 903–910. ucts (dishes, trays, pans, cooking foils, brewage cans…). The major [8] X. Li, Y. Han, Y. Guan, L. Zhang, C. Bai, Y. Li, Aluminum induces osteoblast apoptosis route of exposure to aluminum is through food although a trans- or through the oxidative stress-mediated JNK signaling pathway, Biol. Trace Elem. Res. fi 150 (2012) 502–508. per cutaneous route has also been identi ed (antisweating products, [9] X. Li, C. Hu, Y. Zhu, H. Sun, Y. Li, Z. Zhang, Effects of aluminum exposure on bone vaccines). It is estimated that the aluminum intake can reach up to mineral density, mineral, and trace elements in rats, Biol. Trace Elem. Res. 143 13 mg/day [56,57] although the tolerable weekly intake is only 1 (2011). mg/kgofbodyweight[58]. 95% injested aluminum is eliminated by [10] D. Chappard, in: D. Heymann (Ed.), Bone Cancer: Progression And Therapeutic Ap- proaches, Acad. Press; Elsevier Inc., London 2014, pp. 111–120. the feces whilst the 5% remaining circulate in the blood, bound to [11] J. Denton, A.J. Freemont, J. Ball, Detection and distribution of aluminium in bone, J. transferrin, albumin and other low molecular weight proteins. The Clin. Pathol. 37 (1984) 136–142. circulating aluminum can mainly localize in two preferential organs: [12] M. Kaye, A.B. Hodsman, L. Malynowsky, Staining of bone for aluminum: use of acid solochrome azurine, Kidney Int. 37 (1990) 1142–1147. brain (linked to the phosphoproteins, lipids or phosphate groups of [13] INCA, Energy Operator Manual, Oxford Instruments Analytical, High Wycombe, UK, DNA) [59] and bone (bound to the phosphate groups of hydroxyap- January 2006. atite at the mineralization front) [6,31]. Aluminum localization in [14] J. Huegel, C. Mundy, F. Sgariglia, P. Nygren, P.C. Billings, Y. Yamaguchi, E. Koyama, M. Pacifici, Perichondrium phenotype and border function are regulated by Ext1 and the brain is neurotoxic and is implicated in neurodegenerative dis- heparan sulfate in developing long bones: a mechanism likely deranged in heredi- eases such as Alzheimer, Parkinson and multiple sclerosis [60,61].A tary multiple exostoses, Dev. Biol. 377 (2013) 100–112.

[15] B.M. Zak, M. Schuksz, E. Koyama, C. Mundy, D.E. Wells, Y. Yamaguchi, M. Pacifici, J.D. [41] J. Beuvelot, R. Filmon, Y. Mauras, M.F. Baslé, D. Chappard, Adsorption and release of Esko, Compound heterozygous loss of Ext1 and Ext2 is sufficient for formation of strontium from hydroxyapatite crystals developed in simulated body fluid (SBF) on multiple exostoses in mouse ribs and long bones, Bone 48 (2011) 979–987. poly (2-hydroxyethyl) methacrylate substrates, Dig. J. Nanomater. Biostruct. 8 [16] R. D'Ambrosia, A.B. Ferguson Jr., The formation of osteochondroma by epiphyseal (2013) 207–217. cartilage transplantation, Clin. Orthop. Relat. Res. 61 (1968) 103–115. [42] G. Mabilleau, R. Filmon, P.K. Petrov, M.F. Baslé, A. Sabokbar, D. Chappard, Cobalt, [17] R.T. Ballock, R.J. O'Keefe, Physiology and pathophysiology of the growth plate, Birth chromium and nickel affect hydroxyapatite crystal growth in vitro, Acta Biomater. Defects Res. C Embryo Today 69 (2003) 123–143. 6 (2010) 1555–1560. [18] F.D. Murphy, W.P. Blount, Cartilaginous exostoses following irradiation, J. Bone Joint [43] K.F. Heinrich, Electron Beam X-ray Microanalysis, Van Nostrand Reinhold Co., 1981 Surg. Am. 44 (1962) 662–668. [44] J. Goldstein, D. Newbury, P. Echlin, D. Joy, C. Fiori, E. Lifshin, Scanning electron mi- [19] A. Vallcanera, A. Moreno-Flores, J. Gomez, H. Cortina, Osteochondroma post osteo- croscopy and X-ray microanalysis. A text for biologists, materials scientists, and ge- myelitis, Pediatr. Radiol. 26 (1996) 680–681. ologists. Scanning electron microscopy and X-ray microanalysis, A Text For [20] P. Bhusari, K.I. Kumathalli, S. Raja, A.K. Mishra, Bony exostosis following a traumatic Biologists, Materials Scientists, And Geologists1981. blow: a case report, J. Indian Dent. Assoc. 5 (2011) 285. [45] K. Eyres, E. McCloskey, E. Fern, S. Rogers, M. Beneton, J. Aaron, J. Kanis, Osteoporotic [21] S.V. Dorozhkin, A review on the dissolution models of calcium apatites, Prog. Crystal fractures: an unusual presentation of haemochromatosis, Bone 13 (1992) 431–433. Growth Charact. Mater. 44 (2002) 45–61. [46] M. Duquenne, V. Rohmer, E. Legrand, D. Chappard, N. Wion Barbot, M.F. Baslé, M. [22] M. Corno, A. Rimola, V. Bolis, P. Ugliengo, Hydroxyapatite as a key biomaterial: Audran, J.C. Bigorgne, Spontaneous multiple vertebral fractures revealed primary quantum-mechanical simulation of its surfaces in interaction with biomolecules, haemochromatosis, Osteoporos. Int. 6 (1996) 338–340. Phys. Chem. Chem. Phys. 12 (2010) 6309–6329. [47] P. Mahachoklertwattana, V. Sirikulchayanonta, A. Chuansumrit, P. Karnsombat, L. [23] T. Kiss, P. Zatta, B. Corain, Interaction of aluminium(III) with phosphate-binding Choubtum, A. Sriphrapradang, S. Domrongkitchaiporn, R. Sirisriro, R. Rajatanavin, sites: biological aspects and implications, Coord. Chem. Rev. 149 (1996) 329–346. Bone histomorphometry in children and adolescents with beta-thalassemia disease: [24] A.G.E. Pearse Histochemistry, Theoretical and Applied, Churchill Livingstone, Edin- iron-associated focal osteomalacia, J. Clin. Endocrinol. Metab. 88 (2003) 3966–3972. burgh, 1968. [48] R.S. Weinstein, C.L. Lutcher, Chronic erythroid hyperplasia and accelerated bone [25] M.C. de Vernejoul, A. Pointillart, C.C. Golenzer, C. Morieux, J. Bielakoff, D. Modrowski, turnover, Metab. Bone Dis. Relat. Res. 5 (1983) 7–12. L. Miravet, Effects of iron overload on bone remodeling in pigs, Am. J. Pathol. 116 [49] L. Rioja, R. Girot, M. Garabédian, G. Cournot-Witmer, Bone disease in children with (1984) 377–384. homozygous β-thalassemia, Bone Miner. 8 (1990) 69–86. [26] P. Guggenbuhl, P. Fergelot, M. Doyard, H. Libouban, M.P. Roth, Y. Gallois, G. Chales, O. [50] O. Loréal, T. Cavey, E. Bardou-Jacquet, P. Guggenbuhl, M. Ropert, P. Brissot, Iron, Loreal, D. Chappard, Bone status in a mouse model of genetic hemochromatosis, hepcidin, and the metal connection, Front. Pharmacol. 5 (2014) 128. Osteoporos. Int. 22 (2010) 2313–2319. [51] I. Miyoshi, T. Saito, K. Bandobashi, Y. Ohtsuki, H. Taguchi, Concomitant deposition of [27] R.W. Horobin, Selection, Evaluation And Design Of Biological Stains, E. Horwood, NY. aluminum and iron in bone marrow trabeculae, Intern. Med. 45 (2006) 117–118. Chichester., 1988 [52] C. Exley, Why industry propaganda and political interference cannot disguise the in- [28] B. Pemmer, A. Roschger, A. Wastl, J. Hofstaetter, P. Wobrauschek, R. Simon, H. Thaler, evitable role played by human exposure to aluminum in neurodegenerative dis- P. Roschger, K. Klaushofer, C. Streli, Spatial distribution of the trace elements zinc, eases, including Alzheimer's disease, Front. Neurol. 5 (2014). strontium and lead in human bone tissue, Bone 57 (2013) 184–193. [53] R.A. Yokel, C.L. Hicks, R.L. Florence, Aluminum bioavailability from basic sodium alu- [29] M. Ward, H. Ellis, T. Feest, I. Parkinson, D. Kerr, J. Herrington, G. Goode, Osteomalacic minum phosphate, an approved food additive emulsifying agent, incorporated in dialysis osteodystrophy: evidence for a water-borne aetiological agent, probably al- cheese, Food Chem. Toxicol. 46 (2008) 2261–2266. uminium, Lancet 311 (1978) 841–845. [54] N. Chuchu, B. Patel, B. Sebastian, C. Exley, The aluminium content of infant formulas [30] M. Wills, J. Savory, Aluminium poisoning: dialysis encephalopathy, osteomalacia, remains too high, BMC Pediatr. 13 (2013) 162. and anaemia, Lancet 322 (1983) 29–34. [55] G. Schwalfenberg, S.J. Genuis, I. Rodushkin, The benefits and risks of consuming [31] H.H. Malluche, Aluminium and bone disease in chronic renal failure, Nephrol. Dial. brewed tea: beware of toxic element contamination, J. Toxicol. 2013 (2013) 370460. Transplant. 17 (2002) 21–24. [56] J.A. Pennington, S.A. Schoen, Estimates of dietary exposure to aluminium, Food [32] R. Lillie, R. Mowry, Histochemical studies on absorption of iron by tissue sections, Addit. Contam. 12 (1995) 119–128. Bull. Int. Am. Mus. 30 (1949) 91. [57] Y.C. Rajan, B.S. Inbaraj, B.H. Chen, In vitro adsorption of aluminum by an edible bio- [33] M.C. Faugère, H.H. Malluche, Stainable aluminum and not aluminum content re- polymer poly(gamma-glutamic acid), J. Agric. Food Chem. 62 (2014) 4803–4811. flects bone histology in dialyzed patients, Kidney Int. 30 (1986) 717–722. [58] F. Aguilar, H. Autrup, S. Barlow, L. Castle, R. Crebelli, W. Dekant, K. Engel, N. Gontard, [34] M. Baslé, A. Rebel, Y. Mauras, P. Allain, M. Audran, P. Clochon, Concentration of bone D. Gott, S. Grilli, Safety of aluminium from dietary intake scientific opinion of the elements in osteoporosis, J. Bone Miner. Res. 5 (1990) 41–47. panel on food additives, flavourings, processing aids and food contact materials [35] R.W. Horobin, J.A. Kiernan, Conn's biological stains: a handbook of dyes, stains and (AFC), EFSA J. 754 (2008) 1–34. fluorochromes for use in biology and medicine, Biol. Stain Comm. (2002) 208–209. [59] M. Kawahara, M. Kato-Negishi, Link between aluminum and the pathogenesis of [36] H. Ellis, M. Pang, W. Mawhinney, A. Skillen, Demonstration of aluminium in iliac Alzheimer’s disease: The Integration of the Aluminum and Amyloid Cascade Hy- bone: correlation between aluminon and solochrome azurine staining techniques potheses, Int. J. Alzheimers Dis. 2011 (2011) 276393. with data on flameless absorption spectrophotometry, J. Clin. Pathol. 41 (1988) [60] S.C. Bondy, The neurotoxicity of environmental aluminum is still an issue, 1171–1175. Neurotoxicology 31 (2010) 575–581. [37] J. Fernandez-Martin, P. Menendez, G. Acuna, A. Canteros, C. Gómez, J. Cannata, Stain- [61] C. Exley, T. Vickers, Elevated brain aluminium and early onset Alzheimer's disease in ing of bone aluminium: comparison between aluminon and solocbrome azurine and an individual occupationally exposed to aluminium: a case report, J. Med. Case Rep. their correlation with bone aluminium content, Nephrol. Dial. Transplant. 11 (1996) 8(2014)41. 80–85. [62] J. Connor, S. Menzies, S.S. Martin, E. Mufson, A histochemical study of iron, transfer- [38] S.A. Romanski, J.T. McCarthy, K. Kluge, L.A. Fitzpatrick, Proc. Mayo Clin. Proc., 68, rin, and ferritin in Alzheimer's diseased brains, J. Neurosci. Res. 31 (1992) 75–83. Elsevier 1993, pp. 419–426. [63] P.F. Good, D.P. Perl, L.M. Bierer, J. Schmeidler, Selective accumulation of aluminum [39] J. Goldstein, Scanning Electron Microscopy and X-ray Microanalysis, Third edition and iron in the neurofibrillary tangles of Alzheimer's disease: a laser microprobe Springer, US, 2003. (LAMMA) study, Ann. Neurol. 31 (1992) 286–292. [40] K. Sakurai, H. Eba, K. Inoue, N. Yagi, Wavelength-dispersive total-reflection X-ray [64] D. Chappard, P. Insalaco, M. Audran, Aluminum osteodystrophy and celiac disease, fluorescence with an efficient Johansson spectrometer and an undulator X-ray Calcif. Tissue Int. 74 (2004) 122–123. source: detection of 10–16 g-level trace metals, Anal. Chem. 74 (2002) 4532–4535.