- Email: [email protected]
Back-scattered electron imaging of skeletal tissues
0221-8747183 $3.00 + .OO
Metab. Bone Dis. & Rel. Res. 5, 145-150 (1983)
Copyright 0 1984 Pergamon Press Ltd.
Printed in the USA. All rights reserved.
Back-scattered Electron Imaging of Skeletal Tissues A. BOYDE and S.J. JONES
Department of Anatomy and Embryology, Unwersity College, London, England. Address for correspondence
and reprints:
Dr. Alan Boyde, Department of Anatomy and Embryology, University College, London, Gower
Street WC1E 6BT, England.
Abstract
This paper demonstrates practical applications of BSE imaging to the study of skeletal tissues. The method will be a powerful new tool in the investigation of both normal structure and pathologic conditions of bone and joints.
The use of solid-state back-scattered electron (BSE) detectors in the scanning electron microscopic study of skeletal tissues has been investigated. To minimize the topographic element in the image, fiat samples and a ring detector configuration with the sample at normal incidence to the beam and the detector are used. Very flat samples are prepared by diamond micromilling or diamond polishing plasticembedded tissue. Density discrimination in the image is so good that different density phases within mineralized bone can be imaged. For unembedded spongy bone, cut surfaces can be discriminated from natural surfaces by a topographic contrast mechanism. BSE imaging also presents advantages for unembedded samples with rough topography, such as anorganic preparations of the mineralization zone in cartilage, which give rise to severe charging problems with conventional secondary electron imaging. Key Words: Bone-Mineralization-Density-Scanning tron Microscopy-Back-scattered Electron Imagery.
Materials and Methods Two marn types of preparation of the calcified tissues were made for
BSE imaging. In the first, specimens were made anorganic either by oxygen plasma ashing (in a Nanotech 100 plasmachemistry unit) or by deproteinizing solutions. The plasma ashing method also provides a surface tension-free mode of drying for plastic-embedded tissue, The deproteinizing solutions of either cold 7% sodium hypochlorite or hot (50°C) 25% sodium peroxide must be followed by careful washing with distilled water before blotting to draw water and any residual salt away from the surface of interest, and air drying. In the second type of preparation, a flat surface of bone was prepared either by sawing unembedded tissue or by diamond micromilling or diamond polishing methyl methacrylate-embedded specimens. The latter provides a flat surface with little induced topography or surface deformation. Fractured tissue surfaces can also be employed if topographic relief is not too great. All specimens were coated with either carbon for density Imaging or gold for topographic imaging and examrned in a Cambrrdge Stereoscan S4-10 SEM using one of the following modes:
Elec-
1. Conventional secondary electron imaging at 10, 20, or 30 kV. 2. The BSE mode using a solid-state back-scattered electron detector in a four-segmented ring configuration 3. The converted back-scattered electron (CBSE) mode, in which the usual Everhart-Thornley biased scintillator detector is used but the specimen ISfloated at a small positive potential in the range + 50 to +2oov
Introduction The functional worth of skeletal tissue depends upon many factors. Important among these are the cortical and trabecular bone volumes-that is, the quantity of bone tissue within a bone-and the degree of mineralization of the bone-an element of its quality. The degree of mineralization of cartilage and the pattern of its advance and replacement in subarticular cartilage or epiphyseal growth plates are also of structural concern and reflect the physiologic and pathologic states of the individual. Back-scattered electron (BSE) imaging of calcified tissues in a scanning electron microscope (SEM) provides a unique method of assessing these factors with a range and precision previously unattainable. Used for topographic imaging, tenuously linked mineral particles in cartilage and bone may be imaged without charging artefacts. Applied to flat surfaces, the atomic number contrast is of such sensitivity that minor variations in the degree of mineralization can be detected. Additionally, the bone phase of trabecular bone can be sharply discriminated and analyzed histomorphometrically.
Skeletal material was prepared from a variety of sites and included normal and pathologic human bone, rabbit alveolar Sharpey fiber bone, and cartilage from several functionally diverse locations, includrng rat and human epiphyseal, subarticular, and intervertebral cartilage.
Results and Discussion Detection of the cut surface of a spongey bone sample as a separate phase Clinical evaluation of the presence of osteoporosis depends largely on the stereologic assessment of trabecular bone volume. A cut surface of a whole specimen (rather than either a 145
746
A. Boyde and S. J. Jones: BSE imaging of bones
relatively thick undecalcified section or a thin decalcified section) provides the ideal stereologic plane and is simplest to prepare. Secondary electron images do not, however, enable one to discriminate the cut bone from the remainder and have charging artefacts (Fig. 1). Converted back-scattered electron (CBSE) images enable charge-free visualization of the whole specimen, including the cut bone surface, but inadequate distinction in gray level between the cut and uncut bone (Fig. 2). BSE imaging with a ring detector affords this discrimination (Fig. 3). That this is so depends upon two factors, First, only facets perpendicular to the electron beam and the ring detector will show the highest signal level. Second, the signal from flat facets deep to the cut surface at the sample will be partly intercepted by the regions of the specimen closer to the detector. Using size discrimination for features showing the highest signal level, therefore, it is also possible to analyze the image obtained with an automatic image analyzer, such as a Quantimet. Measurements of the volume of the bone tissue phase in the sample of human trabecular bone illustrated (Figs. I, 2, and 3) were in the range 1%200/o, quite within the expected normal range for iliac crest obtained using other techniques,
Fig. 2. Identical specimen and field as Figure 1. Converted backscattered electron image, and charging no longer present. Field width 2.4 mm.
Measurement of osteocyte lacunar area Osteocyte lacunae differ in size in different types of bone under normal physiologic conditions and in some pathologies and also in their volume proportion within the bone. Much work is conducted using the light microscope to assess the lacunar area (lacunar volume) in the belief that osteocyte lacunae might be enlarged in certain diseases. It is, therefore, important to have a reliable and accurate method of measuring such lacunae and their incidence. The standard SE image is marred by charging zones surrounding each lacuna on the cut bone surface (Fig. 4). Applying a potential of +200 volts to the sample greatly reduces these charging artefacts, and a much more acceptable converted BSE image is obtained (Fig. 5). The solid-state BSE image is also free of this artefact (Fig. 6). Both BSE and CBSE images have been used to measure the area of the osteocyte lacunae exposed on a cut surface. Detection of variation in mineralization number contrast
levels in bone: atomic
It is possible to obtain BSE images of a cut surface of bone that mimic the appearance of microradiographs (Figs. 7 and 8).
Fig. 1. Secondary electron image of human trabecular bone. It is difficult to distinguish the cut phase in this low power view because of charging artefact. Field width 2.4 mm. 30 kV.
However, there is a very important difference between the volumes of bone analyzed by microradiography and by backscattered electron imaging using 30 kV electrons in the SEM. The escape depth for the BSE is of the order of l~rn, whereas it is usual to use 30-100 pm thick sections for microradiography. Thus, the phase discrimination for the BSE method is based upon the analysis of very much smaJler volumes of tissue. It is possible to analyze the image with an image-analyzing computer, such as the Quantimet, and measure the volume proportion of bone having different mineral densities. Thesizes and relative maturity levels of osteones can be determined, and features such as reversal and resting (incremental) lines can be detected (Fig. 9). Because of the simplicity of the preparation, it is probable that this technique will have great value in the rapid assessment of human bone biopsies. Analysis of trabecular bone volume, lacunar volume proportion, and mineralization levels can be made on the same section surface. It is also possible to demonstrate the degree of mineralization of extrinsic (Sharpey) fibers incorporated within bone
Fig. 3. Back-scattered electron image of same specimen and field as Figures 1 and 2. The cut phase is readily differentiated, and its area could be measured automatically to give the trabecular bone volume in the sample. Field width 2.4 mm.
147
A. Boyde and S. J. Jones: BSE imaging of bones
Fig. 4. Secondary electron Image (30 kV) of cut bone surface of burr ran spongy bone biopsy. The edges surrounding the lacunae app ear bright due to charging artefact. Field width 177 pm.
Fig. 5. Same view as Frgure 4 with a potential of + 200 volts apfoIled to the sample, giving a converted back-scattered electron Image with no chargrng. Field width 177 pm.
Fig .6 . Same sample as Figures 4 and 5, solrd-state back-scattered elec:trcIn Image wtth srgnals from all four quadrants of the detector. The are:3C A the osteocyte lacunae exposed in the cut surface can be relriibl\ / detected and measured. Field width 177 pm,
Fig. 7. BSE image of mrcromilled surfaces of methacr) {lateembedded, fluorotic human bone giving a mineral density map tof the cut surface. Very small differences in the level of mrneralization c:an be detected. Freld width 948 urn.
Figs. 8 and 9. BSE images of 1 yrn diamond-polished, methacrylate-embedded human bone from a case of hypophosphataemia. Note reversal and resting (incremental) lines detected in images because of their different levels of mineralization and perilacunar mineralization defect characteristic of this disease. Field widths, Fig. 8, 20 kV, 900 pm; Ftg. 9, 30 kV, 460 pm.
148
A. Boyde and S. J. Jones: BSE imaging of bones
(Fig. 10). Because they mineralize independently of the intrinsic fibers to a large extent, they can be discriminated from them. Rapid formation of Sharpey fiber bone, for example, in the alveolar region of the growing jaws, results in the extrinsic fibers being unmineralized or only partially mineralized. Thus, one gains an indication of the rate of inclusion of the fibers within the bone and a map of their distribution and direction. In the specimen demonstrated here, obtained from the alveolar crest of the upper incisal region of the premaxilla, a further indication of the type of bone is presented by the large size, irregular shapes and distribution, and high volume proportion of the osteocyte lacunae in the polished section surface (Fig. 11).
Localization of early stages of mineralization
of cartilage
The mineral phase of hyaline cartilage is microcalcospheritic in form. These mineral spheres may be entirely separate, as in the early stages of mineralization, or imperfectly coalesced away from the mineralizing front. The discontinuity of the mineral phase poses a problem for the scanning electron microscopist in that charging is difficult to avoid in anorganic specimens, and these have lost the entirely discrete mineral particles. BSE imaging of the intact specimens allows much more accurate localization of the mineral phase in the early stages (Fig. 12), and BSE imaging of the anorganic specimens gives a charge free-image (Fig. 13). The technique also provides excellent imaging in sectioned surfaces of the slowly advancing mineral front at the tidemark region-the calcification front below articuthe junction between calcified lar cartilage (Fig. 14)-and cartilage and subchrondral bone (Fig. 15). The advance in the extent of mineralization of the perilacunar region may be due to the variation in the composition of the matrix in the immediate vicinity of the cell, that is, in the lacunar wall. Detection of levels of mineralization in the different regions of the matrix using atomic number contrast with BSE confirms that the pericellular (capsular) region is more highly mineralized than the interterritorial matrix, presumably in part due to its greater proportion of ground substance.
Fig. 11. Higher power view of same specimen as in Figure 10. The inhomogeneity of the mineral levels in the Sharpey fibers is evident. In some areas this atomic number contrast allows detection of the tncremental layers of the intrinsic fibers. Continuity exists between lacunae of several groups of adjacent osteocytes. Field width 125 pm.
Identification
of diffusion pathways in calcified cartilage
The vitality of the cells in calcified cartilage, which is usually avascular, is dependent upon the diffusion of nutritive substances through the matrix. Calcification has been thought popularly to constitute a barrier to this process and to lead to the death of chrondrocytes. Evidence from tissue culture experiments of cells retained in situ on longitudinal splits through cartilage (Boyde and Jones, 1983) and from analysis of the Kf and Na+ levels in chrondrocytes in all levels of the growth plate (Boyde and Shapiro, 1980) contradicts this view, and better preservation of chrondrocytes for TEM confirms that many of the histologic parameters taken as evidence for the cells’ demise are artefactual (Hunziger et al., 1982). BSE imaging of the architecture of the mineral phase of the cartilage of the growth plate allows excellent visualization of discontinuities in the mineral encirclement of the cells-even in those situations where the dividing walls of matrix between cells of one column are reputed to be mineralized (Fig. 16). The flow of nutrients must be at least as good, and probably is much better, in such calcified cartilage as it is in bone.
Fig. 10. BSE image (20 kV) of lam diamond-polished methacrylateembedded rabbit alveolar bone. The Sharpey fibers are not fully mineralized and extend throughout the total thickness of this interdental septum. The vascular channels are seen as large black spaces, which show one of threeconditions attheir bone border-resting bone border well mineralized, bone border as yet not fully mineralized, and the resorbed, scalloped border. The osteoid is not detected in this image. Small black areas represent osteocyte lacunae. Field width 400 pm.
Fig. 12. BSE image (30 kV) of whole specimen of cartilage giving charge-free topographic contrast of small mineral particles within the collagenous network of the matrix. Field width 17 pm.
149
A. Boyde and S J. Jones: BSE imaging of bones
Fig. 13. Image (30 kV) of anorganic rat epiphyseal cartilage topographic and density information. Field width 102 pm.
Fig. 14. BSE image (30 kV) of longrtudinal section fracture through the trdemark zone, showing level of mineralization below the articular cartilage of femoral head in 82.yr-old male. Field width 336 Wm.
at the mineralizing
front, giving both
Fig. 15. BSE image (30 kV) of longitudinally fractured osteochondral junction below articular cartilage of human femoral head in 81 -yr-old male. The border between the bone (above) and calcified cartilage (below) is distinguished by a textural change rather than a marked change in the degree of mineralization of the two tissues. In this respect, articular cartilage doffers from the temporary growth cartilage. Freld width 339 pm.
Fig. 16. Stereopair: 30 kV BSE image of anorganic rat caudal vertebra growth cartilage. Deficiencies in the mineral encasement of the cartilage cells allow the transference of nutrients. Field height 129 am.
150
A. Boyde and S. J. Jones: BSE imaging of bones
Conclusions
References
BSE imaging can be used to provide charge-free imaging of the mineral phase of both bone and cartilage. Its sensitivity for the detection of atomic number contrast and the simplicity of the preparation of samples for examination mean that we now have a technique for the rapid acquisition of qualitative and comparative and quantitative data from normal and pathologic skeletal tissues.
Boy& A. and Jones S.J.: Scanning electron microscopy of cartilage. In: Carfi/age: Structure, Function and Biochemistry, B.K. Hall, ed. Academic Press Inc., New York, 1983, Vol. 1, Chap. 5, pp. 105148.
This work has been supported by grants from the MRC and SRC. We thank Drs. D. Birkenhaeger-Fraenkel and FL
Boyde A. and Shapiro I. M.: Energy dispersive x-ray elemental analysis of isolated epiphyseal growth plate chondrocyte fragments. Histochemistry 6935-94, 1980. Hunziger E.B.. Hermann W. and Shenk R.K.: Improved cartilage fixation by ruthenium hexamine lrichloride (AHT). A prerequisite for morphometry in growth cartilage. J. Ulffastruct. Res. 61:1-12, 1982.
Acknowledgement:
Steendijk for the provision of bone samples.
Received: June 27, 1983 Accepted: August 18, 1983
L’utilisatlon de back-scattered ekctrons @SE) dans t%tuds en mkroscople 6kctmnique & bakyage des tissus squekttiques a et6 expor&. Four reduire au maximum Is kcteur topographique dans I’imege, des echantillms plats et un d&ecteur annulabe avec rayonnement perpendkulalre s I’echantillon et au d&Mew ont et6 utilises. Des Bchantlllons tres plats ont et6 prepares par mkmmeulage dlamanth ou polissage au diament du tissu inclus en matiere plastique. Le pouvolr de discrimination des densites au seln de I’image est si bon que tee dlfferents nlveaux de densite de I’os mIneralis sont visibles sur I’image. pour 1’0s spongleux non lnclus, les surfaces de coupe peuvent Ctre dlfferenciees des surkces naturelles par un mecanlsme de contraste topographtque. La technique de BSE presente aussl des avantages pour des 6chantllkns non inclus de topographle mal d&ennlrke, comme les preparations anorganlques de la zone de mineralisatlon du cartilage qui donne lieu b de sdrleux probkmes de charge sur lee images de micmscopie electronique P balayage conventionnelk.
Metab. Bone Dis. & Rel. Res. 5, 145-150 (1983)
Copyright 0 1984 Pergamon Press Ltd.
Printed in the USA. All rights reserved.
Back-scattered Electron Imaging of Skeletal Tissues A. BOYDE and S.J. JONES
Department of Anatomy and Embryology, Unwersity College, London, England. Address for correspondence
and reprints:
Dr. Alan Boyde, Department of Anatomy and Embryology, University College, London, Gower
Street WC1E 6BT, England.
Abstract
This paper demonstrates practical applications of BSE imaging to the study of skeletal tissues. The method will be a powerful new tool in the investigation of both normal structure and pathologic conditions of bone and joints.
The use of solid-state back-scattered electron (BSE) detectors in the scanning electron microscopic study of skeletal tissues has been investigated. To minimize the topographic element in the image, fiat samples and a ring detector configuration with the sample at normal incidence to the beam and the detector are used. Very flat samples are prepared by diamond micromilling or diamond polishing plasticembedded tissue. Density discrimination in the image is so good that different density phases within mineralized bone can be imaged. For unembedded spongy bone, cut surfaces can be discriminated from natural surfaces by a topographic contrast mechanism. BSE imaging also presents advantages for unembedded samples with rough topography, such as anorganic preparations of the mineralization zone in cartilage, which give rise to severe charging problems with conventional secondary electron imaging. Key Words: Bone-Mineralization-Density-Scanning tron Microscopy-Back-scattered Electron Imagery.
Materials and Methods Two marn types of preparation of the calcified tissues were made for
BSE imaging. In the first, specimens were made anorganic either by oxygen plasma ashing (in a Nanotech 100 plasmachemistry unit) or by deproteinizing solutions. The plasma ashing method also provides a surface tension-free mode of drying for plastic-embedded tissue, The deproteinizing solutions of either cold 7% sodium hypochlorite or hot (50°C) 25% sodium peroxide must be followed by careful washing with distilled water before blotting to draw water and any residual salt away from the surface of interest, and air drying. In the second type of preparation, a flat surface of bone was prepared either by sawing unembedded tissue or by diamond micromilling or diamond polishing methyl methacrylate-embedded specimens. The latter provides a flat surface with little induced topography or surface deformation. Fractured tissue surfaces can also be employed if topographic relief is not too great. All specimens were coated with either carbon for density Imaging or gold for topographic imaging and examrned in a Cambrrdge Stereoscan S4-10 SEM using one of the following modes:
Elec-
1. Conventional secondary electron imaging at 10, 20, or 30 kV. 2. The BSE mode using a solid-state back-scattered electron detector in a four-segmented ring configuration 3. The converted back-scattered electron (CBSE) mode, in which the usual Everhart-Thornley biased scintillator detector is used but the specimen ISfloated at a small positive potential in the range + 50 to +2oov
Introduction The functional worth of skeletal tissue depends upon many factors. Important among these are the cortical and trabecular bone volumes-that is, the quantity of bone tissue within a bone-and the degree of mineralization of the bone-an element of its quality. The degree of mineralization of cartilage and the pattern of its advance and replacement in subarticular cartilage or epiphyseal growth plates are also of structural concern and reflect the physiologic and pathologic states of the individual. Back-scattered electron (BSE) imaging of calcified tissues in a scanning electron microscope (SEM) provides a unique method of assessing these factors with a range and precision previously unattainable. Used for topographic imaging, tenuously linked mineral particles in cartilage and bone may be imaged without charging artefacts. Applied to flat surfaces, the atomic number contrast is of such sensitivity that minor variations in the degree of mineralization can be detected. Additionally, the bone phase of trabecular bone can be sharply discriminated and analyzed histomorphometrically.
Skeletal material was prepared from a variety of sites and included normal and pathologic human bone, rabbit alveolar Sharpey fiber bone, and cartilage from several functionally diverse locations, includrng rat and human epiphyseal, subarticular, and intervertebral cartilage.
Results and Discussion Detection of the cut surface of a spongey bone sample as a separate phase Clinical evaluation of the presence of osteoporosis depends largely on the stereologic assessment of trabecular bone volume. A cut surface of a whole specimen (rather than either a 145
746
A. Boyde and S. J. Jones: BSE imaging of bones
relatively thick undecalcified section or a thin decalcified section) provides the ideal stereologic plane and is simplest to prepare. Secondary electron images do not, however, enable one to discriminate the cut bone from the remainder and have charging artefacts (Fig. 1). Converted back-scattered electron (CBSE) images enable charge-free visualization of the whole specimen, including the cut bone surface, but inadequate distinction in gray level between the cut and uncut bone (Fig. 2). BSE imaging with a ring detector affords this discrimination (Fig. 3). That this is so depends upon two factors, First, only facets perpendicular to the electron beam and the ring detector will show the highest signal level. Second, the signal from flat facets deep to the cut surface at the sample will be partly intercepted by the regions of the specimen closer to the detector. Using size discrimination for features showing the highest signal level, therefore, it is also possible to analyze the image obtained with an automatic image analyzer, such as a Quantimet. Measurements of the volume of the bone tissue phase in the sample of human trabecular bone illustrated (Figs. I, 2, and 3) were in the range 1%200/o, quite within the expected normal range for iliac crest obtained using other techniques,
Fig. 2. Identical specimen and field as Figure 1. Converted backscattered electron image, and charging no longer present. Field width 2.4 mm.
Measurement of osteocyte lacunar area Osteocyte lacunae differ in size in different types of bone under normal physiologic conditions and in some pathologies and also in their volume proportion within the bone. Much work is conducted using the light microscope to assess the lacunar area (lacunar volume) in the belief that osteocyte lacunae might be enlarged in certain diseases. It is, therefore, important to have a reliable and accurate method of measuring such lacunae and their incidence. The standard SE image is marred by charging zones surrounding each lacuna on the cut bone surface (Fig. 4). Applying a potential of +200 volts to the sample greatly reduces these charging artefacts, and a much more acceptable converted BSE image is obtained (Fig. 5). The solid-state BSE image is also free of this artefact (Fig. 6). Both BSE and CBSE images have been used to measure the area of the osteocyte lacunae exposed on a cut surface. Detection of variation in mineralization number contrast
levels in bone: atomic
It is possible to obtain BSE images of a cut surface of bone that mimic the appearance of microradiographs (Figs. 7 and 8).
Fig. 1. Secondary electron image of human trabecular bone. It is difficult to distinguish the cut phase in this low power view because of charging artefact. Field width 2.4 mm. 30 kV.
However, there is a very important difference between the volumes of bone analyzed by microradiography and by backscattered electron imaging using 30 kV electrons in the SEM. The escape depth for the BSE is of the order of l~rn, whereas it is usual to use 30-100 pm thick sections for microradiography. Thus, the phase discrimination for the BSE method is based upon the analysis of very much smaJler volumes of tissue. It is possible to analyze the image with an image-analyzing computer, such as the Quantimet, and measure the volume proportion of bone having different mineral densities. Thesizes and relative maturity levels of osteones can be determined, and features such as reversal and resting (incremental) lines can be detected (Fig. 9). Because of the simplicity of the preparation, it is probable that this technique will have great value in the rapid assessment of human bone biopsies. Analysis of trabecular bone volume, lacunar volume proportion, and mineralization levels can be made on the same section surface. It is also possible to demonstrate the degree of mineralization of extrinsic (Sharpey) fibers incorporated within bone
Fig. 3. Back-scattered electron image of same specimen and field as Figures 1 and 2. The cut phase is readily differentiated, and its area could be measured automatically to give the trabecular bone volume in the sample. Field width 2.4 mm.
147
A. Boyde and S. J. Jones: BSE imaging of bones
Fig. 4. Secondary electron Image (30 kV) of cut bone surface of burr ran spongy bone biopsy. The edges surrounding the lacunae app ear bright due to charging artefact. Field width 177 pm.
Fig. 5. Same view as Frgure 4 with a potential of + 200 volts apfoIled to the sample, giving a converted back-scattered electron Image with no chargrng. Field width 177 pm.
Fig .6 . Same sample as Figures 4 and 5, solrd-state back-scattered elec:trcIn Image wtth srgnals from all four quadrants of the detector. The are:3C A the osteocyte lacunae exposed in the cut surface can be relriibl\ / detected and measured. Field width 177 pm,
Fig. 7. BSE image of mrcromilled surfaces of methacr) {lateembedded, fluorotic human bone giving a mineral density map tof the cut surface. Very small differences in the level of mrneralization c:an be detected. Freld width 948 urn.
Figs. 8 and 9. BSE images of 1 yrn diamond-polished, methacrylate-embedded human bone from a case of hypophosphataemia. Note reversal and resting (incremental) lines detected in images because of their different levels of mineralization and perilacunar mineralization defect characteristic of this disease. Field widths, Fig. 8, 20 kV, 900 pm; Ftg. 9, 30 kV, 460 pm.
148
A. Boyde and S. J. Jones: BSE imaging of bones
(Fig. 10). Because they mineralize independently of the intrinsic fibers to a large extent, they can be discriminated from them. Rapid formation of Sharpey fiber bone, for example, in the alveolar region of the growing jaws, results in the extrinsic fibers being unmineralized or only partially mineralized. Thus, one gains an indication of the rate of inclusion of the fibers within the bone and a map of their distribution and direction. In the specimen demonstrated here, obtained from the alveolar crest of the upper incisal region of the premaxilla, a further indication of the type of bone is presented by the large size, irregular shapes and distribution, and high volume proportion of the osteocyte lacunae in the polished section surface (Fig. 11).
Localization of early stages of mineralization
of cartilage
The mineral phase of hyaline cartilage is microcalcospheritic in form. These mineral spheres may be entirely separate, as in the early stages of mineralization, or imperfectly coalesced away from the mineralizing front. The discontinuity of the mineral phase poses a problem for the scanning electron microscopist in that charging is difficult to avoid in anorganic specimens, and these have lost the entirely discrete mineral particles. BSE imaging of the intact specimens allows much more accurate localization of the mineral phase in the early stages (Fig. 12), and BSE imaging of the anorganic specimens gives a charge free-image (Fig. 13). The technique also provides excellent imaging in sectioned surfaces of the slowly advancing mineral front at the tidemark region-the calcification front below articuthe junction between calcified lar cartilage (Fig. 14)-and cartilage and subchrondral bone (Fig. 15). The advance in the extent of mineralization of the perilacunar region may be due to the variation in the composition of the matrix in the immediate vicinity of the cell, that is, in the lacunar wall. Detection of levels of mineralization in the different regions of the matrix using atomic number contrast with BSE confirms that the pericellular (capsular) region is more highly mineralized than the interterritorial matrix, presumably in part due to its greater proportion of ground substance.
Fig. 11. Higher power view of same specimen as in Figure 10. The inhomogeneity of the mineral levels in the Sharpey fibers is evident. In some areas this atomic number contrast allows detection of the tncremental layers of the intrinsic fibers. Continuity exists between lacunae of several groups of adjacent osteocytes. Field width 125 pm.
Identification
of diffusion pathways in calcified cartilage
The vitality of the cells in calcified cartilage, which is usually avascular, is dependent upon the diffusion of nutritive substances through the matrix. Calcification has been thought popularly to constitute a barrier to this process and to lead to the death of chrondrocytes. Evidence from tissue culture experiments of cells retained in situ on longitudinal splits through cartilage (Boyde and Jones, 1983) and from analysis of the Kf and Na+ levels in chrondrocytes in all levels of the growth plate (Boyde and Shapiro, 1980) contradicts this view, and better preservation of chrondrocytes for TEM confirms that many of the histologic parameters taken as evidence for the cells’ demise are artefactual (Hunziger et al., 1982). BSE imaging of the architecture of the mineral phase of the cartilage of the growth plate allows excellent visualization of discontinuities in the mineral encirclement of the cells-even in those situations where the dividing walls of matrix between cells of one column are reputed to be mineralized (Fig. 16). The flow of nutrients must be at least as good, and probably is much better, in such calcified cartilage as it is in bone.
Fig. 10. BSE image (20 kV) of lam diamond-polished methacrylateembedded rabbit alveolar bone. The Sharpey fibers are not fully mineralized and extend throughout the total thickness of this interdental septum. The vascular channels are seen as large black spaces, which show one of threeconditions attheir bone border-resting bone border well mineralized, bone border as yet not fully mineralized, and the resorbed, scalloped border. The osteoid is not detected in this image. Small black areas represent osteocyte lacunae. Field width 400 pm.
Fig. 12. BSE image (30 kV) of whole specimen of cartilage giving charge-free topographic contrast of small mineral particles within the collagenous network of the matrix. Field width 17 pm.
149
A. Boyde and S J. Jones: BSE imaging of bones
Fig. 13. Image (30 kV) of anorganic rat epiphyseal cartilage topographic and density information. Field width 102 pm.
Fig. 14. BSE image (30 kV) of longrtudinal section fracture through the trdemark zone, showing level of mineralization below the articular cartilage of femoral head in 82.yr-old male. Field width 336 Wm.
at the mineralizing
front, giving both
Fig. 15. BSE image (30 kV) of longitudinally fractured osteochondral junction below articular cartilage of human femoral head in 81 -yr-old male. The border between the bone (above) and calcified cartilage (below) is distinguished by a textural change rather than a marked change in the degree of mineralization of the two tissues. In this respect, articular cartilage doffers from the temporary growth cartilage. Freld width 339 pm.
Fig. 16. Stereopair: 30 kV BSE image of anorganic rat caudal vertebra growth cartilage. Deficiencies in the mineral encasement of the cartilage cells allow the transference of nutrients. Field height 129 am.
150
A. Boyde and S. J. Jones: BSE imaging of bones
Conclusions
References
BSE imaging can be used to provide charge-free imaging of the mineral phase of both bone and cartilage. Its sensitivity for the detection of atomic number contrast and the simplicity of the preparation of samples for examination mean that we now have a technique for the rapid acquisition of qualitative and comparative and quantitative data from normal and pathologic skeletal tissues.
Boy& A. and Jones S.J.: Scanning electron microscopy of cartilage. In: Carfi/age: Structure, Function and Biochemistry, B.K. Hall, ed. Academic Press Inc., New York, 1983, Vol. 1, Chap. 5, pp. 105148.
This work has been supported by grants from the MRC and SRC. We thank Drs. D. Birkenhaeger-Fraenkel and FL
Boyde A. and Shapiro I. M.: Energy dispersive x-ray elemental analysis of isolated epiphyseal growth plate chondrocyte fragments. Histochemistry 6935-94, 1980. Hunziger E.B.. Hermann W. and Shenk R.K.: Improved cartilage fixation by ruthenium hexamine lrichloride (AHT). A prerequisite for morphometry in growth cartilage. J. Ulffastruct. Res. 61:1-12, 1982.
Acknowledgement:
Steendijk for the provision of bone samples.
Received: June 27, 1983 Accepted: August 18, 1983
L’utilisatlon de back-scattered ekctrons @SE) dans t%tuds en mkroscople 6kctmnique & bakyage des tissus squekttiques a et6 expor&. Four reduire au maximum Is kcteur topographique dans I’imege, des echantillms plats et un d&ecteur annulabe avec rayonnement perpendkulalre s I’echantillon et au d&Mew ont et6 utilises. Des Bchantlllons tres plats ont et6 prepares par mkmmeulage dlamanth ou polissage au diament du tissu inclus en matiere plastique. Le pouvolr de discrimination des densites au seln de I’image est si bon que tee dlfferents nlveaux de densite de I’os mIneralis sont visibles sur I’image. pour 1’0s spongleux non lnclus, les surfaces de coupe peuvent Ctre dlfferenciees des surkces naturelles par un mecanlsme de contraste topographtque. La technique de BSE presente aussl des avantages pour des 6chantllkns non inclus de topographle mal d&ennlrke, comme les preparations anorganlques de la zone de mineralisatlon du cartilage qui donne lieu b de sdrleux probkmes de charge sur lee images de micmscopie electronique P balayage conventionnelk.