Main Scientific Interests

BONE DISEASES AND THERAPIES

The investigation of pathological changes and effects of therapies is one of our main scientific interests. One aim is to relate specific changes in the bone structure/material to specific metabolic or genetic diseases. This could give more insights into the pathologies (such as osteoporosis, osteomalacia…) and create a basis for the prevention and future therapies. It is important to note that some diseases cannot be identified by routine DXA measurements but the investigation of bone material at different hierarchical levels is needed for diagnosis. Another aim of our studies is to clarify the effects of specific therapies (for osteoporosis e.g.) on bone material. During recent years, iliac crest bone biopsies of patients treated with bone anabolic agents (such as sodium fluoride or intermittent parathyroid hormone treatment) or treated with bone resorption inhibitors such as bisphsphonates were studied in this context. The following will give a brief description of these works.

The effect of sodium fluoride (NaF) on bone material:

NaF is a bone anabolic agent used in osteoporosis e.g. to increase osteoblast activity and bone mass. However, NaF is inserted into bone mineral and causes alterations in the chemical properties of bone mineral. We found these chemical changes causing also abnormalities in the collagen/mineral composite. Bone matrix which is formed during NaF treatment can easily be identified by its inhomogeneous mineralization pattern in the backscattered electron image. These alterations are reflected by a shift of the bone mineralization density distribution (BMDD) to higher calcium concentrations. The properties of the collagen/mineral composite material, the nanostructure of the material, was found highly altered due to NaF as observed by an altered size-shape-orientation-distribution of the mineral particles. Abnormally increased particle thickness could be found which was not evident in normal bone material. Deviations from normal could be detected not only for higher dose of NaF but also for low doses of NaF. Similar abnormalities due to NaF were also found in an animal study which could also show that the deviations from normal bone material correlated with lowered mechanical quality of NaF treated bone. (Fratzl et al., J Bone Miner Res 9:1541-1549;1994; Roschger et al., Bone 20:393-397; 1997)

Bone quality after intermittent parathyroid (PTH) treatment:

Parathyroid (PTH) hormone is mainly effecting the osteoclasts. If PTH is continuously released for pathological reasons (hyperparathyroidism e.g.), the stimulation of the osteoclasts causes a decrease in bone mass. However, the intermittent administration of PTH (for instance by a dose of 25 micrograms per day during 18-36 months) shows an anabolic effect on bone mass. Aim of our study was to investigate the mineralization and the nanostructure of the bone tissue formed during PTH treatment. The biopsies taken before and after PTH treatment from each patient (males and females) provided pairwise comparison. We found the typical calcium concentration (the peak of the bone mineralization density distribution) partly lowered in the patients due to treatment. All patients showed a highly increased width of the BMDD (up to 22%) corresponding to a decrease in homogeneity of mineralization due to an increase in the percentage of bone areas with low calcium concentration. This fact can be explained by the enormous increase in bone mass during treatment (thus is a high amount of young tissue with lower degree of mineralization) and reflects normal bone material development. No abnormalities of bone nanostructure could be observed in the newly formed bone during PTH. Size, shape and orientation of the mineral particles were undistinguishable from normal bone. These finding suggest that authentic bone material is formed during PTH treatment which is important for the clinical usage of intermittent PTH. (Misof et al., J Clin Encocrinol Metab 88:1150-1156;2003)

The effect of bisphosphonates on bone material:

Bisphosphonates act as inhibitors of bone absorption (in contrast to NaF or intermittent PTH which both are bone anabolic agents) by intervening the function of osteoclasts. We studied the bone material of iliac crest bone biopsies from osteoporotic patients who were treated by the bisphosphonate Alendronate for two or three years. No change in the nanostructure (this is mineral particle size, shape and orientation) was detected in bone material after treatment. The reduction of bone turnover due to Alendronate, however, caused an increase in homogeneity of mineralization. This decrease in the width of the bone mineralization density distribution (BMDD) was accompanied by a slight but significant increase in the typical calcium concentration in bone from patients treated with Alendronate compared to placebo treated osteoporotic patients. One has to note that the osteoporotic patients were shown to have lowered typical calcium concentration before treatment and treatment shifted their BMDD towards normal values (normal values were obtained from a group of healthy individuals). Similar observations of reduced width of the BMDD were also described for Alendronate treated animals. (Roschger et al., Bone 29:185-191;2001; Roschger et al., Bone 20:393-397;1997)

The bone mineralization density distribution of haemodialysis patients before and after kidney transplantation (NTX) treated with bisphosphonates and nontreated

In general, haemodialysis patients reveal renal osteodystrophy (ROD) characterized by reduced bone quality, which is correlated to other diseases (such as secondary hyperparathyroidism, deviations in VitD-metabolism, PTH-unsensitivity of bone cells, diabetes mellitus) and to treatment with corticosteroids or cytotoxic medicine. High- or low turnover ROD type I to III can be distinguished after histological classification of bone biopsies. After NTX, the bone mass of the patients is further decreased by immunosuppressiva often leading to bone fractures. In this multidisciplinary project, we studied bone mineralization in correlation with clinical parameters as well as the effect of bisphosphonate therapy on the bone material of patients after NTX.

First, the bone biopsies taken before NTX were studied histologically and histomorphometrically and subsequently measured for their bone mineralization density distribution (BMDD). These results were further correlated with clinical data of the patients, such as BMD (hip and spinal), serum and urine parameters, PTH, alkaline phosphatase (ALP), serum-calcium, serum-phosphate, osteocalcin, VitD, CTX and osteoprotegerin. We found that BMD alone could not identify type ROD II or III. However, OPG from ROD type III was significantly reduced compared to low-turnover ROD II patients. OPG can therefore be considered as a new serum marker for the classification of hemodialysis patients. Further, iPTH was significantly increased for ROD III compared to ROD II. The mean and the typical calcium concentration (both obtained from the BMDD) were significantly lower for ROD III compared to healthy bone and were both correlated with OPG in a positive linear way. This can be explained by the inhibition of activity and differentiation of osteoclasts by OPG. A high OPG serum level could inhibit bone resorption and cause an increase of calcium concentration in bone. Histomorphometric parameters, such as osteoblast surface/bone surface and eroded surface/bone surface showed a significantly negative correlation with the typical calcium concentration (Haas et al., Am J Kidney Dis 39:580-586;2002).

Six months after NTX, the calcium concentration was increased for the patients treated with bisphosphonates whereas in the placebo-treated group of patients, the calcium concentration remained unchanged. In both groups of treated and non-treated patients high bone turnover was reduced. This was reflected by a reduction of areas of bone resorption and a lower number of osteoclasts and osteoblasts compared to the patients’ first biopsies. Serological parameters of bone turnover were reduced in the bisphosphonate treated patients. The function of the transplanted kidney was not effected by the bisphosphonates. For conclusion, treatment with bisphosphonates increased trabecular bone calcium concentration after NTX. This beneficial effect was confirmed by an increase in spinal BMD and by the stabilization of the BMD at the femur. (Haas et al., Kidney Int. 63:1130-1136;2003)

Animal models for specific diseases and treatment, studied at the LBIO:

Pathological changes in bone material due to genetic diseases are often studied in animal models (mainly murine models) which reveal an analogous genetic mutation (to the human disease) leading to a symptomatically similar phenotype. The effect of therapies on bone material quality can also be studied by animal models.

During recent years, lots of murine models, such as the osteogenesis imperfecta murine (oim) model were studied at the LBIO for bone material quality. The oim mouse acts as a model for the human brittle bone disease (osteogenesis imperfecta) which is a genetic disease characterized by dramatically reduced mechanical competence of bone. This mouse model has a well defined genetic mutation which causes the collagen to consist of a 1(I)-homotrimers only. We could show that this deviation in the molecular structure of collagen dramatically reduced the mechanical properties of oim tendons: ultimate strain and stress were both reduced by 50% compared to normal tendons (Misof et al., J Clin Invest:100:40-45;1997). Interestingly, it is not only the collagen itself that is effected by the genetic mutation but there are also abnormalities in bone mineralization. Typical calcium concentration and microhardness of bone were found highly increased (Grabner et al., Bone 29:453-457; 2001). We observed also abnormalities in the collagen/mineral composite material, such as reduced mineral particle thickness (Fratzl et al., J Clin Invest 97:396-402;1996; Grabner et al., Bone 29:453-457;2001), suggesting that all of these deviations in collagen and mineral phase contribute to the extreme brittleness of oim bone.

A model of infant hypophosphatasia is the tissue non-specific alkaline phosphatase knockout (TNAP-/-) mouse, which was also studied at the LBIO. This mouse shows reduced growth within the first days after birth, reduced or no gain of weight, together with decreased growth of length and thickness of long bones. In our work, we studied cortical femoral bone from mice aged 8 to 22 days. We found all parameters (mean calcium concentration, homogeneity of mineralization, mean particle thickness and orientation) being spread within a broad range and characteristic for immature bone. The low orientation along the bone long axis gave evidence for a massive disturbance of the collagen structure. These findings were consistent with immunohistochemical staining for osteopontin (an extracellular protein that is important for mineralization). For conclusion, TNAP is not only important for the mineralization in general but this enzyme has an essential regulatory function for the bone matrix. (Tesch et al., J Bone Miner Res 18:117-125;2003)

Further, we studied bone mineralization of mice with an overexpression of VitD-receptors of mature osteoblasts, the OSVDR mouse model. VitD is essential for bone metabolism and lack of VitD causes rickets in children and osteomalacia in adults. The exact mechanisms how VitD acts on the bone cells are not clarified yet. However, there is evidence that VitD acts mainly via the VitD-receptor (VDR) of osteoblasts. The OSVDR mouse exhibits increased mechanical toughness of the long bones together with elevated Young’s modulus. This increase in toughness makes the OSVDR mouse interesting for new therapies of osteoporosis e.g. We studied these mice for changes in bone material due to the genetic mutation. Interestingly, we found slightly but significantly increased bone mineralization density in femoral bone material but this increased mineralization was found much lower than that of oim bone (see above). No abnormalities in the mineral/collagen composite material could be detected. In summary, we could show that the VitD receptor plays an important role for mineralization and toughness of bone and these findings suggest a role of the VitD receptor in future therapies. (Misof et al., Calcif Tissue Int 73:251-257;2003)

BONE GROWTH, DEVELOPMENT, AND AGEING

During the development from the foetus to the adult individual, bone material is adapting to respective mechanical demands by structural and material changes. The investigation of these changes during bone development is an essential scientific subject and is the basis for the diagnosis of pathological deviations from normal development. At the LBIO, we use several modern physical techniques with high spatial resolution to characterize the development of human bone material from micro- to nanometer-size-level.

The mineralization density distribution of trabecular bone from healthy adult individuals

We studied the bone mineralization density distribution (BMDD) of trabecular human bone and found the typical calcium concentration (Ca Peak), the mean calcium concentration (Ca Mean), the shape of the BMDD to be similar for healthy individuals (aged 20-91 years) independent of skeletal site (iliac crest, femoral head or vertebrae), of gender or of ethnic origin. It is striking that the BMDD was found to be constant for healthy individuals suggesting that the mineralization pattern is biologically and mechanically optimized. Even small deviations from the normal BMDD are thought to decrease the mechanical competence of bone and can give evidence for bone diseases. (Roschger et al., Bone 32:316-323;2003)

The development of bone material in the human L4-vertebra

The vertebrae belong to the main weight bearing components of the skeleton. There is lot of literature available concerning the structural development of the vertebrae but the changes which the material undergoes during development were not studied in detail. This was the reason why we studied 46 L4-vertebral bodies from bone-healthy individuals aged 15 weeks post conception to 97 years of age by different methods such as morphometrical analysis, qBEI, and scanning-SAXS (Roschger et al., J Struct Biol 136:126-136, 2001). We focussed on the early development from prenatal to postnatal (from non-weight bearing to weight bearing). A dramatic change in trabecular architecture from a radial pattern (embryonic bone) to a vertical/horizontal pattern (after birth) could be observed. The number of trabeculae shows a maximum during childhood and is subsequently decreasing with age. However, bone area/tissue area (B.Ar./T.Ar.) and the trabecular thickness (Tb.Th.) are both increasing until adolescence. The typical calcium concentration and the mean mineral particle thickness were both rapidly increasing until 4 years of age and their further increase was much slower (both parameters showed a logarithmic increase with age). Additionally, we found the particle orientation parallel to the direction of the trabeculae which gives evidence that the orientation of the particles follows the main stress-directions in the vertebrae (Rinnerthaler et al., Calcif Tissue Int 64: 422-429;1999; Zizak et al, J Struct Biol 141: 208-217;2003).

STRUCTURE/FUNCTION RELATIONSHIP IN SPECIFIC MINERALIZED TISSUES

The macroscopic and microscopic structure and the material properties provide essentially the outstanding mechanical properties (mechanical toughness e.g.) of mineralized, biological tissues (such as bone, cartilage, dentin, mineralized tendon, etc.). From the material physicists view, theses tissues can be considered as a composite material consisting of a highly elastic (collagen in bone) and a hard, brittle component (hydroxyapatite, e.g.). There is scientific interest in understanding how these two components are arranged to lead to a material which reveals the optimum properties of both components. Results of such studies are also essential for the development of bone replacement materials and other new artificial materials.

Bone/cartilage interface

The bone/cartilage interface is crucial for the transmission of the forces acting on the loaded skeleton from the highly elastic material cartilage to the harder tissue bone. In this context, we studied the interface for its mineralization and nanostructure. In general, mineralized cartilage and bone differ in their organic matrix they consist of: collagen type II in the case of cartilage, collagen type I in case of bone. In spite of this difference, both materials can be mineralized by hydroxyapatite particles. At the interface between cartilage and bone, there exists a thin layer of mineralized cartilage. The combination of qBEI and scanning-SAXS at the synchrotron allowed high spatial resolution for the investigation of this transition zone in 3 samples: bone/articular cartilage, bone/intervertebral disk, and bone/ growth cartilage. All samples had in common that mineralized cartilage showed higher calcium concentration than the adjacent bone. The highest calcium concentration was found in growth cartilage, where we found the mineral particle thickness to be larger than in bone. In all other measured tissues the mineral particle thickness was similar for mineralized cartilage and bone. This is surprising since the organic matrices of these tissues are different as mentioned above. The most striking result, however, was the abrupt change of mineral particle orientation at the interface: in mineralized cartilage, the mineral particles were aligned perpendicular to the interface, in the adjacent bone, they were aligned parallel with the interface. For summary, between unmineralized cartilage and bone, a transition zone of mineralized cartilage can be found which is tightly connected to bone. This zone has the important mechanical purpose to inhibit crack generation during loading (compression in general). The differences in mineral particle orientation between mineralized cartilage and bone are reflecting the main collagen orientation in these tissues which is assumed to be defined mainly by the mechanical stress pattern. (Zizak et al., J Struct Biol 141:208-217; 2003)

Dentin

Coronal dentin, the main component of the tooth, which is mechanically loaded during chewing was studied for the structure/function relationship by the combination of several methods with high spatial resolution, such as FTIR, scanning-SAXS, qBEI, nanoindentation at the AFM for the investigation of the enamel/dentin junction. The results obtained from these different techniques could be correlated by means of an overall coordinate-system. Most important, we found a gradient-like behaviour of all measured parameters in dentin adjacent to the dentin/enamel-junction. The correlation of the measured parameters showed that the calcium concentration and the mineral particle thickness were both influencing the elastic behaviour of dentin. These findings suggest that this specific structure of dentin (similar to specific artificial composite materials) inhibits crack deviation from enamel to dentin. (Tesch et al. Calcif Tissue Int 69:147-157;2001)

The nano-composite material of bone from mice

The intrinsic properties of the bone material, thus is the nano-composition of collagen and mineral particles, is highly optimized for mechanical demands. This could be observed in embryonic and adult murine bone. In all studied samples, the mineral particles started to grow in the hole zones of collagen until a thickness of 3nm (Fratzl et al., Calcif Tissue Int 48:407-413;1991). The orientation of the mineral particles differed in two different bone types according to the typical mechanical stress pattern. In the femur, the mineral particles were aligned +/-25° with the long bone axis, whereas in the calvaria no predominant particle orientation could be observed which gives evidence for the optimization for compression in the calvaria. The age-development of the material of the ulna showed clearly the development of the mechanical optimization. In the embryonic bone, we found no orientation of the mineral particles and collagen fibrils, the alignment of the particles formed after birth and reached its maximum one month later. This suggests that weight bearing plays a crucial role in the development of the bone nanocomposite material. (Fratzl et al., J Bone Miner Res 7:329-334;1992)

IN-VITRO BIOMECHANICS OF BONE CELLS, BONE TISSUE, AND COLLAGENOUS TISSUES

The activity of bone cells (and bone remodelling and resorption) is highly dependent on the mechanical forces acting on the loaded skeleton (mainly due to weight bearing). It is not only bone mass that seems to be mechanically triggered but also the structure of the bone material which is mechanically optimized. Our scientific interest in this context focuses on the mechanical stretching of bone cells in-vitro, and on the in-situ investigation of changes in the structure during mechanical straining of the mineralized turkey leg tendon.

The effect of anisotropic mechanical strain on bone cells in-vitro

The mechanical stimulus is essential for bone since it adapts to these mechanical forces. Consequently, it is the mechanical stimulus (together with hormonal and genetic influences) that determines the amount of bone remodelling or resorption. From in-vivo experiments, it is known, that the typical strain amplitudes occurring at the bone surfaces are smaller than 1000 microstrain (one microstrain is corresponding to a change in length D l relative to the original length l 0 of D l/l 0=10 -6). The largest strain amplitudes occurring under physiological conditions are 3000-4000 microstrain (the yield strain of bone is 7000 microstrain). Additionally to the strain amplitude, it is the strain rate that is playing an important role in the stimulation of the bone cells. Dynamic stimuli are more effective than a static stimulation. The number of mechanical stimuli seems to be less important. It is not clarified yet, which bone cells are able to detect the mechanical force and how these cells transfer the signal to the bone cells which are remodelling the bone material. It is assumed that the osteoblasts and the osteocytes are the mechanosensors in bone.

In this context, we have designed a stretching device which enables to study the effect of anisotropic mechanical strain on bone cells in-vitro (Grabner et al., Rev Sci Instrum 71:3522-3529;2000). This apparatus consists mainly of a linear positioning stage positioned within the CO 2-incubator and connected to a PC outside of the incubator. For stretching of the cells (we used preosteoblastic MC3T3-E1 cells), the cells had to be cultured on a polyurethane culture substrate (PUCS), which was then fixed on the moving part of the linear positioning stage and connected to the non-moving part. When the linear positioning stage was moved, the PUCS was cyclically stretched in longitudinal direction and contracted perpendicular to this direction. The strain amplitude of the PUCS and the strain distribution within a culture well on the PUCS was measured by laser-speckle-correlation with an accuracy of 24%. In our experiments, we used the strain amplitude of 6800 +/- 1600 microstrain (mean +/- SD). Further chosen mechanical parameters were: the strain rate of 4900 microstrain/s, 350 cycles per hour. After 48 hours stretching, we found a decrease in cell proliferation (-25%) and an increase in alkaline phosphatase (+200%) compared to unstretched controls. These two parameters suggest that stretching of the cells caused an accelerated differentiation from the preosteoblast to the mature osteoblast phenotype. This mechanism may play a role in the in-vivo mechanotransduction of forces to the response of bone.

The microstructure of the mineralized turkey leg tendon