Jemma G. Kerns

Towards the in vivo prediction of fragility fractures with Raman spectroscopy

Fragility fractures, those fractures which result from low level trauma, have a large and growing socio‐economic cost in countries with aging populations. Bone‐density‐based assessment techniques are vital for identifying populations that are at higher risk of fracture, but do not have high sensitivity when it comes to identifying individuals who will go on to have their first fragility fracture. We are developing Spatially Offset Raman Spectroscopy (SORS) as a tool for retrieving chemical information from bone non‐invasively in vivo. Unlike X‐ray‐based techniques SORS can retrieve chemical information from both the mineral and protein phases of the bone. This may enable better discrimination between those who will or will not go on to have a fragility fracture because both phases contribute to bones mechanical properties. In this study we analyse excised bone with Raman spectroscopy and multivariate analysis, and then attempt to look for similar Raman signals in vivo using SORS. We show in the excised work that on average, bone fragments from the necks of fractured femora are more mineralised (by 5–10%) than (cadaveric) non‐fractured controls, but the mineralisation distributions of the two cohorts are largely overlapped. In our in vivo measurements, we observe similar, but as yet statistically underpowered, differences. After the SORS data (the first SORS measurements reported of healthy and diseased human cohorts), we identify methodological developments which will be used to improve the statistical significance of future experiments and may eventually lead to more sensitive prediction of fragility fractures.

Gold nanoparticles as a substrate in bio-analytical near-infrared surface-enhanced Raman spectroscopy.

“Large” nanoparticles potentially are a good starting point in order to derive informative NIR/IR SERS analysis of biological samples.

Functional adaptation of long bone extremities involves the localized "tuning" of the cortical bone composition; evidence from Raman spectroscopy

Abstract. In long bones, the functional adaptation of shape and structure occurs along the whole length of the organ. This study explores the hypothesis that adaptation of bone composition is also site-specific and that the mineral-to-collagen ratio of bone (and, thus, its mechanical properties) varies along the organ’s length. Raman spectroscopy was used to map the chemical composition of long bones along their entire length in fine spatial resolution (1 mm), and then biochemical analysis was used to measure the mineral, collagen, water, and sulfated glycosaminoglycan content where site-specific differences were seen. The results show that the mineral-to-collagen ratio of the bone material in human tibiae varies by <5% along the mid-shaft but decreases by >10% toward the flared extremities of the bone. Comparisons with long bones from other large animals (horses, sheep, and deer) gave similar results with bone material composition changing across tens of centimeters. The composition of the bone apatite also varied with the phosphate-to-carbonate ratio decreasing toward the ends of the tibia. The data highlight the complexity of adaptive changes and raise interesting questions about the biochemical control mechanisms involved. In addition to their biological interest, the data provide timely information to researchers developing Raman spectroscopy as a noninvasive tool for measuring bone composition in vivo (particularly with regard to sampling and measurement protocol).

Evidence from Raman Spectroscopy of a Putative Link Between Inherent Bone Matrix Chemistry and Degenerative Joint Disease

Osteoarthritis (OA) is a common debilitating disease that results in degeneration of cartilage and bone in the synovial joints. Subtle changes in the molecular structure of the subchondral bone matrix occur and may be associated with cartilage changes. The aim of this study was to explore whether the abnormal molecular changes observed in the matrix of OA subchondral bone can be identified with Raman spectroscopy.

Counteracting bone fragility with human amniotic mesenchymal stem cells

The impaired maturation of bone-forming osteoblasts results in reduced bone formation and subsequent bone weakening, which leads to a number of conditions such as osteogenesis imperfecta (OI). Transplantation of human fetal mesenchymal stem cells has been proposed as skeletal anabolic therapy to enhance bone formation, but the mechanisms underlying the contribution of the donor cells to bone health are poorly understood and require further elucidation. Here, we show that intraperitoneal injection of human amniotic mesenchymal stem cells (AFSCs) into a mouse model of OI (oim mice) reduced fracture susceptibility, increased bone strength, improved bone quality and micro-architecture, normalised bone remodelling and reduced TNFα and TGFβ sigalling. Donor cells engrafted into bones and differentiated into osteoblasts but importantly, also promoted endogenous osteogenesis and the maturation of resident osteoblasts. Together, these findings identify AFSC transplantation as a countermeasure to bone fragility. These data have wider implications for bone health and fracture reduction.

Spatially offset Raman spectroscopy for photon migration investigations in long bone

Raman Spectroscopy has become an important technique for assessing the composition of excised sections of bone, and is currently being developed as an in vivo tool for transcutaneous detection of bone disease using spatially offset Raman spectroscopy (SORS). The sampling volume of the Raman technique (and thus the amount of bone material interrogated by SORS) depends on the nature of the photon scattering in the probed tissue. Bone is a complex hierarchical material and to date little is known regarding its diffuse scattering properties which are important for the development and optimization of SORS as a diagnostic tool for characterizing bone disease in vivo. SORS measurements at 830 nm excitation wavelength are carried out on stratified samples to determine the depth from which the Raman signal originates within bone tissue. The measurements are made using a 0.38 mm thin Teflon slice, to give a pronounced and defined spectral signature, inserted in between layers of stacked 0.60 mm thin equine bone slices. Comparing the stack of bone slices with and without underlying bone section below the Teflon slice illustrated that thin sections of bone can lose appreciable number of photons through the unilluminated back surface. The results show that larger SORS offsets lead to progressively larger penetration depth into the sample; different Raman spectral signatures could be retrieved through up to 3.9 mm of overlying bone material with a 7 mm offset. These findings have direct impact on potential diagnostic medical applications; for instance on the detection of bone tumors or areas of infected bone.

Is the Collagen Primed for Mineralization in Specific Regions of the Turkey Tendon? An Investigation of the Protein–Mineral Interface Using Raman Spectroscopy

The tendons in the turkey leg have specific well-defined areas which become mineralized as the animal ages and they are a thoroughly characterized model system for studying the mineralization process of bone. In this study, nondestructive Raman spectroscopic analysis was used to explore the hypothesis that regions of the turkey tendon that are associated with mineralization exhibit distinct and observable chemical modifications of the collagen prior to the onset of mineralization. The Raman spectroscopy features associated with mineralization were identified by probing (on the micrometer scale) the transition zone between mineralized and nonmineralized regions of turkey leg tendons. These features were then measured in whole tendons and identified in regions of tendon which are destined to become rapidly mineralized around 14 weeks of age. The data show there is a site-specific difference in collagen prior to the deposition of mineral, specifically the amide III band at 1270 cm(-1) increases as the collagen becomes more ordered (increased amide III:amide I ratio) in regions that become mineralized compared to collagen destined to remain nonmineralized. If this mechanism were present in materials of different mineral fraction (and thus material properties), it could provide a target for controlling mineralization in metabolic bone disease.

Millimeter-scale mapping of cortical bone reveals organ-scale heterogeneity.

Raman spectroscopy was used to show that across 10 cm of diaphyseal (mid-shaft) cortical bone the phosphate-to-amide I ratio (a measure of the mineral to collagen ratio) can vary by as much as 8%, and the phosphate-to-carbonate ratio (a measure of carbonate inclusion in mineral crystals) by as much as 5%. The data are preliminary but are important because they reveal a spatial variation at a scale that is much larger than many of the spectral maps reported in the literature to date. Thus they illustrate natural variation in chemical composition that could have been overlooked in such studies or could have appeared as an undue error where the overall composition of the bone was investigated. Quantifying the variation in mid-shaft cortical bone at the millimeter/centimeter scale reduces the possibility of natural heterogeneity obscuring the average bone composition, or being mistaken for experimental signal, and results in an improvement in the sampling accuracy analogous to that obtained by switching from micrometer-size point spectra of bones to spectral images obtained across hundreds of micrometers. Although the study was carried out using Raman spectroscopy, the underlying cause of the variation is ascribed to the variation of the chemical composition of the bone; therefore the findings have direct implications for other chemically specific analytical methods such as Fourier transform infrared spectroscopy or nuclear magnetic resonance spectroscopy.

Assessment of photon migration for subsurface probing in selected types of bone using spatially offset Raman spectroscopy

Bone diseases and disorders are a growing challenge in aging populations; so effective diagnostic and therapeutic solutions are now essential to manage the demands of healthcare sectors effectively. Spatially offset Raman spectroscopy (SORS) allows for chemically specific sub-surface probing and has a great potential to become an in vivo tool for early non-invasive detection of bone conditions. Bone is a complex hierarchical material and the volume probed by SORS is dependent on its optical properties. Understanding and taking into account the variations in diffuse scattering properties of light in various bone types is essential for the effective development and optimization of SORS as a diagnostic in vivo tool for characterizing bone disease. This study presents SORS investigations at 830 nm excitation on two specific types of bone with differing mineralization levels. Thin slices of bone from horse metacarpal cortex (0.6 mm thick) and whale bulla (1.0 mm thick) were cut and stacked on top of each other (4-7 layers with a total thickness of 4.1 mm). To investigate the depth origin of the detected Raman signal inside the bone a 0.38 mm thin Teflon slice was used as test sample and inserted in between the layers of stacked bone slices. For both types of bone it could be demonstrated that chemically specific Raman signatures different from those of normal bone can be retrieved through 3.8-4.0 mm of overlying bone material with a spatial offset of 7-8 mm. The determined penetration depths can be correlated with the mechanical and optical properties of the specimens. The findings of this study increase our understanding of SORS analysis of bone and thus have impact for medical diagnostic applications e.g. enabling the non-invasive detection of spectral changes caused by degeneration, infection or cancer deep inside the bone matrix.