focal point review DEPARTMENT

OF

KAREN ESMONDE-WHITE INTERNAL MEDICINE, RHEUMATOLOGY DIVISION, UNIVERSITY OF MICHIGAN MEDICAL SCHOOL, ANN ARBOR, MI 48109 USA

Raman Spectroscopy of Soft Musculoskeletal Tissues Tendon, ligament, and joint tissues are important in maintaining daily function. They can be affected by disease, age, and injury. Slow tissue turnover, hierarchical structure and function, and nonlinear mechanical properties present challenges to diagnosing and treating soft musculoskeletal tissues. Understanding these tissues in health, disease, and injury is important to improving pharmacologic and surgical repair outcomes. Raman spectroscopy is an important tool in the examination of soft musculoskeletal tissues. This article highlights exciting basic science and clinical/translational Raman studies of cartilage, tendon, and ligament. Index headings: Raman spectroscopy; Cartilage; Ligament; Tendon.

INTRODUCTION oft musculoskeletal tissues, including tendon, ligament, and joints, have a large affect on daily activities. However, these tissues are particularly susceptible to injury and disease. Progressive joint diseases such as osteoarthritis (OA) are major causes of disability and represent a significant socioeconomic burden as lost wages, early retirement, and increased health costs.1 Other rheumatic joint conditions, including gout and rheumatoid arthritis, are also are major causes of disability and affect approximately 4.5 million Americans.2,3 OA is the most prevalent joint disease and a leading cause of disability, affecting 27 million Americans.2,4 Additionally, damage to tendon, ligament, and joint are common in sports injuries with the knee, shoulder, and Achilles’ tendon as the sites most commonly affected.5 The study of musculoskeletal tissues in health and disease is interdisciplinary, requiring the skills of clinicians, engineers, molecular biologists, and chemists. Martin’s book Skeletal Tissue Mechanics provides an excellent introduction to tendon, ligament, and cartilage compositional and mechanical properties.6 Tendon connects bone to muscle and provides resistance to strain. Tendon is a highly dense tissue composed mostly of type I collagen (~85%) with a minor proteoglycan component (1–5%).7 Ligaments connect bone to bone and are also primarily type I collagen with a small component (~3%) of proteoglycans.8 Tendon’s hierarchical structure is shown in Fig. 1, where multiple collagen molecules assemble into fibrils and multiple fibrils assemble into fibers. Collagen fibers are highly oriented, parallel to the long axis of the tendon tissue, with a ‘‘crimp’’ observed at regular intervals. The presence of this crimp is an important mechanical adaptation in tendon, especially at physiological loading

S

Received 12 May 2014; accepted 11 August 2014. E-mail: [email protected]. DOI: 10.1366/14-07592

where the crimp begins to straighten in response to tension. Collagen fibril size and orientation are important contributors to the nonlinear elasticity and viscoelasticity of tendon. A force displacement curve for tendon is shown in Fig. 2, showing the nonlinear response to tension.6 In some reports, the stress-strain curve is presented, and it has a similar shape to the force displacement curve. Physiological loading occurs mostly in the toe region and extends slightly into the linear region, marked by dashed lines in Fig. 2. Crimp straightening is the initial molecular response that occurs in the toe region of low strain, and the collagen chain elongation occurs in the linear region of higher strain. Stress relaxation, hysteresis, and creep are viscoelastic properties observed in tendon. Synovial joints, or diarthrodial joints, are movable joints such as the knee or shoulder. Synovial joints are made up of bone, articular cartilage, meniscus, ligaments, and synovial fluid. Netter’s Clinical Anatomy has excellent anatomy drawings, including several of the knee joint.9 The joint is encompassed by a synovial membrane and surrounded by intra-articular fat pads. Articular cartilage is a multilayered viscoelastic tissue that covers the ends of long bones, and it serves as a load-bearing, lubricative surface in articular joints. The extracellular matrix of articular cartilage consists of four main components: water, type II collagen, the proteoglycan aggrecan, typically bound with a sulfated glycosaminoglycan (GAG), and hyaluronic acid, a non-sulfated GAG.10 Similar to type I collagen, type II collagen assembles into a fiber structure consisting of multiple fibrils. Similar to tendon, cartilage has compositional and molecular structure adaptations at the micro- and tissue level. Collagen orientation at each layer of cartilage, hyaluronic acid entanglement, water permeability, and distribution of chondrocytes are examples of these adaptations. For example, collagen is oriented parallel to the articular surface in the superficial zone, randomly in the middle zone, and perpendicular in the deep zone. As an avascular tissue, cartilage relies on diffusion of nutrients through the synovium and thus has

APPLIED SPECTROSCOPY OA

1203

focal point review

FIG. 1. The multiscale anatomic and composition of tendon. [Reproduced with permission from Ref. 19. Copyright 2011 American Chemical Society.]

a low turnover rate. This low turnover rate is implicated in the slow repair of cartilage and resistance to treatments. Comprehensive models consider the extensive interactions between biological, compositional, and mechanical properties on multiple size scales. Elegant studies of the hierarchical biological, mechanical, and structural properties of bone provide insight into bone strength,

and these principles can be extended to analysis of other musculoskeletal tissues.11–15 Extension of these paradigms to soft musculoskeletal tissues can be seen in basic science OA research. New models of OA and other joint diseases consider a model wherein the joint is considered to be a self-contained functional musculoskeletal organ composed of subchondral bone, articular

FIG. 2. Force-displacement curve of tendon collagen. Modified from original figure in Ref. 6.

1204

Volume 68, Number 11, 2014

cartilage, menisci, ligaments, synovium, and intraarticular fat.16 Musculoskeletal tissues have exquisite spatial heterogeneity of compositional, mechanical, and biological properties that can be observed on the micro-level up to the tissue level. Interestingly, micro-level heterogeneity does not always predict tissue-level heterogeneity. Traditional methods of measuring tissue composition typically required tissue homogenization and could not inform on spatial heterogeneity. Thus, there is a need for analytical techniques that can characterize not only individual tissues, but also the tissue interfaces. Ideally, new analytical approaches would be minimally invasive and non-destructive, enabling multiple analyses on the same tissue, and provide highly specific information at the required spatial resolution. Improved understanding of tissue composition and intermolecular interactions could inform on repair strategies and guide development of engineered tissues that mimic normal tissue function. Vibrational spectroscopy, infrared and Raman, can make important contributions in addressing these issues. The high spatial resolution, non-destructive nature, and specificity of Raman microspectroscopy can capture tissue compositional heterogeneity at the molecular and tissue size scales. Raman mapping and imaging provides a high spatial resolution map of compositional tissue properties, such as collagen secondary structure and fibril orientation, which cannot be obtained using standard molecular biology techniques. Translation of Raman spectroscopy into the clinical setting, using fiberoptic instrumentation, can potentially provide tissuelevel heterogeneity measurements and complement established clinical imaging modalities. A combined Raman microspectroscopy and fiber-optic Raman approach may demonstrate relationships between microlevel properties and tissue-level properties. This review highlights recent applications in Raman spectroscopy of soft musculoskeletal tissues, with the acknowledgment that this field has a basis in previous infrared research. Basic science and translational studies are reviewed. Finally, I daydream about exciting future possibilities and provide perspective on the challenges facing translation of Raman spectroscopy into clinical use.

INSTRUMENTATION Raman spectroscopy, as microscopy or fiber-optic instrumentation, has been used to examine soft musculoskeletal tissues. A recent review of Raman spectroscopy in biomineralization extensively reviews Raman instrumentation and includes a discussion of sampling volume and spectral interpretation of bone collagen bands.17 These concepts and instrumentation can be applied toward tendon, ligament, and cartilage. An introduction to spatially offset Raman spectroscopy by Macleod and Matousek18 reviews the concepts and instrumentation that enables deep tissue Raman spectroscopy. The recent paper by Masic et al.19 also provides the theoretical basis of polarized Raman

spectroscopy. For an introduction to coherent antiStokes Raman spectroscopy (CARS), the reader is directed to the seminal 1999 paper by Zumbusch et al. and a 2004 review by Cheng and Xie.20,21 Fruediger provides an excellent introduction to the principles and instrumentation of stimulated Raman scattering (SRS).22 Data Interpretation. Collagen structure and crosslinking in mineralized and unmineralized tissues are affected by the local physiochemical environment, age, disease, and post-translational signaling. Collagen was deemed ‘‘a most difficult system’’ by Frushour and Koenig, and interpretation of collagen Raman and infrared spectra remains a vibrant research topic.23 Table I provides band assignments for tendon, cartilage, and ligament spectra. Collagen, as type I in tendon, ligament, and bone, or type II in cartilage, is the dominant molecular species in musculoskeletal tissues. As such, collagen bands are consistently found in these tissues. Spectra of type I and type II collagen overlap significantly and, in most practical circumstances, cannot be distinguished by their Raman spectrum. Tendon and ligament spectra are dominated by type I collagen bands. In addition to type II collagen bands, cartilage spectra also contain a band ~1063 cm-1 corresponding to the sulfated glycosaminoglycan chondroitin sulfate. Although there are certainly non-collagenous proteins, glycosaminoglycans, and lipids in tissue, their contributions in spontaneous Raman spectra are weak and generally overlap with collagen bands. However, lipid components can be detected using SRS or CARS. Important spectral features include the amide I envelope centered at ~1665 cm-1, the CH2/CH3 stretch at ~1448 cm-1, amide III envelope at 1220–1280 cm-1, and proline/hydroxyproline C-C at 820–880 cm-1. Band assignments and their interpretation, particularly for the amide envelopes, are based on Raman studies of collagen, poly-proline, poly-hydroxyproline, glycosaminoglycans, and tissues.24,25,23,26–29 Early biophysical Raman studies of polypeptides complemented other biophysical analyses of these materials, such as X-ray diffraction, circular dichroism, and infrared spectroscopy. These studies provided the first data on collagen secondary structure and macrostructure of the collagen fiber. Acid/base denaturation or deuterium exchange were later added to the suite of biophysical Raman tools, and this approach was employed by Frushour and Koenig23 to demonstrate that collagen ~1248 and 1271 cm-1 bands arose from amide III vibrations and not amino acid side chains. Interestingly, the Frushour and Koenig paper focuses its analysis on the amide III envelope with little discussion about the amide I envelope. Although not implicitly mentioned in the paper, a close inspection of the gelatin Raman spectra used to characterize the 1248 and 1271 cm-1 bands may be useful in also interpreting the source of bands in the amide I envelope. Gelatin amide I and amide III bands disappear almost entirely under acid or base denaturation, indicating that bands in

APPLIED SPECTROSCOPY OA

1205

focal point review TABLE I. Raman band assignments for tendon, ligament, and cartilage are based on literature reports of collagen, amino acids, glycosaminoglycans, and tissues.25,23,27,58 Raman shift (cm-1) 822 850 875 938 1002 1063 1220–1280 1448 1620–1700

Assignment

Macromolecule

Tendon

Ligament

Cartilage

mC-C mC-C mC-C C-C stretch, a-helix Ring breathing OSO3-, symmetric stretch Amide III CH2/CH3 Amide I

Collagen Hydroxyproline Hydroxyproline Collagen Protein Chondroitin sulfate Collagen amide Organic content Collagen amide C=O

X X X X X

X X X X X

X X X

X X X

X X X X X X X X X

these regions do not arise from side chains. Raman spectra of deuterated gelatin shows that the 1248 and 1271 cm-1 band positions were unaffected by deuteration, suggesting that these bands do not arise from amino acid side chains. However, the band maximum shifts from 1670 to 1664 cm-1 in the amide I envelope and suggests contributions from side chains. Given these conflicting data, how are we to interpret the amide I envelope? I would agree with Frushour and Koenig, who warned against a simplistic interpretation of the amide I envelope and acknowledge that amide I bands may arise not only from the polypeptide backbone but also from side chains. Continuing research in the interpretation of amide I and amide III, for infrared and Raman spectroscopy, is a vibrant topic of study.30–33 In particular, there is robust discussion in the infrared literature regarding interpretation of the amide I envelope as a spectroscopic marker of collagen cross-linking, with possible implications in Raman spectral interpretation. The Raman community’s understanding and use of the amide I envelope as a spectroscopic marker of collagen cross-linking has historically relied upon the infrared literature of collagen crosslinking, mostly how it pertains to bone. In 2001 Paschalis et al. reported infrared spectroscopic characterization of mature pyridinoline (Pyr, at ~1660 cm-1) and immature divalent (at 1690 cm-1) cross-links in bone and other collagen-based materials.34 Other infrared studies by independent groups, reviewed in 2011, confirm the original interpretation.35 Two papers published in 2011 reported on infrared measures of bone collagen cross-linking using the lathyrism rat model. In the lathyrism rat model, exogenous delivery of b-aminoproprionitrile (bAPN) inhibits lysyl oxidate-mediated crosslink formation. The 2011 study by Paschalis et al. examined the resulting cross-links by infrared spectroscopy and biochemical analysis because their earlier studies had found the altered cross-links ratio was confined to newly formed bone while the rest of the bone appeared unaffected.36 Infrared measure of collagen cross-linking was described using a ‘‘collagen maturity’’ metric, as measured by the 1660/1690 cm-1 area ratio. Cross-links were also measured by biochemical tests that involved tissue digestion and fractionation of cross-linked amino acids

1206

Volume 68, Number 11, 2014

by partition chromatography. Immature cross-links were assayed by ion-exchange chromatography and mature cross-links quantified by reverse-phase, high-performance liquid chromatography with fluorescence detection. bAPN treatment resulted in a statistically significant (P , 0.05) increase in the ratio of Pyr/divalent crosslinks, as measured by both biochemical and spectroscopically. Infrared measures at the periosteal and newly formed bone surfaces also revealed statistically significant decreases in collagen maturity, which appears to be primarily driven by a disproportionate decrease in the 1690 cm-1 area. A 2011 study by Farlay et al.37 also examined bone collagen crosslinks in a rat model of lathyrism using infrared spectroscopy and chromatography/mass spectrometry analysis. However, they concluded that the 1660/1690 cm-1 ratio was not related to collagen cross-links. It is important to note differences in experimental approaches that may have produced the apparently incongruous results: bAPN dose and duration, anatomic site examined, age of bone examined, biochemical method sensitivity, and study design. Neither study published an a priori or post hoc power analysis, even if one was performed as part of the study, so it is difficult to assess if non-statistical significance was the result of a small study cohort. In the 2011 PLoS ONE study,37 five lathyritic and five normal radius specimens (out of 20 total) were examined by infrared, and perhaps a follow-up study would achieve statistical significance if the entire cohort of specimens were examined by infrared. There is ongoing Raman work to validate the historical interpretation of amide I for collagen cross-linking, and two exciting conference abstracts deserve mention. A recent abstract by Gamsjaeger et al. provides the first experimental validation studies by reporting Raman identification of trivalent collagen cross-links.38 An abstract by McNerny et al.39 provides the first Raman results of assessing bone collagen maturity in bAPNtreated mice and found that bAPN treatment increased the 1660/1683 cm-1 ratio. Calcein labeling of the examined bone enabled identification of newly formed sites in the bone tissue, and Raman results indicated lower collagen maturity at new tissue sites when compared to existing bone tissue.39 Combination of this

FIG. 3. Raman spectrum of mouse tendon collected by the author using Raman microspectroscopy instrumentation (k = 785 nm).

technique with biochemical analysis would provide compelling validation of Raman spectroscopy for measuring collagen cross-linking. Band intensity ratios, band width, and band position provide input for univariate analysis. The most commonly used ratios for analysis of collagen secondary structure are the amide I components, 1680 cm-1/1660 cm-1, and amide III components, 1270 cm-1/1240 cm-1. For mineralized tendon, Raman metrics include tissue mineralization (PO4-3m1/matrix bands) where the matrix marker can be the amide I envelope, the CH2/CH3 envelope, or the proline bands at 850 and 875 cm-1, mineral stoichiometry (CO3-2m1/PO4-3m1, 1070 cm-1/958 cm-1), and mineral crystallinity (width of PO4-3m1 at 958 cm-1). Multivariate analysis treatment can produce a more sophisticated analysis, and there are myriad reports of techniques such as principal components analysis, least-squares regression, and linear discriminate analysis in analyzing Raman spectra and images. Shaver presents an overview of chemometrics for Raman spectroscopy in the 2001 Handbook of Raman Spectroscopy: From the Research Laboratory to the Process Line.40 Specific applications in biomedical Raman spectroscopy are reviewed by Reddy and Bhargava in the 2010 book Emerging Raman Applications and Techniques in Biomedical and Pharmaceutical Fields.41

APPLICATIONS A recent review of biomedical Raman microscopy focused on experimental advances in spontaneous Raman, surface-enhanced (SERS) and tip-enhanced (TERS) Raman, CARS, and SRS.42 Intracellular biochemical composition, nanoparticle tracking, and whole cell

imaging were highlighted applications. The literature in tendon, ligament, and cartilage spans the molecular level to fiber-level to tissue-level analyses. Additional Raman studies of synovial (joint) fluid complement joint tissue studies. Raman Spectroscopy Applications in Tendon and Ligament. The Raman spectrum of tendon, shown in Fig. 3, is dominated by collagen bands. Tendon is an excellent source of type I collagen, and fibers can be easily harvested. Thus, most Raman collagen studies have been performed using tendon collagen fibers. Collagen composition and orientation in tendon are rich targets for Raman spectroscopic analysis. Collagen composition can be observed by normal Raman spectroscopy. Additional orientation metrics can be obtained by polarized Raman microscopy. Studying tendon collagen under strain provides dynamic insight into tendon’s hierarchical molecular adaptations to mechanical strain. For this very reason, most of the modern Raman studies of tendon collagen incorporate a Raman analysis into mechanical loading tests. The first Raman report of tendon was by Frushour and Koenig in 1975.23 They compared the spectrum of bovine Achilles tendon collagen, skin collagen (solid and solution state), and gelatin and constituent amino acids. In addition to producing band assignments that are considered to be the gold standard in Raman collagen studies, the authors hypothesized about the orientation effects on the spectrum. They underscored the difficulties in studying collagen, especially in interpreting the amide III and amide I envelopes. The authors warned against a straightforward interpretation of the amide envelopes as corresponding to polar and non-polar

APPLIED SPECTROSCOPY OA

1207

focal point review regions in the protein chain, even though this interpretation can be applied to globular protein spectra. This original work laid the groundwork for modern Raman studies of tendon collagen in strain, orientation effects of amide envelopes, and tendon collagen orientation. Wang et al. reported Raman microscopy of rat tail tendon (RTT) in 2000. RTT specimens were affixed to a specially designed stage equipped with a fixed jaw and a moving jaw. Specimens were elongated by micrometer. After each elongation, three to five Raman spectra were collected. Strain was estimated by the relative separation of the fixed and moving jaws. Raman band position and full-width half maximum (FWHM) of the 822 and 879 cm-1 bands were plotted against applied strain (reported in % strain). The FWHM and position 822 cm-1 band, corresponding to collagen backbone C–C, decreased with applied strain. The FWHM and position of the 879 cm-1 band, from side chain carbonyl groups, increased with applied strain. The data suggested that collagen side chain carbonyl groups were compressed and the collagen backbone deformed under tension. Polarized Raman spectroscopy for studying collagen orientation and anisotropy in tendon emerged in 2010.19,33,43,44 Janko et al. established that polarized Raman spectroscopy can be used to measure the orientation of tendon collagen.43 Amide I C=O groups showed a strong polarization dependence, while the amide NH2 (1251 cm-1) and CH2/CH3 groups (1451 cm-1) did not. The anisotropy of the amide I ~1668 cm-1 and amide III ~1271 cm-1 bands indicated that amide groups are oriented along the long axis of the fibril. Bonifacio et al. showed a polarization dependence on amide III band intensities and suggested caution when interpreting the amide III envelope because it may not exclusively reflect collagen secondary structure.33 Masic et al. extended Janko’s work to the study of multiscale strain-dependent collagen orientation. 19 They examined rat tail tendon using high magnification (603, 1.0 NA) Raman microscopy and in situ under strain. For the in situ studies, tendon specimens were placed in a chamber with controlled humidity and temperature, and a custom microtensile tester 50N load cell produced a displacement of 1 lm/sec. A lower magnification objective (203, 0.4 NA) was used for in situ stress-strain Raman spectroscopy. Stress was measured in the cross-sectional area of tendon using confocal microscopy. Chemical images and Raman orientation maps were generated. Figure 4 shows example orientation maps around the fibril crimp. On the molecular level, the collagen orientation is similar in toe and heel regions of the stress-strain curve, but it changes in the linear region of the stress-strain curve. At the tissue level, the fibril crimp straightens by the heel region, and the chain continues to elongate in the linear region. Follow-up theoretical work and polarized Raman spectroscopy of model peptides confirmed the original findings in tendon.43

1208

Volume 68, Number 11, 2014

The shoulder is an anatomic site at high risk for tendonitis and injury. Older models describe the bone-totendon interface, also called the tendon enthesis, as four distinct zones: tendon, fibrocartilage, mineralized fibrocartilage, and bone. However, mechanical studies of tendon do not show failure at any particular zone and indicate a gradient of functional properties. Raman investigation of the enthesis demonstrates a compositional gradient of collagen and mineral properties, supporting newer models of how tendon attaches to bone.45–48 Raman spectroscopy of the supraspinatus tendon enthesis identified a mineralization gradient across the tendon-bone insertion site, with a gradient in mineral crystallinity.45 The authors hypothesized that tendon mineralization and the mineral crystallinity may help to understand the mechanical properties of tendon in health. In a later study, Monte Carlo and finite element models were developed to understand the relationship of mineral volume and mineral orientation in the collagen fiber on tendon stiffness and hierarchical fiber-to-tissue mechanics. Theoretical work complemented Raman microscopy and polarized light microscopy on the supraspinatus enthesis of the rat shoulder.46 Longitudinal studies of developing tendon enthesis by Raman, microcomputed tomography, histopathology, and transmission electron microscopy confirmed a mineralization gradient and suggested that enthesis development is linked to endochondral bone maturation near the insertion site.47 A significant effect of muscle loading on enthesis mineral stoichiometry, mineral crystallinity, and collagen orientation was later reported.48 Figure 5 shows that muscle unloading affects the histological and compositional properties of the enthesis. Raman characterization of functional adaptations has been extended to the mandible enthesis.49 Identification of accumulated collagen fragments (hyalinization) in degenerated human tendon was reported by Penteado et al.50 The degree of tendon hyalinization increases with degeneration. Raman examination of human supraspinatus tendon revealed increased intensity of collagen amino acids bands and amide I envelope broadening associated with tendon hyalinization. PCA could not adequately perform a class separation, but dendrogram analysis could separate normal from hyalinized tendon. Thus, Raman may be able to provide a ‘‘yes/no’’ answer for tendon hyalinization. Normal Raman and SERS were used to examine alterations to collagen structure in the supraspinatus tendon of the rotator cuff that results from shock wave treatment.51 Shock wave treatment of tendon has been shown to enhance normal healing, but the biochemical effects on the tissue are poorly understood. Rat and bovine tail tendon were subjected to shock wave treatment. A silver colloid suspension was added directly to the tissue specimens for SERS. Control SERS studies were performed on alanine, glycine, and proline. Normal Raman tissue spectra contained significant fluorescence background, precluding comparison be-

FIG. 4. Polarized Raman mapping of the collagen fiber orientation in unstretched, fully hydrated rat tail tendon. (A) Optical microscopy image of the analyzed region where the crimp structure of collagen is visible. (B) Map obtained by fitting 13 Raman images collected with different polarization angles of the incident laser light. The direction of arrows indicates the orientation of collagen fibers, their length represents the amplitude of the fitting curve, and the color code represents the average intensity of the amide I band (parameter a). (C–E) Magnified regions of interest reveal specific structural characteristics at the level of tissue. Collagen fiber orientation changes in (D) corresponding to the crimp (at about 50 lm). (F) Frequency plot of the c parameter with respect to the rat tail fiber, direction z. [Reproduced with permission from Ref. 19. Copyright 2013 American Chemical Society.]

tween unenhanced and SERS spectra. Tissue SERS spectra showed changes related to the amino acids of collagen. Ligaments have only recently been examined by vibrational spectroscopy techniques, but I anticipate that they will be rewarding research topics for both the FT-IR and Raman research communities. As a potential tool during ligament repair surgeries or monitoring regenerated tissue quality, Raman or FT-IR spectra provide a noninvasive measurement of ligament composition in normal conditions or under tension. Raman has a possible application in the surgery suite to help surgeons set the proper ligament tension, thus stabilizing the knee joint and ensuring long-term success of anterior cruciate ligament (ACL) repair. A preliminary study by Winchester et al. indicates that unpolarized Raman spectroscopy can

monitor tensile and compressive forces in the collagen chain.52 They used a custom-built Raman system with free-standing optics to examine mink and rabbit ligament. Because the composition of ligament is similar to that of tendon, its Raman spectrum is similar to a tendon spectrum. Raman bands in the ligament spectrum corresponding to the C-C stretch of the collagen backbone (822 cm-1), hydroxyproline (879 cm-1), amide I (1672 cm-1), and CH2/CH3 bending (1417 cm-1) were examined. The authors found that the 822 and 1417 cm-1 band position decreased with applied tension, indicating strain on the collagen backbone. The 879 and 1672 cm-1 band position increased with applied tension, indicating side chain compression. These preliminary results are encouraging because they confirm earlier results by Wang et al.49 on tendon and extended Raman spectroscopy for ligaments.

APPLIED SPECTROSCOPY OA

1209

focal point review

FIG. 5. Raman microprobe analysis of mineralized fibrocartilage within the enthesis of Botox-treated shoulders demonstrate the effects of muscle loading on enthesis formation. (A–C) Representative von Kossa (mineral) and toluidine blue (cells and organic material) staining of mineralized enthesis sections show decreased cellularity and mineralization in specimens with reduced muscle loading. (D) Tissue mineralization, determined by the ratio of the phosphate mineral 960 cm-1 peak to the phenylalanine matrix 1002 cm-1 peak, was not significantly changed by muscle unloading. (E) However, carbonate substitution into the apatite matrix increased at P56 by muscle unloading specimens relative to normal controls and with animal age. (F) Crystallinity of the mineral was decreased (i.e., FWHM was increased) at P56 in muscle unloaded specimens compared to loaded controls (*P , 0.05; scale bar = 50 lm). [Reproduced with permission from Ref. 48. Copyright 2012 Elsevier.]

Unfortunately, there are no literature reports of follow-up studies by the Winchester group. There is a recent FT-IR imaging paper on the ligament-bone interface by Spalazzi et al., and I look forward to more developments in this field.53 Raman Spectroscopy Applications in Joint Tissues: Cartilage, Meniscus, Subchondral Bone, and Synovial Fluid. Infrared and Raman spectroscopies have improved our understanding of compositional and molecular structure properties of cartilage and other joint tissues. Early FT-IR studies have laid the groundwork in experimental approach, data analysis, and technology development. The readers are directed to a review of the FT-IR work, which covers the literature up to 2008.54 There are new FT-IR applications in building

1210

Volume 68, Number 11, 2014

multivariate data models, fiber-optic probes, understanding cartilage loss in animal models, and measuring cartilage compressive stiffness.55–57 Raman studies have been performed in calcified cartilage and subchondral bone, in healthy or animal OA models, since 2006.58–63 There are also Raman applications in monitoring tissue quality in engineered cartilage tissue and cartilage tissue reshaping procedures.64–67 We focus on Raman analysis of unmineralized joint tissues and synovial fluid and include recent studies using multi-photon Raman techniques, including CARS and SRS, which provide new details on tissue composition and molecular structure of cartilage extracellular matrix molecules. The Raman spectrum of cartilage, shown in Fig. 6, bears strong resemblance to tendon and ligament spectra, owing

FIG. 6. Raman spectrum of human cartilage collected by the author using Raman microspectroscopy instrumentation (k = 785 nm).

to the type II collagen. A band at ~1063 cm-1, arising from the sulfate group in chondroitin sulfate, is unique to cartilage and distinguishes its spectrum from the spectrum of other soft tissues. The effects of collagen orientation are known in FT-IR spectra and polarized light microscopy but have not yet been studied by polarized Raman spectroscopy.68,69 Preparation of cartilage tissue requires attention to detail. We, and others, have noted that the Raman spectra of articular surfaces are dominated by bone signal, even at sites with visually intact cartilage.58,59,62 The thin cartilage layer, cartilage optical scattering, and strong Raman bone signal may prevent collection of a pure cartilage spectrum. The layer of cartilage is quite thin, 1–2 mm in the lower joints, and varies with anatomic site, age, and disease.70 The optical properties of cartilage are dominated by optical scattering in the forward direction, and photons are scattered very efficiently from cartilage to bone.59,71 Bone mineral is a strong Raman scatter, and the bone collagen signal strongly overlaps with cartilage collagen. Aside from the 1063 cm-1 band from cartilage GAGs, there is little to spectroscopically distinguish cartilage from bone. In semi-confocal microscopy and fiber-optic instrument configurations, we have found isolating cartilage Raman signal in spectra from intact articular surfaces challenging. For our experiments, we assumed spectral interferences from bone if we observed a moderate-to-strong band at ~958 cm-1 corresponding to m1 of bone mineral phosphate. Even though an overlying cartilage layer does not significantly affect collection of subchondral bone signal, overlapping cartilage collagen bands may

complicate spectral interpretation of the amide envelopes. Excising cartilage from the joint enables cartilage Raman spectroscopy without interference from the underlying bone. The excised tissue can then be fixed, embedded, and sectioned or cryosectioned to enable analysis in the individual cartilage zones. If fresh tissue is examined, care must be taken to ensure tissue hydration. Raman mapping of articular cartilage zones was demonstrated in 2010 by Bonifacio et al.69 Excised porcine cartilage was fixed and embedded, and 100 lm sections were cut perpendicular to the articular surface. Resulting Raman spectra were processed by a variety of univariate and multivariate tools including principle components analysis, partial least squares regression (PLSR), and cluster analysis. A reference dataset of 60 pure component spectra included 20 polarized Raman spectra of collagen. Raman intensity maps corresponding to DNA, chondroitin sulfate, collagen, and noncollagenous proteins showed high DNA levels in the superficial and middle zones, and ubiquitous collagen and dense chondroitin sulfate in the deep zone. Noncollagenous proteins were found most concentrated in pericellular and territorial regions with low collagen. PLSR provided reasonable quantification of components, and the model may be improved by including more pure component spectra of other glycosaminoglycans and types of collagen. Interestingly, the group found extracellular calcium carbonate in the middle zone of cartilage. In an elegant polarized Raman study, Lim et al.72 examined biochemical, histological, and structural alterations of porcine cartilage subject to impact loads of 15, 20, and 25 MPa. Preliminary work in human

APPLIED SPECTROSCOPY OA

1211

focal point review osteoarthritic knee meniscus by Papaspyridakou et al. showed apatite and calcite at arthritic sites, and identification of these inorganic minerals may be another spectroscopic marker of diseased tissue.73 Mansfield et al. demonstrated multi-modal spectroscopy and imaging of cartilage using two photon fluorescence, CARS, SRS, and second-harmonic generation imaging.74 Excised fresh full-thickness equine cartilage specimens were placed between two coverslips and hydrated with saline. Imaging and microscopy examination of the specimen was carried out using custom-built and commercially available instrumentation. CARS images of chondrocytes reveal the presence of lipid in pericellular zones, with lipid droplets mostly comprised of unsaturated fatty acids. SRS images of mineral components in calcified cartilage and bone reveal a secondary tidemark in calcified cartilage. Figure 7 shows SRS, two-photon fluorescence, and second-harmonic generation images of calcified cartilage and bone. Synovial fluid is an important component in the articular joint.75 Synovial fluid exhibits shear-dependent viscoelastic properties, owing to the entanglement properties of hyaluronic acid, provides a lubricative surface for cartilage, and imparts resistance to compressive forces.76 Synovial fluid undergoes extensive chemical changes in early-stage joint disease including increased concentration of inflammatory cytokines, depolymerized hyaluronic acid, reduced fluid pH, inorganic crystals, and collagen fragments. Some of these chemical changes manifest in altered fluid properties including rheology and visual features such as color and clarity.76,77 Joint swelling and pain are typically the first symptoms of joint degeneration. Joint aspiration is a minimally invasive treatment for arthritic conditions, involving removal of diseased synovial fluid followed by intra-articular injection of a viscosupplement or steroid.78 Because joint aspiration is routinely performed in patients presenting with swollen and painful joints, the aspirated synovial fluid is a convenient source for analyses and may provide the first indications of arthritic conditions. Biochemical synovial fluid markers in arthritic conditions, including osteoarthritis, rheumatoid arthritis, gout, and pseudo-gout, are topics of intense research. Two approaches are reported in the literature. The first approach is a proteomic analysis where the entire protein profile is characterized by mass spectrometry.79–81 The Raman spectrum of synovial fluid may be considered a molecular fingerprint of the proteome, even if individual proteins may not be identified. A combined drop deposition/Raman spectroscopy technique was used to examine synovial fluid from 40 patients requiring knee joint aspiration.82 Approximately half of the patients also had radiological evidence of joint degeneration. The Raman spectra of synovial fluid proteins in patients with radiological evidence of joint degeneration had a decreased abundance of ordered a-helical secondary

1212

Volume 68, Number 11, 2014

structure and unsupervised k-means clustering analysis of band intensity ratios resulted in a classification with 74% sensitivity and 71% selectivity. A later analysis of the data was performed by multivariate techniques (K. Esmonde-White, unpublished data). Weightings from three selected principle components were input into a leave-one-out cross-validated linear discriminate analysis to predict the positive/negative score from each individual spectrum’s principal components weightings using the known positive/negative scores for the training data of the other N-1 spectra. The resulting 199 predicted positive/negative scores, corresponded to several spectra per patient. A majority-vote algorithm was used to classify the patient positive/negative score. The multivariate analysis resulted in a classification with 91% sensitivity and 64% selectivity. Another approach is to identify a single biomarker or a group of biomarkers.77,83,84 In addition to specific cytokines or protein-like markers, inorganic crystals are intensely investigated because they are chemically unique biomarkers. The sensitivity of Raman spectroscopy for inorganic minerals and crystals can provide highly specific analyses of pathological crystals associated with crystal deposition diseases. Crystal deposition diseases such as gout, calcium pyrophosphate deposition (pseudogout), or Milwaukee shoulder syndrome are characterized by the presence of pathological inorganic crystals in joint tissue or synovial fluid. Possible crystals include monosodium urate (MSUM) in gout, calcium pyrophosphate (CPPD) in pseudo-gout, and basic calcium phosphate (such as apatite) in osteoarthritis or rheumatoid arthritis. If a crystal deposition disease is suspected clinically, aspirated synovial fluid will be examined under polarized light microscopy (PLM) for the presence of crystals. Crystal features such as shape or birefringence are used to identify the type of crystals present. Despite it being the gold standard for synovial fluid crystal analysis, PLM has poor sensitivity because it is affected by user expertise, low crystal concentrations, and the presence of non-pathological crystals such as cholesterol. Other analytical techniques including atomic force microscopy have been used, and Yavorskyy et al. reviewed the analytical technology for synovial fluid crystal analysis.85 Raman spectroscopy may improve synovial fluid analysis because the spectral specificity enables unambiguous crystal identification. In 1991, the first report of Raman identification of pathological crystals was reported by McGill.86 Akkus et al.87 has produced elegant work translating a ‘‘point-and-shoot’’ Raman instrument for synovial fluid crystal analysis. In their first publication, they described a sample preparation method to isolate crystals from synovial fluid and characterized the Raman limits of detection for MSUM and CPPD. They also obtained human synovial fluid from 35 patients and examined the fluid for MSUM or CPPD by Raman spectroscopy and polarized light microscopy.87 This feasibility study demonstrated MSUM and CPPD detection at clinically relevant levels, and the Raman results

FIG. 7. Stimulated Raman spectroscopy (SRS) images of mineralized cartilage and subchondral bone. (A, B) SRS imaging of the carbonate and phosphate mineral components reveal secondary tidemarks within calcified cartilage and suggest that there is little difference in the mineral composition in the calcified cartilage and subchondral bone. (C) SRS imaging of lipids (CH2) shows a uniformity in lipid distribution in the calcified cartilage, but highly localized concentrations within bone channels. (D) A merged image of the tissue, composed of the mineral and matrix components in panels A–C. (E, F) Two-photon fluorescence (TPF) and second-harmonic generation (SGH) imaging clearly delineates the boundary of calcified cartilage and subchondral bone. [Reproduced with permission from Ref. 74. Copyright 2013 Wiley-VCH Verlag GmbH & Co. KGaA.]

APPLIED SPECTROSCOPY OA

1213

focal point review

FIG. 8. A custom-design and built microfiltration cartridge enables removal of organic material in synovial fluid, enabling sensitive Raman detection of inorganic crystals. [Reproduced from Ref. 88. Copyright 2014 Royal Society of Chemistry.]

agreed with PLM in 32 out of 35 cases. The authors warned that the results show a higher degree of agreement in normal synovial fluid than diseased synovial fluid. Raman and PLM both indicated ‘‘normal’’ in the 25/25 patients with normal synovial fluid. However, the Raman and PLM results agreed in only 7/10 patients with diseased synovial fluid, with the Raman analysis generating one false positive and one false negative. Detailed description of a rapid, low-cost, point-of-care Raman instrument for synovial fluid analysis was reported in 2014.88 The authors described their approach, which includes enzymatic digestion of hyaluronic acid and other organic material, filtration, and Raman spectroscopy. A customized microfiltration system was designed to concentrate and isolate crystals in a defined volume. The microfiltration cartridge, Fig. 8, is loaded into a small-footprint (~25.4 3 30.48 cm) custombuilt cost-efficient Raman device (CARD), shown in Fig. 8. The CARD is capable of MSUM and CPPD analysis at sub-clinical concentrations. Representative Raman data are shown in Fig. 9 for a pathological synovial fluid sample containing MSUM. Chemical mapping of MSUM using the CARD identified locations with high, medium, and low levels of MSUM. CARD-derived maps were confirmed by scanning electron microscopy. Follow-up

1214

Volume 68, Number 11, 2014

work optimizing the wavelength and photobleaching protocols demonstrated that 785 nm is well suited for synovial fluid measurements.89

CONCLUSIONS AND FUTURE DIRECTIONS Tendon, ligament, and joint tissues are important in maintaining daily function. They can be affected by disease, age, and injury. Slow tissue turnover, hierarchical structure and function, and nonlinear mechanical properties present challenges to diagnosing and treating soft musculoskeletal tissues. Understanding these tissues in health, disease, and injury is important to improving pharmacologic and surgical repair outcomes. Raman spectroscopy is an important tool in the examination of soft musculoskeletal tissues. This article highlighted exciting basic science and clinical/translational Raman studies in cartilage, tendon, and ligament. Composition, molecular structure, and orientation information provided by Raman spectra complement histological and mechanical studies. There are many exciting future directions in this field that will improve our fundamental understanding of these dynamic tissues, characterize biomimetic materials, and identify damaged tissue in a clinical environment. From a basic science perspective, the principles

FIG. 9. (A) Microscopy image of materials deposited onto the microfilter (B) and (C) scanning electron microscopy images of the microfilter confirm the presence of monosodium urate crystals. (D, E) Raman maps of the same microfilter reveal sections of high, medium, and low concentration of monosodium urate spectra. [Reproduced from Ref. 88. Copyright 2014 Royal Society of Chemistry.]

developed to characterize the functional grading of tendon can be applied toward cartilage. Cartilage has two functional gradings that involve mineral gradients: uncalcified cartilage to calcified cartilage and calcified cartilage to subchondral bone. The number of functional gradings in the osteochondral interface increases to five if we also consider interzonal variations in uncalcified cartilage. A recent FT-IR imaging study on the cartilagebone interface examined collagen, proteoglycan, and mineral distribution and collagen integrity as a function of cartilage zone and age.90 As expected, FT-IR images revealed zone-dependent compositional heterogeneity with proteoglycan and collagen most concentrated in the deep zone. Calcified cartilage was found to have an exponential mineralized gradient-similar to that of the tendon-bone enthesis. Gamsjaeger et al. orally presented results from a high-resolution Raman microscopy study of articular cartilage and subchondral bone from the femoral head at the SPEC meeting in 2010. Nonlinear

spectroscopic imaging of these transitions may be able to provide additional high-resolution studies.61,74 These new developments are exciting, and I look forward to more work in this area. There are many possible avenues for translation of Raman spectroscopy into the clinic. Akkus et al. have demonstrated one application in a point-of-care crystal analysis system. The low cost of the CARD instrument and highly diagnostic Raman identification of crystals may enable joint aspiration at clinics where dedicated equipment is cost prohibitive and rheumatology expertise is limited. Another avenue is intra-operative Raman spectroscopy to potentially improve surgical treatments for joint diseases or injuries.59,91 We developed prototype Raman arthroscope probes and performed preclinical studies in human cadaveric knee joints, osteochondral tissue phantoms, and finite element analysis simulations of light propagation in the knee joint.59 In another study, Esmonde-White and Morris

APPLIED SPECTROSCOPY OA

1215

focal point review have demonstrated feasibility of rapid, non-contact in vivo Raman measurements in the surgical suite during ACL repair surgeries.92 The Raman instrumentation is compatible with sterility, laser safety, and time constraints of the surgical suite. High signal-to-noise exposed bone measurements were collected in as little as 15 sec. Extension of Raman spectroscopy to arthroscopy and arthroplasty may improve patient outcomes by identifying sclerotic subchondral bone, grading cartilage damage, or determining optimal sites for ligament reattachment. Exciting preclinical work and clinical studies have demonstrated technological feasibility. One important consideration for future Raman arthroscopy studies is that the joint is constantly irrigated with sterile phosphate buffered saline during the surgery to remove debris and slightly swell the joint cavity to enable better visualization of joint structures. The irrigation has important consequences on collection efficiency. The first step is adaptation of in silico models to a water-to-tissue environment instead of an air-totissue environment. Future prototype optical and probe casing designs for a Raman arthroscope should be compatible with an immersion environment. Coupling the reduced limits of detection and sensitivity of SERS with the non-invasive capability of transcutaneous Raman spectroscopy was reported independently in 2010 by Stone et al.93 and Yuen et al.94 A coupled surface-enhanced spatially offset Raman spectroscopy instrument may enable non-invasive visualization of soft musculoskeletal tissues.95 Additional studies are underway on SERS-active nanoparticles to address in vivo safety and obtaining reliable SERS enhancement in tissues where optical scattering and absorption are significant.96–98

6. 7.

8.

9. 10.

11.

12.

13.

14. 15. 16. 17.

18. 19.

20.

ACKNOWLEDGMENTS

21.

I thank Prof. Ozan Akkus and his research group for helpful discussions. I also thank Prof. Blake Roessler, Prof. Michael Morris, Dr. Francis Esmonde-White, and Daphne Esmonde-White for their support.

22.

1. D. Symmons, C. Mathers, B. Pfleger. Global Burden of Osteoarthritis in the Year 2000. Geneva, Switzerland: World Health Organization, 2000. 2. R.C. Lawrence, D.T. Felson, C.G. Helmick, L.M. Arnold, H. Choi, R.A. Deyo, S. Gabriel, R. Hirsch, M.C. Hochberg, G.G. Hunder, J.M. Jordan, J.N. Katz, H.M. Kremers, F. Wolfe, National Arthritis Data Workgroup. ‘‘Estimates of the Prevalence of Arthritis and Other Rheumatic Conditions in the United States: Part II’’. Arthritis Rheum. 2008. 58(1): 26-35. doi: 10.1002/art.23176. 3. E. Myasoedova, C.S. Crowson, H.M. Kremers, T.M. Therneau, S.E. Gabriel. ‘‘Is the Incidence of Rheumatoid Arthritis Rising? Results from Olmsted County, Minnesota, 1955–2007’’. Arthritis Rheum. 2010. 62(6): 1576-1582. doi: 10.1002/art.27425. 4. U.S. Department of Health and Human Services, Centers for Disease Control and Prevention. ‘‘Prevalence of Doctor-Diagnosed Arthritis and Arthritis-Attributable Activity Limitation—United States, 2010–2012’’. Morb. Mortal. Wkly. Rep. 2013. 62(44): 869-873. 5. B.S. Miller, K.K. Briggs, B. Downie, J.R. Steadman. ‘‘Clinical Outcomes Following the Microfracture Procedure for Chondral

1216

Volume 68, Number 11, 2014

23. 24.

25.

26.

27.

28.

29.

Defects of the Knee’’. Cartilage. 2010. 1(2): 108-112. doi: 10.1177/1947603510366575. R.B. Martin. Skeletal Tissue Mechanics. New York, NY: Springer, 1998. T.W. Lin, L. Cardenas, L.J. Soslowsky. ‘‘Biomechanics of Tendon Injury and Repair’’. J. Biomech. 2004. 37(6): 865-877. doi: 10.1016/j.jbiomech.2003.11.005. M.Z. Ilic, P. Carter, A. Tyndall, J. Dudhia, C.J. Handley. ‘‘Proteoglycans and Catabolic Products of Proteoglycans Present in Ligament’’. Biochem. J. 2005. 385(Pt 2): 381-388. doi: 10.1042/BJ20040844. J.T. Hansen, F.H. Netter. Netter’s Clinical Anatomy. Philadelphia, PA: Saunders/Elsevier, 2010. 2nd ed. C. Kiani, L. Chen, Y.J. Wu, A.J. Yee, B.B. Yang. ‘‘Structure and Function of Aggrecan’’. Cell Res. 2002. 12(1): 19-32. doi: 10.1038/sj.cr.7290106. C.J. Hernandez, T.M. Keaveny. ‘‘A Biomechanical Perspective on Bone Quality’’. Bone. 2006. 39(6): 1173-1181. doi: 10.1016/ j.bone.2006.06.001. M.L. Bouxsein, E. Seeman. ‘‘Quantifying the Material and Structural Determinants of Bone Strength’’. Best Pract. Res. Clin. Rheumatol. 2009. 23(6): 741-753. doi: 10.1016/j.berh.2009.09.008. D. Felsenberg, S. Boonen. ‘‘The Bone Quality Framework: Determinants of Bone Strength and Their Interrelationships, and Implications for Osteoporosis Management’’. Clin. Ther. 2005. 27(1): 1-11. K.J. Jepsen. ‘‘Systems Analysis of Bone’’. Wiley Interdiscip. Rev.: Syst. Biol. Med. 2009. 1(1): 73-88. R. Muller. ‘‘Hierarchical Microimaging of Bone Structure and Function’’. Nat. Rev. Rheumatol. 2009. 5(7): 373-381. D.T. Felson, T. Neogi. ‘‘Osteoarthritis: Is It a Disease of Cartilage or of Bone?’’ Arthritis Rheum. 2004. 50(2): 341-344. K. Esmonde-White, F. Esmonde-White. ‘‘Raman Spectroscopy in Biomineralization’’. Biomineralization Sourcebook. Boca Raton, FL: CRC Press, 2014. Pp. 59-71. N.A. Macleod, P. Matousek. ‘‘Deep Noninvasive Raman Spectroscopy of Turbid Media’’. Appl. Spectrosc. 2008. 62(11): 291A-304A. A. Masic, L. Bertinetti, R. Schuetz, L. Galvis, N. Timofeeva, J.W. Dunlop, J. Seto, M.A. Hartmann, P. Fratzl. ‘‘Observations of Multiscale, Stress-Induced Changes of Collagen Orientation in Tendon by Polarized Raman Spectroscopy’’. Biomacromolecules. 2013. 12(11): 3989-3996. doi: 10.1021/bm201008b. A. Zumbusch, G. Holtom, X. Xie. ‘‘Three-Dimensional Vibrational Imaging by Coherent Anti-Stokes Raman Scattering’’. Phys. Rev. Lett. 1999. 82(20): 4142-4145. doi: 10.1103/PhysRevLett.82.4142. J.-X. Cheng, X.S. Xie. ‘‘Coherent Anti-Stokes Raman Scattering Microscopy: Instrumentation, Theory, and Applications’’. J. Phys. Chem. B. 2004. 108(3): 827-840. doi: 10.1021/jp035693v. C.W. Freudiger, W. Min, B.G. Saar, S. Lu, G.R. Holtom, C. He, J.C. Tsai, J.X. Kang, X.S. Xie. ‘‘Label-Free Biomedical Imaging with High Sensitivity by Stimulated Raman Scattering Microscopy’’. Science. 2008. 322(5909): 1857-1861. doi: 10.1126/science.1165758. B.G. Frushour, J.L. Koenig. ‘‘Raman Scattering of Collagen, Gelatin, and Elastin’’. Biopolymers. 1975. 14(2): 379-391. W.B. Rippon, J.L. Koenig, A.G. Walton. ‘‘Raman Spectroscopy of Proline Oligomers and Poly-L-proline’’. J. Am. Chem. Soc. 1970. 92: 7455-7459. M.J. Deveney, A.G. Walton, J.L. Koenig. ‘‘Raman Spectra of Imino Acids and Poly-L-hydroxyproline’’. Biopolymers. 1971. 10(4): 615-630. S.C. Goheen, L.J. Lis, J.W. Kauffman. ‘‘Raman Spectroscopy of Intact Feline Corneal Collagen’’. Biochim. Biophys. Acta. 1978. 536: 197-204. R. Bansil, I.V. Yannas, H.E. Stanley. ‘‘Raman Spectroscopy: A Structural Review of Glycosaminoglycans’’. Biochim. Biophys. Acta. 1978. 541: 535-542. P. Tarakeshwar, S. Manogaran. ‘‘Proline and Hydroxyproline Zwitterions—An Ab Initio Study’’. J. Mol. Struct. Theochem. 1997. 417(3): 255-263. R. Ellis, E. Green, C.P. Winlove. ‘‘Structural Analysis of Glycosaminoglycans and Proteoglycans by Means of Raman Microspectrometry’’. Connect. Tissue Res. 2009. 50: 29-36.

30. Z. Chi, X.G. Chen, J.S.W. Holtz, S.A. Asher. ‘‘UV Resonance RamanSelective Amide Vibrational Enhancement: Quantitative Methodology for Determining Protein Secondary Structure’’. Biochemistry. 1998. 37(9): 2854-2864. 31. S.U. Sane, S.M. Cramer, T.M. Przybycien. ‘‘A Holistic Approach to Protein Secondary Structure Characterization Using Amide I Band Raman Spectroscopy’’. Anal. Biochem. 1999. 269(2): 255-272. 32. J.T. Pelton, L.R. McLean. ‘‘Spectroscopic Methods for Analysis of Protein Secondary Structure’’. Anal. Biochem. 2000. 277: 167-176. 33. A. Bonifacio, V. Sergo. ‘‘Effects of Sample Orientation in Raman Microspectroscopy of Collagen Fibers and Their Impact on the Interpretation of the Amide III Band’’. Vib. Spectrosc. 2010. 53(2): 314-317. 34. E.P. Paschalis, K. Verdelis, S.B. Doty, A.L. Boskey, R. Mendelsohn, M. Yamauchi. ‘‘Spectroscopic Characterization of Collagen CrossLinks in Bone’’. J. Bone Miner. Res. 2001. 16(10): 1821-1828. 35. E.P. Paschalis, R. Mendelsohn, A.L. Boskey. ‘‘Infrared Assessment of Bone Quality: A Review’’. Clin. Orthop. 2011. 469(8): 2170-2178. doi: 10.1007/s11999-010-1751-4. 36. E.P. Paschalis, D.N. Tatakis, S. Robins, P. Fratzl, I. Manjubala, R. Zoehrer, S. Gamsjaeger, B. Buchinger, A. Roschger, R. Phipps, A.L. Boskey, E. Dall’Ara, P. Varga, P. Zysset, K. Klaushofer, P. Roschger. ‘‘Lathyrism-Induced Alterations in Collagen Cross-links Influence the Mechanical Properties of Bone Material Without Affecting the Mineral’’. Bone. 2011. 49(6): 1232-1241. doi: 10.1016/j.bone.2011.08.027. 37. D. Farlay, M.-E. Duclos, E. Gineyts, C. Bertholon, S. Viguet-Carrin, J. Nallala, G. Sockalingum, D. Bertrand, T. Roger, D.J. Hartmann, R. Chapurlat, G. Boivin. The Ratio 1660/1690 cm-1 Measured by Infrared Microspectroscopy is not Specific of Enzymatic Collagen Cross-Links in Bone Tissue. PLoS One. 2011. 6(12): e28736. 38. S. Gamsjaeger, S. Robins, D.N. Tatakis, K. Klaushofer, E.P. Paschalis. ‘‘Identification of Trivalent Collagen Cross Links by Raman Microspectroscopy’’. Bone. 2012. 51(6): S18. doi: 10.1016/j. bone.2012.08.055. 39. E. McNerny, B. Gong, M. Morris, D. Kohn. Assessing Spatial Variance in Collagen Maturity: Importance of Tissue History. San Antonio, TX: Orthopaedic Research Society, 2013. 40. I.R. Lewis, H. Edwards. Handbook of Raman Spectroscopy: From the Research Laboratory to the Process Line. Boca Raton, FL, USA: CRC Press, 2001. 41. R.K. Reddy, R. Bhargava. ‘‘Chemometric Methods for Biomedical Raman Spectroscopy and Imaging’’. In: P. Matousek, M.D. Morris, editors. Emerging Raman Applications and Techniques in Biomedical and Pharmaceutical Fields. Berlin, Germany: Springer, 2010. Pp. 179-213. 42. K.A. Antonio, Z.D. Schultz. ‘‘Advances in Biomedical Raman Microscopy’’. Anal. Chem. 2014. 86(1): 30-46. doi: 10.1021/ ac403640f. 43. M. Janko, P. Davydovskaya, M. Bauer, A. Zink, R.W. Stark. ‘‘Anisotropic Raman Scattering in Collagen Bundles’’. Opt. Lett. 2010. 35(16): 2765. doi: 10.1364/OL.35.002765. 44. L. Galvis, J.W.C. Dunlop, G. Duda, P. Fratzl, A. Masic. ‘‘Polarized Raman Anisotropic Response of Collagen in Tendon: Towards 3D Orientation Mapping of Collagen in Tissues’’. PLoS One. 2013. 8(5): doi: 10.1371/journal.pone.0063518. 45. B. Wopenka, A. Kent, J.D. Pasteris, Y. Yoon, S. Thomopoulos. ‘‘The Tendon-to-Bone Transition of the Rotator Cuff: A Preliminary Raman Spectroscopic Study Documenting the Gradual Mineralization Across the Insertion in Rat Tissue Samples’’. Appl. Spectrosc. 2008. 62(12): 1285-1294. 46. G.M. Genin, A. Kent, V. Birman, B. Wopenka, J.D. Pasteris, P.J. Marquez, S. Thomopoulos. ‘‘Functional Grading of Mineral and Collagen in the Attachment of Tendon to Bone’’. Biophys. J. 2009. 97(4): 976-985. doi: 10.1016/j.bpj.2009.05.043. 47. A.G. Schwartz, J.D. Pasteris, G.M. Genin, T.L. Daulton, S. Thomopoulos. ‘‘Mineral Distributions at the Developing Tendon Enthesis’’. PLoS One. 2012. 7(11): e48630. 48. A.G. Schwartz, J.H. Lipner, J.D. Pasteris, G.M. Genin, S. Thomopoulos. ‘‘Muscle Loading is Necessary for the Formation of a Functional Tendon Enthesis’’. Bone. 2013. 55(1): 44-51. doi: 10.1016/j.bone.2013.03.010.

49. L. Wang, Y.X. Su, G.S. Zheng, G.Q. Liao, W.H. Zhang. ‘‘Healing Masseter Entheses of Mandibular Reconstruction with AutograftRaman Spectroscopic and Histological Study’’. Int. J. Oral Maxillofac Surg. 2013. 42(7): 915-922. doi: 10.1016/j.ijom.2012.12. 010. 50. S.C.G. Penteado, B.P. Fogazza, C. da Silva Carvalho, E.A. Loschiavo Arisawa, M.A. Martins, A.A. Martin, H. da Silva Martinho. ‘‘Diagnosis of Degenerative Lesions of Supraspinatus Rotator Cuff Tendons by Fourier Transform-Raman Spectroscopy’’. J. Biomed. Opt. 2008. 13(1): 014018. doi: 10.1117/1.2841017. 51. J.J. Carcamo, A.E. Aliaga, E. Clavijo, M. Branes, M.M. CamposVallette. ‘‘Raman and Surface-Enhanced Raman Scattering in the Study of Human Rotator Cuff Tissues After Shock Wave Treatment’’. J. Raman Spectrosc. 2012. 43(2): 248-254. doi: 10.1002/jrs.3019. 52. M.W. Winchester, L.W. Winchester, N.Y. Chou. ‘‘Application of Raman Scattering to the Measurement of Ligament Tension’’. 2008 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 1-8. 2008. Pp. 3434-3437. 53. J.P. Spalazzi, A.L. Boskey, N. Pleshko, H.H. Lu. ‘‘Quantitative Mapping of Matrix Content and Distribution Across the Ligament-toBone Insertion’’. PLoS ONE. 2013. 8(9): e74349. doi: 10.1371/journal. pone.0074349. 54. M.D. Morris, M.V. Schulmerich, K.A. Dooley, K.A. Esmonde-White. ‘‘Vibrational Spectroscopic Imaging of Hard Tissue’’. In: R. Salzer, H.W. Siesler. editors. Infrared and Raman Spectroscopic Imaging. Weinheim, Germany: Wiley-VCH, 2008. 55. A. Hanifi, H. McCarthy, S. Roberts, N. Pleshko. ‘‘Fourier Transform Infrared Imaging and Infrared Fiber Optic Probe Spectroscopy Identify Collagen Type in Connective Tissues’’. PLoS One. 2013. 8(5): e64822. 56. A.M. Croxford, S. Whittingham, D. McNaughton, K.S. Nandakumar, R. Holmdahl, M.J. Rowley. ‘‘Type II Collagen–Specific Antibodies Induce Cartilage Damage in Mice Independent of Inflammation’’. Arthritis Rheum. 2013. 65(3): 650-659. doi: 10.1002/art.37805. 57. L. Rieppo, S. Saarakkala, J.S. Jurvelin, J. Rieppo. ‘‘Prediction of Compressive Stiffness of Articular Cartilage Using Fourier Transform Infrared Spectroscopy’’. J. Biomech. 2013. 46(7): 1269-1275. 58. K.A. Dehring, N.J. Crane, A.R. Smukler, J.B. McHugh, B.J. Roessler, M.D. Morris. ‘‘Identifying Chemical Changes in Subchondral Bone Taken from Murine Knee Joints Using Raman Spectroscopy’’. Appl. Spectrosc. 2006. 60(10): 1134-1141. 59. K.A. Esmonde-White, F.W.L. Esmonde-White, M.D. Morris, B.J. Roessler. ‘‘Fiber-Optic Raman Spectroscopy of Joint Tissues’’. Analyst. 2011. 136(8): 1675-1685. 60. T. Buchwald, K. Niciejewski, M. Kozielski, M. Szybowicz, M. Siatkowski, H. Krauss. ‘‘Identifying Compositional and Structural Changes in Spongy and Subchondral Bone from the Hip Joints of Patients with Osteoarthritis Using Raman Spectroscopy’’. J. Biomed. Opt. 2012. 17(1): 017007. doi: 10.1117/1.jbo.17.1.017007. 61. J.C. Mansfield, C.P. Winlove. ‘‘A Multi-modal Multiphoton Investigation of Microstructure in the Deep Zone and Calcified Cartilage’’. J. Anat. 2012. 220(4): 405-416. doi: 10.1111/j.1469-7580.2012.01479.x. 62. R.A. de Souza, M. Xavier, N.M. Mangueira, A.P. Santos, A.L.B. Pinheiro, A.B. Villaverde, L. Silvera Jr. ‘‘Raman Spectroscopy Detection of Molecular Changes Associated with Two Experimental Models of Osteoarthritis in Rats’’. Lasers Med. Sci. 2013. 1-8. 63. J.G. Kerns, P.D. Gikas, K. Buckley, A. Shepperd, H.L. Birch, I. McCarthy, et al. ‘‘Evidence from Raman Spectroscopy of a Putative Link Between Inherent Bone Matrix Chemistry and Degenerative Joint Disease’’. Arthritis Rheumatol. 2014. 66(5): 1237-1246. doi: 10.1002/art.38360. 64. G. Leroy, L. Fleurisse, C. Creusy, S. Mordon. Raman Microspectrometry of Laser Irradiated Cartilage. Lake Buena Vista, FL: American Society for Laser Medicine and Surgery, 2005. 65. M. Heger, S. Mordon, G. Leroy, L. Fleurisse, C. Creusy. ‘‘Raman Microspectrometry of Laser-Reshaped Rabbit Auricular Cartilage: Preliminary Study on Laser-Induced Cartilage Mineralization’’. J. Biomed. Opt. 2006. 11(2): 024003-024008. 66. N. Ignatieva, O. Zakharkina, G. Leroy, E. Sobol, N. Vorobieva, S. Mordon. ‘‘Molecular Processes and Structural Alterations in Laser Reshaping of Cartilage’’. Laser Phys. Lett. 2007. 4(10): 749-753.

APPLIED SPECTROSCOPY OA

1217

focal point review 67. M. Pudlas, S. Koch, C. Bolwien, H. Walles. ‘‘Raman Spectroscopy as a Tool for Quality and Sterility Analysis for Tissue Engineering Applications like Cartilage Transplants’’. Int. J. Artif. Organs. 2010. 33(4): 228-237. 68. Y. Xia, N. Ramakrishnan, A. Bidthanapally. ‘‘The Depth-Dependent Anisotropy of Articular Cartilage by Fourier-Transform Infrared Imaging (FTIRI)’’. Osteoarthritis Cartilage. 2007. 15(7): 780-788. 69. A. Bonifacio, C. Beleites, F. Vittur, E. Marsich, S. Semeraro, S. Paoletti, V. Sergo. ‘‘Chemical Imaging of Articular Cartilage Sections with Raman Mapping, Employing Uni- and Multi-variate Methods for Data Analysis’’. Analyst. 2010. 135(12): 3193-3204. 70. D.E.T. Shepherd, B.B. Seedhom. ‘‘‘‘Thickness of Human Articular Cartilage in Joints of the Lower Limb’’. Ann. Rheum Dis. 1999. 58(1): 27-34. doi: 10.1136/ard.58.1.27. 71. J.F. Beek, P. Blokland, P. Posthumus, M. Aalders, J.W. Pickering, H.J.C.M. Sterenborg, M.J. van Gemert. ‘‘In Vitro Double-IntegratingSphere Optical Properties of Tissues Between 630 and 1064 nm’’. Phys. Med. Biol. 1997. 42(11): 2255-2261. 72. N.S.J. Lim, Z. Hamed, C.H. Yeow, C. Chan, Z. Huang. ‘‘Early Detection of Biomolecular Changes in Disrupted Porcine Cartilage Using Polarized Raman Spectroscopy’’. J. Biomed. Opt. 2011. 16(1): 017003-01700310. 73. P. Papaspyridakou, D. Papachristou, P. Megas, G. Kontoyannis, M. Orkoula. Unraveling the Chemical Changes Induced on Human Meniscus by Osteoarthritis Using Micro-Raman Spectroscopy. Poster presented at: 9th Panhellenic Scientific Conference in Chemical Engineering. Athens, Greece; May 23-25, 2013. 74. J. Mansfield, J. Moger, E. Green, C. Moger, C.P. Winlove. ‘‘Chemically Specific Imaging and In-Situ Chemical Analysis of Articular Cartilage with Stimulated Raman Scattering’’. J. Biophotonics. 2013. 6(10): 803-814. doi: 10.1002/jbio.201200213. 75. G.D. Jay, J.R. Torres, M.L. Warman, M.C. Laderer, K.S. Breuer. ‘‘The Role of Lubricin in the Mechanical Behavior of Synovial Fluid’’. Publ. Natl. Acad. Sci. U.S.A. 2007. 104(15): 6194-6199. doi: 10.1073/pnas.0608558104. 76. H. Fam, J.T. Bryant, M. Kontopoulou. ‘‘Rheological Properties of Synovial Fluids’’. Biorheology. 2007. 44(2): 59-74. 77. S.R. Brannan, D.A. Jerrard. ‘‘Synovial Fluid Analysis’’. J. Emerg. Med. 2006. 30(3): 331-339. 78. N. Garg, L. Perry, A. Deodhar. ‘‘Intra-articular and Soft Tissue Injections: A Systematic Review of Relative Efficacy of Various Corticosteroids’’. Clin. Rheumatol. 2014. doi: 10.1007/s10067-0142572-8. 79. D.S. Gibson, M.E. Rooney. ‘‘The Human Synovial Fluid Proteome: A Key Factor in the Pathology of Joint Disease’’. Proteomics-Clin. Appl. 2007. 1(8): 889-899. 80. R. Gobezie, A. Kho, B. Krastins, D.A. Sarracino, T.S. Thornhill, M. Chase, P.J. Millet, D.M. Lee. ‘‘High Abundance Synovial Fluid Proteome: Distinct Profiles in Health and Osteoarthritis’’. Arthritis Res. Ther. 2007. 9(2): R36. 81. S.Y. Ritter, R. Subbaiah, G. Bebek, J. Crish, C.R. Scanzello, B. Krastins, et al. ‘‘Proteomic Analysis of Synovial Fluid from the Osteoarthritic Knee: Comparison with Transcriptome Analyses of Joint Tissues’’. Arthritis Rheum. 2013. 65(4): 981-992. doi: 10.1002/art.37823. 82. K.A. Esmonde-White, G.S. Mandair, F. Raaii, J.A. Jacobson, B.S. Miller, A.G. Urquhart, B.J. Roessler, M.D. Morris. ‘‘Raman Spectroscopy of Synovial Fluid as a Tool for Diagnosing Osteoarthritis’’. J Biomed. Opt. 2009. 14(3): 034013. 83. J. Cibere, H. Zhang, P. Garnero, A.R. Poole, T. Lobanok, T. Saxne, V.B. Kraus, A. Way, A. Thorne, H. Wong, J. Singer, J. Kopec, A. Guermazi, C. Peterfy, S. Nicolaou, P.L. Munk, J.M. Esdaile. ‘‘Association of Biomarkers with Pre-Radiographically Defined and Radiographically Defined Knee Osteoarthritis in a Population-Based Study’’. Arthritis Rheum. 2009. 60(5): 1372-1380.

1218

Volume 68, Number 11, 2014

84. B. Schmidt-Rohlfing, M. Thomsen, C. Niedhart, D.C. Wirtz, U. Schneider. ‘‘Correlation of Bone and Cartilage Markers in the Synovial Fluid with the Degree of Osteoarthritis’’. Rheumatol. Int. 2002. 21: 193-199. 85. A. Yavorskyy, A. Hernandez-Santana, G. McCarthy, G. McMahon. ‘‘Detection of Calcium Phosphate Crystals in the Joint Fluid of Patients with Osteoarthritis—Analytical Approaches and Challenges’’. Analyst. 2008. 133(3): 302-318. 86. N. McGill, P.A. Dieppe, M. Bowden, D.J. Gardiner, M. Hall. ‘‘Identification of Pathological Mineral Deposits by Raman Microscopy’’. Lancet. 1991. 337: 77-78. 87. X. Cheng, D.G. Haggins, R.H. York, Y.N. Yeni, O. Akkus. ‘‘Analysis of Crystals Leading to Joint Arthropathies by Raman Spectroscopy: Comparison with Compensated Polarized Imaging’’. Appl. Spectrosc. 2009. 63(4): 381-386. 88. B. Li, S. Yang, O. Akkus. ‘‘A Customized Raman System for Point-ofCare Detection of Arthropathic Crystals in the Synovial Fluid’’. Analyst. 2014. 139(4): 823-830. doi: 10.1039/c3an02062b. 89. S. Yang, B. Li, M.N. Slipchenko, A. Akkus, N.G. Singer, Y.N. Yeni, O. Akkus. ‘‘Laser Wavelength Dependence of Background Fluorescence in Raman Spectroscopic Analysis of Synovial Fluid from Symptomatic Joints’’. J. Raman Spectrosc. 2013. 44(8): 1089-1095. doi: 10.1002/jrs.4338. 90. N.T. Khanarian, M.K. Boushell, J.P. Spalazzi, N. Pleshko, A.L. Boskey, H.H. Lu. ‘‘FTIR-I Compositional Mapping of the Cartilageto-Bone Interface as a Function of Tissue Region and Age’’. J. Bone Miner. Res. 2014. doi: 10.1002/jbmr.2284. 91. G. Pezzotti, N. Sugano. ‘‘Cartilage Regeneration and the Role of Vibrational Spectroscopy in Future Joint Arthroplasty’’. In: A. Bianco, I. Cacciotti, I. Cappelloni, editors. Bone and Materials for Bone Tissue Engineering. Clausthal-Zellerfeld, Germany: Trans Tech Publications, 2013. Pp. 121-133. 92. F.W.L. Esmonde-White, M.D. Morris. ‘‘Validating In Vivo Raman Spectroscopy of Bone in Human Subjects’’. In: N. Kollias, editor. Photonic Therapeutics and Diagnostics IX. Proc. SPIE , 2013. P. 85656K. 93. N. Stone, K. Faulds, D. Graham, P. Matousek. ‘‘Prospects of Deep Raman Spectroscopy for Noninvasive Detection of Conjugated Surface Enhanced Resonance Raman Scattering Nanoparticles Buried Within 25 mm of Mammalian Tissue’’. Anal. Chem. 2010. 82(10): 3969-3973. doi: 10.1021/ac100039c. 94. J.M. Yuen, N.C. Shah, J.T. Walsh, M.R. Glucksberg, R.P. Van Duyne. ‘‘Transcutaneous Glucose Sensing by Surface-Enhanced Spatially Offset Raman Spectroscopy in a Rat Model’’. Anal. Chem. 2010. 82(20): 8382-8385. doi: 10.1021/ac101951j. 95. S.E. Bohndiek, A. Wagadarikar, C.L. Zavaleta, D. Van de Sompel, E. Garai, J.V. Jokerst, S. Yazdanfar, S.S. Gambhir. ‘‘A Small Animal Raman Instrument for Rapid, Wide-Area, Spectroscopic Imaging’’. Proc. Natl. Acad. Sci. U.S.A. 2013. 110(30): 12408-12413. doi: 10.1073/pnas.1301379110. 96. S.T. Sivapalan, B.M. DeVetter, T.K. Yang, T. van Dijk, M.V. Schulmerich, P.S. Carney, R. Bhargava, C.J. Murphy. ‘‘OffResonance Surface-Enhanced Raman Spectroscopy from Gold Nanorod Suspensions as a Function of Aspect Ratio: Not What We Thought’’. ACS Nano. 2013. 7(3): 2099-2105. doi: 10.1021/ nn305710k. 97. X. Yang, A.P. Gondikas, S.M. Marinakos, M. Auffan, J. Liu, H. HsuKim, et al. ‘‘Mechanism of Silver Nanoparticle Toxicity Is Dependent on Dissolved Silver and Surface Coating in Caenorhabditis elegans’’. Environ. Sci. Technol. 2012. 46(2): 1119-1127. doi: 10.1021/es202417t. 98. Y.-C. Yeh, B. Creran, V.M. Rotello. ‘‘Gold Nanoparticles: Preparation, Properties, and Applications in Bionanotechnology’’. Nanoscale. 2012. 4(6): 1871. doi: 10.1039/c1nr11188d.