1.

Introduction

Osteoarthritis (OA) is a painful, debilitating joint disease characterized by progressive degeneration of the articular cartilage. OA affects approximately one third of people in the United States over 18 years of age¹ and is considered the leading cause of long-term disability among adults.² The joint degeneration that is characteristic of OA occurs most frequently in the hand, foot, knee, and hip joints, and frequently results in the need for total replacement of the joint. A significant complication in the treatment of OA is the inability to diagnose the disease process at an early stage. While macroscopic damage related to end-stage OA is easily recognized by current radiographic techniques, early stages of OA that involve changes in the molecular components of the cartilage, but no obvious mechanical damage, are much more difficult to identify.

The principal macromolecular components of cartilage are the fibrillar protein type II collagen and the major proteoglycan (PG) aggrecan. Aggrecan consists of glycosaminoglycan side chains covalently attached to a protein core. Pathological cartilage turnover involves the degradation of these macromolecules triggered by either cellular or mechanical signals.^{3 4 5 6 7 8 9} Early stage OA has been characterized by an increase in tissue hydration (tissue swelling), collagen breakdown,^{7 8 9 10 11 12 13} and by chondrocyte upregulated synthesis of specific metalloproteinases (MMPs), primarily collagenases MMP 1 and 13, and aggrecanases 1 and 2 (ADAMTSs 4 and 5),^{7 8 9 10 11 12 13} that initiate the break down of collagen and PG, respectively. This process leads to irreversible degradation of the densely packed type II collagen fibrils in the cartilage superficial zone and the eventual loss of the PG aggrecan from the extracellular matrix.^{11 14 15 16}

A recent study in our laboratory has demonstrated that it is possible to monitor changes in the molecular structure of cartilage components at the articular surface of late-stage human osteoarthritic knee joints using Fourier transform infrared spectroscopy (FTIR) in conjunction with an infrared fiber optic probe (IFOP).¹⁷ We hypothesized that the observed molecular changes were related primarily to the breakdown of the collagen network. In the current study, we further investigate the origin of these spectral changes using FTIR imaging spectroscopic (FT-IRIS) analysis of collagenase-treated articular cartilage combined with immunohistochemical analysis. The biochemical action of collagenase is to cleave the collagen molecule within the triple helical domain at a single site 3/4 distance from the N-terminal end, resulting in 3/4- and 1/4-length fragments that can be detected by an antibody specific for these fragments.¹⁸ Comparison of the spectral imaging data from these in vitro collagenase-treated samples to the spectral data previously acquired by IFOP from human OA cartilage provides support for the potential use of the IFOP to monitor collagen breakdown in an in vivo environment.

2.

Materials and Methods

3.

Results

Changes in the surface morphology of the superficial zone of the collagenase-treated cartilage explants were seen after both 15 and 30 min of treatment [Figs. 1(b) and 1(c)]. Nontreated (control) explants had intact superficial zones [Fig. 1(a)]. Immunohistochemical evaluation using the Col2-3/4C _short antibody resulted in positive staining for the collagenase-treated sections only (Fig. 2). Positive staining was observed through most of the superficial zone for the 15-min treated specimens (depth of penetration 250 μm±20 μm), and through the superficial and into the transition zone for the 30-min specimens (depth of penetration 350 μm±30 μm) (Fig. 2).

Figure 1

Microscope images of bovine cartilage sections on the BaF ₂ window before FTIR data acquisition: (a) untreated, (b) 15-min collagenase treatment, and (c) 30-min collagenase treatment.

Figure 2

Antibodies show positive stain (green) for collagenase activity in 15- and 30-min treated specimens.

A comparison of the spectral parameters (amide II/1338 cm⁻¹ area ratio and amide III-1238 cm⁻¹/1227 cm⁻¹ peak ratio) [Fig. 3(a)] between the treated and untreated specimens mirrored the differences previously found between normal and osteoarthritic cartilage. With collagenase treatment, the amide II/1338 cm⁻¹ area ratio increased for both the 15-min and 30-min treatment times (p<0.05) [Fig. 3(b)], while the 1238 cm⁻¹/1227 cm⁻¹ peak ratio decreased for both the 15-min and 30-min treatment times (p<0.05) [Fig. 3(c)]. The polarization images and spectra obtained in the superficial zone of the nontreated samples (control) showed that the collagen fibrils were highly oriented parallel to the articular surface (Fig. 4). With collagenase treatment, the amide I/amide II area ratio decreased [Figs. 5(a) and 5(b)] in this region, indicating reduced orientation in the direction parallel to the articular surface (p<0.010). In addition, FT-IRIS analysis below the area of collagenase activity (as measured immunohistochemically) revealed no statistical difference in the collagen orientation between the treated and untreated specimens [Fig. 5(c)]. Also note that for the parameters analyzed, the larger standard deviations for the untreated groups reflect the greater zonal variation within the untreated cartilage.

Figure 3

(a) IR imaging spectra of untreated, 15- and 30-min collagenase-treated specimens. (b) Amide II/peak 1338 area ratio change with collagenase treatment. (^* p<0.05 versus control as determined by one way ANOVA.) (c) 1238 cm⁻¹/1227 cm⁻¹ peak ratio change with collagenase treatment. (^* p<0.05 versus control as determined by one way ANOVA.)

Figure 4

Polarized FT-IR images of the amide I/II area ratio in the superficial zone of untreated and collagenase-treated cartilage.

Figure 5

(a) Polarized IR spectra from the superficial zone of an untreated and 30-min treated specimen. (b) Amide I/II area ratio change within the superficial zone with collagenase treatment. (^* p<0.01 versus control as determined by one way ANOVA.) (c) Amide I/II area ratio change below the regions that showed positive staining for collagenase activity [one way ANOVA determined no statistically significant difference among the three groups (p=0.928)] .

4.

Discussion

A key factor in the development and progression of osteoarthritis is the enzyme-specific destruction of the articular cartilage extracellular macromolecules type II collagen and aggrecan (PG). The matrix metalloproteinase collagenases (MMPs 1, 8, and 13) have been well documented as the primary source for the in vivo degradation of collagen.^{30 31} In the current study, the effect of collagenase on type II collagen at the articular surface of mature bovine cartilage explants was successfully monitored by infrared spectroscopic analysis, and in situ collagen cleavage was confirmed by the use of immunofluorescent staining with the Col2-3/4C _short antibody. These results suggest that collagenase activity on type II collagen does result in IR spectral changes; specifically, an increase in the amide II/1338 cm⁻¹ area ratio (p=0.024), a decrease in the 1238 cm⁻¹/1227 cm⁻¹ peak height ratio (p=0.025), and a decreased in the polarized amide I/II area ratio (p=0.010). These findings corroborate our previous results obtained with the IFOP, where parallel differences were observed between normal and osteoarthritic (severely degraded) human cartilage.¹⁷ It is important to stress at this point that the focus of this study was to relate IFOP spectral changes seen in late-stage OA to the effect of collagenase on the collagen molecule, which is why relatively strong collagenase treatment conditions were utilized. Our next step (currently underway) is to determine the threshold for IFOP detection based on lower concentrations of collagenase at shorter treatment times.

As put forward in the earlier study,¹⁷ the decrease in the 1338 cm⁻¹ collagen absorbance, which arises from CH ₂ side chain vibrations and is sensitive to the order of the triple helix, would account for the increased amide II/1338 cm⁻¹ area ratio. This absorbance band has also been shown to decrease in intensity as the collagen denatures in other studies.^{24 25} Similarly, the change in the amide III absorbance contour with collagen degradation, quantitated by a decrease in the 1238/1227 cm⁻¹ peak intensity ratio, parallels the spectral differences found between normal and OA cartilage. In this case, however, assignment of specific molecular changes to exact frequencies has proved challenging due to the highly complex nature of the amide III vibrational mode.^{26 27 28} Nonetheless, the current study supports those observed spectral changes being associated with the unraveling of the collagen triple helix structure due to enzymatic activity.

To accurately correlate our FT-IRIS findings with collagenase-treated areas, the specimens prepared in this experiment were stained for collagenase-generated neoepitopes using the Col2-3/4C _short antibody. All of the treated sections showed positive staining for collagenase activity through most of the superficial zone. FT-IRIS images generated from the polarized amide I/amide II area ratios enabled visualization of the disrupted orientation that occurs with collagen cleavage. For treated specimens, analysis of this data showed an inverse relationship between collagen organization (orientation) and collagenase treatment times. Control specimens (negative staining for collagenase activity) had the highest superficial zone amide I/amide II area ratios [Fig. 5(b)], which corresponds to the collagen fibrils having an orientation parallel to the articular surface.²¹ As an internal control, spectral analysis below the area of collagenase activity revealed no change in the amide I/amide II area ratio between the treated and untreated specimens, further evidence of the sensitivity of this parameter to collagen structure. These findings clearly present us with a new method for detecting collagen orientation changes in histological sections.

The use of IR spectroscopy to detect surface collagen architecture and molecular changes holds promise as an early diagnostic technique, since such changes may precede macroscopic surface alterations such as ruptures and fibrillations that occur with the progression of OA. Also encouraging is the fact that these molecular changes have previously been reported with the use of an infrared fiber optic probe.¹⁷ Currently, several IR fiber optic techniques are being investigated for clinical use.^{32 33} These systems have the potential to be minimally or noninvasive, nondestructive, and may ultimately allow for early detection of molecular changes in disease states. With respect to cartilage, such a system would ideally be used during arthroscopic procedures for in situ determination of articular cartilage surface integrity, and could also be used to assess the quality of tissue-engineered cartilage both before and after implantation. The current study validates the origin of the molecular changes in degenerative cartilage previously detected by IFOP, and supports its potential as a novel tool for monitoring and treating OA and other joint diseases.