Cartilage is a dense connective tissue which functions as a load bearing material in synovial j oats and provides a smooth surface for joint articulation. It is composed of a highly charged solid phase (the extracellular matrix), an electrolyte fluid, and relatively few cells. Many of the structural and functional properties of cartilage are dependent on the two major constituents of the extracellular matrix:​ glycosaminoglycans and collagen.

The goal of this project was to develop NMR spectroscopy techniques to provide nondestructive and noninvasive determinations of 'cartilage structural and functional integrity. As described below, sodium NMR was used to measure sodium content of cartilage from which fixed charge density (FCD) was calculated, giving an estimate of glycosaminoglycan content. Magnetization transfer NMR was used to evaluate collagen content and structure. These techniques were verified and then used to study changes in matrix composition following perturbations which mimic physiologic and pathophysiologic states.

Because glycosaminoglycans (GAGs) have a net charge under physiological conditions, GAG concentration in the tissue is directly related to tissue fixed charge density (FCD). In turn, FCD is a determinant of the concentration of sodium in tissue water. The NMR method of determining GAG concentration involves detection of the NMR signal from sodium nuclei (the magnitude of this signal is directly proportional to the number of sodium ions in the sample). Sodium concentration is calculated from sodium and water content. GAG concentration can then be calculated using Donnan equilibrium theory. We have verified the NMR visibility of sodium in cartilage by comparison of sodium measured by NMR to sodium measured by inductively coupled plasma emission spectroscopy. Using NMR sodium measurements in calf articular cartilage, we have calculated FCD on the order of that measured using other techniques. For calf epiphyseal cartilage, FCD varied with the position of the sample within the tissue, in a manner consistent with tissue GAG content determined using a biochemical assay. Preliminary NMR images of intact ulnar epiphyseal cartilage demonstrated similar variations in sodium content. NMR measurements for cartilage exposed to baths of differing salt composition, pH, or ionic strength demonstrated the ability of this technique to track changes in tissue sodium content. FCD was also observed to decrease when cartilage was depleted of GAGs by exposure to trypsin. These results demonstrate the ability of sodium NMR to nondestructively measure FCD in cartilage and to follow changes in FCD with changes in matrix charge and composition.

To examine collagen, an NMR experiment known as magnetization transfer (MT) was performed. With a type of MT known as saturation transfer, the protons on macromolecules are perturbed with a radiofrequency pulse until the magnetic moments are randomized. These protons are then given time to exchange magnetization with bulk water protons, leading to a randomization of the magnetic moments in the water. Observation of the water signal at this point results in less signal relative to the case without MT. The amount of relative signal decrease, expressed as Ms/Mo, provides an indication of the amount of macromolecules present and of the interaction between macromolecular and water protons. MT experimental parameters were characterized for collagen suspensions and cartilage. Based on these results, the saturation pulse length in subsequent experiments was chosen to be at least 5T1 (typically 12 sec for cartilage, 16 sec for collagen model systems), saturation pulse power was 12 kT (set with 1 msec 180 degree pulse), and frequency offset was 6 kHz. Ms/Mo did not change significantly in samples after freezing. Repeated measurements of Ms/Mo on the same sample varied by less than 5% (SD/mean). We have shown that the MT effect is dependent on collagen concentration in collagen suspensions, collagen gels, soluble collagen suspensions, and cartilage, with an approximately logarithmic relationship. However, at concentrations on the order of those seen in vivo, MT is relatively insensitive to changes in concentration. In addition, we have demonstrated that factors other than concentration affect the MT signal in collagen suspensions and cartilage. For example, Ms/Mo was -20% greater for highly crosslinked insoluble collagen suspensions than for soluble suspensions of equal collagen content. Ms/Mo also varied among cartilage from different sources with similar collagen content. We have shown that an intact triple helix is important for MT;​ loss of the triple helix with thermal denaturation of soluble collagen suspensions resulted in substantially increased Ms/Mo. Furthermore, treatment with enzymes or low pH led to measurable differences in Ms/Mo which were not explained by differences in collagen concentration. They may be attributable to changes in matrix charge, structure and/or collagen fibril hydration (although changes in fibril hydration with trypsin and low pH produced conflicting results). These data demonstrate the ability of MT to detect changes in collagen concentration and collagen structure.

Osteoarthritis is a degenerative disease of cartilage, particularly common among the aging population, which results in diarthrodial joint failure. The ability to fully understand the disease and formulate sensible preventive and therapeutic strategies has been hampered by the inability to assess the in vivo functional status of cartilage. Clinical diagnosis of degenerative diseases like osteoarthritis is typically only possible once the disease has progressed to some advanced state in which gross structural changes have occurred. We have shown that sodium and magnetization transfer NMR can be used to study cartilage matrix composition, specifically the glycosaminoglycan and collagen constituents. In the future, these techniques should be applicable in a clinical setting for early diagnosis and monitoring progression of degenerative diseases.