Research Overview

Mary Cowman
Associate Professor of Biochemistry

General Information

Tel: 718.260.3054
Room: RH 805



Hyaluronan Polysaccharide Structure and Interactions

Our laboratory has an interest in the biochemical and biophysical characterization of glycosaminoglycans, especially hyaluronan. The research has been supported by the National Institutes of Health and by industry.

Figure 1. The covalent structure of a tetrasaccharide of hyaluronan.

 Figure 2. Space-filling model of a hexasaccharide of hyaluronan.

The structure of the glycosaminoglycan hyaluronan is shown in Figures 1 and 2. Hyaluronan is a high molecular weight linear anionic polysaccharide, with the repeating structure poly[(1-->3)-2-acetamido-2-deoxy-b-D-glucosamine-(1-->4)-b-D-glucuronic acid]. It has widespread occurrence in the extracellular matrix of animal tissues. It is a major contributor to the biomechanical properties of tissues, helps control tissue hydration and water transport, and serves as the structural backbone of cartilage proteoglycan assemblies. Via specific cell surface receptor proteins, hyaluronan affects numerous biological processes, such as development, tumor metastasis, and inflammation. The rheological properties of hyaluronan solutions are of special interest with respect to the functions of liquid connective tissues such as the eye vitreus and joint synovial fluid.

We have developed methods for the enzymatic degradation of glycosaminoglycans and the high resolution chromatographic fractionation of the fragments (Cowman et al., 1981; Cowman, 1985; Hittner and Cowman, 1987). By gaining inspiration from the DNA field, we developed novel polyacrylamide gel electrophoretic methods to analyze the size and charge heterogeneity of the fragments (Cowman et al., 1984a). We subsequently devised a sensitive detection method based on a combination of fragment precipitation in the gel matrix by dyes followed by silver staining (Min and Cowman, 1986). We have used this technique to determine exact molecular weight distributions for hyaluronan fractions of low molecular weight and low polydispersity, for comparison with apparent molecular weights determined by light scattering (Figure 3) (Turner et al., 1988).

For high molecular weight (up to about 8 million) hyaluronan, we developed an agarose gel electrophoretic method to determine molecular weight distribution (Figure 4) (Lee and Cowman, 1994), which is difficult to accomplish by gel permeation chromatography above a molecular weight of about 3 million. In collaboration with Louis Rosenfeld (New York Medical College), our method was applied to demonstrate hyaluronan degradation by peroxynitrite, which may be produced in tissues under inflammatory conditions (Li et al., 1997). It was also employed in a collaborative project with Paul W. Noble (Johns Hopkins University) to determine the size of low molecular weight hyaluronan that induces the expression of genes for inflammation mediators (Noble et al., 1996; McKee et al., 1996). The sensitivity of the electrophoretic method has been utilized in an industrially sponsored study (Biomatrix, Inc.: collaborators Endre A. Balazs, Arnold I. Goldman, Philip A. Band) of hyaluronan molecular weight distribution in knee joint synovial fluid from normal and osteoarthritic patients (Band et al., 1999; Balazs et al., 1999).



Figure 3. Polyacrylamide gel electrophoresis separation of enzymatically digested hyaluronan and purified hyaluronan segments. Adjacent bands arise from hyaluronan segments differing in size by one disaccharide unit. The number of disaccharides is indicated. Taken from Turner et al. (1988).
Figure 4. Densitometric scans of high molecular weight hyaluronan fractions separated by agarose gel electrophoresis. The electrophoretic mobility is linearly related to log M in the range 0.2-8 x 106. Taken from Lee and Cowman (1994).

The solution properties of hyaluronan have been examined by numerous methods. Using circular dichroism (Cowman et al., 1981) and 1H-NMR (Cowman et al., 1984b) spectroscopy, we showed that a proposed hydrogen bond linking the acetamido group with the carboxyl group could not occur directly (although a water-mediated bond is possible). Using both solid state and solution 13C-NMR spectroscopy (Feder-Davis et al., 1991;Cowman et al., 1996; Cowman et al., 2001), we showed that the four-fold helical form adopted by Na-hyaluronan in the solid state is not maintained in aqueous solution (Figure 5), but that the polymer and its oligosaccharide fragments have segmental motions on the nanosecond time scale (Figure 6).


Figure 5. 75 MHz 13C NMR spectra of HA, comparing solid state (top), DMSO solution (middle), and aqueous solution (bottom). Taken from Cowman et al. (1996).


Figure 6. Dependence of sugar ring average 13C NMR T1 at 75 MHz on HA oligosaccharide chain length (tetrasaccharide to octasaccharide) and residue position. Chains are tethered to the right axis by their reducing ends. Taken from Cowman et al. (2001).

These data were part of a larger effort by numerous other researchers in the field, which point to a semiflexible solution conformation, with a dynamically formed and broken set of hydrogen bonds between adjacent sugars. Further supporting data for this model was obtained by analysis of the viscometric behavior of hyaluronan (Turner et al., 1988; Cowman and Matsuoka, 2000). Short fragments generated by enzymatic degradation and fractionated to a very low degree of polydispersity were analyzed by electrophoresis to determine exact molecular weight averages. The intrinsic viscosity of each fraction was determined, and the relationship between the intrinsic viscosity and molecular weight was analyzed, as shown in Figure 7.


Figure 7. Experimental dependence of intrinsic viscosity on molecular weight for hyaluronan in 0.15 M NaCl. Low M region data from Turner et al. (1988), and high M region data from Balazs (1965). Taken from Cowman and Matsuoka (2000).

The transition from low to high molecular weight behavior is not, in fact, sharp, but the extrapolation of the limiting behaviors to an apparent crossover point has the advantage of identifying the smallest chain which behaves like the random coil polymer. Shorter chains behave as extended worms. The smallest coiled chain has a hydrodynamic diameter that corresponds approximately to the contour length divided by p. This behavior is not unique to hyaluronan, but can be seen for other semiflexible polymers.

In collaboration with Shiro Matsuoka (Polytechnic Institute of NYU), we studied the theoretical dependence of polymer solution viscosity on the concentration and the molecular weight (really, intrinsic viscosity, as shape has a strong influence). This led to a new and quite general expression for polymer solution viscosity (Matsuoka and Cowman, 2000; Kwei et al., 2000). The specific viscosity was shown to depend on a polynomial in the product c[h], in a manner similar to the Martin equation, but with only the first four terms required. As shown in Figure 8, the equation proved to be equally applicable to well-behaved solutions of flexible, semi-flexible, and perfectly rigid polymers. Hyaluronan was no exception, which implies that it is well-behaved. This was something of a surprise, as we had earlier found evidence for self-association in solutions of low molecular weight hyaluronan fragments (Turner et al., 1988). A survey of the literature showed that many other researchers had commented about ill-defined "trouble" with low molecular weight hyaluronan, especially with losses on filtration. The molecular weight dependence of the self-association may reflect the extended conformation adopted, or may simply reflect the higher molar concentrations at which the low molecular weight samples are usually studied.


Figure 8. Dependence of the specific viscosity on the coil overlap factor for flexible, semi-flexible, and rigid polymers. Cowman and Matsuoka (2001).

Atomic Force Microscopy of Hyaluronan

Atomic force microscopy (AFM) has given us new insight into hyaluronan structure and behavior. In our initial studies (Cowman et al., 1998a; 1998b; 2000a, 2000b), we observed extended chains of hyaluronan, with a very large apparent persistence length. Figures 9 and 10 are typical of this type of result.



Figure 9. Hyaluronan extended by molecular combing. 1500nm x 1500nm
Figure 10. Extended hyaluronan chain on hydrated mica surface. 700nm x 700nm


Images such as Figures 9 and 10 proved to be difficult to reproduce, until we ascertained the reason for the extended structure's stability (Cowman et al., 2000a, 2000b). When mica is freshly cleaved, then allowed to hydrate at ambient humidity, it builds a layer of structured water at the surface. Contact angle measurements confirm its highly hydrophilic nature. Applying hyaluronan to such a surface allows the polysaccharide chains to lie on this ice-like water layer. The chains adhere well enough to remain extended after molecular combing. It is possible to see the water layer in Figure 10, where we used light tapping for imaging.

In contrast, when chains are applied to freshly cleaved mica, the hyaluronan lies within the surface water layer. It is more difficult to image (we often need to use contrast enhancement), and the chains are less adherent to the surface. Figure 11 shows combed hyaluronan that has begun to recoil. Figure 12 shows a more fully relaxed hyaluronan chain, whose conformation is in accord with expectations for the persistence length of hyaluronan in aqueous salt solutions. The chain appears to have a loose helical bias, with a radius of curvature for the helical turns on the same order of magnitude as that seen in the viscometric analysis of the smallest coiled hyaluronan, or about 15 nm.


Figures 11 and 12. Relaxed hyaluronan chains on freshly cleaved mica. The chains appear to lie within the thin water layer on the surface. Left, 2000nm x 2000nm. Right, 1600nm x 1600nm.

The formation of hairpin structures, as shown in Figures 13 and 14, was also observed in a number of images. This probably illustrates one form of condensation of the chain, when association is favored under the conditions of partial dehydration. It also may reflect twisting of chains as they are deposited on the mica surface from a moving water droplet.


Figure 13. Hairpin formation in hyaluronan. 726nm x 726nm. Figure 14. Coiled hyaluronan with hairpin turn to reverse chain direction. 1000nm x 1000nm
Figure 15 shows condensed chains of hyaluronan, which have adopted the form of short thick rods. This image was obtained for hyaluronan applied to a mica surface from a relatively high concentration solution (0.5 mg/mL). In Figure 16 the condensed hyaluronan worm-like chains appear to wrap around each other, and form twisted ropes.



Figure 15. Condensed hyaluronan in the form of shortened rods or tubes. 3000nm x 3000nm.
Figure 16. Condensed hyaluronan in the form of shortened and intertwined rods or tubes. 1000nm x 1000nm.

Figures 17 and 18 show an alternative mechanism of condensation. Here long fibrous associations are formed.


Figures 17 and 18. Fibrous association of hyaluronan chains. Left, 1300nm x 1300nm. Right, 300nm x 300nm.

Using AFM, we have observed a wide variety of forms for hyaluronan. The loosely coiled form is most likely to reflect the conformation of hyaluronan in dilute solution. The nearly straight or extended forms are a product of the surface interaction and sample application technique. The condensed forms illustrate the versatility of hyaluronan structure, and may be particularly relevant to the understanding of hyaluronan in concentrated solution or in the confines of the extracellular matrix.

References Cited:

  • Balazs, E.A., Cowman, M.K., Goldman, A.I., Band, P.A., Lee, H.G., and Moreland, L. (1999). "Correlation of Knee Pain with Synovial Fluid Properties in Osteoarthritis Patients." New Frontiers in Medical Sciences: Redefining Hyaluronan, Padua, Italy
  • Band, P.A., Goldman, A.I., Cowman, M.K., Moreland, L., Lee, H.G., and Balazs, E.A. (1999). "Correlation of Knee Pain with Synovial Fluid Properties in Osteoarthritic Patients." EULAR99, meeting of the European League Against Rheumatism, Glasgow, Scotland
  • Cowman, M.K., Balazs, E.A., Bergmann, C.W., and Meyer, K. (1981). "Preparation and Circular Dichroism Analysis of Sodium Hyaluronate Oligosaccharides and Chondroitin." Biochemistry 20, 1379-1385.
  • Cowman, M.K., Slahetka, M.F., Hittner, D.M., Kim, J., Forino, M., and Gadelrab, G. (1984a). "Polyacrylamide Gel Electrophoresis and Alcian Blue Staining of Sulphated Glycosaminoglycan Oligosaccharides." Biochem. J. 221, 707-716.
  • Cowman, M.K., Cozart, D., Nakanishi, K., and Balazs, E.A. (1984b). " 1H-nmr of Glycosaminoglycans and Hyaluronic Acid Oligosaccharides in Aqueous Solution: The Amide Proton Environment." Arch. Biochem. Biophys. 230, 203-212.
  • Cowman, M.K. (1985). "Preparation and Characterization of Enzymatically Derived Oligosaccharides and Segments from Glycosaminoglycans." in New Developments in Industrial Polysaccharides, V. Crescenzi, I. Dea, and S. Stivala, eds., Gordon and Breach, New York, pp. 233-253.
  • Cowman, M.K., Hittner, D.M., and Feder-Davis, J., (1996). "13C-NMR Studies of Hyaluronan: Conformational Sensitivity to Varied Environments." Macromolecules 29, 2894-2902.
  • Cowman, M.K., Liu, J., Li, M., Hittner, D.M., and Kim, J.S. (1998a). "Hyaluronan Interactions: Self, Water, Ions.", in The Chemistry, Biology and Medical Applications of Hyaluronan and its Derivatives, T.C. Laurent, ed., Wenner-Gren International Series No. 72, Portland Press, London, pp. 17-24.
  • Cowman, M.K., Li, M., and Balazs, E.A. (1998b). "Tapping Mode Atomic Force Microscopy of Hyaluronan: Extended and Intramolecularly Interacting Chains." Biophys. J. 75, 2030-203.
  • Cowman, M.K. and Matsuoka, S. (2000). "Understanding the Intrinsic Viscosity of Hyaluronan" at Hyaluronan 2000 Conference, Wrexham, Wales. In press for 2002, in Hyaluronan, J.F. Kennedy, G.O. Phillips, P.A. Williams, V.C. Hascall, eds., Woodhead Publishing, Cambridge.
  • Cowman, M.K., Li, M., Dyal, A., and Balazs, E.A. (2000a) "Tapping Mode Atomic Force Microscopy of the Hyaluronan Derivative, Hylan A." Carbohydr. Polym. 41, 229-235.
  • Cowman, M.K., Li, M., Dyal, A., and Kanai, S. (2000b) "Tapping Mode Atomic Force Microscopy of Hyaluronan and Hylan A", In press for 2002, in Hyaluronan, J.F. Kennedy, G.O. Phillips, P.A. Williams, V.C. Hascall, eds., Woodhead Publishing, Cambridge.
  • Cowman, M.K., Feder-Davis, J., and Hittner, D.M. (2001) "13C-NMR Studies of Hyaluronan. 2. Dependence of Conformational Dynamics on Chain Length and Solvent." Macromolecules 34, 110-115.
  • Cowman, M.K. and Matsuoka, S. (2001). "Model for Viscosity of Hyaluronan Solutions", American Chemical Society, San Diego, CA.
  • Feder-Davis, J., Hittner, D.M., and Cowman, M.K. (1991). "Comparison of Solution and Solid State Structures of Sodium Hyaluronan by 13C-NMR Spectroscopy.", in Water-Soluble Polymers: Synthesis, Solution Properties, and Applications, S.W. Shalaby, C.L. McCormick, and G.B. Butler, eds., ACS Symposium Series 467, American Chemical Society, Washington, D.C., pp. 493-501.
  • Hittner, D.M. and Cowman, M.K. (1987). "High Performance Gel Permeation Chromatography of Glycosaminoglycans: Column Calibration by Gel Electrophoresis." J. Chromatogr. 402, 149-158.
  • Kwei, T.K., Nakazawa, M., Matsuoka, S., Cowman, M.K., and Okamoto, Y. (2000) "The Concentration Dependence of Solution Viscosities of Rigid Polymers." Macromolecules 33, 235-236.
  • Lee, H.G. and Cowman, M.K. (1994). "An Agarose Gel Electrophoretic Method for Analysis of Hyaluronan Molecular Weight Distribution." Anal. Biochem. 219, 278-287.
  • Li, M., Rosenfeld, L., Vilar, R.E., and Cowman, M.K. (1997). "Degradation of Hyaluronan by Peroxynitrite." Arch. Biochem. Biophys. 341, 245-250.
  • Matsuoka, S. and Cowman, M.K. (2000). "Viscosity of Polymer Solutions Revisited" at Hyaluronan 2000 Conference, Wrexham, Wales. ", In press for 2002, in Hyaluronan, J.F. Kennedy, G.O. Phillips, P.A. Williams, V.C. Hascall, eds., Woodhead Publishing, Cambridge.
  • McKee, C.M., Penno, M.B., Cowman, M., Bao, C., and Noble, P.W. (1996). "Hyaluronan (HA) Fragments Induce Chemokine Gene Expression in Murine Alveolar Macrophages. The Role of HA Size and CD44." J. Clin. Invest. 98, 2403-2413.
  • Min, H. and Cowman, M.K. (1986). "Combined Alcian Blue and Silver Staining of Glycosaminoglycans in Polyacrylamide Gels: Application to Electrophoretic Analysis of Molecular Weight Distribution." Anal. Biochem. 155, 275-285.
  • Noble, P.W., McKee, C.M., Cowman, M., and Shin, H.S. (1996). "Hyaluronan Fragments Activate an NF-kB/I-kBa Autoregulatory Loop in Murine Macrophages." J. Exp. Med. 183, 2373-2378.
  • Turner, R.E., Lin, P., and Cowman, M.K. (1988). "Self-Association of Hyaluronate Segments in Aqueous NaCl Solution." Arch. Biochem. Biophys. 265, 484-495.