A novel strategy for anti-viral intervention of hepatitis B virus (HBV) through the disruption of the proper folding and transport of the hepadnavirus glycoproteins is described. Laboratory reared woodchucks chronically infected with woodchuck hepatitis virus (WHV) were treated with N-nonyl-deoxynojirimycin (N-nonyl-DNJ), an inhibitor of the endoplasmic reticulum (ER) alpha-glucosidases. The woodchucks experienced significant dose dependent decreases in enveloped WHV, resulting in undetectable amounts in some cases. The reduction in viremia correlated with the levels of hyperglucosylated glycan in the serum of treated animals. This correlation supports the mechanism of action associated with the drug and highlights the extreme sensitivity of the virus to this type of glycan inhibitor. At N-nonyl-DNJ concentrations that prevented WHV secretion, the glycosylation of most serum glycoproteins appeared unaffected, suggesting great selectivity for this class of therapeutics. Indeed, this may account for the low toxicity of the compound over the treatment period. We provide the first evidence that glucosidase inhibitors can be used in vivo to alter specific steps in the N-linked glycosylation pathway and that this inhibition has anti-viral effects.

5625Correction. In the article "Molecular basis of adult-onset and chronic GM2 gangliosidoses in patients of Ashkenazi Jewish origin: Substitution of serine for glycine at position 269 of the a-subunit of (3-hexosaminidase" by Barry H. Paw

The magnetic spin-lattice relaxation rates of solvent water nuclei are known to increase upon addition of diamagnetic solute protein. This enhancement of the relaxation rate is a function of magnetic field, and the orientational relaxation time of the protein molecules can be deduced from analysis of the field-dependent relaxation rates. Although the nature of the interactions that convey information about the dynamics of protein motion to the solvent molecules is not established, it is known that there is a contribution to the relaxation rates of solvent protons that plays no role in the relaxation of solvent deuterons and 17O nuclei. We show here that the additional interaction arises from a cross-relaxation process between solvent and solute protons. We introduce a heuristic three-parameter model in which protein protons and solvent protons are considered as two separate thermodynamic systems that interact across the protein-solvent interface. The three parameters are the intrinsic relaxation rates of each system and a cross-relaxation term. The sign of the latter term must always be positive, for all values of magnetic field, in order for magnetization energy to flow from the hotter to the cooler system. We find that the magnetic field-dependence of the cross-relaxation contribution is much like that of the remaining solvent proton relaxation, i.e., about the same as the deuteron relaxation field dependence. This finding is not compatible with the predictions of expressions for the cross-relaxation that have been used by other authors, but not applied to data over a wide range of magnetic field strength. The model predicts that the relaxation behavior of both the protein protons and the solvent protons is the sum of two exponentials, the relative contributions of which would vary with protein concentration and solvent isotopic composition in a fashion suggestive of the presence of two classes of protein protons, when there is in reality only one. This finding has immediate implications for the interpretation of published proton relaxation rates in complex systems such as tissues; these data should be reexamined with cross-relaxation taken into account.

Cross-polarization magic-angle-spinning nuclear magnetic resonance spectroscopy has been used to determine insect cuticle composition and cross-link structure during sclerotization or tanning. Unsclerotized cuticle from newly ecdysed pupae of the tobacco hornworm, Manduca sexta L., had a high protein content with lesser amounts of lipid and chitin. Concentrations of chitin, protein, and catechol increased substantially as dehydration and sclerotization progressed. Analysis of intact cuticle specifically labeled with carbon-13 and nitrogen-15 revealed direct covalent linkages between ring nitrogens of protein histidyl residues and ring carbons derived from the catecholamine dopamine. This carbon-nitrogen adduct was present in chitin isolated from cuticle by alkaline extraction and is probably bound covalently to chitin. These data support the hypothesis that the stiffening of insect cuticle during sclerotization results primarily from the deposition of protein and chitin polymers and their crosslinking by quinonoid derivatives of catecholamines.