The therapeutic potential of enzyme replacement in lysosomal storage disorders has remained largely unfulfilled, perhaps because of negative reactions to the initial disappointing results. Despite the existence of several animal models that can be utilized to explore solutions to the problems of exogenous enzyme targeting, the interest in ERT prevalent during the 1970's seems to have subsided to be replaced by active interest in bone marrow transplantation (BMT, Krivit and Paul [1986]). This is a logical approach to enzyme replacement in storage disorders of the RE system, and indeed some encouraging results have been obtained. However, in addition to having high morbidity and mortality, in the ultimate analysis BMT presents the same targeting problems as conventional ERT. In our opinion, these problems can be solved more easily in the case of ERT by exploiting the existing cellular uptake mechanisms and infusing enzymes whose structure has been suitably modified by simple biochemical manipulations. Accordingly, we have explored a methodology that takes advantage of negative charges on the cell surface to obtain nonspecific but effective membrane binding of beta-hex coupled to the highly positively charged PLL, followed by internalization and routing to the lysosomes. This system increases uptake of exogenous enzyme by some neurons in vitro and possibly in vivo, but its efficiency depends on the cells' endocytic activity that, in the case of neuronal soma, apparently is low. Thus, we have chosen as recognition marker for specific neuronal uptake a nontoxic fragment of TTx that is efficiently taken up by these cells. The initial results are encouraging; they support our contention that effective enzyme replacement methodologies can be devised, and encourage us to continue our work in this direction. Finally, recombinant DNA techniques are now being applied to a number of LSD, and the genes for several of the pertinent enzymes have been or are being isolated. In addition to representing a first step towards gene replacement therapy, the results of this work will permit the generation of large amounts of human enzymes from bacteria by recombinant DNA methods, thus obviating the problem of enzyme supply for ERT. Since human lysosomal enzymes obtained from bacteria will be nonglycosylated, to obtain cell uptake it will be necessary to resort to the type of modifications that we are trying to develop at this time, i.e., covalent linkage to moieties that allow non-glycosyl-mediated cellular uptake. Thus, our work on beta-hex may provide a model for biochemical manipulations of bacterially produced enzymes applicable to several LSD.