Long-term gene therapy in the CNS: Reversal of hypothalamic diabetes insipidus in the Brattleboro rat by using an adenovirus expressing arginine vasopressin

Messenger RNAs occur within the axons of magnocellular hypothalamic neurons known to secrete oxytocin and vasopressin. In Brattleboro rats, which have a genetic mutation that renders them incapable of vasopressin expression and secretion and thus causes diabetes insipidus, injection into the hypothalamus of purified mRNAs from normal rat hypothalami or of synthetic copies of the vasopressin mRNA leads to selective uptake, retrograde transport, and expression of vasopressin exclusively in the magnocellular neurons. Temporary reversal of their diabetes insipidus (for up to 5 days) can be observed within hours of the injection. Intra-axonal mRNAs may represent an additional category of chemical signals for neurons.

Adenoviruses are attracting increasing attention as general purpose mammalian cell expression vectors, as recombinant vaccines, and potentially as vectors for gene therapy. Not only is the adenovirus genome relatively easy to manipulate by recombinant DNA techniques, but adenovirus vectors are relatively stable, grow to high titers, and can transduce a variety of cell types in cell culture and in vivo. Vectors can be designed that are either replication competent or replication defective and, in the latter case, are highly efficient at delivering and expressing genes in mammalian cells without resulting in cell killing. Methods are described for growing, titrating, and purifying adenoviruses, for extracting viral DNA from purified virions and from infected cells, for rescuing inserts of foreign DNA into the viral genome, and for assessing expression of inserted genes in adenovirus vectors.

Seven mouse monoclonal antibodies (IgGs) were produced against rat neurophysins (NPs). Three were specifically directed against vasopressin-associated NP (NP-AVP), and four were specific for oxytocin- associated NP (NP-OT). These specificities were observed in liquid phase assays, immunoblot, and immunoprecipitation experiments. Homozygous Brattleboro rat tissues and extracts, which do not contain vasopressin or NP-AVP, did not react with the anti-NP-AVP antibodies but reacted with high affinity to the anti-NP-OT antibodies. In immunoprecipitation assays the antibodies brought down the appropriate NPs as well as their precursor molecules synthesized in vivo with no detectable cross-reactivity. In solid phase assays where the antigens were presented in a different manner, there was a significant cross- reactivity of the anti-NP-AVP antibodies with NP-OT. The extent of this cross-reactivity in solid phase correlated with the cross-reactivities of the antibodies observed in immunocytochemical studies. These solid phase (and immunocytochemical) data demonstrated that liquid phase specificities and absorption controls of antibodies are inadequate to assess their immunocytochemical (solid phase) specificities. Posterior pituitary extracts from the mouse and frog, as well as purified NPs from the rat, cow, and human were studied for their cross-reactivities to two of the antibodies, PS 36 and PS 45. In liquid phase assays the anti-rat NP-OT antibody, PS 36, reacted only with rat and mouse NPs and did not cross-react with NPs from any of the other species. In contrast, the anti-rat NP-AVP antibody, PS 45, was cross-reactive across species lines including an NP-like antigen extracted from frog posterior pituitaries. Immunoblot staining with these antibodies showed heterogeneity of NP-AVP and NP-OT in the rat posterior pituitary. Analysis of the epitopes for PS 36 and PS 45 indicated the antigenic determinants were located near amino acid positions 80 to 81 in NP-OT and 75 to 86 in NP-AVP, respectively.