Identification of candidate proteins binding to prion protein.

Prion diseases are disorders of protein conformation that produce neurodegeneration in humans and animals. Studies of transgenic (Tg) mice indicate that a factor designated protein X is involved in the conversion of the normal cellular prion protein (PrPC) into the scrapie isoform (PrPSc); protein X appears to interact with PrPC but not with PrPSc. To search for PrPC binding proteins, we fused PrP with alkaline phosphatase (AP) to produce a soluble, secreted probe. PrP-AP was used to screen a lambdagt11 mouse brain cDNA library, and six clones were isolated. Four cDNAs are novel while two clones are fragments of Nrf2 (NF-E2 related factor 2) transcription factor and Aplp1 (amyloid precursor-like protein 1). The observation that PrP binds to a member of the APP (amyloid precursor protein) gene family is intriguing, in light of possible relevance to Alzheimer's disease. Four of the isolated clones are expressed preferentially in the mouse brain and encode a similar motif.

Selective neuronal targeting in prion disease.

The pattern of scrapie prion protein (PrP(Sc)) accumulation in the brain is different for each prion strain. We tested whether the PrP(Sc) deposition pattern is influenced by the Asn-linked oligosaccharides of PrP(C) in transgenic mice. Deletion of the first oligosaccharide altered PrP(C) trafficking and prevented infection with two prion strains. Deletion of the second did not alter PrP(C) trafficking, permitted infection with one prion strain, and had a profound effect on the PrP(Sc) deposition pattern. Our data raise the possibility that glycosylation can modify the conformation of PrP(C). Glycosylation could affect the affinity of PrP(C) for a particular conformer of PrP(Sc), thereby determining the rate of nascent PrP(Sc) formation and the specific patterns of PrP(Sc) deposition.

Conversion of truncated and elongated prion proteins into the scrapie isoform in cultured cells.

The only known component of the infectious prion is a posttranslationally modified protein known as the scrapie isoform of the prion protein, PrPSc. Upon limited proteolysis, a protease-resistant fragment designated PrP 27-30 is formed. Using in vitro mutagenesis, we examined the role of the N and C termini in the formation of PrPSc in persistently infected, mouse neuroblastoma (ScN2a) cells. Neither deletion of amino acids 23-88, which are also removed by proteinase K in the formation of PrP 27-30, nor deletion of the five octapeptide repeats within this region altered synthesis of PrPSc. Elongation of PrP with one, two, four, or six octapeptide repeats in addition to the five found in wild-type PrP did not alter the synthesis of PrPSc. Truncation of the C terminus was accomplished by substituting a translation stop codon for the predicted glycosylinositol phospholipid (GPI) anchor-attachment signal corresponding to amino acids 231-254. Expression of this C-terminal PrP mutant in ScN2a cells produced PrPSc that appeared to lack a GPI anchor. We conclude that neither the GPI anchor nor the N-terminal 66 amino acids are required for the synthesis of PrPSc as measured by the acquisition of limited resistance to proteinase K digestion. Whether these truncated or elongated PrP molecules are competent to participate in the formation of infectious prions remains to be established.

The gene for the rat heat-shock cognate, hsc70, can suppress oncogene-mediated transformation.

In cells transformed by mutant mouse p53 plus ras, the former protein is found to be complexed with the heat-shock protein cognate hsc70. To determine whether hsc70 can directly affect neoplastic transformation, nonestablished rat embryo fibroblasts (REF) were transfected with rat genomic hsc70 DNA in conjunction with various oncogenes. We report here that the hsc70 gene could efficiently suppress focus induction by mutant p53 plus ras, as well as by myc plus ras. No inhibitory effect of hsc70 was detectable in assays monitoring the ability of REF to be immortalized by mutant p53, arguing against a nonspecific deleterious effect of the hsc70 genomic clone on REF survival and proliferation. Lines generated in the presence of the hsc70 plasmid produced augmented levels of hsc70. Plasmids encoding only short NH2-terminal fragments of hsc70 could also, in some cases, partially reduce oncogene-mediated focus formation. However, a maximal inhibitory effect required the production of a functional hsc70 protein. The data presented here raise the possibility that hsc70 may be directly involved in the modulation of oncogene-mediated transformation.

Simian virus 40 can overcome the antiproliferative effect of wild-type p53 in the absence of stable large T antigen-p53 binding.

In simian virus 40 (SV40)-transformed cells, a tight complex is formed between the viral large T antigen (large T) and p53. It has been proposed that this complex interferes with the antiproliferative activity of p53. This notion was tested in primary rat fibroblasts by assessing the ability of SV40-mediated transformation to be spared from the inhibitory effect of wild-type (wt) p53. The data indicate that relative to transformation induced by myc plus ras, SV40-plus-ras-mediated focus formation was indeed much less suppressed by p53 plasmids. A majority of the resultant cell lines made a p53 protein with properties characteristic of a wt conformation. Furthermore, cell lines expressing stably both SV40 large T and a temperature-sensitive p53 mutant continued to proliferate at a temperature at which this p53 assumes wt-like properties and normally causes a growth arrest. Surprisingly, at least partial resistance to the growth-inhibitory effect of wt p53 was also evident when transformation was mediated by an SV40 deletion mutant, encoding a large T which does not bind p53 detectably. In addition to supporting the idea that SV40 can overcome the growth-restrictive activity of wt p53, these findings strongly suggest that at least part of this effect does not require a stable association between p53 and large T.

Analyses of transplanted murine tumors for HSV DNA sequences.

Meignier et al. (1986) report the results of exposure of C57BL/6NCr mice to vaginal plugs containing live or inactivated herpes simplex virus 1 or 2 (HSV-1 or HSV-2) or recombinant viruses 5 times a week for up to 114 weeks. Genital organs showing abnormalities were transplanted into nude mice. Of 33 transplants, 13 produced subcutaneous tumors in nude mice and 12 were subsequently transplanted into C57BL/6NCr mice. We report that the DNA extracted from coded tumor tissues of nude mice and from normal viscera of the same rodents did not hybridize with HSV-1 and HSV-2 DNA probes representing the viral genomic regions shown previously to be capable of morphologically transforming cells in culture. The sensitivity of the assays was such that we could detect 0.5 copies of the HSV sequences of complexity equal to or greater than 1 Kbp per cell DNA equivalent. To control for the sensitivity of the assays in the actual hybridizations, the tumor-cell DNA was also hybridized with a beta-globin mouse DNA probe. A striking feature of these control hybridizations was the detection of beta-globin polymorphism in some nude mouse tumors. The beta-globin polymorphism allowed us to conclude that the analyzed tissues contained significant amounts of the tumor cells occurring in the C57BL/6NCr mice.