Retroviral Superantigens and Type 1 Diabetes Mellitus

Matters Arising diagnosed for less than three weeks), and four healthy Retroviral Superantigens and Type 1 adult controls. All subjects were of Caucasoid extrac-Diabetes Mellitus tion. RT-PCRs, with the Conrad et al. (1997) protocol and reagents, generated a band of the expected size (489 bp) for IDDMK 1,2 22 with U3-R-poly(A) primer from In 1994, Conrad et al. reported initial results that rekin-all samples, but in both the presence and absence of dled the debate about the involvement of an infectious RT. The U3-R-poly(A) primer also generated a 489 bp agent in type 1 diabetes. They analyzed the T cell (anti-band from genomic DNA samples. The RT-PCR product gen) receptor repertoire of T lymphocytes that had in-hybridized with the U3-R probe, as reported by Conrad vaded the pancreatic islets of two recently deceased, et al. (1997). However, DNase treatment of the RNA acutely diabetic patients. Antigen receptors carrying samples prevented generation of the PCR product and the V␤7 variable region were shown to be strikingly en-its detection by Southern blot. We could not achieve riched in the T cells, evoking the involvement of a super-specificity for RNA, by extending 5Ј the U3-R-poly(A) antigen (SAG). Recently, Conrad et al. (1997) (see also primer from four to 11Ts and/or by subtracting eight of Benoist and Mathis, 1997) described the isolation of a the 3Ј nucleotides, whereas the Pittsburgh and London hitherto unknown endogenous HERV-K10-like retroviral laboratories (see below) detected viral RNAs with modi-genome, IDDMK 1,2 22, related to mouse mammary tumor fied primers, although these were also not disease spe-virus, whose env gene encoded a SAG. The SAG stimu-cific. It should be noted, however, that the various modi-lated V␤7-bearing T cells, but not those that expressed fied primers were not identical. other V␤ genes. Transcripts derived from, or at least Conrad et al. (1997) added RT and PCR reagents to related to, the IDDMK 1,2 22 genome were detectable the same reaction tube. When we added the PCR re-by RT-PCR in the plasma of 10/10 newly diagnosed type 1 agents after the RT reactions, the yield of the U3-R-diabetes patients, but not from 10 age-matched, nondia-poly(A) product significantly increased, but again it was betic control individuals. In order to detect IDDMK 1,2 22 not specific for RNA. The increased yield was most likely RNA, Conrad et al. (1997) performed RT-PCR on total due to separation of Pwo, a DNA polymerase with 3Ј–5Ј …