Loss of heterozygosity in certain human embryonal tumours implicates a tumour-suppressor gene at chromosome 11p15.5 and selective loss of maternal alleles suggests that this gene is paternally imprinted. The human H19 gene maps to 11p15.5, is expressed in differentiating fetal cells and is paternally imprinted. We report here that two embryonal tumour cell lines, RD and G401, showed growth retardation and morphological changes when transfected with an H19 expression construct. More importantly, clonogenicity in soft agar and tumorigenicity in nude mice were abrogated in the G401-H19 transfectants. In addition to demonstrating its tumour-suppressor potential, this transfection system should help structural and functional studies of the enigmatic H19 gene.

To test the potential role of H19 as a tumour suppressor gene we have examined its expression and DNA methylation in Wilms' tumours (WTs). In most WTs (18/25), H19 RNA was reduced at least 20-fold from fetal kidney levels. Of the expression-negative tumours ten retained 11p15.5 heterozygosity: in nine of these, H19 DNA was biallelically hypermethylated and in two cases hypermethylation locally restricted to H19 sequences was also present in the non-neoplastic kidney parenchyma. IGF2 mRNA was expressed in most but not all WTs and expression patterns were consistent with IGF2/H19 enhancer competition without obligate inverse coupling. These observations implicate genetic and epigenetic inactivation of H19 in Wilms' tumorigenesis.

Medical Sciences. In the article "t(11;22)(q23;q11.2) in acute myeloid leukemia of infant twins fuses MLL with hCDCrel, a cell division cycle gene in the genomic region of deletion in DiGeorge and velocardiofacial syndromes" by Maureen D. Megonigal, Eric F. Rappaport, Douglas H. Jones, Terrence M. Williams, Brian D. Lovett, Kara M. Kelly, Paul H. Lerou, Thomas Moulton, Marcia L. Budarf, and Carolyn A. Felix, which appeared in number 11, May 26, 1998, of Proc. Natl. Acad. Sci. USA (95, pp. 6413-6418), the authors wish to note the following corrections. On page 6414, column 2, line 4, the text should read: "The 24-l ligation reaction mixture contained 0.5 g" not "0.05 g" as printed. Also, on page 6415, column 1, line 56, the text should read: "AmpliTaq DNA polymerase, all four dNTPs (each at 200 M)" not "(each at 250 M)" as printed.Medical Sciences. In the article "Potent inhibition of human immunodeficiency virus type 1 replication by an intracellular anti-Rev single-chain antibody" by Lingxun Duan, Omar Bagasra, Mark A. Laughlin, Joseph W. Oakes, and Roger J. Pomerantz, which appeared in number 11, May 24, 1994, of Proc. Natl. Acad. Sci. USA (91, 5075-5079), the undersigned authors wish to note the following: "The sequence of the heavy chain of the D8 anti-Rev single chain variable fragment (SFv) has been reanalyzed and found to be not what was reported in the article. It is likely that the coding DNA was derived from the fusion partner cell line used to make the original D8 hybridoma, and not from the heavy chain gene expressed by the B cell precursor to this hybridoma. There is a deletion in the framework region 3 (FR3) leading to a framehsift in CDR3 and downstream regions of the heavy chain gene. Individual nucleotide differences make this sequence very close to that described [Thammana, P. (1994)