We have examined the structure and expression of the products associated with the t(2;13)(q35;q14) translocation associated with alveolar rhabdomyosarcoma. The chromosome 13 gene (FKHR) is identified as a member of the fork head domain family of transcription factors characterized by a conserved DNA binding motif. Polymerase chain reaction analysis demonstrates that a 5'PAX3-3' FKHR chimaeric transcript is expressed in all eight alveolar rhabdomyosarcomas investigated. Immunoprecipitation experiments detect the predicted fusion protein. These findings indicate that the t(2;13) generates a potentially tumorigenic fusion transcription factor consisting of intact PAX3 DNA binding domains, a truncated fork head DNA binding domain and C-terminal FKHR regions.

When suspension cultures of human promyelocytic leukemia cells (line HL60) were treated with 12-O-tetradecanoylphorbol 13-acetate

We have determined that PAX3 (found previously to be mutated in Waardenburg syndrome) is the chromosome 2 locus rearranged by the t(2;13)(q35;q14) translocation of the paediatric solid tumour alveolar rhabdomyosarcoma. The rearrangement breakpoints occur within an intron downstream of the paired box and homeodomain-encoding regions. Upstream PAX3 sequences hybridize to a novel transcript in t(2;13)-containing lines. Cloning and characterization of this novel transcript indicate that the translocation juxtaposes the PAX3 DNA binding elements with chromosome 13 sequences, suggesting formation of a hybrid transcription factor. Therefore, PAX3 gene alterations are associated with two completely unrelated human diseases.

SummaryWe have examined at the molecular level the CDR3 and adjacent regions in peripheral blood B lymphocytes of normal individuals. A total of 111 sequences (12-28 sequences from six individuals) were obtained after cloning of the polymerase chain reaction-amplified segments into plasmids or phage . The average length of the VDJ joining was 109 nucleotides, with a range from 79 to 151 . Approximately 75% of the sequences were in frame when translated into amino acids . Among the JH segments, J H4 was found most frequently (in 52 .5% of the sequences), and JH 1 and JH 2 segments the least frequently (-I% of the clones) . A polymorphic JH6 gene with a one-codnn deletion accompanied by a base change was present in two of six patients. Preferential breakpoints were found for JH 2, J.3, J 4, and Jx5, although the breakpoints of JH6 were distributed more heterogenously.In -90% of the cases, significant homology of the D regions with published D sequences was found . Preferential usage of a particular coding frame was observed in in-frame sequences utilizing DA, D21/9, and DMl segments. However, in general, all coding frames of germline D genes were used to generate CDR3s . Eight sequences that have a DN1-like D sequence with two base changes at the same positions were identified, suggesting the likely existence of a new germ line D gene belonging to the DN families. Using probes specific for a particular CDR3, the frequency of a specific B cell clone in the peripheral blood of normal individuals was estimated to be at most as high as 1/20,000 .T he most variable region of the immunoglobulin heavy chain is the third complementarity determining region (CDR3) (1, 2) . This region spans the junction between the variable (V ) diversity (D), and junctional 0,,) segments in the rearranged IgH genes (1, 2) . The hypervariability of this region is due to the combinatorial assortment of the many V , D, and JH segments that are utilized to generate a particular CDR3, to the imprecise joining mechanisms that include deletion of bases from the potential coding regions of each segment to be joined (3), and the addition of new bases that can be enzymatically added at the point of joining (N regions) (4) . Finally, somatic mutations of the rearranged region can contribute to the production of higher affinity antibodies (5) .The nucleotide sequences of all the human D genes, estimated to be -30 in number (6), have not been fully defined, and questions remain about the relative usage of different D genes used in VDJjoinings during development and in adult individuals . Furthermore, the characteristics of the extent of base excision and addition, including the identification of preferred sequence boundaries for the V., D, and J regions, have not been well delineated. The relative frequency of inframe translation products reflecting productive rearrangements, and the possibility of specific translation frames being preferred for particular D gene families have not been determined on a large sample size.To address these questions, we ha...