No abstract

The nucleotide sequence of four anti‐(Glu60‐Ala30‐Tyr10)n (GAT) monoclonal gamma 1 heavy chain mRNAs was determined from codon 10 to 120. This sequence overlaps with the NH2‐terminal amino acid sequence, allowing elucidation of the complete protein sequence encompassing regions VH, D and JH. These sequences, which are highly conserved, indicate that anti‐GAT antibodies expressing the same public idiotypic specificities represent a paucigene system, which uses at least two D‐J combinations leading to functionally similar hypervariable regions involved in the recognition of the dominant Glu‐Tyr determinant. D regions are encoded by D genes which are closely related either to the D‐SP2 or the D.FL16 germ line gene cores.

Curvature and flexibility are structural properties of central importance to genome function. However, due to the difficulties in finding suitable experimental conditions, methods for studying one without the interference of the other have proven to be difficult. We propose a new approach that provides a measure of inherent flexibility of DNA by taking advantage of two powerful techniques, X-ray crystallography and nuclear magnetic resonance. Both techniques are able to detect local curvature on DNA fragments but, while the first analyzes DNA in the solid state, the second works on DNA in solution. Comparison of the two data sets allowed us to calculate the relative contribution to flexibility of the three rotations and three translations, which relate successive base pair planes for the ten different dinucleotide steps. These values were then used to compute the variation of flexibility along a given nucleotide sequence. This allowed us to validate the method experimentally through comparisons with maps of local fluctuations in DNA molecule trajectory constructed from atomic force microscopy imaging in solution. We conclude that the six dinucleotide-step parameters defined here provide a powerful tool for the exploration of DNA structure and, consequently will make an important contribution to our understanding of DNA-sequence-dependent biological processes.

and 1D6partement d'Immunologie de l'Institut Pasteur, 28. rue du Docteur Roux, 75724 Paris Cedex 15, France Communicated by M.FougereauHeavy and light chain variable regions from eight monoclonal Ab2 (anti-idiotypic) antibodies of the GAT antigen, a (Glu60 Ala30 Tyr10)n co-polymer, have been analyzed by direct mRNA sequencing. Three mAb2s were directed against private idiotopes and used various VH-D-JH and Vk-Jk combinations. By contrast, the five 'anti-public' mAb2 antibodies used a very restricted repertoire, including all gene segments. Interestingly, within their D regions, Glu-Glu-Tyr or Tyr-Tyr-Glu sequences were reminiscent of the original (GAT) antigen and may act as possible internal images. A striking observation was that two mAb2 antibodies shared the same V-D-J sequence although derived from separate fusions. As this D sequence, 33 nucleotides long, has not been described so far, it is suggested that it may be encoded by a new germline D gene, acting as a crucial regulatory element in a GAT germ-line idiotypic network. An alternative model that may lead to the construction of this D segment by 'odd' rearrangements from pre-existing already reported sequences is also presented.

NH2-terminal amino acid sequences of heavy and light chains of seven poly(Glu'Ala30Tyr'0) (GAT) specific hybridoma products derived from DBA/2 and (DBA/2 X BALB/c)Fl hybrid mice and those of BALB/B polyclonal antibodies have been determined over the first 40 residues. Comparison of these sequences with those of nine other GAT or poly(GluWMaW) (GA) specific hybridoma products previously reported allowed the following conclusions. (i) Sequences of hybridoma H and L chains are present in the pool of polyclonal antibodies. (ii) The public CGAT (or pGAT) idiotypic specificities are strictly confined to antibodies exhibiting limited heterogeneity with regard to both the variable heavy (VH) and the variable K (Va) sequences that may be accounted for by one and two germ-line genes, respectively. (iii) The public idiotypic specificities GA-1, expressed by some anti-GAT and most anti-GA antibodies, make use of the same (or similar) VH germ-line genes as the CGAT or pGAT antibodies butpossess a distinctive VK sequence. (iv) Antibodies expressing neither of the alternative public specificities mentioned above appear to be more heterogeneous and express VH and V,, sequences that were found to differ from the basic structures defining the CGAT (pGAT) or GA-1 correlates. It is concluded that CGAT (or pGAT) and GA-1 public idiotypic specificities are germ-line markers of both VH and Vi regions, an observation in agreement with previously reported serological data.In various inbred strains of mice, the random terpolymer (Glu-60 Ala30Tyr'0)n (GAT) elicits specific antibodies, most of which express crossreactive, or public, idiotypic specificities. These were initially defined by using heterologous anti-idiotypic antisera prepared against polyclonal anti-GAT antibodies from DL.LP (1, 2) or BALB/c (3, 4) mice. The -guinea pig antiserum against D1.LP anti-GAT antibodies identified idiotypic markers that were called CGAT (1, 2), whereas idiotypic specificities of BALB/ c anti-GAT antibodies detected by a rabbit antiserum were termed pGAT (3,4). For simplification, it may be considered that both reagents identify similar or identical public idiotypes, hereafter referred to as pGAT or CGAT idiotypes.