Search

Background: The immunoglobulin framework has been mutagenized to engineer recombinant libraries of proteins as potential diagnostics and novel catalysts, although the often shallow binding cleft may limit the utility of this framework for binding diverse small organic molecules. By contrast, the glutathione S-transferase (GST) family of enzymes contains a deep binding cleft, which has evolved to accommodate a broad range of hydrophobic xenobiotics. We set out to determine whether GST molecules with novel ligand-binding characteristics could be produced by random mutagenesis of segments of the binding cleft., Results: We have identified two ligand-recognition segments (LRSs) in human GST P1, which are near the active site in the folded protein, but have characteristics indicating that the integrity of their sequence is not essential for the overall structure or activity of the protein. Libraries of GST P1-derived proteins were produced by substituting randomized sequences for an LRS or inserting random sequences into an LRS. The recombinant proteins in the libraries, collectively designated as 'glubodies,' generally retain enzymatic activity but differ markedly both from each other and from the parent enzyme in sensitivity to inhibition by diverse small organic compounds. In some instances, a glubody is inhibited by completely novel structures., Conclusions: We have shown that a non-antibody framework can be used to create large libraries of proteins with a wide range of binding specificities for small organic molecules. The glubodies provide a rich source of data for correlating the structural and functional features of proteins relevant to ligand binding. The criteria applied for identifying an LRS in GST P1 are generally applicable to other protein frameworks.

Two full-length cDNA clones encoding rabbit glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were isolated from a lambda gt10 rabbit spleen cDNA library and sequenced. As predicted from the open reading frame (ORF) in vitro translation of a sense orientation GAPDH cDNA clone yielded a protein product with a molecular mass of 37 kDa. Rabbit GAPDH exhibits a high degree of homology to the mouse, rat, hamster, chicken and human GAPDH on both the nucleotide (nt) and amino acid (aa) levels.

During the switch in expression of an immunoglobulin class, the gene segment encoding the constant region of the heavy chain is replaced in a way that leads to a deletion. Three different models of how this deletion is generated have been proposed: recombination between homologs, unequal sister chromatid exchange, and looping out and deletion. While none of the predicted recombination products of the first two models have been found, the products of the looping out--inversions and circular DNA--have been isolated. Thus looping out and deletion appears to be the appropriate model to explain the genetic events leading to the immunoglobulin heavy chain class switch.

The class of immunoglobulin is defined by the constant region of its heavy chain. When a B lymphocyte switches the class of heavy chain it produces, the constant region of mu-type heavy chain is replaced; this occurs through a DNA rearrangement that brings the gene segment encoding the new constant region close to the VDJ segment encoding the variable region. The pre-B-cell line 18-81, which switches from heavy chain mu to gamma 2b production in culture, occasionally abnormally rearranges the heavy chain locus so that DNA sequences between the switch regions of mu and gamma 2b are inverted. Because looping-out is an intermediate step in generating an inversion, the switch rearrangement could occur by looping-out and deletion. Provided that recombination is reciprocal, this would produce a circle of DNA. Indeed, circular DNA molecules have been isolated as products of rearrangement among gene segments encoding the variable regions of the T-cell receptor and of the immunoglobulin heavy chain and light chain. But whereas the breakpoints for the variable region rearrangement are precisely defined, the breakpoints for any given heavy chain class switch are scattered over a length of greater than 6 kilobases, including both switch regions. We have now isolated circular DNA containing the sequences deleted by class-switching, thereby showing that the immunoglobulin heavy chain class switch occurs through looping-out and deletion.

Immunoglobulin genes are generated during differentiation of B lymphocytes by joining gene segments. A mouse pre-B cell contains a functional immunoglobulin heavy-chain gene, but no light-chain gene. Although there is only one heavy-chain locus, there are two light-chain loci: kappa and lambda. It has been reported that kappa loci in the germ-line configuration are never (in man) or very rarely (in the mouse) present in cells with functionally rearranged lambda-chain genes. Two explanations have been proposed to explain this: (a) the ordered rearrangement theory, which postulates that light-chain gene rearrangement in the pre-B cell is first attempted at the kappa locus, and that only upon failure to produce a functional kappa chain is there an attempt to rearrange the lambda locus; and (b) the stochastic theory, which postulates that rearrangement at the lambda locus proceeds at a rate that is intrinsically much slower than that at the kappa locus. We show here that lambda-chain genes are generated whether or not the kappa locus has lost its germ-line arrangement, a result that is compatible only with the stochastic theory.

The authors have developed a method to measure the rate of spontaneous mutations taking place in IgH, the gene encoding the immunoglobulin heavy chain. When an amber chain-termination codon mutates to a sense codon, translation of the polypeptide chain will be completed, and mutant cells producing the heavy chain can be detected with a fluorescent labelled antibody. The protocol used is the compartmentalization test which minimizes any effect of selection. In subclones of the pre-B lymphocyte line 18-81, the spontaneous mutation rate in the part of IgH encoding the variable region is somewhat greater than 10(-5) mutations per base pair per generation. This supports the hypothesis that hypermutation is not dependent on cell stimulation by an antigen. In a hybrid between a cell of this line and a myeloma (which represents the terminal stage of the B-cell lineage), the mutation rate was too low to be determined by this test, less than 10(-9). When the same loss to gain procedure system was used with an opal chain-terminating codon in the part of IgH encoding the constant region (C mu), a high rate of reversion by deletion was found. Long (more than one exon) and short (less than one exon) deletions occurred at rates of 1.7 x 10(-5) and 1.4 x 10(-7) per generation, respectively. It is thought that the high rate of deletion is not related to somatic hypermutation but rather to DNA rearrangement during the heavy-chain class switch, which is occurring in these pre-B cell lines. The point mutation rate was too low to be detected above the background of deletion mutants, less than 5 x 10(-8). The immunoglobulin mutator system works weakly, if at all, on two other, nonimmunoglobulin, genes tested: B2m (beta 2 microglobulin) and the gene for ouabain resistance.

When termination codons were introduced into exons of the gene for Ig mu chain, steady-state levels of mu mRNA were reduced, both at the pre-B cell stage and at the plasma cell stage. A termination codon in the variable region gene segment and a termination codon in the second exon of the constant region gene segment had effects of similar magnitude. When the termination codon was deleted, the original level of mRNA was restored. The rate of mu gene transcription was the same whether or not a termination codon was present. Therefore, the termination codons must reduce the amount of the mRNA by reducing its stability. Since the introduced termination codons prematurely terminate translation and, in so doing, change the ribosome load on the mRNA, we conclude that mu mRNA stability is conferred in part by ribosomal protection from enzymatic degradation. We propose that the differences in mu mRNA stability during B lymphocyte differentiation are due to different amounts of ribosomes available for translation.

Insulin receptors were solubilized from rat liver microsomes by the nonionic detergent Triton X-100. After gel filtration of the extract on Sepharose CL-6B, two insulin-binding species (peak I and peak II) were obtained. The structure and binding properties of both peaks were characterized. Gel filtration yielded Stokes radii of 9.2 nm (peak I) and 8.0 nm (peak II). Both peaks were glycoproteins. At 4 degrees C peak I showed optimal insulin binding at pH 8.0 and high ionic strength. In contrast, peak II had its binding optimum at pH 7.0 and low ionic strength, where peak I binding was minimal. For peak I the change in insulin binding under different conditions of pH and ionic strength was due to a change in receptor affinity only. For peak II an additional change in receptor number was found. Both peaks yielded non-linear Scatchard plots under most of the buffer conditions examined. At their binding optima at 4 degrees C the high affinity dissociation constants were 0.50 nM (peak I) and 0.55 nM (peak II). Sodium dodecyl sulfate/polyacrylamide gel electrophoresis of peak I revealed five receptor bands with Mr 400 000, 365 000, 320 000, 290 000, and 245 000 under non-reducing conditions. For peak II two major receptor bands with Mr 210 000 and 115 000 were found. The peak II receptor bands were also obtained after mild reduction of peak I. After complete reduction both peaks showed one major receptor band with Mr 130 000. The reductive generation of the peak II receptor together with molecular mass estimations suggest that the peak I receptor is the disulfide-linked dimer of the peak II receptor. Thus, Triton extracts from rat liver microsomes contain two receptor species, which are related, but differ considerably in their size and insulin-binding properties.

It is established that somatic mutation is an important source of antibody diversity in vivo. It is also established that Igh-V gene segments are hypermutable in vitro. This is not a completely satisfactory situation. While there is no reason to believe that Igh-V genes are not hypermutable in vivo as well, direct experimental evidence is lacking. Perhaps experiments with transgenic mice will soon fill this gap. It is not so clear how much higher than normal the rate of hypermutation is. As far as we are aware, there are no direct measurements of mutation rates per base pair per cell generation in mammals, certainly not for lymphocyte cell lines. For a variety of reasons, it is difficult to measure very low mutation rates. The general consensus is that the normal rate should be somewhere between 10(-10) and 10(-12) mutations per base pair per cell generation. Therefore, an experiment designed to directly determine a rate using the compartmentalization test would involve hundreds of cultures, each containing at least 10(9) cells. It is not a trivial problem to find one or a few mutants among so many cells. It is simple to study mutation to resistance to a drug, for example, ouabain or azaguanine, but, as we discussed, there are technical and conceptual pitfalls. The vast excess of dead cells influences the growth of a few mutant cells, particularly in lymphocyte cell lines. Even if this problem could be solved, the mutation rate so obtained would be "per gene(s)" and not "per base pair". The problems associated with cytotoxic agents can be avoided by immunofluorescence methods in conjunction with selective cloning or cell sorting. Using these techniques, we have carried out extensive experiments to determine whether the immunoglobulin mutator system acts, at least partially, on genetic elements other than those in or near the heavy chain variable region gene segment. For an opal termination codon in a heavy chain constant region gene segment, the rate of reversion was less than 10(-7) per base pair per cell generation. This upper limit was fixed by the high rate of small deletions at the heavy chain locus. For an allotype mutation at B2m, the gene encoding beta 2 microglobulin, the rate of mutation was less than 10(-8). This upper limit could be lowered by at least two orders of magnitude by using a high-speed cell sorter.