Instability of immunoglobulin genes in S107 cell line

Immunoglobulin hypermutation in cultured cells

Studies of endogenous and engineered Ig genes in mice have begun to reveal some of the cis-acting regions that are involved in the somatic hypermutation of variable regions in vivo. These studies suggest that the initiation of transcription plays a role in this process. However, it will be difficult to identify and manipulate the individual genetic elements and the trans-acting proteins that regulate and target the mutational events using solely in vivo assays. These studies would be greatly facilitated if constructs containing the genetic elements that are essential for V-region mutation could be transfected into cultured cells and undergo high rates of V-region mutation in vitro, and if permissive and non-permissive cell lines could be identified. Such in vitro systems would also allow a detailed molecular and biochemical analysis of this process. Here, we discuss some of the in vitro systems that have been developed and use data from our own studies in cultured cells to illustrate the potential benefits of studying V-region hypermutation in model in vitro systems.

An immunoglobulin mutator that targets G.C base pairs

Hypermutation can be defined as an enhancement of the spontaneous mutation rate which the organism uses in certain types of differentiated cells where a high mutation rate is advantageous. At the immunoglobulin loci this process increases the mutation rate > 10(5)-fold over the normal, spontaneous rate. Its proximate cause is called the immunoglobulin mutator system. The most important function of this system is to improve antibody affinity in an ongoing response; it is turned on and off during the differentiation of B lymphocytes. We have established an in vitro system to study hypermutation by transfecting a rearranged mu gene into a cell line in which an immunoglobulin mutator has been demonstrated. A construct containing the mu gene and the 3' kappa enhancer has all the cis-acting elements necessary for hypermutation of the endogenous gene segments encoding the variable region. The activity of the mutator does not seem to depend strongly on the position of the transfected gene in the genome. The mutator is not active in transformed cells of a later differentiation stage. It is also not active on a transfected lacZ gene. These results are consistent with the specificity of the mutator system being maintained and make it possible to delineate cis and trans mutator elements in vitro. Surprisingly, the mutator preferentially targets G-C base pairs. Two hypotheses are discussed: (i) the immunoglobulin mutator system in mammals consists of several mutators, of which the mutator described here is only one; or (ii) the primary specificity of the system is biased toward mutation of G-C base pairs, but this specificity is obscured by antigenic selection.

Immunoglobulin variable region hypermutation in hybrids derived from a pre-B- and a myeloma cell line

Somatic mutation of the variable (V) regions of immunoglobulin genes occurs in vivo at rates that have been estimated to be between 10(-3) and 10(-4) per bp per generation. To study this process in vitro, the 18.81 pre-B-cell line and hybrids derived by fusing 18.81 to the NSO myeloma fusion partner were transfected with a mu heavy-chain construct containing a nonsense mutation in the V region (Vn) or the constant region (Cn). Mutation was quantitated by reversion analysis using the ELISA spot assay to detect single cells secreting IgM. Fluctuation analysis revealed that V-region mutations spontaneously occurred in 18.81 cells at an average rate of 5.8 x 10(-6) per bp per cell generation and in selected 18.81-NSO hybrids at greatly increased rates of 1.6 x 10(-3) to 5.8 x 10(-4) per bp per generation. The Vn construct also reverted frequently in transgenic mice, indicating that it contained sufficient information to mutate at high rates both in vivo and in vitro. Sequence analysis of reverted genes revealed that reversion was due to point mutations. Since the rates and nature of the mutations that are occurring in these transfected genes are similar to those reported in vivo, it should be possible to use this system to identify the cis-acting sequences and trans-acting factors that are responsible for V-region somatic hypermutation.

A well-differentiated B-cell line is permissive for somatic mutation of a transfected immunoglobulin heavy-chain gene

pSV2neo plasmids containing an IgM heavy-chain gene with nonsense mutations in either the variable (V) or the constant (C) region were transfected into four differentiated mouse plasma cell lines: S107 and the NSO fusion partner (myeloma cell lines) and 2C3 and 36.65 (hybridoma cell lines). The frequencies of reversion of the nonsense mutations in multiple independent transfectants were determined with the spot ELISA and rates of reversion were calculated by fluctuation analysis. Mutations in both V and C regions were confirmed by sequence analyses. In the S107 cell line, spontaneous point mutations occurred in the V region at a rate of approximately 5 x 10(-5)/bp per cell generation, > 400-fold higher than the rate of V-region mutation in the NSO cell line and considerably higher than the rates in 2C3 and 36.65 hybridoma cell lines. These studies suggest that S107 is a relatively permissive cell line in which V-region mutations can occur constitutively, even though it represents a late stage of B-cell differentiation. Further, the results show that the construct used contains sufficient information in its flanking and coding sequences to allow a relatively high rate of V-region mutation, at least in the S107 cell line.

In vitro activation of a nonproductive immunoglobulin allele by a single base pair insertion

We have shown that one of the alleles in a hybridoma was nonproductive because the sequence in the N region of the heavy chain caused it to be out of frame and to terminate prematurely. This allele became productive, and the gamma 1 heavy chain that it encoded was secreted when an adenosine was inserted to produce an open reading frame. This event occurred at a very low frequency following mutagenesis of the cultured cells. This result suggests that similar sorts of events must occur in vivo when both productive and nonproductive alleles undergo frequent mutations and that new genes may be expressed in hybridomas as they are being subcloned.

Isolation of new nonsense and frameshift mutants in the immunoglobulin mu heavy-chain gene of hybridoma cells

In order to expand the experimental material available for genetic and biochemical analyses of the natural immunoglobulin genes, we have isolated a variety of mutant mouse hybridoma cell lines. Some of these mutants have partial or complete deletions of the mu gene. Other mutants have nonsense or frameshift mutations in the exons encoding the variable and the second and third constant region domains of the mu heavy chain. When combined with earlier mutant data, this collection of genotypically and phenotypically tight mutants of known sequence spans most of the 10 kb of the mu gene, providing material for a variety of studies of genetic recombination and mRNA metabolism.