Full Article

Pseudoprimers Cause Somatic Hypermutation of Ig Genes

A. M. Olovnikov

Institute of Biochemical Physics, Russian Academy of Sciences, ul. Chernyakhovskogo 5-94, Moscow, 125319 Russia; fax: (095) 214-9269; E-mail: am@olovnikov.msk.ru

Received November 6, 1998
It is proposed that hypermutation of the Ig gene is based on the use of "misprimers" (MPs) capable of competing with a true primer for the DNA. The MP is a product of the cleavage of the nascent transcript. Each MP has an erroneous base on its 3´ terminus. Wobbling of the unpaired 3´ end of the MP forces the DNA polymerase to make an error.
KEY WORDS: antibody, hypermutation, pseudogenes, adaptive mutation

Hypermutation of the variable (V) region of rearranged immunoglobulin (Ig) genes is important for creating high affinity antibodies. In spite of long-term endeavors, the exact mechanism of hypermutation is still elusive [1-3]. Here I propose the hypothesis that hypermutation is based on the usage of special “pseudoprimers”, or, in other terms, “misprimers” (MPs), during replication of the V region of the Ig gene. It is known that RNA polymerase may work with errors [4, 5], whereupon the transcription complex stops (the phenomenon is known as RNA polymerase “pausing”) and cleaves a fragment from the nascent RNA [6]. One can assume that there should be an erroneous nucleotide at the very 3´ terminus of this RNA fragment. I suppose that this oligoribonucleotide, being complementary to the sense strand of its own gene, should be able to compete for its DNA template with the true RNA primer produced by a primase, and that is why it is designated here as a misprimer (MP), i.e., pseudoprimer. The unpaired 3´ terminus of a MP, when the MP substitutes by itself the true primer at the template, is able to initiate the erroneous inclusion of non-complementary 5´ terminal deoxyribonucleotide into the nascent Okazaki fragment. Mutation itself appears in a growing DNA replica due to the wobbling of the unpaired 3´ end of the MP. This wobbling forces the DNA polymerase to make an error, producing mutation (point substitution or, rarely, deletion or insertion) at the 5´ terminus of the DNA in the nascent Okazaki fragment. It is supposed that just such kind of mutations remain unnoticed by the correction system during the joining of DNA fragments into the united lagging strand. This process underlies the mechanism of Ig gene hypermutation.

To ensure success of “misprimer mutagenesis” of a given gene, transcription of the gene should immediately precede its replication because of the short lifetime of any MPs in the presence of RNases. MP-dependent mutagenesis is coupled predominantly with replication of the DNA lagging strand, thus resulting in asymmetry of the mutation frequency within DNA strands [1, 3]. Nevertheless, the corresponding MPs can, in principle, compete also with those true primers, which are used during synthesis of the leading strand.

The proposed novel type of mutagenesis may also be responsible for the appearance of the so-called "adaptive mutations" emerging in microorganisms under extreme for them conditions in the course of intense transcription of some genes [7]. In multicellular organisms, transcripts from pseudogenes could intentionally perform the role of mutagenic MPs in the course of replication of structural genes that are homologous to the corresponding pseudogenes. In addition, misprimer mutagenesis might serve as a fundamentally new factor of carcinogenesis.

An independent issue is the question, which processes happen in cells of various origin and differentiation in the course of creation of MPs? What in common in this sense do B lymphocytes and bacteria have? To answer this question, I would like to propose one more hypothesis aimed at explaining the pathways of origin of MPs under so distinct conditions.

Whereas the above suggested major idea of the present communication deals with the immediate cause of transcription-related mutagenesis, the second, complementary hypothesis proposed below is aimed to answer the following question. What could force the transcriptional machinery to make errors? It is known that transcriptional factors, acting in complex with RNA polymerase, decrease the frequency of its errors [4-6]. Taking this into account, one can assume that quite different pathways of distortions of the optimal protein content of a transcriptosome could reduce the fidelity of the work of RNA polymerase and increase the frequency of appearance of erroneous transcripts as potential sources of MPs. For microorganisms, extreme for them conditions of cultivation could be the regulatory factors changing the pattern of transcriptosome proteins. This would cause the appearance of mutations in many different transcribed and copied genes, rather than in one gene. This would bring a certain set of mutations, a part of which originating in non-random genes could be beneficial for survival of the population of microorganisms. As for B lymphocytes, the same principle, namely the deliberate distortion, for the sake of creation of MPs, of the normal mode of the transcriptosome activity may be kept, however radically modified. In B lymphocytes, it is necessary to introduce mutation into the rearranged Ig gene, rather than in any transcribing genes. Therefore, it is utterly undesirable to alter the protein pattern of transcriptosomes in B lymphocytes, since in the latter case the undesirable MPs would appear in the course of transcription of non-Ig genes. One could suppose that in case of B lymphocytes, a possibility to modify the transcriptosome in a strictly localized manner was realized. It was shown that artificial substitution of immunoglobulin enhancers for other ones decreases the efficacy of Ig hypermutation, whereas replacement of the promoter does not exert such an effect [1-3]. Hence, the Ig gene enhancers could apparently be somehow responsible for the appearance of errors in the course of transcription of the V region of Ig genes. Recently I put forward a proposal about the existence of a special mechanism of ionic modulation of gene expression in eukaryotes [8]. Let us assume here that in B lymphocytes a very short-term and local influx of Ca²⁺ ions takes place in close vicinity of the V region of the Ig gene. This ionic shot, which acts locally and lasts for a short time, will produce a strictly localized and short-term distortion in the functioning of the transcriptosome. The above mentioned problem could apparently be solved in this way. Flanking of the V region of the Ig gene by long spacers is, probably, necessary just for prevention of the adjacent genes from the action of the locally and deliberately modified transcription machinery.

This work was supported by the Russian Foundation for Basic Research (Grant 98-04-48654).

REFERENCES

1.Milstein, C., Neuberger, M. S., and Staden, R. (1998) Proc. Natl. Acad. Sci. USA, 95, 8791-8794.
2.Storb, U., Klotz, E. L., Hackett, J. Jr., Kage, K., Bozek, G., and Martin, T. E. (1998) J. Exp. Med., 188, 689-698.
3.Winter, D. B., and Gearhart, P. J. (1998) Immunol. Rev., 162, 89-96.
4.Jeon, C., and Agarwal, K. (1996) Proc. Natl. Acad. Sci. USA, 93, 13677-13682.
5.Thomas, M. J., Platas, A. A., and Hawley, D. K. (1998) Cell, 93, 627-637.
6.Gu, W., and Reines, D. (1995) J. Biol. Chem., 270, 30441-30447.
7.Hall, B. G. (1998) Genetica, 102/103, 109-125.
8.Olovnikov, A. M. (1997) Int. J. Dev. Biol., 41, 923-931.