Many Copies of V Genes and Only a Few C Genes
The patterns of sequence variability of the first 100 residues of the L and H chains and the constancy of the remainder of these chains suggested that the genes for L and H proteins were split into two parts. The V regions could then be encoded by any one of a set of different V region segments; upon differentiation into lymphocytes, one particular V re-gion segment would be connected to the appropriate constant region segment to form a gene coding for an entire L or H chain. Such a proposal should be readily verifiable by the techniques of genetic engineering.
Cloning DNA coding for the V and C regions was not difficult. Since antibodies are secreted, they are synthesized by ribosomes attached to the endoplasmic reticulum of lymphocytes. Therefore, isolation of poly-A RNA from membranes of mouse myeloma cells yielded an RNA fraction greatly enriched for antibody messenger. This was then con-verted to cDNA with reverse transcriptase and cloned. Sequencing of the clones provided positive identification of those that contained the immunoglobulin genes. While this approach provided DNA copies of the mature RNA that is translated to form immunoglobulin, it did not directly indicate the status of the portion of the genome that encoded the RNA.
Southern transfer experiments done by Tonegawa and by Leder using fragments of cloned cDNA as a probe showed three facts about the genes involved with immunoglobulin synthesis. First, any single V region sequence possesses detectable but variable homology with up to about 30 different V region segments in the genome. The total number of V region coding segments in the genome can then be estimated from the numbers of such segments sharing homology to more than one V region probe. Results of such experiments indicate that the genome of the mouse contains from 100 to 1,000 kappa V region genes, with the most likely number being about 300. The second result from the Southern transfer experiments was that the genome contains much smaller num-bers of the C region genes. The mouse genome contains four of the lambda subclass C region L genes and only one of the kappa subclass. The astounding third result obtained from Southern transfers was that the genome in the vicinity of a V and C region gene undergoes a rearrangement during differentiation from a germ line cell to a lympho-cyte. That is, the size of DNA fragment generated by digestion with a restriction enzyme and containing homology to a V region differed between DNA extracted from embryos and DNA extracted from the myeloma source of the cDNA V region clone.
Answering deeper questions about the structure and rearrangements of the genes involved with antibody synthesis required additional clon-ing. Copies of undifferentiated V and C region genes could be obtained by cloning DNA isolated from mouse embryos. The differentiated copies could be obtained by cloning from myelomas. DNA extracted from mouse embryos and myelomas was cloned in lambda phage, and the plaques were screened for those containing DNA homologous to the V or C region from the cDNA clone. The results of these experiments in conjunction with the Southern transfer experiments showed that in the germ line cells, the C and V region genes are situated far from one another. In a myeloma cell, however, a DNA rearrangement has placed the expressed V and C region genes beside one another. After the rearrangement, they are separated by only about 1,000 base pairs. An R-loop between light chain mRNA and the rearranged DNA shows the V and C region genes separated by an intervening sequence plus an additional portion of V region gene called V’. This codes for the leader sequence that is required for secretion of the polypeptide.