Negative Regulation of Activation-Induced Cytidine Deaminase Protein Prevents Aberrant Somatic Hypermutation and Lymphomagenesis.

Abstract 94

Somatic hypermutation (SHM) is a natural process that introduces point mutations into immunoglobulin (Ig) genes during antibody affinity maturation. During the first phase of SHM, the enzyme activation-induced cytidine deaminase (AID) converts cytosine (C) to uracil (U) to result in a U-G mismatch. Spontaneous U-G mismatches are normally corrected by high-fidelity DNA repair pathways. However, during the second phase of SHM, U-G mismatches are processed by low-fidelity DNA repair pathways to yield mutations. These second phase pathways are initiated by recognition of the uracil by uracil DNA glycosylase (UNG) and MSH2/MSH6. As a DNA mutator, AID poses a direct threat to genomic integrity, a notion supported by reports demonstrating that aberrant targeting of AID contributes to translocations and point mutations of proto-oncogenes associated with B cell malignancy. Transgenic mice with constitutive and ubiquitous expression of Aid (Aid-Tg) provide a unique system for investigating the poorly understood dynamics of Aid targeting and DNA repair. In these animals, development of B and T lymphocytes is indistinguishable from that of wild-type littermates. However, within 5 to 7 months, a subset of Aid-Tg mice develop monoclonal T cell lymphomas with high levels of SHM of the T cell receptor locus (Tcr), cMyc, Pim1, Cd4 and Cd5 genes. Interestingly, these animals do not develop B cell malignancies, suggesting that B cells have a specific mechanism for regulating Aid activity and/or high-fidelity versus low-fidelity DNA repair of non-Ig genes. To further investigate this mechanism, we crossed Aid-Tg mice with Ung^−/−Msh2^−/− double knockout mice. In the absence of Ung and Msh2, Aid-generated U-G mismatches are not repaired and are simply replicated, leaving a footprint of Aid targeting in the form of C/G to T/A transition mutations. Our plan was to compare Aid targeting (Aid-Tg Ung^−/−Msh2^−/−) and DNA repair (Aid-Tg) patterns between ‘normal' splenic B and T cells, prior to the development of genomic instability and malignant transformation. Splenic B and T cells from 4-month-old Aid-Tg and Aid-Tg Ung^−/−Msh2^−/− mice were obtained by fluorescence activated cell sorting. Genomic DNA was prepared and a ∼1-kilobase (kb) region spanning the first 1.5-kb downstream of the major promoter of cMyc, Pim1 and H2afx was PCR-amplified and sequenced (∼40 to 60-kb/gene). Surprisingly, mutation frequencies of all genes were found to be extremely low in all cell types (range, 8.4 × 10⁻⁶ to 7.8 × 10⁻⁵ mutations/bp) and well below the mutation frequencies of 4 × 10⁻⁴ to 9 × 10⁻⁴ mutations/bp we have previously seen for these genes in germinal center B cells from Ung^−/−Msh2^−/− mice. In fact, these mutation frequencies approach the background mutation frequency attributable to the sequencing procedure of 1.6 × 10⁻⁵ mutations/bp (previously determined from Aid^−/− mice). Additional sequencing in the T cells also did not demonstrate any mutations in Cd4 or Tcrb, contrasting the high mutation frequencies found in tumors from Aid-Tg mice. To explore the possible cause of these unexpected results, we proceeded with further analysis of Aid expression in our system. Resting splenic B and T cells were obtained from wild-type and Aid-Tg mice and divided into one group that was used to make resting total cellular RNA and protein lysates and a second group that was activated ex vivo with lipopolysaccharide and IL4 (for B cells) or anti-CD3/CD28 beads and IL2 (for T cells). Real-time RT-PCR demonstrated high and equal expression of Aid transcript in resting and activated Aid-Tg B and T cells that was ∼3 to 4-fold higher than activated wild-type B cells. Analysis of Aid expression by Western blotting demonstrated that Aid protein is expressed at equal levels in activated wild-type and Aid-Tg B cells but is undetectable in resting wild-type B and T cells, resting Aid-Tg B and T cells, and activated Aid-Tg T cells. Thus, the Aid transgene is highly transcribed but likely fails to initiate SHM of known target genes in splenic B and T cells because Aid protein is absent. Our findings suggest that, in this model, B and T cells have an internal mechanism for negatively regulating Aid protein, preventing aberrant SHM and malignant transformation. Further investigation is needed to determine whether Aid protein is regulated at the level of mRNA translation or protein stability and how this mechanism breaks down during the pathogenesis of lymphoid malignancy.