DNA stability works on many different levels: the need for the organism to survive past reproduction, the need for providing a suitable genetic pool for a species, as well as allowing for changes in the DNA to ensure adaptability. There are numerous different ways how this level of stability is achieved. Our work has focused on a class of enzymes, known as DNA deaminases. These proteins are actively recruited to DNA and deaminate cytosine to uracil. One of the best characterised DNA deaminases is AID (Activation Induced Deaminase), which is required for the maturation of antibodies during an immune response. At the genetic level deamination causes a coding change, as uracil is read as thymidine rather than cytosine. At the biochemical level, this lesion causes an alteration in the DNA structure, which in turn recruits members of the DNA repair machinery to the lesion. The outcome of the DNA deaminase induced lesion, repair, recombination, or mutation is dependent on the physiological state of the cell.
As human beings are living longer and past their reproductive age, DNA instability has begun to manifest itself in the population as cancer. Understanding how DNA instability is initiated, controlled or repaired has become the cornerstone of understanding cancer biology. Initiation of DNA instability is usually derived from exogenous sources (such as UV light or smoking) or endogenous sources (such as mistakes during DNA replication or natural oxidative damage). On the other hand, it is still not well understood how non-DNA damaging agents - such as hormones - can initiate DNA lesions that lead to cancer.
Hormonal Regulation of DNA deaminases
Because of their function and enzymatic activity, it is evident that DNA deaminases must be tightly regulated. A number of recent reports have begun to dissect how the sub-cellular distribution of AID is regulated, our own efforts have also highlighted other means by which AID is controlled. Transcriptional regulation though, is one of the more prominent pathways that we have analysed. In B cells, E-box proteins, NFκB and Pax5 have been shown to bind to the AID promoter, but it remained unclear which pathways would activate AID expression in non-lymphoid cells (such as germ cells). In our recent work, we have identified that the hormone signaling pathway can regulate AID transcription independently of other signals. Ligand bound hormonereceptors can bind directly to the AID promoter and alter AID protein production, which in turn influences downstream events such as mutation and translocation. While oestrogen will activate AID transcription, progesterone will inhibit AID mRNA formation; this antithetical regulation indicates that AID is part of normal hormonal homeostasis. While the majority of our work was done in B cells, AID production could be activated up to 25 fold in non-immune tissue, such as ovaries.
The consequence of oestrogen induced AID production in B cells is excessive antibody diversification, which may lead to autoimmunity. Interestingly, it has long been known that B cell autoimmunities such as lupus are ten times more likely in women then in men, and that the levels of oestrogen can alter the overall efficacy of the immune system.
Oestrogen and DNA lesions
The panpleiotropic activities of hormones are well documented, and there is not an organ or system that is not influenced by them. On the other hand, already 50 years ago it was observed that hormones such as oestrogen can enhance, if not even cause, cancer. Yet there has been very little progress in understanding how oestrogen causes DNA lesions that lead to oncogenic mutations and translocations. There is limited evidence that oestrogen metabolites can damage DNA or even induce mutations. Our finding that the DNA deaminase, which can act as a DNA mutator, is activated by oestrogen not only in B cells but in a number of different hormonally responsive tissues, proposes a new theory (Figure 1). Here, oestrogen (even at physiological levels) will enter a cell, bind to the oestrogen receptor, transport into the nucleus, and locate to the AID promoter (or that of other DNA deaminases). This will recruit the transcription machinery, activate AID mRNA production, and initiate protein translation. This can cause misexpression of AID and allow AID to target non-physiological loci, giving rise to mutations in oncogenes and tumor suppressor genes; with some of the AID induced lesions even initiating translocations (e.g. Burkitt lymphoma, c-Myc/immunoglobulin H gene - Figure 2). Thereby, our work has shown for the first time how oestrogen can have genotoxic activity in vertebrates (only vertebrates express DNA deaminases), without causing DNA damage itself, and how oestrogen can induce DNA instability leading to cancer.
Figure 1. Direct induction of DNA instability by oestrogen. The hormone oestrogen (red circle) enters the cell and is bound by a dimeric oestrogen receptor (blue). Upon ligand binding the complete complex translocates into the nucleus and binds the AID promoter, to activate transcription. Enhanced AID protein will then deaminate cytosine residues in the genome of the stimulated cell, which can lead to mutation and translocation, and possibly cancer.
Figure 2. Schematic (left) of how AID induces c-Myc/IgH translocations. AID targets the IgH switch region (yellow) and c-Myc promoter. DNA repair then allows for the translocation, placing the c-Myc oncogene under IgH promoter control. PCR identifications (right) of c-Myc/IgH translocations from DMSO or oestrogen treated B cell cells.
In the past, tamoxifen (a synthetic oestrogen analogue) has been used to antagonise the activity of oestrogen. In our experiments tamoxifen can inhibit the mutator activity of oestrogen, but on its own tamoxifen can also act as an agonist of the oestrogen pathway. This could explain why tamoxifen treatment can lead to secondary tumour formation; by activating DNA deaminases and thereby mutating DNA.
Outlook
The discovery that oestrogen can induce DNA instability, by directly activating DNA deaminases, will provide new opportunities for identifying cancer targets and preventative measures. Our current work is dissecting the molecular mechanisms of how AID induces DNA lesions within and outside the immunoglobulin locus. Proteomics and genetic screens will identify co-factors, which alter AID activity and thereby serve as targets during prolonged oestrogen treatment. This work will have a direct impact on cancer patients and their treatments.
For a list of refereed research papers, see Publications (in navigation on left).
