Comment on Margalit et al, page 1599
Margalit and colleagues demonstrate that BCL6 expression is increased by p53, thus discovering a new feedback loop associating these 2 proteins—a finding that has numerous implications for our understanding of normal B-cell physiology and lymphoma pathogenesis.
The BCL6 proto-oncogene functions as a sequence-specific transcriptional repressor that is necessary for germinal center (GC) formation. It was originally identified by virtue of its involvement in chromosomal translocation affecting band 3q27.1 These translocations, present in approximately 40% of diffuse large B-cell lymphomas (DLBCLs), may lead to deregulated BCL6 expression by placing the intact coding sequence of the BCL6 under the influence of a potent heterologous promoter. In addition, the 5′ noncoding regulatory region of the BCL6 is targeted by somatic mutation, some of which result in deregulated BCL6 expression.2,3 Since the biological function of BCL6 in GCs is presumably to favor sustained proliferation of B cells by modulating the transcription of genes involved in cell-cycle regulation, proliferation, activation, and differentiation, it was hypothesized that the deregulated increase of BCL6 expression may lead to lymphomagenesis. Recently reported constitutive expression of BCL6 in mice lymphocytes did recapitulate the pathogenesis of human DLBCL.4 In addition, recent studies further elucidated the potential mechanism of BCL6-induced lymphogenesis by demonstrating that BCL6 down-regulates p53 expression and hence may allow lymphoma cells to escape apoptosis in response to DNA damage.5 In this issue of Blood, Margalit and colleagues report on a new interaction between the guardian of the genome—p53 and BCL6.
While examining the transcriptional program regulated by p53 using DNA arrays, the authors found up-regulation of BCL6 expression. Evaluation of the BCL6 5′ regulatory region, frequently deranged by translocations and mutations, disclosed the presence of a putative p53 response element (p53RE). In a stepwise fashion, the authors demonstrate in vitro and in vivo binding of the p53 to the BCL6 p53RE as well as activation of transcription by binding to this response element. Furthermore, the authors demonstrated p53-mediated chemotherapy-induced increase in BCL6 mRNA and protein expression. These findings complement the previous observation of Phan and Dalla-Favera,5 and suggest the presence of a regulatory loop in which p53 increases expression of BCL6, which in turn suppresses the expression of p53 (see figure). Similar loops exist for BCL6 and BLIMP1,6 in which each partner suppresses the other, as well as for p53 and its negative regulator MDM2. The identification of this new p53-BCL6 feedback loop is clearly a major step toward understanding how the balance between these key proteins promotes the development of lymphoma and how the GC milieu may permit the occurrence of chromosomal breaks and somatic mutations. However, an important and yet unaddressed question is whether the 2 portions of the proposed regulatory loop (eg, BCL6-induced repression of the p53 and the p53-induced expression of BCL6) are functional in the same cell, or whether the functionality of each branch of the loop is cell-context dependent. This question is raised by the following observations: (1) in normal GC cells there is high expression of BCL6 and low expression of p53;5 (2) while Phan and Dalla-Favera5 demonstrated a decrease in BCL6 and increase in p53 protein levels in response to chemotherapy in GC-derived lymphoma cells, Margalit and colleagues showed chemotherapy-induced increase in both the p53 and BCL6 in non-GC–derived cells. Addressing this question as well as proving that each of the branches of the proposed p53-BCL6 regulatory loop is crucial for the development of B-cell lymphomas will be the next important step. ▪