Functional analysis of KLF2 and its lymphoma-derived mutants
Splenic marginal zone lymphoma is a low-grade B cell neoplasm, which originates from the splenic marginal zone B cells. Whole exome studies show recurrent somatic mutations in KLF2, and genes in the NOTCH2, NF-κB, BCR and TLR signalling pathways – pathways that are essential for the development and maintenance of marginal zone B cells. Of these, KLF2 mutations are the most frequent (20-42%) and are thought to lead to the inactivation of KLF2. KLF2 is a zinc finger transcription factor involved in a variety of cellular processes, including negative regulation of NF-κB activation. Functional assays show that KLF2 can suppress NF-κB activation, through both the canonical and non-canonical pathways. However, mutations in KLF2 lead to a loss in NF-κB suppression, which may promote the lymphomagenesis by facilitating differentiation of marginal zone B cells and their homing to the splenic marginal zone.
We hypothesised that KLF2 may also modulate other pathways that are important in marginal zone B cell development, namely the NOTCH2 signalling pathway. In vitro reporter assays showed that wild type KLF2 suppresses the NOTCH2 signalling pathway. Introducing mutations that are recurrently seen in splenic marginal zone lymphoma patients into KLF2 resulted in aberrant regulation of NOTCH2 signalling. All three C-terminal truncation mutants and 8/10 lymphoma derived missense/ indel mutants showed a loss of NOTCH2 suppression. Interestingly, KLF2 C274Y, C279Y, H292Y, H296Y and S287P mutants displayed enhanced reporter activities, suggesting gain of function. Wild type KLF2 was also found to be capable of suppressing the activation of the MAPK/ ERK signalling pathway and both truncation and missense mutations in KLF2 resulted in a loss of this repression in a similar manner to the NOTCH2 pathway, with several mutants, including KLF2 C274Y, C279Y, H292Y and H296Y showing gain of function activity. This, along with previous studies, suggests that KLF2 may be acting as a global negative regulator of transcription. Mutations in KLF2 can lead to a loss of this repressive activity and, in the case of several missense mutants, additionally show a gain of function, which may contribute to lymphomagenesis.
Confocal imaging analysis revealed remarkable differences in the subcellular localisation between wild type and mutant KLF2. While wild type KLF2 localises in the nucleus, we showed that truncations in KLF2 disrupted its nuclear localisation signals resulting in predominant localisation throughout the cytoplasm. Of the ten missense/ in frame deletion mutants investigated, seven displayed a “speckled” pattern of nuclear expression, with those that showed gain of function reporter activities producing this speckled pattern at a significantly higher rate. Further analysis suggested colocalisation between mutant KLF2 and SRSF2, a component of the spliceosome, suggesting that KLF2 may also be involved in splicing and RNA processing activities.
Gene expression profiling showed that wild type and mutant KLF2 regulate many of the same genes that are involved in regulating B cell development and differentiation. Additionally, WNT signalling was identified as being differentially expressed when mutant KLF2 is expressed in comparison to wild type KLF2. Gene set enrichment analysis also showed that the WNT, BCR, chemokine, MAPK and JAK-STAT pathways were all significantly impacted by both wild type and mutant KLF2. Interestingly, KLF2 missense mutants showed additional effects on the TLR and NOTCH2 signalling pathways, in particular, many positive regulators of these pathways are upregulated with mutant KLF2 and downregulated with wild type KLF2. In conclusion, these studies have shown that KLF2 may be acting as a global negative regulator, particularly of pathways that are essential in the development of marginal zone B cells. Mutations in KLF2 not only abolish this suppression but, in several cases, lead to gain of function activities. These mutations caused aberrant localisation of KLF2, which suggested that KLF2 may be interacting with the spliceosome and its regulators to control gene expression.