Targeted disruption of the basic Krüppel-like factor gene (Klf3) reveals a role in adipogenesis.

Krüppel-like factors (KLFs) recognize CACCC and GC-rich sequences in gene regulatory elements. Here, we describe the disruption of the murine basic Krüppel-like factor gene (Bklf or Klf3). Klf3 knockout mice have less white adipose tissue, and their fat pads contain smaller and fewer cells. Adipocyte differentiation is altered in murine embryonic fibroblasts from Klf3 knockouts. Klf3 expression was studied in the 3T3-L1 cellular system. Adipocyte differentiation is accompanied by a decline in Klf3 expression, and forced overexpression of Klf3 blocks 3T3-L1 differentiation. Klf3 represses transcription by recruiting C-terminal binding protein (CtBP) corepressors. CtBPs bind NADH and may function as metabolic sensors. A Klf3 mutant that does not bind CtBP cannot block adipogenesis. Other KLFs, Klf2, Klf5, and Klf15, also regulate adipogenesis, and functional CACCC elements occur in key adipogenic genes, including in the C/ebpalpha promoter. We find that C/ebpalpha is derepressed in Klf3 and Ctbp knockout fibroblasts and adipocytes from Klf3 knockout mice. Chromatin immunoprecipitations confirm that Klf3 binds the C/ebpalpha promoter in vivo. These results implicate Klf3 and CtBP in controlling adipogenesis.

Human KLF17 is a new member of the Sp/KLF family of transcription factors.

The Sp/KLF transcription factors perform a variety of biological functions, but are related in that they bind GC-box and CACCC-box sequences in DNA via a highly conserved DNA-binding domain. A database homology search, using the zinc finger DNA-binding domain characteristic of the family, has identified human KLF17 as a new family member that is most closely related to KLFs 1-8 and 12. KLF17 appears to be the human orthologue of the previously reported mouse gene, zinc finger protein 393 (Zfp393), although it has diverged significantly. The DNA-binding domain is the most conserved region, suggesting that both the murine and the human forms recognize the same binding sites in DNA and may retain similar functions. We show that human KLF17 can bind G/C-rich sites via its zinc fingers and is able to activate transcription from CACCC-box elements. This is the first report of the DNA-binding characteristics and transactivation activity of human KLF17, which, together with the homology it displays to other KLF proteins, put it in the Sp/KLF family.

KAI1 promoter activity is dependent on p53, junB and AP2: evidence for a possible mechanism underlying loss of KAI1 expression in cancer cells.

A molecular mechanism to explain reduced KAI1 expression in invasive and metastatic tumour cells remains elusive. In this report, we extend an earlier study in bladder cells to confirm that a 76 bp region of the KAI1 promoter (residues -922 to -847), with binding motifs for p53, AP1 and AP2, is required for high level activity of a KAI1 reporter in prostate cancer cell lines. Gel shift and supershift experiments supported binding of p53, junB and heterodimers of AP2alpha/AP2gamma or AP2beta/AP2gamma to this sequence. Introduction of mutations into specific motifs demonstrated an essential requirement for p53 and junB to reporter activity, and that functional synergy between these two factors enhanced activity. A further elevation of reporter activity required AP2. Roles of individual p53, junB and AP2 proteins, as well as functional synergy between p53 and junB, were confirmed in transfection experiments. Western blotting analysis showed that an absence of wild-type p53, and/or a loss of junB and AP2 protein expression, correlated with downregulation of KAI1 mRNA levels in a series of prostate cancer cell lines. A loss of p53 function and/or expression of junB, combined with reduced expression of specific AP2 proteins may underly downregulated KAI1 expression in tumour cells.

Pentaprobe: a comprehensive sequence for the one-step detection of DNA-binding activities.

The rapid increase in the number of novel proteins identified in genome projects necessitates simple and rapid methods for assigning function. We describe a strategy for determining whether novel proteins possess typical sequence-specific DNA-binding activity. Many proteins bind recognition sequences of 5 bp or less. Given that there are 4(5) possible 5 bp sites, one might expect the length of sequence required to cover all possibilities would be 4(5) x 5 or 5120 nt. But by allowing overlaps, utilising both strands and using a computer algorithm to generate the minimum sequence, we find the length required is only 516 base pairs. We generated this sequence as six overlapping double-stranded oligonucleotides, termed pentaprobe, and used it in gel retardation experiments to assess DNA binding by both known and putative DNA-binding proteins from several protein families. We have confirmed binding by the zinc finger proteins BKLF, Eos and Pegasus, the Ets domain protein PU.1 and the treble clef N- and C-terminal fingers of GATA-1. We also showed that the N-terminal zinc finger domain of FOG-1 does not behave as a typical DNA-binding domain. Our results suggest that pentaprobe, and related sequences such as hexaprobe, represent useful tools for probing protein function.

CCHX zinc finger derivatives retain the ability to bind Zn(II) and mediate protein-DNA interactions.

Classical (CCHH) zinc fingers are among the most common protein domains found in eukaryotes. They function as molecular recognition elements that mediate specific contact with DNA, RNA, or other proteins and are composed of a betabetaalpha fold surrounding a single zinc ion that is ligated by two cysteine and two histidine residues. In a number of variant zinc fingers, the final histidine is not conserved, and in other unrelated zinc binding domains, residues such as aspartate can function as zinc ligands. To test whether the final histidine is required for normal folding and the DNA-binding function of classical zinc fingers, we focused on finger 3 of basic Krüppel-like factor. The structure of this domain was determined using NMR spectroscopy and found to constitute a typical classical zinc finger. We generated a panel of substitution mutants at the final histidine in this finger and found that several of the mutants retained some ability to fold in the presence of zinc. Consistent with this result, we showed that mutation of the final histidine had only a modest effect on DNA binding in the context of the full three-finger DNA-binding domain of basic Krüppel-like factor. Further, the zinc binding ability of one of the point mutants was tested and found to be indistinguishable from the wild-type domain. These results suggest that the final zinc chelating histidine is not an essential feature of classical zinc fingers and have implications for zinc finger evolution, regulation, and the design of experiments testing the functional roles of these domains.

Identification of regulatory regions within the KAI1 promoter: a role for binding of AP1, AP2 and p53.

The mechanism underlying loss of KAI1 gene expression in invasive and metastatic tumour cells is unknown. A possible scenario could involve altered expression or function of protein factors normally involved in regulating KAI1 transcription. To explore this possibility, we have initiated a study to characterise regulatory elements of the KAI1 promoter, using as a model, two bladder cancer cell lines (BL13 and HT1376) expressing high levels of endogenous KAI1 messenger RNA (mRNA). Transfection experiments using reporter plasmids with progressive KAI1 promoter deletions, identified a 76 bp region upstream of the transcription initiation site which contained putative binding motifs for AP2, p53 and AP1, as essential for reporter activity. DNA-binding studies using nuclear extracts from both cell lines, showed that AP1 and AP2 formed specific complexes with oligonucleotides containing KAI1 promoter motifs. Mutation of either motif abrogated reporter activity and abolished specific complex formation. In BL13 cells (endogenous wildtype p53), but not in HT1376 cells (endogenous mutant p53), mutation of the p53-binding motif also abrogated reporter activity and abolished specific complex formation in gel shift assays. These data suggested that a combination of AP2, p53 and AP1 binding to specific motifs within the KAI1 promoter might be required for high level promoter activity and that loss of expression or function of these factors might contribute to loss of KAI1 expression in invasive tumours and tumour cell lines. To explore this possibility, we examined levels of these proteins in nuclear extracts of BL13 and HT1376, as well as three bladder cancer cell lines which expressed little or no KAI1 mRNA. Our data suggested that a loss of KAI1 mRNA was not simply due to absence of AP2, AP1 or p53 expression.