Research Highlights

Histones on LSD

Covalent histone modification, such as phosphorylation, acetylation, methylation and ubiquination, marks the transcription state of genes. These markings are dynamic and reflect the constant reprogramming of gene expression in response to various cellular signals. Histone methylation was thought to be stable and to change only with histone turnover, but recent studies provide evidence for the existence of enzymes that remove methyl groups from modified lysines and arginines in histones. LSD1 (also known as BHC110) has been shown to specifically demethylate modified Lys4 of histone H3 in isolation but has no activity toward modified H3 in the context of the intact nucleosome. LSD1 is part of a multiprotein corepressor complex that also contains histone deacetylases and a protein called CoREST. Lee et al. show that in the corepressor complex, LSD1 has a higher demethylase activity toward modified H3 Lys4 than does LSD1 alone. In addition, the complex demethylates nucleosomal H3 Lys4. This activity requires the CoREST component, which enhances the association between LSD1 and the nucleosome. In a related study, Metzger et al. show that LSD1 associates with the androgen receptor to stimulate transcription of target genes in response to androgen. Notably, the LSD1–androgen receptor complex demethylates a different modified lysine of H3, Lys9, to activate transcription. These results support the idea that histone demethylases can regulate the level of histone methylation. They further show that both the activity and the specificity of LSD1 are fine-tuned by its association with either CoREST or the androgen receptor. Discovering how this is achieved will now require additional studies. (Nature, advance online publication 3 August 2005, 10.1038/nature04021 and 10.1038/nature04020) HPF

Degrading Wee1

Upon entering mitosis, the cyclin-dependent kinase's (CDK's) inhibitory kinase, Wee1, is downregulated primarily through proteasome degradation mediated by the E3 ubiquitin ligase SCF^β-TrCP. β-TrCP is the substrate-recognition component of the ligase, which binds its substrates through a conserved phosphodegron motif (PD). Phosphorylation of Ser53 and Ser123 of Wee1 by polo-like kinase 1 (Plk1) and CDK, respectively, is required for SCF^β-TrCP-mediated degradation. Whereas pSer53 lies within a PD similar to that required for β-TrCP binding, pSer123 does not, so its role in Wee1 destabilization has been unclear. Watanabe et al. now show that CDK phosphorylation of Ser123 promotes Wee1 degradation in a few different ways. pSer123 can help recruit Plk1 to Wee1 for Ser53 phosphorylation but is not essential for this activity. Modeling of the sequences flanking Ser123 on the basis of the structure of a β-catenin phosphopeptide–β-TrCP complex suggests that pSer123 and a second phosphorylation site at Ser121 participate in β-TrCP interactions, but competition assays show that pSer121 is more important for binding. In fact, pSer121 and the sequences adjacent to it are part of a second PD suitable for β-TrCP recognition. The authors found that pSer123 is required to prime Ser121 for phosphorylation by the ubiquitously expressed serine/threonine kinase CK2. Inhibition of CK2 slows Wee1 degradation and mitotic progression. Thus, Ser123 phosphorylation by CDK facilitates SCF^β-TrCP-mediated Wee1 turnover by interacting directly with β-TrCP and by promoting the creation of two PDs for β-TrCP recognition. These results suggest that the pSer121 PD helps maintain basal levels of Wee1 during interphase, but that the second PD at Ser53, created when Plk1 is activated at the beginning of mitosis, promotes more efficient degradation of Wee1. (Proc. Natl. Acad. Sci. USA, published online 16 August 2005, 10.1073/pnas.0500410102) MM

Stopping at the start

MicroRNAs (miRNAs) repress gene expression by base-pairing with the 3′ untranslated regions of target messenger RNAs. Although the exact mechanism of this repression is unknown, it could be accomplished either by inhibiting translation or by inducing proteolysis of the nascent polypeptide. In a recent paper, Pillai et al. explored how miRNAs inhibit protein accumulation in mammalian cells, using reporter mRNAs whose translation is regulated by two different mechanisms. They found that the let-7 miRNP regulates protein accumulation by inhibiting translation rather than by proteolysis. Using a variety of experiments to pinpoint the step of translation affected by the let-7 miRNP, the authors show that initiation of translation at the 5′-terminal m⁷G cap is repressed by this miRNP but that initiation of translation at an internal ribosome entry site is not. The data indicate that the let-7 miRNP interferes with the function of the m⁷G cap and possibly with its recognition by eIF4E. The authors also investigated the cellular localization of the mRNAs that accumulate after the inhibition of translation. They show that these mRNAs, miRNAs and Argonaute proteins accumulate in cytoplasmic processing bodies, where they are stored or possibly degraded. This work sheds new light on the mechanism of gene silencing by miRNA and reaffirms the importance of processing bodies in miRNA-mediated repression. (Science, published online 4 August 2005, 10.1126/science.1115079) DM

let miRNA degrade

In Caenorhabditis elegans, two small temporal RNAs, lin-4 and let-7, encode miRNAs that are central in controlling the timing of gene expression and differentiation during development. lin-4 regulates larval development by inhibiting the expression of LIN-14 and LIN-28, proteins that specify developmental switches between the early larval stages. let-7 regulates lin-41, which encodes a protein controlling temporal cues to divide or terminally differentiate. Both the lin-4 and let-7 miRNAs are expressed just before the downregulation of their targets and are proposed to regulate translation in a manner dependent on the amount of base pairing between the miRNAs and the 3′ untranslated regions of their target mRNAs. However, the mechanism of miRNA-mediated regulation is not clear. Bagga et al. have analyzed genetically defined targets of lin-4 and let-7 in C. elegans and found that regulation of mRNA levels may also contribute to the inhibition of protein expression. The authors show that expression of lin-4 results in reduced levels of lin-14 and lin-28 mRNA. Furthermore, let-7 promotes the degradation of lin-41 mRNA, and this requires the 3′ untranslated region of lin-41 and 5′→3′ exonucleases. These nucleases recognize exposed 5′-monophosphate–containing mRNAs generated by decapping or endonucleolytic cleavage, and they colocalize in cytoplasmic processing bodies with decapping enzymes, Argonaute proteins and mRNA targets of miRNAs. The authors propose that lin-4, let-7 and perhaps other miRNAs control gene expression by directing their targets to processing bodies for degradation. The authors suggest that the outcome of any particular miRNA–target interaction likely depends on many factors besides the level of base pairing between the two RNAs. (Cell, published online 11 August 2005, 10.1016/S0092867405008019). EJ

Research highlights written by Hwa-ping Feng, Evelyn Jabri, Michelle Montoya and Dorothy Moore.