Location, location, location – no more! The unravelling of chromatin remodeling regulatory aspects of plant metabolic gene clustersписьмо
Аннотация: 'These data … provide the first evidence that this mechanism is important in the regulation of plant gene clusters and elevate our understanding of the regulation of these clusters beyond a mere description of the physical proximity of their constituent genes.' Gene clusters, also commonly known as 'operons', are a common feature of prokaryotes but are also present in eukaryotes. The term operon was first used in the seminal paper of Jacob and Monod on the discovery and characterization of the lactose utilization gene cluster, lac operon, of Escherichia coli (Jacob et al., 1960). True operons not only contain genes of similar function and common regulatory control but are also transcribed as a single polycistronic mRNA from a single promoter (Jacob et al., 1960). Gene clusters have been reported in fungi, insects, plants, vertebrates and yeast (Hurst et al., 2004), however, those found in eukaryotes tend to be transcribed as independent mRNAs (see, for example, Collemare et al., 2008). Although clustering of paralogous genes with high-sequence homology is common in plants, clustering of nonhomologous genes of similar function has only relatively recently been described (Frey et al., 2009). Indeed, such clusters have to date been confined to specialized metabolism. Clusters of genes encoding specialized metabolic pathways have been described in a wide range of species including oat, rice, cassava, opium poppy, sorghum, tomato, potato, lotus and Arabidopsis (Boycheva et al., 2014). Before discussing the significance of the link between chromatin remodelers and plant gene clusters it seems prudent to provide a short review of the metabolic pathways for which plant gene clusters have been described. Given that these have been comprehensively reviewed recently (Boycheva et al., 2014) we will not discuss them in detail; however, it is important to grasp the diversity of the pathways involved (see Fig. 1). The first such gene cluster identified in plants was that involved in the biosynthesis of the benzoxazinoid 2,4-dihydroxy-7-methoxy-1,4-benzoxazin-3-one (DIMBOA) in maize. Initially, five genes (Bx1–5) were identified on chromosome 4; however, a sixth gene Bx8 was later found to co-localize and Bx6 and Bx7 are further away on the same chromosome (Boycheva et al., 2014). Importantly, both DIMBOA and its precursor 2,4-dihydroxy-1,4-benzoxazin provide pathogen resistance and deter herbivory (Sue et al., 2011). Interestingly, wheat and rye contain partially clustered forms of the pathway, as do some wild barleys, but in cultivated barley this locus has been lost (Sue et al., 2011). Similarly, genes responsible for the defense-related cyanogenic glycosides linamarin, lotaustralin and rhodiocyanoside have been found to be clustered in lotus and cassava whilst genes encoding dhurrin biosynthesis are clustered in sorghum (Boycheva et al., 2014). More recently, a 10-gene cluster for noscarpine, an important antitumor agent, in opium poppy (Winzer et al., 2012) and seven or nine genes of the steroid glycalkaloids – antinutritional compounds which can cause neurological and intestinal dysfunction – in potato and tomato, respectively (Itkin et al., 2013). Most of the clusters characterized to date, however, encode pathways of terpene synthesis with clusters being found in rice, Solanum species, oat, lotus and, as already mentioned, Arabidopsis (Boycheva et al., 2014). Unfortunately, we only have space to focus on the last species on the list here, despite the fact that, unlike thalianol and marneral, the metabolic function and bioactivities of many of the other compounds are known rendering them potentially considerably more interesting. The thalinol and marneral gene clusters are involved in the biosynthesis and subsequent modification of these two triterpenes (Field & Osbourn, 2008; Field et al., 2011), both of which derive from 2,3-oxidosqualene. The corresponding clusters, like that found in oat and already described, contain oxidosqualene cyclases – thalianol synthase and marneral synthase, respectively. Thalianol is further converted to thalian-diol by the concerted action of hydroxylase and desaturase activities, whilst the function of a BAHD acyl transferase which co-expresses with these three genes is as yet unclear (Field & Osbourn, 2008). In the case of marneral, the reactions are oxidase and desaturase, respectively (Fig. 1; Field et al., 2011). Intriguingly, overexpression of either thalianol or marneral synthases results in a dwarf phenotype whilst their overexpression provokes root growth and delays flowering (Field & Osbourn, 2008; Field et al., 2011). Despite these results suggesting an important function of these metabolites, mechanistic insight into the regulation of their biosynthesis has, to date, been lacking. However, it is important to note that this is not specific to the Arabidopsis examples we are focusing on here but rather a general short-coming of work on clustered plant genes. As a first step to assessing the contribution of chromatin to the regulation of the thalianol cluster, Nützmann & Osbourn (2014a) evaluated the levels of the thalianol hydroxylase gene across a range of Arabidopsis thaliana mutants defective in histone modifications and chromatin remodeling, finding it altered in expression in five of the mutants – most noticeably in the arp6 mutant (Nützmann & Osbourn, 2014b). ARP6 is part of the SWR1 chromatin remodeling complex required for the incorporation of H2A.Z into nucleosomes. This is an exciting finding given that H2A.Z deposition activates the DAL gene cluster of yeast (Meneghini et al., 2003). Subsequent experiments on arp6 revealed that the expression of all four genes of the cluster, but not of the genes that directly flank the cluster, were reduced as was the accumulation of thalianol. These data led Nützmann & Osbourn (2014a) to investigate the hta9/hta11 Arabidopsis mutant defective in two of the three A. thaliana H2A.Z genes and demonstrated to phenocopy apr6. Similar results were obtained as for apr6 thus implicating H2A.Z in ARP6-mediated cluster regulation. Chromatin immunoprecipitation (ChIP) experiments using transgenic HTA11:GFP targeted at the transcriptional start sites of the thalianol cluster genes revealed significantly higher deposition of HTA in tissues in which the cluster was expressed. Significantly, this deposition spanned the entire gene cluster but was not found in genes flanking the clusters and as such differs from the pattern of deposition seen in nonclustered eukaryotic genes (Guillemette et al., 2005). Importantly, similar results were found for the other gene cluster characterized in Arabidopsis – the marneral cluster. The use of micrococcal nuclease accessibility assays followed by quantification of undigested DNA facilitated the identification of nucleosome positions and stability. In seven of eight nucleosome positions analyzed, an increased occupancy in the arp6 mutant roots was found in comparison with the control. These data thus suggest that the ARP-mediated H2A.Z incorporation into nucleosomes within the gene clusters leads to localized opening of the chromatin structure and thereby facilitates cluster expression (depicted in Fig. 2). As such they provide the first evidence that this mechanism is important in the regulation of plant gene clusters and elevate our understanding of the regulation of these clusters beyond a mere description of the physical proximity of their constituent genes. The results presented in Nützmann & Osbourn (2014a) are not without precedence in plants since the Osbourn group has previously demonstrated that the avenacin cluster in oat alternates between a highly condensed and open chromatin structure on the transition from repressed to active states (Wegel et al., 2009). That said the results of Nützmann & Osbourn (2014a) are crucial in providing a handle by which to leverage underlying mechanistic aspects of the regulation of these gene clusters. Several important specific and general questions remain open. The major Arabidopsis specific questions relate to the nature of the transcription factors that act on the open chromatin structures of the thalianol and marneral cluster and the precise function of these triterpenoids in planta as well as understanding the mechanisms which allow basal expression of the genes encoded by these clusters in the arp6 mutant. In addition, the generality of chromosome remodeling as a regulatory mechanism needs to be tested in the many other species for which gene clusters have been described. Given the close resemblance of the Arabidopsis and yeast regulatory mechanisms it is highly tempting to extrapolate that the other clusters will be similarly controlled; however, this needs to be empirically established. Finally, since chromatin regulation has also been shown to affect the synthesis of compounds such as phenylpropanoids, glucosinolates and gibberellins (Nützmann & Osbourn, 2014b and references cited therein; Sarnowska et al., 2013), which are not encoded by clustered genes, the wider role of this mechanism in modifying metabolism should be borne in mind in studies intent on comprehensively understanding or manipulating plant specialized metabolism. Work on secondary metabolism in the laboratory is funded by the European Commission's Directorate-General for Research within the 7th Framework Program (FP7/2007-2013) under grant agreements 270089 (MULTIBIOPRO), and by a Deutsche Israeli project administered by the Deutshe Forschungsgemeinshaft (Project FE 552/12-1).
Год издания: 2014
Авторы: Alisdair R. Fernie, Takayuki Tohge
Издательство: Wiley
Источник: New Phytologist
Ключевые слова: Plant Gene Expression Analysis, Plant Molecular Biology Research, Plant biochemistry and biosynthesis
Другие ссылки: New Phytologist (HTML)
PubMed (HTML)
PubMed (HTML)
Открытый доступ: closed
Том: 205
Выпуск: 2
Страницы: 458–460